Breeding and Improvement

~MsGRAW-HILLPUBLICATIONS IN THE
AGRICULTURAL SCIENCES
R. A. BRINK, Consulting Editor
BREEDING AND IMPROVEMENT

OF FARM ANIMALS
McGRA W-HILL PUBLICATIONS IN-THE
AGRICULTURAL SCIENCES
R. A. BRINK, CONSULTING EDITOR
ADRIA:\"CE AND BRISON' Propagation of Horticultural Plants

AHLGRE);J . Forage Crops
Al'iDERi'\O:-O . Dis('ases of Fruit Crops
BROWN' Cotton
CARROLL AND KRIDER' Swine Production
CRUESS' Commercial Fruit and V('getable Products
DICKSON . Dis('ases of Field Crops
ECKLES, Cm.ms, AND MACY . Milk andl'Iilk Products
ELLIKER . Practical Dairy Bacteriology
FERXALD AND SHEPARD' Applied Entomology
GARD'NER, BRADFORD, AND HOOKER' The Fu~I::lIll('ntals of Fruit Production'
GUSTAFSOX . Conservation of the Soil
GCSTAFSOX . Soils and Soil Management
HAYES, IMMER, A;I;D S:lHTH . Methods of Plant Brf'cdin!(
HEALD' Manual of Plant Discases
HEALD' Introduction to Plant Pathology
HERRINGTO!'l . Milk and Milk Processing
HUTT' Genetics of the Fowl
JE!'INY . Fact.ors of Soil Formation
JULL' Poultry Husbandry
LAURIE AND RIES . Floriculture
LEACH' Insect Transmission of Plant Dis('ases'
MAYNARD' Animal Nutrition
:\iETCALF, FLI;I;T, AXD METCALF' Destructiv(' and Useful Insects
XEVEXS . Principles of Milk Production
PATgRSON . Statistical Technique in Agricultural Research
PETERS AND GRt':lnlER . Livestock Production
RATHER AND HARRISON' Field Crops
RICE AND ANDREWS' Bre('ding and Improvl'mcnt of Farm Animal8
ROADHOUSE AND HEXDERSO!'l . The l\iarket-milk Industry
ROBBINS, CRAFTS, AND RAYNOR' \V('('d Control
RCHILLETTER AND RICHEY' Textbook of Gl?npra],!'F~lifmTl'
STEIXHAUS . Principles of Insect PathologJ1 .,~~ ••
Tno;'\IPsoN . Soils and Soil Fertility '.
THO~!p'SON . Veg('table Crops
WALKER' Diseases of Veg('table C~ops
WALKEll . Plant Pathology .
WILSO!'l . Grain Crops
WOLFE AXD KIl'PS . Production of Fidd Crops
The late Leon J. Cole was Consulting Editor of this series from 1937 to 1948.
Th('re are also the relatcd scries of McGraw-Hill Publications in the Botanical Science~,
of which Edmund W. Sinnott is Consulting Editor, arid in the Zoological Sci~ces, of
which Edgar J. Boell is Consulting Editor. Titl('s in the Agricultural Scien!es were
publishrd in these series in the period 1917 to 1937,
Breeding and lInprovemellt
of
Farm Animals,
VICTOR ARTHUR RICE ,.

Professor oj'Animal llusbandry
Cni11ersily of ll.fassachusetts
AND
FREDERICK
/ , NEWCOMB ANDREWS
Professor of Animal Husbandry
Pllrdue University
With Chapter on
Selection in Meal Anirn.o,ls
BY
EVERETT JAMES WARWICK

Geneticist, Bureau of Animal Industry, U.S. Department of Agriculture
and Professor of Animal Husbandry, University oJ Tennessee
.\.PAU C~l~"" U •••• ~ l

Acet M.l i,-13~ t
Vate: If .I·~
' ....
NEW YORK TORONTO LONDON
McGRAW-HILL BOOK' COMPANY, INC.

1951
BREEDING AND IMPROVEMENT OF FARM ANIMALS
Olpyright, Hl26, 1934, 1942,1951, by thEl McGraw-Hill Book Company, Inc.

Copyright renewed 1954 by Victor Arthur Rice. Printed in the United
Htates of America. All rights reserved. This book, or parts thereof, may
not be reproduced in any form without permission of the publishers.
PREFACE TO THE FOURTH EDITION
In the Preface to the First Edition, published in 1926, we said, "This
book has been written primarily for use as a textbook of animal breeding.
It treats the subject of breeding practice from the scientific standpoint
in so far as this is possible at the present time. Breeding is an art to be
learned only by practice, but knowledge of principles supplies the only
firm foundation for its practice. Superior animals will be more numerous
when breeders know why as well. as how."
Part of the Preface to the Second Edition, published in 1934, reads,
"Great progress has been made in all three of the fields included in the
scope of this book since its publication in 1926. In the field of reproduc-
tive physiology significant discoveries have been made in matters pertain-
ing to the endocrine control of sex manifestations as well as in those
dealing with the maintenance of genital health and normal activity.
Great forward strides have been made also in the science of genetics and
new phases of this subject have been developed. These include among
others the artificial induction of mutations by radiation or heat; the
development of the theory of genic balance; and the cytological demon-
stration of crossing over. The mechanism of transmission is now pretty
thoroughly understood and the chromosomal interpretation of inheritance
established beyond reasonable doubt. Of great significance, also, are the
new approaches and points of view that have come to the fore in the art
of breeding. Rules of thumb are fast passing into the discard and breeders
are now basing their operations on sound physiological and genetic
principles. "
The Preface to the Third Edition, published in 1942, is included in full
on the following pages.
In the present revision, the basic organization remains as it has been
from the start, viz., (1) a general introductory section, followed by sec-
tions dealing with (2) reproductive physiology, (3) genetics, and (4)
selection. During the past eight years, significant advances have been
made in all these fields: a broadened recognition of the fundamental
aspects of animal agriculture together with breed advancements, and
shifting emphasis among the various classes of livestock as indicated in
Section I, Animals and Man Up to the Present; further unraveling of the
intricate interactions of the hormonal regulation of sex manifestations
ane! tlJ.e widespread use of artificial insemination as portrayed in Section
v
~-.- --
Vl PREFACE TO l'HE FOURTH EDl'l'ION
II, Mechanisms of Reproduction; continuing delineation and refinement

of the basic l.aws of inheritance and their application to livestock improve-
ment as shO\m in Section III, Mechanisms of Heredity; and a concerted
experimental attack on the problems and implications of the various
systems of breeding as related to the main f.unction of animal breeding-
more successful selection as demonstrated in Section IV, The Art of
Breeding.
For ultimate success in creating more .efficient and beautiful animals,
one must possess an understanding of the basic principles underlying
reproductive physiology and inheritance and then gather, interpret, and
apply performance facts about his animals so that the hereditary deter-
miners for desirable qualities may sluwly but surely replace those for
undesirable qualities now present in the germ cells of the animals cur-
rently constituting his herd or flock.
It is the authors' hope that this book may provide both the knowledge
and the inspiration to accomplish this most worth-while and interesting
task.
We are again indebted to many friends, authors, and publishers for
kindly criticisms, helpful suggestions, and permission to quote and to
reproduce illustrative materials. We are happy to acknowledge our
special indebtedness for helpful suggestions to Doctors B. B. Bohren and
A. E. Bell and Professor R. B. Cooley of Purdue University; Dr. W. A.
Craft of the Regional Swine Breeding Laboratory; Doctors J. E. Nordby
and C. E. Terrill of the Regional Sheep Breeding Laboratory; and to
Professor \V. A. Cowan of the University of Massachusetts.
Our very deep appreciation is hereby extended to Dr. E. J. Warwick
for his generous and penetrating suggestions throughout our labors and
for his willingness to contribute the chapter on selection in meat animals.
For invaluable clerical assistance, we wish to thank Mrs. O. R. Rudde-
forth and Mrs. W. A. Cowan of Amherst, Massachusetts, and Mrs.
Thelma Boesch of Lafayette, Indiana.
We shall always welcome constructive criticism.
VICTOR ARTHUR RICE

.FREDERICK NEWCOMB ANDREWS
AMHERST, MASS.
LAFAYETTE, IND.
October, 1950
PREFACE TO THE THIRD EDITION
In the early history of the Land Grant Colleges there was an almost
complete lack of technical agricultural science supported by experi-
mental data, with the result that curriculums ·were built largely out of
liberal arts materials. Today ,ve have reached the other extreme with
so much well-supported technical agricultural science data available
that the student of agriculture has all too little time for liberal or cultural
courses. Most agricultural students probably get a smattering of repro-
ductive physiology in general courses in anatomy and physiology and a
smattering of genetics in general courses in biology; but since the limita-
tions of time so often prevent the inclusion of separate courses in repro-
ductive physiology, animal hygiene, genetics, embryology, etc., and since
the inclusion of all these courses would exclude just so many more cultural
courses, the author has proceeded on the basis that undergraduate
students in animal husbandry should have one complete course in animal
breeding with reproductive physiology, genetics, and the art of breeding
brought together into one complete whole. He has also endeavored to
make this something more than just a technical book on breeding by
giving a bit of the historical background involved and by indicating
some of the broader implications of the place of man and his animals on
this planet; but he has tried to leave to the reader the matter of drawing
philosophical deductions.
This book is comprised of four sections. Section I begins with the
present status of animal breeding and then traces it back to its probable
beginnings. The origins of both man and his animals are investigated,
as well as the broader aspects of the evolutionary process that apparently
brought both into existence.
Section II deals with the processes involved in the physiology of
reproduction both from the scientific and the practical angle. The
breeder should be acquainted with the normal functioning of the genital
systems in order that his herds and flocks may be maintained at a high
level of reproductive efficiency.
Section III is devoted to a study of the principles of heredity. The
various types of behavior of the chromosomes and genes, passed on to
each new individual by means of the germ cells of its parents, in the
perpetuation of ancestral traits or the creation of variations from the
older :patterns, are considered in some detail. Although the direct
manif"stations of many of these principl~s which are explicitly detailed
vii
Vl11 PREFACE TO THE THIRD EDITION
in lower organisms are difficult or impossible of exemplification in the

higher species, it is a VIrtual certainty that they are at work. Therefore
a full understanding of the basic principles is necessary if the breeder is
to handle, in the most intelligent and profitable manner, his practical
problems of selection.
Section IV is devoted to what is often called the art of breeding, which
may be summed up in one word selection. This term has both a present
and a future connotation, since a breeder is constantly selecting from
among the best animals making up his herd or flock on the basis of their
pedigrees and performance, and he is also anticipating future selection
by planning matings to create new animals of a preconceived pattern.
Selection, therefore, involves the various systems of breeding both
among related and unrelated individuals. In the final chapter the reader
can glance backward over the trail that animal breeders and scientists
have blazed and look forward to try to foresee" the shape of things to
come."
It is the author's hope that this book will be stimulating and helpful
to students of breeding both in classrooms and on livestock farms.
Many friends have assisted the author in the preparation of this book;
among them Drs. H. H. Plough and Charles L. Sherman of Amherst Col-
lege; Dr. Harry L. Shapiro of the American Museum of Natural History,
New York; Mr. Bradford Knapp, Jr., of the Bureau of Animal Industry,
Washington, D.C.; and Drs. G. Chester Crampton, Johr. B. Lentz,
Frank H. Hays and Mr. Floyd Johnson of the }\Iassachusetts State Col-
lege. Special thanks are due Dr. F. N. Andrews of Purdue Cniversity,
Dr. Arthur B. Chapman of the University of Wisconsin, and Drs. Hugh C.
"McPhee, Ralph W. Phillips, and Ralph G. Schott of the Bureau of
Animal Industry, Washington, D.C., for their suggestions and construc-
tive criticisms. The author is also greatly indebted to Mr. Talcott
Edminister for assistance in reading proof, and to Miss Evelyn Day, who
in some ingenious manner deciphered the original manuscript and put it
through a typewriter.
The author is indebted to Prof. R. A. Fisher and to Messrs. Oliver &
Boyd, of Edinburgh, for permission to reprint Tables 29 and 31 from their
• book Statistical Methods for Research Workers, 8th edition (1941).
Grateful acknowledgment is also made to various authors and pub-
lishers for permission to use certain of their materials.
If the reader is curious as to the sources of the ideas herein expressed,
he may discover them.in great measure through consulting the refer-
enees listed at the close of each chapter.
VICTOR ARTHUR RICE
AMHERST, MASS.
March,1942
CONTENTS
PREFACE TO THE FOURTH EDITION . v
PREFACE.TO THE THIRD EDITION . . YI
SECTION I. ANIMALS AND MAN UP TO THE PRESENT

I. Animal Breeding-Pres~nt and Past. . . 1
II. Early Man and Animal Domestication 38
III. Animal Origins and ProgreslSion. . . . . 68
SECTION II. MECHA~ISMS OF REPRODUCTION

IY. The Male's Part in Reproduction. . 98
Y. The Female's Part in Reproduction. 134
VI. Reproductive Efficiency. . . . 173
VII. Lowered Fertility and Sterility. . . 198
VIII. Pregnancy and Parturition. . . . . 226
IX. :Mammary Development and the Initiation of Lactation. 249
X. Artificial Insemination. . . . . . . . . . . . . . . . 260
SECTION III. MECHA~ISMS OF HEREDITY

XI. History and Problems of Genetics. . . 290
XII. The Principles of Heredity. . . . . . 304
XIII. The Principles of Heredity-(Continued) 339
XIV. The Principles of Heredity (Continued) 364
XV. The Principles of Variation . . . . . 390
XVI. The Principles of Variation (Continued) . 418
XVII. Sex Determination . . . . . . . . . . 433
SECTION IV. THE ART O}" BREEDING

XVIII. Systems of Breeding-Unrelated AniRlals . 452
XIX. Systems of Breeding-Related AnimaJs . 485
XX. General Considerations in Selection. .528
ix
x CONTENTS
XXI. Selection in Dairy Cattle . 580

XXII. Selection in Meat Animals. 644
XXIII. Selection in Horses . . . 708
XXIV. Retrospect and Prospect. . 736
ApPENDIX. Livestock Record Associations . 757
NAME INDEX . . 761
SUBJECT INDEX. 767

To
---
All Who Love Fine Animals
'- -
An artist, modeling in plastic clay or conjuring with marble,
brings forth a conception that the world acclaims a triumph.
He deals, however, with his materials direct, and they respond
instantly to his slightest touch, as he toils toward a precon-
ceived ideal. There is no resistance to his manipulations.
What, then, should be our estimate of the work of one who
has first to conceive the figure in 'his brain; whose only tools
are the laws of heredity, selection, inbreeding, outcrossing, and
alimentation; whose only materials are flesh and blood, unap-
proachable except by indirection; who battles ever against the
stubborn forces of atavism or reversion to ancestral forms;
who seeks, and succeeds in producing, a creature pulsating with
life, exquisitely fashioned, down to the minutest detail, not only
a thing of beauty in itself-whieh artists try, sometimes with
ill success, to reproduce on canvas or in bronze-but a creation
that serves as well the highest utilitarian purpose?
The breeder of animals (or plants) directs the spark of life
itself. The possibilities of his art are almost infinite.-A. H.
, SANDERS.
THE NATIONAL GEOGRAPHIC MAGAZINE

December, 1925
SECTION I
Animals and Man Up to the Present
CHAPTER I
ANIMAL BREEDING-PRESENT AND PAST
Successful animal breeding depends primarily on the proper choice of

parents for the next generation. The A, B, C's of it are, therefore,
Ancestors Better Chosen. Since any farm animal arises from the union
of one egg and one sperm, we can epitomize the whole problem by saying
that whenever better eggs are fertilized by better sperm, we will have
better animals-and there is absolutely no other way to get them. So, it
is apparent that the breeder's task is that of ascertaining as precisely as
possible what the germ cells of his animals contain in the way of hereditary
determiners of potentialities for the next generation. To know the details
of the workings of the reproductive organs as they perform their two
functions of (1) bridging the gap between the generations and (2) setting
up the specifications for the next generation is quite obviously the
breeder's first task. Fortified with this basic knowledge, he can then
tackle his second job of dovetailing systems of breeding and selection
into a more efficient tool for creating improved livestock.
Animal breedin~ is o~e of the largest industries in the United States.
The total national income in 1950 was 235.6 billion dollars. Farm
income was 27.9 billion, and of this, 15.5 billion was income from livestock
and livestock products. Animals of some type are to be found on about
90 per cent of our farms in the United States. Livestock breeding
involves nearly 200 million farm mammals, 700 million poultry and other
birds, in addition to an innumerable host of dogs, cats, white rats; also
foxes, ferrets, and other fur bearers, not to mention among others guinea
pigs, guppies, and goldfish. Animal breeding for some is a hobby, but
many thousands of people depend upon it for their livelihood. In terms
of food, clothing, and the maintenance of soil fertility, every person in the
United States has a stake in its success.
Hobbyist and realist are trying in most instances to shape the external
form of their animals according to some preconceived plan and at the
same time to include within this desirable exterior form some combination
of mental and physiological characters that will have a high commercial or
1
2 BREEDING AND IMPROVEMENT OF FARM ANIMALS
aesthetic value. The horseman wants speed and stamina; power,

patience, and persistence; easy gait and companionability. The cattle-
man wants a compact animal that will fatten rapidly into a carcass of
good quality; the dairyman wants a "typy" cow that will yield a large
TABLE I.-NUMBER AND AVERAGE VALUE PE~ HEAD OF THE VARIOUS CLASSES OF
FARM LIVESTOCK IN THE UNITED STATES, 1880--1950
Horses on farms Mules on farms Dairy cattle
Year Number Number Number I

(000 Av. value (000 Av. value (000 Av. value
omitted) omitted) omitted)
1880 10,903 $ 53.74 1,878 $ 61. 74 11,754 $ 23.31

1890 15,732 69.27 2,322 77.61 15,000 22.30
,
1900 17,856 43.56 3,139 51.46 16,544 31.30
1910 19,972 107.70 4,239 119.98 19,450 35.40
1920 20,091 96.45 5,651 148.29 21,455 81.51
1930 13,742 69.98 5,382 83.93 23,032 82.70
1940 10,616 77.43 4,321 114.53 25,334 57.22
1950 4,763 43.30 1,990 82.00 24,579 218.00
I
Cattle other than

milk cows I Sheep I Hogs including pigs
Year
Number Number Number
(000 Av. value (000 Av. value (000
Av. value
omitted) omitted) omitted)
1880 31,593 $ 15.75 44,867 $ 2.18 44,327 $ 4.40

1890 45,014 15.16 42,693 2.29 48,130 4.80
1900 43,195 24.67 45,065 2.97 51,055 5.36
1910 39,543 19.20 46,939 4.06 48,072 9.05
1920 48,945 39.99 37,328 10.59 60,159 20.00
1930 39,971 40.38 45,477 9.00 55,705 13.45
1940 43,435 30.86 48,473 6.33 58,312 7.79
1950 59,600 136.00 28,065 26.40 1 65 ,028 38.20
amount of milk of a high quality over a lengthy period. The sheepman

wants quality and quantity in both carcass and wool; the hog man as
much high-quality pork as he can get from each sow in his herd each year.
The dog fancier wants looks, smartness, companionability, and sometimes
utility. All of them want docility, longevity, freedom from disease,
.
regularity of breeding, etc.
Can these various combinations be had in single animals? The answer
ANIMAL lJREEDING-~'J)I{!i)/$f!JN']' AND PAS']'
obviously is "yes" because we have had such animals in the past and have
some of them today. Can we so breed our animals that they not only
\vill have these desirable combinations of qualities but also will be able to
pass them on to their offspring? The procedure for accomplishing the
latter purpose is now known, and the purpose of this book will be to
demonstrate the principles underlying reproduction, hereditary trans-
mission, and variation and to show how selection involving the various
systems of breeding can be made to yield certain definite results.
Livestock Trends in the United States.-In Table 1 will be found an
enumeration of the various farm mammals over the past 70 years. It is
obvious from Table 1 that mechanical power has rapidly been supplanting
horse and mule power since 1920. 'Whether machines will entirely sup-
TABLE 2.-ToTAL NU~1BERS AND VALUE OF FAR~I MAIIUIALS AND HU:VIAN POPULATION,
UNITED STATES, 1880-1950*
Year ::\0. of all farm

livestock
I Value of all farm
livestock
IHuman population,
United Stat est
1880 120,990,000 $ 1,576,636,520 50,156,000

1890 165,286,000 2,418,774,100 62,948,000
1900 170,315,000 2,581,751,910 75,994,000
1910 178,355,000 4,649,803,510 91,972,000
1920 196,875,000 7,383,500,260 105,711,000
1930 185,464,000 5,995,084,680 122,775,000
1940 196,491,000 4,904,307,000 131,669,000
19,JO 184,025,000 16,712,631,000 153,000,OOOt
* Horses, mules, cattle, sheep, swine.

t To even thousands.
t Estimate.
plant horses and mules as power units, the future will tell. Only from
one-fourth to one-half as many work horses and mules are being produced
as would be needed to maintain their numbers. On the other hand, there
seems to be a growing interest in the horse for pleasure purposes. Dairy-
cattle numbers have shown a steady increase over the last 70 years.
Meat animals have had a slight actual increase and, in addition, they now
mature more rapidly than formerly, the quicker turnover actually mean-
ing more product from a given number of animals. Many factors, of
course, influence the numbers of animals kept on farms from year to year.
The magnitude of the total value of American livestock and the meager-
ness of the average value stan.d in marked contrast to each other. Both
of these points indicate opportunity for the livestock preeder ..
Table 2 shows the trend in total numbers and total value of farm live-
stock in the United States by decades for the past 70 years and the trend
"'),
in human population. It shows that the breeding of livestock is one of

the most important types of endeavor in this country.
TABLE 3.-AuCTION-SALE PRICES OF PUREBRED CATTLE
I
Breed 1920 1924 1928 1932 1936 1940 1944 1948
- - - --- _ - - - - - _ - - --- --- ---
Ayrshire ............... $359 $124 $184 $108 $156 $146 $196 $321
Guernsey ............... 508 380 334 162 252 225 358 415
Holstein ................ 372 192 218 105 164 173 327 439
Jersey .................. 354 206 211 126 161 147 251 373
Aberdeen-Angus ......... 705 150 170 106 161 209 336 533
Hereford ............... 416 133 210 105 151 195 317 497
Shorthorn ... , .......... 603 160 192 104
I 137 171
TABLE 4.-shLK AND FAT PRODUCTION OF PURE BREEDS*
\ No. of Av. No. Av. pounds Av. pounds

Breed
records of days of milk of fat
Holstein (AR) ................ 60,375 315 16,440 570.0

Holstein (HIT) ............... 220,195 ....... 11,213 391.4
Brown Swiss (ROP) ........... 1,329 3052X 10,341.57 424.63
Brown Swiss (ROP) ........... 591 3053X 11 ,901.93 483.75
Brown Swiss (ROP) ........... 897 3652X 11 ,626.02 474.35
Brown Swiss (ROP) ........... 1,337 3653X 13,752.34 551.85
Brown Swiss (HIR) ............ 14,132 3052X 9,159.78 362.84
Brown Swiss (HIR) ............ 999 305 3X 11,653.4 464.88
.'.
Guernsey ..................... 22,361 365 11 ,014 541
Guernseyt······· ............. 3,351 3053X 9,723 470
Guernseyt·········· .......... 17,420 3052X 8,036 390
Guernsey (HIR) .............. 50,240 ....... 8,159 396 I
Dutch Belted ................. 100 365 10,844.32 414.09
Devon ....................... 46 365 7,484.4 339.99
Red Poll ..................... 1,915 ....... 8,692.88 373.88
Jersey (HIR) ................. 22,385 ....... 7,070 374
Ayrshirei·········· ........... 10,118 ....... 9,846 401
Milking Shorthorn ............. 22,580 348 8,972 357.08
American Dairy Cattle Club ... '1 1,985 305 10,584 437
* New England Homestead, Springfield, !\lass., Sept. 24, 1949.

t Carried calf 175 days or more of record. Guernsey AR averages of records conlpleted during: last
5 years.
t Ayrshire Association has discontinued computing records on a herd-test yearly basis. Their
practiee now is to compute a new average each year.
Value of Purebred Animals.-Table 1 shows the low value of the aver-

age animal,in each class. Now let us look at the brighter, upper side of
the picture. Table 3, for example, gives some auction-sale prices of plll'e-
bred cattle.
ANIMAL BREEDING-PRESENT AND PAST 5
On the average the monetary value of purebreds always exceeds that

of grade animals. It is sometimes argued that grades are practically as
good as purebreds, and the fact that the grades generally owe their desir-
able qualities to the" blood" of registered animals is entirely overlooked.
In commercial production, grade animals will probably ahyays fill a very
important place, but they will probably continue to depend on the pure-
bred for their improvement, although this is not necessarily so. Actually,
less than 5 per cent of the animals in any class are purebred and registered.
The great bulk of our animals are unregistered purebreds or grades. The
purebred breeder, therefore, faces a great and continuing challenge in
planning his breeding operations so that he will be producing sires which
are capable of improving grade herds. The practice of artificial breed-
ing, now gaining so rapidly in dairy-cattle operations, should be a great
boon to the commercial producer, since if properly organized and admin-
istered, it should remove one of the stumbling blocks to his progress,
namely, that of securing better sires.
Table 4 gives some production figures for the various breeds of dairy
cattle.
Still another aspect of the possible rewards in the purebred business
is shown by the following tabulation of some anction-sale record p-rices
for livestock.
ALL-TI~1E AUCTION RECORDS l
Aberdeen-Angns-male, Prince Eric of Sunbeam, $40,000; female, Erianna B.

6th, $25,000.
Ayrshire-male, Netherhall Swanky Dan, $20,000; female, Low Milton Queen
of Hearts, $10,277.
Brown Swiss-male, Colonel Harry of Judd's Bridge, $23,500; female, James
Chloe of Judd's Bridge, $11 ,500.
Guernsey-male, Gardenville Coronation King, $45,~0; female, Shuttlewick
Levity, $25,000.
Hereford-male, WHR Helmsman 89th, $61,000; female, WHR Lady Lill
15th, 3769212, $20,000.
Holstein-Friesian-male, Prince Aggie of Berylwood, $110,000; female, Pabst
Korndyke Cornflower, $30,000.
Jersey~male, Sybil's Gamboge, $65,000; female, Wonderful Dreaming Givia,
$21,000.
Polled Hereford~male, AFL Choice Domino 6th, $35,000; female, Trumain
Domino 63rd, $10,000.
Red Poll-male, Billy Charmer, $2,000; female, Dandy Lou, $1160.
Shorthorn-male, Pittodrie Upright, $61,335; female, 8th Duchess of Geneva,
$40,600.
Belgian-male, Farceur, $47,500.
1 New England Homestead, Springfield, Mass., Sept. 25, 1948,.p. 30.
Clydesdale-male, Baron of Bucklyvie, $47,500.

Percheron-male, Philix, $9{)OO; female, Iolanthe, $3500.
Suffolk-American sales, Boxted Confider, $3,700; female, Rose Mary, $1,225.
As shown in Table 5, most of our purebred breed associations in the
United States 'sere formed about 1875-1885. Importations of purebred
livestock into this country preceded this, to be sure, but improved breed-
ing was given great impetus by the formation of the breed associations.
Table 5 also shows the number of animals of the different breeds that
have been registered up to the present.
Some people maintain that our primary problem in animal husbandry
is not that of better breeding but rather one of better feeding and manage-
ment. Several experiment stations have purchased cows with records
from dairymen and increased their production by 2,000 or 3,000 lb. as a
result of better feeding and management. The average cow produces
about 5,000 lb. of milk yearly. Is her actual inheritance 7,000 or 8,000
lb. ? Could be, but is it? Disease also takes its toll in all our classes of
livestock. Studies show that about 50 per cent of the dairy cows which
leave herds annually do so because of low production and disease. Early
death, failure to grow rapidly, thus delaying first calving in dairy cattle
and arrival at market in fat stock, too long intervals between parturitions,
too high a percentage of herd replacements yearly, and low production or
rate of growth through poor feeding and management cost animal agri-
culture untold millions yearly. Of 100 pigs born, about 56 eventually
get to market. This little pig went to market, this little pig stayed
home--period. Certainly there is great need for better feeding and
management-"half the breeding goes down the throat." But all this
being true does not in any way lessen the need for better breeding in
terms of production, type, longevity, resistance to disease, regularity of
breeding, etc.
Foods of Animal Origin.-A rapidly increasing population does not
necessarily mean the extinction of any particular class of livestock. It
should, however, mean the rapid passing of inefficient members of all
classes. Thickly populated European countries still produce meat, milk,
wool, and power. The actual number of animals in the United States,
together with their increasing productivity, mean that our population is
eonsuming an increasing amount of animal products. The population
of the United States was about 5 million in 1800 and has grown to about
147 million in 1949. It has been variously estimated that the total pop-
ulation that can be provided with home-grown food in the United States
i8 200 up to 331 million. This is based on maintaining the present dietary
standard. The average per capita consumption of dairy products is still
far below the nutritional optimum, for our people consume on the average
TABLE 5.-NUMBER OF REGISTRATIONS IN THE VARlQUS PURE BREEDS*
Breed Membership 1948 All-time

registrations registrations
Cattle Associations
Hereford .................... 1 16,000 352,888 5,825,000

Holstein ................... . 39,516 168,338 4,154,304
Shorthorn .............. ' ... . 23,842 49,674 3,170,463
Jersey ................ ". : .... '.1 3,445 66,766 2,233,025
Guernsey ........... , .. . 3,693 96,895 1,616,326
Aberdeen-Angus ............ '1 15,000 84,078 1,158,372
Ayrshire ................... . 7,733 26,113 492,459
Polled Hereford ............. . 7,000 34,110 356,732
Brown Swiss ............... . 2,568 22,625 296,396
Polled Shorthorn ............. 1 2",447 10,326 223,935
Milking Shorthorn ........... 1 7.890 24,650 200,000
Red Poll .................... \ 3;000 4,274 181,942
Brahman .................. . 992 15,145 109,917
American Galloway ......... . 24 648 51,701
Devon ..................... . 75 365 32,590
American Dairy Cattle ...... . 200 2,000 19,000
Dutch Belted ............... . 35 50 7,377
Red Danish ................ . 155 10 3,500
American Scotch Highland ... . 16 434 434
Kerry and Dexter ........... . No report
40 (1948) 150
Sheep Associations
Shropshire ................. . 12,160 16,523 1,117,000

Hampshire ....... . 6,400 28,390 880,482
Rambouillet. 2,054 . 9,704 514,136
Oxford ....... . 1,450 3.179 193,950
SOuthdown ........ . 1,234 8,172 192,033
Cotswold .............. . 1,100 375 126,366
Corriedale .......... . 1,320 10,904 92,669
Dorset ................ " . 500 3,487 79,135
Lincoln ........ " ., .' .. No report
135 (1948) 764 68,132 (1948)
Cheviot .................... . About 700 2,345 55,000
Texas Delaine-Merino ....... . 188 2,120 41,478
American Suffolk ........... . 344 5,921 41,412
Columbia .................. . 288 3,390 28,990
Romney ................... . 200 1,608 28,608
National Suffolk ............ . 300 3,750 26,055
Karakul ................... . 491 1,005 19,275
Black Top-National. ........ .
Delaine-Merino ............. . 24 110 1,081
American and Delaine-Merino No report
8 BREEDING AND IMPROVEMENT OJ? FARM ANIMALS
TABLEl5.-NUMBER OF REGISTRATIONS IN THE VARIOUS PURE BREEDS. *-(Cont'inued)
Breed Membership I 1948 All-time

I registrations registrations
Horse Associations
Per~heron .............. '1 6,500 251 247,893

Thoroughbred .......... . 50 8,375 108,375
Saddle ......... " " .... '. 500 4,477 79,477
Belgian ................. / 4.450 350 62,739
Jack and Jennet. . . . i 2,200 66 37,028
Tennessee \Valking .. 1.056 3,319 33,642
United States Trotting ... 7.500 3,536 27,323 (since 1938)
Shetland Pony .... . 681 849 26,111
Clydesdale ......... . 130 45 25,460
Shire ............... . 1.100 16 21,959
American Quarter ... . 4.500 3,849 17,895
Morgan ............. . .500 469 17,751
National Quarter ... " Over 3,000 3,041 13,833
Palomino (Texas) ....... ' 3.000 1.000 8,071
Palomino (California) ... . 600 685 5,420
Suffolk ................ . 120 9
Arabian ......... . 650 600 5,200
Morocco Spotted ....... . 500 active 94 1,390
Welsh Pony ............ . 30 46 1,260
Albino ............. " .' . 150 73 900
Appaloosa .............. 1 204 314 850
Pinto ................... , No report
535 (1948) 95 625 (1948)
~- ----------'----------'------ ,-~---.-------
Swine Associations
--,-- - -----_.-_.
United Duroc. .. . ...... 1 6.760 100.391 !2,603,068
Poland-China ........... 1 II ,400 30.800 '2,208.637
Hampshire .............. 1 9,000 50.217 776,875
Chester White ........... 3,782 23,861 699,694
Berkshire ........ , ...... 11 ,500 24,000 635,317
National Spotted Poland-
China ................ 12,995 48,216 625,061
O.LC .................... 1,176 16,254 270,024
National Hereford ....... 2,041 9,200 80,833
Yorkshire ............... 800 5,912 61.890
Chester White Record .... 2,363 2,332 29,916
Essex ................... 90 200 14,765 (1948)
Inbred Livestock ......... 551 3,890 10,427
National Mule Foot ...... 150
Kentucky Red Berkshire. 15 112 2,352
Tamworth .............. 80 1,400 1,533
Goat Associations
American Angora .. 624 7,156 232,000
American Milk Goat .... . 1,700 4,036 93,387
tAmerican Goat ........ . 1,000 3,550 35,000
Rabbit and Cavy Associations
American Rabbit and

Cavy Breeders ........ .
I 8,745 20,201 47,894
* New En~\,and Homestead, Springfield, Mass., Sept. 24, 1949.
t P"rebreds. A as does not register grades.
about 2.2 glasses of milk; H oz. of butter (2 pats), ;~ oz. of cheese per
person per day, and about. 2 dishes of ice cream per person per week. In
addition, there is an average consumption of about 1 can of evaporated
milk per person every 3 weeks, plus a small amount of sweetened con-
densed milk.
The 1948 per capita consumption of meat was as follows: beef, 63.4 lb. ;
veal, 9.4 lb.; lamb, 5.0 lb.; pork products, 68.3 lb.; a total of 146 lb. per
capita per year. This is lower than it is in several countries, higher than
that in many other countries. Whether or not it is the optimum, the
writer does not know, but he is confident that meat would be in greater
demand if the means for its purchase could be made available. In terms
of the 21 meals that most of us eat each week, our present consumption
0.9
0.8
~>-:-,. ..... ~..I';,e
..........
0.7
0.6
.........s~
.......~.c
-....---.... ---- ....
/O~- ...
0.5
V I~_
--_
" ..................
,-- :--- -_
_._ - ,..--
0.4
;::::::-.:::::: ~I
0.3
0.2
0.1
- - - :::;:::-- H.!!!_-
se5and~~ __
Dairy faffle
--, =--- r-- "_-
"
__
°1880 1890 1900 1910 1920 1930 1940 1950
FIG. 1.-Graph showing number of various classes of livestock per capita in the United
States, 1880-1950.
of meat would provide about as follows: for 3 meals we could have 7.5 oz.
of beef and veal; for 1 meal we could have 1.5 oz. of lamb; and for 3 meals
we could have 7.0 oz. of pork, leaving 14 meals each week without any
meat. Or the 45 oz. of meat (2 lb. 13 oz.) we consume on the average
each week would give us 1 serving of about 8 oz. for some meal on each
of 6 days of the week, leaving 1 day for fish or eggs, or maybe fish-eggs.
It would be possible to maintain a much larger population on primary
food products (grain, etc.) because of the unavoidable losses in all animal
feeding and the fact that no return is receiyed from that portion of the
animal's feed which goes for maintenance and which amounts to approx-
imately one-half. The inclusion of animal products in the diet, however,
makes it much easier to provide a balanced diet with all the necessary
nutritional elements, including minerals and vitamins, plus the additional
i.mportant fact of making the diet, to most of us at least, much more
palatable and enioyable. Much energy is stored here on earth in a form
not suitable for human consumption, and there is the added considera-
tion of soil productivity, both of which make animal raising indispensable.
The graph (Fig. 1) shows the trends in the number of different animals
per capita in the United States during the past 70 years.
Table 6 gives a picture of the trend in meat consumption per capita in
the United States since 1900.
The trends indicated by Fig. 1 and Table 6 certainly do not indicate
the passing of animal foods from our diet. Meats are palatable and
nutritious and will probably be utilized as long as they can be secured
and paid for. At the present time, milk y,·ould seem to be an indispen-
sable article of diet, and Fig. 1 shows that dairy cattle are maintaining
their numbers in relation to population. If they are increasing in average
yield of production, there is certainly no less milk being consumed per
capita by an increasing population. .
Dietary standards, coarse feeds, and climatic and soil conditions would
:;;eem an effectual barrier against the passing of livestock. Increasing
population means an increased demand for food of all kinds; hence no
time should be lost in raising the efficiency of food production to the
TABLE 6.-MEAT CONSUMPTION PER CAPITA, UNITED STATES

In Pounds
Year Beef Pork * Lamb Veal All meats

---~
1900 67.0 65.3 6.5 5.2 144.0

1910 70.5 66.9 6.4 7.2 151.0
1920 59.1 63.9 5.4 8.0 136.4
1930 48.9 69.8 6.7 6.4 131.8
1940 55.2 64.4 6.6 7.3 133.5
1948t 63.4 68.3 5.0 9.4 146.1
* Excluding lard.
t Tentative.
highest possible level. In such a program there will surely be found a

place for all the really efficient animals that can be produced.
The place of farm animals in converting rough feeds into usable form
for the human animal, in improving our American diet, and in conserving
and increasing soil fertility has long been realized. But an increasing
spirit of awareness of the major role which animals are destined to play
now pervades, and livestock production is now often referred to as animal
agriculture. We now see that our so~called surpluses of grains can best
he stored in animal hides. This at one stroke provides a reserve of food
for emergency and a better fed populace and is in full keeping with the
idea of an economy of abundance .
•
The data in Tables 1 and 2 show that the number of dairy cattle, beef
cattle, sheep, and swine compared to human population has steadily
declined, starting at a high of 2.64. of these animals to each person in the
United States in 1880 and decreasing by 10-year intervals through 2.40
in 1890; 2.05 in 1900; 1.67 in 1910; 1.59 in 1920; 1.34 in 1930; 1.33 in
1940; and 1.16 in 1950. Increasing production in dairy cattle, faster
maturity, and meatier carcasses in the meat classes have, of course, tended
to offset the decrease in numbers.
The Bureau of Agricultural Economics has reduced all food-producing
livestoGk (meat animals, dairy cows, and poultry) to a hog-equivalent
basis. They report that in 1920 we fed 1.67 head of producing livestock
per capita; in 1930 the figure had fallen to 1.55; in 1940 it stood at 1.53;
and in 1948-1949 it ,vas 1.41. There are two possible ways of storing
grain-in dead storage or in animal hides. When stored in the latter
way, we have an emergency food supply, we have a chance to increase
our soil fertility, and we have a better nourished populace consuming
more milk, meat, and eggs.
H. E. Babcock estimated in 1948 that for the proper nourishment of
150 million people, we would need an additional 10 million dairy cows,
11 million beef animals, 8 million sheep, 20 million hogs, and 120 million
poultry. It is obvious that we need increases both quantitatively and
qualitatively in all our classes of livestock. To an extent greater than
the average person realizes does the future welfare of the United States
hinge and depend upon these gains in animal agriculture.
Present Status and Problems.-Many animals are unprofitably low
or inefficient producers because they are poorly fed and cared for. If the
average horse were well fed and cared for and were thus able to work and
breed at maximum efficiency, if the average cow, ewe, and sow were per-
forming at somewhere near the maximum set by their inheritance, live-
stock farming would be much more profitable than it now is. This
involves also the control or elimination of our- various and sundry live-
stock pests and diseases, which take an annual toll of millions of dollars
in terms of lowered individual and breeding efficiency. Likewise, our
civilization would be on a much higher plane if each human individual
were performing more closely to his inherited capacities. The problem
facing us, therefore, as far as our animals are concerned, consists of fred-
ing and managing what we have more efficiently as well as raising the
level of capabilities through selective breeding.
In one sense a treatise on the breeding and improvement of farm ani-
mals seems rather futile :without its corollary, the breeding of better men,
for if our human inheritance and environment were better and we could
induce our people to live and work more nearly to the upper limit of their
1:2 BREEDING AND IMPROVEMENT OI" F'ARM ..ANJMALS
!.-~p'abilities,
thel'e would be less need to wTite books about Qreedi:ng better
livestock. An encouraging angle, however, is ,the fact that the principlelS
underlying the creation and maximum productivity of livestock are
identical with the principles underlying these things in man himself.
If we can put the principles to work generally ,,·ith our livestock, they will
perhaps gradually become incorporated in our thinking about ourselves.
In our animal breeding '\\-e must, of COUJ;se, start from where we now
are, with 1yhat has come do,yn to us from the past. The genetic task in
the field of breeding, therefore, is twofold: (1) to find out what we have
genetically, in other words, to analyze; and (2), to raise the average level
FIG. 2.-Jersey Bull, Count St. George. First-prize bull in 1882. (C01l1'lesy of American
Jersey Cattle Club.)
of production and efficiency all along the line by making better combina-
tions of genetic materials, in other words, to synthesize.
There can be little question about the fact that most of our animals
at the present time are mixed in their inheritance. We shall learn later
that inheritance is due to the actions and interactions of discrete bodies
called genes, which are fo~nd associated in groups, each group of genes
making up a chromosome. The chromosomes (ch?'oma the Greek word
for color, and soma the word for body) are found in the nuclei of cells.
Each species of plant and animal has a definite, permanent number of
chromosomes. If a member Qf the Shorthorn breed of cattle got the
gene, or determiner, for red color (R) in a chr0mosome from its sire and
its allele for white color (r) in a chromosome from its dam, its genetic
m~ke-up would be Rr and its color roan (i.e., RR = red, Rr = roan, and
rr = white, there being no dominance between genes Rand r). Such an
animal cannot breed true because it ",rill pass along to each of its offspring
either the chromosome containing the gene R or the one containing the
gene r.
This illustrates the two basic concepts of modern breeding knowledge:
(1) the mel}hanism of inheritance is found in discrete bodies (genes) or, in
other words, is particulate; and (2) each offspring gets only one-half of
the genetic material which its parent possesses; in other words, the
inheritance is halved each time it is passed to an off5pring. So an Rl'
(roan) animal cannot breed true because it cannot give an identical gene
(in this case, determiner of color) to all its offspring. This animal is not
pure for color, speaking genetically, or, in technical terms, is heterozygous
FIG. 3.-Jersey Bull, Brampton Standard Sir, a modern sire noted for the excellent type
of his offspring. (Courtesy of American Jersey Cattle Club.)
as far as color is concerned. An RR (red) or a rr (white) animal is pure

genetically (for color) or, in technical terms, homozygous (for color) and
will, therefore, breed true in the sense of passing an identical gene to each
of its offspring.
Our breeds of livestock are pure, or relatively so, for certain distin-
guishing color markings, for horns or for no horns, etc., but they are not
pure for the genes that determine commercially valuable characteristics.
-They have been made pure or homozygous for their superficial trade-
marks by the process of weeding out those animals which were themselves
not pure as evidenced either in their own make-up or in that of their off-
spring, a relatively simple process in characteristics determined by one or
a few pairs of genes acting on distinct and easily recognizable traits.
They are not pure (homozygous) for speed and stamina, rapidity and
,efficiency of gain and quality of carcass, quality and quantity. of meat
d wool, amount and quality of milk, etc., simply because we have not
14 BREEDING AND IMPROVEMEN'l' OF FARM ANIMALS
yet ruthlessly culled out those which do not measure up to given standards
of production and transmission. We have all sorts of shapes and qualities
and productiveness in our livestock, and the animals cannot yet be
counted on to transmit as they themselves look or behave. They are
heterozygous (mixed) in their genetic make-up and hence cannot breed
true or uniformly.
The greatest need in animal breeding today is for records of perform-
ance so that we may learn definitely how our animals are transmitting.
When we have developed better ways of measuring performance, and
physiological studies of different hormone levels have provided new clues
FIG. 4.-Jersey Cow, Khedive's Primrose. First-prue cow in 1882. (Courtesy of American
J eraey Cattle Club.)
to probable performance, we can intelligently weed out the unfit, breed

only from the fit, and so, gradually, purify our stock in terms of their
commercially desirable characters.
Breeding Opportunities.-Although it cannot be denied that well-nigh
perfect specimens of the various classes and breeds of livestock have been
produced and that in practically all breeds the prices paid for individuals
have risen well into the thousands of dollars, the fact remains that oppor-
tunity is still 'Written large before animal breeders. A few nearly perfect
specimens have been produced, and record after record has been estab-
lished only to be replaced by a better one after a short while.
The record in milk production during 1 year is 41,943 lb. (Carnation
Ormsby Madcap Fayne, H.F.), in butterfat 1,614 lb. (Melba 15th of
Darbalara, Milking Shorthorn) ; in speed at the trot for 1 mile, 1 minute
•
55~ seconds (Greyhound), at the pace, 1 minute 55 seconds (Billy

Direct).
The record in horse pulling contests for heavyweight teams is 4,225 lb.
tractive pull, which is equivalent to starting a load of 54,870 lb. on a
wagon on granite-block pavement 15 to 20 consecutive times, or more
than equal to pulling 10 plows cutting furrows 14 in. wide and 6 in. deep,
in ordinary corn-belt black-loam soil. The record for lightweight teams
is 3,525 lb. tractive pull, which is equivalent to starting a load of 45,779 lb.
on a wagon on granite-block pavement for the same number of times, or
more than equal to pulling 8 plows cutting furrows 14 in. wide and 6 in.
deep, in ordinary corn-belt black-loam soil.
FIG. 5.-Jersey Cow, Sybil Design Etta. Gra nd Champion Jersey cow, 1947-1948.
(Courtesy of American Jer8ey Cattle Club.)
The record in the ton-litter contest stands at 5,117 lb., the combined
weight attained by an Illinois litter of 17 pigs in 180 days.
The outstanding Rambouillet sire, Prince of Parowan, owned by the
Bureau of Animal Industry, produced seven annual fleeces of wool that
averaged 31.5 lb. per fleece, unscoured, an outstanding record of long-
time wool production. This wool yielded about 40 per cent of clean wool,
or 12.6 lb. of actual wool, per fleece. A large percentage of several hun-
dred of the best Rambouillet sheep in the flocks of the Western Sheep
Breeding Laboratory, Dubois, Idaho, are descendants of this ram.
The ability of the goat to produce milk is evidenced by the fact that
the highest official test on record in the United States was made by a
French Alpine doe, Little Hill Pierrette's Lady Penelope, that produced
,632.3 lb. of miik in 10 months, an average for the period of 7.2 qt. a day,
a figure not far below the average production of all dairy cows in America.
16 BREEDING AND IMPROVEMENT OF FARM ANiMALS
An idea of the opportunity for improvement within a particular breed

of beef cattle by breeding and selection may be gained from some of the
Bureau of Animal Industry record-of-performance data with steers,
which show a range in efficiency all the way from 12.01 to 24.47 lb. of live
weight gained for each 100 lb. of total digestible nutrients consumed.
The number of days from birth to 900
lb. live weight varied from 412 to 591.
With dual-purpose cattle, the
breeder is confronted simultaneously
with the two variables-milk produc-
tion and beef production. The varia-
tion in efficiency with which beef is
produced by the dual-purpose animal
is similar to that in the beef breeds.
In the bureau's work with milking
Shorthorns, the variation in efficiency
covers a range of from 11.49 to
21.38, and the number of days from
birth to 900 lb. live weight extend
from 358 to 532. Milk production
varied from less than 5,000 to as high
as 12,000 lb.
All these records, when compared
with the average performance of our
FIG. 6.-William Duthie of Collyrue, a livestock, indicate the tremendous
Shorthorn breeder who held annual sales
for 39 years and sold 1,022 head, mostly opportunities still existing in the
bull calves, for a grand total of $1,238,- field of breeding. These top-notch
646. (CourtlJ8Y of Breeder's Gazette.)
animals resulted from certain com-
binations of hereditary units plus, of course, excellent care, feeding,
training, and general management. As we shall learn later, there are
literally billions of \vays in which the hereditary chromosomes can be
recombined. Our first job is to learn the genetic make-up in our live-
stock through careful record keeping and then to recombine the units into
more desirable combinations through systems of breeding and selection.
Breeding is the improvement of animals through rational selection of
germ plasm and full development of resulting somatoplasm. A cursory
glance at the qualifications requisite in one who is to improve animals
removes at once any surprises at the scarcity of such breeders. Men like
Bakewell, Cruickshank, Col. Taylor, Lothrop Ames, Gentry, Duthie,
Marsh, and a score of others of greater or lesser fame are, comparatively
speaking, very rare because of the size of the task and the multiplicity of
details with which a breeder must deal.
Successful breeders are equipped with keen powers of observation,
are thorough students of animal form and function who are able to select
and mate wisely, are good farmers and intelligent feeders ,yho can bring to
full expression the inherent potentialities of their animals, and they are
not deficient in business acumen, which makes possible the accumulation
and wise use of capital. They are the type of men who inspire loyalty
and fairness from those with whom they come in contact. In short,
successful breeders are usually hard-working, practical idealists who
appreciate the privilege of working with the Creator in molding nature's
forms more closely after some ideal by means of records of performance,
systems of breeding, and selection.
The Purebred Era.-The rise of the present-day pure breeds of domes-
tic livestock dates back a comparatively short time. Practically all of
them arose during the eighteenth and nineteenth centuries. At that
time, breeders began to be guided by more definite ideals of type, to have
greater recourse to assortative matings (like to like) which soon led to the
necessity of various degrees of inbreeding, which tended to fix the char-
acteristics they desired. In terms of genetics, they took a very hetero-
zygous strain, inbred it, which sorted it out into homozygous form, and
then disposed of those undesirable animals which no doubt carried double
sets of recessive factors. Such methods are still efficacious, but it is a
question whether or noL they are being used as much as they might
,veIl be.
The term "purebred" may in some instances be something of a mis-
nomer. For example, the term" purebred dairy cow" should imply, it
,would seem, that the animal in question is capable of producing large
quantities of milk. It does not, of course, imply anything of the sort but
only that the animal's sire and dam were purebred. Our dairy breeds
are relatively pure for color, size, etc., but they are not yet pure for high
milk production, though, of course, by proper methods of selection they
can be made so. It would be much better to speak of these animals as
"registered" rather than as "purebred" and to use this latter term only
in its genetic sense. The term purebred, however, has gained such a place
through common usage, that it would be futile to oppose its use. The
student, nevertheless, should be on his guard and cognizant of the various
shades of meaning of the term purebred.
The first herd book, for keeping Thoroughbred pedigrees, was a private
affair started in England in 1791 to be followed by another private herd
book for Shorthorns in 1822. None of our purebred registry associations
in the United States is as yet eighty-five years old. Pedigrees for the
Thoroughbred horse and the Jersey cow were first kept in America in
1868, for the Corriedale sheep in 1915. There were 11 of the more impor-
tant breed associations formed in the decade 1870-1880; 13 in the next

decade, 1880-1890; 6 in the next, 1890-1900; 2 in the next, 1900-1910;
and 3 in that from 1910-1920. The average date for the founding of the
cattle associations was about the year 1870, for horse and swine associa-
tions about 1880, and for the sheep associations about 1890. However,
new breeds are still being formed and new breed associations established.
Men with common interests in the same breed of livestock banded
themselves together into breed associations for two prime reasons. One
of these was to protect the purity of the breed in order that it might.
retain its distinctive external trade-marks or to keep out the genes that
determined characters not wanted. This \vas to be accomplished by a
system of registration open at first to animals that met these certain trade-
mark qualifications. This practice in due time was to be superseded by
the stipulation that only the offspring of registered animals were to be
eligible for registration. Because no stipulations regarding abUity to
fatten or to milk, etc., were or ever have been set up, it is not to be
wondered at that poor producers were originally admitted into the select
circle of the purebreds and that there are still some to be found there.
For many years any horse could be admitted to the Standardbred Reg-
ister for horses on successfully meeting certain requirements for speed.
The only criterion for the admittance to registration of a calf of the dairy
breed::; is that its sire and dam be registered animals of the same breed
and the calf show no disqualifying characteristics. In the one case the
criterion was performance with no questions asked about parentage. In
the other it is parentage with no questions asked about performance.
Perhaps the first system was wrong, certainly the second is, and the best
system would include stipulations regarding both performance and
pedigree.
Some of the breed associations, having realized that much poor germ
plasm is to be found in all breeds, are now struggling to devise means of
selective registration. Rules that will be genetically sound and commer-
cially fair as well as practicable for this sort of thing are very difficult to
formulate. If the difficulties prove insurmountable, an alternative
method would be something similar to the herd test in the dairy breeds.
In this system every female in the herd must be tested, i.e., her milk
weighed and its butterfat percentage determined, and the poor producers
thus can be eliminated from the herd average by surrendering their regis-
tration papers. If this practice could be made widespread and all the
facts published, the breed would gradually purge itself of its poor germ
plasm, assuming that no dairyman would use a son of a bull which had
demonstrated his own lack of the genes for high production through the
low production of his daughters. Such a system is very slow, perhaps too
slow and a bull may have many daughters as well as many sons in service
befo;e his own poor inheritance has had time to show itself in the first
5 to 10 daughters' production. Low official-type classification also
requires the surrendering of registration certificates or the prevention
of further registrations.
The other prime function of the breed associations was that of pro-
moting its particular breed. This led to some very worth-while compe-
tition between the breeds and some, perhaps, that was not very much
worth while. Much of this early breed-promotion work was based on
superficialities, largely perhaps because the real basis for animal improve-
ment was at the time unknown. Excessive fat was utilized to produce
even contours, cows were forced to
abnormally high and probably uneco-
nomical milk production in order to
" sell" the breed. Breeds were
judged, by the unthinking, in terms
of a few superlative accomplishments
rather than in terms of their average
level of merit. Demand for pure-
bred sires was very brisk with the
result that every breed retained much
poor germ plasm, and a lot of it
is there to this day. To get a pure-
bred sire seemed to be the easy and
quick way out of all our breeding
difficulties. This attitude is found in 8o<.vtuoftWa.: u.s Department of Agriculture
all fields of human endeavor. We FIG. 7.-Sources of the American farm
dollar in 1948. (Courte8yof U.S. Depart-
want to get to our goal by a quick ment of Agriculture.)
and .easy method, so we take various
and sundry short cuts, but, unfortunately, they seldom or never lead us
to our goal. Breeding better livestock is a slow, tedious process. If we
use proper methods, we can hope to make some progress, but we should
beware of short cuts and panaceas.
The breed associations also set the ideals for their breeders largely
through the awards made at livestock shows. The breed associations
have always exercised their right to have something to say about who
shall judge at livestock shows. Many of them publish lists of acceptable
judges, and some of them conduct judging schools to train judges in breed
ideals and to encourage uniformity in placings in different sections of the
country. In brief, they try to mold their breeds after acceptable stand-
ards of perfection. In all this up to the present, there has been no com-
pulsion. Each breeder is free to follow his own bent, to test or to show
or not to do so. Any breeder is free to attempt to change the presently

accepted breed type, and the types have been changed many times, pre-
sumably in accordance with changing market demands.
Our purebred breed associations have made invaluable contributions
to better livestock. That they should not have made mistakes would
have been miraculous. Their good far out,veighs the bad up to the
present, but they still have much hard work to do. Their final task will
consist of providing the mechanism for proving the producing and trans-
mitting merit of the animals in their respective breeds.
The title of this section, "The Purebred Era," should not be construed
as meaning that the writer believes the purebred has had its day and will
pass on. The work of the purebred breed associations over the past 85
years has laid a firm foundation on which a great superstructure of really
beau6ful and efficient animal types can and wjJJ be built. It has been
our purpose to pay our respects to the untiring work that the men making
up these associations have done and also to point out the ultimate fallacy
of protecting breed purity and promoting the use of purebreds without
taking the necessary steps to purge the breeds of their undesirable germ
plasm by feasible systems for proving breeding merit.
That genetic variations in efficiency in converting feed into milk,
butterfat, meat, wool, or power exist among our animals is now well-
established through carefully controlled experimental work. Probably
there are also genetic differences in disease resistance, maternal charac-
teristics, quality of product, fertility, ease of milking, etc. The next task
of the breeder and breed associations is to devise the recording machinery
so that these inherited differences may be ascertained and the facts made
available to all.
Rise of the Breeds.-For 200 years after the beginning of the settle-
ment of North America, the early settlers and later immigrants brought
livestock to these shores, as none of the larger farm animals were here
when the~ came. These were the native stock of their respective native
lands. There were no breeds as we use the term today, although local
varieties of animals have probably been continuously developed in all
parts of the world for the past few thousand years. Certainly that was
the case in early America, and some of these creations have persisted
'1nd developed into well-established breeds.
The early settlers needed transportation, especially when the country
began to expand westward, and naturally turned to the horse. Through
selection, there was gradually developed a sturdy, sure-footed, intelligent.
and easy-riding horse that was later to provide one of the foundation
stones for the American Saddle Horse, the latter breed also having
infusions of Thoroughbred, Morgan, and Trotting blood. The need for
a means of travel was met in a practical, matter-of-fact way, and to these

b sic fundamental traits were added later the refinements of form and
s:yle to make some of man's finest creations, the American Saddle Horse,
the Tennessee Walking Horse, the Quarter Horse. The story can be
repeated for the American Trotter and Pacer, for the Chester 'White,
Duroc, Poland-China, and Hampshire breeds of swine, and for a few
breeds of sheep, and for one breed of cattle.
We have seen that the mean date for the founding of our livestock breed
associations in America was about the year 1880. This was preceded
by a period of about 50 years of final preparation. The average mean
date for the introduction from other parts of the world of the local vari-
eties of horses that later became recognized as distinct breeds was about
1820, for cattle 1830, and for sh.eep and swine 1840. During the years
from 1830 to 1880, when the breed associations Were formed, strains of
red, black, white, belted, or spotted hogs were developed; likewise strains
of trotting, saddle, general-purpose, and draft horses; of black, red, black
and white, red and white, light or dark-fawn cattle; and of sheep of vary-
ing sizes and qualities of meat and wool. Through a gradual process of
selection, involving both external trade-marks and intrinsic abilities to
render some sort of service to man, the stage was set for the formation of
breed associations, each group of men with somewhat similar likes and
ideals banding together so as to be able to trace pedigrees more accurately
and to merge their efforts of promotion.
All our breeds, foreign or American formed, are of relatively recent
ongm. All of them arose by the same process of choosing certain com-
binations of characters from the general inheritance that makes up cattle
or horses, swine, sheep, goats, dogs, etc. Our breeds arose from a grand
and glorious mixture, and as yet we have purified them only for their less
intrinsically valuable qualities.
The exact origin of a breed, old or new, can never be accurately known.
This point needs no elaboration for the old breeds. nut suppose that we
set out to create a new breed by crossing Guernseys and Holsteins. In
the first-cross animals We would know that one-half of their genetic mate-
rial had come from one breed, one-half from the other; but when we mate
two of these crossbreds, we have no way of telling what each one is trans-
mitting. Assuming there are 60 chromosomes in cattle, then the hybrid
has 30 each from the Guernsey and Holstein breeds, or 60 in all. Each
individual, however, transmits o'nly half of its own genetic material to
each of its offspring, so that in one case the first-cross animal might
transmit 15 chromosomes from each of its two original sources. It might
transmit 16-14 or 17-13 or 18-12 or any combination up to 30-0, but
there would be absolutely no way by which to determine what the com-
22 BREEDING AND IMPROVEMENT OF FARM ANI MALS '.
t·
bination was. So the exact origin in terms of genetic material for a breed
either old or new can never be known. _
Before the advent of breeds, it is logical to suppose that the main
criterion for selection was utility. The halo attached to or developed
around the term purebred was responsible perhaps for relegat' 'proved
merit to something less than the main criterion in selection.' his has
occurred, it is a distinct loss and an impediment that should.. Ispensed
with as quickly as possible. The breeders themselves are 1'f~ 'Ones who
should take the lead in accomplishing this result.
Beginnings of Breeds in England.-During the eighteenth century,
agricultural progress in England quickened. By 1700 from one-third to
FIG. 8.- West llighland cows, of great antiquity and indigenous to Great Britain. (From
Thompson, Outline of S cience, Geo. N ewnes, Ltd.)
one-half of the arable land was still cultivated on the open-field system
with its heavy hand laid against more progressive and efficient methods.
"No individual owner could attempt to improve his flock or his herd,
when all the cattle and sheep of the village grazed together on the same
commons." Common lands with movable landmarks and unfenced land
with crops often despoiled by passing cattle and sheep led to continuous
quarreling and litigation. This type of farming was especially hard on
livestock, as the commons were generally overstocked. Writers spoke of
"half-starved, greyhound-like sheep," of "animals bearing ' a closer
resemblance to living skeletons than anything else," and of a motley
mixture of all the different breeds of cattle and sheep at present known in
the island, many of which are diseased, deformed, small, and in every
respect unworthy of being bred from. It was claimed that 5 acres of
individually owned pasture would have been worth more than pasture
rights over 250 acres of common. Enclosing started about 1450 and
ANIMAL BREEDING-PRESEN7' AND PAST 23
progressed at different rates in different parts of England, sometimes by
agreement, sometimes by acts of Parliament.
With the change from open field to enclosure, better methods of farm-
ing became possible. The great pioneer and the" greatest individual
improver" that English agriculture had ever known was Jethro Tull, who
published a book, "Horse-hoeing Husbandry," in 1733. He developed a
drill for planting seed in rows and practiced clean cultivation to destroy
weeds and conserve moisture. The rank and file scoffed at his ideas,
some of which became established in Scotland before they did in England;
but large landowners gradually adopted his suggestions, so that they
eventually spread over England revolutionizing the whole field of agri-
cultural practice.
With the coming of field cultivation of clover and artificial grasses,
sometime after 1600, and of roots somewhat later, a great impetus was
given to agriculture and livestock breeding. Winter feed could now be
had, more livestock kept, more manure produced, better crops grown,
still more livestock kept, etc., over and over. This provided both the
necessity and the possibility for improving livestock. There had grown
up a great variety of short-wool sheep in many sections of England ..
There were also a number of long-wool breeds, and in general their type
was more uniform than that found among the short-wool sheep. Some
attempts 'were made in the latter part of the seventeenth century to
improve the breeds, but they were generally made in terms of fancy
points rather than practical ones.
In cattle, size alone seemed to rule in selection rather than quality of
carcass or earliness of maturity, though power at the yoke and milking
quality were early criteria for selection. Here again each county had its
varieties, more or less true to some color, length and set of horns, etc.
Among these breeds were the Shorthorn, Middlehorn, Longhorn, Devon,
Hereford, Sussex, Pembroke and Red Glamorgan, 'Vest Highlands,
Ayrshire, Galloway, and Angus. Perhaps the quest for size reached its
zenith in the Lincolnshire Ox, standing 19 hands high and measuring
4 yd. from his face to his rump. "Stock breeding as applied to both
cattle and sheep, was the haphazard union of nobody's son with every-
body's daughter." Prizes were offered for the animal with the longest
legs, necessary enough perhaps at the time.
Robert Bakewell of Dishley (1725-1795) had the imagination to
picture the future needs of a growing population in terms of meat and set
about creating a low-set, blocky, quick-maturing type of both sheep and
beef. He also worked with the heavy Blackhorse, shortened his back,
brought him closer to the ground, replacing height and weight with
activity and strength. Bakewell paid little or no attention to fancy
points but bred for animals that would weigh heavily in the best jointr:;
and show efficient feed conversion and quick maturity.
His success was due largely to three factors. (1), He had a definite
aim in mind and bred for it consistently. Joints preserved in pickle
hung from his walls, and skeletons of his most famous animals adorned
his halls. (2), Bakewell divorced himself from the common practice of
crossing breeds, which tends to dissipate good qualities, and adopted in
its p~ace ~he prac~ice of "breedi~g the best to the ~est" r~~less of
relatIOnshIps. ThIS meant a consIderable amount of mbreeding,,"a prac-
tice generally taboo in the England of that day. Finally, he let for fancy
prices, rather than sold, his males; in other words, he got his neighbors to
"prove" the transmitting abilities of his sires for him.
It was with the old Leicestershire sheep that Bakewell's success was
outstanding. From large heavy-boned and -framed animals with little
or no propensity to fatten quickly, Bakewell gradually brought them
closer to earth, gave them blockiness of form, fineness of bone, and quick-
fattening propensities, which set them off from their forebears as a new
variety. Other breeders ,,,ith animals better fitted to their respective
localities were soon following Bakewell's example with the Leicestershire,
~nd it was not long before Leicesters were closely rivaled if not
surpassed.
vVorking with Lancashires, or Craven Longhorns, Bakewell evidently
went too far in establishing better beef qualities at the almost total
expense of milk. Since the region, Leicestershire, was also a dairy region,
this was fatal, and the Durham Shorthorns bred by the Collings along
the lines laid down by Bakewell largely supplanted the Lancashires.
Breeders of other breeds soon followed the trail blazed by Robert Bake-
well. Animal breeding of this sort became fashionable and respectable
and enlisted the services of many country gentlemen.
How great was the change can be seen in the weights of animals at the
famous Smithfield Market. In 1710, beeves had averaged 370 lb.;
calves, 50 lb.; sheep, 28 lb.; lambs, 18 lb.; whereas in 1795 they were
800, 148, 80, and 50 lb., respectively. Individual initiative encouraged
by enclosures, the introduction and use of turnips and clover, the labors
of agriculturalists like Tull and Townshend helped enormously to bring
about these changes. But the main influence was that of Robert Bake-
well, whose imagination, initiative, and courage put a firm foundation
under improved methods of livestock breeding. Animals and man had
progressed slowly together for 10,000 or 20,000 years. Latent in both
was the promise of a better tomorrow. To Bakewell above all other men
belongs the credit for loosening the shackles that held animals back.
Using the crude principle of "like begetting like," he molded animal
ANIMAL BREEDING--PHJ;JSEN'l' AND PAST 25
inheritance into a preconceived pattern through close breeding and

selection. This system, then as now, leads to purity or homozygosity.
When other breeders adopted Bakewell's methods, the foundations of
our modern purebreds were laid.
Late Middle-Ages Agriculture of England.-During the Middle Ages
(500-1500) in England, as elsewhere, rotation and improvement of crops
and improved breeding methods were not a necessity because virgin soil
was abundant and worn-out lands could be deserted for new. Increasing
population and the establishment of settlements Were later to make
improved husbandry a dire necessity. Thus ,vild field-grass husbandry
waS supplanted by the permanent separation of arable land from pasture-
land. The former was the system in vogue during feudal days in Eng-
land from the eleventh to the fifteenth centuries. At this time there
were four types of land: (1) the demesne or lord's land, (2) the free land
of the military service or rent-paying men, (3) the unfree land worked
by various classes of bondsmen, and (4) the common pastures and
untilled wastes. Each manor was self-sustaining, and there was but
little trade in agricultural products. "For such ready money as he
needed, the lord looked mainly to the produce of his livestock."
Horse farms appear in the accounts of some of the manors in the
thirteenth century, probably supplying the "great horse" for military
purposes. Cart and plow horses were to be found on most farms and
were often worked along with oxen. The latter were steadier, easier
on the harness and plow gear. The winter keep of oxen was less than that
of horses. They needed less care, were less liable to sickness, and on
death yielded both a hide and a carcass. Attendants of both classes
slept in the barns with their charges.
Cattle fared badly in these early days. Pastures, never very plentiful,
were generally overgrazed so that fresh-butchered meat was seldom
eaten. Aged animals and worn-out oxen were prepared after a fashion
through the summer for slaughter in the fall and then eaten fresh, or
salted for winter use. Winter feed was scarce with the result that spring
generally found the cattle in a very thin ·weakened condition.
Dairy cattle fared little better and produced four-fifths of their yield
during the 24 weeks of summer. In the winter milk was worth three
times its summer price. Calves were poorly grown because of the human
demand for milk. Much use was made of all sorts of cheeses and much
summer butter was preserved for winter use.
To the medieval agriculturalist, sheep were the sheet anchor of farming.
Their chief product was not meat, milk, or hides, but wool. From
earliest times, districts like Shropshire, Leominster, and the Cotswolds
had been famous for their wool. The high prices paid for rams is evi-
26 BREEDING AND IMPROVEMENT OP FARiv[ ANIMALS
dence of an attempt at better breeding, though the presence of scab and

rot made sheep raising a risky venture. The shepherd's task was one of
the most important ones on the farm, he was respected above most others,
and lame shepherds were in great demand because a la'me man was not
likely to overdrive his sheep.
Pigs were held in high favor and valued as more profitable than a cow.
They were also invaluable as scavengers. Turned loose, they were self-
supporting during all except 3 months of the year. At farrowing and
fattening time they received some extra care and feed. In. type they
were rangy, lop-eared, and agile. What selection was practiced was
based largely on the animal's ability to fend for itself, "root hog or die."
To sum up the status of livestock in thirteenth-century England, ,ve
have seen that types of animals to fit the contemporary need had been
developed, but that means of adequate feeding or very high ideals of
better breeding were sadly lacking. Nevertheless an adage of the time
ran, "He that hath sheep, s\vine, and bees, sleep he, wake he, he may
thrive."
The open-field system, including the arable land, the meadow, and
the permanent pastureland, prevailed quite generally in England at some
time or other, and examples could stm be found up to the eighteenth
century. The manor system of farming tended to break up between 1300
and 1485, largely because the lords became more interested in sport and
war than in improving farming practices. Famine followed, owing to
lack of interest by the large owners and the lack of resources of lesser men.
Whatever improvements were registered were largely the contribution of
monks, winter was a season to be dreaded. Gradually, and for a variety
of reasons, the manor system with its graduated mutual dependence
began to break up and was supplemented by a system that involved
greater individual independence.
The Black Death (1348-1349) hastened the breakup of the manors as
it increased the lord's land but greatly decreased the available labor
supply. Rents fell, wages rose, and extensive legislation designed to aid
the landlord could not stem the tide of manorial breakup. More and
more small separate farms came into being that were let out for money
rents to individual occupiers. Freeholders, leaseholders, renters, and
free laborers increased at the expense of the class of vmeins.
Thus, shortly before 1500, feudalism in England practically ceased to
exist-the Middle Ages were supplanted by modern times. Feudalism
was followed by enclosure. Open fields and pasture· commons ,vere
enclosed, sounding the death knell to the old medieval agrarian partner-
ships and signifying the birth of individual ownership. The rise of a very
profitable wool trade as well as the increased demand for meat turned
much arable land into pasture, much of the change being unwarranted.
Land now came to be thought of in terms of what rent it could earn; in
other words, agriculture became commercial.
Grass holdings were enlarged at the expense of tillage in order to
accomodate the sheep-farming craze. In the process many small farmers
were ruined and many once happy villages became "deserted." Laws
passed to preserve farm homes and buildings and to require tillage were
easily circumvented, and preachers and versifiers inveighed against sheep
farmers and the sheep themselves. But the greater efficiency of enclosed
land, owned and worked by an individual, over open land, jointly owned,
gradually made itself felt, and a new era of progressive agricultural
development opened for England. The price for enclosing some %
million acres was the throwing out of work of around 35,000 persons.
That there was much suffering can be assumed from the numerous rural
uprisings of the time, but there were also attempts to deal fairly and
adjudicate liberally.
Unaccompanied by a contemporaneous development of arable land,
enclosures caused a growing scarcity of employment. Tenant farmers,
copyholders, squatters, cottagers, and laborers suffered alike, and the
rise in the cost of living pushed many to the brink of starvation. Some
individuals waxed rich through enclosure, so that the gro'wing unequal
distribution of wealth brought many abortive attempts at legislative
relief for the unfortunate. Petty crimes, kidnapping, livestock stealing,
and similar misdemeanors grew in the rural districts, though a goodly
part of it was doubtless perpetrated by disbanded soldiers, adventurers,
and other ne'er-do-wells.
During the reign of Elizabeth (1558-1603) "wool was the chief source
of the wealth of traders and of the revenues of the crown" and controlled
the foreign policy of England. Short wool had serious competition from
the Spanish Merino. Ryeland was the preeminent short-wool breed,
its center at Leominster and "Lemster ore" the equivalent of "golden
Fleece." The Cotswolds led the long-wool breeds with Lincolnshires
close behind.
On the whole, there was not much change in the sixteenth- as compared
with tlli:) thirteenth-century agriculture in England. Some gains were
registered, e.g., the greater use of iron in the implements used, but arti~
ficial grasses and roots were still unknown, hops the only new crop.
Seasonable operations remained about the same, with magic and super-
stition still controlling plantings and breeding operations. Sheep 'were
considered the "most profitable cattle a man can have," though without
turnips their full value could not be realized. Ewes were still milked,
and the writers of the time have "nothing to say about the improvement
of breeds of cattle for the purposes that they serve." The general-
utility animal was still the ideal. The combining of crop production with
livestock feeding was yet to come.
Medieval Continental Agriculture.-In Western Europe conditions
were chaotic following the breakup of the Roman Empire, which started
in the third century and was complete by the seventh.
The work of reconstruction in Western Europe began in the eighth
century . Various rulers, and particularly the Christian church, provided
the main impetus. Methods and tools were crude and results in terms of
effort meager, but progress was made and the hope of economic security
was again revived. Thousands of monasteries were established all over
Western Europe, and each of them provided a nucleus for agricultural and
livestock-breeding endeavors. Large herds of cattle, horses, sheep, and
swine were again established, horticulture and winegrowing again came
into prominence.
Even after their adoption of the rudiments of civilized living, the
majority of the Celts, Anglo-Saxons, and Germans \Vere largely herdsmen,
and in central Germany cattle were still a means of exchange up to the
eighth century. The ratio of small animals like hogs, sheep, and goats
to larger ones like cattle and horses ranged from 5 up to 9: 1.
The period from 900 to 1200 saw the rise and the establishment of
feudalism in Western Europe. Land at that time was practically the only,
capital and it came to be controlled by the ecclesiastical and military
classes. Communal ownership of land had changed first to small individ-
ual holdings. These were gathered together into great blocks by powerful
organizations or individuals. As the saying had it at that time, "no lord
without land, no land \vithout a lord." There were still some free men
scattered about, but one-third to one-half of all the land, for instance,
was owned by the Church and another large percentage by powerful
knights and barons. This land was given out as fiefs to be worked by
slaves, serfs, half-free, and free peasants. St_ated returns went to the
lord, and generally the workers also owed the lord or the monastery a
stated number of days of work per week.
Nothing, of course, was known of scientific agriculture, and the lot
of the peasant and his animals was not a happy one, a horse being worth
100 sous and a peasant 38 sous in the eleventh century. Fallowing was
used in place of crop rotation and the cultivation of roots and grasses.
Gradually centralized feudal states emerged, nationalism developed, and
the monarchies began to revive. This led to extensive colonization all,
bver Europe from the eleventh to the fourteenth centuries. With this
came some attention to agriculture, irrigation, development of grasslands
and fodder crops, and attempts to foster an empirical veterinary art and
to improve livestock. Studs were set up to try to improve the horse for
both war and commerce, and some sections became noted as mule-breed-
ing districts. Cattle also received more attention from the standpoint of
meat and milk as well as draft. Small animals, however, continued to
dominate, sheep taking the first rank. Millions of sheep were to be found
in the Low Countries, France, and Germany, as well as huge migratory
flocks in Spain and Italy. There was improved cattle breeding in
Scandinavia by the twelfth century, and this region was exporting butter,
cheese, fats, lard, and hides. The art of fattening cattle on roots and
legumes was first developed in the Low Countries.
The eastern portion of the Roman Empire (the Byzantine) fared better
following the crisis of the fifth o~ sixth centuries. For 1,000 years, it
held off or quickly recovered from the onslaughts of the barbarians. It
was realized that the two pillars of resistance were agriculture and
military art. Through an intelligent system of agricultural colonization,
hundreds of thousands of outsiders, including barbarians, were absorbed
into this Eastern Empire. Agriculture, animal husbandry, and horti-
culture flourished, early Greek and Roman agricultural practices were
revived, thereby creating the wealth that made this part of the world the
envy of all men during the Middle Ages. In the early part of the period
the land was held largely by individual owners, but by the tenth or
eleventh century most of tlle lands had gotten into the hands of the
Church or of large landholders. Slavery, however, was abolished, and
the serfs were accorded privileges and rights, which stabilized rural life
with the result that disciplined labor gave to the Eastern Empire its
chief economic supremacy. 1
The East Roman Empire thus accomplished, during the early Middle Ages, a
task of supreme importance. It had received undiminished the heritage of Rome
and added to it. It left a profound mark upon every kind of work. It succeeded
in colonizing the Christian lands of Eastern Europe and it civilized the barbarians,
calling them to the fruitful labours of peace. It gave a powerful impulse to
every form of economic activity, and carried the production of wealth to the
highest point. If in the social order it only half succeeded in protecting free
labour and free property against the exploitation and usurpations of the aristo-
cratic classes, it yet suppressed slavery, and strove with all its strength to mainJ
tain the middle class, both urban and rural. Thus it placed itself in the van of
civilization, whose great traditions it carried on, and it was partly to Byzantium
1 BOISSONNADE, P., "Life and Work in Medieval Europe," pp. 60--1n, Alfred A.
Knopf, Inc., New York, 1927.
30 BRE};JDING AND IMPROVENlEN '£ OF FARM ANIMALS
that the West in its turn went to school to be prepared for its own civilizing
mission.
Roman Agriculture and Animal Husbandry.-The seven separate hills
of Rome, settled at first by humble herders and hunters, were eventually
drawn together into the Septimontium (League of the Seven Hills) and
destined to become the ruling center of the greatest empire the world
had ever known. Agriculture in the surrounding region kept pace with
the growth of the city and came to enjoy a real prosperity. - The unpaid
militia, consisting to a large extent of small farmers, was often called upon
to protect the region from invaders, but the campaigns were seldom of
long enough duration seriously to affect agricultural practice. This is
not true, however, of the period from the First Punic. War (against
Carthage) in 264 B.C. down to the time of the death of Augustus, A.D. 14.
FIG. 9.-Statuettes of Roman animals in bronze, first century B.C. to third century A.D.
(Courtesy of 7'he Metropolitan Museum of Art.)
These were the times of Roman wars of expansion, and they were long
and bloody. The middle-class farmer bore the brunt, waoS slaughtered
by thousands, with the result that a great deal of the property fell into
the hands of a few rich owners. In addition, thousands of the conquered
were brought to Rome and sold as slaves to work the land. The earlier
type of Roman agriculture gradually found itself unable to compete with
outlying conquered territories in the matter of supplying grain. The
type of agriculture, therefore, shifted to greater numbers of livestock,
especially horses, sheep, and hogs, and to increases in orchards, vine-
yards, and gardens. The country became almost deserted except for
foreign slaves and barbarians, the drainage works fell into ruin, and the
old unhealthfulness returned. l
1 TOUTAIN, J., "The Economic Life of the Ancient World," p. 234, Alfred A. Knopf,
Inc., New York, 1930.
This transformation of rural economy led in the end to a social phenomenon of

great importance in the history of Rome. The countryside was deserted by most
of the small landowners and free tenants, who came crowding into the towns, and
especially to Rome. This influx of poor people, ruined by the new condition of
farming, enormously increased the mass of the Plebeians, which was at the same
time swelled by increasing throngs of freedmen of foreign origin, Greek and
Oriental. There was no longer a balance in the citizen body between the rural
elements, which had once been hard-working, productive, and level-headed, and
the urban proletariat, which grew more and more accustomed to call upon the
State, the magistrates, candidates for office and ambitious men of every kind
for its food and amusement-panem et circenses. Rome ceased to be the capital
of an essentially agricultural people, whose wealth was mainly based on landed
property and agricultural resources; it became a turbulent agglomeration, in
which industry, trade, and money-~ealing assumed an importance hitherto
unknown.
The formation of the Roman Empire near the beginning of the Chri;;-
tian Era brought Gaul, Spain, Africa, Greece, and Egypt under Roman
influence. Rome demanded economic, political, and religious loyalty
but allowed old native traditions, habits, and sentiments to persist, pro-
'Tided they did not interfere with the loyalties she demanded. Economic
life in these conquered areas went on much as before, though quickened
by inclusion in a greater united whole. Roman legions and fortifications
held back the barbarian hordes on all sides and gave a comparatively
long period of relative peace throughout the Empire and in Rome itself,
when the civil wars were extinguished.
So far as we can learn, there were no great agricultural innovations or
livestock improvements either in Italy proper or in the other parts of the
Empire during this time. Of a slow, steady evolutionary process, how-
ever, we have reason to be assured. Selection, no doubt, went on in live-
stock breeding, and the long period of peace during the first two centuries
of the Christian Era was bound to have had its effects. We get some
indication of the relative importance of various types of agriculture in the
Empire from· the fact that Columella's 13 books De re rustica include no
whole book devoted to grain growing, 272 books devoted to winegrowing,
about 1 each to fruit trees, including the olive, and to gardens, while
livestock occupies 4 books.
The Empire consisted of cities with surrounding agricultural lands,
orchards, and vineyards. Beyond were to be found the pastoral nomads
with their flocks and herds; and in the hinterland, the barbarians. Each
region of the Empire adapted its crops and animals to its particular topo-
graphical and climatic conditions. Variety there was to be sure, in lands,
in peoples, in methods. Although there were large estates in all parts
32 BREEDING LiND IMPROVEMENT OF FARM ANIMALS
of the Empire worked by slaves or tenants, the real backbone of the fabu-
lous old Roman Empire was undoubtedly to be found in the sturdy, indi-
vidual small farmer, businessman, and trader. He it was who created
the wealth and provided the sinews that made the Empire. When this
class, for whatever reason, was forced out of existence, the" grandeur
that was Rome" quickly faded.
Beginning with the third century, the power of Rome began to wane.
Slowly but surely the ordered, progressive life in a ,yell-tended country-
side and in hundreds of cities with fine public buildings, temples, and
homes was undermined by anarchy, insecurity, and discord at home and by
successive waves of barbarians from the north and east. Imperial edicts
proved a poor substitute for the individual efforts of free men. Thus
the structure of the Empire was weakened from within and rendered
highly vulnerable to attack by tribes of vandals which gloried only in
destruction, murder, and rapine. So thoroughly did they carry out their
mission of destruction that by the seventh century a once productive
countryside had largely reverted to waste, the once proud cities were in
ruins, and the population cut in half. There was barely enough courage
and initiative left to start the long, tedious process of rebuilding, which
was to consume the next 600 or 700 years. The effect on livestock breed-
ing was particularly severe. Herds and flocks were wantonly destroyed
and their half-starved remnants turned back to a state of nature.
Grecian Agriculture and Animal Husbandry.-Before the time of Alex-
ander, agriculture and stockbreeding were one of the chief sources of
,,'ealth in Greece and her colonies. Their stables and byres contained
horses, cattle, sheep, swine, and goats, which provided food, clothing,
milk, and power. These early Greeks also kept bees, and the poultry
yard contained geese, ducks, guinea fowl, and pigeons as well as hens.
These people also consumed various and sundry products of the field,
grains, vegetables, and fruits, and their banquets and religious ceremonies
called for the liberal use of wine. In the early days of Greece, the stand-
ard of value was measured in terms of cattle, and among the prizes sug-
gested by Achilles for the funeral games in honor of Patroclus were,
besides mares and mules, a tripod valued at 12 oxen and a slave girl
valued at 4 oxen. There was considerable barter involving the exchange
of cattle for metals. The Greek poems of the period contain many refer-
ences to wealth reckoned in terms of flocks and herds, just as in later
Roman times the word for money, pecnnia, was derived from peens,
cattle. The earliest known Greek coins date from the seventh century
B.C.
Economic activity at this time was mainly agricultural and pastoral,
pursuits which provided most of the wealth. Through the period from
A.NIMAL BREEDING-PRESENT AND PAST 33
the eleventh to the fourth centuries B.C., some improvements in details of

agriculture and animal husbandry are recorded, but there was no radical
change in agricultural equipment or methods and no evidence of any
marked development in stockbreeding practices. Much of the work was
done by slaves, though there were also individual farmers who owned and
operated their farms as ,,·ell as some farms that were run by hired lahor.
FIG. 1O. -Stat uette of Greek horse in bronze, fifth century B.C. (Courtesy of The ~Melro
,olilan Museum of Art.)
When Alexander of Macedon led the Greeks to the conquest of Asia

inor and Egypt, a new and rich agricultural economy fell under Greek
domination. Here they found a flourishing livestock industry: 30,000
mares with 300 stallions in one stud in Syria, large herds of camels farther
east between Syria and the Euphrates Valley. Great herds of oxen and
horses were found in the well-watered districts from the Mediterranean
to the Black Sea with flocks of sheep in the higher plateaus. Even greatel
apicultural and livestock wealth awaited them in Egypt in spite of the
misrule of the Persians during the two preceding centuries. Egypt now
became the granary of the Mediterranean world in addition to supplying
4M)untless other vegetable and aromatic products as well as fruits.
Livestock too abounded in Egypt: horses, cattle, sheep, goats, poultry,

of various sorts, and the real beast of burden, man's fellow ,,'orker in the
fields and on the roads, the lowly ass. In addition Egypt supplied all
sorts of wild game, gazelle, antelope, wild ox, goat, hare, hyena, lion,
hippopotamus, crocodile, and various waterfowl. Thus a whole new
world was opened up by the Greek conquest. There were huge temple
properties and royal domains in both Asia Minor and Egypt, the former
apparently worked by slaves and the latter by royal farmers, \\'ho were
somewhat of the nature of serfs and tenants, as well as by salaried
workers. In addition, there were large estates worked by tenants and
FIG. ll.-Wall painting of Egyptian bull during XVIII dynasty. about 1400 B.C. (COu.,·_
te8yof The Metropolitan Mu.seum of A,·t.)
probably some small, free landholders. Greek agriculture and animal

husbandry certainly took on a greater diversity during the centuries that
saw Greek influence spread to Asia Minor, Syria, and Africa, but there
is little evidence of any marked improvements in the techniques of either.
Beneficial variations were no doubt seized upon and in some cases main-
t.ained or increased, but the general level of accomplishment in stock-
breeding remained about where it had been.
Oriental Agriculture and Animal Husbandry.-Colored drawings on the
walls of the tombs clustered around the Pyramids of Giza depict the life
of the people who lived in Egypt about 3000 B.C. The lord of the estate
is pictured looking out over his fields, the oldest agricultural scene now
ANiMAL BREE'DING-PRESENl' AND PAST 35
kno,,'n. Cattle are shown being used to pull a crude wooden plow; others
are shown grazing in pasture and still others being milked (after their
hind legs had been tied to prevent kicking). Donkeys too are much in
evidence, but the horse was apparently unknown to the Egyptian at this
time, not appearing there until about 1700 B.C. At the Plain of Shinar,
later knmvll as Babylon, carvings and drawings from about the same
time show oxen at the pImy, sheep and goats at pasture, and donkeys
pulling ,,-heeled carts and chariots, the wheel appearing here for the first
time. Horses made their appearance in Western civilization about 200
B.C. when they were brought to Babylon by the conquering mountain
I - , • .,-,-:- ~ ,.
FIG. 12.-WaU painting of Egyptian hunter during XVIII dynasty, about 1400 B.C.
(Courtesy of The Metropolitan Museum of Art.)
Kassites, ,,·ho in turn had gotten the domesticated horse from tribe::; to
the north and east. To the north of Babylon ,ms another early civiliza-
tion, the Assyrian, with its capital at Nineveh. These people too had
herds of oxen, sheep, and goats and used donkeys for draft purposes.
They also had domesticated horses which, together with their weapons of
iron, they used very effectively in their conquests of neighboring regions.
The peoples who made the Babylonian, Assyrian, and Chaldean empires
came largely from the grassland fringes of the Arabian desert.
In the north was a second fount of nomadic conquerors located in the
gasslands stretching from the lower Danube along the north side of the
Black Sea through southern Russia and into Asia. This was a white race
ealled Indo-European, and most of the people who will read this book are
their descendants. Their original home was the grassy steppe east and
northeast of the Caspian Sea. They were nomadic tribes following their
flocks and herds on horseback and using horses for draft. These people
were the ancestors of the Greeks and Romans and of most of the peoples
of Western Europe and America. Their language group is Aryan, which
includes English, German, Latin, Greek, old Persian, and East Indian,
although these, people themselves were without writing. The Medes and
the Persians were but the Eastern wing of this movement of original
Northern nomads.
Here the historical record fades out. What preceded it in the field of
agriculture and animal husbandry can only be surmised from occasional
relics. This historical record we have traced backward from the present,
a period covering about 5,000 years. We have noted some progress, but
the tempo of change was very slow. We are perhaps justified in assuming
that the period from the earliest do~estication of animals up to this
beginning of the historical record occupied a much longer time.
Summary.-We have tried in this chapter to start from the known
facts about our livestock situation of today and trace it backward to its
historical beginnings. In these few pages we have covered 5,000 years
of man's struggle from barbarism to a certain state of civilization. Much
was accomplished, but a vast deal more still remains to be done. With-
out any sure knowledge, except what he could observe in the gross, man
has made tremendous strides forward in breeding better livestock, aided
materially, of course, by better housing and management and especially
by better feeding. For marked progress in animal breeding, we need
knowledge of principles, ideals, and good feeds together with peace or at
least freedom from invasion. All these except the first were finally pres-
ent in England during the eighteenth century, and it was there and then
that the foundations of our modern breeds were laid.
References
ALLEN, L. F. 1884. "American Cattle," Orange Judd Publishing Co., Inc., New
York.
ALLEN, R. L. 1847. "Domestic Animals," Orange Judd Publishing Co., Inc., New
York.
BIDWELL, P. W., and FALCONER, J. I. 1925. "History of Agriculture in Northern
United States 1620-1860," Carnegie Institution of Washington, Washington,
D.C.
BOISSONNADE, P. 1927. "Life and Work in Medieval Europe," Alfred A. Knopf,
Inc., New York.
BRANDT, L. 1945. "The Reconstruction of World Agriculture," W. W. Norton
& Company, New York.
BREASTED, J. H. 1935. "Ancient Times," Ginn & Company, Boston.
BURCH, G. I. 1946. "Population Bulletin, Vol. 2, No.5," Population Reference
Bureau, Washington, D.C.
LRMAN, H. J., and TUGWELL, R. G. 1934. "Upon Field Husbandry in New
England," Columbia University Press, Kew York.
LRRIER, L. 1923. "The Beginnings of Agriculture in America," McGraw-Hill
Book Company, Inc., New York.
JRTLER, W. H. R. 1909. "A Short History of English Agriculture," Oxford
University Press, New York.
A-UBENY, C. 1857. "Lectures on Roman Husbandry," Oxford University Press,
New York.
~NLE, LORD. 1927. "English Farming Past and Present," Longmans, Green &
Co., Inc., New York .
•rmers in a Changing World. U.S. Dept. Agr. Yearbook, 1940.
JRNAS, C. C., and FURNAS, S. M. 1943. "The Story of Man and His Food," The
New Home Library, New York.
RAS, N. S. B. 1925. "History of Agriculture," Appleton-Century-Crofts, Inc.,
New York.
A-RRISON, F. 1917. "Roman Farm Management," The Macmillan Company,
New York.
is, J. J. 1936. "Domestic Animals of Mongolia," review by Dr. D. Kislovsky,
Jour. Hered., January, 1938.
EILSON, N. 1936. "Medieval Agrarian Economy," Henry Holt and Company,
Inc., New York.
OTESTEIN, F. 'V. 1945. "Food For The World," University of Chicago Press,
Chicago.
~ARSON, F. A., and PARLBERG, D. 1944. "Food," Alfred A. Knopf, Inc., Kew York.
- - and HARPER, F. A. 1945. "The 'Vorld's Hunger," Cornell University
Press, Ithaca, N.Y.
lENTICE, E. P. 1935. "Breeding Profitable Dairy Cattle," Houghton Mifflin
Company, Boston.
IIOMPSON, W. S. 1944. "Plenty of People," Jacques Cattell Press, Lancaster, Pa.
1942. "Population Problems," 3d ed., McGraw-Hill Book Company, Inc.,
New York.
CHAPTER II
EARLY MAN AND ANIMAL DOMESTICATION
Man is a late arrival on the planet Earth. When he arrived, he found

the earth already inhabited by myriad forms of living things. He was
apparently just another in a long series of creations extending back for
hundreds of millions of years to the beginning of life, just another mani-
festation of the great stream of life that far transcends man's own imme-
diate interests. Before him, in time, 'had arrived many of the other
mammals as well as the birds, reptiles, amphibia, fish, and a whole host
of invertebrate forms.
It is estimated that life has existed on this planet for about 1,200
million years. Up to about 500 million years ago, the only animal forms
were creatures without backbones, the invertebrates. The vertebrates,
animals with backbones and an internal skeleton, appeared about 480
million years ago, the first land-living forms about 350 million years ago,
the mammals about 160 million years ago, Homo sapiens (wise man)
about 25,000 years ago, and we have something of a historical record of
his doings for about 5,000 years.
The record yielding the above facts is 'written in the rocks of the
earth's crust. The time intervals are calculated on the basis of the rate
of breakdown of certain rare minerals and for more recent periods by
counting the annual glacial varves. Even the ancients must have sensed
something of the span of time when they wrote "A thousand years in
Thy sight are but as yesterday when it is past."i In these trying times
facts such as the above should give us new courage. In a ridiculously
short span of time man has made tremendous strides forward. To say,
as is sometimes done, that human nature does not change is but to show
one's ignorance of the facts. Present man is probably very different
intellectually, morally, and spiritually from his forebears of 50,000 years
ago. In a few hundred years, recent man has come largely to understand
and to be the master of his natural world. True, the much more difficult
task of becoming master of lilmself yet remains, but in the light of the
accomplishments of the stream of life over the past billion years this does
not seem to be an impossible task.
The first skeleton of "modern" man appears in the geological record
1 Psalm 90: 4.
38
EARLY MAN AND ANIMAL DOMESTICATION 39
about 25,000 years ago. Starting with Pithecanthropus erectus (walking

ape man), who dates back from ~~ to 1 million years, an incomplete and
rather sketchy series of prehuman forms runs on down and finally emerges
as modern man. Man is included in the Primates, a group of animals
comprising the lemurs, monkeys, apes, and man. This does not mean
that man has descended from monkeys but probably that man, monkeys,
and the great apes are all offshoots of a common progenitor whose identity
is still uncertain. The apes, therefore, are our cousins and not our direct
ancestors.
The primates as a class are not of very recent origin, for fossil remains
of the lemurs are known from the Oligocene and Eocene periods, i.e., 35 to
55 million years ago. It is impossible to say when the stock that was
eventually to form man split off from the parent stock, though it is
believed that both man and the modern apes are descendants of a common
progenitor in the Cenozoic period. N either is it known where, geo-
graphically, the split occurred, though either Asia or Africa seems the
most probable, or whether there was one split or several in scattered
localities.
The close relationship between the great apes and man is attested by
the striking similarities in all their physical characteristics. One can
easily point out differences between a gorilla and a man, but a detailed
study would show very much more similarity than difference between
these species. The monkeys and apes excel in certain respects-man in
others. The most striking variations in man are, of course, the upright
gait, which frees the hand to become a grasping, tool-making and -using
organ, and the development of the brain. Man's brain is the finest
creation of the evolutionary process and is not to be thought of as some-
thing supernatural, miraculous, or of sudden appearance. It is a slow,
progressive development of evolution and still manifests a wide range of
. variation.
Early Forerunners of Man. Pithecanthropus. erectus.-This, the ape
man of Java, was discovered by Dr. E. Dubois of Holland on the island
of Java in 1891. The find was made 45 ft. below the present level of the
terrain in association with 27 different varieties mainly of extinct mam-
mals but including the deer, boar, elephant, and hippopotamus. The
find consisted of a skull cap, a left thighbone, two upper molar teeth, a
lower premolar, and a fragment of a lower jaw and is generally dated as
of early to middle Pleistocene Age. The skull cap is thick-walled and has
a cranial capacity of about 900 cc., which is somewhat below the lower
limit of present-day human skull capacity, and the brain gives evidence
of being closer in form to the human than to the ape. The skull cap
indicates a low cranial vault and a narrow receding forehead, and it has a
'4-0 BREEDING AND IMPROVEMENT OF FARM ANIMALS
TABLE 7.-COSMIC OR ASTRONOMIC TIME CHART*
Eras Periods Time, years

! Characteristic life
Psycho zoic ... Recent 25,000 Man
Pleistocene 1,000,000 Ice Age

Pliocene 7,000,000 Emergence of man
Miocene 19,000,000 Culmination of mammals
Cenozoic, Oligocene 35,000,000 Modern mammals domi-
4 per cent nant
Eocene 55,000,000 Archaic mammals, first
horse
Cretaceous 95,090,000
i
First grasses
Culmination of reptiles
Mesozoic, Comanchian 120,000,000
1
First flowering plants
Giant reptiles, dinosaurs
11 per cent
Jurassic 155,000 ,000
1
First birds, first mammals
First modern trees
Triassic 190,000,000 Rise of dinosaurs
,Permian 215,000,000 Rise of reptiles

~ Rise of insects
Pennsylvanian 250,000,000
Ferns and seed ferns
Tennessean 275,000,000 Rise of sharks
Mississippian 300,000,000 Rise of echinoderms
~ First land plants
Devonian 350,000,000
First land animals
Palaeozoic, ;)
30 per cent \ Silurian 390,000,000
1
First lungfishes
}~incinnatian
First scorpions
425,000,000 First corals
Ordovician 480,000,000 First armored fishes
Canadian 490,000,000
Ozarkian 505,000,000 Trilobites dominant
Acadian 525,000 ,000 First shellfish
\Taconic 550,000,000 First known marine fauna
Proterozoic,
25 per cent . .............. . ............ . Primitive invertebrates
Archaeozoic,
30 per cent . .............. . 1,800,000,000 Earliest life
* LOOMIS, F. B., "The Evolution of the Horse," p. 24, Marshall Jones Co., Boston, 1926.
EARLY "MAN AND ANIMAL DOMESTICATION 41
very well-developed supraorbital ridge that is suggestive of the great

African apes. The thighbone indicates an upright, bipedal posture, and,
judging from this bone, the creature stood about 5~~ ft. in height. The
two molar teeth are now thought to be those of an orang, although another
much more humanlike tooth was found in the same stratum sometime
later, "which, if it really belonged to Pithecanthropus, adds another to
this creature's manlike characteristics.
Whether Pithecanthropus was a direct ancestor (missing link) or just
a collateral relative is still a question. At any rate, he still remains the
lowest known member of the family Hominidae. Since no tools or
artifacts were found associated with these bones, we can guess nothing as
to the degree of culture that members of this race had attained, though it
seems safe to assume that any cuhure they may have possessed ,,,as of a
very primitive type.
Another find (by Dr. G. H. R. von Koenigswalde) was made in Java
in 1936. This was the skull of an infant about one to two years of age,
though the teeth, which ,,,ould give the best indication as to age, were not
found. The capacity and characteristics of this skull seem to place it in
the Pithecanthropus group of the lower to middle Pleistocene, although
some authorities would designate it as a skull of an infant Solo man of
upper Pleistocene age. In 1937, Dr. von Koenigswalde found an adult
Pithecanthropus skull in Java including four teeth of undoubted human
form though of great size. This skull has an estimated capacity of about
750 cc. More recently a fourth specimen of Pithecanthropus Was found
of the same general type.
Sinanthropus pekinensis (Peking Jfan).-The finding of human teeth
in cave deposits near Peking, China, in 1926 and 1927 led to further search
in this region for human remains. The search was rewarded in 1929 when
Dr. W. C. Pei of China found a nearly complete skull. Another skull
was found in 1930, and since that time many fragmentary skulls, teeth,
and parts of skeletons have been discovered. In 1936, three more skulls
were discovered, including some of the facial bones. The first skull had a
cranial capacity of something over 900 cc., and the capacity of the second
was somewhat greater. The three latter skulls have capacities of 1,200,
1,100, and 1,050 cc., the last probably a female.
The general form of these skulls is similar to that of Pithecanthropus,
the teeth are of human type and larger than those of recent man though
there is still no chin. Associated animal fossils indicate that these men
lived at about the same time as the Java man, viz., during the middle
Pleistocene; i.e., 500,000 to 17~ million years ago. It seems established
that the Peking man used fire and had crude implements and tools made
of chipped stone or the bones of animals. The greater brain capacity
42 BREEDI NG AND I MPROVEMENT OF FARM ANIMALS
and manlike teeth perhaps place Sinanthropus a little higher in the scale
than Pithecanthropus.
Eoanthropus dawsoni (Piltdown M an).-This discovery was made by
Charles Dawson in 1911-1912. It consisted of several fragments of a
human skull, much like that of a modern except for its excessive thickness
of bone, and , in close proximity, the right half of a lower jaw containing
two molar teeth. This portion of jaw is very similar to that of a chim-
panzee, a creature, however, never found fossilized in England. Later
finds, near Piltdown, of skull fragments and teeth tend to confirm the
once-doubted fact that the skull and jaw belonged originally to the same
creature. The cranial capacity seems to be about 1,200 cc., and the fore-
head not only lacked the supraorbital ridge but was quite vertical. The
FIG. l3.-Prehistoric man; Pithecanthropus, Piltdown, Neanderthal, ero-Magnon.

(Courtesy of The American Museum of Natural Hiatory.)
apelike lower jaw has no chin but its jaw socket is formed as in Homo
sapiens. A chipped flint and a large bone implement were found in the
deposit that is generally accepted as being of early Pleistocene. These
fossils have caused a great deal of argument among anthropologists, and
it is still not clear where the Piltdown man fits in the general scheme of
man's forerunners.
Homo heidelbergensis (Heidelberg Man).-This find , of a lower jaw,
was made by Professor Schoetensach at Mauer, near Heidelberg, Ger-
many, in 1907 in river sand 79 ft. below the present surface in a stratum
that had yielded many bones of Pleistocene animals. This individual
is thought to have Eved 150J OOO to 300,000 years ago. The jaw, while
very heavy, has teeth that are entirely human. No artifacts were found
associated with this jaw, so that nothing can be guessed as to the cultural
status prevailing during his time. Again it is impossible to assign a
EARLY MAN AND ANiMAL DOMESTICATION 43
definite place to this undoubted forerunner of man. It should be noticed

that we have herewith used for the first time the genus name for man-
Homo. That is to say heidelbergensis appears human enough to be placed
in the same genus with present man-but he is not yet called sapiens
(wise).
Homo neanderthalensis (Neanderthal Man).-We now reach a type
concerning which something more definite may be said. First of all,
Neanderthal man spread over the continent of Europe, into the Channel
islands, and into Palestine. He had a well-defined flint-tool industry,
used fire, probably buried his dead, lived contemporaneously with the
reindeer, wild horse, cave bear, and woolly mammoth during a time
extending from 200,000 down to 25,000 years ago.
The first known Neanderthal.skull was found in 1848 at Gibraltar,
though it was not recognized as such or studied until 1906. It was that
of a female with a brain capacity of about 1,300 cc. The discovery at
Neanderthal occurred in 1856. This skull has a very low vault, heavy
supraorbital ridges, and many unusual features in the arm and leg bones.
One of the most complete skeletons thus far discovered was found at
the cave of La Chapelle-aux-Saints in France in 1908. This was an
almost complete skeleton ,yhich had been buried in the floor of the cave.
It has been studied in complete detail by Prof. M. Boule of Paris. Other
Neanderthal skeletons or parts thereof have been discovered in various
parts of France, Germany, Italy, Yugoslavia, Palestine, and in the
Crimean Peninsula on the Black Sea. As would be expected, there is
considerable variation in these skeletons, but no basis as yet exists for
breaking neanderthalensis up into several "races."
The general type is that of a short, heavy-boned man. The brain
capacity, both sexes considered, varies from about 1,100 to 1,600 cc. The
head is long but does not extend very high above the ears. There is 3,
huge supraorbital ridge, a retreating forehead, large eye orbits, a wide
and rather prominent (not flat) nose, long upper lip, and little or no chin.
The forearm and shin are short in comparison with the arm and thigh,
the thumb is opposable, the big toe not (just as in modern man), and
there are several apelike characters about the feet. Neanderthalensi8
probably walked more uprightly than many artists have represented.
This hurried survey of man's forerunners has brought us down to the
last phase of the Old Stone or early stages of the New Stone Age. Our
last representation of preman, neanderthalensis, lived during the Old
Stone or Paleolithic Age. He had come a long way from Pithecanthropus
in physical beauty and probably had progressed mentally as well. So
far as is known, he was still a hunter and food gatherer. He had, hOIY-
ever, learned to make and use fire, which, together with his earlier tool-
making, was an important step in his ('lmancipation from environmental

bondage. No one knows how he lived, but undoubtedly it was a rather
precarious existence---killing or catching enough game to piece out the
roots, berries, and seeds which he could find, and all the time in danger
himself of being devoured. To learn how best to hunt or fish, which
foods would nourish and which kill was a long slow process; but, when
once learned, it was undoubtedly quickly absorbed into the cultural
tradition and passed along to each new generation. Considerable skill,
courage, and cooperative effort were undoubtedly needed to hunt the
mammoth, bison, wild horse, and reindeer that roamed over the steppes
and plains. These hunters undoubtedly followed the herds as they
migrated, north to Russia and Siberia-south to the Danube basin. We
get a glimpse of the fauna of the upper Paleolithic through the fact that
these peoples carved animals in stone or ivory, drew and painted them
on the walls of their caves, and modeled them in clay.
These hunters and food gatherers were content to live on what nature
provided though undoubtedly they progressively improved their methods
of seizing what nature offered. The first great forward step-that from
food gathering to food producing-had to await the arrival of a more
ambitious, more intelligent man than H. neanderthalensis and the wait
was not to be long.
Homo sapiens (Modern Man).-Where sapiens came from and when
he arrived is not definitely known. We see neanderthalensis survi ving
for perhaps a few thousand years following the last glacial period, say
until 25,000 years ago. Then he seems suddenly to have disappeared
and to have been replaced by H. sapiens. This occurred in the last stages
of the Paleolithic, which merged with the Neolithic or New Stone Age,
about 10,000 years ago. Speculations regarding the origin of early
sapiens and his relation to the modern races of man have been numerous;
actual data in the case are about negligible. Did he (i.e., H. sapiens)
come in from Asia or from Africa to blend whh or totally supplant
neanderthalensis'l Or did the latter evolve into sapiens? All we can do
at present is to ask the questions and hope the future may provide an
answer.
The "Grimaldi Race."-Near Monaco on the Mediterranean coast in
the so-called "Grimaldi caves" there were discovered in 1901 two skele-
tons, one of a woman, the other of a fourteen-year-old boy. These
skeletons are of a distinct type. Both of them were about 5 ft., 2 or
3 in. tall. The leg and arm bones and more particularly the head and
facial bones are strongly suggestive of negroid characteristics. They
have caused considerable argument in anthropological quarters, but the
bulk of opinion now inclines to the belief that they were not representative
EARLY MAN AND ANIMAL DOMES1'ICATION 45
of a distinct" Grimaldi race" but rather an early and primitive example
of a late Paleolithic stock-the Cro-Magnon.
Cro-Magnon Jv1 an.-This was the name assigned to five skeletons that
were found buried in the Cro-Magnon rock shelter in the Dordogne,
France, in 1868. Other finds of this type have been made in various
parts of Western Europe. This was probably a fairly tall race, with
long heads, a high vertical forehead, short, broad face, prominent nose
and chin. The shin and forearm were long in comparison with thigh and
upper arm. This type of man, living in the late Paleolithic or Old Stone
Age, had a well-developed culture. Certainly they made quantities of
flint and bone tools and were excellent artists and sculptors. Their ori-
FIG. l4.- Neolithic man of the Campignian Stage. Mural painting by Charles R. Knight.
(Courtesy of The American Museum of Natural History.)
gin is not yet known, and it is believed that they intermingled 'with other
stocks, and perhaps traces of them are still to be found in men living
today. What a fascinating thing it would be if you and I could trace
just one line of our pedigree back to the year, say, 13,000 B.C. just to
make the wish moderate!
Several other fossil remains of man have been discovered, viz., Wadjak
man (Dr. E. Dubois) and Solo man (W. F. F. Opperworth) from Asia;
Rhodesian man CA. S. Woodward) and Boskop man (Dr. R. Broom) from
Africa. But these, together with several other finds, it has not been
possible to place with any degree of satisfaction in the general pedigree
of mankind or his collaterals.
In all this length of time, from perhaps 1 million years ago down to
10,000 or 12,000 years ago, man traveled his difficult way without the
aid of a settled agriculture and animal husbandry. The wonder is of
course that he survived at all. He lived side by side with many other
..!(j BREEDING AND IMPROVEMENT OF FARM ANIMALS
mammals, some extant, some extinct. Probably for vast periods of time
man has used animals for food and their skins for clothing and shelter.
This Paleolithic hunting stage has persisted in some places right down
relatively to the present, e.g., in Australia, Africa, and America.
Hunting Stage.-The exact times and places of the domestication of
plants are, of course, unknown, and there has been much debate as to
whether agriculture preceded or followed the domestication of animals.
Opinion at the present leans to the following sequence: first hunting;
then a hoe or hand culture, then animal domestication; then agriculture
or an animal-power culture; and, finally, the extensive development of
herds and flocks into a pastoral culture, although herding may have
preceded agriculture in certain areas.
Certainly, we are justified in assuming the hunting stage to have
preceded all forms of agriculture or animal husbandry. Man lived at
this time on what roots, berries, and seeds he could find and on such
insects, animals, and fish as he could catch. Gradually, we can imagine,
he improved his techniques for finding sources of food and utilizing these
natural sources so that they would return a better yield. Likewise his
techniques of hunting and fishing improved. He learned to trap and·
spear fish, to make ingenious traps and pits into which animals could be
lured or driven, and to disguise himself under the skin or hide of the
animal he was hunting in order to permit a closer approach. In this
stage it seems safe to assume that the men generally provided the meat·
portion of the diet, while the women largely supplied the vegetable prod-
ucts. This has led to the assumption that women were probably the
real instigators of agriculture.
During this stage of man's development, he had to make the best of
his own particular environment. Some regions, no doubt, provided
ample game and fish and little food of a vegetable nature, while in some
other regions the reverse situation was encountered. Man being omnivo-
rous, this type of enforced adaptation would present no very great
difficulties. There were undoubtedly wide variations during this early
hunting or food-gathering stage of man's existence. It has sometimes
been assumed that these hunters were extensive roamers, but it seems
more logical to think of each group confining itself to a relatively small
area where they would not come into direct conflict with other groups.
Under these conditions, children are a distinct liability, and it seems
probable that groups increased at a very slow rate.
Early Plant Culture.-Somewhere man took his first decisive step that
was eventually destined to lead him up out of savagery. He had been
utilizing such vegetable products as nature had provided him just as
he found them. Now, however, he begins to try to control this source
EARLY MAN AND ANIMAL DOMESl'ICATION 47
of sustenance. He takes wild grasses, wheat, rye, oats, corn, and various
bulbs, tubers, and trees and tries in his crude way to control and assist
their growth by scratching and preparing the soil, removing weeds,
protecting them from animals, and, finally, probably after seeing their
more luxuriant growth on dung and refuse heaps, by actually feeding the
plants. Thus in many parts of the world and under· a great variety
of conditions our grains, tubers, and fruits were brought under man's
conscious control. Later the creation of one region found its way to a
new locality where it sometimes was added to the local accomplishments,
and sometimes it totally superseded and displaced the old. This momen-
tous step in man's history occurred after the last glacial period, probably
in the early part of the Neolithic.
Closely associated with agricultl~re was the art of storing and preserv-
ing food. Just when it developed is not proved, though man had no
doubt been observing similar activities on the part of animals for a long
time. In fact, to have survived at all it seems that man must early
have learned how to store up food to tide over periods of shortage. At
any rate, after the domestication of plants, there would be more to
preserve, and any process that causes man to take forethought is a
considerable stimulus to civilization, for it embraces the feature of man's
nature which more than any other thing differentiates him from other
animal forms. This earliest stage of plant culture (sometimes called
"hoe culture") does not imply the use of domestic animals.
The Metal Ages.-The paleolithic, or Old Stone Age, with its crude
stone and bone implements, gradually merged into the Neolithic, or
New Stone Age, with its polished stone and bone tools and implements,
about 10,000 years ago. Sometime during the Paleolithic Age man
learned to use fire, but he had no agriculture and probably no domestic
animals except perhaps the dog. Sometime during the Neolithic, agri-
culture and animal domestication came into being, and these peop1es
learned how to build shelters of wood.
The ancient stone ages were finally superseded by the various ages
of metals. Metals reached Southeastern Europe soon after 3000 B.C.
but did not penetrate to Northern and Western Europe until nearly
2000 B.C. The first metals used were copper and its alloy of tin, bronze.
This was followed about 1000 B.C. by the Iron Age. Thus the three
great ages of stone, bronze, and iron followed and blended into each
other, although their dates in different localities may have differed by a
thousand years or more, so that Bronze Age here may have been Iron Age
there. With the development of better and more durable tools, the
improvement of agricultural and animal-husbandry practices, civilization
developed new impetus.
Domestication of Animals.-Just when the domestication of anim~ls

began is unknown; it certainly was long before there was any thought
of, or means for, recording historical da_ta. Domestication very likely
began at the end of the Old Stone Age and received decided impetus
during the New Stone Age. Each domestic animal has h&d a rather
restricted and humb~rigin, yet through constant selection for one or
another character all of the multifarious forms have evolved-some
useful, some ornamental, some having both qualities, and some
neither. Selection always has been and always will be the key to animal
bre;ding.
Although the exact beginning of the domestication of animals is
unknmvn, it seems logical to infer that, with the possible exception of
the dog, it arose when man ceased to be purely a hunter and adopted a
more or less settled mode of living. This new mode of living demanded
that food be close at hand. This in turn necessitated plant and animal
culture. From that remote time to the present, man has been growing
plants and animals, selecting promising variations, and eventually estab-
lishing pure strains of plants and pure breeds of animals. Plants and
animals have both reached a stage of marked efficiency through long
periods of breeding. Superior animals were no doubt developed by each
of the ancient civilizations and suffered retrogression with the decline of
the civilization of which such development was a product.
Though its exact time, origin, and mechanism are, and perhaps always
will remain, wholly unknown, animal domestication continues to be a
lively field for discussion. It is true, also, that if we knew the pedigrees
of all our animals back to the very origin of the mammals, we might
not be any better fitted to breed good livestock today. Still we like to
try to piece the scattered details together into a theory that seems to
have some degree of plausibility. Of one thing we can probably be
certain, viz., that Neolithic man did not hatch full-blown the idea that
animals could help him in his difficult struggle for existence. This is
undoubtedly true for the reason that at that time swine did not fatten
rapidly, sheep produced no wool, cows gave only enough milk for their
calves, and the wild horse was probably quite intractable. True, animals
have been of untold benefit to man, but their potentialities have heen
revealed bit by bit over a period of thousands of years.
All our larger farm animals belong to one order, the ungulates or
hoofed animals; cattle, sheep, and goats to one smaller group, Pecora.
Why did man select just the animals which he did for domestication?
Did he try to domesticate many others and fail? True, when we list
all man's domestic animals, we have quite an array-horse, ass, cattle,
yak, water buffalo: sheep, .goat, swine, camel, llama, elephant, dog, cat.
reindeer, guinea plg, rabblt, rat, mouse, poultry, turkey, duck, goose,
guinea, silkworm, honeybee-and who knows that other conquests will
be made in the future?
For successful domestication, of course, animals had to have the
proclivity of breeding in captivity as "well as a certain degree of tract-
ability. The dog has had a varied history under domestication. It is
thought that he started out as a scavenger living on the outskirts of
'man'R settlements. From this he grew to be a pet, companion, herder,
and protector, an ally in the chase, a source of food, clothing, and tractive
power. No other animal has filled so many human n0eds as the dog.
The reindeer, on the other hand, probably owes his domesticity to his
fondness for human urine, which he _apparently considers quite a delicacy
as a source of salt. Many fowl were apparently domesticated for magic
and religious purposes and at the time laid very few eggs. Some tribes
domesticated the pigeon just because they liked its looks. In other
words, many animals have been domesticated for a variety of reasons,
none of 'which was practical. Even today some native tribes keep vast
herds of cattle merely for the pleasure of possession.
On the contrary, travel in the desert without the aid of the camel or
in the arctic without the reindeer or dog would have been difficult, if not
impossi.ble, as these animals often provide milk as well as transportation
and often feed themselves in the bargain. On the contrary, the Egyp-
tians made no use of their hogs and sheep as food, and the Chinese used
cattle for thousands of years as beasts of burden but never used their
milk.
It seems probable that there were several centers of domestication,
but where they were is not known. It is generally accepted that the
dog was domesticated, probably from the wolf \vith whom he will still
mate successfully, in the late Paleolithic or early Neolithic. There is
no safe or accepted chronological order for the domestication of the
larger farm animals. Drawings of cattle, sheep, hogs, and donkeys
occur in Egyptian writings dating back to 3000 or 4000 B.C. The horse
is mentioned in Babylonian writing about 2300 B.C. and appears in Egypt
about 1700 B.C. Camels are mentioned in Mesopotamia in 1100 B.C.
How much earlier the actual domestication may have taken place is, of
course, unknown.
Influence of Domestication on Civilization.-Many factors have
influenced the development of civilization, such as the development of
speech and writing, the use of tools, the discovery of fire, and the harness-
inr of steam and electricity. Along with these we must also place the
development of agriculture and the domestication of animals. Certain

it is that without agriculture and domestic animals man would of neces-
sity still be a nomad and hunter. The domestication of plants and
animals furnished the magic" open sesame" that beckoned man out of
his dirty, pinched cave to establish a glorious countryside and later to
build great cities. By lifting physical labor from his shoulders, it also
freed his mind for the development of science as wen as for the con-
templation of the beautiful. Domestic animals have long been man's
most dependable friends and strongest allies in his task of conquering
the world in which he lives. All the energy on this planet originates in
the sun and the majority of it is stored here in forms that man cannot
use directly, e.g., grass, straw, stalks, etc. Domestic animals are merely
machines that man has adapted to. convert these stores of otherwise
unusable energy into useful forms.
Before he domesticated plants and animals, man was a savage moti-
vated largely by selfishness, with little incentive to provide for the future.
Like little children the savage is inclined not to look beyond his immediate
interests and pleasures. He lives in the present with little thought for
the future. Coming into possession of animals changed this, for their
future wants must be foreseen and provided for if they are to be of any
real service. Flocks and herds must also have had a marked tend-
"'ncy to broaden the sympathies of the savage--in short, to lead him
out of savagery and start him on his journey toward civilization. Our
civilization is greatly indebted to domestic animals, though for the most
part we are inclined to take their contribution as a matter of course.
There is no instance of any race or tribe of man having reached a high
state of civilization without the aid of domestic animals, and the leading
and conquering nations have always been far advanced in the arts of
husbandry. Domesticated animals have nourished as well as reinforced
man's sinews, quickened his mind, and broadened his sympathies, so that
it is hardly too much to say that the development of civilization had to
wait upon the domestication of animals, even though civilization may
always have gotten its start in grasslands or centers of grain culture
which could be carried out by unaided human labor.
The Dog.-Available evidence indicates that the dog was the first
animal domesticated by man, to be followed next by the even-toed
ungulates, or artiodactyls, cattle, sheep, and swine, and, finally, by a
representative of the perissodactyls, or odd-toed ungulates, the horse.
Besides being first in time the dog still holds first place as man's com-
pamon. Cuvier, the great French naturalist, wrote:
The dog is the most complete, the most remarkable, and the most useful COll-
quest ever made by man. Every species has beeome our property; eaeh indi-
EARLY MAN AND ANIMAL DOMESTICATION .')1
vidual is altogether devoted to his master, assumes his manners, knows and
defends his goods, and remains attached to him until death; and all this proceeds
neither from want nor constraint, but solely from true gratitude and friendship.
The swiftness, the strength, and the scent of the dog have created for man a
powerful ally against other animals and were, perhaps, necessary to the establish-
ment of society. He is the only animal which has followed man through every
region of the earth.
The following portion of a poem by Walter Scott depicting the faith-

fulness of the dog was inspired by the story of a young man who lost his
life by falling from one of the precipices of .!VIt. Helvellyn. Three months
later his remains were discovered at the bottom of a ravine, and hi,;
faithful dog, almost a skeleton, still guarding them.
Darkgreen "'as the spot: mid the brown mountain heather

Where the pilgrim of nature lay stretched in decay;
Like the corpse of an outcast, abandoned to weather,
Till the mountain winds wasted the tenantless clay;
Nor yet quite deserted, though lonely extended,
For, faithful in death, his mute favorite attended,
The much loved remains of her master defended
And chased the hill~fox and the raven away.
How long didst thou think that his silence was slumber?
When the wind waved his garments how oft didst thou start?
How many long days and long weeks didst thou number
Ere he faded before thee, the friend of thy heart?
The dog, an unguiculate, i.e., having claws or nails rather than hoofs
(ungulate), belongs to the order of mammals known as the Carnivora,
meaning flesh eaters. There are two suborders of the Carnivora, the
aquatic carnivores, or Pinnepedia, such as the seals and walruses, and
the land carnivores, or Fissipedia, including the dogs, cats, bears, etc.
The Carnivora have small incisor and large canine teeth with the anterior
molars or premolars usually modified for shearing. The Fissipedia
include the following families, the Viverridae (genets), Felidae (cats),
Hyaenidae (hyenas), Mustelidae (weasels), Procyonidae (raccoons),
Ursidae (bears), Protelidae (earth wolves), and the Canidae (dogs,
wolves, etc.).
The chief genus of the family Canidae is that called Canis. In this
genus are to be found modern domesticated dogs, Canis familiaris; as
well as the wolves, C. lupus, etc.; the jackals, C. aureus, etc.; and the
foxes, C. decussatus, etc. The chief differences between these four species
of Canis lie in the shape of the eye and its pupil. The dog, the wolf,
and the jackal are interfertile, and their gestation periods are of similar
52 BREEDI NG AND IMPROVEMEN T OF FARM A N I MALS
length. The dog has five toes on the front feet and four on the hind feet.
The typical dentition in this family is:
Incisors Canines Premolars Molars
3 1 4 3
a- T 4 "3
IOrigin and Domestication of the Dog.-The ancient ancestry of the

dog is very little known. Owing to the closeness of the anatomical and
skeletal features of the dog and the wolf, it is quite generally stated
FrG . I5.-St. Bernard (Hercuveen Chum) and Chihuahuas (La Rex Doll Pon Pon and
La Rex Doll Snowflake), the results of nature's variation and man's selection. (Under-
wood and Underwood.)
that the dog is an offshoot of the wolf tribe. Possibly savage hunters
captured wolf puppies and took them home to amuse the children, and
in this way through the course of time the wolf gradually lost his ferocity
and aloofness and learned how to bark joyously at his master's approach.
The jackal of the Old World, a smaller, more yellowish wild dog and
much less ferocious than the wolf, is believed by some to be the progenitor
of the dog. This strain is also much more easily tamed than is the wolf.
gome also have suggested that the dog arose as a result of a cross between
wolves and jackals. Nothing definite, hO\vever, is known regarding the
dog's progenitors.
Various breeds of dogs existed during the Roman classical period,
and some of thes~ are preserved in sculpture. Dogs are also represented
on the Egyptian monuments as far back as 3400 B.C. Skeletal remains
of a canine animal are embedded in the Danish middens of the Neolithic,
or New Stone Age. The dog has, therefore, been in existence for at least
5000 to 8,000 years.
'tJe dog may first have served a.s a source of food for ancient men in
times of famin~This in itself would have been a powerful selective
agency,-asthe le'ast valuable would have been the first sacrificed. At
a later period, before the time of the advent of agriculture and a more
or less settled mode of existence, he no doubt became an invaluable ally
in the chase. Later still, certain stra:ins were used for draft as are the
Eskimos and other dogs of today. (}hlOUgh it_ all, the dog has continued
to grow in man's affection no doubt because of his sagacity and faith-
fulness and also becauslLillQLe than any other ani;naI the dog is both
capable anacIesirous of returning any evidence of affection. .
There are now recognized six distinct groups of dogs, viz., the grey-
hound, hound, mastiff, spaniel, terrier, and wolflike. Mason's" Dogs of
All Nations" lists 183 breeds of dogs from 39 countries.\ The variations
in the different breeds of dogs as to size, form, color, degree of harshness
and length of coat, skull and jaw shape, length of body and of leg, tem-
perament, vitality, and educability cover a startlingly wide range. All
breeds and types of the domestic dog have come, however, from not more
than two or three wild progenitors in the comparatively short space of a
few thousand years. All of them from the huge St. Bernard to the tiny
Mexican Chihuahua are the results of nature's variations plus man's
selection.If
The Horse.-The name horse is equivalent to the Anglo-Saxon hors,
which means swiftness, and it is logical to suppose that this genus was
able to survive the vicissitudes of time and enemy attack only because
of its speed and mobility.
The only surviving genus of the family Equidae, ..to which the horse
belongs, is that designated as Equus. There are at present four living
types: horses, asses, half asses, and zebras. Besides the domesticated
horses, Equus caballus, there is one wild type found in Mongolia, E.
przhevalski. The domestic ass is called E. asinus and has at least two
wild varieties, africanus and somaliensis. In addition, there are three

Asiatic half asses, E. kiang, E. onager, and E. hemionu8. Among zebras,
which are striped animals, there are three types, E. zebra, E. 5urchelli,
and E. grevyt', and in addition several varieties of E. quagga (now extinct).
All these species of horses, asses, and zebras are capable of producing
hybrids among themselves but the hybrids are, as far as is known, gen-
erally sterile.
Equus has two close relatives among the perissodactyls (uneven
number of toes), viz., the tapirs and the rhinoceroses. The tapirs are
rather heavy, thickset animals with stout limbs having four toes on the
forefoot and three toes on the hind foot. The nose and upper lip are
elongated into a flexible, mobile snout, the ears small and erect, and
the tail short. The dental formula in this genus is ~, f' ~,~. The
rhinoceroses are very heavy, thickset animals with very short, stout
limbs and three completely developed toes on each foot. The head is
large with one or two horns on the median line of the face or nose. The
ears are erect, eyes small, and the skin very thick and set off into sections
of indurated plates. The incisors and canines are generally lacking,
0043
..
glvmg a d entaIf ormu1a 0 f 0' 0' 3' 3'
The members of the genus Equus are generally taller and swifter
animals than their near relatives, the tapirs and rhinoceroses, their
greater height being due to a lengthening of the cannon bones and middle
toe. The head and tail are also longer, and the body has a much denser
hairy covering. There are wide differences in the genus Equus in size,
color, tail, ear, and face characteristics. The dentition of the genus is
3 133
3' l' 3' 3' The horse E. caballus is distinguished by a long flowing
mane and forelock, smaller head, broader feet, shorter ears, longer
limbs, heavier tail, and also by four callosities, one on each limb on the
inner surface above the "knees" of the forelimbs and below the hocks
of the hind limbs. E. przhe!!alski has a short erect mane. The asses
and zebras have narrow hoofs, large ears, chestnuts or callosities on the
front limbs only, a short mane, and short hairs on the tail. The wild
asses or half asses are intermediate with fairly long ears, a short mane,
and only the front feet narrow.
\\Origin and Domestication of the H orse.-The horse was probably the
last animal to be domesticated by man, bufhis immediate ancestry as
well as the date of his domestication is stil!,...§. matter: of dispute. It
seems probable that at least-two or perhaps three wild types have made
their contrihution to our domestic horse. One of these was no doubt
~... .. .... . .
_;:"._- ~
........:-f~:.:~··:l?· :...: ~~'- .~.'t" :"

-of
~, , -;.-
~
_'
~. I
1,
':,
EARLY MAN A.ND AN1MAL DOMESTJCATlON 55
the so-called "steppe horse" now known as the fossil representative of

Przbeva~~ This was a small, sturdy, short-legged horse with
a. lllod~y long, heavy head. Another was the so-called "desert
horse," standing as did the steppe horse about 13 hands high and cor-
responding closely to the now extinct tarpan or Mongolian horse. This
strain was somewhat more slender than the steppe horse and had a
shorter head. The third contribution was that of the "forest horse,"
a type standing about 15 hands high with longer but stout limbs and
having a long, narrow head and long body. It seems probable that all
three types made their contribution and that by crossing and selection
lPm. 16.-The "Ta.rpan," a. Mongolian wild hor 'c from which the horses of Western Europe
ave been derived. (F,.om Thompson, Outline of Science, Geo. N ewnes, Ltd. )
the foundations of our modern breeds were laid. The horse was appar-
ently domesticated separately in both Asia and Europe, probably earlier
• Asia. A Turanian folk tamed the Przhevalski horse around 3000 B.C.
The earliest record of the horse dates back to Paleolithic times about
l"rAU~.V\.'" years ago. Around an open camp at Solutre in France are found
remains of several thousand horses, indicating that they may have
as a source of food. In later Paleolithic times, rock carvings of
horse were made but do not show him harnessed, ridden, or attached
any sort of vehicle, so that we assume the horse was not yet fully
-cIomesticated.
The earliest trace of the horse hitched to a chariot goes back to about
iOOO B.C. in Greece, whereas the first Egyptian records of the horse
,
date from about 1600 B.C. These were small horses about 13 hands
high, similar to Przhevalski's horse. The horse ev.idently grew in size
and importance in Persia and Mesopotamia during the intervening years
and after about 750 B.C. began to serve as a mount. Mounted horsemen
were first given a place in the Olympic games in 648 B.C. The Arabs
did not use horses until after the time of Christ.
Horse Ass
]Gang Zebra
l<' IG. 17.-The horse and some of his relatives. (Courtesy New York Zoolorrical Society.)
The development of the Arabian horse by the Bedouins of the desert

is generally recognized as the first great achievement in animal breeding
of which there is any definite record. Because the Bedouin's safety
often hinged upon the speed and endurance of their horses, these people
surrounded their breeding operations with an air of superstition and
mystery, much of which has persisted in certain quarters to the present
day. The Arabian breed probably was used later to re:fine the large
"cold-blooded" horses of the countries to the west and north, particu-
larly of France and England. The Arabs, as well as Barb and Turk
horses, were also used to cross with the racehorse of England, thus laying
the foundation for the Thorou~hbred.
For many centuries ~__or.se has been man's foremost helpmate in

work, war, and play. He rendered incalculable assistance in conquering
the New W orltr,"America. From 4,336,000 horses in 1840, the year of
the first agricultural census in the United States, his numbers steadily
increased to 21,500,000 in 1918. Since then he has been replaced to a
considerable degree by machines, automobiles, trucks, and tractors, so
that in 1950 his numbers had dwindled to about 6,000,000. This free-
ing of land from the production of feed for horses has complicated the
situation of agriculture tremendously, and there seems now to be a definite
increase in the number of horses being used for pleasure. What the
future holds in regard to mechanization no one can prophesy, but it can
be said with little fear of successful contradiction that our present civiliza-
tion rests quite largely on the sturdy legs and broad backs of our four-
footed friends and not the-least among these is E. caballusr)
Cattle.-The Pecora, or true ruminants, is made up of the following
families: the Cervidae (deer), Qjraffidae (giraffes), Bovidae, to which
all types of cattle belong, as wen as sheep, goats, and true antelopes.
Both sexes in this family are usually horned. The horns are hollow and
grown on bony cores and arise from the frontal bones. The second and
fifth digits are rudimentary or absent, the third and fourth fully devel-
oped. The dental formula for this family is ~, ¥, ~,~. These animals
are ungulates or hoofed and belong to the group known as artiodactyls,
or even-toed. There are one or more enlargements for food storage
along the esophagus, usually three, and the members of this family
ruminate or chew their cuds.
The genus Bos includes the taurine group, Bos taurus (cattle); and
B. indicus (humped cattle); the bibovine group, B. gaurus (the gaur);
B. frontalis (the gayal); and B. sondaicus (the banteng); the bisontine
group, B. grunniens (the yak); B. bonasus (the European bison); and
B. bison (The American bison); the bubaline group, B. caffer (the African
buffalo); B. bubalis (the Indian buffalo); B. mindorensis (the Mindoro
buffalo); and B. depressicornis (the Celebes buffalo). There are many
varieties under most of these species, both living and extinct. As far
as known, all these species are generally interfertil.e and produce fertile
hybrids, though the males are often sterile.
These close relatives of our modern domesticated breeds of cattle
form a very interesting and widespread group of mammals. Many of
them have become domesticated. Of the Bibovine group, the gaur (B.
gaurus) is the largest of the present-day representatives and is found
in India, Burma, and all the Malayan regions. It has somewhat flattened
horns, a short tail, white lower extremities, or "stockings," fine silky
58 BREEDI NG AND I M PRO VE M E N T OF FARM ANI MALS
hair, and long spines on the anterior thoracic vertebrae forming a high
set of withers, the bull standing about 16 hands high. It is a somewhat
shy, forest- and hill-loving animal. The gayal is somewhat smaller
with short horns and" stockinged" feet and is easily domesticated. The
banteng is the smallest representative of this group, has a white disk
from the hocks to the tail setting, and is somewhat longer legged. The
banteng frequents the lowlands as well as the hills.
Gayal African Cape Buffalo

American Bison Ayrshire Bull
FIG. l8.-Some relatives in the genus Bos, (Cow'tesy of New York Zoological Society.)
The bison tine group includes the yak of Tibet as well as the European
and American bison, the latter often erroneously called a buffalo. The
yak inhabits the wild, cold regions of Tibet and Siberia, and it is said that
travel there without his help as a pack animal would be impossible.
The yak has a skirt of long hair hanging from his shoulders, flank, thighs,
and tail. He is smaller than the gaur, standing about 14 hands high,
and is not" stockinged" with white. The bison is characterized by very
heavy forequarters and light rear quarters. His head, neck, chest, and
shoulders are covered with moderately long, dark hair and like the yak
he has 14 or 15 pairs of ribs. Great herds of bison once ranged over
most of Europe and a goodly portion of North America.
The bubaline group, or true buffalo, includes four species at present.
The African buffalo is noted for his very heavy down curving horns,
which almost meet in the mid-line of his face, very unlike the Indian
buffalo with his wide, out- and upcurving horns. The Indian buffalo
is also larger, standing 15 to 16 hands high. Both the Philippine and
Celebes buffaloes are much smaller in stature, and in the latter the
horns grow straight back and upward from the forehead resembling the
horns of an antelope. The buffalo is generally found in low, swampy
regions.
'()..Origin and Domestication of Cattle.i-It seems probable that cattle
were domesticated during the New Stone Age in Europe and Asia from
more than on~. There are two types of domestic cattle now
living, B. in'"iHCUs, the humped cattle of tropical countries, and B. taurus,
of the more temperate zones. Humped cattle were domesticaQ as
early as 2100 B.C. Cattle play an important part in Greek mythology,
were sacred animals in many older civilizations, and their slaughter
was therefore forbidden. The great ox or aurochs, B. primigenius,
which Caesar mentioned in his writings, is generally considered to be
one of the progenitors of our modern-day breeds. This was a very large
animal and described by Caesar as "approaching the elephant in size
but presenting the figure of a bull." The wild park cattle of Britain are
considered by some to be the direct descendants of B. primigenius.
Another progenitor of our modern breeds is B. longifrons, a smaller type,
with somewhat dished face. This is the Celtic Shorthorn which has
never been found except in a state of domestication ~nd was the only
ox in the British Isles until 500, when the Anglo-Saxons came bringing
with them the great ox or aurochs of Europe.
It is doubtful whether any of our present-day European or American
breeds trace solely to either one of these ancient types. It seems much
more probable that our present breeds are the results of various degrees
of crossing between them. The cattle of India and Africa, B. indicus,
are characterized by a lump of fleshy tissue over the withers sometimes
weighing as much as 40 or 50 lb. They also have a very large dewlap
and the voice is more of a grunt than a low. They are thought to be
descended from the wild Malayan banteng.
It seems prQbable that savage man first used members of the family
Bovidae as a source of food. Real domestication began when these
animals were used as draft animals, probably in the 1rst faltering steps
of the beginning of agriculture or the ti~e of the soil. In their \vild
state there was little tendency to store excess fat on the body, as this
would have been a hindrance rather than a help under the conditions
then existing. Milking qualities also were just sufficient for the rearing
of the~. As civilization developed, feed became more abundant,
methads af caring far livestack impraved, and the latent passibilities

far rapid grawth, fat storage, and milk praductian began to, be realized
under man's selection.; ~
That the ax played an impart ant part in man's aesthetic develapment
is attested by its use in architecture and interiar mural decoration as
well as its frequent use as a subject af paetic fancy. The ax assumed
great religiaus importance in many ancient civilizatians, and the best
members of the breed were sacrificed to, prapitiate the gods. They
were crawned with wreaths and honored in other ways during the
pageants and halidays, and to same extent we perpetuate this custam
in aur fairs and expositions today. The Romans used the term pecunia
for maney, a ward derived fram pecus meaning cattle, and in ancient
times a man's wealth was figured in t.erms af his cattle passessions.
Down through the ages variaus members of the genus Bas have can-
tinued to be one of man's most useful allies. Withaut their cantributian
of pawer, meat, milk, and inspiration it is difficult to, visualize the passi-
bility of an advancing civilization. The fallawing is G. L. Bickersteth's
translation of Carducci's paem, "The Ox": ' .-;:::::
I love thee, holy ox; a soothing sense
Of power and peace thou lodgest in my heart.
How solemn, like a monument, thou art,
Watching the pastures fertile and immense!
Or 'neath the yoke with calmness how intense

Dost thou to man's quick toil thy aid impart!
He shouts and goads thee; patient of the smart,
Thine eyes, slow turning, claim more reverence.
From thy broad nostrils, black and moist, doth rise

Thy breath in fragrant incense; like a psalm
Swells on the air thy lowings' joyful strain.
Austerely sweet are thy grave emerald eyes,
And in their depths is mirrored, wide and calm,
All the divine green silence of the plain.
Sheep and Goats.-These twa genera of the family Bovidae are very
closely related, so closely in fact that a naturalist never speaks lightly of
separating the sheep from the goats. Modern breeds are easily enough
distinguishable, but the task is sometimes very difficult in their wild
relatives. The genus Ovis includes the sheep and its wild relatives,
whereas gaats ana their kind make up the genus Capra. Sheep are
to be distinguished fram gaats by glands in bath fareTeet'and hind feet,
-
by the absence of a true beard, and by the absence af the strong gaaty
odor in males. There are also marked differences in the skulls, and
. the horns -g;nerally spiral in opp~ directions, the right horn of the
sheep to the right like a corkscrew, and the goats to the left. The sheep
gets its Latin name Ovis from the Sanskirt a~ying to keep or to
guard.
Our domestic sheep have a great variety of wild relatives. Lydekker
divides these into four groups: (1) the bighorn (Ovis canadensis) of North
America and Kamchatka, (2) the argali (0. poli) of Central Asia, (3) the
urial (0. vignei) of Asia and the mouflon (0. musimon) of Asia Minor
and Europe, and (4) the bharal (0. nahura) of Little Tibet and the
Barbary (0. tragelaphus) of North Africa.
The bighorns are a large breed of wild sheep standing 38 to 42 in.
at the withers. As the name indicates, the horns are very large and are
noted for the small size of the transverse ridges and the prominence of
the outer anterior angle of the horn. The horns curve back, out, down,
and forward in a bold sweep. The upper part of the body is covered
with dark hair, the lower portion with a gray or dirty white hair. They
have a very noticeable caudal disk of light-colored hair extending in a
radius of about 87-2 in. from the root of the short tail. They make a
square track, for the interior and exterior edges of the toes are nearly
parallel. The front of the toe is also beveled, which accounts for the
remarkable climbing power of this group of mountain-dwelling wild
sheep. The group has a very shallow lachrymal pit in the skull.
The argali, or Marco Polo's sheep, is a large species standing 40 to
48 in. high at the withers. They have strongly marked transverse ridges
on the horns and a less pronounced outer anterior angle. The horns
are very long (circle and a quarter) and make a wide sweep. The
color of this group is very similar to that of the bighorn, including the
caudal disk. These however, have a short mane on the throat. These
wild sheep inhabit the grassy slopes of The Pamirs and never attain to
very high altitudes. There are several closely related varieties in this
group differing somewhat in size, color and shape of horns, habitat, etc.
The urials are a smaller race of sheep generally found in high altitudes.
They stand about 3 ft. high and have shorter horns than the bighorns
or the argalis. The general color is brownish gray with a white belly.
The throat has a short beard or mane. There are several differing
varieties in this group also, including the mouflon of Sardinia and Corsica.
This is a small species of brownish-gray color with lighter colored under-
line and legs.
The bharal in many ways stands intermediate between sheep and
goats. Its horns lack the deep transverse ridges of the sheep and are a
dark olive-green color like the goats. It has a longer, therefore more
goatlike tail, and it lacks the lachrymal gland on the face. On the other
hand, it lacks the strong odor of goats, also lacks a beard, and like the
sheep has glands between the toes of all four feet. The bharal stands
about 30 to 36 in. high and is a dark slaty-blue color. There is no long
hair on the throat or tail. The skull characteristics of the bharal are
goatlike rather than sheeplike.
The wild relatives of the goat include the ture, the pasang, generally
regarded as the true progenitor of all our modern domesticated breeds
Rocky Mountain Bighorn Moufion

Punjab Soutbdown
FIG. 19. -The sheep and some of its cousins. (Courtesy of New York Zoological Society.)
of goats, the ibex, the markhor, and the tahr. Here again we find a
very wide variation in superficial characteristics among the mountain-
dwelling members of the genus Capra, a detailed description of which
cannot be entered into here.
/( Origin and Domestication of Sheep.-The sheep probably originated
in Europe and the cooler regions of Asia not further back than the
Pleistocene or later Pliocene. Remains of a sheep or goatlike animal
have been found at the sites of the Swiss lake dwellings oCNeolithic
times. Sheep ~ to have been derived from the antelopelike
animals allie~azelles because of certain similarities of the molar
teeth. It seems certain that our modern breeds trace back to at least
.two remote ancestors, the mouflon of Europe (0. musimon) and the
Asiatic urial (0. vignei).
The sheep, most important of wool-bearing animals, was originally,
as in the case of some breeds still, a 1mtfY"animal with an underfur of
wool. No doubt people living in cold climates who used skins as clothing
were the first to begin the selection of sheep for wool production. As in
all our domesticated animals, there is wide variation among sheep.
Some, like the African long-legged and Abyssinian maned sheep, bear hair
instead of wool, some have spiral horns 2 ft. or more in length, others
Fro. 20.-The Moufion-a wild sheep lacking uniform color and having an exterior coat
of hair. (From Coffey, Productive Sheep Husbandry, J. B . Lippincott Company.)
110 horns at all. The tail of the common, domesticated sheep is long
and slender; in some other strains, it is loaded \yith fat and is 1 ft. in
width, whereas still others have merely a vestige of a tail. The last
IIOrt often carry huge patches of fat on the rear quarters, the stored
fat in all cases serving to tide the animal over periods of food shortage.
The hunia, a tall, long-legged sheep, is used in India as a fighting animal,
f,leing pitted against its fellows or other animals.
Goats are also versatile in characteristics, yielding the underfur, for
Cashmere shawls, and mohair, milk, meat, and draft power and also
. ~ m~of clearing up brush land because of their fondness
all sorts Grtender shoots. Ii .
Swine.-8wlne are ungulates belonging to the suborder artiodactyls
They belong to the family Suidae. The Dicotylidae, or
peccaries, and the Hippopotamidae, or hippopotamuses, are closely

related families, these three families comprising the Suina. These
animals have tubercles on the molar teeth, and in them there is not found
a complete fusion of the third and fourth metapodials to form a "cannon
bone." The nose is elongated into a more or less mobile snout. The
typical dental formula for this group is ~, i, ~, i, though this is not com-
plete in all forms.
The genus Sus includes, besides the domesticated pig, several wild
species. Among them are Sus scrofa, the European wild boar; S. crista-
tus, the Indian wild boar; S. andamanensis of the Andaman Islands; S.
salvanius, from the foot of the Himalaya; S. vittatus, S. verrucosus,
and S. barbatus of the Malayan region; S. africanus and S. procus of
Africa. The foregoing species vary greatly in size, color, length of coat,
length of tail, presence or absence of crest on the neck, size of tusks,
number of mammae, etc. The Indian wild boar stands 30 to 40 in.
high, whereas S. salvanius measures only 11 in. over the shoulder. The
tusks of the wild boar are several inches long, those of the red river hog,
S. africanus, only as long as those of the domestic boar, and those of
S. salvanius are very short. The young of all the above-mentioned
species are born with longitudinal dark strips on the body which dis-
appear as the animal matures.
Other wild relatives of the pig include the genus Babirussa of Celebes
and Burma as well as the genus Phacochaerus. In the former genus the
upper tusks of the male do not have their summits abraded by wearing
against the sides of the lower ones as in the genus Sus, but instead the
upper tusks arise close to each side of the face and pass right out through
the top of the nose, never entering the mouth cavity, thence sweeping
backward and upward, frequently touching the forehead, after which
they curve forward, thus resembling horns rather than teeth. These
tusks differ from those of the wild boar in not having a covering of
enamel. They often measure 14 in. or more in length. These animals
are rather "leggy" and almost devoid of hair.
The genus Phacochaerus, or wart hogs, of Africa have a head very
large in proportion to the body with eyes placed far back on the face.
They derive their name from the wartlike cutaneous lobes that project
from each side of the face. Their upper canines are even more greatly
developed than those of the Babirussa, curving upward and inward and
being large disproportionately to the size of the skull. In older animals
all the teeth except the four tusks and the four last molars disappear.
The molars, however, measure more than 2 in. in length from front to
rear, the lessened number of teeth being compensated by the over-
development of those remaining, which assures ample grinding surface.
This development also occurs in the case of the adult elephant, except
that in this species the lower tusks are also wanting. "Vild pigs are
found only in the Old World, their place in America being taken by the
peccaries, a distinct family.
Wild Boar and Sow

Siamese Sow and Pigs
FIG. 21.-Probable progenitors of modern swine. (Courtesy of Lon(Jmans, Green & Co.)
'" Origin and Domestication of Swine.~t appears that our modern breeds,
'so domesticus, have descended from at least two wild stocks; the N orther~
European breeds from the wild boar S. scrofa, and those 'Of Southern
Europe, Asia, and Africa from one of the Malayan pigs, possibly the
collared pig S. vittatus. The former was a larger, coarser animal through-
out than the latter and had a denser covering of hair. .
Present-day breeds are · no doubt the result of varying degrees qf
crossing between the parent stocks and their offspring. It seems probable
that the pig was domesticated later than cattle and sheep and previous
to the horse. Selected for his ability to fatten rapidly and economically,
the pig is foremost in converting feed into flesh and second to none in the
economy oIAffi~ meat production. Most of the breeds of hogs
found in America are of our own breeders' making; e.g., the Duroc-
Jersey, the Poland China, and the Chester White breeds are strictly
American creations.
In addition our breeders have developed the American Saddle Rorse,
the Standardbred Horse, the Morgan Horse, a~ few breeds of sheep.
For the most part, however, the-American breeder's task has been simply
to i~~rove f9reign breeds and to render them more suitable to ou~
condItIOns. ;I
Summary.-We have tried in this chapter to outline the events leading
up to the appearance of Homo sapiens on this planet. He came upon the
scene by slow and probably painful stages. Finally, some 10,000 to
25,000 years ago, he emerged, and his progress has been quickening ever
since. We surveyed his early hunting, plant-culture, and, finally, animal-
domestication phases and attempted to trace the origins of his four-
footed servants. Each successive stage encouraged and made possible
a larger human family. By the year 1800, man numbered 750 million.
Then came the agricultural and industrial revolutions. Food became
relatively plentiful, and man's procreative powers responded so well that
during the next 140 years population on the earth trebled, reaching a
total of over 2 billion. The world has been filling up, perhaps too fast.
There are no longer frontiers, but in some parts of the world we have
caught a glimpse of the abundance that should be the reward of every
man of intelligence and energy. Science has now placed in our hands
the tools for producing abundance. We have yet to decide what to do
with them.
References
ALLEN, R. L. 1847. "Domestic Animals," Orange Judd Publishing Co., Inc., New
York.
Anonymous. 1928. "The Indispensable Sheep," Nall. Geog. Mag., 53:512-528.
BREASTED, J. H. 1935. "Ancient Times," Ginn & Company, Boston.
CARTER, WILLIAM H. 1923. "The Horses of the World," The National Geographic
Society, Washington, D.C.
CLARK, W. E. L. G. 1934. "Early Forerunners of Man," William \Vood & Com-
pany, Baltimore.
DAVENPORT, E. 1910. "Domesticated Animals and Plants," Ginn & Company,
Boston.
EDITORS, 1949. "Dogs of America," Life 26:5.
FINLAY, G. F. 1925. Proceedings of the Scottish Cattle Breeding Conference, Oliver
& Boyd, Ltd., Edinburgh and London.
HOOTON, E. A. 1931. "Up from the Ape," The Macmillan Company, New York.
KEITH, A. 1938. "New Discoveries Relating to the Antiquity of Man," J. B.
Lippincott Company, Philadelphia.
LOOMIS, F. B. 1926. "The Evolution of the Horse," Marshall Jones Co., Boston.
LULL, R. S. 1928. "Ancient Man," Doubleday & Company, Inc., New York.
LYDEKKER, R. 1912. "The Horse and Its Relatives," George Allen & Unwin, Ltd.,
London.
1912. "The Sheep and Its Cousins," E. P. Dutton & Co., Inc., New
York.
1912. "The Ox and Its Kindred," Methuen & Co., Ltd., London.
MILLER, C. J. 1903. "Dogs of All Nations," Illustrated Editions Co., Inc., New
York.
MORGAN, J. J. M. 1925. "Prehistoric Man," Alfred A. Knopf, Inc., New York.
MORSE, E. W. 1910. The Ancestry of Domesticated Cattle, U.S. Dept. Agr. Rpt.
Bur. Anim. Indus. 27.
RIGGS, E. S. 1932. "The Geological History and Evolution of the Horse," Field
Museum of :Natural History, Chicago.
SANDERS, A. H. 1925. "The Taurine \Vorld," The National Geographic Society,
Washington, D.C.
SEMPLE, A. T. 1931. "The Origin of Domestic Cattle," Natural History, 31 :287-
299.
SHALER, N. S. 1895. "Domesticated Animals," Charles Scribner's Sons, New York.
WILDER, H. H. 1926. "The Pedigree of the Human Race," Henry Holt and
Company, Inc., New York.
ZIRKLE, C. 1935. "The Beginnings of Plant Hybridization," University of Pennsyl-
vania Press, Philadelphia.
CHAPTER III
ANIMAL ORIGINS AND PROGRESSION
Animals are organisms that are capable of sensation and voluntary

movement. Some are one-celled, others many-celled; some move in the
air, some on land, some under "oater; some have their supporting struc-
tures, if any, on the outside like an overcoat and are called invertebrates;
some are supported by a bony or cartilaginous inner framework and are
called vertebrates, species of the former far outnumbering those of the
latter. Some, like man, have a very highly developed brain and sensory
system for recording environmental stimuli; others, like the clam, ar~
relatively immune to environmental changes. All must ingest food
materials and oxygen and rid themselves of waste products in order to
live. All are endowed with the power and "desire" to perpetuate their
kind through some form of reproduction. Although they show the
widest sort of diversity, there seems to be a thread of unity binding all
living creatures into one organic whole. They all come into being,
grow to maturity, reproduce their like, and, finally, die in very much
the same fashion.
Early Ideas on the Genesis of Living Forms.-Man has been pondering
the question of the whence, why, and whither of himself and all other
animate nature for many thousands of years. These questions probably
did not concern him in his prehuman state, any more than they con-
cerned you and me when we were infants. With the dawn of reason and
beginning of the capacity for reflective thinking, however, man began
to wonder about the origin of and relationship among living forms.
Until the advent of science with its inductive method of reasoning, he
could but guess and speculate on these questions.
Science is nothing but the discovery and arrangement of separate
facts into logical patterns of relationships or of cause and effect, the
ascertaining of general laws from separate facts. Millions of facts,
classified into many categories, indicate relationship between all living
forms. For a full appreciation of the problems of breeding and inherit-
ance, an understanding of the general scope of animal relationships
furnishes the best medium.
Centuries ago it was obvious that the sun revolved around the earth, .....
men saw it happen every day. It was just as obvious that the earth
68
ANIMAL ORIGINS AND PROGRESSION 69
was flat. In this same category of the obvious was the production of
living matter from lifeless materials; maggots and flies from decaying
meat, worms from manure, etc. This idea of spontaneous generation of
living organisms from dead material is, therefore, of ancient origin and
is probably still extant in some regions.
Most of the early Greek philosophers felt the need of offering some
explanation of the genesis of living forms. Thales (seventh and sixth
century B.C.) was the originator of the doctrine of abiogenesis, or the
production of living forms from lifeless matter. He taught that life
develops from amorphous slime under the influence of heat. Anaxi-
mander (611-547 B.C.) taught that the earth existed first in a fluid state,
which later solidified in spots to form dry land, and that living creatures
formed in the ,vater came out onto the land in a gradual fashion and
acquired the appendages and other bodily mechanisms which would
make terrestrial life possible.
Anaxagoras (510-428 B.C.) was the first to suggest the idea that
gradual development was not the result of the workings of blind force
but was directed by some "Intelligence," a theory known today as
teleology. His idea was that the germs of all animate and inanimate
things existed first in chaos, but that gradually soil was laid down,
covered with air, and this in turn by ether. The germs of plants floating
in the air finally settled on the earth and covered the earth ,,,ith vegeta-
tion. Thereupon the germs of animals and men, which he supposed
were floating in the ether, settled onto the earth, grew, and prospered.
All of this, according to Anaxagoras, followed an intelligent plan created
by some designer; i.e., the process was not inherent in the materials but
needed an outside directing force.
Empedocles (500?-430? B.C.) had the notion that parts of organisms,
heads, arms, legs, etc., existing first in separate states, were finally brought
together into ,,,hole organisms through the triumph of love over hate,
and that those which had made a lucky combination of suitably adjusted
parts were able to survive and perpetuate themselves in their offspring,
leaving others less fortunate in their acquisitions, to extinction. Here
\ve meet for the first time the general idea of the survival of the fittest.
Epicurus (341-270 B.C.) and his later Roman apostle Lucretius (99-
55 B.C.) taught that plants and animals arose on the earth from the
action of the sun's heat on the moist earth, or, as Lucretius phrased it in
his materialistic poem "On the Nature of Things," " Nothing from
nothing ever yet was born." Lucretius anticipated Dalton and modern
chemistry in his reference to the atomic nature of matter as well aH
Mendel and modern genetics in his reference to hereditary bodies within
the mother.
70 BREEDING AND IMPROVEMENT OF FAR},! ANIMALS
Aristotle (384-322 B.C.), one of the most renowned of the Greek

philosophers, believed, as did Anaxagoras before him, that some intelli-
gence, entelechy, or "soul of living things" directed the process of the
active generation of living forms from passive matter. This soul property
was thought to be present in increasing degrees in earth, water, air,
and fire, so that the earth produced plants, water produced aquatic
animals, air gave rise to terrestrial organisms, and fire the supposed
inhabitants of celestial bodies. Aristotle maintained that plants and
many animals, e.g., worms, insects, crabs, and other marine forms and
even man, might arise spontaneously, all the former from the dew,
slime, manure, sweat, meat, decaying trees and fruit or seawood, or
from the moist earth, whereas man's first appearance was thought to
have been in the form of a worm. Aristotle's teachings dominated
men's thinking down to the time of science, and the acceptance of his
ideas by St. Augustine (A.D. 354-430) established spontaneous generation
as a dogma of the Church. St. Augustine and others of the early Church
fathers, however, were not literalists in regard to the Mosaic account of
creation but interpreted the 6 days of creation to be periods of time much
in excess of 24 hours.
Aristotle's works were translated by the Arabs in the ninth century
and found their way into Spain a century later. The revival of these
pagan Greek ideas caused considerable friction in the ·Western European
Christian world of the b,oelfth and thirteenth centuries, but a synthesis
of Greek scientific and Christian theological ideas was finally achieved
by the great churchman St. Thomas Aquinas (1215-1274).
Unlike Aristotle, who wanted to understand and eventually control
nature for beneficial human ends, the churchmen of the ::\1iddle Ages
started with some acceptable belief. Their ideal was the contemplation
of truth arrived at through intuition and faith. They believed that they
knew the purpose and meaning of existence and that the function of
nature and of man was to glorify God. The Scholasticism of St. Thomas
was a synthesis of Greek reason and Christian faith on the theory that
reason and faith are not antithetical but supplementary, and that faith
can go beyond reason and solve problems for which reason is inadequate.
Intuition was thus judged to be superior to reason, and for 600 years this
idea, togethBr with the authoritarianism founded on Aristotle's works,
prevented scientific inquiry into the nature of the universe and man. All
inquiry into the nature of things thus became heresy, and for nearly
600 years anyone who deviated from this pattern was forced to recant.
But the flame of curiosity was not entirely extinguished, and the noted
Italian Leonardo da Vinci of the fifteenth century, scientist as well as
painter, pioneered in the field of paleontology, one of the pillars on which
the theory of the gradual diversification of living forms now rests. In

the sixteenth century, Giordano Bruno (1548-1600) espoused the Greek
idea of gradual development and left among other maxims, "The investi-
gation of Nature in the unbiased light of reason is our only guide to
truth." He was finally burned at the stake for his refusal to recant hi8
heretical beliefs.
From the liberal ideas of the early Church fathers, the pendulum
swung, by the middle of the sixteenth century, clear over to the extreme
literal position. Separate creation and the general over-all authority of
the Bible both as spiritual guide and scientific textbook were firmly
entrenched, and all doubters or too persistent seekers of nature's truths
by means of reason or experiment were labeled heretics and forced to
recant or be liquidated.
In the last half of the sixteenth century, and still more in the seven-
teenth century, curio8ity and experimentation began to make some head-
way, but spontaneous generation held its place in the minds of men like
Harvey (1578-1657), Descartes (1596-1650), Newton (1643-1727),
Buffon (1707-1788), and even Lamarck (1744-1829).
Redi (1626-1697) was the first to show t,hat decaying meat or fish
could not produce maggots if flies were prevented from laying their eggs
thereon. Leeuwenhoek (lG32-1723), the Dutch naturalist, revealed,
through his invention of the microscope, a whole new world of tiny living
forms, \,"hich tended to strengthen the belief in spontaneous generation
although he himself considered the forms of organisms infesting rotting
meat, etc., to be air-borne. Spallanzani (1729-1799), the Italian scien-
tist, Schwann (1810-1881), the German anatomist, and many others
devised experiments to try to disprove spontaneous generation in mate-
rials uncontaminated by organisms and were generally, though not
ahmys, successful. It remained for Louis Pasteur (1822-1895), the great
French chemist, to provide, by means of his bent-necked flasks, the final
obituary to this age-old idea of spontaneous generation of living organisms.
There are three possible ways for accounting for life on the planet
Earth. Spontaneous generation in its ancient sense seems to have been
finally ruled out. To think that life was wafted to this planet from some
other heavenly body only moves the problem back one step where it is
entirely insoluble, and modern astronomical conceptions lend no support
to the plausibility of the idea. Still a third possibility is to say that life
has existed and will exist eternally. This again is impossible of proof and
not very satisfactory to a practical mind.
Modem Ideas of the Beginning of Life.-From Pasteur's work came
the final proof that fully formed living organisms cannot be formed in an
instant and de novo in any sort of a culture. By some this was interpreted
to mean that there could never be or have been a transition from non-
living to living material. We know, however, that dead bodies of animals
and plants are finally broken down into inorganic compounds or into the
elements. The question is whether the reverse process could ever have
taken place. It is now coming to be generally agreed that it not only
could but did so happen, not that we now have final proof or that the
biochemist can now make life.
The most probable general steps in the process seem to be about as
follows: (1) the pulling out of material from our sun by the near passage
of another star some 1 or 2 billion years ago and the centralization of this
material to form our planetary system, including the earth with its central
molten metal core covered by a layer of igneous rocks and enveloped in
an atmosphere of aqueous vapor; (2) formation of hydrocarbons by the
interaction of carbides of iron and other metals and water; (3) formation
of ammonia by the interaction of metal nitrides and water; (4) production
of numerous high-molecular compounds through condensation, polymeri-
zation, and oxidation-reduction reactions of the derivatives (alcohols,
aldehydes, ketones, organic acids, etc.) of the primitive hydrocarbons and
nitrogen-bearing compounds to form simple compounds of a protein
nature; (5) formation of a primary colloidal system; (6) natural selection
of those colloidal systems with a highly developed physicochemical
organization surrounded by a semipermeable membrane capable of repro~
ducing themselves and also capable of continued growth and variation.
This is admittedly a very sketchy picture. Detailed knowledge of the
internal structure of the simplest living forms is not now available.
Nature has been made to yield many of her secrets, and it seems not
improbable that the true nature of living matter will sometime be dis··
covered. Granted the formation of even the simplest living and repro-
ducing forms together with the principle of variation, an orderly progres-
sion of living forms from the simplest to the most complex becomes
thinkable.
One-celled to Simpler Living Forms.-Students as ,veIl as laymen are
sometimes prone to look upon single-celled organisms as the smallest
and simplest manifestation of life, but this is an erroneous view. The
single-celled plant or animal is actually very complex and endowed with
the capabilities of nourishing itself, of getting tid of its waste products,
and of reproducing. The planet Earth is thought to have existed for
about 2 billion years, and it seems probable that, after life arose from
inorganic materials, it took several hundred million years to work up to
the level of single-celled plants or animals.
Let us then start with single-celled forms but reverse the usual process
and try to work back to the original form of life, though we should be
,
\
wary and not expect to get a necessarily definite answer at the present.
The single-celled organism, e.g., amoeba, yeast, etc., as already indicated,
is a complex structure and is organized in a definite way. It consists of
a cell wall surrounding cytoplasm in which are located chloroplasts,
chromoplasts, leucoplasts, chrondriosomes, centrosome, fat droplets,
mineral deposits, yolk granules, and a nucleus, the latter often containing
dark-staining bodies called chromosomes. It has now been established
that the chromosomes are the bearers of hereditary potentialities in the
higher plants and animals and, therefore, form the bridge of inheritance
bet,,'een succeeding generations. The chromosomes are aggregates of
genes arranged in linear series, whereas the genes themselves are probably
complex organic molecules.
As we start down the scale in size, We come first to the bacteria. There
are many thousands of these forms and in general they are rodlike, round,
or spiral-shaped. We are inclined to think of them as enemies to man as
those which cause typhoid, pneumonia, and other diseases surely are, but
there are thousands of them that are harmless to man and many more
which are definitely beneficial, e.g., those soil bacteria which fix nitrogen
from the air, those which convert alcohol to acetic acid, or milk curd to
cheese.
Ordinary bacteria range in size from about 500 to 750 m,u. Even the
highest powers of the microscope cannot at present disclose much con-
cerning their internal structure, although we have a right to believe that
they do have a very definite and specific structure. They produce
variations that are definitely heritable, and, although we are unable to
see a nucleus and its chromosomes within them, it seems a fair assumption
that they do actually have some such mechanisms. Stoughton (1929)
and Barnard (1930) have shown that at the time of fission the scattered
chromidia of certain bacteria aggregate into an ill-defined mass within
the organism, which then divides to produce two cells. Here apparently
the genetic or hereditary materials are scattered, \vhereas in the one-called
organisms, discussed above, they have become definitely and permanently
organized into chromosomes.
It is known that living entities of a diameter below 300 or 400 m,u
exist, although they cannot be seen even with the most powerful micro-
scope we now have. These materials are called filtrable viruses or filter
passers and are the cause of certain diseases such as smallpox, yellow
fever, rabies in man, and foot and mouth disease in cattle and swine;
but, as with the bacteria, there may also be beneficial ones. Some of
them have been photographed in recent years by means of the shorter
wave ultraviolet light and more recently with the electron microscope.
Although they cannot be seen, they undoubtedly are living materials, for
they can produce their like on suitable culture media. In size the viruses
are thought to scale down to about 10 mfL.
In the same range of size are also found the bacteriophages, or bac-
terium eaters. These agents live on young cultures of bacteria 'which
they cause to disintegrate or dissolve. As the amount of bacteria
decreases, the amount of phage increases. As the viruses can live only
on living culture tissue so can the various phages live only on suitable
bacteria. The phages are thought to be much smaller than most viruses,
probably from 25 to 60 mf,t. Because both the viruses and the phages
have very specific properties and reactions with their respective hosts, it
seems logical to infer that they have definite organization and the ability
to transmit their characteristics to their offspring perhaps by means of
genes. The fact that ,the known phages and viruses depend upon other
organic forms higher in the scale of life as sources of food material may
be advanced as an argument against their having preceded the higher
forms in time. It is a valid objection. Two escapes from the dilemma
are possible: (1) that earlier forms of phages and viruses were not so
limited but endowed with the power of subsisting on simpler organisms,
compounds, or directly on inorganic materials and (2) that the present
parasitic forms of phages and viruses are simply examples of these general
classes of organisms which have changed from earlier habits into that of
being parasitic on higher forms.
Recently an American biochemist, W. M. Stanley, succeeded in
crystallizing the virus of tobacco-mosaic disease. Other viruses have
since been secured in a crystalline or partially crystalline form. Various
nonliving materials like cane sugar as well as complex proteins and
enzymes also have the property of forming crystals. Protein molecules
scale down to around 4 mf,t, whereas starch and sugar molecules are
thought to be of about 2 and 1 mf,t, respectively. From this point, we
proceed down to individual atoms of unknown size and finally reach what
is considered to be the basis of all matter in the universe, protons, elec-
trons, and neutrons.
The best guess at the moment seems to indicate that under suitable
conditions various elements and compounds on the surface of the earth
or in the warm seas became organized into very simple forms of living
material with the power of nourishing and reproducing themselves. Very
likely this process occurred in a variety of places, and, for all we know
to the contrary, this form of spontaneous generation may still be going
on. The materials so formed were probably relatively stable, but with a
tendency for polymerization as well as for various rearrangements of
their atoms and molecules to produce somewhat different forms. From
these earliest living forms, there perhaps eventually came to be organi-
zations of materials resembling the gene.
Existing singly at first, the genes perhaps gradually combined and
took on a protective covering resembling at this point chromosomes.
Later, cytoplasmic materials and a cell wall may have been added, and
after a few hundred million years development may have reached the
one-celled amoebalike stage. Growth, aggregation, variation and differ-
entiation, and specialization of function proceeded, yielding finally the
hundreds of thousands of plants and animals that at present inhabit the
earth. It hardly seems necessary to say that we need not expect to find
traces of the earliest forms of life, because they were too small for us to
see and also because of their fragile nature. The developmental process,
then, may perhaps logically be thought of as progressing from the ele-
ments to organic substances, to genes, to chromosomes, to single-celled,
and finally to multiple-celled organisms and plants.
One-celled to More Complex Living Forms.-It is beyond the scope
of this chapter to do more than point out the general road that progressive
development apparently has traveled. The first step ·was perhaps the
formation of various hydrocarbons. Next may have come the con-
version of the above-formed carbohydrates into an amino acid by the
addition of nitrogen and the conversion of the amino acids into a protein
through fusion of many amino-acid molecules into one finally complex
molecule. Somewhere these minute and relatively simple molecular
aggregates must have become endowed with "life," especially in the sense
of being able to reproduce themselves as the genes apparently do today.
The reverse of the process described in the previous section might
lead from this genelike structure to a microscopically visible one-celled
organism after the passage of hundreds of thousands of years. From
this one-celled stage we pass to a colony of cells such as the Gonium, or
free-swimming colony of 16 cells, or Volvox, in which the cells are arranged
in a hollow sphere with (1) division of labor and (2) differentiation into
soma and germ cells. The next step is the invagination of one side of
the hollow-sphere stage forming a hollow sack two cells deep with the
outer cells specialized for locomotion, sensation, and protection and the
inner ones for digestion. Here the same opening serves as mouth and
vent. N ext we find the opening continuing through the body with a
separate mouth and vent. Segmentation has taken place, the worm
being an example of this stage.
N ow we pass to the forerunner of the vertebrates with well-developed
gill clefts and an outpouching of the gut region which will one day pro-
duce the notochord and eventually the backbone of higher types. The
cartilaginous ring at the mouth of the fishes gives us our first indication of
jaws, although the vertebrate jaw develops from the first set of gill bars.
We also find here our first true division into two sexes, which will persist
right up through the higher animals, including man. Here we should
mention too the Dipnoi, or lungfishes, which are intermediate between
the fish type and amphibian, though they have only a two-chambered
heart. The first land-living forms are believed to have come from the
lobe-finned ganoids, which like the Dipnoi had an incipient lung.
Amphibians such as frogs and salamanders now appear. They spend
their early life in water and later life on land. They have a three-
chambered heart and, in place of fins, five-fingered limbs connected to
the body by pectoral and pelvic girdles. After them come the reptiles,
some gigantic in size though so well adapted to their environment that
they could not change when it did and so became extinct. Here the gill
clefts are still retained in the embryo but the young have lost gills, and
the eggs are laid on land rather than in the water. In the higher forms,
traces of the fourth heart chamber appears.
From this reptilian type were developed both bird and mammalian
forms. Archaeopteryx was a primitive bird type in which the teeth
were still retained and the three fingers of the forelimbs were still distinct
and used for climbing. In modern birds the teeth have been lost, and
the fingers are united by a ,yeb. At the summit of the tree of life comes
the Mammalia, to which class man himself belongs.
The lowest mammalian forms are those like the spiny anteater and
duckbill mole. In these forms the egg is still laid outside the body and
the digestive tract and urogenital organs empty into a cloaca as in
reptiles and birds. Hair has now replaced scales, and a diaphragm sepa-
rates the abdominal from the thoracic cavity. The brain is not deeply
wrinkled as yet, and the nutritive glands are scattered over the belly
region of the female and give out a fluid that is licked up by the young.
The second stage in the mammalian group is represented by the
marsupials, such as the kangaroos and opossums. The young are born .
immature and are placed in a pouch on the mother's belly where they
complete their embryonic growth. These species are viviparous and have
separate openings for urogenital and alimentary tracts as well as a dis-
tinctly four-chambered heart.
The highest stage that the process has yet reached is that of the
placental mammals. In these the embryo is retained for a longer time
in the mother's womb and connected to the mother for nourishment at
the placenta. Here the brain is more convoluted or creased. The
mammae, or teats, are well-developed for suckling and arranged over the
belly region usually in two rows, although in the highest types they
become restricted to the breast region. Representative types here are

the ungulates or hoofed animals such as horses, cattle, sheep, swine, etc.,
the carnivores such as cats, dogs, wolves, tigers, etc., rodents such as
rats, squirrels, etc., anthropoids such as apes, gibbons, gorillas, orangs,
chimpanzees, and, finally, man himself.
In the anthropoids we find a continued shortening of the snout region
and the gradual assumption of an upright position, first evidenced through
the use of the arms in swinging in the trees. The increasing size of the
brain further shortened the snout, caused the further arching upward of
the skull, and eventually threw the face into a similar plane to that
of the vertebral column. The gorilla spends most of his time in the trees,
although he does walk clumsily on his hind legs, but his long arms touch
the ground frequently-a necessity if he is to keep his balance. The
gibbon, a more primitive form than the gorilla, also walks on the hind
legs and uses the arms for balance.
The upright position largely frees the front legs from supporting the
body and enables the hand to develop as a grasping organ, to pick up
and examine food, and to bring it to the mouth, which is now so greatly
flattened that it is not so well-suited for picking up food as in long-
snouted animals. This also frees the hand for development as a tool-
using member, and even the higher apes are able to learn to use simple
implements wielded by the hand. The baboons frequently use sticks to
pry up stones in search of food. Chimpanzees use stones to crack nuts
and sticks to pry things they cannot move ,vith their fingers.
Organic Evolution.-The process sketched in the preceding section is
generally known as organic evolution. By definition it is "the theory that
the various types of plants and animals have arisen by descent with
modification from other pre-existing types." Because man is apparently
one of the latest products of the evolutionary process, our evidence for
it, is indirect. From the researches of physicists, chemists, paleontolo-
gists, and biologists, we can assume that matter in our solar system has
progressed through a succession of stages from electronic to atomic to
molecular to colloidal-organic to living forms. The most interesting
phase of the process from our standpoint is the development of mind.
This arose gradually and has reached its highest present level in man. It
has provided the means for increasing control over and independence
from the environment and at last enabled us to catch a glimpse of the
processes of the universe. Contrary to some opinion, as our knowledge
broadens, our reverence deepens.
This view is diametrically opposed to the supernatural theory of
separate creation of each species or breed. \Ve cannot here go into the
philosophical implications involved in these two contrasting theories.
Each individual should think out his own philosophy of life rather than
accept someone else's. There are two doors leading into the office where
the writer is now working, and, although from a scientific standpoint, it
might be important to determine which door one used for entrance, after
all, it is not a very important moral or ethical question whether one
entered by one door or the other. That one arrives at his destination is
certainly of greater ethical importance than his method of travel. The
mystery, as well as the beauty, of life is here with us now and we are part
of it. How these qualities and we ourselves arrived, we do not exactly
know, and perhaps we never shall. The question as to who made the
universe and why may finally prove to be rather overlarge for our human
faculties, bu.t this need not necessarily plunge us into a state of cynicism,
boredom, peevishness, or hostility. We are not here concerned with the
question as to "why" life is now found on this planet; nor are we" primarily
concerned with its mode of origin or its ultimate end. The first and the
last of these questions rightfully belong to the field of philosophy.
Science, which is but another name for organized knowledge, is con-
cerned with the mode of origin of living things, but as yet it has no def-
inite answer to this problem. The theory of evolution accepts life as a fact
and tries to trace its course through the ages. Once given a living entity
that is not immutable but rather prone to vary, the task of the biologist
is to ascertain the causes of variation and finally to shape and to mold
them to more desirable human ends.
The concept of an evolutionary process is not new, for in its broad
outlines it was sensed by early Hindu and Greek philosophers. They
merely had the vague notion but were able to adduce no proof. The
early fathers of the Christian church seem to have been rather liberal-
minded men, but the dogma of the special creation of species and the
literal interpretation of the Mosaic account of creation soon came to be
the accepted belief. During the difficult Middle Ages, little or no thought
could be devoted to questions of this nature. There are occasional ref-
erences to the general subject in the later writings of men like Bruno,
Francis Bacon, Leibnitz, Linnaeus, and Kant, although most of the men
of these times who thought about the matter at all believed in the
separate creation and the immutability of species.
j Buffon (1707-1788), a French naturalist, greatly enlarged the concept
of mutability through direct environmental means, an idea which he
developed in later life after having earlier subscribed to the idea of
immutability. He is a precursor of Lamarck and perhaps also of Darwin
in regard to pangenesis as well as the struggle for existence and the sur-
vival of the fittest. Coupled with his idea of change due to environ-
mental influences was his further implied belief that these so-called
"acquired" characters were heritable. Buffon's chief part seems to have

been ·that of suggesting broad new vistas, rather than filling in details,
asking questions for his followers to answer, or opening up new fields for
conjecture and experimentation. To him goes the credit for beginning
the consideration of organic development from the observational and
inductive standpoint.
/Erasmus Darwin (1731-1802), an English physician and poet and a
grandfather of Charles Darwin, is one of the most imposing figures in
this field of human inquiry. He borrowed and enlarged on many old
ideas concerning evolution and made distinctive new contributions of his
own. He was the first to stress the idea that evolution has been operating
from the time of the first primordial life. To him new forms were but
the flowering of potentialities originally deposited in the life stream by
the Creator and called into being by the necessity for adaptation. He
foreshadows the work of his grandson Charles in allusions to man as an
offspring of lower quadrupedal ancestors, in some of which the thumb
had accidentally come to lie opposed to the fingers, giving a hand that
led finally to the great development of the human mind. He assumed
a simple living filament endowed with irritability and sensitivity as the
progenitor of all that lives. The first living material was endowed with
potentialities for new developments that were called out by the organism's
response to pleasure and pain. He followed Buffon and anticipated
Lamarck in his belief in the inheritance of acquired characteristics.
/Lamarck (1744-1829), a French botanist and zoologist, can be right-
fully called the father of the modern theory of evolution O\ving to his
having been the first to devise a phyletic tree to include all plants and
animals. He was the first to state that all animals formed a branching
series of related forms, shading into each other but with many gaps due
to the intermediate forms not yet having been discovered. He believed
that small masses of mucilaginous matter took on living properties and
became elemental plants or animals.. His theory of evolution pro-
pounded in 1809 has three main points: (1) that the environment, directly
in plants but indirectly in animals through the medium of the nervous
system, calls out changes in the organism, (2) that the use or disuse or
parts leads, respectively, to their further development or to their atrophy,
and (3) that these s~-called "acquired" characteristics are inherited.
The main idea to be noticed here is Lamarck's belief that the environ-
ment is the principal cause of change. This is similar in its general
outline to Buffon's earlier belief that something external to the organism
brings about changes rather than something within the organism.
Geoffrey St. Hilaire (1772-1844), another French zoologist, subscribed
to Lamarck's idea that the environment caused the developmental
80 BRRRDING AND IMPROVRMENT OF FARM ANIMALS
process but by means of changes brought about in the embryonic 01'

germinal condition rather than in the adult. He thus, in a sense, antici-
pated Weismann's idea of germinal variation and causation as well as
De Vries's idea of evolution by means of large jumps or mutations, for his
work in the field of teratology made him acquainted with sudden depar-
tures from type that, he reasoned, might give rise to new species.
Goethe (1749-1832), the great German poet and philosopher, was, in
his earlier days, a scientist working in the fields of botany and anatomy,
contributing original papers on the metamorphoses of plants and sug-
gesting, erroneously, that the skull 'vas of vertebral origin. He vigorously
opposed the idea of the fixity of species and upheld that of descent with
modification, or of the ascent of man from some lower form. Xn the
decade between 1780 and 1790, Goethe provided the beginnings of one of
the pillars on which the theory of evolution rests, viz., that of comparative
anatomy and embryology. Osborne yields him the honor of being the
first man to conceive evolution in the modern sense and says that the
transfer of his affections and abilities to the field of literature retarded
the demonstration of the law of evolution by half a century. He first
offered the correct solution to the riddle of vestigial structures as remnants
of those which were necessary further down the scale of life and fore-
shadowed the dual nature of the process--its centripetal, conservative,
hereditary force and its centrifugal, progressive, variational force.
Cuvier (1769-1832) was a champion of the fixation of species and a
stout opponent of transformism. He, nevertheless, was the founder of
paleontology, which furnishes still another of evolution's supporting
pillars. He explained the series of different forms in the earth's crust
as due to catastrophic extinctions and later migrations, whereas some of
his students and followers said they were each due to a separate creation.
Cuvier's exalted place in the science of his day served to retard the
acceptance of the idea of mutability, and his brilliant pupil Louis Agassiz
was never convinced of the reality of transformism.
Several noted embryologists of the early nineteenth century stressed
the mutability of species and called particular attention to the many
similarities between different animals in their early embryonic states and
to the fact that the higher animals pass through stages which form the
terminal point for lower animals. Many other writers in the first half
of the nineteenth century had inklings or convictions about" creation by
evolution"; and two of them, Wells in 1813 and Matthew in 1831,
recognized the principle of natural selection, which was to form the
cornerstone of Charles Darwin's theory as expressed in "The Origin
of Species."
Opposition to Mutability.-We should here remind ourselves that in
this brief review of the history of men's thinking about animal origins up
to the time of Darwin, we have said nothing of the general opposition to
the theory of gradual change. True there had been martyrs to this
.,<Teneral idea back in the fifteenth and sixteenth centuries, and many others
had been forced to recant. The general populace, however, knew little
or nothing about such things. Life here was hard for the mass of man-
kind, but they had what seemed a sure promise of a better one to come.
The Roman Church and its Protestant derivatives alike frowned upon
any views of man's relation to the universe other than those given in the
Mosaic account. So although the general idea of change had had a long
history and its development had been contributed to by scores of men
for over 3,000 years, yet mankind as a whole was both ignorant of what
had transpired and bound by certain dogmas directly antithetical to the
idea of evolution. But into this hostile, prejudiced atmosphere the idea
did finally come in 1859. It has had a stormy passage but is now
accepted by most educated people, for it seems to offer the best available
solution to the age-old problem of the whence of animal forms.
Darwinism.-Charles Darwin (1809-1882), to \vhom reference has
already been made, Was an English naturalist. He had been educated
in the universities of Edinburgh and Cambridge, studying both medicine
and theology. He practiced in neither field, however, and wrote later,
"My scientific tastes seem to have been certainly innate . . . I consider
that all I have learnt of any value has been self-taught." The foregoing
pages will have indicated that Darwin was not the father of the idea of a
gradual progressive change. He might perhaps more correctly be called
the attending physician who brought the conception safely into the world,
or as Butler said, "Darwin's chief glory is not that he discovered evolu-
tion but that he made men believe in it, and what glory," he added,
"could be greater than this?"
Darwin sailed on H.M.S. Beagle as naturalist on a round-the-world
trip lasting from 1831 to 1836. On this voyage he had the opportunity
of studying the rich fauna and flora of South America and several island
archipelagoes and was struck by the manner in which animal types
shade into each other as well as by the distinctive forms found on separate
islands. He began this journey with a belief in separate creation, but
the facts he observed caused him gradually to give up this explanation for
that of the mutability of species. On his return to England in 1837, he
began to collect facts on plant and animal variation and during the next
year, while reading Malthus's "Essay on Population," the idea of the
struggle for existence that is constantly going on in nature as an explana-
tion of the great variety of plant and animal species came to him as it
had previously come to Wells, Matthew, and perhaps many others.
Darwin's thesis involved four points: (1) that organisms vary, i.e.,
are not exactly like their parents, which no one can successfully deny; (2)
that these varie,tions are or may be hereditary, i.e., passed along-to future
descendants, which again is a commonplace; (3) that there is a continual
struggle for existence in nature; and (4) that those best fitted to survive
in the given environment will be the most apt to survive and should,
therefore, leave the most descendants, which seems to be a very logical
assumption. He thought that the working of the above-named four
principles would account for the great variety of forms of plants and
animals which now inhabit the earth or have inhabited it in the past.
Darwin accepted evolution as a working hypothesis, accepted the
obvious facts of variation and hereditary transmission of potentialities,
and adduced a tremendous amount of data that seemed to support his
thesis that natural selection or, to use Spencer's term, "the survival of
the fittest," is the leading actor in this drama of survival. Darwinism
is not the process of gradual development itself but is one suggestion as
to how evolution works. Discussions among biologists, therefore, regard-
ing the correctness or incorr~ctnegs of Darwinism are discussions regard- .
ing the mechanism of evolution,' not questionings of the process itself,
which practically all· biologists and -other scientists accept as a very
probable--as the most probable--fac~.
Darwin saw that organisms vary.' He had observed the keen com-
petition to survive in both the plant and animal world, and, when these
two ideas are added, the sum is inevitably natural selection. As has
been said, the title of Darwin's book should have been the" Survival"
rather than the" (Origin' of-Species." He accepted variation as he.saw
it without attempting to explain its underlying causes. He accepted the
older ideas of the influences of use and disuse of parts in causing develop-
ment or atrophy. He concurred in the belief that the environment could
cause heritable variations. He made no distinction between germinal
and environmental variations. He believed to a certain extent at least
in the inheritance of acquired characters and formulated a mechanism
for its operation in his theory of pangenesis, i.e., gemmules from flll parts
of the organism being deposited or registered in the germ cells ready to be
passed along to the next generation.
It would have been truly miraculous if Darwin had made no mistakes.
Little was known at the time about germ cells and their cytology.
Chromosomes, now the accepted bearers of hereditary potentialities,
were undreamed of, and naturally the synthesis of cytology and genetics
into a firm foundation for biological thought was lacking.
Darwin did not distinguish between genetic and environmental varia-
tions but depended on a "strong principle of inheritance" that pre-
sumably included both. Later research has shown that the germ cells,
ova ~d spermatozoa, are the bridge of inheritance between parents and
their offspring, and numerous attempts to demonstrate the inheritance
of acquired characteristics (environmental variations) have failed: Dar-
win accepted all variations as potentially heritable and was concerned
with the mechanism that allowed some variants t o survive and breed
successfully, whereas others failed. His conclusion was that the environ-
ment acted as the selecting agency, preserving some variations while
FIG. 23.- Charles Darwin. (Courtesy of The Jo w'nal of Heredity.)
destroying others. This is known as natural selection. He built · his

theory on the cumulative action of a multitude of small variations and
looked with disfavor on the possible contribution of big jumps, saltations,
or mutations.
Darwin's success was due to his care and patience in gathering facts,
• task to which he devoted 25 years, and to his formulation of a workable
&eory into which the facts would fit. He used the inductive method of
..ae:J.enlce, from facts to general principles rather than the deductive method
principles to facts. He believed that variations occur at random and
Heaven to ~ 'forfend him from Lamarck's nonsense of a tendency
84 BREEDING AND IMPROVEMENT OF' FARM ANIillIALS
to progression or from the slow willing of animals"; in other words, he

did not believe in some outside design.
A. R. Wallace, another English naturalist and contemporary of Dar-
win, reached a conclusion very similar to Darwin's at about the same
time, although he had not marshaled so much data to support his thesis as
had Danvin and therefore like a true gentleman and scientist yielded
first place in this field to Darwin.
Natural Selection.-Thus far no better theory for the appearance of
new species has been suggested than Danvin's natural selection among
naturally arising variations. Danvin did not try to account for the
origin of new heritable variants that might in time become new species.
Concern over natural selection, adaptation, and the building of hypo-
thetical classifications probably did much to retard inquiry into the basic
facts regarding the origin and mechanism of hereditary variations and
their mode of transmission. It was not until the twentieth century,
through the synthesis of chemical, physical, genetic, and cytological
sciences, that we were to begin to have a clear conception of the basis of
living matter and its gradual development through germinal changes.
There have been many attempts to deny the validity of the corner-
stone of Darwin's theory, natural selection, and likewise many attempts
to replace it with other theories. These in general take two forms: one
the substitution of orthogenesis, or progression from natural causes in a
straight line, for Darwin's random variation; and the other an appeal to
an internal, mystic directing force within the living material.
For the emergence of complex from simple animal forms two things
are necessary: (1) living material and (2) some variation. These are
obvious to any observing person, and there has yet to be suggested a
better selective agency than the one proposed by Darwin. In many
species of plants and animals, nature's passive selection has now been
superseded by man's active selection. Our task as livestock breeders
consists of learning more about the workings of living material, mecha-
nisms of heredity and variation, and of devising systems of breeding and
selection so that the inherent possibilities of our stock may have the
opportunity to make themselves manifest.
Evidence for Evolution.-Lack of space prevents a lengthy discussion
of the various lines of evidence which seem to indicate the validity of the
general theory of evolution. In outline, they are as· follows:
1. Classification or Taxonomy. The fact that all the 800,000 described
species of animals can be arranged into 11 phyla and the 250,000 known
species of plants into 4 phyla, the whole resembling a branching tree.
2. Comparative Embryology. The fact that all embryos start as single-
celled zygotes, progress along similar lines of development, and that those
ANIMAL ORIGINS AND PROGRESBION 85
of the higher species pass hurriedly through stageS which are the end
product of organisms lower down the evolutionary ;'lcale.
3. Comparative Anatomy. The fact that the brttin, heart, and other
organs show a progressive development from lower to higher species, that
all mammals have the same bony framework, the bones in the flipper of a
whale, the wing of a bird, the front leg of a horse, and the arm of a man
being similar in general form and articulation.
4. Vestigial Structures. The fact that man during his embryonic life
develops a set of gill clefts and arches, a tail, and a fairly heavy covering
of hair, all of which disappear before birth, and that adult man has ear
and tail muscles, an appendix, and many other vestigial structures which
arc well-developed and functional lower down the animal scale.
5. Paleontology. The fact that fossil remains of animals show a
progressive development from lower to higher in ascending strata of the
earth's crust.
6. Zoogeography. The fact that species grow increasingly divergent
in form in widening circles from their point of origin.
7. Blood Tests. The fact that the blood of closely related animals
shows small incompatibilities as measured by amount of precipitation,
whereas distantly related ones show great incompatibilities.
8. Observation and Experiment. The fact that all of our breeds of
farm animals have been made by a process of selection during about the
past 200 years and that recently a successful cross was made between the
radish and the cabbage giving a true breeding form having some charac-
teristics of both parents but no longer being fertile with either parent.
Evolution of the Horse.-The evolution of the horse is more com-
pletely kno·wn from fossils than is that of any other species of animal.
Evidences of horses are found in both the New and Old Worlds as far
back as Eocene time. Loomis, the leading American student of the
fossil horse, held that the species arose in ·Western North America and
migrated across ,,,hat is now Bering Strait into Asia and Europe. They
were once populous in South America as well and have been completely
wiped out in both continents at different times. Prehistoric horses
were very abundant in both North and South America but were not
present here when Columbus made his discovery of this continent.
The American representative of Equus in Eocene times Was Eohippus.
This was a small animal 10 to 20 in. tall, with four toes on the front feet
and three toes on the hind feet. The horse representative of the Oligo-
cene was Mesohippus, a three-toed horse, standing [Lbout 24 in. high. On
through the Miocene, Pliocene, and Pliestocene, the evolutionary proce~s
reveals successive adaptations to perpetuate the species from the stand-
point of both nutrition and reproduction.
86 BREEDIA'G AND IMPROVEMENT OF FARM ANIMALS
As the horse changes from a denizen of the swamp to that of the

forest and later to that of the open prairie, his teeth grow longer, stronger,
and more roughened to suit the gradual change to harsher grasses,
Accompanying this change comes a great change in the stru~ture of the
feet and legs. The cannon bones, metacarpals and metatarsals, lengthen,
the middle toe grows larger and stronger, and the other toes gradually
disappear. This permits greater and greater speed, on which this
species had to depend for survival. In the modern horse, only the
middle or third toe remains, \vith rudiments (splints) of the second and
fourth toes. The Eocene horse with four toes recounts the first of these
changes. It is confidently hoped the original five-toed horse will one
day be discovered. Figure 24 shows the progressive changes in the teeth
and forelimbs of the horse from Eohippus to the modern Equus.
Mechanism of Evolution.-Granted living forms, capable of survival
and reproduction, and granted relative but not absolute stability, an
evolutionary process would seem to be inevitable. The obvious question
at this point would appear to be, what is the mechanism of the process,
how did it work? The answer logically involves two factors, one the
internal features and the other the influence of the external environment.
We have assumed that life originated in minute and relatively simple'
forms. Because the modern science of genetics has shown that inherit-
ance at the present time depends upon the make-up of the chromosomes,
meaning thereby their content of genes, it again seems logical to suppose
that this has been the system of inheritance from the very origin of life
onward.
We do not at the present moment know the chemical content or
structure of any gene. After proper staining, the chromosomes in all
the higher species can be readily observed. These bodies always have
a definite form and number in each species. They appear superficially
to be made up of units strung together much like beads on a string.
These individual units are the genes. Studies of the giant chromosomes
found in the salivary glands of Drosophila melanogaster when in the·
larval stage have shown that the genes are discrete bodies or regions of
the greater whole, the chromosome. Cytological investigations have
revealed also that demonstrable departures from the normal in mature
pomace flies are correlated with evident changes in the form or arrange-
ment of the discrete units making up the chromosomes, the genes. These
and many other features of correspondence between genetic development
and cytological form will be discussed later when we discuss the chrom-
osomal theory of inheritance. Suffice it to say here that this evidence
has now reached the point where it can no longer be doubted. In other
words the modern breeder now has a workable theory of inheritance
whose broad outlines appear applicable to all forms of life and whose
details have already been worked out in many of the lower, simpler,
more rapidly multiplying forms. And this theory rests solidly on the
genes as the hereditary units involved in the transmission of potentialities
from parent to offspring.
The mechanism of inheritance between individuals today is the gene.
The mechanism of evolution, inheritance between all living forms in
all time, is undoubtedly also the gene. For this to be true, the gene
must be able to reproduce itself exactly. It does this today at every
cell division, it has done it billions of times in your own body, it has
undoubtedly been doing it since the earliest living forms first appeared.
In addition to reproducing itself exactly, the gene must also be capable
of occasionally producing some slight change as it reproduces itself.
This power has also been demonstrated in today's living forms, and
such changes have been given the name mutations. There would seem
to be no right or necessity for denying this power to living forms from
their very inception.
This, in brief form, is the content of modern evolutionary thought.
In other words, its mechanism is to be found in changes in the genes.
Of necessity, only those changes which were relatively slight would
survive. Any living form, from a single gene to a horse, is a very deli-
cately balanced mechanism. Any major change would probably be
lethal. A horse's legs might grow a little longer, and, if this little Was
not too much to prevent the mouth from reaching to the ground, the
horse would survive; if too great, it would be handicapped in securing
food. Likewise, small changes in genes are possible of survival, large
ones probably not. The gene or the horse can adapt itself to small
changes, not to large ones. In time, the summation of many small
changes \yould, of course, cause the organism to vary considerably
from its origInal state This will be recognized as the heart of the
Darwinian theory of the origin of new forms. It is still the most logical
explanation of that process.
The stream of life is, therefore, the germ plasm, and the germ plasm
is the sum of the genes .. Perhaps the genes existed first as naked bodies,
next became grouped into aggregates, next took on cytoplasm to assist
in their growing complexity of function. We might liken the whole
process to radiating assembly lines. The hub or core is the simplest
living form, the gene. One assembly line adds but little material as
the genes pass along it, and there eventuates at the end of the line an
amoeba. Another adds more and different materials, so that there roll
off this line marine forms like fish. Yet another line produces amphibia,
another reptilm" another birds, and yet another mammals. The workers
ANLtfAL OIUGIl'·;S AAD PROGRESSION 89
along these assembly lines are the genes, their raw materials chemical
elements or compounds, their finished products all the millions of forms
of plants and animals that have lived or exist now.
Role of the Environment.-The other feature of this general problem
of the mechanism of evolution is concerned with the environment.
Various roles have been ascribed to this feature in the past. As noted
earlier, Lamarck and others before and since his time have given first
place to the environment as the causative factor of variation. That
the environment may have been very active in causing variations in
the minute and relatively naked early living (germinal) material seems
quite probable, after the hereditary material had encased itself in a
protective coat of cytoplasm seems less likely. Just where the dividing
line may have come is hard to say. Recent research has shown that the
germ cells of forms like D. melanogaster are highly subject to change due
to direct treatment by varying dosages of X rays or of heat. It seems
unnecessary to doubt that the relatively better protected germ cells of
higher forms might be subject to similar experimental influences, though
perhaps to a lessened degree. That the germ cells can be changed by
such processes seems well established, that the changes can be induced
in an advantageously directed manner is not established. Owing to the
delicate balance between any organism and its environment, most of
the changes seem to be deleterious rather than beneficial.
The general role of the environment, at least in later stages of the
developmental process, seems to have been that of a passive selective
agent rather than the direct causative agent. Probably both natural
genetic variability and that induced by the environment have functioned
from the start. In its earlier stages, changes were perhaps more largely
due to the environment; in its later stages, mostly to genetic variations,
natural changes in the genes, although for any variation there is pre-
sumably some specific cause. This natural genetic mutability is capable
of at least two sorts of explanation. Either it is induced by some outside
supernatural force, or it is inherent in the relatively unstable nature of
the material itself.
The mechanism of inheritance is to be found in the germinal mate-
rials, the genes. The process of progressive change is best viewed as
that of interactions between the genes and their environment, both
cellular and general. In the higher forms, the environment seems to
act largely in a passive manner, somewhat perhaps in the nature of a
sieve, allowing some varieties better fitted for their particular environ-
ment to survive and leave offspring, whereas others are unable to do so.
lt also seems plausible to think that variations in a certain direction
might lead to still others in the same direction, a view known as ortho-
genesis. The last man born arose through a gradual developmental

process beginning as one undifferentiated zygote, ending as an organism
made up of billions of cells organized into a definite pattern that sets
him off from all his fellows as a unique individual. Logic as well as
fact, as far as known, dictate that the first man was made by a similar
process.
Mechanism of Development.-Any organism, plant or animal, is
composed of cells, one, hundreds, thousands, or billions. In the year
FIG. 25.-Diagram of a typical cell, with structures commonly present. at, attracti n
sph.ere; eh, chromatin network; ei, cell inclusions; em, cell membrane; en, centrosome; .
cy, cytoplasm; t, linjn thread; n, nucleus; nt, nucleolus; nm, nuclear membrane; n8, nuclear
sap; p, plastids; pdiv, plastid dividing; v, vacuole. (From Shull, Principles 0/ Animal
Biolouy.)
1665, Hooke observed the cellular structure of cork. In 1838, two Ger-
man scientists, Schleiden and Schwann, propounded the cell theory for
both plants and animals. In its very briefest form, the cell doctrine
states that all plants and animals are made up of units called cells. In
both the plant and the animal kingdoms are found all degrees of com-
plexity from single-celled organisms, like algae and amoebae, to highly
specialized structures, such as plants and farm animals, which consist
literally of billions of cells.
A cell is a bit of living protoplasm in which are to be found vacuoles,
plastids, a centrosome, and a nucleus, the whole sometimes surrounded
by a wall and sometimes naked. The really important part is the con-
tent of the cell, the sum total of which is called protoplasm. This includes
the nucleus, centrosome, and the remaining portion called the cytoplasm.~
The body of any multicellular organism is made up of cells (and cell

products) that have arisen through cell division from one original cell
called a zygote. The zygote is the product of the union of two reproduc-
tive cells, viz., an ovum from the female and a spermatozoon from the
male.
Following the fertilization of an ovum by a spermatozoon, the result-
ing zygote begins to grow by means of cell division. This one original
cell divides into two cells, these two into four, etc., until the complete
organsim, consisting of billions of cells in the higher animals, is produced.
The complex individuals found among the higher animals are composed
of a great diversity of kinds of cells that have arisen through speciliza-
tion. All the organs of an animal body are composed of cells more or
less specialized in form and function for definite purposes-lung cells
for breathing, brain cells for thinking, etc. Reproduction is likewise
affected through the activity of certain specialized cells.
Classification of Cel1s.-For clarity, and because of certain funda-
mental differences, the cells making up any organism are grouped into
two categories. One of these, called the somatoplasm, includes all the
ceUs necessary to the well-being of the particular individual whose com-
ponent parts they are. The other relates to the cells immediately con-
cerned in the production of future individuals and is called the germ
plasm. It is essential that the student differentiate clearly between
these two. True, the germ cells are housed in and nourished by means
of the soma or body cells, but differences in cell division as well as in
function place the former in a class by themselves. In farm animals
a new individual arises through cell division from a zygote, which is an
ovum or egg of the female that has been fertilized by a spermatozoon
from the male.
Both ova and spermatozoa are cells with the characteristic cytoplasm
and nucleus. The ova are produced in the ovaries of the female. Ova
are usually larger than the spermatozoa because of the stored-up food
material in the yolk that must serve to nourish the developing individual
until the latter becomes firmly attached in the uterus of the mother.
The spermatozoa, produced in the testicles of the male, are also specialized
for their function of seeking out and fertilizing the ova. They have a
characteristic shape, very unlike that of the ova, being made up of three
parts, viz., the head, containing the nucleus; the body, containing the
centrosome; and the elongated tail, which gives the spermatozoa their
motility. The chromosomes found in the nuclei of the germ cells furnish
the "bridge of inheritance" between parent and offspring.
Chromosomes and Genes.-The theory was advanced many years
ago that certain definite entities in the nucleus of the germ cell brought
about the development of certain characteristics in the fully developed

individual. Subsequent investigation in cytology has proved the cor-
rectness of this hypothesis. In the higher animals, there is a marked
distinction between soma and germ plasm, and also between the cyto-
plasm and the nucleus of the germ cells. It is in the nucleus that t he
determiners for the individual's characteristics are located. The nucleus
itself is made up of two principal parts, the linin network and the chro-
matin material. This latter goes through various stages in fulfilling its
functions . In a resting state, it is strung out in a long thread supported
on the linin network. As the cell
starts to divide, the chromatin
thread shortens and thickens and,
finally, breaks up into the number
of chromosomes characteristic of
the species.
Each species of plant and ani-
mal has a characteristic number
of chromosomes. This number is
found in the nuclei of all the cells
of the body that have a nucleus,
with the exception of the germ
FIG. 26.- Diploid chromosome complement
cells at certain stages as will be
at metaphase in root tip of Vieia faba . noted later. In the human, the
Spindle-attachment regions of a ll chromo- number of chromosomes is 48; in
someS near equator. Nucleolus-forming re-
gions are by the "secondary constrictions" cattle, 60; in Drosophila, 8; in
above and below. In this strain one pair of corn, 20; and in wheat, 16. The
short chromosomes has co nspicuous "con-
strictions" also (left and right). X 1750. chromosomes themsel ves are
(From Sharp, Introduction to CytoloGY.) divided into two categories, viz.,
autosomes and sex chromosomes, the latter being so named because
they carry the principal determiners for sex as well as for certain
other characteristics. The chromosomes occur in a variety of shapes,
though for the most part they are rod like or oval in shape. Abundant
cytological and experimental breeding data have now been diligently
studied, and these data have proved that each chromosome consists of
a great many constituents known as genes, which are the indivisible
units of inheritance.
It is supposed that certain of the filterable viruses which are assumed
to be the causative agents in certain diseases, such. as hog cholera, are
composed quite largely of nuclear or chromatin material with very little
or no cytoplasm present. As indicated above, it seems most probable
that life must have originated in somewhat similar minute bodies and
that from this minute origin all the varied forms of liying things hILV e:
-
·1
arisen by means of heredity and variation. The most logical reason

for the differences between individuals would seem to be that their
chromatin elements differ. The same is undoubtedly true of species,
class, and phyletic differences. During cell division the chromatin
becomes stretched out into a long thread, and at more or less regular
intervals the thread exhibits regions of thickening. In other words,
the chromosomes look like strings of beads. These beads are the genes
TABLE 8. ~CHROMOSOME N U11BERS IN MAN AND F AR:\I lVLuIMALS
Species Number Authority
Man .................. : ......... 48 Evans and Swezy, 1928

Horse........ .. . ............... . 60 Painter, 1924
Cattle....... ................... 60 Krallinger, 1927
\ Sheep........ .. ...... ...... .. ... 54 Berry, 1938
Swine........................... 38 Krallinger, 1931
Goat...... . . . . . . . . . . . . . . . . . . . . . . 60 Berry, 1938
or factors which bring about the development of all of our individual

characteristics. Each chromosome is always reformed from the same
group of genes, and each gene always occupies a definite locus on its
particular chromosome.
Somatic Mitosis.-Somatic mitosis or cell division is the process
whereby one cell, through growth and subsequent division into approxi-
mately equal halves, produces two cells. Somatic mitosis is a continuous
process starting with one cell in the resting state and ending with two
cells in the resting state. For clarity the process is often divided into
four phases, as follows: prophase, metaphase, anaphase, and telophp,se.
In a typical cell in 'the resting state, the chromatin of the nucleus
is strung out into long threads presenting the appearance of a tangled
skein. In the prophase, a shortening and thickening of the chromatin
and its division into the number of chromosomes characteristic for the
species, together with the arrangement of the chromosomes in an equa-
torial plane, take place. The nuclear wall also disappears during this
preparatory stage. By this time the centrosome, which lies just outside
the nucleus, has divided, and one-half of it has proceeded to each, pole
of the cell with the accompanying arrangement of the asters attached
to the chromosomes. In the metaphase, the chromosomes split length-
wise (or reproduce themselves) into two equal parts, and these parts
start to move to the opposite poles of the cell. In the anaphase, the
two groups of chromosomes complete their migration, one group to each
pole of the cell. 'In the telophase, the cell wall constricts, forming two
94 BREEDIN G AND I MPROVEMENT OF FARM ANI MA LS
2 3
4 .5 6
!) 10
FIG. 27.- Diagram ol mitosis in a lour-chromosome species, 1-5, propJ.ase; ~8: meta-
phase; 9, anaphase; 10, t elophase.
new daughter cells, each with cytoplasm, centrosome, and nucleus"\vith

its chromatin, in place of the one original parent cell.
Such cell division with differentiation is characteristic of the rise
of a many-celled organism from a single-celled zygote. At times this
process stops short and a dwarf results; and at other times it continues
beyond the normal, the individual with an extra number of cells being a
giant. Various types of abnormalities arise from causes urikno\Yn.
fOe.",. 10.S 10.10 10.15
8tlf.i.),
~
ro.17 10.18 10.19
10.22
10.23
10.30
70.25 70.40
10.50
7t a.IfI.
_ _if~!? ~
28.-The division of a living cell in a culture of chick embryo cells. Above are the
at which the drawings were taken, the whole series representing 1 hour. (From C. C.
The Mechani8m 0/ Creative Evolution, The Macmillan Company.)
force that specifies when, where, and how differentiation is to take

PMra<:e is not definitely known. Perhaps it is the result of response to
stimuli, perhaps to pressure of other cells, perhaps to the influ-
...."...,_._- of organizing centers. Up to the present, differentiation is still
mystery, but the wonder should arise not at the fact that
abnormalities are produced hut, rather, that the whole
complicated process from the one to the billions. with consequent differ-
entiation is ever accomplished perfectly.
Origin of Germ Cells.-The mechanism of inheritance resides in the
germ cells, the only direct contribution of parents to their offspring
in the higher species. The mode of origin of germ cells is a very impor-
tant consideration. As we shall see in detail in the next two chapters,
they are produced in testicles and ovaries, but they should be thought
of as direct lineal descendants of the fertiliied ovum that has also given
rise to the entire somatoplasm of the individual now producing ge;m
cells. At sexual maturity, germ cells "Yvill begin to be produced by the
germ plasm, but the chromosomes and genes in these germ cells are not
derived in the final analysis from the body of the parent (somatoplasm)
but from the same source from which the body of the parent has come,
1,iz., the fertilized ovum or zygote.
Summary.-We have tried in this chapter to trace the broad outlines
of ,he process of development of animals from its probable inception in
ultramicroscopic forms on up through its varied manifestations to its
final eventuation in the highest mammals. We have tried to correlate
the processes of heredity and variation, as we know them to manifest
themselves today in our breeding animals, with the greater process of
the production of new species and races. "\Ve have seen that life appar-
ently first manifested itself in some very simple forms, and that by
means of heredity and variation plus differential survival all the thousands
of higher forms have gradually made their appearance in an orderly'
progression. Thus it has appeared that all living forms are related, or,
if not all tracing back to. one original living thing, they all must trace
back to similar original manifestations of life. Finally, we outlined the
process of somatic mitosis by means of which a new individual gradually
arises from a fertilized egg.
References
Books
BELL, E. T. 1938. "Man and His Lifebelts," Reynal & Hitchcock, Inc., New York.
DARWIN, C. 1859. "The Origin of Species," A. L. Burt Company, New York.
DOBZHANSKY, T.· 1941. "Genetics and the Origin of Species," Columbia University,
Press, New York. •
GASKELL, A. 1928. "What Is Life?" Charles C Thomas, Publisher, Springfield, Ill.
HURST, C. C. 1932. "The Mechanism of Creative Evolution," The Macmilla1.l
Company, New York.
LINDSEY, A. W. 1929. "Textbook of Evolution and Genetics," The Macmillan
Company, New York.
LOOMIS, F. B. 1926. "The Evolution of the Horse," Marshall Jones Co., Boston.
MASON, F. 1928. "Creation by Evolution," The Macmillan Company, New York.
MONTAGUE, WILLIA~[ P. 1940. "The Ways of Things," Prentice-Hall, Inc., New
York.
I1NI.MAL ORIGINS AND PROGRESSION 97
MOORE, R "The Origin and Nature of Life," Henry Holt and Company, Inc., New
York.
MORGAN, T. H. 1932. "Scientific Basis of Evolution," W. W. Norton & (Jompany,
New York.
MOULTON, F. R. 1937. "The \Vorld and Man as Science Sees Them," Univer~ity of
Chicago Press, Chicago.
NEWMAN, H. H. 1925. "Evolution, Genetics and Eugenics," University of Chicago
Press, Chicago.
OPARIN, A. I. 1938. "The Origin of Life," translated by S. Morgulis, The Macmil-
lan Company, New York.
OSBORN, H. F. 1929. "From the Greeks to Darwin," Charles Scribner's Sons, New
York.
PENROSE, S. B. L. 1941. "Philosophy for Lowbrows by One of Them," 'Yhitman
Publishing Company, Racine, Wis.
RANDALL, J. H., JR. 1926. "The Making of the Modern Mind," Houghton Mifflin
Company, Boston.
SCOTT, 'Y. R 1917. "The Theory of Evolution," The .:\facmillan Company, New
York.
SErFRIZ, 1VILLIA'If. 1936. "Protoplasm," :'.lcGraw-Hill Book Company, Inc., New
York.
SHARP, LESTER W. 1934. "Introduction to Cytology," 3d ed., McGraw-Hill Book
SHULL, A. FRANKLIN. 1936. "Evolution," McGraw-Hill Book Company, Inc.,
Kew York.
SIMPSON, G. G. 1949. "The Meaning of Evolution," Yale University Press, New
Haven.
TYLER, J. M. 1912. "Man in the Light of Evolution," Appleton-Century-Crofts,
Inc., New York.
SECTION II
Mechanisms of Reproduction
CHAPTER IV
THE MALE'S PART IN REPRODUCTION
The science of physiology seeks to explain the how and the why of all
the varied reactions or responses of living organisms to external or internal
stimuli and to determine the function of each organ of the body, as well as
the interrelations existing among all parts of the body. The physiology
of reproduction seeks to explain the part that each of the specialized
organs of this system plays in the animal's ability to reproduce its kind.
From the time of the early Greeks on through the Dark Ages, the works
of Galen, an anatomist of the second century, served as the authority
in anatomical and physiological science. This was in keeping with the
general tendency of the time to believe what someone else had said,
rather than to question and to seek to test whether what was asserted
was actually true. Galen used the method of observation but made his
observations on animals alone, applying the results in a general way to
man.
In 1315, an Italian professor, Mondino da Luzzi by name, published a
work on human anatomy based on dissection of human bodies rather than
on Galen's authority. It remained, however, for Vesalius, a Belgian born
in 1514, to break the allegiance to old authority and to set up experimental
inquiry in its place.
After the establishment of the experimental method, in which Harvey,
an Englishman, played a leading part, all that remained needful was the
perfection of method. Up to the seventeenth century, restricted progress
was all that could be expected, for it was not until then that men such as
Hooke and Grew in England, Malpighi in Italy, and Swammerdam and
Leeuwenhoek in Holland began to develop compound microscopes for
study of minute phenomena of all kinds. Physiology is founded upon
the function of organs, but it was not until 1838 that men began to under-
stand even the structure of organs. This development started with the
founding of the cell doctrine of Schleiden and Schwann in that year. -
Preformation and Incasement.-The older thought and belief regard-
ing reproduction were embodied in the "preformation theory." This
98
•
THE MALE'S PART IN REPRODUCTION 99
was, in effect, a theory that the whole body with all its parts was already
contained in the egg or sperm of an animal but in such a minute and
transparent form that it could not be detected and that, therefore, the
whole of development was nothing more than a growth, or an unfolding,
of the parts that were already "infolded."
Closely connected with this theory was that
of incasement. According to this theory, the
embryonic ovary was supposed to contain the
ova of the following generation; these again
the ova of the next, etc., ad infinitum.
Early Discoveries in Embryology.-Wolff,
in 1759, after careful observations, proved
that embryonic development did not consist
of an unfolding of the preformed organs but
was a series of new constructions, one part
arising after another, all of them simple in
their early stages, though perhaps highly
specialized later. The substantiation of his
finding by Oken, in 1806, stimulated activity
in the field of embryological research. Baer
was the most successful of these investigators.
His contribution led to the discovery of the
three germinal layers on the germ disk of all
the higher animals, viz., the ectoderm, the
mesoderm, and the endoderm.
From the ectoderm arise later the nervous

aystem. important parts of the sense organs, the
outer layer of the skin, and some others; from the FIG. 29.-Hartsoeker's con-
endoderm, the lining of most of the digestive tract ception of human sperm in
such outgrowths of the digestive tube as the accordance with the theory of
preformation (1694).
liver, pancreas, and the lungs; and from the
mesoderm, the muscles, bones, blood system, and most of the urogenital
eyBtem.l
Baer also first discovered the human ovum, a tiny cell in the Graafian
follicle 7125 in. in diameter.
Ten years after Baer had given a firm foundation to embryological science by
theory of germ layers, a new task confronted it on the establishment of the
. . . .IIIIIIIIJa"I..... theory, in 1838. What is the relation of the ovum and the layers which
-._.. .,...., from it to the tissues and cells which compose the fully developed organism?
\ HAECKEL, E., "The Riddle of the Universe," p. 58, Harper & Brothers, New York,
The correct answer to this difficult question was given about the middle of this
century by two distinguished pupils of Johannes Muller-Robert Remak, of
Berlin, and Albert Ki:illiker, of Wiirzburg. They showed that the ovum is at
first one simple cell, and that the many germinal globules, or granules, which
arise from it by repeated segmentation, are also simple cells. From this mul-
berrylike group of cells are constructed first the germinal layers, and subsequently,
by differentiation or division of labor, all the different organs. Ki:illiker has the
further merit of showing that the seminal fluid of male animals is also a mass of
microscopic cells. Thus it was proved that both the materials of generation, the
male sperm and the female ova, fell in with the cellular theory.1
Forms of Reproduction.-From the standpoint of nature, there are
three main considerations for any succession of individuals: (1) The
individual must be born, (2) it must grow to maturity, and (3) it must
reproduce in order that the cycle may be endless. Reproduction is the
most vital consideration in nature. It simply cannot fail, and the adap-
tations to prevent failure are almost endless.
There are two general types of reproduction, viz., asexual and sexual.
ASEXUAL
Fission. Cell division.
Budding. A small part being pinched off.
Vegetative. Regeneration of a whole from a part.
SEXUAL
Parthenogenesis. Female hatching young from unfertilized eggs.

Paedogenesis. Reproduction by young or larval animals (may be
parthenogenetic) .
Hermaphroditism. Both sexes in one individual.
Unisexualism. Requiring combined function of two individuals of
opposite sex.
In all the higher animals, unisexuality prevails, the union of male and
female germ cells, produced by the two separate entities, being necessary
for the production of a new individual. All the marvelously specialized
organs, in both male and female, for accomplishing this result have had a
common origin, and, although the differentiation probably began millions
of years ago, there still remain many similarities between the male and
the female genitalia. The reproductive organs of one sex have their
homologues in the other, although each has been specialized for its respec-
tive function. The genital ridge, the first indication of sex organs ip the
embryo, is much alike in both male and female of the higher specie~for a
considerable time before specialization begin~.
lIbido
It will be the purpose in this and the following chapter to study the
principles underlying reproductive physiology. It is not the purpose to
attempt to develop veterinarians, however desirable that might be from
a breeding standpoint, but to impart to the breeder intimate knowledge
of the principles underlying reproduction, in order that he may practice
his profession wisely and understand and counsel with the veterinarian
when it becomes necessary to call upon the latter.
Embryological Development of the Genitalia.-The urogenital system
of the higher animals is formed for the most part from mesodermal tissue.
This double system, designed both to rid the body of certain waste prod-
ucts and to provide a means for reproducing the species, has had a very
interesting evolutionary development.
The earliest stage in this development results in the formation of the
paired pronephric tubules, or pronephros. This is the functional kidney
in Amphioxus and some lampreys. It forms in 7 to 14 somites toward
the cephalic end of the embryo from the intermediate cell mass of the
mesodermal somite (nephrotome). Mesially this forms an invagination
opening into the coelom, with which blood vessels communicate and waste
products transfuse into these invaginated areas. The free or distal end
of this section of each of these somites canalizes, the tubular ends bend
backward, anastomose, and form a duct that runs posteriorly and empties
into the cloaca, the pronephric duct. These structures are found in
early embryonic stages of larval fishes, amphibians, reptiles, birds, and
mammals but are transitional.
The second kidney formed is the mesonephros. As the primitive
pronephric tubules degenerate, there are formed behind them about 80
mesonephric tubules that empty into the already formed prone ph ric duct.
This second formation is known as the mesonephros or W olffian body and
serves the higher organism as a temporary excretory organ, although in
man it is entirely degenerate in a four-month embryo. The old prone-
phric duct is now called the mesonephric or W olffian duct. For some
larval fish and amphibians, the mesonephros is the permanent kidney.
The true kidney is the third stage in this evolutionary process. The
ureter and collecting tubules of the permanent kidney grow out from the
caudal end of the mesonephric duct, while the secretory tubules and
Bowman's capsule are formed from the caudal end of the nephrogenic
cord.
The growth of these tissues in the successive somites of the early embryo
causes a longitudinal ridge known as the urogenital ridge to form in the
body cavity on either side of the median line. This ridge soon divides
into a lateral mesonephric fold and a median genital fold. In the former
are lQcated the W olffian duct, already spoken of, and there later forms,
by invagination, another hollow tube, the Mullerian duct. The mesial

portion of the urogenital fold-the genital fold-is of peculiar interest
because it is destined to form either testicles or ovaries. This genital
fold in the early ''leeks of embryonic life extends from near the diaphragm
back toward the posterior end of the embryo. Later the anterior portion
of the body grows relatively faster, so that the fold assumes a more
caudal position, and this together with differential growth causes the
genital fold to assume a beanlike shape. Thus are formed the gonads,
or precursors of testes and ovaries.
The human embryo, therefore, at six weeks has a pair of undiffer-
entiated gonads and two sets of tubes, W olffian and Mullerian. Its sex
has been previously determined by the chromosome content of the egg
and sperm, but there has as yet been no differentiation of specific male
or female parts. Such differentiation begins to be noticeable at about
six weeks when the embryo is about H to H in. in length.
If the embryo is destined to become a male, the gonads complete their
development into testes, which pass down into the pelvic cavity and
about a month before birth into the scrotum. The Wolffian duct (old
pronephric) differentiates into the long, variously differentiated tube
which will eventually carry the spermatozoa from the testes out through
the penis. The Mullerian duct on the other hand atrophies and, except
for two small vestiges (at the top of the testicle and embedded in the
prostate gland), disappears.
If the embryo is destined to become a female, the gonads develop
into ovaries, the Mullerian duct into the uterine tubes, uterus, cervix,
and part of the vagina, while the W olffian duct, again except for two
small nonfunctional vestiges in the ovary, disappears. In some cases the
sex-differentiating mechanism fails to function properly and varying
incomplete stages of both sexes appear in the individual, popularly, but
erroneously, called hermaphrodites.
The gonad consists of a central core of connective tissue, or stroma,
with blood vessels and nerves and is covered with a layer of germinal
epithelium. Columns of cells (Pflugers egg cords) from the germinal
epithelium grow down into the stroma. In the male these cords give
rise to the convoluted seminiferous tubules that after birth canalize into
tubules in which the spermatozoa eventually are formed. The sus-
tentacular cells of Sertoli, or nurse cells, are formed within the tubules
from the undifferentiated epithelial cells of the sex cords, while the
interstitial cells that lie outside and around the tubules are formed from
the mesenchymal stroma. After the fifth week, a dense layer of con-
nective tissue, the tunica albuginea, develops between the germinal
epithelium and the sex cords, and the epithelium is reduced to a single
layer of flat cells.
In the female the process is very similar, except that the first group of
egg cords seems to disappear, after which another group of egg cords
passes down into the stroma of the ovary. These egg cords break up
and form the many thousands of immature follicles that are present in
the ovaries at birth.
It is seen from the above that the primordial germ cells are present in
an embryo at a very early age. Granted normal health, there is nothing
that a breeder can do to change them or their hereditary units in any way.
From a genetic point of view, an animal will transmit just the same when
it is a two-year-old as when it is a ten-year-old. vVe can trace these
early germ cells' precursors in the embryonic testes or ovaries back to the
original fertilized egg or zygote that gave rise to this particular embryo.
It is seen, therefore, that the manner in which an animal will transmit
is determined very early, actually when the specific egg was fertilized
by the specific sperm to produce this specific animal. The new being
got certain determiners of hereditary characteristics from both of its
parents. At sexual maturity, it will be ready to pass on a sample half of
this hereditary material to each of its offspring. After fertilization has
once taken place, the breeder is powerless to change the genetic make-up
of the developing fetus. His only chance for animal improvement comes
before this event, and his task is to devise ways of learning the genetic
content of the germ cells of his animals so that he may know what he
actually has from a genetic standpoint in the animals in his herd or flock.
The genital tubercle develops in the mid-line between the umbilical
cord and anus at about the third or fourth week of embryonic life. On
its caudal slope there appears a small urethral groove. In the male this
groove closes over, and it eventually extends all the way through the
penis and the glans. Rounded ridges (labio-scrotal swellings) appear on
each side of the base of the phallus and in the male hollow out to form the
two-chambered scrotum or sac in which the testes are carried. In the
female the phallus lags in development and eventually forms the clitoris,
while the labio-scrotal swellings form the labia majora, or external lips
of the vulva.
The Testis.-In mammals, with the exception of the whale, elephant,
rhinoceros, and seal which do not have scrota, the testes are present in
the scrotum at or shortly following birth. Their basic anatomy is estab-
lished during embryonic and fetal development, but their functiona!
development awaits the proper environmental and endocrine stimulation.
The testis is covered by a dense connective-tissue capsule, the tunica
by invagination, another hollow tube, the Mullerian duct. The mesial

portion of the urogenital fold-the genital fold-is of peculiar interest
because it is destined to form either testicles Of ovaries. This genital
fold in the early weeks of embryonic life extends from near the diaphragm
back toward the posterior end of the embryo. Later the anterior portion
of the body grows relatively faster, so that the fold assumes a more
caudal position, and this together with differential growth causes the
genital fold to assume a beanlike shape. Thus are formed the gonads,
or precursors of testes and ovaries.
The human embryo, therefore, at six weeks has a pair of undiffer-
entiated gonads and two sets of tubes, Wolffian and Mullerian. Its sex
has been previously determined by the chromosome content of the egg
and sperm, but there has as yet been no differentiation of specific male
or female parts. Such differentiation begins to be noticeable at about
six weeks when the embryo is about Ys to ~~ in. in length.
If the embryo is destined to become a male, the gonads complete their
development into testes, which pass down into the pelvic cavity and
about a month before birth into the scrotum. The Wolffian duct (old
pronephric) differentiates into the long, variously differentiated tube
which will eventually carry the spermatozoa from the testes out through
the penis. The Mullerian duct on the other hand atrophies and, except
for two small vestiges (at the top of the testicle and embedded in ~the
prostate gland), disappears.
If the embryo is destined to become a female, the gonads develop
into ovaries, the Mullerian duct into the uterine tubes, uterus, cervix,
and part of the vagina, while the Wolffian duct, again except for two
small nonfunctional vestiges in the ovary, disappears. In some cases the
sex-differentiating mechanism fails to function properly and varying
incomplete stages of both sexes appear in the individual, popularly, but
erroneously, called hermaphrodites.
The gonad consists of a central core of connective tissue, or stroma,
with blood vessels and nerves and is covered with a layer of germinal
epithelium. Columns of cells (Pflugers egg cords) from the germinal <-
epithelium grow down into the stroma. In the male these cords give
rise to the convoluted seminiferous tubules that after birth canalize into
tubules in which the spermatozoa eventually are formed, The sus-
tentacular cells of Sertoli, or nurse cells, are formed within the tubules
from the undifferentiated epithelial cells of the sex cords, while the
interstitial cells that lie outside and around the tubules are formed vom
the mesenchymal stroma. After the fifth week, a dense layer of ~
nective tissue, the tunica albuginea, develops between the germinal
...
THE MALE'S PART IN REPRODUCTION 103 •
epithelium and the sex cords, and the epithelium is reduced to a single
layer of flat cells.
In the female the process is very similar, except that the first group of
egg cords seems to disappear, after which another group of egg cords
passes down into the stroma of the ovary. These egg cords break up
and form the many thousands of immature follicles that are present in
the ovaries at birth.
It is seen from the above that the primordial germ cells are present in
an embryo at a very early age. Granted normal health, there is nothing
that a breeder can do to change them or their hereditary units in any way.
From a genetic point of view, an animal will transmit just the same when
it is a two-year-old as when it is a ten-year-old. We can trace these
early germ cells' precursors in the embryonic testes or ovaries back to the
original fertilized egg or zygote that gave rise to this particular embryo.
It is seen, therefore, that the manner in which an animal will transmit
is determined very early, actually when the specific egg was fertilized
by the specific sperm to produce this specific animal. The new being
got certain determiners of hereditary characteristics from both of its
parents. At sexual maturity, it will be ready to pass on a sample half of
this hereditary material to each of its offspring. After fertilization has
once-taken place, the breeder is powerless to change the genetic make-up
of the developing fetus. His only chance for animal irrwrovement comes
before this event, and his task is to devise ways of learning the genetic
content of the germ cells of his animals so that he may know what he
actually has from a genetic standpoint in the animals in his herd or flock.
The genital tubercle develops in the mid-line between the umbilical
cord and anus at about the third or fourth week of embryonic life. On
its caudal slope there appears a small urethral groove. In the male this
groove cle>ses over, and it eventually extends all the way through the
penis and the glans. Rounded ridges (labio-scrotal swellings) appear on
each side of the base of the phallus and in the male hollow out to form the
two-chambered scrotum or sac in which the testes are carried. In the
female the phallus lags in development and eventually forms the clitoris,
while the labio-scrotal swellings form the labia majora, or external lips
of the vulva.
The Testis.-In mammals, with the exception of the whale, elephant,
rhinoceros, and seal which do not have scrota, the testes are present in
the scrotum at or shortly following birth. Their basic anatomy is estab-
lished during embryonic and fetal development, but their functiona!
development awaits the proper environmental and endocrine stimulation.
The testis is covered by a dense connective-tissue capsule, the tunica
albuginea, from which connective-tissue extensions, the septa, radiate

and divide the testis into lobules. Within each lobule are the seminif-
erous tubules, long coiled ducts in which the spermatozoa will eventually
be produced. The seminiferous tubules converge at the apexes of the
lobules where they straighten out and become the tubuli recti, and in
FIG. 30.-Generative and urinary organs of bull. 1, scrotum; 2, spermatic chord; 3, testi-
cle; 4, epididymis (globus major and minor); 5, vas deferens; 6, vesicula seminalis; 7, mem-
branous portion of urethral canal covered by W'ilsou's muscle; 8, part of prostate gland
covered by Wilson's muscle; 9. Cowper's gland; 10, accelerator urinae muscle; 11, penis;
12, cut suspensory ligaments of penis; 13, sheath, laid open; 17, ureters. (From U.S.
Department of Agriculttt1'e, Diseases of CaUle.)
turn enter the rete testis, a system of inegular anastomosing cavernous

spaces. From the rete testis there is a series of ductules (about 20 in
man) which emerge on the surface of the testis and enter the head of the
epididymis. Around and between the seminiferous tubules there are
numerous blood vessels, nerves, and connecti~e tissue and the inter-
stitial or Leydig cells.
The testis has two distinct functions. The older, phylogenetically, is

probably the production of spermatozoa. This function is carried on in
the seminiferous tubules. The second function is the production of the
male hormone, testosterone. The male hormone is necessary for the
development and function of the male genital organs and for such second-
ary sexual characteristics as sexual drive. facial hair in the human, and
FIG. 31.-Semischematic figure showing small segment of the wall of an active seminif-
erous tubule. The sequence of events in the production of spermia is indicated by the
numbers. A spermatogonium (1) goes into mitosis (2) producing two daughter cells (2a
and 2b). One daughter cell (2a) may remain peripherally located as a new spermatogonium
eventually coming to occupy such a position as lao The other daughter cell (2b) may grow
into a primary spermatocyte (3) being crowded meanwhile nearer the lumen of the tubule.
When fully grown the primary spermatocyte will go into mitosis again (4) and produce two
eecondary spermatocytes (5, 5). Each secondary spermatocyte at once divides again (6, 6)
producing spermatids (7). The spermatids become embedded in the tip of a Sertoli cell
7a), there undergoing their metamorphosis and becoming spermia (8), which when mature
are detached into the lumen of the seminiferous tubule. (From Patten, Embryology of the
Pig, 2d ed. , The. Blakiston Company.)
the brilliant feather patterns of some birds. The hormonal function is

carried on by the interstitial cells.
The mature active testis is a very busy and efficient organ. Since
millions and even billions of sperm are ejaculated at each copulation,
the surface area of the germinal epithelium is necessarily large. It has
been estimated that the seminiferous tubules of man, if laid end to end,
would extend about 1,000 ft: and in the bull nearly 3 mi. At birth the
tubules contain a basal layer of loosely arranged cells called
spermatogonia and the Sertoli or nurse cells to \yhich the sperm will
temporarily attach themselves during the final stages of development.

There is no central canal in the tubules at this time. As sexual devel-
opment progresses, the spermatogonia divide by ordinary mitosis and
form primary spermatocytes from which are derived secondary sperma-
tocytes, spermatids, and spermatozoa in their proper turn. As this
process progresses, the testis increases in size, the seminiferous tubules
increase in diameter, a central canal appears in each tubule, and the
FIG. 32.-Seminiferous tubule of mature rat testis. Observe the almost diagrammatic
arrangement of the cells and the abundance of all cell types.
sperm eventually pass through the tubuli recti into the rete testis, the
efferent ducts, and thence to the epididymis. .
In order that the number of chromosomes typical of the species may b~
maintained, and yet a means of variation provided, sperm proliferation
is accompanied by a process called miosis. The spermatogonia, or sperm
mother cells, grow and enlarge and may divide to form either other
spermatogonia or primary spermatocytes. The spermatogonium remains
at the wall of the tubule, while the other, the primary spermatocyte,
begins to move toward the lumen. This latter cell goes through two
rapid cell divisions. Previous to the first cell division synapsis (fusing
of like memhers of pairs of chromosomes) occurs.
~I.
~f
•. '
108 BREEDING AND IMPROVEMEST OF FARM AXIMALS
During synapsis each of the fused homologous chromosomes splits or

reproduces, each one thus coming to consist of two sister chromatids,
making a bundle of four chromatids, or the tetrad. The paired chroma-
tids twist about each other at this time, and sometimes a section of a
chromatid is replaced by a section from one of the members of the
homologous pair of sister chromatids. This interchange of chromatin
material between nonsister chromatids will be discussed fully in a later
chapter when we are dealing ,yith crossing over.
Synapsis is follo,ved by disjunction of the fused pairs of chromatids,
one pair now moving to each of the two poles of the cell. Upon the
completion of this migration of chromatids, the cell wall constricts,
forming two secondary spermatocytes, each with one-half the normal
_ _::....c-----Primary spermatocyte
secreted by wall of tubule
Reduction division
including synopsis
Secondory spermotocytes
with one-half the normal
number of chromosomes
Spermatids
~--1f---Spermatozoa ready to
travel out from the testicles.
10 fertl'lize an ovum
FIG. 34.-Schematic seminiferous tubule showing spermatogenesis.
number of chromosomes for the particular species, each in the form of

a fused pair of chromatids. This division is spoken of as the heterotypic
division, because the two resulting cells are not identical with the parent
cell. The two secondary spermatocytes now go through a homotypic
division, one of each of the original four chromatids being found in each
of the resulting four spermatids. The spermatids in turn become elon-
gated, the nucleus passes to one end, and the opposite end stretches out
into a long filamentous tail, the lashing of which provides the sperm with
its motility. The newly formed spermatozoa then become attached to
the large nurse or Sertoli cells, which project toward the lumen of the
seminiferous tubules. The fully developed spermatozoa are found even-
tually in the lumina of the tubules through which they make their exit
from the testes. \
Hormonal Regulation of the Testis.-The primary regulating agents
of the testis are the gonadotropic hormones produced by the anterior
pituitary gland. If these hormones fail to be produced in sufficient
amounts in the young animal, sexual maturity does not occur. Failure
'l'HE MALE'S PAR'l' I N REPRODUC'l'ION 109
of gonadotropic hormone production in the mature animal is followed by

both spermatogenic and endocrine dysfunction of the testis. There are
two gonadotropic hormones. One is called the follicle-stimulating hor-
IDone (FSH) and the other the luteinizing hormone (LH). These
gonadotropins also regulate ovarian function, and their names are actu-
ally descriptive of their effects in the female. Some workers refer to the
luteinizing hormone as the interstitial-ceil-stimulating hormone (ICSH).
Both hormones are proteins and have been isolated from the anterior
pituitary in relatively pure form. Both FSH and LH produce enlarge-
FlO. 35.-1. Immature fowl testis, composed primarily of spermatogonia. 2. ~1ature fowl
testis showing abundant germinal cells of all types including sperm.
ment of the testis when injected, but each acts in a different manner.
The development and maintenance of the seminiferous tubules and
spermatogenesis are dependent primarily on FSH. The interstitial cells
he testis, which secrete the male hormone, are stimulated by LH.
Although the male hormone is concerned chiefly with the accessory
structures of the male genitalia and the secondary sexual characteris-
tics, it may also assist in the development and function of the seminiferous
tubules. Thus it appears that the normal development of the seminifer-
OUS tubules depends upon FSH, LH, and the male hormone.
Age and Testis Function.-There are wide species, strain, and individ-
ual differences in the age at which spermatozoa are produced. Studies
at the Purdue Experiment Station (Hogue and Schnetzler, 1937) showed
that in the chicken full spermatogenesis was attained between the
eighteenth and thirtieth weeks and was then maintained more or less
continuously throughout reproductive life. The pattern of spermato-
genesis was somewhat different in the ring-necked pheasant, a species
which normally breeds in the spring. It was found that in some males
the testes contained sperm at about fourteen weeks of age but that the
testes then regressed to a resting stage until spring. Full spermato-
genesis was thus not attained until the males were about forty weeks of
age (Kirkpatrick and Andrews, 1944). Phillips and Andrews (1936)
found that sperm were present in bull testes at thirty-two weeks and in
the boar at twenty-one weeks. The testes of purebred, inbred, cross-
bred, and hybrid boars have been studied at Purdue (Andrews et al.,
1949). Sperm were consistently present in one of the Duroc lines as
early as 110 days and in another line did not appear regularly until 140
days. The age at which sexual maturity occurs is of considerable impor-
tance in the development of new lines of breeding. Delayed or incom-
plete spermatogenesis may reflect itself in lowered prolificacy and
fertility. Precocious sexual maturity may interfere with the growth
process, for the production of large amounts of the male or androgenic
hormone may cause the long bones to cease grmving at too early an age.
Seasonal Periodicity.-In most birds in this hemisphere the production
of both sperm and ova is strictly a seasonal phenomenon. The chicken,
however, has been carefully selected on the basis of reproductive capacity
for many years and is now regarded as a continuous breeder. Wheeler
and Andrews (1943) have shown, however, that in the Barred Rock the
greatest numbers of sperm per ejaculate are produced between December
and April. The domestic turkey and goose are definitely seasonal breed-
ers. In the majority of mammals sperm are produced only at certain
seasons of the year. Seasonal spermatogenesis is characteristic of wild
species, but man and the domesticated mammals have long been regarded
as continuous breeders. Only in recent years has it been recognized that
there are seasonal differences in the rate of spermatogenesis or in semen
quality, even though sperm are produced throughout the entire year. It
is assumed by many that sexual drive is indicative of the degree of
spermatogenesis and, for example, that the willingness of rams to mate
throughout the year is an indication that sperm are being produced.
This is not true, since sperm are formed in the seminiferous tubules, a
region which is anatomically separate from the hormone-secreting portion
of the testis. Sexual drive can be stimulated in the castrate male by the
injection of the androgenic hormone, but it is impossible for such an
animal to produce sperm. McKenzie and Berliner (1937) clearly showed
that in the ram, under normal Missouri conditions, spermatogenesis
and sp.men quality were maximum in the fall and minimum in J\lly
THE MALE'S PART IN REPRODUCTION III
and August. Shropshire rams were more affected by season than

Hampshires.
A 20-year study of breeding efficiency in the Purdue dairy herd showed
that fertility was maximum in March, April, May, and June and mini-
mum in July, August, and February. Part of this seasonal difference
,vas attributed to the effect of season on semen quality (Erb et al., 1942).
The lowest conception rate was obtained in the winter and spring in
three herds of cattle located at different latitudes in eastern Canada, and
the highest fertility occurred in the summer and fall (Mercier and Salis-
bury, 1947). A companion study of the seasonal variations in fertility
of 125,000 cows and 71 bulls in New York State showed that younger
and older cattle were most affected by season. The poorest conception
rate was observed in the winter. It was concluded that the amount of
daylight had a definite influence on fertility at that latitude and that
the response to light varied with the age of the animal (Mercier and
Salisbury, 1947).
Effect of Light and Temperature.-The most obvious characteristics
of the seasons are differences in light, temperature, and in some areas,
precipitation. Plants are especially responsive to these variations, and
florists make routine use of controlled light and temperature in the
production of out-of-season crops. Animals are probably not quite so
responsive to light as plants. However, the ancient Chinese are reported
to have known that canaries could be induced to sing out of season if
candles were placed in front of their cages. We know now that singing
was induced because of the stimulatory effects of light on sexual activity.
Rowan's (1929) work with the junco clearly showed that sperm produc-
tion could be brought about in ~ ovember and December by increasing
the length of day, even though the birds were at the same time subjected
to subzero temperatures. The normal mating season of the junco is in
May. Poultrymen have long made use of artificial light to alter patterns
of egg production. It is generally agreed that in chickens artificial light
usually does not increase the total number of eggs but makes possible
their production at a time of year when eggs are most valuable (fall and
winter). Turkeys can be induced to produce both eggs and sperm earlier
in the winter than normal by the use of supplementary illumination.
The effects of quality of light are incompletely known. Bissonnette
(1931) found that white and red light were most effective and that green
light inhibited testis function in the starling.
It is obviously more difficult to alter the environment of the larger
farm animals and the effects of light on spermatogenesis are incompletely
known. As previously pointed out, the seasonal differences in fertility
of cattle may be explained on the basis of light at some latitudes.
The effects of temperature on spermatogenesis are probably subject to

wide species variations. In birds, including the domestic fowl, it is
probably correct to generalize that light has much more effect on sperma-
togenesis than temperature. McKenzie and Berliner reported a marked
decline in semen quality when rams were placed in a warm room during
the winter. Mills (1939) has summarized data on seasonal variations on
human conception rate in various North American cities. In San Fran-
cisco, where there is very little seasonal variation in temperature, there
are practically no monthly fluctuations in conception rate. In Tampa,
Fla., and Charleston, S.C., there are significantly fewer conceptions in
July and August. Mills points out that human fertility is maximum
when the mean monthly temperature is between 40 and 65°F. Fertility
declines when the mean temperature rises above 70°F. or falls below 40°F.
In most parts of the United States high environmental temperatures
eoincide with maximum length of day, and it thus becomes difficult to
separate the effects of the two on animals maintained under normal
conditions.
It is clear that environmental alteration does affect the reproductive
process. The mode of action of the external environmental factors is at
least partially understood. The removal of the anterior pituitary gland
brings about a complete cessation of either testicular or ovarian activity
and seasonal reproductive cycles cease. The injection of the anterior
pituitary gonadotropic hormones reestablishes gonad activity. It
appears, therefore, that the anterior pituitary is the chief regulating
agent. The several possibilities by which environmental factors may
influence gonad function have been summarized by Turner (1948). In
some species, the ferret, the optic nerve must be intact if light is to affect
the anterior pituitary, whereas in the duck the optic nerve is not essential.
There are marked seasonal differences in general metabolism which may
conceivably affect the gonads directly or indirectly by means of the
anterior pituitary. Body heat production rises as environmental tem-
perature declines. Much of this is due to changes in activity of the
thyroid gland. In the human it has long been known that both hypo-
and hyperthyroidism are associated with lowered fertility, and there is
evidence that this may also be true of farm animals. Reineke (1946)
found that thyroprotein, a compound which contains the thyroid hormone
thyroxine, will correct certain types of infertility in bulls, and Shaffner
and Andrews (1948) reported that experimental hypothyroidism reduced
semen quality in the fowl.
The effects of season upon the quality of feedstuffs and general plane
of nutrition should not be overlooked. It is perhaps not a coincidence
that fertility is highest during the months when pastures are most Iuxuri-
ant. Throughout much of the United States the nutritive status of

animals declines during the late winter months, and the stimulating
effects of spring pastures on milk production, growth, fertility, ar~d gen-
eral health are well known. This is due to the fact that some of the
constituents of feeds, e.g., vitamin A, deteriorate during storage. Vita-
min A is essential for normal fertility, and pasture is one of the richest
source of this vitamin.
The Hormonal Functions of the Testis.-As previously stated, the
testis is concerned with the production of spermatozoa and the secretion
of the male hormone by the inter-
stitial or Leydig cells. Although the ..,0
testicular hormone was not identi-
fied chemically until 1935, Berthold
had suggested as early as 1849 that
the testis produced a substance \\'hich
was responsible for the development
of the secondary sexual character-
istics. The term male hormone is a HO
broad one and is not limited to a Androsterone
specific compound. Although occa-
sional reports have been made that
the testis produces more than one
hormone, the best evidence available
indicates that only one is secreted
by the testis itself. This testicular
hormone is called testosterone and is o
the most potent of the male hormones. Testosterone
The male hormones as a group are FIG. 36.-Structural formulas of two of
the most common male hormones.
called the androgens or androgenic
hormones. Testosterone, then, is a specific androgen and is produced by
the testis. The androgens are widely distributed in animal cells and
secretions. Androsterone, an androgen about one-sixth as potent physio-
logically as testosterone, is usually present in significant amounts in the
urine of males. .Androg~Ils can also be produced by the ovaries and the
l!4!:.~nal ~ort~!n the human female it is not uncommon for the adrenals
to secrete excessive~amounts of androgens and for such male character-
istics' as facial hair to develo.Q.. In the hen it is not unusual for rather
~e4__Blale s.econdary sexual -;;-hange;;- to occur. Extensive comb
growth, crowing, and male sexual behavior may be observed. These are
du; to increased androgen production, the source of which is usually the
rudimentary ovary of the fowl, or the adrenal cortex. It is thus clear
that the androgens are_pot limit.!l_Q _t&_mle.-8eX, that tissueR other than
the testis can produce them, and that the physiologic and anatomic
manifestations of sex may differ from the genetic sex of the individual.
Many androgens which do not appear as such in the body have been
synthesized in the laboratory.
In the intact organism, hormones are usually secreted by highly
specialized cells and are distributed by means of the blood and lymph to
other parts of the body where they produce their specific effects. In
general terms, the functions of the various hormones are to provide a
means of chemical coordination of the tissues and organs which make
up the organism as a whole. Testosterone enters the capillaries which
supply the interstitial cells and is transported to all body cells, Some of
these cells are very responsive to this particular hormone, e.g., the comb
of the cock, and tissues such as those found in the small intestine are
little affected. Exogenous androgens can be very effective if properly
administered. Most of the male hormone compounds are relatively
ineffective if administered orally. The most effective me_thods of ad:rnin-
i~tration are the subcutaneous injection of the androgen in oil solutiQp. or
in the form of dense pellets which can be implanted beneath the skin.
The Male Hormones and Secondary Sexual Characteristics.-The
androgens have many functions and somewhat diverse effects in the
normal male. In general, the effects of the male hormones are exactly
opposite to the changes which occur following castration. The testicular
hormone is essential for the normal growth, development, and fun-ction
of the male genital organs. The seminal vesicles, prostate and Cowper's
glands are dependent on this hormone. Withggt it they undergo very
littlB development and produce relatively little secretion. The~rowth
of the penis, the process of erection, and the development and mainte-
nance of the scrotum are all regulated by the androgens.
\ I- Sexual behavior, sex drive or libido, is dependent to a large extent on
.r,- the androgens. In-man and the c~m;~n farm animals sex dri;e is
present more or less continuously from puberty until senescence. As
previously explained, the _presence of sex drive does not indicate that
sperm are_ being produced. The ability to copulate is not necessarily
lost following castration. The psychological explanation for this is that.
behavior patterns, when once established, are not immediately lost even
when their physiological basis is changed. However, in species which
are strictly seasonal breeders, hormonal and spermatogenic activity of
the testes are closely correlated and sexual drive is seasonal in nature.
It has long been known that the characteristic male and female second-
ary sexual characteristics are dependent upon the gonads, since these
characters fail to develop following prepubertal castration. The second-
ary sex characters are highly developed and have been much studied in
birds. In some species, the English sparrow, the feather patterns are
not altered by castration or androgen injection. In the domestic fowl
genetic and hormonal factors are both involved and in general the male
~ dependent upon the male hormone. TheJ.:1lsk_§_of the boar, .1
horns and crest of the b.1lll, ..horns of the rarn, and comb of the cock. are
either reduced in size or lacking if castration is performed at an early a~.
Nehher masculinity nor femininity are absolute. The physiologist
explains this partially on the basis of relative differences in the sex
hormones, the psychologist on the basis of training and experience, and
the geneticist on the basis of gene composition. Whatever the mech-
anism, there is agreement that sex is relative. The ox is a much more
tractable animal than the bull, and this knowledge has been made use of
for centuries. The eunuch is characterized by a high-pitched voice, a
small chest, beardless face, and is often described as lacking in aggressive-
ness and intellect.
Miscellaneous Effects of the Male Hormone.-Hundreds of studies on
the effects of the androgens on various tissues and body functions have
been made. A few of the many effects of these compounds are therefore
mentioned. The androgens both stimulate and inhibit general body
growth. The effects vary with species and level of hormone dosage.
They may increase metabolic rate, increase red-blood cell formation in
castrate males, be concerned with pigmentation of the skin, baldness,
fat-deposition patterns, nervousness, and numerous other conditions.
They have been used experimentally in the treatment of such diverse
diseases as angina pectoris, cancer, mastitis, and ringworm of the scalp.
It is not implied that the androgens have primary effects in the situations
mentioned but that the chemical regulation of the organism by hormones
is truly a complex mechanism.
Castration.-The removal of the testes is one of the oldest of surgical
procedures. In ancient times, men destined for the priesthood were
castrated and prisoners of war and the vanquished were often thus
treated. The effects of castration depend somewhat on the age at which
the operation is performed. Prepubertal castration of the farm animals
produces: (1) permanent sterility, (2) increased skeletal development due
to prolonged growth of the long bones, (3) failure of normal development
of the genital organs, (4) absence or retardation of the secondary sexual
characteristics, (5) improved carcass quality because of increased fat
deposition and retardation of such characters as heavy shoulders and
neck, (6) absence or retardation of sex drive, (7) varying psychic changes
depending upon the species involved. After sexual and body growth
have been attained, the outward evidences of castration may be less
marked. The most usual effects are (1) atrophy of the genital organs,
(2) loss or diminution of libido, (3) increased fat deposition, (4) decreased
basal metabolic rate, and (5) decreased aggressiveness.
The term castration implies the surgical removal of the testes; how-
ever, several methods of securing the effects of castration without the
removal of the gonads are now practiced by livestock men. The most
common method in complete removal is to incise or remove the lower
portion of the scrotum and then remove the testes. In Europe, and in
parts of the United States, castration is frequently performed with a type
of instrument called the Burdizzo forceps. These are double-hinged
pincers capable of great pressure and are made in various sizes depending
on the type of animal involved. One spermatic cord at a time is located
in the scrotum and placed between the forceps without cutting the
scrotum. When pressure is applied, the spermatic cord and vas deferens
are severed by crushing. When properly done, this method is very
effective. The testicle immediately begins to atrophy because its blood
supply has been cut off and all the effects of castration are produced.
This method is called bloodless castration because the scrotum remains
intact. Infection is rare and difficulties with flies are avoided. A new
method, called elastration, was introduced to the United States from
Australia in 1947. Castration is accomplished by placing a special heavy
elastic band around the scrotum of lambs at a few days of age. The
pressure cuts off the blood supply to the testes and scrotal tip, and both
atrophy. This method is reported to have been favorably received by
Western sheepmen.
The castration of lambs and pigs by conventional methods is a simple
procedure and should be carried out by stockmen at the proper time.
The problem is more difficult in cattle and colts and should be performed
by a veterinarian, unless the owner has thoroughly learned the art.
The cardinal features are (1) sanitation: clean hands, clean instruments,"
and clean premises; (2) incisions which are large enough and located so
as to ensure complete drainage of the wound; (3) the complete removal
of both testes; (4) prevention of excessive loss of blood; (5) proper
restraint of the animal; (6) the prevention of undue excitement of the
animal.
Lambs are easily castrated by cutting off the tip of the scrotum with a
knife or large pair of shears. When pressure is applied to the scrotum,
the testes are easily grasped. In young lambs they are best removed by
pulling with gentle pressure until the spermatic cord breaks. Pigs are
usually held on their backs by an attendant. Two incisions the length
of the scrotum about Yz in. from the median lines are made, the testes
squeezed out and removed by pulling until the cord breaks. Cutting the
spermatic cord with a sharp knife will cause excessive bleeding. In young
THE J[ALE'S PART IN REPIWDUC'1'ION 117
animals it is best severed by traction. In older animals it should he

crushed with an emasculatome.
If recently castrated animals are confined to dirty premises, infection,
sometimes tetanus, may occur. In warm weather a fly repellant should
be used.
Lambs and pigs are usually castrated at two to four weeks of age,
calves at six to twelve weeks, and colts at about one year of age.
Scrotum.-The testes are formed in the abdominal cavity, and in some
species remain there throughout life. In the course of evolutionary
development they gradually came to lie at the posterior end of the body
pressing against the body wall. A continuation of this process finally
led to an outpouching at this region forming a sac, the scrotum, outside
the body proper. It is here that the testicles are found in mammalian
farm animals. In many other animals (e.g., birds) they remain in the
abdominal cavity permanently, whereas still others (e.g., woodchuck)
show an intermediate phase with the testicles descending into the scrotum
during the breeding or rutting season and being withdrawn into the body
cavity when this season is completed.
In addition to serving as a covering for the testicles, the scrotum also
functions as a thermoregulatory mechanism. This function of the
scrotum is shown by actual differences in temperature, as demonstrated
by Moore and Quick (1924) and Phillips and McKenzie (1934), the
temperature in the scrotum being from 1 to goC. lower than that of the
abdominal cavity. It has been shown by several workers that insulation
or the application of heat to the testicles results in a degeneration of the
spermatogenetic tissue, the production of abnormal sperm, and temporary
sterility. These results are also known to follow periods of intense fever.
Experimental transposition of the testes to the abdominal cavity leads
to the same result, and it has been shown that cooling of the abdomen
after transposing the testicles retards the rate of spermatogenic degenera-
tion. If this degeneration has not been too severe or prolonged, the
testicles recover their spermatogenic function when all conditions return
to normal. Moderate lowering of the temperature of the testicles has
been shown to have no deleterious effect on spermatogenesis.
The thermoregulatory function of the scrotum is accomplished through
the medium of the tunica dartos muscle, as shown by Phillips and
McKenzie (1934). This muscular coat consists of fibroelastic and smooth
muscle and consists of two layers, an outer layer applied closely to the
skin of the scrotum and a 4eavier inner layer. This muscle is highly
sensitive to changes in temperature, drawing the testicles close up to the
body at low temperatures and reversing the process at high temperatures.
As shown by Phillips and Andrews (1936) the development of the dartos
118 BREEDING AND IMPROVEMENT OF FAR.
anatomically and physiologi~ally parallels the develop

especially their spermatogemc development. 'W ork at P the t
.
that castratIOn 0
f prepu bertal rats prevents the develop ~h
estes'
thermoregulatory function, and castration of mature rats is fo oWn
the loss of the function (Tyrell et al., 1942). The injection of testo e
either produced or maintained the thermoregulatory function, depen
on the age at which castration was performed (Almquist and Andrew,
1944).
In other words, the gradual transition of the incompletely developed
seminiferous tubules of a newborn animal into fully formed tubules
endowed with the power of producing mature viable spermatozoa, which
occurs during the first few months post partum, is accompanied by the
development of the dartos and its transition into a functional mechanism
for keeping the temperature of the testicles within physiologically desir-
able bounds.
If the testicles remain in the abdominal cavity in any species where
they should normally descend into the scrotum, a condition known as
cryptorchidism, they usually lack their spermatogenic function but retain
their endocrine function. In some animals, e.g., the elephant and the
anteater, the testicles remain in the abdominal cavity; whereas in others,
e.g., most of the rodents and the hedgehog, they descend periodically
during the breeding season into the scrotum, whence they are again with-
drawn into the abdominal cavity at the end of the breeding season.
In the stallion, bull, boar, and ram the testicles communicate through
the inguinal canal with the pelvic cavity, where the accessory organs and
the glands are located.
Any malformation of the genital organs should be regarded as a serious
or disqualifying defect of a herd sire. If an animal is bilaterally crypt-
orchid, he will usually be sterile. Fertility may be relatively normal if
one testis is present in the scrotum. It is well known that cryptorchidism
can be inherited, therefore any animal with this characteristic, or from
a line where cryptorchidism is prevalent, should not be used for breeding
purposes. In commercial livestock production, the occurrence of crypt-
orchidism increases the difficulties of castration, results in decreased
carcass quality when one of the testes is missed at the time of castration,
and decreases the value of purebred animals that otherwise might be
sold as breeders.
Sometimes the ing~al canal fails to close properly and allows the
viscera to enter the s<l'fptum. This condition is known as scrotal hernia
and is known to be heritable.
Epididymis.-The convoluted seminiferous tubules converge at the
mediastinum testis (absent in Equidae) and there join severally to form
from 6 to 24 efferent ducts. These in turn unite into one duct, thus
giving rise to the epididymis, a greatly coiled tube varying from a few
feet to several hundred feet in length in some species. The epididymis
is lined with tall columnar, ciliated epithelial cells surrounded by a muscu-
lar wall, and this in turn is covered by the serosa. One arises near the
top of each testis where it is coiled to form the head of the epididymis.
It eventually passes down the side of the testis, in a part called the body,
to the bottom of the testis where it forms the tail of the epididymis.
This tube continues the passageway for the egress of spermatozoa from
the seminiferous tubules. The passage of sperm through the epididymis
requires 4 to 7 days in the rabbit and 5 to 6 days in the ram (Anderson,
1945). It appears that passage is accomplished by the ciliated cells in
the vasa efferentia and epididymis, by testicular pressure and by contrac-
tions of these structures which occur during ejaculation.
It is well known that spermatozoa recovered from the testis do not
have the same capacity to fertilize ova which sperm in the vas deferens
have. The work of Young (1929) clearly showed that a process of
maturation of the spermatozoa occurs in the epididymis. Sperm recov-
ered from the head of the epididymis produced approximately half as
many conceptions as those taken from the tail region. Many studies on
epididymal function have been conducted in recent years. It is generally
agreed that the epididymis provides the most favorable medium for
the maintenance of sperm viability. The accumulation of harmful
metabolites is apparently at a minimum, and the reaction is such that
the sperm mature and remain in a resting state until ejaculated. It has
been reported by several investigators that epididymal bull sperm are
more resistant to cold shock than are ejaculated spermatozoa (Lasley
and Mayer, 1944). The functions of the epididymis are (1) a passage
and storage place for sperm, (2) the secretion of a small amount of fluid
which contributes to the semen, (3) maturation of the spermatozoa. The
maintenance of normal epididymal function is dependent upon the
presence of the testis hormone. In the absence of the testes, epididymal
sperm have a reduced length of life as measured by the maintenance of
sperm motility.
Vasa deferentia.-These tubes are a continuation of the epididymides,
continuing the passage for the spermatozoa to the urethra. The walls
of the vasa deferentia are thick and consist of layers of mucosa, sub-
mucosa, muscularis, and fibrosa or serosa. The lumina are small, giving
the tubes a firm cordlike character. They are lined with ciliated colum-
nar epithelial cells. In some species there is an enlargement (ampulla
of Henle) near the termination of each vas deferens. This is absent in
the boar, quite small in the dog and cat, and large in the bull, ram, and"':
being over 1 in. in diameter and often exceeding 6 in. in length in Equidae.
Glands of the vas deferens add to the total content of the semen. The
vas deferens, together with longitudinal strands of smooth muscle, blood
vessels, and nerves all encased in a fibrous sheath make up the spermatic
cord (two of them) that passes up through the inguinal canal into the
pelvic cavity.
Sterility can be produced by cutting or tying the vasa deferentia.
This operation, vasectomy, is legalized in some states for the sterilization
of the feeble-minded, criminally insane, and certain other conditions.
Sterility results because sperm are prevented from entering the urethra.
Spermatogenesis continues in the testes and the hormonal function is
normal. Because some glands, c.g., the pancreas, atrophy if their
excurrent ducts are closed, it was hypothesized that following vasectomy
sperm formation would eventually cease. It was lil~e,\'i3e reasoned that
the atrophy of the seminiferous tubules would allow the interstitial cells
to produce more hormone and that sexual reiuvenation could be brought
about. Information which would tend to disprove this was available
as early as 1830, but the so-called rejuvenating operation was widely
performed in Europe in the 1930's. Experiments on many species have
clearly shown that sperm formation does not cease. The sperm which
are produced undergo resorption within the seminiferous tubules and
the epididymis.
Ejaculatory Ducts.-These are formed through a union of the vas
deferens and the proximal portion of the seminal vesicle on each side.
They open into the floor of the urethra through small slitlike apertures
between which is found, in man, the uterus masculinus, which is the
homologue of the uterus and vagina of the female. In some species, c.g.,
the boar, there are separate openings for the vasa deferentia and the sem-
inal vesicles into the urethra. On the floor of the prostatic portion of tL.e
urethra is found a small elevl1tion known as the crista urethr~, or caput
gallinaginis, ·which contains erectile tissues and thus prevents the semen
from passing backward into the bladder at time of service.
Seminal Vesicles.-These are the largest of the accessory glands of
reproduction in the male. They are located in the pelvic cavity at the
ends of the vasa deferentia, being an outgrowth of the latter at their
urethral end, and each connects with the urethra by means of a duct.
McKenzie, Miller, and Baugess (1938) report as follows 1 on the seminal
vesicles of the boar:
The glands are tortuous, elongated, hollow bodies, with very irregular, branched
lumina and numerous out-pocketings. The wall consists of a thin external con-
I McKENZIE, F. F., MILLER, J. C., and BAUGEss, L. C., The Reproductive Organ8
and Semen of The Boar, Mo. Agr. Expt. Sta. B1il. 279, 1938, p. 36.
THE MALE'S PARI' IN REPRODUG1'IOlv' ]21
nective tissue sheet, of a thin middle layer of muscle, and of a mucous membrane
resting upon a thin submucous layer. The mucous membrane forms an elaborate
. system of thin, high primary folds which branch into secondary and tertiary folds.
These project far into the lumen, anastomosing frequently with one another,
thereby forming many irregularly shaped cavities of different sizes. These
cavities are separated from one another by thin branching partitions, and all
opening into a larger cavity.
The epithelium lies on a thin vascularized connective tissue supported by
muscle strands. Although showing some variation, the epithelium is simple
columnar in nature with some areas pseudostratified. The nuclei are round or
oval shaped and located at or near the base. Secretion granules are present
above the nuclei and on the free surface, drops or bulblike formations appear.
These are cast into the lumen, forming the secretion product. The fluid has a
gray, opaque color, a medium viscosity, and a pH of approximately 6.7.
There is no central duct of the seminal vesicles. Instead, there are several
large ducts branching and anastomosing irregularly, which finally converge into
one excretory duct. The duct from each seminal vesicle enters the urethra as a
slit-like opening, close to, but ventro-lateral to the vasa openings.
The seminal vesicles vary in size according to the animal and the
species, being in the horse 6 to 8 in. long and about 2 in. in diameter.
They are lacking in the dog and cat. The function of the seminal
vesicles, although not fully understood, is thought to be the secretion
~ fluid that furnishes a medium for transport of the spermatozoa.
This secretion is thick, alkaline, and globulin-containing. The vesicles
;"vere formerly thought to serve as a storage place for s ermatozoa hence
,their name, but this t eory has hyep di§p~. The seminal vesicles
are not essential for reproduction as evidenced by the fact that their
removal does not produce sterility.
Prostate.-At the neck of the bladder, surrounding or nearly surround-
ing the urethra and connecting with it by two rows of openings on either
side of the ejaculatory ducts, is the prostate gland. This gland is a
composite of many small compound tubuloalveolar branching glands.
McKenzie et al. (1938) studied the prostate in the boarl and wrote:
The prostate gland in the boar is a yellowish multilobular gland. It is com-

posed of two parts; the body or bulb, closely attached to the dorso-Iateral surface
of the pelvic urethra at its anterior extremity, and the pars disseminata which is
attached to the body but embedded in the wall of the pelvic urethra beneath the
body of the gland. The body of the gland weighs from 15 to 25 grams and
measures about 5 cm. in length by 4 cm. across by 0.5 cm. thick. The ratio of
prostate body weight to the live weight of the animals was about 1 :8,000.
1 Ibid., p. 42.
122 BREEDING AND IlvIPROVEMENT OF FARM ANIMALS
The glandular tissue is firm, without apparent storage space and contains no
secretion which can be expelled by pressure. Incomplete sections which have
not been made in a known direction, present a bewildering mass of secretory
tubules, follicles, connective tissue, smooth muscles, blood vessels, nerves and'
lymphatics. The gland is a composite of many small compound tubulo-alveolar
glands, giving rise to numerous excretory ducts which open into the urethra
independently on its dorsal wall. ,
The prostate is very irregular in form. Large 'branching cavities, narrow
ducts, and alveoli appear to be massed together in an irregular manner. There
is no distinct basement membrane and the glandular epithelium rests upon a
layer of connective tissue. The epithelium varies from simple or ps~udostratified
columnar in the smaller alveoli to cuboidal or even squamous in the larger cavi-
ties. Numerous secretory granules can be seen in the cytoplasm, and cytoplas-
mic drops appear to be attached to the free end of the epithelial cells.
There is abundant interstitial tissue which appears to consist of dense connec-
tive tissue with collagenous fibers and elastic network, and many smooth muscles
arranged in strands of varying thickness. There is also a connective tissue
capsule about the periphery of the gland.
The secretion of the prostate is viscid and contains proteins and salts.
The prostatic secretion is alkaline, tends to cleanse the urethra prior to
and at the time of ejaculation, and provides bulk and a suitable medium
for transportation of the spermatozoa. In some animals (rodents) the
abundant secretion of the seminal vesicles is coagulated in the vagina by
an enzyme in the prostatic fluid, and thus the ejaculate of the male
becomes a more or less solid plug in the vagina of the female.
Cowper's Glands.-Cowper's glands (bulbo-urethral), two in number
and corresponding to the glands of Bartholin in woman, are situated in
the urethral muscle (except in the boar) on either side of the pelvic por-
tion of the urethra, with which each communicates by means of a small
duct. They are lacking in the dog and very greatly developed in the
boar, a species in which they may extend the entire length of the pelvic
floor and reach a diameter of 1% in. The function of these glands and
the urethral glands is thought to be that of producing an alkaline secre-
tion for the purpose of neutralizing and cleansing the urethra before the
passage through it of the main bulk of the semen.
Penis.-The remaining portion of the male genitalia consists of the
penis. This organ, besides conveying urine to the exterior, has the
additional function of conveying spermatozoa into the female genital
tract. The penis is made up of muscular and erectile tissue that becomes
engorged with blood during the process of erection.
Farm animals exhibit a wide variation in the structure of this organ.
In the horse, the end of the penis is rounded, and the organ nearly fills
the vagina ~t copulation. In the bull, the diameter is much smaller, with
\
'THE MALE'S PART IN REPRODUCTION 123
the external urethral orifice situated in the urethral papilla at the end and
. on the left side. The urethral papilla of the bull represents a vestigial
filiform appendage. This is well developed in the ram in the form of a
slender appendage, which protrudes beyond the glans penis in a twisted
manner. The penis of the boar is not provided with an appendage or
lateral papilla, and the opening to the exterior is situated at the end of the
organ and in the center.'
Friction on the glans penis or on the homologue in the female, the
clitoris, finally leads to a sexual orgasm during copulation that is in part a
reflex action, for it can take place after the spinal cord has been tran-
sected. Ejaculation by the male is effected by a series of muscular con-
tractions beginning in the vasa efferentia and involving the epididymis,
vasa deferentia, seminal vesicles, prostate, and Cowper's glands. The final
discharge of the semen is brought about by the rhythmical contractions
of the bulbocavernosus and ischiocavernosus muscles, which, proceeding
progressively in waves, result in the ejaculation of the semen from the
~xternal orifice of the penis.
N ow, in resume, it may be said that the essential organs of reproduc-
tion in the male are the testes, whose functions are to produce sperma-
tozoa and the male hormone; the accessory glands, the prostate, the
seminal vesicles, the Cowper's glands, whose function it is to furnish a
medium for the spermatozoa to travel in on their way to impregnate the
female; and the epididymides, vasa deferentia, and urethra, the parts of a
long tube that is concerned with the maturation and passage of the
spermatozoa. The blood vessels, nerves, and muscles function in bring-
ing about erection and ejaculation.
Physicochemical Properties of Semen.-Semen is the entire dis-
charge of the male during normal ejaculation. It consists of two general
portions. The cellular elements, the spermatozoa, are produced by the
testis. The liquid portion of the semen, the plasma, consists of the secre-
tions of the seminiferous tubules, epididymides, vasa deferentia, seminal
vesicles, prostate, and Cowper's glands and the diffuse glandular cells
which are present in certain parts of the urethra. As shown in Table 9
there are great differences in the volume of semen and numbers of sperma-
tozoa in the different classes of animals. When semen was centrifuged
and the volume of packed cells determined, it was shown that approxi-
mately 15 per cent of fowl semen, by volume, is composed of sperma-
tozoa. Bull semen contains about 10 per cent sperm by volume (Shaffner
and Andrews, 1943).
Thousands of chemical determinations have been made on the semen of
various species and on the secretions of the various parts of the genital
system. One 1)£ the chief purposes of such determinations has been to
gain information which would lead to the development of suitable artificial

media for sperm. This approach has been of value but has many limita-
tions. As biochemists well know, the determination of elements and
simple compounds tells very little about the organization of complex
biological systems.
TABI,E 9.-S0ME QUANTITATIVE DATA ON SEMEN AND SPER~IATOZOA *
Volume per
Sperm per
ejaculate, in Total sperm
Animal cubic
cubic in ejaculate
millimeter
centimeters
Boar ............. ' . . .... . .... I 200 100,00~ 20,000,000,000

Stallion .............. ..... . . . 100 60,000 6,000,000,000
Bull ...................... ... 3-4t 800,000. 3,000,000,000
Ram ..................... ... . 0.8 1,000,000 800,000,000
Dog ......................... 7.0 5,000,000t 35,000,000,000t
Cock ......................... 0.6 3,000,000t 1,800,000,000+
* Adapted from Table 3. p. 36, U.S. Dept. Agr. Ctr. 567.
t About 1 teaspoonful.
:j: Estimates.
Numerous studies have been made of the heads of spermatozoa which

have been separated from the semen, washed, and the tails removed by /
centrifugation. It is known that the sperm heads contain the basic
protein, protamine, and certain inorganic bases in combination with .
nucleic acids. It is likely that these basic proteins are similar to the
composition of the genes themselves, and it is also likely that it will be '
many years before their structure and composition are understood.
The semen itself is chiefly water, the amount differing with species and
the relative concentration of sperm per unit volume of semen. The
amount of water varies from about 90 per cent in man to about 98 per
cent in the dog.
One of the limiting factors in the survival of single cells, such as blood
cells, sperm, or bacteria, is the osmotic pressure of the medium. It has f.::
long been known that this is remarkably constant in fluids such as blood
and milk, and in recent years many studies of the osmotic pressure of
semen have been made. The most common method of doing this is by
measurement of the depression of the freezing point; the temperature at ' .
which semen freezes is compared to the freezing point of water. The
average depression of bull semen is 0.609°C.; ram semen, 0.641 °C.;
stallion semen, 0.595°C.; boar semen, 0.616°C.; and rabbit semen,
0.574°C. (Anderson, 1945). By the use of such methods Salisbury et al. ..
(1948) have shown that the osmotic pressure of fresh, normal bull semen
THE MALE'S PART IN REPRODUCTION 12:")
is similar to that of cattle blood and that the formula of the sodium ei.trate
buffer which would be isotonic with blood is different from that which
has been widely used in the preparation of yolk-citrate dilutor for semen.
To be isotonic with blood, heated sodium citrate solutions should contain
2.9 g. NaaC6H507.2H20 per 100 ml. of glass distilled water. The actual
preparation of this dilutor will be discussed in a subsequent chapter.
The determination of the pH and buffering capacity of semen has been
TABLE lO.-CHEMICAL COMPOSITION OF ~ORMAL HUMAN SEMEN, PROSTATIC FLUID,
AND SEMINAL VESICLE SECRETION*
Semen Prostatic fluid Seminal vesicle fluid
No. \ High \ Low IAv. No.1 High Low I I Av.\ No·1 High Low I I
Av.
Per liter of fluid
pH ........................... 9 7.36 6.9 7.19 3 6.6 6.33 6.45 2 7.32 7.26 7.29
Water, g .•.......••... " ...... 13 944 891 918 5 936 927 932 2 900 880 890
Sodium, mM .................. 14 133 100 117 5 158 149 153 1 103
Potassium, roM ............. 12 27.4 1722.9 6 61.4 28.7 48.3 2 21.2 14.3 17.8
Calcium, mM .............. 3 7.15 5.3 6.22 3 32.7 28.7 30.2
Total CO" mM ........... 7 33.2 19.2 24 3 5.4 3.1 4.2·
Chloride, mM .............. ... 31 57.3 28.3 42.8 8 46.1 34.8 38.1
Acid soluhle phosphorust ... 8 32.3 17.2 23.8 1.77 0.651 09 7 19.8 9.65 14.7
Specific gravity ................ 6 1039 16/
1031 1035 14 1027 2 1038 1036 1037
101811022
Per 100 ce. of fluid

I
Total nitrogen, mg ............. 34 1225 560 913 14 511 295 41G 3 1343 1233 1284
Nonprotein nitrogen, rug ........ 12 130 73 96 6 90 3053.6 1 .. .. .... 99
Total protein by difference. g .... 12 6.85 3.29 4.50 6 2.93 1.66 2.17 1 .... .... 7.78
Total protein, grav-inletric, r, ..... 11 7.74 4.30 5.80 2 2.64 2.46 2.55 1 .... ....
9.04
Globulins, g .......... ........ 6 I- 2.43 0.76 1.20
Glucose,t mg .................. 6 369 203 295 12 48 Trace 16.4 5 625 275 390
Ascorbic acid, t mg ...... ..... U .. . .... 12.8 1U . ... .....
0.54 9 . .. . ... 4.66
Inorganic phosphorus, § rug ...... ., .... .... 45
Spermine phosphorus, § mg .. , ... .. .... .... 22.5
Urea,§ mg ......... . . .. ..... .. . , .. . ... 72
Lactic acid, § mg. ..... ..... . .. ., .., . .... 95
Cholesterol, § mg. " .... .. .... . ... 80 .. 618 86 Ii
I
* Huggins, C., Scott, W. W., and Heinen, J. H., 1942.

t Data of Huggins and Johnson (1933).
:I Data of Berg, Huggins, and Johnson (1941).
§ Data of Goldblatt (1935).
II Data of Moore, Miller, and McLellan (1941).
used routinely in semen investigations. The pH of semen is affected by a

wide variety of factors such as species, temperature, age of semen, rela-
tive amounts of fluid from the different accessory glands, frequency of
service, season of the year, and numerous others. In farm animals the
.,
pH is seldom less than 6 or more than 8 and tends to be confined to a

narrow range around 7. The average pH of bull semen during a calendar
year was 6.87 and was 7.04 for fowl semen during a similar period (Indi-
ana Station). Smith and Asdell (1940) showed that bull semen is well
buffered in the pH range 4 to 5.5 and 9 to 10, but relatively poorly
buffered between pH 6 and 9.
The most common inorganic constituents of semen are sodium, potas-
sium, calcium, magnesium, phosphorous, and chlorides. The actual
amounts vary widely with species and the amounts of secretion from the
various accessory glands. For detailed analyses in the boar the reader
is referred to McKenzie, Miller, and Baugess.
Because of the importance of glucose and lactic acid in ordinary cellular
metabolism, these organic constituents of semen have received the most
attention. The wide range in these compounds reported by most investi-
gators suggests that they may be influenced by many factors. One of
the most recent studies of ram semen showed that the glucose content of
26 samples from 7 rams was 516.3 mg. per cent at the time of ejaculation
with a standard deviation of 138.4 and a coefficient of variation of 26.8
per cent. The lactic acid content of fresh semen was 120 mg. per cent
(Moore and Mayer, 1941). As summarized by Anderson (1945), bull
semen has been reported to contain 100 to 495 mg. per cent glucose and
40 to 75 mg. per cent lactic acid and stallion semen, 0 to 116 mg. per cent
glucose and 4.6 to 71.5 mg. per cent lactic acid. The production of acid
from glucose or other substrates occurs rapidly in semen following ejacu-
lation. This process is called glycolysis and is apparently due to the
action of cell enzymes. At ooe. glycolysis is practically absent, but
goes on rapidly at body temperature. Although there is a correlation
between rate of glycolysis and sperm motility, this method of evaluating
the actual fertility of semen has not yet been of practical importance.
Semen has been reported to contain from 1.26 to 6.8 g. of total protein
per 100 mI., depending on the species. In the human the greater part of
the seminal proteins are proteoses. About 60 per cent of the proteins
can be dialyzed through cellulose membranes impermeable to blood
serum proteins. About 30 per cent of the total protein in human semen
is globulin. Huggins et al. (1942) reported that in the human, semen
protein is decreased by frequent ejaculation and subnormal male hormone
production.
Anatomy of the Spermatozoa.-Spermatozoa were first observed by
Ham, a student of Leeuwenhoek, in 1677, but their significance in
inheritance was not realized at that time. About the middle of the
eighteenth century, Spallanzani, by filtering semen, showed that the
spermatozoa were the essential factor in reproduction, for the filtrate
THE MALE' S PART I N REPRODUCTION 127
proved to be impotent. Shortly after Kalliker had shown that the

spermatozoa arise from the testes, Barry, in 1843, observed the conjuga-
tion of the sperm and egg in the rabbit.
The size of spermatozoa varies considerably in various animals but in
any case they are extremely minute, that of the
human being 7'SOO in. in length. Savage and
"his ~workers state the head length of sperma-
tozoa of the bull to be 9.7 p. and those of the
stallion 6.16 p.. Phillips reports the head length EAD
of ram spermatozoa to be 8.15 p.. After
puberty, spermatozoa are being continually
produced in the testicles of the higher mammals
at varying rates, and those not ejaculated are
resorbed.
There are many species differences in sperm
morphology. The general pattern in man and
the farm mammals is very similar. Normal
sperm consist of a head, neck, body, and a tail
portion. The head, in profile, is flattened, but
as ordinarily seen under the microscope is
rounded or oval. It consists chiefly of nuclear
material and is covered by a thin layer of
cytoplasm. A caplike structure, the acrosome,
is frequently observed on the anterior surface
of the head. The head is of great importance
because it contains the chromosomes.
The neck is very short and contains the ante-
rior centrosome. The body is about the same FIG. 37.-Generalized anat-
omy of the spermatozoon.
length as the head. It contains the posterior
centrosome at the neck end. A fibrillated filament extends from the
posterior centrosome to the end of the tail. There are several sheaths
around this filament in the body of the sperm. The tail is made up of
two regions. . The thicker portion is called the chief piece and the remain-
der, the end piece, is the bare continuation of the axial filament. The
movement of the sperm is lilrought about by the lashing of the tail.
Physiology of the Spermatozoa.-Although spermatozoa are relatively
aUnple cells morphologically, much is yet to be learned about their
physiology·.
The function of the spermatozoa is the fel'tilization of the ovum. The
process of fertilization ordinarily takes place in several distinct steps:
il) insemination, the deposition of the sperm in the female genitalia,
the movement or tr~nsport of the spermatozoa to the Fallopian tubes,
(3) the penetration of the cell wall of the ovum by a spermatozoon, (4)
the loss of the body and tail piece of the sperm, and (5) the union of the
male and female pronuclei. The process of fertilization is not complete
'.Intil the pronuclei of each gamete have united and the species number of
1. Human. 2. Sheep. 3. Chicken. 4. R!tt. 5. Guinea pig.

}'IG,38.-Species differences in spermatozoa.
FIG. 39.-Normal bull sperm. :.\Iagnification approximately 200X.
chromosomes has been restored. It is generally said that a sperm has

two functions, viz., to contribute the paternal set of chromosomes and to
initiate cell division on the part of the ovum. It has long been known
that in some of the lower organisms, cell division can be mechanically
induced. In this case the egg develops parthenogenetically and exhibits
THE MALE'S PART IN REPIWlJUGTION 129
only the maternal characteristics. Pincus and coworkers have been able
to induce cell division and embryological development in the rabbit by
altering the environment of the rabbit ovum.
Sperm Motility.-The most outstanding characteristic of spermatozoa
is their ability to move. It is generally accepted that sperm which have
lost their motility are incapable of fertilizing an ovum. The presence of
motility, however, does not guarantee that the fertilizing capacity still
exists. It is likewise well known that the more rapidly sperm move the
shorter their length of life. Some of the basic problems in the preser-
vation of semen for use in artificial insemination are the reduction of
sperm motility during storage, the maintenance of the capacity for
motility, and the reestablishment of vigorous activity after the sperm have
been introduced into the female genitalia. It is not unusual for the
motility of bull spermatozoa to persist for 3 weeks under laboratory con-
ditions, but it is very unusual for conception to occur in cattle with
semen which has been stored for 10 days. Shaffner (1942) was able to
restore motility in fowl sperm which had been held at -79°C. for 14
months. Although motility was excellent, fertility was not induced.
Sperm Survival in the Female Genital Tract.-The fact that there are
very great species differences in sperm survival in the female genitalia has
resulted in many unsubstantiated beliefs concerning m~ and the farm
animals. In most mammals it is very unlikely if sperm have a useful
life of more than 36 to 48 hours. In fact, most animal-breeding special-
ists recommend that natural service or artificial insemination should
precede the time of ovulation by only a few hours. In cattle, a species in
which ovulation occurs about 14 hours after the end of estrus, the best
results are obtained when service is given about 20 hours before ovulation
occurs. In the chicken fertile eggs are commonly produced for 3 weeks
following a single insemination and have been known to occur for as long
as 32 days. Of the mammals, the bat seems to be in a special class. In
some species of bats highly motile sperm have been found as long as 159
days following isolation from the male. Although, under natural con-
ditions, bats may copulate during the winter and spring, it has been
shown that sperm deposited in the fall are capable of fertilizing ova
produced in the spring (Wimsatt, 1944). Among insects, the queen
bee has been reported to lay fertile eggs for 7 years following the last
insemination.
In spite of the above facts, we must realize that the store of energy in a
tiny sperm is not overlarge and that, when the seminal fluid is deposited
in the female genital tract of the higher animals, the spermatozoa are
immediately plunged into violent activity. It would seem reasonable
to expect, therefore, that their tenure of life would be brief. This
undoubtedly is generally the case, and we are safe in assuming that Wv

can reckon their functional activity in hours rather than days. Because
it has been shown also that the egg is generally capable of fertilization
for a period of but a few hours following ovulation, it becomes obvious
that the breeder's task is to see to it that viable sperm are deposited in the
female genital tract at about the flame time that the egg is released from
the ovary.
Rate of Travel of Spermatozoa.-The older conceptions regarding the
rate at which spermatozoa travel in the female genital tract were that it
was a matter of several hours before they reached the distal end of the
ovarian tubes. These ideas have had to undergo considerable revision
in the light of later findings. When placed in proper media, mammalian
spermatozoa have been clocked at speeds of from 1 to 4 or 5 mm. per
minute with the average at about 3 mm. per minute. Many workers
have found that the sperm reach the junction of the uterus and tube in
periods as short as 10 minutes. As reviewed by Hartman (1939),
Whitney (1927) recovered sperm in the ovarian bursa 18 minutes after
coitus in the bitch, Lewis and Wright (1935) \vithin 15 minutes in the·
mare, and Phillips and Andrews (1937) within 30 minutes in the ewe.
More recent work in the ewe indicates that 20 minutes is sufficient for
sperm to reach the upper portion of the Fallopian tubes (Schott and
Phillips, 1941). Chicken sperm have been recovered in the upper part
of the oviduct in 26 minutes (Mimura, 1939). Cattle sperm require as
little as 2>1 to 4 minutes to reach the oviduct (Van Demark and Moeller,
1950).
In some species it is possible for semen to be deposited directly into or
through the cervix at the time of copulation. This is true of the horse
and swine. However, even in the mare and sow, at least a portion of the
semen is deposited in the vagina. The anatomy of the cow, ewe, and
human is such that semen must be deposited in the vagina at copulation.
Transport of Spermatozoa in the Female Genitalia.-Many hypoth-
eses have been proposed to account for the movement of sperm from the
vagina to the distal ends of the Fallopian tubes. Since sperm are highly
motile, they could move through the tract unaided. This is very
unlikely. As previously discussed, in species in which the semen is
deposited through the cervix, sperm should require less time to reach the
Fallopian tubes. It has been suggested that muscular activity of the
vagina and cervix at the time of copulation may be a factor and that at
the time of the female orgasm the cervix may dilate and/or aspirate
semen into the uterus. In some species copulation is followed by the
development of a semen plug. The semen of the boar contains much
coagulated material which could possibly act as a mechanical barrier pre-
venting the loss of sperm from the vagina. It is well known that the uterus
undergoes rhythmic contractions and that the activity is usually greatest
during estrus. It seems probable that uterine contractions are largely
responsible for transporting sperm' through the uterus and Fallopian
tubes. Inert carbon particles, when mixed with semen, travel at approxi-
mately the same rate as sperm.
As previously mentioned, it is essential that sperm maintain the power
of motility if fertility is to be preserved. It would appear logical to
theorize that once sperm have been transported through the uterine horns
and Fallopian tubes that they contact the ovum as the result of their
swimming action and that contact with the ovum is maintained by the
motile sperm.
The various means of evaluating semen quality, the dilution, preserva-
tion, and use of semen for artificial insemination will be discussed in
another chapter.
Summary.-The male's part in reproduction consists of the production
of spermatozoa and their deposition in the female genital tract at the
. proper time in relation to ovulation. The sperm are formed in the testis
in the seminiferous tubules. They then pass through a series of small
tubules into the epididymis where they undergo final maturation, and
thence into the vas deferens where they are stored prior to ejaculation.
The testis, in mammals, is ordinarily maintained in a rather vulnerable
organ, the scrotum. This is essential if sperm are to be formed normally,
since spermatogenesis cannot be completed in most mammals unless the
testes are maintained at temperatures lower than those of the body
cavity. In addition to the production of sperm the testis secretes the
male sex hormone, testosterone, which is responsible for sex drive and the
male secondary sexual characters. Male hormone and sperm production
are usually synchronized, but may be independent under certain condi-
tions. Testicular function depends primarily upon the gonadotropic
hormones of the anterior pituitary" gland. However, the maintenance of
the male in a state of maximum reproductive efficiency requires a favor-
able environment, adequate nutrition, freedom from disease or physio-
logical imbalance, and wise breeding management.
References
Books
. ALLEN, E. 1939. "Sex and Internal Secretions," The Williams & Wilkins Com-
pany, Baltimore.
AREY, L. B. 1937. "Developmental Anatomy," W. B. Saunders Company,
Philadelphia.
DUKES, H. H. 1942. "The Physiology of Domestic Animals," Comstock Publishing
Company, Inc., Ithaca.
HADLEY, F. B. 1949. "Principles of Veterinary Science," W. B. Saunders Com-

pany, Philadelphia.
MARSHALL, F. H. A. 1922. "The Physiology of Reproduction," Longmans, Green
& Co., Inc., New York.
- - - and HALMAN, E. T. 1932. "Physiology of Farm Animals," Cambridge
University Press, N ew York.
MILLS, C. A. 1939. "Medical Climatology," Charles C Thomas, Publisher,
Springfield, Ill.
PATTEN, B. M. 1931. "The Embryology of the Pig," The Blakiston Company,
Philadelphia.
SISSON, S., and GROSSMAN, J. D. 1938. "The Anatomy of the Domestic Animals,"
W. B. Saunders Company, Philadelphia.
TURNER, C. D. 1948. "General Endocrinology," 'V. B. Saunders Company,
Philadelphia.
WEISMAN, A. 1. 1941. "Spermatozoa and Sterility," P:lul B. Hoeber, Inc., Harper
& Brothers, New York.
Bulletins and Papers
ALMQUIST, J. 0., and ANDREWS, F. N. 1944. Effect of Testosterone Propionate on
Thermo-regulatory Function of Rat Scrotum, Anat. Rec., 89:125-133.
ANDERSON, J. 1945. The Semen of Animals and Its Use for Artificial Insemination,
Imp. Bur. Anim. Genet., Edinb.
ANDREWS, F. N., WARWICK, E. J., and WILEY, J. R. 1949. The Development of the
Testis in the Boar, Unpublished Data.
BISSONNETTE, T. H. 1931. Studies on the Sexual Cycle in Birds. V. Effect of
Light of Different Intensities upon the Testis Activity of the European Starling
(Sturnus Vulgaris), Physiol. Zool., 4:542-574.
ERB, R. E., ANDREWS, F. N., and HILTON, J. H. 1942. Seasonal Variation in
'Semen Quality of the Dairy Bull, Jour. Dairy Sci., 25 :815-826.
HARTMAN, C. G. 1939. Ovulation, Fertilization, and the Transport and Viability
of Eggs and Spermatozoa. In ALLEN, E., DANF'ORTJi, C. H., and DOISY, E. A.,
"Sex and Internal Secretions," The Williams & Wilkins Company, Baltimore.
HOGUE, R. L., and SCHNETZLER, E. E. 1937. Development of Fertility in Young
Barred Plymouth Rock Males, Poultry Sci., 16 :62-61.
HUGGINS, C., SCOTT, W. W., and HEINEN, J. H. 1942. Chemical Composition of
Human Semen and of the Secretions of the Prostate !1nd Seminal Vesicles, Amer.
Jour. Physiol., 136:467-473.
KIRKPATRICK, C. M., and ANDREWS, F. N. 1944. Development of the Testis in 1\e
Ring-necked Pheasant, Anat. Rec., 89 :317-324.
LASLEY, J. F., and MAYER, D. T. 1944. A Variable Physiological Factor Necessary
for the Survival of Bull Spermatozoa, Jour. Anim. Sci., 2 :129-135.
McKENZIE, F. F., and BERLINER, V. R. 1937. The Reproductive Capacity of
Rams, Mo. Agr. Expt. Sta. Res. Bul. 265.
- - - , MILLER, J. C., and BAUGESS, L. C. 1938. The Reproductive Organs and
Semen of the Boar, Mo. Agr. Expt. Sta. Res. Bul. 279.
MERCIER, E., and SALISBURY, G. W. 1947. Fertility Level in Artificial Breeding
Associated with Season, Hours of Daylight, and the Age of Cattle, Jour. Dairy
Sci., 30 :817-826.
_ - - and - - - . 1947. Seasonal Variations in Hours of Daylight Associated
with Fertility Level of Cattle under Natural Breeding Conditions, Jour. Dairy
Sci., 30:747-756.
MIMURA, H. 1939. On the Mechanism of Travel of Spermatozoa through the

Oviduct in the Domestic Fowl, with Special Reference to the Artificial Insemina-
tion, Sonderabdruck aU8 Okajimas Folia Anatomica Japonica Band., 17 :459-476.
MOORE, B. H., and MAYER, D. T. 1941. The Concentration and Metabolism of
Sugar in Ram Semen, Mo. Agr. Expt. Sfa. Res. Bul. 339.
:\fOORE, C. R., and QUICK, W. J. 1924. The Scrotum as a Temperature Regulator
for the Testis, Amer. Jour. Physiol., 68:70-79.
PHILLIPS, R. W., and ANDREWS, F. N. 1937. The Speed of Travel of Ram Sperma-
tozoa, Anat. Rec., 68:127-132.
- - - , and - - - . 1936. The Development of the Testes and Scrotum of the
Ram, Bull, and Boar, Mass. Agr. Expt. Sta. Bul. 331.
- - - and McKENZIE, F. F. 1934. The Thermo-regulatory Function and Mecha-
nism of the Scrotum, Mo. Agr. Expt. Sta. Res. Bul. 217.
REINEKE, E. P. 1946. The Effect of Synthetic Thyroprotein on Sterility in Bulls.
In ENGLE, E. T. "The Problem of Fertility," Princeton University Press,
Princeton, N.J.
ROWAN, W. 1929. Experiments in Bird Migmtion. I. Manipulation of the
Reproductive Cycle: Seasonal Histological Changes in the Gonads, Boston Soc.
Nat. Hist. Mem., 89:151-208.
SALISBURY, G. W., KNODT, C. B., and BRATTON, R. W. 1948. The Freezing Point
Depression of Bull Semen and Its Relation to the Diluter Problem, Jour. Anim.
Sci., 7 :283-290.
SCHOTT, R. G., and PHILLIPS, R. W. 1941. Rate of Sperm Travel and Time of
Ovulation in Sheep, Anat. Rec., 79 :531-540.
SHAFFNER, C. S. 1942. Longevity of Fowl Spermatozoa in Frozen Condition,
Science, 96 :337.
- - - and ANDREWS, F. N. 1948. The Influence of Thiouracil on Semen Quality
in the Fowl, Poultry Sci., 27 :91-102.
- - - and - - - . 1943. The Determination of the Concentration of SpermatozO:\
in Fowl and Bull Semen, Anat. Rec., 86:99-107.
SMITH, S. E., and ASDELL, S. A. 1940. The Buffering Capacity of Bull S3mprt,
Cornell Vet., 30:499-506.
TYRRELL, W. P., ANDREWS, F. N., and ZELLE, M. R. 1942. Effect of Postpubertal
Castration on Thermo-regulatory Function of Rat Scrotum, Endocrinology,
31 :379-383.
VAN DEMARK, N. L., and MOELLER, A. N. 1950. Spermatozoan Transport in the
Reproductive Tract of the Cow, Jour. Dairy Sci., 33:390.
\\THEELER, N. C., and ANDREWS, F. N. 1943. The Influence of Season on Semen
Production in the Domestic Fowl, Poultry Sci., 22 :361-367.
WIMSATT, W. A. 1944. Further Studies on the Survival of Spermatozoa in the
Female Reproductive Traet of the Bat, Anat. Rec., 88 :193-204.
YOUNG, W. C. 1929. A Study of the Function of the Epididymis. II. The
Importance of an Aging Process in Sperm for the Length of the Period during
Which Fertilizing Capacity Is Retained by Sperm Isolated in the Epididymis of
the Guinea Pig, Jour. Morph. and Physiol., 48:475-491.
CHAPTER V
THE FEMALE'S PART IN REPRODUCTION
In all the higher species of animals, fertilization and embryonic devel-

opment take place within the female genital tract understanding the
structure and physiological functions of these organs is a prime requisite
for anyone 'vho would succeed in the art of livestock breeding. Enor-
mous financial losses are incurred annually by breeders because of the
difficulty so often experienced of getting females to conceive and, although
it is unjust to lay all the blame at the door of the females, their share
of the total fault is no doubt considerable. "Knowledge is power," and
an understanding of the processes involved in fertilization, gestation,
parturition, and lactation would often enable a breeder to avoid the
fatalities that so often threaten individuals before as well as after birth.
Ovaries.-Mammalian females possess two ovaries, homologous to the
testicles of the male; and, like the latter, the ovaries arise from the inter-
mediate portion of the mesodermal somites and have a double function:
(1) the production of ova or eggs and (2) the secretion of hormones. In
other words, the ovary performs an ovogenetic as well as an endocrine
function. The ovaries are generally spoken of as the primary organs of
reproduction in the female because of their ovogenetic function and
because they are the source of the female sex hormones. The develop-
ment and functional activity of the remaining portions of the female
genitalia are dependent upon the second function of the ovary, viz., its
endocrine function. An internal secretion of the ovary stimulates the
growth of the uterus and vagina and other hormonal secretions from
the corpus luteum, which appears cyclically in the ovary, sensitizes the
uterine wall for the reception of the fertilized egg, maintains the preg-
nancy, and assists at parturition. These secretions together with others
from the placenta and anterior pituitary stimulate mammary develop-
ment and lactation following parturition. In some species ovogenesis is
practically continuous after puberty. In others it occurs once, twice,
or a few times a year.
In the higher animals, these glands are situated in the abdominal or
the pelvic cavity suspended in the broad lig~ment. Their position varies
somewhat in different species and in different individuals. 'Fhey ~
i!:Stached to the broad ligament on one side at the so-called "bjJl!~'
134
THE FEMALE'S PART I N REPRODUCTION 135
From this point the tissues of the broad ligament spread out and pass
through the ovary, forming the ovarian stroma, made up of fibrous
connective tissue. Toward the surface, the stroma becomes interspersed
with glandular tissue that forms the ovarian cortex. Surrounding the
ovary is a rather dense layer of fibrous tissue called the tunica albuginea
and outside this is found a modified peritoneum, the inner porti~ of
which consists of a single layer of cuboidal cells and which is known as
thejlerminal epithelium. It is from the latter that the ova arise.
In the mare, the ovaries are somewhat bean-shaped and 2 to 4 in. in
the longest diameter; in the cow, 1 to IH in. ; in the ewe, H to 1 in.; and
in the s~omewhat larger than in the ewe and more it~regular, owing
UTERINE
HOR N--=~IL+
CORPUS
LUTEUM
FIG. 40.-Diagram of the genital organs of the cow.
to the many protuberances of the follicles and corpora lutea which give
them a shape somewhat like a bunch of grapes.
In the domestic fowl. and other avian species, only the left ovary is
functional. During early embryonic life there are two female gonads, but
the right remains as a functionless rudiment. During the egg-produc-
ing period the right ovary contains clusters of yellow or reddish-yellow
spheres. These vary in size from the fully developed yolk to microscopic
ova. Although they differ from mammalian ova in that they contain a
large amount of yolk, they serve the same general function. It is
interesting to observe that the chicken, the most highly developed of the
farm animals from the reproductive standpoint, has only one functional
ovary.
The ovary consists of glandular, connective, and nerve tissues, blood
and lymph vessels, and many thousands of follicles in various stages of
development. The last, which arise from the germinal epithelium, are
found all through the stroma of the ovary, and each contains an immature
ovum that, when matured, is liberated and may be fertilized by a sperma-
tozoon and produce a new individual. It has been estimated that there
are upward of 75,000 follicles in the ovaries of a heifer calf at birth. Only
a very small percentage of these follicles would ever ripen, even though
FIG. 41.-Portion of generative organs of cow. 1. os uteri; 2, right cornu; 3, ovary; 4,

ovarian ventricle; 5, fimbriated end of oviduct; 6, oviduct. (From WiUiam8, Diseases oj
the Genital Organs of Domestic Animals.)
the cow was never bred and came in heat regularly once in 3 weeks for
10 to 15 years. In this case, the ovaries would gradually become fibrous
because of the deposition of the small plugs of connective tissue, the end
result of the absorption of corpora lutea. The newer concept of a con-
tinual production of new ova from the germinal epithelium, similar to
spermatogenesis in the male, would involve the production of even
greater numbers of follicles, most of which atrophy and are absorbed
THE FEMALE'S PART IN REPRODUCTION 137
before reaching maturity. This continuous process of ovogenesis from

the layer of germinal epithelium is comparable to the continuous process
of spermatogenesis in the male. It is now claimed to exist in the mouse,
rat, guinea pig, dog, cat, monkey, and mare, and is probably the g~neral
method of egg production of all the mammals. In either case, nature
has attempted to ensure the perpetuation of the race by providing a
superabundance of both spermatozoa and ova.
The ovarian follicle at the beginning consists of the egg cell sur-
rounded by a single. layer of epithelial cells Surrounding this is a
specialized layer of ovarian connective-tissue stroma known as the theca.
FIll. 42.-The generative organs of the mare. 1, left ovary; 2, Fallopian tube; 3, left horn
of uterus; 4, right horn of uterus; 5, body of uterus; 6, broad ligament; 7, vagina; 8, abdomi-
nal wall; 9, left kidney; 10, left ureter; 11, urina.ry bladder. (F1'om Leisering, AUa8 of the
Anatomy of Domesticated Animal8.)
As growth proceeds, the epithelial cells multiply and push out away from
the ovum forming an antrum, which becomes filled with a yellowish,
alkaline, albuminous fluid, the liquor folliculi The ovum itself remains
on a small hillock of epithelial cells, the cumulus oophorus. There is
some variation in the size of the ova of different mammalian species, but
they measure on the average about 0.13 mm. This is greatly in excess
of the size of the sperm, which has little or no cytoplasm. The egg, on
the other hand, has a nucleus with its contained chromosomes of about the
same size as the head of the sperm, and in addition it has a considerable
amount of stored cytoplasm, or yolk, in which the nucleus is suspended.
This makes the egg several thousand times the size of the sperm, but
dam is actually no more important than the sire from a hereditary
"""'''~l'''''Jl'llt, for both egg and sperm have a nucleus of equal size in which
138 BREEDING AND IMPIWVEMEN7' OF FARM ANIMALS
are found an identical number of chromosomes (with the possible excep-

tion of sex chromosomes in some species). The stored yolk of the egg is
Rresumably designed to provide energy f;r, the developing zygote for
..t.he first few days following fertilization and until the zygote can pass
down the ovarian tubes into the uterus and find another sourc of en;;'gy
an materials for growth. The developing follicle passes toward the
periphery of the ovary, finally protruding from the ovary \0 the extent
of half its size or more preparatory to being liberated. This maturing
of the ovum in the ovary occurs at
more orIessregularintervalsinfarm
animals and is usually accompanied
or preceded by estrus, or heat.
This cyclic development of a
mature ovum begins at puberty,
under the stimulation of the anterior
pituitary hormones. As noted in
the previous chapter, the develop-
ment of ova from Pfliigers egg '
cords begins early in embryonic
life. These embryonic follicles,
however, and those from birth to
puberty usually become atretic.
Puberty marks the time when a
follicle succeeds in reaching full
maturity "\,tith the attendant matur-
ing of the egg and reduction of the
number of chromosomes to one -half.
FIG. 43.-Mouse ovary showing follicles of It is now thought that waves of
different sizes and types.
follicles start developing cyclically
throughout the reproductive life of the female, one or a few of them
reaching full maturity, the others being resorbed preceding the next wave.
In the production of a functional ovum, or egg, the primordial egg cell
in the follicle goes through two rapid cell divisions. Previous to the first
division, synapsis occurs; and the net result of the first division, which
takes place in the ovary in mammals, except perhaps in the bitch, is a
reduction in the number of chromosomes to one-half the number charac-
teristic of the species and the very unequa, division of the cytoplasm.
This is known as the heterotypic division and is followed by a s~cond
homotypic cell division in which the chromosome number is not further
reduced. The one mature ovum now passes into the peritoneal cavity
of the broad ligament and then into the uterine tube.
Starting with a primary germ cell, or oocyte, in the ovary, the reduction
division, including synapsis, and the formation of tetrads is the same as

found in the male, except that in the female, instead of two secondary
spermatocytes, there are formed one secondary oocyte and one non-
functional polar body, each of which, however, has received one-half the
J'ta. 44.-Maturation in ova of gonadotropin-stimulated mouse ovaries. Observe the

ebromosomes and the maturation spindles in the metaphase stage. Magnification approxi-
1200 X.
renrOInatlll material and contains one-half the normal number of chromo-

In effect, the nucleus, rather than the whole celi, divides. The
body differs from the secondary oocyte in having received none
the yolk or cytoplasm. These two, the secondary oocyte and the
body, then divide, the oocyte giving rise to one ovum or egg and
more polar body. The first polar body may give rise to two polar
The ovum is the female gamete that is capable of being fertilized
by a spermatozoon. The three polar bodies are nonfunctional and are

later absorbed. So, whereas in the male there are produced four func-
tional spermatozoa from each primary spermatocyte, from the primary
oocyte of the female in the higher forms there arises but one functional
ovum and two or three nonfunctional polar bodies.
The purpose of the reduction division in the formation of germ cells,
in both the male and the female, is to keep the number of chromosomes
for the species constant. When an ovum with one-half the normal num-
ber of chromosomes is fertilized by a spermatozoon, also with one-half
the normal number of chromosomes, the summation of the two halves
restores the normal number of chromosomes in the zygote that is to
produce the new individual. No doubt, some cases of sterility in hybrids
are due to the fact that the two species that were crossed had different
characteristic numbers of chromosomes, and, although fertilization and
subsequent development are possible, it is nevertheless impossible for
the hybrids to produce functional spermatozoa or ova, because the nor-
mal pairing of chromosomes and subsequent cell mechanism is interfered
with.
So far as known, the distribution of chromosomes in germinal mitosi~
is at random. The solid chromosomes in Fig. 45 represent those con-
tributed by the male, those in outline by the female. Following synapsis
and disjunction, one member of each pair will proceed to one pole and
be found in one cell, the other member of each pair will go to the other
pole and be found in the other cell. The fact that one member of a
certain pair goes to one cell has no influence on the distribution of any
of the other chromosome pairs; in other words, the distribution is inde-
pendent, or random, it being understood that one or the other member
of each pair must normally go to each daughter cell. Figure 89 repre-
sents all the possible combinations of the eight chromosomes of Drosoph-
ila. As will be pointed out later, random distribution is one of the most
potent causes of variation.
We meet here again the all-important fact that any of the higher
organisms passes on to each of its offspring a sample hrtlf of its entire
inheritance, one member of each pair of its chromosomes. Our whole
purpose in making a study of breeding principles will be to learn how
our animals may be made more pure in their genetic content of desirable
~
genes, so that it will make no particular difIerence which member of each
pair of chromosomes any offspring may chance to get.
Hormonal Regulation of the Ovary.-The normal development and
function of the ovaries and the testes are dependent upon the gonado-
tropic hormones of the anterior pituitary gland. '.' If these hormones are
not secreted in proper balance in the mature female, the heat periods
\
may be irregular and normal follicular development and ovulation may

not occur. AB previously explained, there are two anterior pituitary
gonadotropins. The follicle-stimulating hormone (FSH) is concerned
primarily with the growth of the follicle. The luteinizing hormone (LH)
l'lG. 45.- Diagram of maturation or miosis of germ cells in animals. The process begins
with the third row of cells. Maternal chromosomes white, paternal black. All chromo-
pomes in the fertilized egg received from the mature egg are thereafter maternal, and those
received from the spermatozoon are thereafter paternal, regardless of what they were in
the mature germ cells. (After Shull.)
is chiefly responsible for ovulation or rupture of the mature follicle and

the development of the corpus luteum. There is some disagreement in
different research laboratories as to the exact time relationships which
are involved, but there is reasonable agreement in regard to the basic
principles of the mode of action of the gonadtropins in the female.
In the ovary of the sexually immature female there are hundreds of

small follicles. These are formed both prior to birth and puberty and
continue to be produced even after the surgical removal of the anterior
pituitary gland. Accompanying or probably causing puberty, FSH is
secreted by the anterior pituitary in increasing amounts. Under the
influence of this hormone a follicle, or follicles, grows rapidly. As the
follicle grows under the influence of FSH, the anterior pituitary secretes
minute but gradually increasing amounts of the luteinizing hormone.
The LH cau~s the follicle to produce estrogen, the female sex hormone,
and the female comes in heat (estrus). The increased secretion of
estrogen by the follicle results in decreased production of FSH by the
! pituitary and increased secretion of LH. When LH is present in the
blood in sufficient amounts, it brings about rupture of the mature follicle
and the development of the lutein cells which form the corpus luteum.
As the corpus luteum forms, it secretes the hormone progesterone, which
is concerned with the final preparation of the lining of the uterus for
pregnancy. Progesterone also inhibits the secretion of the gonadotropic
hormones by the anterior pituitary gland. If pregnancy is not estab-
lished, the corpus luteum eventually loses its function, ceases to secrete
progesterone, and FSH is again produced in increased amounts and a
new ovarian cycle is begun. If pregnancy is established, the ovaries
usually cease to produce new follicles and ova until after parturition.
In many seasonal breeders there is only one ovarian cycle each year.
In cattle, on the other hand, the cycle is repeated at intervals of 19 to 21
days throughout reproductive life, unless pregnancy interferes. In
summary, follicular development, estrus, and ovulation depend upon the
interaction of the gonadatropic hormones FSH and LH and the two
ovarian hormones estrogen and progesterone.
The Sexual CyCle.-The outstanding characteristic of the reproduc-
tive process in the female is its rhythmicity. All of the vertebrates
which have been studied have definite reproductive cycles. There are
wide species differences in the cycles, but each species has some basic
pattern.
In most mammals the cycle is called the estrous or estrual cycle.
Some animals, such as the fox, have only one heatperiod each year, and
are called monestrous. The farm animals all have more than one cycle
per year and are called polyestrous. Females, such as the ewe, which
have a limited reproductive period are called seasonal breeders in con-
trast to the cow which is a continuous breeder. The estrous cycle can
be divided into several distinct phases. The most easily recognized
phase is the period of sexual receptivity, or heat, and is technically known
as estrus. The stage following estrus is called metestrus. It is during
this period that the corpus luteum forms and secretes progesterone
which brings about the final development of the uterus for pregnancy.
If pregnancy does not occur, the corpus luteum regresses and there is a
short period of genital rest called diestrus. Diestrus is followed by
proestrus, which is characterized by follicular growth and generally
heightened reproductive activity. Follicular growth is accompanied
by the secretion of estrogen which gradually induces estrus. During
heat the follicle usually reaches its maximum size, and ovulation occurs
shortly before or just following the end of estrus. In seasonal breeders
the genitalia have a long period of quiescence which is called anestrus.
In cattle and swine the first part of lactation is usually accompanied by
a period of sexual inactivity called lactational diestrus.
TABLE ll.-THE REPRODUCTIVE CYCLE IN FARM ANIMALS
Length of estrual Length of ges-

Length of estrus
Species
cycle, days
--~--I----~----I
I Usual time of
ovulation
tation, days Age at
puberty,
months
Av. Range Av. Range Av. Range
_ _- - - - - - - - - - - - - 1 - - - - 1 - - - - - - - - - - - 1 · - - -
Mare....... 21 10--37 5-6 days 1-37 days 24-48 'h 0 U r s 336 310--350 10--12
before end of
estrus
Sow...... .. 21 18--24 2-3 days 1-5 days Usually second 112 111-115 3-7
day of estrus
Ewe....... . 16 14-20 30 hours 20--42 hours 1 hour before 150 140-160 4-8
end of estrus
Goat .... , .1 20 12-25 36-48 hours 20--80 hours Near end of 151 140--160 4-8
estrus
Cow. ...... 119-20 16-24 I 15-20 hours 8-30 hours 14 hours after 281 274-291 4-8
end of estrus
I I
Because small quantities of blood sometimes appear in the genital
secretions of the cow and bitch, it is frequently said that menstruation
occurs. This is not the case, and the terms menstruation and menstrual
cycle should be reserved for the Primate. Menstruation is the shedding
or elimination of a large portion of the uterine lining (endometrium) at
regular intervals-usually each 26 to 30 days. In the Primate ovulation
occurs at approximately the midpoint of the menstrual cycle. Thus
ovulation would be most apt to occur 14 days after the beginning of
menstruation in a female with a 28-day menstrual cycle. Primates,
unlike other mammals, do not have a limited period of sexual receptivity
and thus no outward manifestations which can be easily correlated with
ovulation. The farm animals are characterized by the presence of a
very limited period of estrus which is closely correlated with ovulation.
144 BREEDING AND IMPROVEMENT OP PARLlI ANIMALS
FACTORS INFLUENCING THE SEXUAL CYCLE
Age.-Sexual maturity is a gradual process. The appearance of heat

does not necessarily me~m that normal fertility has been attained. Dur-
ing the transitional ppriod of puberty the sexual cycles may at first be
irregular. Ovulation may occur in the absence of estrus, and estrus with-
out ovulation is not uncommon. After full sexual maturity has been
attained, the sexual cycles are usually regular throughout life. In the
human, the reproductive processes usually cease in later life; this is called
thc menopause. In the farm animals old age itself is rarely the cause of
reproductive failure. The age at puberty is approximately the same as
previously given for the male. In sheep, however, the seasonal factor is
important. Whereas sperm first appear in the ram at five or six months
of age, sexual cycles are not initiated in the ewe until fall. Thus a lamb
born in January would probably not exhibit sexual activity until Sep-
tember. There are, of course, wide species and individual variations in
the time of puberty. Some of these will be discussed in the chapter on
Reproductive Efficiency.
Seasonal Periodicity, the Breeding Season.-The effects of season on
reproductive activity are more easily recognized in the female than in
the male. As previously explained, this is sometimes due to the fact that
some males may exhibit marked sex drive in the absence of spermato-
genesis, e.g., the ram. In seasonal-breeding females estrus is not manifest
except during the breeding season.
Most of the wild mammals are monestrous. The dog and the cat
usually have two estrual cycles annually. Sheep are polyestrous, but in
the X orthern Hemisphere the cycles are usually limited to the period
between August 15 and January 30. Goats are very similar to sheep,
the breeding season usually extending from September through Decem-
ber. Estrus occasionally occurs in sheep and goats in the spring and
summer months, but planned breeding can rarely be expected during
this period. Cattle, swine, and horses exhibit reproductive activity
throughout the year but, as explained in the previous chapter, seasonal
tendencies still persist. In the Western range states cattle breeding is
largely a seasonal operation. Environmental and nutritional conditions
are such that it is most satisfactory to have the calves born in the spring.
For obvious reasons, the calving period is somewhat later in the Northern
areas than in the South and Southwest. For very practical reasons
swine breeding in the corn belt is seasonal in nature. When the two-
litter system is used, the sows are bred to farrow in the spring and fall.
The majority of swine men using the single-litter system farrow their
HO\\"" in the spring. Thu~, although hoth rattle and swine ean reproduce
throughout the year, man has done much "to keep them on a seasonal
schedule. Mares, if well cared for, are capable of continuous reproduc-
tive activity. Animals which are not well fed, and some which are, are
most apt to exhibit estrous behavior during the spring and summer
months. When horse production was more important than at present,
by far the greatest number of mares were mated in the spring and
summer.
Light and Temperature.-The mode of action of light and temperature
upon reproduction in the female is probably rather similar to that
described in the male. The effects of light in stimulating fall and winter
egg production in chickens are well
known. Although the total annual
egg production is increased but
slightly, the hens are induced to lay
in the fall and winter when the prices
of eggs are highest. It seems clear
that the effects of light are upon the
anterior pituitary gland and its secre-
HO tion of the gonadotropic hormones.
Estrone In the chicken both egg production
and fertility are lowest during the
summer. It is difficult to separate
the effects of light and temperature,
but it has been hypothesized that
decreased thyroid activity during the
warm months may indirectly reduce
HO reproductive efficiency. In human
Alpha - estradiol gynecology it is frequently reported
:FIG. 47.-Structural formulas of two of the that the correction of subnormal
common female hormones.
thyroid activity is followed by
pregnancy. Although conclusive experimental proof of such hypotheses
is lacking, they are worth consideration. Temperature, light, and
"
humidity are very difficult to control in large-animal experimentation.
Studies with goats by Bissonnette (1941) and by Sykes and Cole (1944)
strongly suggest that increased light followed by decreased light, in an
attempt to duplicate seasonal changes, will induce out-of-season breed-,
ing activity.
The Hormonal Functions of the Ovary.-The endocrine activity of the
testis is simple by comparison with the ovary. In the male the andro-
genic hormone is produced at a more or less continuous rate throughout
the reproductive period. In strictly seasonal breeders, or' course, the
male hormone is produced in maximum amounts only during the breed-
ing season. The ovary produces at least two distinctly different hor-
mones. These are produced only at definite times during the sexual
cycle, and the levels of secretion of each hormone vary during the
cycle.
Estrogen.-It has been known since the 1890's that atrophy of the
uterus following castration could be prevented by an ovarian graft.
Long and Evans (1922) and Allen (1922) suggested that the substance
responsible for the maintenance of the uterus was produced by the
Graafian follicle, and this was proved by Allen and Doisy in 1923. Many
workers regard these studies as the beginning of modern endocrinology.
The hormone produced by the Graafian follicle is commonly referred to
as the female sex hormone and is one of a large number of the estrogenic
hormones. During the early 1920's it was thought that this was the only
ovarian hormone.
It is now known that at least 10 different estrogens appear in animal
tissues, secretions, blood, or urine, and that many others with estrogenic
activity can be prepared synthetically. Estrogens have been isolated
from the ovary, placenta, testis, and adrenal cortex and from various
plants. They are found in large amounts in the urine of both males and
females and such unlikely sources as petroleum and the slime from the
bottom of the Dead Sea.
The first estrogen to be crystallized and chemically identified was
estrone. This hormone was isolated from the urine of pregnant women
by Doisy et al. in 1929. The most active natural estrogen, and the one
which is probably secreted by the Graafian follicle, is estradiol. It is
very similar to estrone in structure but has considerably more physiologic
activity.
The natural estrogens have little effect unless they are administered by
injections, e.g., subcutaneously or intramuscularly. During the 1930's
many compounds which do not appear in nature but which have great
estrogenic activity were synthesized. Many of these compounds are
dissimilar chemically from the natural estrogens, and some are very
effective when administered orally. One of the best known is diethyl-
stilbestrol. It can be synthesized at low cost, is more active physiologi-
cally than most of the natural estrogens, is orally active, and has few
undesirable side effects. Diethylstilbestrol is widely used in human and
animal medicine for the correction of reproductive disorders. In addition
to its effect on the genital system it has been shown to be capable of
inducing fattening in chickens (Andrews and Bohren, 1947, and others)
and of increasing growth in cattle and sheep (Dinnusson et al., 1949, and
Andrews et al., 1949).
Progesterone.-The physiologic importance of the corpus luteum has
long been known, although the substance responsible for the effects was
not isolated until 1934. Beard suggested in 1897 that this structure
was necessary during pregnancy and would likewise affect ovulation and
the estrous cycle (Turner, 1948). The general descriptive name of the
corpus luteum hormone is progestin and the crystalline hormone itself is
called progesterone. Progesterone is a steroid hormone closely related
chemically to both'the natural estrogens and androgens. It is produced
cyclically during the estrous cycle, reaching a peak during the late
metestrous stage of the cycle. It is secreted in large amounts during
early pregnancy and can apparently
be produced by both the corpus
Juteum and placenta.
Androgen.-It is clear that the
ovary, in addition to secreting estro-
gen and progesterone, can produce
significant quantities of androgen.
Diethylstilbestrol Whether the ovary normally secretes
androgen in small amounts is not
known, but there is no doubt of its
androgenicity under certain special
conditions. When ovarian grafts
have been made into the ears of cas-
0/ trate rats and mice, sufficient andro-
Progesterone gen has been produced to maintain
FIG. 48.-Structural formulas of diethyl- the male accessory glands (Turner,
stilbestrol and progesterone. 1948). The temperature of the graft
site may be of some importance, since the ear has a lower temperature
than the normal abdominal ovarian location.
Effects of the Ovarian Hormones on the Genital Organs.-The most·
easily recognized effect of the estrogens is their stimulatory action on the
vagina. This effect is the basis of the most commonly used biologic
test for estrogenic activity. Prepubertal castration of the rat or mouse
prevents the development of the vagina. The opening of the vaginal
canal is prevented or greatly delayed, and the vagina has only a very thin
lining of epithelial cells. The injection of estrogen in such an animal
causes precocious opening of the vagina and the development of a very
thick lining of epithelial cells. These cells are produced in great quan-
tities by the germinal epithelial layer. As new cells are produced, the
innermost cells are sloughed off and appear as large scaly cells with
indistinct nuclei. These cells can be obtained for' study by introducing
a swab into the vagina and making a smear on a glass slide for microscopic.
.
study. The presence of large numbers of such large scaly cells is indica-
-
THE FEMALE'S PART IN REPRODUGTIOlv' 149
tive of an estrogenic effect. This method is very accurate for the de tee-
tion of ~nute quantities of estrogen and is widely used in biology and
medicine. Likewise, the condition of the vagi1lal epithelium in normal
animals is indicative of the stage of the estrual or menstrual cycle and
has wide application in clinical studies.
The development of the uterus is dependent upon both estrogen and
progesterone. Development is initiated by estrogen and completed by
progesterone. Estrogen injection of the castrate is followed by increased
proliferation and activity of the epithelial cells lining the uterus and of
the smooth muscle cells of the myometrium. The activity of the uterine
glands is ordinarily stimulated first by estrogen and is then continued by
progesterone during the luteal phase of the cycle. Uterine activity, as
measured by the rhythmic contractions of the uterus, is maximum during
estrus, and the uterus is relatively quiescent when progesterone is being
secreted by the corpus luteum.
As previously discussed, the corpus luteum is present and functional
only at certain stages of the estrous cycle and during pregnancy. The
primary function of its hormone, progesterone, is the final development
of the uterine epithelium in preparation for implantation of the blastocyst.
Progesterone is concerned with the maintenance of pregnancy following
implantation. In the cow, removal of the corpus luteum from the ovary
during pregnancy, especially during the first half of pregnancy, will
nearly always result in abortion. In "the mare, the ovaries may be
removed during early pregnancy without causing abortion. This is
assumed to indicate that the presence of a functional corpus luteum in
the ovaries of pregnant animals is essential in some species and not in
others. It is known that the placenta can produce progesterone, estro-
gen, and even gonadotropins, and this may explain some of the species
differences in dependency on the ovaries during pregnancy.
The Ovarian Hormones and Secondary Sexual Characteristics.-Like
the male, the female also exhibits characteristics peculiarly associated
with her sex and referable, as are the development and functional activi-
ties of the accessories, to hormones secreted by the ovaries. This is one
of the main functions of the ovaries.
These characteristics are to be found largely in the general body form
and are generally not so striking as those found in the male. The typical
male form is heavy or well developed anteriorly, less well developed
posteriorly. In the female, the exact opposite prevails. The mamma-
lian female develops a wider pelvis and a general feminine type expressed
in finer bone and more refined fe2.tures than the male. The deposition
of fat differs from that in the male, the voice is higher pitched, and she
differs also in psychological traits.
Miscellaneous Effects of the Ovarian Hormones.-The estrogens have

very definite effects on the skeletal §Y-stem.--..Althoogh skeletal growth
is dependent chiefly on -the growth hormone of the anterior pituitary
gland, both estrogen and androgen stimulateOfinhibit gro~vth, depending
-on:-the levels at which they are present. Preeocious secretion of the
sex hormones tends to bring about closure of the epiphyseal junctures
of the long bones and skeletal growth stops. In the castrate, because
of the removal of the sex-hormone influence, skeletal growth may con-
tinue beyond the normal limits. The injection of both estrogens and
androgens at certain optimum levels has been shown to increase growth
rate.
As will be discussed in a later chapter, the estrogens are concerned in
the development of the mammary system. Numerous secondary effects
have been attributed to the estrogens. Depending on the species and
the type and level of estrogen administration, they have been reported
to affect fat deposition, calcium metabolism, nervous activity, metabolic
rate, blood-cell formation, vascular phenomena, and many other body
activities.
Muscular activity of the uterus is greatest during the phase of maxi-
muin -estrogen ~eretiun. The injecti"on of progesterone, or its secretion
during the luteal phase of the estrual cycle, is followed by uterine quies-
cence. The reason for this sequence of events appears to be concerned
with the transport of sperm in the female genitalia during estrus. The
less motile stage would seem to be conducive to the implantation of the
developing blastocyst. In humans with histories of habitual abortion,
progesterone is frequently administered because of its depressing effects
on uterine motility and its stimulatory action on endometrial growth.
There are obviously many different causes of abortion, and progesterone
could not be expected to have beneficial effects where some specific cause-,
such as brucellosis in cattle and swine, is present. Like estrogen, proges-
terone is concerned with development of the mammary glands.
Estrous Behavior.-It is frequently assumed that estrus, being the
only phase of the estrous cycle which is e~sily recognizable, needs no
further discussion. Since the widespread adoption of artifical insemina-
tion, it has become increasingly apparent that many livestock breeders
cannot recognize estrus and do not appreciate its full significance. In the
farm animals there is a definite relationship between estrus and ovulation.
If service or artificial insemination is carried out in relation to a particular
stage of estrus, breeding efficiency can usually be increased. Under
natural conditions females probably mated several times during estrus,
thus increasing the likelihood of fresh sperm being present in the uterus
THE FEMALE'S PART n·/ REPRODUCTION 151
at the time of ovulation. If mating occurs only once during heat, the
time relatJonships are of vital importance.
In all the farm animals estrus or heat is recognized by the willingness
of the female to accept the male. In the ewe and mare the usual outward
signs are the fact that the female stands and will allow the male to mount.
Mares react very vigorously when not in estrus and may severely injure
the stallion. For this reason, it is recommended that mares be tried to
a stallion in some sort of a teasing stall or chute. Sows may occasionally
exhibit both male and female mating behavior, but the estrous animal is
usually recognized by her willingness to accept the male. Of the farm
animals, cows are most likely to exhibit both male and female mating
behavior. This has led to numerous errors in artificial insemination
associations, and the "wrong" cow has often been kept in the barn for
the insemination. Ordinarily, the truly estrous cow is the one which is
mounted by other cows. When one female is approached by several
cows, there should be little doubt that she is in heat. Fortunately, there
are some physiologic as well as behavioral signs of estrus. Estrus is
usually characterized by increased genital secretion and discharge due to
stimulation by the estrogenic hormone. In cattle and swine there is
frequently some swelling of the vulva and urination may be frequent.
Following the discovery of estrogen, it was concluded that this hor-
mone was solely responsible for estrous behavior. The experimental
study of estrus indicated that androgens and progesterone, as well as
estrogen, may be involved. In the spayed guinea pig sexual receptivity
depends upon the secretion of estrogen followed by the secretion of small
amounts of progesterone. In this species it has been shown that proges-
terone is secreted by the Graafian follicle prior to ovulation (Dempsey
et al., Hl36). This has recently been shown to be true in the rat, mouse,
and hamster (cited by Cole, 1948). Cole (1948) has found that in the
ewe androgen, in combination with equine gonadotropin, was quite
effective in inducing estrus. Since androgens are present in significant
amounts in the female of several species, this reaction might not be limited
to the ewe.
Ovulation.-Ovulation is the freeing of the ovum by the rupture of the
follicle in the ovary. As indicated irr-the previous section r this takes '
place in animals durIng the estrus or heat period, except those, of course,
which do not ovulate spontaneously. Many theories have been proposed
to account for the freeing of the egg from its follicle; e.g., pressure from
within the follicle, pres~ure induced from the blood supplY-of the ovary,
ml,lSQ~ffi-the--(')¥ary or the follicle, and the action of
proteolytic enzYlJli)S in the liquor folliculi.
Most animals, including all the .farm animals,-o.vulate spontaneousl

That is, ovulation does not depend'upon copulation. The rabbit differs
in that some type of stimulation, such as copulation, mechanical or
electrical stimulation of the cervix, or other afferent stimuli, is necessary
FlG. 49.-0vulation in the rat. 1. Ovum detached from germinal hill and approaching
stigma. 2. Rupture of the stigma and passage of the ovum th.rough it. (Courtesy of Dr.
Richard J . Blandau, University of Washin(Jton, School of Medicine).
if ovulation is to occur. In the species in which the mechanisms of

ovulation have been studied, whether they ovulate spontaneously or are
similar to the rabbit, it appears that the primary regulation is by hormones.
As previously mentioned, ovogenesis and follicular formation can
proceed in a limited way in the absence of the gonadotropic hormones.
THE Fl!JM ALE' S PART IN REPROD UCTION 153
However, the gonadotropins are needed for the development of Graafian

follicles lnd ovulation. The most rapid phase of follicular growth
appears to occur under the influenc~f ESll- In experimental animals
under the prolonged influence of FSH the follicles become large and
cystic and do not rupture. In normal animals it is thought that as
follicular growth proceeds the anterior pituitary gland secretes increased -
amounts of LH. and that QYlIlatiQR oeem:s_when a certain balance between ';;$'
FSH and LH is reached. In the rabbit, copulation stimulates the release ~s /.f
of-;-nterior pituitarygonadotropins apparently as the result of transmis-
don of the stimulus to the pituitary through the nervous system.
FIG. 50.-Cystic ovaries in a nonpregnant sow. This animal exhibited irregular estrual
periods and had been bred repeatedly during a 5-month period without the establishment of
pregnancy. (Courtesy 0/ Dr. A. V. Nalbandov, Department 0/ Animal Science, University
0/ Illinois.)
The process of ovulation is a gradual one and not in any sense explosive
or cataclysmic. The ripened follicle protrudes beyond the surface of
the~vary, one or several areas of the follicular wall become avascularized,
thin, and more or less transparent until the wall finally ruptures at one
point. The follicular fluid flows through the ruptured area carrying the
ovum with it.
During the final stages of follicular development the chromosomes in
the egg are reduced to one-half the somatic number by the budding off
of a polar body containing the chromosomes but little or no cytoplasm.
As the time of ovulation approaches, the germinal hill (discus proligerus
or cumulus oophorus) breaks up, freeing the ovum in order that it -ffill
not be trapped in the follicle.
There have been many suggested schemes in regard to the alternation
154 BREEDING AND IMPROVEAfENT OF FARM ANIMALS
of ovarian function. The theory that one ovary produces male and the
other female offspring is obviously unworkable in mammals in view of
what is known of the genetic determination of sex. There is no evidence
to substantiate the idea that the ovaries alternate in egg production.
Alternation sometimes occurs, but the usual pattern of ovarian activity
is a random one. The removal of one ovary usually has little or no
effect on subsequent fertility.
Breeding in Relation to Time of Ovulation.-The time of ovulation is
of very great practical importance to livestock breeders. Relatively
little was known about the time or process of ovulation before 1930, and I
in cattle the matter had received little attention prior to 1940. The
"average" time of ovulation of the various classes of farm animals is '
known, but as in all biological phenomena there is considerable individual
variation. There is as yet no practical method for predicting exactly
when a particular animal will ovulate. In animals which ordinarily
produce only one egg during the heat period, the breeder's problem is to
ensure that large numbers of vigorous fresh sperm will be present in the
Fallopian tubes at the time of ovulation. In the sow, the theoretically
optimum time would be shortly before the majority of the ova are
liberated. In the other farm animals fertilization is an all or none
phenomenon, since only one ovum is ordinarily involved. In the sow
pregnancy may occur even though the percentage of eggs fertilized may
be very low. Breeding swine either too early or too late in the heat
period may have an effect on litter size.
Cows and ewes should be mated during the latter part of the heat
period. The optimum time is shortly after the middle of estrus. How-
ever, since heat is relatively short in those species, random breeding at
any time during heat will result in fair breeding efficiency. In the sow
and mare the longer heat period may influence fertility. At the Purdue
Station it is recommended that gilts be bred early during the second day
of heat and that sows be mated sometime during the second day. Sows
tend to stay in heat longer than gilts and can be mated a little later.
These recommendations are made with the assumption that the breeder
will try the animals each day to determine the presence of estrus. In
swine, if the boar is not being used too heavily, conception rate and litter
size can usually be improved by mating the females more than once'
during heat. Mares have the lowest conception rate of all the farm
animals. This is due to the limited survival time of sperm in the female
and to the uncertainty of the length of heat and the time of ovulation in
any particular mare. Studies by the writer showed that the highest
conception rate occurred when the mares were either naturally or artifi-
cially inseminated each day beginning on the second day of heat and '.
.~ , , " r,.'
.. "!
II
THE FEMALE'S PAR'l' IN REPRODUCTION 155
continuing until the end of estrus. This system is not practical under
farnf conditions. Breeding during the third, fourth, and fifth days of
heat produced excellent results. Some breeders report equal success by
breeding every other day beginning on the third day of heat. If a mare
can be bred only once, the fourth day of heat is usually the best.
Vitality of Ova.-The survival of sperm within the female reproductive
organs is long by comparison with the viability of ova. The early work
of Lewis at the Oklahoma Station in 1911 is classic for its type. Of 13
sows force-bred a total of 34 times following the cessation of estrus, only
1 became pregnant. Lewis correctly concluded that the ovum does not
retain its vitality for more than a few hours following ovulation. More
recent studies have shuwn that not only is the life of the ovum short but
that developmental abnormalitie~ may be related to the age of the egg
at the time of fertilization. In the guinea pig Blandau and Young (1939)
found that when fertilization was delayed for 8 hours following ovulation
there was an increase in the number of sterile inseminations, a decrease
in litter size, and an increase in abnormal pregnancies. No normal
development followed insemination more than 20 hours after OVUlation,
and complete sterility occurred after 32 hours. Similar results were
obtained in the rat (Blandau and Jordan, 1941). Records of cattle
inseminations have consistently shown that very poor conception occurs
when insemination is carried out 12 hours or more following the end of
heat (Trimberger and Davis, 1943).
Superovulation.-Following the discovery of the gonadotropic prin-
ciples, numerous attempts to induce estrus and ovulation have been made.
For the most part the studies have been concerned with the induction
of normal estrual cycles and ovulation rates typical of the species. How-
ever, a number of investigations of superovulation, the production of
large numbers of ova, have been carried out.
It was shown that certain gonadotropic materials would produce super-
ovulation in rats and mice, that many of the eggs were fertilized, that
there was a relatively high rate of embryonic mortality, but that larger
than normal litters could be produced (Engle, 1927; Cole, 1937; Evans
and Simpson, 1940).
This problem has been investigated in both cattle and sheep at the
University of Wisconsin by Casida et al. (1940-1944). Twenty-four ewes
treated with follicle-stimulating extracts had a total of 576 corpora
lutea, and 357 ova were recovered. All the ewes were artificially insemi-
nated. Seven of the animals had no fertilized ova, and 17 of them
yielded from 2 to 19 ova in varying stages of cleavage. The responses
obtained varied with the gonadotropins used and the stage of the estrual
cycle at the time of treatment (Murphree et al., 1944).
156 BREiZVING AND IMPROVEMENT OF FARM A N IMALS
Following the demonstration that superovulation and fertilization

could be obtained, a study of the survival of induced multiple pregnancies
was made (Casida et al., 1944). Superovulation was induced in 25 ewes.
It was found that at 2 to 5 days following inseminaHon there were 9.2
normal embryos per ewe, 3.4 normal embryos at 14 to 27 days and an
average of 0.8 normal embryos at 30 to 37 days of gestation. No lambs
were produced. Similar results
had previously been obtained with
cattle (Casida et al., 1943).
There is considerable doubt as to
the desirability of multiple preg-
nancies in most species of farm
animals. In the horse, twins or
triplets are very undesirable.
Twins are seldom successfully raised
and most frequently both die. In
cattle twins occur once in every 50
or 60 births. The high incidence of
the freemartin condition in twins
of opposite sexes, the increased
mortality rate of twins, and the
fact that there seem to be more
difficulties in cows which produce
twins, all combine to make multiple
births of questionable value in
practical cattle production.
Multiple births are considered
desirable in both sheep and swine,
FIG. 51.-Superovulation in the rabbit. but even in these species there are
Uterus of rabbit 13 days after termina-
tion of treatment with gonadotropins there problems. Under range conditions
were 33 placental sites of which 18 contained single lambs are often preferred
normal embryos, 4 contained degenerate
embryos, and the rest were empty. (Cour- because of their greater size and
tesy of Drs. Casida, Warwick, and Murphree, vigor. In some cases, high pro-
University of Wisconsin .)
lificacy of s",jne has been associated
with reduced birth weight of the pigs and thus a lower rate of survival.
This problem is not general in swine, but is mentioned because there
are limits to prolificacy.
It is possible that the capacity of the uterus, in regard to size, cil'cu-
lation efficiency, nutritive and excretory capacity, may be lilniting factors.
AJ3 pointed out by the Wisconsin group, superovulation may result in the
production of abnormal ova, capable of fertilization but not of normal
intrauterine development.
Castration.-This operation in the female is generally called spaying.

It has not been practiced to as great an extent as in the male and is
becoming rare in practical livestock production. One of the reasons is
that any operation which involves entering the abdominal cavity necessi-
tates a skilled surgeon and always involves risk. The chief reason for
the decline in spayed animals is that the results are not so striking as in
castration of the male. Many studies of its value have been made; the
results at the Purdue Station are typical, although not original. Spayed
beef heifers tend to have better finished carcasses than normal females,
but they do not grow so rapidly nor do they utilize feed so efficiently.
It is doubtful if the practice is economically sound.
This terminates, temporarily, the discussion of the primary organs of
reproduction in the female, the ova,ries. We will return to them again
when fertilization and implantation are dis.cussed and in the consideration
of fertility, sterility, and breeding efficiency in subsequent chapters.
Fallopian Tubes.-The Fallopian tubes, ovarian tubes, or oviducts,
act as excretory ducts of the ovaries, conveying ova from the ovaries to
the uterus. They are not, however, in direct continuity with the glands
but rather partly in continuity with and partly attached to them, both
ovaries and oviducts being suspended in a fold of the broad ligament, the
mesosalpinx. Their anterior ends open into the peritoneal cavity, and
posteriorly they join the horns of the uterus or the uterus proper.
Their analogues in the male are the vasa deferentia. The Fallopian
tubes consist of an inner mucosa and outside this a circular and a longi-
tudinal muscularis and an external serous coat. They are firm to the
touch and have a diameter of about 0.1 in. in the cow. The fimbriated,
or funnel-shaped, end of the oviduct nearest the ovary is called the
infundibulum. The middle portion consists of the dilated ampulla and
the constricted isthmus, whereas the portion entering the horn of the
uterus is called the intramural portion. The mucosa of the oviduct is
thrown up into folds and is richly supplied with lymphatics and thin-
walled blood vessels, indicating a secretory activity. The epithelial
lining of the oviduct is often ciliated, and the ciliary motion is chiefly
toward the uterus. Spermatozoa that are deposited in the vagina or
uterus at copulation pass up these tubes to their upper end, where
fertilization usuaily takes place as soon as the ovum is liberated.
Experimental removal of one ovary and tying off of the opposite ovi-
duct has' been followed by pregnancy. In some way, in other words,
the egg produced by one ovary moved across the peritoneal cavity to
the other oviduct and on down to the uterus. Dukes suggests that this
may be due to the action of the cilia in the infundibulum setting up a
.
current itJ. the.., peritoneal fluid, the cilia beating toward the uterus .
158 BREEJ)ING AND IMPROVEMENT 01<' FARM ANIMALS
The fertilized ovum or zygote passes down one of these tubes in from
3 to 4 days in domestic animals to find lodgment in a horn or in the body
of the uterus. For this purpose, aid is perhaps derived in some species
from the cilia or microscopic fingerlike projections from each of the cells
that line the inside of the oviduct and by a waving motion tending to
push the ovum downward. In other species, however, cilia are lacking
in the oviducts, so that the downward passage of the zygote is probably
here referable to the peristaltic action of the musculature of the tubes,
even though this is made up of small waves rather than one continuous
wave. Like the rest of the female genital tract, the oviducts and their
glands become more active at estrus. Fallopian tubes, the lumina of
FIG. 5~.-Uterine lining of the mare. 1. First day after foaling, endometrium specialized
for pregnancy. 2. First day of foal heat, endometrium not yet repaired following foaling.
which are smaller than the lead in a pencil and greatly convoluted, act
somewhat in the nature of a seal, thus preventing anything harmful
passing down into the pregnant uterus. It can readily be seen, too, that,
if infection of any sort once found lodgment here, it would be exceedingly
difficult to combat or to eliminate. The most usual type of sterilization
operation in woman consists of the severing and tying off of the oviducts
so that sperm are not able to come into contact with the egg.
Uterus and Horns.-The 9viducts in the higher animals terminate in
the horns of the uterus, and the two horns terminate in the body of the
uterus. It is here, of course, that the embryo undergoes its prenatal
development. In the human, the horns of the uterus are practically
lacking. In the mare the body of the uterus is nearly equal in length to
that of the horns, whereas in the cow, ewe, and sow the body is short a.nd
... ,.
the horns long. The horns, or cornua, are long and straight in the dog
and cat, l~ng and coiled or folded in the sow, and long and curved, some-
what like a ram's horns, in the cow, ewe, and goat. The horns and
uterus proper are lined with a highly vascularized mucosa bounded by
two layers of muscle fibers, the inner circular and the outer longitudinal
layers. This muscular structure provides the uterus with the degree of
elasticity necessary for pregnancy and also provides the means for the
final expulsion of the fetus at time of parturition.
The external covering of the horns and uterus is the serosa. The
mucosa of the uterus is highly vascularized and contains many glands
and lymph spaces. During sexually mature life it is undergoing rhythmic
changes due to the heat periods, the inner surface being renewed period-
ically preparatory to the release of eggs from ovaries. The glands
increase in size and activity at this time also. The fertilized egg descends
the oviduct and finds lodgment in or on the uterine mucosa. Parts of
the inner uterine wall become modified, forming the maternal part of the
Illacenta by means of which the embryo receives its materials for growth.
This may be a circular or disklike area, as in woman and the mare, a
zone or band, as in the bitch or cat, or rows of cotyledons or caruncles, as
in the cow and ewe.
Cervix.-The uterus terminates posteriorly in the cervix, or neck of
the womb. This is a very thick-walled portion of the genital tract. It
is lined with simple epithelial cells that secrete copious amounts of mucus
into the small canal passing through the cervix. The thick muscular
walls are thrown into many folds that expand at time of parturition to
allow passage of the fetus and then retract in a few days to their con-
tracted state. The cervix forms an effective seal at the posterior end of
the uterus.
In the mare, the cervix is 2 to 3 in. long and 17'2 in. in diameter.
The cervical canal is straight and never fully closed in the nonpregnant
state. In the cow it is about 4 in. long with a wall 1 in. or more in thick-
ness and is always tightly contracted except at parturition. The canal
through the cow's cervix is spiral, folds of tissue being thrown back on
each other making it difficult to dilate or to insert an instrument of any
sort. In the ewe, the cervix is about l}~ in. in length, and the lumen is
tightly contracted and difficult of entrance as in the cow, whereas in the
sow it is about 4 in. long, partly occluded by rounded prominences on
, its interior and has no intravaginal projection as found in the mare, cow,
and ewe.
The character of the mucus secreted by the cervix varies considerably:.
at different times of the reproduction cycle. During estrus, the mucus
is thin and watery, the cervix also somewhat relaxed. Roth these con-
ditions render passage of the semen into the uterus easier. Following
heat, the mucus thickens and remains in this condition until the next
heat period. Following the onset of pregnancy, the cervical mucus
becomes very thick, forming a more or less solid and permanent plug
that seals the cervix and thus protects the developing embryo. Disturb-
ance of this plug is often followed by abortion, but the reaction here is
quite variable among both individuals and species. Owing to its stlUC-
ture, especially in the cow, pathological conditions in the cervix are
particularly hard to treat.
Vagina.-From the cervix posteriorly to the urogenital sinus or vesti-
bule, the female genital tract is termed the vagina. In farm animals it
Jew. 53.- Changes ill the vaginal epithelium of the mare during the estrual cycle. 1. First
day of estrus. 2. Seventh day of estrus , observe increased number of epithelial cells and
flattening of surface cells. 3. Five days after estrus.
lies horizontally below the rectum and above the bln.dder. This portion
of the tract is relatively simple both in structure and function. It consists
anatomically of the usual three layers of tissue: mucosa, muscularis, and
serosa. The epithelial lining of the vagina is very responsive to cyclic
sexual changes, and these changes have been intensively studied and
correlated with the phenomena of estrus, ovulation, etc., in laboratory
animals and in woman. During estrus the cellular epithelial lining of
the vagina proliferates, becomes highly stratified. During interestrus it
is less active and consists of fewer layers of cells. In the mare the vagina
is about 6 to 8 in. long; in the cow, 8 to 10 in.; in the ewe, 3 to 4 in.; in
the sow, 4 to 5 in. The vagina generally becomes longer during the
pregnant state.
The function of the vagina is twofold: (1) to receive the intromittent
THE FEMALE'S PAWl' I N REPRODUCTION 161
organ of the male in copulation and pl'obably in most cases the ejaculate
also, and (2) to provide a passage for the fully developed fetus at term.
Vulva and Clitoris.-The female genital tract terminates posteriorly in
the vulva or external orifice. With the onset of heat, or estrus, these
structures become congested and enlarged, particularly in the sow,
somewhat less so in the cow, the mare, and the ewe. The clitoris, the
small erectile organ homologous to the penis in the male, is situated just
inside the portion of the vulva farthest removed from the anus. Suffi-
cient stimulation of the clitoris results in a se)o,,'ual orgasm in the female.
Fertilization.-Hormones from the anterior pituitary cause a number
Second polar body
FIG. 54.- A rare photomicrograph of the early fertilization process in the rat taken with the
phase microscope. (Coul·tesy of Dr. Richard J . Blanda", Uni'IJersity of Washington, School
of Medicine.)
of follicles to begin their final rapid growth and maturing processes. F S:.H
Eventually one or several (depending on individuality and species)
follicles become fully mature, a process which includes the reduction in
the number of chromosomes to the haploid number through the budding
off of a polar body containing one member of each pair of chromosomes.
The follicle finally ruptures, and the egg is liberated during estrus. At
this time the female will accept service by the male, and billions of
spermatozoa are introduced into the distal end of the vagina, into the
• cervix, or possibly in some species directly into the uterus. In the course
of a very few hours masses of spermatozoa have been transported through
the uterus and tubes to the upper end of the latter.
• The ovum drops, or is "Swept, into the infundibulum, and one sperma-
tozoon immediately pierces it. This stimulates the ovum to throw off the
second pglar .-~ Immediately following, the tail of the sperm is

--aropp(;d or discarded, the male and the female pronuClei approach each
other within the ovum, the centrosome biougn-t·-iii-oy·tIie middle piece
of the sperm divides, a portion moving to each pole of the cell, and the
paternal and maternal chromosomes finally become arranged in the
equatorial plane. This process is known as fertilization and is completed
with the fusion of the male and female pronuClei and th~ -arrangement of
the chromosomes in the equatorial plane.- It }.I; probable that fertil-
ization is completed within a few hours after service by a normal male
on a female in estrus; other sperm may penetrate the zona pellucida but
only one enters the ooplasm to effect fertilization.
In recent years~~m~;nan ova, especially from the rabbit, have been
secured and have been fertilized in test tubes, the process having been
observed. They can then be inserted into the uterus of a properly
stimulated female, and embryonic development will proceed normally to
term in the foster mother. Pincus (1939) has recently caused.develop-
ment of the rabbit ovum without stimulation of the sperm. This is the
long-sought mammalian parthenogenesis and occurred, according to
Pincus, because of an early restoration of the diploid number of chromo-
somes through nuclear division without cell division.
In the multiparous animals it is seldom that all the eggs liberated are
fertilized and implanted, the loss in the sow having been determined to
average 20 to 30 per cent and to run as high as 50 per cent. Multiple
ovulations in the cow and mare are also known to be more numerous
than multiple births, which in these species is a good thing. It is better,
too, from a practical standpoint, for a sow to have 10 or 12 good-sized
vigorous pigs than 20 or 25 small, weak ones, the large majority of which
generally die very early. .
The fertilized egg, or zygote, immediately begins (1) to grow by cell
division and (2) to dlilscenctthe~anopiaii tuheTnto a horn or the uterus
-proper: - The egg is huge compared with other body cells because of its
stored yolk. Its early segmentation does not involve an increase in
size but rather an increase in cell numbers and a consequent reduction in
the size of the cells to that characteristic of other body cells of the specie·s.
Cell divisions proceed at first at the rate of one or two a day. The
passage down. the tube is probably effected by both ciliary and muscular
.activities of the oviduct and takes from 3 to 4 days. The oviduct also
produces a secretion that may help to nourish the egg in its passage
downward. Normally the mass of cells reaches a horn or the uterus
proper (though ovarian, abdominal, and tubal pregnancies occur now
and then) and becomes attached to the uterine mucosa where embryonic
development proceeds.
\.
Attachment generally occurs in the larger farm animals in the horn of

the uterus corresponding to the ovary that contains the corpus luteum,
but many cases of migration and implantation in the opposite horn have
been recorded. In multiparous animals like the sow, implantations occur
at more or less regular intervals throughout the two horns, regardless of
the fact that most of the eggs may have been produced by one ovary or,
as demonstrated by Warwick, after removal of one ovary. The exact
mechanism regulating the site of implantation is at present unknown.
Corpus Luteum.-The corpus luteum (yellow body) is an intermittent
or periodic gland of internal secretion that is formed 'Q::.it.mn.th(j follicT~
after ovulation; Upon rupture of the folliCle and liberation of the ovum,
there ensues a rapid vascularization of the granuloSfl, and theca cells that
surrounded the follide and sometimes an escape of bloocfIrom the theca
into the follicular cavity. :The cavity closes over and with the retained
blood clot is known as the corpus hemorrhagicum, which is the earliest
stage in the development of the corpus luteum. The connective-tissue
cells of the theca folliculi, which have arisen from the ovarian stroma
surrounding the follicles, gradually grow in and partially fill the follicular
cavity, the blood clot gradually undergoing absorption. The granulosa
cells of the undischarged follicle hypertrophy. The rapidly multiplying
cells of the theca,_unc!e~.stirnlilation.ofLH.Jr~he anterior pituitary,
absorb yellow lipoid pigment and fat droplets, which give to the .corpus
luteum nscnaracfenstlc yellow c01or;-which ;;aries, hO'wever, in differe~t
species:--Blood vessels are carried in with the ingrowing tissue from the
theca, making the corpus luteum a highly vascularized gland~ The
corpus luteum grows quickly, soon filling the follicular space and usually
protruding from the ovary, its later growth being due to enlargement of
the cells rather than to cell multiplication. .
If the a~_es, th~~_rpus l~t_~l!lll_~Q.!l_ti!?Jles<O enlarge until
about the mIddle of the gestatIOn period.. It then often begins to shrink
in size and to decrease in secretory activity until it finally ~~nerates
and becomes the ~.E~ albicans. In some species it is' not '~~h~fly
absorbed until a considerable time after parturition. If conception does
not occur, the corpus luteum does not grow so large, and progesterone
secretion de..~~ines before the onset of the next estrual period.
The functi~;_;--of the corpus luteum hormone, progesterone, are as
follows: (1) pI;ogestational changes in the endometrium of the uterus
. that are neceSsary to se;sitize the uterine wall before implantatIon of
the morula can take place as well as the early development of the placenta,
(2) the regll!.!:ttion of the estrolls...cycle by preventing estrus and ovulation
until the corpus luteum reaches a certain stage of atrophy, (3) the main-
tenance o~4) to assist in the relaxation of the pclvis -at
164 BREEDIl'W AND IMPROVEMENT OF FARM ANIMALS
partur~d (5) some participation, it seems, in the development of

---ttleIilammary gland,_ -----_
~
The amouiit of luteal tissue necessary to lead to successful implanta-

tion has been determined for some species. It has been shown in many
species that removal of the corpus Iuteum shortens the sexual cycle by
quickly inducing the next heat period, which in the cow generally follows
within 48 hours.
In many species re~oval of the corpus luteum during pregnancy results
in the absorption of the fetus or an abortion within 1 to 10 or 12 days.
If performed very soon after fertilization, abortion or absorption is
practically certain to follow. Some species can evidently safely dispense
with the corpus luteum much earlier than others, and in those cases it is
thought that the placenta may then take over some of the functions of
the missing corpus luteum, for the relaxation of the pelvic muscles occurs
normally as does the development of the mammary glands.
Placenta.-The placenta consists of two parts: (1) the maternal
placenta, which is of maternal origin, and (2) the fetal placenta, which is
of fetal origin. The fertilized egg, or zygote, obtains its nourishment at
first from its own stored yolk and perhaps also from its investing enve-
lopes, the zona pellucida and the corona radiata. Next it subsists on
materials secreted by the uterine glands, absorbed directly at first and
obtained bter by means of the yolk sac. Finally, it becomes embedded
in or attached to the uterine ,vall. The placenta is developed through
which food materials pass inward to the embryo, and waste materials
pass out\'v'ard from it by means of the placental blood vessels.
Following segmentation, which produces a solid mass of cells (morula
stage), a central cavity is formed within the mass of cells (blastula stage.)
Then there ensues an invagination (gastrula stage), and the embryo
proper develops from the invaginated cells. The portion remaining out-
side is called the trophoblast. At first each blastomere is nourished
independently, but, following invagination, the trophoblast is concerned
wholly with the nourishment of the embryo that is to be elaborated. The
fusing of the external wall of the allantois, an outpouching from the hind-
gut region, with the trophoblast produces the chorion, or afterbirth:
The trophoblast is highly vascularized by outgrowths from the blood-
vessels of the allantois.
In the pig, the blastocysts are spherical for 10 days, but then they
rapidly elongate, and by the fourteenth day they together fill the whole
length of the uterus. Foldings appear in the trophoblast that fit into
folds of the uterus, which increase the area of contact. The trophoblast
fits closely to the surface epithelium of the uterus and sends protoplasmic
processes between the cells and perhaps down to the underlying layer of
\
\
dilated capillaries. Villi do not form in the pig, and the uterine mucosa
has no special cotyledonary areas. The placenta of the pig is, therefore,
of the contact type (semiplacenta) in contrast to the burrowing type (true
placenta), e.g., human. Materials secreted by the uterine glands are
absorbed by the trophoblast and transported to the embryo by means of
the allantoic blood vessels.
In the mare the lymphatic system of the uterine mucosa is greatly
developed, and villi are found in the allantois that fit into tiny crypts of
the uterine wall.
FIG. 55.-Pregnant uterus with cotyledons. A, uterus; BB, maternal cotyledon; D, fetal
cotyledon. (From U.S. D epartment of Aoriculture, Diseases of Cattle. )
In the cow and the ewe there are many special points of attachment
between the chorion and the uterine wall. These are called cotyledons
(buttons), and these species are, therefore, of the polycotyledonary type.
There are 70 to 130 cotyledons in the ewe and cow. During pregnancy
the cotyledons become spongy, owing to increased vascularization and
the enlargement of the lymphatics. Villi containing blood vessels from
the allantois grow out from the cotyledonary areas of the trophoblast and
become embedded in the tissues of the cotyledons of the uterus. The
uterine glands in the intercotyledonary areas as well as the surface
epithelium also secrete copiously during pregnancy. This secretion,
together with its contained formed constituents (red-blood corpuscles,
leucocytes, fat globules, salts, glycogen, etc.) from the blood and lymph
systems, is known as uterine milk. It seems probable that the exchange
of oxygen and carbon dioxide is effected in the cotyledons, where glands
are absent and where the maternal and fetal blood systeII+g come into
close contact.
The function of the placenta is, of course, the elaboration of the
materials necessary for fetal development and the development of a
transporting system for conveying food materials to the embryo or fetus
as well as \Va. te products away from it. It should be remembered that
,
o o
o
FIG. 56.-Embrronic calf at the two-cell stage. (Courtesy Bureau of Dairy Industry, U.S.
Department of Agriculture, and Carnegie Instit ution of Waahington.)
the mother contributes no blood as such to the embryo. If we think of

an island out in midstream as the embryo, then the intervening river
and boats plying back and forth might be thought of as the placenta
foetalis and the sandy beach of the mainland as the placenta uterina.
The placenta performs an endocrine function as well; and in some
species secretes estrogens, progesterone, and gonadotropic hormones.
Embryological Development.-As noted previously, fertilization, or the
union of the male and female pronuclei, each with its haploid number of
chromosomes, generally occurs in the Fallopian tube following ovulation.
The zygote within the smrounding capsule, or zona pellucida, immedi-
ately starts its descent to the uterus. Cell division is inaugurated by the
entrance of the sperm and continues during the descent. The time .. ,
...
....,
'PRE FEMALE' S PART I .V REPROD UCTION 167
required for the developing zygote to traverse the oviduct probably varies
somewhat in different species but consumes a period ranging from 1 or 2
up to several days. During this time the egg lives and grows upon its own
stored nourishment found in the yolk.
Upon entrance into the uterus, the group of cells, probably in the
morula stage, takes lodgment upon or in the uterine wall, absorbing
nourishment for growth from the uterine secretions until its own develop-
ing eJl.'traembryonic membranes can set up a connection with the blood
FIG. 57.-Stages in the implantation of the guinea pig ovum. 1. Note bipolarity of the
ovum and beginning of attachment to the uterine epithelium. 2. Observe loss of nuclei in
the uterine epithelium. 3. Implantation approximately one-half completed. 4. Implan-
tation almost completed. (Courtesy of Dr. Richard J. Blandau, University of Washington,
School of ]I{ edicine.)
vessels of the mother's uterus. It should be understood that the nutrition

of the embryo is indirect after the initial stages. The embryo lives first
upon the stored nutrients in .the yolk of the egg, next upon the uterine
secretions, then upon materials carried to the placenta in the mother's
blood stream, then, in mammals following parturition, upon milk secreted
also from the mother's blood, and, finally upon food that the young has
learned how to secure for itself. The embryo never gets any blood as
such from its moj;her. It builds up its own blood, a liquid tissue, in the
same manner that it builds up all the other tissues of its body, bone,
muscle, nerve, etc., from materials supplied from the mother's blood
stream.
The momla, or mulberry like, stage is succeeded by the blastula, or
hollow-sphere, stage, then through an infolding of a portion of the cells
the gastrula stage is formed. The outer portion is called the trophoblast,
FIG. 58.-Fetal calf within its membranes. (From U.S. Department of A{l1'wulture, Disea8es
of Cattle.)
the embryo proper being retained outside the blastocoele. A yolk sac is
developed dorsally, and, althoug!:l little yolk is present in the egg of
mammals, the yolk sac functions for about 3 weeks in the cow, sheep, and
pig, and about 6 weeks in the mare, as the medium for supplying nourish-
ment to the embryo. Folds grow out from all sides of the embryo and
grow up over it, fusing and enfolding it. This is the amnion, which
becomes filled with a watery fluid in which the embryo floats, thus being
THE FEMALE'S PART IN REPRODUC'l'ION 169
protected from concussion. From the hind gut there forms an out-
pouching that grows out and enfolds the amnion and its contained
embryo. This is the allantois, which is also filbd ,,,ith a watery fluid.
Its outer wall fuses with the outer wall of the blastocoele, forming the
placenta. Food materials are carried down into the maternal placent&,
there transfuse into the capillaries of the allantoic circulation, and art
carried to the embryo by means of the blood vessels through the umbili
calor navel cord. Waste products of the embryo are eliminated by thE'
reverse process. In ruminants, the connection between chorion and
uterine mucosa occurs at the cotyledons, or buttons.
Fro. 59.- Bovine monster. Embryonic abnormalities of this type are the exception rather
than the rule.
Embryonic development is a complicated and marvelous process. The

zygote consists of one minute lmdifferentiated cell. The baby or calf
or pig consists literally of billions of cells that have all come from the
original zygote. Some have been differentiated into long-muscle cells,
others 'into secreting cells (liver, pituitary, etc.), others into bone cells,
still others into brain cells. We know little at present concerning the
cause of this specialization and of the directing forces involved. It seems
peculiar from a genetic standpoint, for all the cells of the body have
similar chromosomes. Internal stimuli within the cytoplasm of the cells
as well as external stimuli in the different regions (organization centers)
of the embryo are no doubt operative in bringing about the transforma-
tion from the one undifferentiated cell to its billions of differentiated
descendants.
Recapitulation.-The ovary, under the influence of the anterior pitui-

tary gonadotropins, produces mature Graafian follicles and ova. The
ovary also secretes the female sex hormone, estrogen, in sufficient quantity
to cause estrus. Mating ordinarily occurs during heat, and toward the
latter part of heat the ovum is liberated from the Graafian follicle as the
result of the ovulatory process. Immediately following ovulation a
corpus luteum is formed in each ruptured follicle and the hormone
progesterone is produced. This is essential if the uterus is to be properly
prepared for the establishment and maintenance of pregnancy. If the
animal has been mated, the spermatozoa immediately start their journey
toward the ovary. They pass through the cervix, body, and horns of
the uterus and into the distal ends of the Fallopian tubes where the.~T
await and fertilize the ovum or ova. The ovum is transported by the
follicular fluid as it oozes from the ruptured follicle into the infundibulum
and thence into the Fallopian tube where it must be fertilized'; by a sperm
within a fmv hours if fertilization is to occur. Ordinarily only one
spermatozoon pierces the wall of the ovum and only one male pronucleus
can unite with the female pronucleus if normal development is to follow.
The fertilized ovum now starts its journey down the Fallopian tube and
becomes attached to a suitable area of the uterine horn or };lody for its
prenatal development. All these events are under endocrine control and
require the most exact morphological and physiological changes and
coordination if fertilization and normal intrauterine development are to
take place. Embryonic differentiation and fetal development proceed
in accordance with the species pattern and end, in ma~mals, with
parturition and lactation.
Chronological Order of Reproductive Physiological Processes.-It is
essential for the student and breeder to keep in mind the various physio-
logical processes in the female that are concerned in reproduction. In
order of their appearance they are: (1) maturation of the follicle, (2)
estrus, (3) copulation, (4) rupture of the follicle, (5) fertilization, (6) for-'
mation of corpus luteum, (7) implantation, (8) gestation, (9) parturition, .
(10) lactation, (11) absorption of the corpus luteum.
Summary.-The female's part in reproduction consists of the secretion
of viable ova or eggs in the ovary, their rupture into the infundibulum of
the uterine tubes or oviducts, where each is joined by a sperm that con-
tributes paternal determiners of potentialities and incites the marvelous
process of embryonic development which will eventuate in a few weeks
or months in a new and unique individual. To use again our analogy of,
an assembly line, the ova originate at the far end of a big tube; they very
quickly (if at all) pick up a sperm and then pass into a roomy cavity
where all the marvelous parts are slowly fashioned and fitted together
into a completely new individual. As soon as the job is completed, the

front door of the uterus, the cervix, which after admitting the sperm was
tightly locked to prevent any harmful intrusions, is thrown open and the
new individual emerges into a new and strange world. To be successful
the breeder must thoroughly understand these processes and the factors
affecting any of them for good or ill, in order that breeding operations
may be properly timed and conditions made optimum for the eventual
bringing into being of superior animals.
References
Books
GILBERT, M. S. 1938. "Biography of the Unborn," The Williams & Wilkins Com-
pany, Baltimore.
HAMMOND, J. 1940. "Farm Animals, Their Breeding, etc.," Longmans, Green &
Co., Inc., N ew York.
- - - 1927. "The Physiology of Reproduction in the Cow," Cambridge Univer-
sity Press, New York.
PATTEN, B. M. 1931. "The Embryology of the Pig," The Blakiston Company,
Philadelphia.
TURNER, C. D. 1948. "General Endocrinology," IV. B. Saunders Company,
Philadelphia.
ALLEN, E. 1922. The Oestrous Cycle in the Mouse, Amer. Jour. Anat., 30:297-371.
- - - and DaISY, E. A. 1923. An Ovarian Hormone: Preliminary Report on Its
Localization, Extraction, and Partial Purification, and Action in Test Animals,
Amer. Med. Assoc. Jour., 81 :819-821.
ANDREWS, F. N., and BOHREN, B. B. 1947. Influence of Thiouracil and Stilbestrol
on Growth, Fattening, and Feed Efficiency in Broilers, Poultry Sci., 26 :447-452.
- - - , BEESON, W. M., and HARPER, C. 1949. The Effect of Stilbestrol and
Testosterone on the Growth and Fattening of Lambs, Jour. Anim. Sci., 8:578-582.
BISSONNETTE, T. H. 1941. Experimental Modification of Breeding Cycles in Goats,
Physiol. Zool., 14 :379-383.
BLANDAU, R. J., and JORDAN, E. S. 1941. The Effect of Delayed Fertilization on
the Development of the Rat Ovum, Amer. Jour. Anat., 68:275-291.
- _ - and YOUNG, W. C. 1939. The Effects of Delayed Fertilization on the
Development of the Guinea-Pig Ovum, Amer. Jour. Anat., 64:303-329.
CASIDA, L. E. et al. 1943. Effects of Pituitary Gonadotropins on the Ovaries and
the Induction of Superfecundity in Cattle, Amer. Jour. Vet. Res., 4 :76-94.
- - - , WARWICK, E. J., and MEYER, R. K. 1944. Survival of Multiple Pregnan-
cies Induced in the Ewe Following Treatment with Pituitary Gonadotropins,
Jour. Anim. Sci., 3:22-28.
COLE, H. H. 1948. The Hormonal Control of Estrous Behavior, Western Jour. of
Surgery, Obstet. and Gynec., 66 :503-506.
1937. Superfecundity in Rats Treated with Mare Gonadotropic Hormones,
Amer. Jour. Physiol., 119 :704-712.
DEMPSEY, E. W., HERTZ, R., and YOUNG, W. C. 1936. The Experimental Induc-
tion of Oestrus (Sexual Receptivity) in the Normal and Ovariectomized Guinea
Pig, Amer. Jour. Physiol., 116:201-209.
DINUSSON, W. E., ANDREWS, F. N., and BEESON, W. M. 1950. The Effects of

Stilbestrol, Testosterone, and Thyroid Alteration on Growth and Fattening of
Beef Heifers, Jour. Anim. Sci. (in press).
DOlSY, E. A., VELER, C. D., and THAYER, S. 1929. Folliculin from the Urine of
Pregnant Women, Amer. Jour. PhYS1:ol., 90 :329-330.
ENGLE, E. T. 1927. Pregnancy Following Superovulation in the Mouse, Soc.
Expt. BioI. and M cd. Proc., 25 :84-85.
EVANS, H. M., and SIMPSON, M. E. 1940. Experimental Superfecundity with
Pituitary Gonadotropins, Endocrinology, 27 :304-308.
LEWIS, L. L. 1911. The Vitality of Reproductive Cells, Okla. Agr. Expt. Sta. Bul. 96 ..
LONG, J. A., and EVANS, H. M. 1922. The Oestrous Cycle in the Rat and Its
Associated Phenomena, Calif. Univ. Mem., 6:1-148.
MURPHREE, R. L., et al. 1944. Potential Fertility of Ova from Ewes Treated with
Gonadotropins, Jour. Anim. Sci., 3 :12-21.
SYKES, J. F., and COLE, C. L. 1944. Modification of Mating Season in Sheep by
Light Treatment, Mich. Agr. Expt. Sta. Quart. But., 26 :250--252.
TRIMBERGER, G. W., and DAVIS, H. P. 1943. Conception Rate in Dairy Cattle by
Artificial Insemination at Various Stages of Estrus, Nebr. Agr. Expt. Sta. Res.
Bul.129.
CHAPTER VI
REPRODUCTIVE EFFICIENCY
Reproductive efficiency connotes the optimum effective use of the

reproductive powers of all the animals in a herd or flock. In the female
it is the regular production of offspring over a period of years; in the male,
the successful fertilization of the largest number of ova with the fewest
possible services. In the following chapter ,ye will describe some of the
more common types and causes of sterility and lowered fertility. Unfor-
tunately, reproductive efficiency is not limited only by the conditions
which we will discuss. All herds and flocks have some abnormal
individuals in which the failure of reproduction can be accounted for.
Likewise, in all herds and flocks, all of the apparently normal healthy
females do not become pregnant following a single service by a normal
healthy male. There is good evidence that these differences in reproduc-
tive efficiency are often traceable to differences in breeding management
and general care. "\Ve should never overlook the fact that our animals
are individuals and that the best management of one may not necessarily
be the best for another. For the most part livestock management entails
the application of common sense directed both by previous experience
and by a knowledge of scientific principles. The more thoroughly the
principles are understood and the greater the accumulation of experience,
the more efficient should be the management. Without good judgment
or common sense, neither knowledge nor experience, nor both, will avail
to render a livestock-breeding enterprise profitable.
Standards of Reproductive Efficiency.-One of the objectives of every
livestock breeder is to produce a pregnancy as the result of each service.
In species characterized by single births, as in cattle, this would mean
one offspring per service if the process were 100 per cent efficient. In
swine, the- standard is higher, and we might arbitrarily designate the
production of 10 pigs per service as 100 per cent efficiency. This degree
• of efficiency is occasionally attained, but for the livestock population as
a whole the process is far less efficient. It is the authors' opinion that
nearly any livestock breeder can increase reproductive efficiency at least
10 per cent by the adoption of currently accepted breeding practices.
This in.crease can be expected in normal animals without the use of any
special drugs or therapeutic regimes and involves chiefly the judicious
173
174 BREEDING AND IMPROVEMEN1' OP PARM ANIMALS
use and care of sires and the mating of females at the optimum time in
respect to ovulation.
The incidence of absolute sterility in cattle has been reported to range
from 5 to 7.2 per cent. When those animals ,vith complete reproductive
failure are considered separately from those which eventually conceive,
the breeding efficiency of the "fertile" animals can be determined.
Trimberger and Davis (1943) have summarized the data of 16 different
investigations of the breeding efficiency at natural service of nearly 23,000
cattle. In one study of 1,801 Co\YS only 1.32 services per conception
were required. Approximately 78 per cent of the cows conceived on
the first service, 15 per cent on the second, 4.5 per cent on the third, and
1.7 per cent on the fourth or more services. These data were reported in
1918 and to the authors' knowledge, have not since been equaled in a
large group of cattle. At the other extreme are the results of a study of
1,151 cows in which the average services per conception was 2.52. This
difference of only one more service per pregnancy may not seem great.
When the data are examined, it is seen that only 47.4 of the cows con-
ceived on the first service, 22.4 per cent on the second, 12.7 per cent on
the third, and 17.5 per cent required 4 or more services. Delay in the
establishment of pregnancy increases the time interval between calvings,
decreases the number of offspring per animal, and is accompanied by
decreased milk production. When the data from all sources were sum-
marized, it was found that 1.79 services per conception were required.
This figure is fairly typical of the average cattle herd in the United States.
In occasional well-managed, disease-free herds 70 to 80 per cent of the
cows may conceive on the first service. We believe that this degree of
efficiency can be attained if the best practices are used. In poorly
managed herds, or in animals in which there is a high incidence of genital
diseases, breeding efficiency may be less than 40 to 50 per cent. Data on
16,555 cows showed that 64.3 per cent conceived on the first service, 19.9
and 8.5 per cent conceived on the second or third services. App~oxi
mately 7 per cent required four or more services. These animals were
all naturally mated. The breeding results which are being obtained in
artificial insemination associations throughout the United States are
very similar to natural service. As a general rule animals which fail to
conceive following repeated natural service to a fertile sire cannot be
expected to become pregnant when artificial insemination is used.
The breeding efficiency of beef cattle under range conditions is usually
expressed in terms of the annual calf crop, since the bulls are allowed to
run with the cows during the breeding season. The calf crop on the
range may be as low as 40 per cent and occasionally as high as 90 per cent.
Baker and Quesenberry (1944) studied the breeding and calving records
\
\
REPRODUCTIVE EFFICIENCY 175
at the U.S. Range Livestock Experiment Station, Miles City, Mont., fot"
an 18-year period. The average calf crop during 4,753 cow years wa:-;
83.1 per cent. The calf crop varied from 66.1 per cent following the
severe drought of 1936 to 92.5 per cent in 1939. An analysis of the data
did not show age to be significantly correlated with fertility, but it was
revealed that over 50 per cent of the shy breeding cows could be identified
by four years of age.
Swine have generally been considered to be the most fertile of all the
farm animals. In fact, during the 1930's the prolificacy of swine was
considered as one of the leading economic problems of the livestock
industry. The data which are available on the breeding efficiency of
swine do not tend to support the first statement. Phillips (1939) reported
that of 250 sows selected at random from a Bureau of Animal Industry
swine herd 16.8 per cent failed to produce litters. Of those which far-
rowed, 84 per cent conceived on the first service, 12 required two services,
2.5 required three, and 1.5 per cent required four services. In a subse-
quent report, Phillips and Zeller (1941) presented data on conception
failures in 1,354 breeding seasons in six breeds. They found that con-
ception failures ranged from 13.9 to 36.9 per cent. By estimating the
number of ova produced by the different breeds at the time of ovulation
and comparing with actual breeding and farrowing data, these workers
have attempted to account for the losses of ova which occur. Their
estimates of the per cent of ova lost are as follows: "(a) failure of con-
ception, 36.4 per cent, (b) loss from conception to parturition, 19.9 per
cent, (c) loss from parturition to weaning, 14.6 per cent, and (d) loss at
parturition, 2 per cent. This leaves 27.1 per cent of the ova which have
the chance to be fertilized represented by live pigs at weaning." When
the fertility of swine is considered from this standpoint, reproductive
efficiency seems very low. It has been known for many years that in
pregnant sows examined in the slaughterhouse only about 70 per cent
of the ova which have been produc()d are represented by normally devel-
oping fetuses (Hammond, 1921). Thus, when the fertility of swine is
considered in terms of females which do not conceive, those in which few
ova are fertilized, the high mortality of fertilized ova, and losses at
parturition and prior to weaning, it is evident that a serious and challeng-
ing problem exists.
Sheep are commonly flock mated, and the number of services per
pregnancy is therefore difficult to ascertain. Fertility is usually expressed
in terms of the number or percentage of lambs born to each 100 ewes in
one breeding season. Breed differences, plane of nutrition, and general
management practices undoubtedly have a great influence on the lamb
crop. Under farm-flock conditions in Great Britain, lamb crops of 154
per cent in flushed ewes and 140 per cent in nonflushed ewes have been
reported (Nichols, 1926). As reported by Phillips (1939) in the United
States the lamb crop of flushed ewes kept under farm conditions was
reported as 164 per cent and nonflushed 143 per cent. In range flocks
it has been estimated that the average lamb crop is 70 to 80 per cent.
During a 10-year period the average services per pregnancy were 2.9
(Shropshire), 2.9 (Southdown), 2.5 (Hampshire), and 1.4 (Karakul) in
the U.S. Department of Agriculture flock at Beltsville.
Reproductive failure is more common in the horse than in any other
farm animal. Most of the data which are available express fertility in
terms of the per cent of mares which foaled during a particular period.
Mares are frequently mated more than once during a single heat period,
and a large per cent are bred during two or three successive heat periods
before they conceive. If the per cent of services which resulted in the
birth of living foals was determined, the efficiency would be extremely
low.
Data gathered during a 6-year period on the breeding performance of
390 stallions and 28,241 Clydesdale mares showed that 52 per cent of the
mares foaled. During a .)-year period 42.3 per cent of 3,640 Thorough-
bred mares bred to 43 stallions foaled (Robinson, 1921). In no class of
animal is knowledge and skill of more importance than in the horse. In
the authors' experience the foal crop can usually be increased from the
50 per cent level to 70 to 75 per cent by proper breeding management
(Andrews and McKenzie, 1941). In areas where horse breeding is of
importance, as in the Thoroughbred-producing regions, it has been pos-
sible to produce living foals from as many as 80 per cent of the mares
bred each year.
The Influence of Age on Reproductive Efficiency.-Since age can be
accurately measured and is usually closely correlated with growth rate
and sexual development, it is not surprising that many practical rule-of-
thumb recommendations in regard to age and time of breeding have been
developed. At least two important points should be considered by the
livestock breeder in the application of such recommendations. All ani-
mals must be regarded as individuals. Animals of the same age may
differ widely in weight and degree of development and thus differ in
readiness for conception. A second problem for consideration is that
sexual maturity, or puberty, is not a single point but extends over a
considerable period of ,time, depending on the species concerned and
other factors. In the human, for example, Mills (1939) estimated that
there is a lag of about 6.5 years between the onset of the menses and the
establishment of pregnancy in married Filipino girls. In rats Babcock
et al., (1940) reported that the age at puberty in the female is about 65
days but the normal age for breeding is about 100 days. Even when
breeding in rats is delayed until 100 days, however, there is a time lapse
of 11.7 days between opportunity of mating and the establishment of
pregnancy (Asdell et al., 1941).
The problem of age and breeding efficiency has been widely investigated
in cattle and, to a lesser extent in swine and sheep. One of the earliest
investigations of this problem was begun by Mumford at the Missouri
Experiment Station with swine in 1909. Sows bred at 218 days of age
farrowed 8.68 pigs per litter, and reached a mature weight of 415 lb.
Females bred at 479 days farrowed 8.36 pigs per litter and reached a
mature weight of 401 lb., whereas those not mated until 838 days farrowed
only 5.62 pigs per litter and reached a mature weight of 384 lb. The
rate of growth of the early-bred gilts was lowered, presumably because of
the drain of lactation, but the growing period was lengthened (McKenzie,
1928). A more recent study by Stewart (1945) showed that litter size
increased with the age of the gilt from 9 months until 15 months but did
not increase beyond this point. Under practical conditions swine are
usually bred to farrow when from 12 to 14 months of age.
There is not complete agreement as to the influence of age on the
fertility of dairy cattle, probably due to differences in cattle-management
practices through the years and in different areas of the country, and
in types of cattle observed. There is fairly general agreement that the
breeding efficiency of heifers is lower than that of cmys calving more than
once and that fertility declines in animals more than 9 or 10 years old.
A study at Cornell University involving 41 bulls and over 12,000 services
to Holstein cows showed that the average breeding efficiency of all cows
artificially inseminated was 48.2 per cent. It was maximum (50.6 per
cent) in cows 4 to 6 years of age and minimum in cows 1 to 3 years of age
or more than 10 years (46.5 per cent in both groups). In the bull, the
only outstandingly high figure ,vas 56.1 per cent breeding efficiency in the
2-year-old group and an average of 48.2 per cent in bulls ranging from 1
to 12 years of age (Tanabe and Salisbury, 1946). A study of breeding
efficiency in the University of Nebraska herd between 1896 and 1934
showed that cows under 2 years of age were less fertile than cows from 2
to 10 years old and bulls under 2 years of age had the highest breeding
efficiency (Morgan and Davis, 1938). A summary of the fertility of
dairy bulls in the Purdue herd between 1920 and 1940 showed that bulls
from 21 to 33 months of age had the highest efficiency (73.4 per cent),
but a small number of 10- to 12-year-old bulls still had a breeding effi-
ciency of 64 per cent. The data on older bulls are often biased because of
a tendency to eliminate older animals for many reasons other than lowered
fertility and to retain some bulls of known lowered breeding efficiency
but of outstanding blood lines.
It appears that the effects of age on the fertility of beef cattle are
rather similar to those of dairy cattle. A study of about 1,300 Hereford
cattle on the San Carlos Indian Reservation showed that when artificial
insemination was used 2.37 inseminations per calf were required in ani-
mals 2 to 3 years old and only 1.36 inseminations per calf born in cattle
5 to 6 years of age. Fertility appeared to decline gradually after the
sixth year (Lasley and Bogart, 1943).
Sheep are seasonal breeders and are ordinarily not mated until their
second breeding season at 18 to 21 months of age. Ewes are ,,·ell grown
out by this time, and the age factor is of relatively little importance.
However, Briggs (1936) showed that ewe lambs may be bred during their
first breeding season when about 9 months of age. Such animals grew
more slowly than those bred at their second breeding season but reached
the same ultimate size. In general, there is an increase in lambing rate
up to 3 to 6 years and a decline in lamb crop after this time (McKenzie
and Terrill, 1937). It has been shown that lambs and yearlings have
shorter estrual periods and a lower ovulation rate than mature ewes.
In the mare, breeding efficiency is lowest in animals entering the
breeding herd for the first time and in mares bred at foal heat. Mares
which are ~well grown out can be bred at 2 years of age, but most animals
are 30 to 36 months of age before being placed in the breeding herd
(Andrews and McKenzie, 1941).
Failure of females to reach their expected mature size is frequently
attributed to the premature establishment of pregnancy. This excuse
is invariably given as an explanation for the lack of size of a particular
individual in a herd. The data which are available for farm animals
indicate that lactation places greater demands upon an animal than
pregnancy. If pregnant and lactating animals are properly fed, their
ultimate size appears to be unaffected, even though the rate of growth
may be reduced. This general problem has been extensively studied in
rats at Cornell University, and although there are certain dangers in
interpreting rat data in terms of farm animals, the results are of con-
siderable interest.
It has been shown (1) that castrated female rats grew faster than bred
or virgin rats; (2) that rats bred at normal age and their young killed
immediately so that they were pregnant most of the time but not lactating
grew more rapidly and for a much longer time than virgins or those bred
early or late; (3) that early-bred and lactating females were somewhat
retarded in growth due to lactation but that they eventually reached the
same adult size as those bred later; (4) rats bred at normal age (ahollj,
100 days and somewhat after puberty) grew most rapidly next to the
castrates and those bred at normal age with young killed, and eventually
reached the greatest weight; (5) that rats bred late were retarded in
growth and did not reach the weights attained by castrates, those bred
early, or at the normal time; (6) that virgin rats were most retarded in
growth and attained the smallest size at maturity (Bogart et ai., 1940).
From the standpoint of reproductive efficiency it appears that early
and regular reproduction is a stimulus to reproduction. Rats which were
never bred showed irregularities and cessation of estrual cycles earlier
than any of the mated groups. It was suggested that breeding promotes
a greater harmony between the gonads and the other endocrine glands
(Asdell et ai., 1941).
Relation of Breeding and Time of Ovulation.-As previously stressed,
the chief objective of the livestock breeder is to time the breeding act in
such a way that large numbers of viable spermatozoa will be present in
the female genitalia at the time of ovulation. Such planned mating
should be regarded as essential in the mare, since breeding within the
first few hours of heat is almost certain to result in a sterile mating. At
the other extreme is the chicken; in this species one natural or artificial
insemination results in approximately 90 per cent fertility for an entire
week.
In order that breeding and ovulation may be timed, it is assumed that
the ovulation time of the species is known. In sheep the problem appears
to be less acute than in the other farm animals. The heat period is
relatively short, ovulation occurs near the end of estrus, and rams running
with ewes usually mate more than once during heat.
Trimberger and Davis (1943) have shown that in dairy-cattle, breeding
at the start of estrus results in much lower efficiency than at the middle
of heat and that if insemination is carried out 12 hours or more after the
end of heat, the conception rate is low. The fact that some cows con-
ceived when bred as late as 36 hours after the end of heat probably
indicated that these animals ovulated later than the average cow.
Practical recommendations for the various species will be made in a
subsequent section.
BREEDING MANAGEMENT OF THE MALE

Service for Immature Males.-Although many studies have been made
of the testicular development of males of various species, and the influence
of such factors as age, nutrition, and season upon semen quality, little
reliable information is available as to the optimum rate of use of males
during the transitional period from puberty until somatic maturity has
TABLE 12.-BREEDING RESULTS IN FEMALES BRED EXPERIMENTALLY AT VARIOUS

STAGES OF ESTRUS AND IN FEMALES BRED ARTIFICIALLY IN THE UNIVERSITyHERD*
Cows conceiving
Number of from one service
Time of service
cows bred
No. %
--~
Start of estrus ...... ... . . . .............. , ....... 25 11 44.00

Middle of estrus ................................ 40 33 82.50
Bred at middle of estrus and reb red in 24 hours .... 25 21 84.00
End of estrus ................................... 40 30 75.00
Artificial routine breeding ........................ 194 123 63.40
6 hours after estrus ended ....................... 40 25 62.50
12 hours after estrus ended .......... ..... .... . . .. 25 8 32.00
18 hours after estrus ended .............. . . .. . . ... 25 7 28.00
24 hours after estrus ended ...... ...... . 25 3 12.00
36 hours after estrus ended .... .... . . .... . ..... 25 2 8.00
48 hours after estrus ended ... ............. 25 0 0.00
* Trimbe:r'ker and Davis, 1943.
been reached. The data which are available seem to suggest that the
ram and bull, species characterized by relatively small semen volume and
high sperm concentration, can be used relatively frequently "without
evidence of permanent damage to reproductive performance. In the
stallion and boar, species characterized by large semen volume and low
sperm concentration, frequent use is followed by easily recognized dele-
terious changes in semen quality. In the absence of clear-cut experi-
mental evidence, the authors draw on their own experience and that of
practical livestock breeders in order that some concrete suggestions may
be made. . To err is human, and in this case the 'writers have attempted
to recommend conservatively.
As previously stressed, there can be no substitute for the understanding
of each sire's individual capabilities. In the farm animals sperm are
produced in limited numbers long before skeletal and muscular growth
are completed. It stands to reason that a certain degree of size must be
attained before copulation can be performed with physical safety. The
breeding habits of males should be established, not allowed to develop in
a hit-or-miss fashion. The frustration of immature males in unsuccess-
fully attempting to serve animals which are too large, which are improp-
erly restrained or in an unsuitable location, may have a lasting effect on
the mental attitude of the sire. Males which fall because of slippery
footing or receive head injuries because of use in low-ceilinged barns may
be both physically and psychologically injured. The first practical sug-
gestion, then, is that no male be used for service unless he is mated with a
REPRODUCTIVE EFFICIENCY .i81
physically suited female. Second, the sire must exhibit sufficient sex
drive to complete the sexual act. Some males develop slowly in this
respect, and others are apt to be restrained in the presence of man.
Impatience and punishment at this time have no place in the intelligent
breeding program. It is our opinion that care in the establishment of
good breeding habits is just as important as the initiation of a heifer to
the milking parlor or the high-strung Thoroughbred to the saddle.
If a sire is willing and able to give service, if these abilities are main-
tained following such use, and if the percentage of females which conceive
is approximately the average breeding efficiency of the species, it "would
seem as though the male ,,,ere being wisely used.
Bulls.-Widespread experience in artificial-breeding associations indi-
cates that the average mature bull can be used at "'eekly intervals prac-
tically indefinitely without affecting semen quality. In most cases two
successive ejaculations are obtained at each weekly collection period.
During the early use of artificial insemination some bulls remained in
useful service when ejaculated on alternate days or even every day, but
such rate of service is now considered undesirable. Under natural breed-
ing conditions a well-developed yearling bull can be expected to serve
at least 25 cows between his first and second years. It would seem wise
to distribute these services throughout the year, or at least not to concen-
trate several servicE;ls per week into a short breeding season. Evidences
of a decline in sex drive or lo:,;s of "weight should be considered warning
that the system of management is unsatisfactory. In view of the data
showing that maximum fertility is usually attained at t,yO years of age
in bulls, it seems logical to expect animals to be capable of natural service
at least once a week after this time. It is well known that bulls can be
used two or three times daily without affecting fertility if they are allowed
several days of sexual rest afterward. Many laboratory tests of the
effects of repeated ejaculation on semen quality have been made. In one
such test 24 ejaculations were obtained during a 27-hour period. The
concentration of sperm remained constant until the fifteenth mating,
after which it fell sharply. However, after an 18-hour rest sperm con-
a
centration rose to high level again (Anderson, 1945). Many bulls have
been ejaculated at intervals of 3 or 4 days for periods of a year with no
observed harmful effects. Other bulls, which have given excellent result~
when used in natural service in small herds, have proven worthless when
attempts have been made to collect at weekly intervals for artificial
insemination.
Rams.-Under practical conditions, ram Iambs are not ordinarily used
until they are about eighteen months of age. However, lambs are capable
of limited service during their first fall 'when about nine months old.
When the earliest possible progeny testing is of importance in genetic

studies, a limited number of females can be mated to spring rams. In
such cases, artificial insemination has the advantage of extending the use
of young sires far beyond their natural capabilities.
It has been found very difficult to exhaust the sperm stores of mature
rams. In one exhaustion test a particular ram after 42 consecutive
ejaculations within a 9-hour period still ejaculated 100 million sperm
(Anderson, 1945). Studies at the Missouri Station showed that there
are wide individual differences in the semen production of rams, that
complete sperm exhaustion is difficult to obtain, and that when the sperm
are depleted, a short rest period is sufficient for restoration (McKenzie
and Berliner, 1937). At two and one-half years of age a ram should be
capable of serving 40 or 50 females, and older rani\..have frequently
produced satisfactory results when mated with 80 or 90 ewes annually.
Boars.-There appears to be a discrepancy between laboratory studies
on semen production of the boar and the breeding practices used by
successful swine producers. McKenzie et al. (1938) showed that when
boars were ejaculated at intervals of 24 hours or less, semen volume,
spenn concentration and total numbers, and duration of sperm motility
all declined. Two of three boars used at 12-hour intervals temporarily.
refused to mate after three or four services. Russian workers concluded
that boars should not be used more than twice, and preferably once, in
24 hours. Yearling boars, it was reported, should not be used oftener
than eyery other day if they are to be used for as long as 2 weeks (Ander-
son, 1945). An Indiana survey by Smith (1937) showed that swine
breeders were agreed that mature boars can serve two or occasionally
three sows daily. Nearly all the producers questioned reported that
yearling boars could be used as frequently as older males without affecting
breeding efficiency. Personal questioning of successful s,vine breeders
by the author during the past 5 years has revealed that as many as four
or five sows have been bred to one boar on a particular day without
affecting fertility or litter size.
It is felt that some reservations need to be made in regard to the
frequency of use of boars. The incidence of 10 to 20 per cent of non-
pregnant sows in otherwise successful swine herds and the frequent
occurrence of litters of two to four pigs cannot but raise the question as
to the role of the boar in this problem. Current studies at the Indiana,
Illinois, l\fissouri, and Wisconsin stations have shown that many sows
which failed to conceive following repeated services on farms have
become pregnant on the first service when taken to the experiment
stations mentioned. It cannot be concluded at present that this is
attributa'ble to the boar, but it does suggest that the optimum use of
boars may fall somewhere between the recommendations of the research

worker and the swine producer.
Stallions.-There are wide individual and breed differences between
stallions in age at seAlial maturity and in sexual capacity. In general,
stallions should not be used until they are two years of age, and one
service per week during the breeding season is usually not excessive.
Stallions three or more years of age can be used more heavily. Here
again, there is a difference of opinion in the optimum rate of use. It
seems evident, on the basis of experience, that stallions can occasionally
give up to four or five services per day, provided sexual re -t is given
after such use. In the authors' experience, mature stallions can be
FIG. 60.-Exercising paddocks for bulls at the University of Massachusetts. It is essential

that old bulls secure sufficient exercise.
expected to give six or seven services per week during a breeding season
of 45 days \\"ithout affecting breeding efficiency. Other workers have
found that some stallions can be used twice daily for extended periods
without harmful effects (Anderson, 1945).
General Management of Males.-Because the maintenance of maxi-
mum breeding efficiency of the male, from a purely physiological stand-
point, to say nothing of his influence from a genetic standpoint, can and
does contribute so much to the economic welfare of his owner, it would
seem that the question merits a lot more attention than it often gets.
The male is often kept in that part of the barn that does not lend itself
to any other use or in some place on the farm where he " 'ill not interfere
\\;th other livestock operations. This location may vary, in the case
of the bull, from a stanchion at the end of the barn, an abandoned
horse stall, or a muddy triangle between the barn and the silo. Bulls all
too frequently receive the hay which the cows have previously sorted
over and have no access to pasture or any source of green feed.
We will subsequently discuss the nutritional requirements for repro-
duction but will make no recommendations of specific rations for males.
By way of summary, our present state of knowledge suggests that the
quantitative and qualitative requirements for reproduction do not exceed
those for the growth of young animals or for the maintenance of older
animals in a state of good health. It should be obvious that the ration
should contain a balance of carbohydrates, proteins, and minerals, and
should supply the vitamins known to be essential for the various classes
of livestock. The value of good pasture for breeding animals cannot be
oyeremphasized. It is all too true that the scrub bull allowed to run
with the herd is often more fortunate in this respect than the valuable
purebred sire confined to a small exercise lot. It has been known for
many years that fat sires are often sexually indifferent and may, in
addition, be of lowered fertility. It is still not clear whether the fatness
is the cause of lowered fertility or 'whether decreased testicular function
results in increased fat deposition. Regardless of the cause, there seems
little to be gained by the maintenance of a high degree of finish in breeding
animals and much to be lost. The degree of fatness is, in some cases,
due to feeding and management practices. Confinement of the sire in
close quarters and the liberal use of carbohydrates is almost certain to
result in fat deposition. The occurrence of obesity in the human may
sometimes be of endocrine origin, but the consensus of medical opinion
is that the chief cause is overeating. In the case of farm animals we can
frequently attribute it to overfeeding. .p'9i'"
The rapid development of artificial insemination has done much to
focus our attention on the nature and causes of reproductive disturbances.
Numerous attempts have been made to develop rations which will increase
semen production and improve semen quality. As previously mentioned,
it still appears that there are no special nutrients concerned only with
fertility. Carbohydrate and fat deficiencies are extremely unlikely to
occur in farm animals. The amount and quality of protein required by.
breeding males is still controversial. It has been reported that high
protein intakes are both helpful and detrimental in bulls. Studies by
Branton et al. (1947) have shown little difference in the fertility of bulls
receiving 12, 16, and 20 per cent total protein. It has been reported that
:,;ources of animal protein such as skim milk, fish meal, and casein will
improve fertility. Cornell experiments in which bulls were fed digestible
nutrient levels of 100, 120, and 140 per cent of the Morrison dry-cow
maintenance requirements showed no difference in fertility (Reid, 1949).
UEPlWDUC'l'lV E EFJ!lG'lENCY 185
It is possible that specific nutrients or hormones or other chemical sub-

stances may eventually be found which will increase fertility beyond
p~esent levels, but until such substances are known, the emphasis should
be on good feeding and management rather than on patented nostrums.
The exercise requirements of breeding animals is a favorite but contro·
versial subject of livestock men. There is little disagreement that breed··
ing males should be active, willing, and able to perform service quickly.
The fat , sluggish male that will not perform service is a trial to even the
most phlegmatic owner. The maintenance of animals in the desired
state is most easily accomplished by providing large paddocks which
contain good forage. In this way the animal receives valuable nutrients
FlG. 61.- A method for exercising bulls in use at experimental farm, Beltsville, Md.
(Courtesy of B ureau of Dairy Industries, U .S. Department of Agrwulture.)
and is provided an opportunity to exercise at will. Most animals can

regulate their exercise needs. The amount necessary for one may be
quite unsuited to another. A prominent and youthful university presi-
dent has been quoted to the effect that whenever he feels the urge to
exercise he lies down until it wears off. One of the most valuable results
of a suitable paddock is the beneficial effect which it has on the health of
the feet and legs. Animals which are confined in close, damp quarters
may develop foot rot, overgrown hooves, and joint disturbances. It is
tragic to observe a valuable and prepotent sire which cannot perform
.service because of feet and leg damage.
Numerous methods of forced exercise have been employed. There is
little doubt that stallions are maintained in better general health and are
more tractable when they perform regular work. This can be overdone.
Stallions have sometimes been given so much road work in an attempt

to restore fertility that they have been physically exhausted. When
large bull studs were first established for artificial insemination, all bulls,
regardless of age, condition, or breed, were given a certain amount of
exercise daily. It was soon observed that forced exercise could have
undesirable as well as beneficial effects. There is no substitute for the
recognition of individual requirements by an alert herdsman. There
seems little wisdom in feeding large amounts of carbohydrates only to
later walk them off on a treadmill, when the animal could better be
maintained in a fit condition by judicious feeding.
Not a great deal is now known about equine, bovine, ovine, or porcine
psychology. We have known for some time that certain animals did
not mate readily, that moving sires to new locations "was often followed
by long or short periods of sterility. The underlying causes are not yet
well understood, but they merit careful thought and experimentation.
We have long known that males not trained to use a breeding rack in
early life are often taught with great difficulty, if at all, in later life.
N ow, in artificial inseminating work, we are running into much the
same problem of males refusing to serve while an attendant stands close
by with an artificial vagina ready for use. Our highly bred animals are
rather delicately balanced organisms from a nervous standpoint, and we
have much to learn along these lines before we can get 100 per cent
service (in the general sense of that term) from them.
Our males generally serve eagerly when young if conditions are made
suitable. Our task is to retain that quality in them as long as we desire
to use them. We know too, contrary to some rather extensive opinion,
that genetically a male will transmit the same at any age-one or tllenty
"-that the genes do not change "with age, experience, production, J)r any
other extraneous influences.
Males should be purchased subject to their ability, willingness, and
success in impregnating sound, healthy females. It would probably be
wise also before laying down a good price for a breeding male to have a
veterinarian examine his semen, because what we are really buying when.
we purchase a male is semen, containing viable germ cells, and these in
turn contain, we hope, the genes that will bring about improvement in
our herd or flock. For some types of genital diseases the only way to tell
if a male is diseased is through using him. Preferably a male should be
bred to a limited number of females and then held out of service for a
few weeks to see if gestation is proceeding normally.
Besides making the external conditions conducive to the success of a
young male's first service, strict attention should also be given to the
internal conditions of the female genital tract. If, as should always be
the case, the young male has come from healthy parents, has grown up
in a clean environment, and has been furnished suitable feed, he should
reach puberty strong and uninfected. If this is true, one cannot be too
careful in guarding him against infection. The male should be examined
by a competent veterinarian and, if judged sound, he is ready for service.
He should be allowed to serve only healthy females. The pernicious
practice of allowing healthy young males to serve females that have been
unable to conceive from one or several previous services cannot be con-
demned too severely, though it is often practiced, even when the female
has a purulent discharge. Infection is all too prevalent and must be
guarded against continually. A. clean herd can be had if due care is
exercised, and, when this end is once attained, its health should be
guarded diligently in every detail.
Over a period of years your sires will make or break you, so select them
intelIigently and treat them understandingly.
BREEDING MANAGEMENT OF FEMALES

There is no single plan of breeding management for the various species
of farm animals nor is there only one satisfactory system for each class of
animals. In a nation as large and as diverse as the United States, to say
nothing of conditions in other parts of the ·world, local conditions must be
given primary consideration. The needs and the conditions of the pro-
fessional dairymen supplying fluid milk for the Nmv York City market
are entirely different from those of the beef producer in eastern Montana.
Calves dropped in January in a New York State barn have an excellent
chance of survival, whereas in the northern range states calving must be
confined to the spring months. For similarly good reasons, hand VB.
herd or flock mating, artificial VB. natural service, lard type VB. bacon
type, wool VB. mutton breeds are all questions which must be settled on
their merits.
Cattle Problems.-Most cows come in heat at regular intervals
throughout the entire year, but evidences of the seasonal breeding habits
of primitive cattle still persist. In the larger dairy herds which are
engaged in the production of milk for fluid consumption, breeding opera-
tions are ordinarily distributed throughout the year. In less intensive
dairy areas and in the beef herds of the Western range states the breed-
ing season is usually limited to the late spring and early summer months.
The average length of the entire estrual cycle is 19 to 20 days, and heat
itself usually lasts 16 to 20 hours. In most cows ovulation occurs from
10 to 18 hours after the end of heat. An understanding of the estrual
cycle of each cow is necessary if maximum breeding efficiency is to be
attained.
When heat is first observed in the morning, it is recommended that

service be given during the afternoon of the same day, since breeding in
about the middle of the heat period has been shown to give optimum
results. When it is known that estrus has begun in the afternoon, it is
desirable to breed during the late evening, since the cow may be out of
heat the next day. In such cows, if heat persists the following day, a
second service on the second day may increase the conception rate.
The fact that cattle ovulate following the end of heat makes postestrual
insemination possible. This has led some herdsmen to believe that cows
can be inseminated at any time without respect to heat. This false
impression can have adverse effects on qreeding efficiency. If insemina-
tion is carried out within 6 hours after the end of heat, the chances of
pregnancy are fairly good, although not so good as following service at
mid-estrus. Under practical farm conditions there is nothing to be gained
by two services in quick succession. There is some evidence that cows
with exceptionally long heat periods or those in which ovulation may be
delayed may benefit from two services at different times during heat.
Natural service at mid-estrus followed by artificial insemination shortly
after the end of heat may be helpful.
There are a few reproductive characteristics of cattle which differ from
those of other animals. Cows have shorter heat periods than the other
farm animals. It is easy, therefore, to fail to detect heat. When cattle
appear to miss heat periods and have cycles of 38 to 42 days, there is a
good chance that a heat period has occurred without detection. The
importance of a system of regular observation and recording of the heat
periods cannot be overemphasized. The fact that cattle ovulate after
heat has ended also puts them in a separate class. Obviously not all cows
ovulate at the same time with respect to the termination of estrus.
Abnormal ovulation time may cause infertility in otherwise normal cows.
As suggested, breeding more than once may be helpful in such animals.
When unusual ovarian acti vity is suspected, a specially trained veterinar-
ian can verify it by rectal examination.
The interval between calving and the first heat period and its signifi-
cance should be understood. Beef cattle usually come in heat sooner
than dairy cattle. The average interval for all cattle is from 40 to 60
days but may vary from 6 days to 6 months or longer. When heat
periods have not been resumed within 80 or 90 days, an examination to
determine the cause should be made. The manual removal of the corpus
luteum, if this is the case, can be easily accomplished.
Cattle are subject to several reproductive disturbances. The frM-
martin condition is more or less unique in cattle; 90 per cent of the heifers
born twin to a bull are sterile. For this reason, but not the only one,
twinning is not a desirable trait in cattle. The occurrence of nympho-

mania and cystic ovary is more common in cows than in other animals,
although it does occur in mares. Cattle have a higher incidence of
genital diseases than other farm animals. Brucellosis is a national
problem, and trichomoniasis and infectious vaginitis can be troublesome
when established. The occurrence of abortion or any abnormal genital
discharge should be investigated at once.
Swine Problems.-Sows do come in heat throughout the year and
there appear to be no seasonal limitations to fertility. However, when
sows farrow at all months and pigs of all ages are present on individual
farms, there are certain swine-disease problems which must be overcome.
Sanitation must be given close attention and infectious diseases, such
as hog cholera, dysentery, and gastroenteritis, must be guarded against.
In the corn belt, sows are usually bred to farrow twice a year, in the
spring and fall. In areas where the one-litter system is practiced, most
sows are bred for spring farrowing.
The average length of heat is 2 or 3 days, and sows usually have longer
heat periods than gilts. However, heat may vary from 1 to 5 days, and
it is desirable to know its onset and termination by trying the sows daily.
The total length of the estrual cycle is about 21 days.
It has been commonly believed that ovulation usually occurs on the
second day of heat. However, there is much to be known about the
time of ovulation as well as its pattern, whether most of the ova are shed
during a short interval, and what the range between the first and last
ova is. The best information which is available seems to indicate that
sows ovulate during the second day of heat. We have recommended,
therefore, that gilts be bred early during the second day of heat and
so>vs sometime during the second day. Here again, practical swine
breeders have some difference of opinion as to breeding time, and many
report that sows can be mated at any time during heat with equal success.
As mentioned previously, the fact that swine have a much lower breeding
efficiency than has been thought should be good reason for more care in
breeding management. The occasional female that stays in heat only
one day should obviously be mated early during that day. The sow
which stays in heat three or four days will undoubtedly benefit from
service on both the second and third days. When a gilt and a sow are to
be bred to the same boar, and both come in heat at the same time, it
would seem logical to breed the gilt on the morning of the second day
and the sow during the afternoon. If it can be avoided, the boar should
not be used for successive services only a few minutes apart.
It is well known that some sows come in heat a few days after farrowing,
hut arp, seldom mated at this time. The next heat period ordinarily
does not occur until 3 or 4 days after the pigs are weaned. However,
sows which have small litters may exhibit estrus while lactating and if
the pigs are not weaned at the usual time will eventually resume estrual
cycles.
There has been some practical interest in the establishment of preg-
nancy in lactating sows. Experiments by Robeson at the Ohio Station
in 1918 showed that the separation of the pigs from the sows for several
nights, but letting them run together during the day, would induce heat.
A more recent study by Cole and Hughes (1946) showed that the injection
of 1,000 I. U. of equine gonadotropin between the thirty-ninth and sixty-
eighth days of lactation would induce heat in 3 to 7 days. This method
has practical applications, especially in causing animals which farrow late
to come in heat and thus concentrate the next farrowing season in a
shorter period of time.
Among the more important swine-breeding problems which n~ed solu-
tion are the embryonic and fetal death losses and the high early post-
natal deaths. The most conservative estimates indicate that at least
5 to 20 per cent of the fertilized ova fail to develop and that 10 to 30 per
cent of the nel,-born pigs die within the first week of life. Brucellosis is
the only specific genital disease of swine which is widespread. Its dangers
to man and the economic losses which it causes in swine make it a serious
threat to the swine industry. It can best be guarded against by the
introduction of only tested stock from herds of known disease status.
Sheep Problems.-Sheep are peculiar in that they have as yet success-
fully resisted man's attempts to eliminate the seasonal breeding pattern.
Most breeds of mutton sheep exhibit estrus from August through January.
In the Midwest, maximum fertility is usually attained during late Sep··
tember, and Hampshires and Rambouillets tend to become pregnant a
little sooner than Shropshires and Southdowns. In areas where hot-
house lamb production is practiced, Dorset X Merino ewes have been
popular because of their tendency to exhibit estrus during the summer.
Breeding practices in the range states vary with local conditions. In
the Southwest and coast states ewes are frequently bred to drop their
lambs from January through March. In colder areas most ewes are
lambed in May. It is apparent that the breeding time is limited not only
by the seasonal patterns of the sheep themselves but by economic and
geographic considerations.
The desirability of twins may depend upon the nature of the sheep
project. Single Iambs are usually larger and grow more rapidly than
twins, whereas twins which survive produce more lamb meat per ewe.
Because of differences in vigor, many range producers prefer single lambs.
As will be discussed, whether or not flushing affects the size of the
lamb crop, there are no disadvantages to improving the nutritional status

of e\ves priori to breeding. Excess fatness prior to lambing is to be
avoided, but the ewes must be fed a well-balanced ration if pregnancy
disease is to be avoided.
Ewes remain in heat about 30 hours and ovulate shortly before the end
of heat. When hand mating is practiced, breeding during the latter part
of heat is preferred. The majority of ewes are bred by rams running
with the flock, and there is no doubt that this method produces entirely
satisfactory results jf the ram is of normal fertility. It is desirable that
the wool in the tail region be clipped before breeding in order to facilitate
mating. In farm flocks, painting the brisket of the ram with used motor
oil to which a dye has been added will mark the ewes which are bred.
In this way a fairly accurate breeding record can be kept. Also, in farm
flocks; the ram is sometimes turned in with the ewes for only a few hours
daily. In this ,Yay his energies can be conserved, and he can be fed
separately if he is losing ,,·eight.
Genital diseases in sheep are rare, and there are few problems which
cannot be overcome by the application of knowledge which is already
at hand.
Horse Problems.-The decline in the use of the horse as a source of
farm power has resulted in an almost complete lack of interest in horse
breeding by farmers. At the present time specialized horse-breeding
operations are carried on almost exclusively by the producers of saddle,
running, and light-harness horses.
Although 'rell-fed mares come in heat throughout the year, most mares
are bred to foal in the spring. The relatively long heat period (4 to 6
days) of the average mare and the relatively short life of stallion sperma-
tozoa poses a problem of timing which is greater than in any other farm
animal.
The serious horse breeder 'will find it essential to try mares for the
occurrence of estrus every day, or at least every other day, during breed-
ing season. Mares which have been mated should be tried frequently
and not just at the t,Yenty-first day after breeding. As previously
explained, the average interval between heat periods is 14 to 16 days,
and the reoccurrence of heat is often missed because of poor timing.
The average mare ovulates 24 to 48 hours before the end of heat, but
it is impossible to predict this in advance. It is therefore agreed that
most mares should be bred more than once during heat if the rate of
conception is to be improved. We believe that the optimum system is
the natural or artificial insemination of each mare daily, beginning on
the second day of heat and continuing until heat ends. Some successful
breeders recommend breeding on the third, fourth, and fifth days of heat,
Hl2 BREEDING AND IMPROVEMENT OF FARM ANIMALS
and others breed every other day, beginning on the third day of estrus.
If only one mating is possible, the fourth day of heat is probably the best.
When stallions traveled from farm to farm, mares were usually bred
once without respect to the day of heat. Under these conditions it is
not surprising that only 40 to 50 per cent of the females bred conceived.
The chief problem in practical horse breeding is, therefore, one of
having copulation and ovulation coincide. It has been believed for
centuries that the best time to breed mares is during foal heat on the
ninth day after foaling. This concept is unsound, not only because all
mares do not ovulate on this magic ninth day, but because there is good
evidence that genital diseases may be spread by breeding too soon after
foaling. It appears that navel ill in the foal may actually begin its
development when mares are bred during foal heat. Data from various
horse-breeding areas indicate that fewer mares conceive ",hen bred at
foal heat than if mating is delayed until the second estrus period after
parturition.
Breeder and Veterinarian Relations.-There are three possible points'
of view concerning the relations between breeders and veterinarians.
One of these is that the veterinarian should be called to take care of
every departure from normality in animals as regards health (or acci-
dent), no matter how slight. The other extreme is that an experienced
herdsman can entirely dispense with the services of veterinarians and
diagnose and treat all health abnormalities in his herd or flock. The
writers think that both extremes are wrong and that there is a middle
ground, somewhat variable between individuals, which is better for all
concerned. If a cow scratches her udder or steps on a teat or a colt
cuts himself on a barbed-wire fence, an avenue of infection is opened up
that may result harmfully or even fatally. If the damage is great enough
to require surgery or suturing, by all means the veterinarian should be
called. If it is not, the herdsman can dress the wound, keep it clean,
and bring about healing.
If an animal loses its appetite and becomes nervous and feverish, it
may be due to a more 0,1' less trivial digestive upset or to constipation,
which a dose of salts or castor oil together with more careful feeding for
a few days would correct. It might, on the other hand, be the beginning
of an acute case of colic or the onset of some fatal disease, the symptoms
of which are totally unknown to the herdsman, and in these latter cases
it would have been much better to secure the services of the veterinarian
as soon as the trouble was discovered.
Sometimes the male becomes a bit slow and uncertain in service, or
the females do not conceive. The herdsman may change the ration,
fltart to feed. minerals or vitamins, or try this or that panacea suggested by
,
\
well-intentioned neighbors or friends. Perhaps the herdsman himself

has had a bit of experience and training in massaging ovaries, breaking
out cysts, and so on. ¥any herdsmen with training and experience can
do these things and do them well. For many others no amount of avail-
able training or experience seems to suffice. They just don't have the
knack. If you are in the first-mentioned class, you are fortunate, and
you probably also have the good sense to be on your guard at all times
and to call upon your veterinarian when things look as though they are
beyond your own power to deal with successfully. If you are one who
lacks the training, experience, or knack in handling such things, you will
be well advised to call on your veterinarian as soon as you notice some-
thing abnormal. And you will be still better advised not to let neighbors
or others probably no better fitted than yourself start prescribing for or
"treating" your animals.
1.ikewise, the time of parturition is a critical one for both mother and
Dffspring, and again many herdsmen are capable of straightening up a
fetus and of helping the dam to make a safe delivery. Other herdsmen
are more or less helpless at this time if things are not normal. The same
parallel might be pointed out regarding difficulties and abnormalities
following parturition.
The veterinarian is all too often put in a bad hole to start with by
being called only after a case has gotten really serious or by having not
only the original trouble to contend with but also having to straighten
out as best he can the bungling attempts of some not-too-intelligent
animal nurse or midwife. From a purely selfish financial standpoint,
overlong delay in securing the help of the veterinarian is generally
"penny-\vise and pound-foolish." The first prerequisite for success in
animal breeding is the general and the reproductive health of our animals.
The need for veterinary services is a variable thing on all farms, and its
need arises earlier under this owner or manager than it does under that
one. Probably every herd needs veterinary attention at times, and it
is for each man to determine his own "when." The extent to which the
owner or herdsman can take care of the simpler health cases of his animals
also varies ·within rather wide limits, and each man must learn his own
abilities and more especially his own limitations. A good veterinarian
in any case is an invaluable help to herd health, and, because this influ-
ences productivity and profits to such a large extent, the breeder, for
his own good, must study the problem and use veterinary services in such
a manner as will return the greatest profit to him. It certainly does not
pay to be penurious in this field. It certainly does not pay to Ul-le
"quacks" of any description. It certainly does not pay to remain
ignorant of our own capabilities and limitations to render aid to our
animals, and, finally, and most important, it does not pay to delay over-.
long in getting needed medical care for abnormal, sick, or injured animals.
A good veterinarian is about the best and most valuable friend a good
breeding establishment can have. He merits fairer treatment than has
often been his lot in the past.
If a man is serving as herdsman or manager of a livestock-breeding
farm, he will do well to arrive at a very definite understanding with the
owner as to the procedure to be followed regarding the use of veterinary
service. Are you as herdsman or manager expected to supply veterinary
as well as animal-husbandry services? Few men are thoroughly trained
in both fields, and the lack of training is almost sure to show up sooner
or later and usually with some cost to the owner. Are you to provide
what medical service you feel you are fitted to, and are you free to call
on whatever veterinarian you wish to whenever you think you need his
services? Are you to provide animal-husbandry services only and to
notify the owner whenever you think veterinary services are called for?
Is a veterinarian to be employed to make regular health examinations and
to be entirely responsible for both the general and genital health of the
herd or flock?
TABLE 13.-REPRODUCTIVE PHENOMENA
ILength
Age at Xormal Age Length of ges- Rebred.
puberty, breeding first of heat tat ion after
months season bred period period, parturition
days
---
Horses .. 10-12 April-July 3 years 5-6 days 336 24-30 days
after foaling
Cattle. 4-8 Any time 15-30 months 16-20 hours 281 60-90 days
Sheep. 4-8 September- 18 months 30 hours 147-152 Next fall
December "
Swine .. 3-7 December 9 months 2-3 days 112 3-5 days after
and June weaning pigs
All these systems may work under certain circumstances. but there
will be much less loss and misunderstanding if a general policy can be
adopted and followed. If you are the owner of the animals, the decision
as to health policy is yours and should be made, and, when made,
adhered to.
The situations referred to are largely of an emergency nature. All
livestock men should also inform themselves in regard to the symptoms,
prevention, and control of diseases which affect the entire livestock
industry. Such communicable diseases as tuberculosis, brucellosis, hog
cholera, mastitis, hoof-and-mouth disease, and several others are of

national importance, and numerous local and national programs for the
elimination of these diseases have been developed. Purebred breeders,
particularly, should familiarize themselves with the rules and regulations
in regard to livestock sales, exhibitions, and interstate shipments. The
local veterinarian, the state veterinarian, and The Bureau of Animal
Industry are all concerned with the preservation of our livestock in the
best possible state of health, not only for economic livestock production
but for the prevention of diseases of animals transmissible to man.
In recent years, in both human and animal medicine, there has been
increased emphasis on the prevention or early diagnosis of disease and
disorders. It is obviously more satisfactory to prevent such disturbances
than to cure them, and it is likewise easier to correct them in their early
stages than after they have become well established.
,Many cattle breeders have adopted the practice of routine genital
examinations of all breeding animals. The herd is usually visited at
monthly intervals by a veterinarian who is especially trained for such
work. Animals which have not resumed their estrual cycles at the
expected time following calving are examined. The cause is often readily
apparent and easily corrected, e.g., a persistent corpus luteum. Cows
with irregular heat periods, animals which have been bred repeatedly but
have failed to conceive, and those with any evidence of abnormalities are
carefully examined. Cows which have been bred and which have not
returned in heat are examined to see if they are pregnant. All too often,
animals which have not returned in heat are sold as non breeders only
to be found gravid at the time of slaughter. The keeping of adequate
records and the regular examination for pregnancy will eliminate such
errors. In other cases, animals which have little or no likelihood of
becoming pregnant are kept in herds in the hope that they \yill conceive.
The elimina60n of such animals will save much time and expense. The
detection of an infectious genital disease or other disturbances in their
early stages may allow for easy correction and safeguard the other animals
in the herd.
Summary.-In this chapter we have discussed the actual practice of
breeding animals and have considered data on the breeding efficiency
which can be expected in average disease-free animals. We have empha-
sized the fact that breeding efficiency can usually be increased by the
intelligent application of knowledge which is available, but we have
likewise pointed out that more knowledge is needed and that there are
differences of opinion as to the effectiveness of certain practices. As
discussed in the following chapter, there are numerous recognizable causes
of lowered fertility and sterility. In this chapter we have tried to
account for the differences in the fertility of apparently normal animals.

Species differences and the effects of age, sexual use, time of mating, and
above all, the individual characteristics of animals, have been considered.
Reproductive efficiency can be maintained only by constant care in the
selection, purchase, or sale of healthy animals and by managing them
to the best of our ability and knowledge. Although it has not been
proved to the satisfaction of scientists, there is much to support the idea
that the breeding performance of the adult may be influenced by events
during embryonic or prepubertal development. The task and problems
of the livestock breeder are never ending and involve not just the animal
but the soil, the crops, the seasons, and the microorganisms which
influence all living material.
References
Books
MILLS, C. A. 1939. "Medical Climatology," Charles C Thomas, Publisher, Spring-
field, Ill.
SMITH, \V. W. 1937. "Pork Production," The Macmillan Company, New York.

Imper. Bur. Anim. Breeding and Genet. Edinb.
ANDREWS, F. N., and McKENZIE, F. F. 1941. Estrus, Ovulation, and Related
Phenomena in the Mare, Mo. Agr. Expt. Sta. Res. Bul. 329.
ASDELL, S. A., BOGART, R., and SPERLING, G. 1941. The Influence of Age and Rate
of Breeding upon the Ability of the Female Rat to Reproduce and Raise Young,
N.Y. (Cornell) Agr. Expt. Sta. Mem. 238.
BABCOCK, M. J., et al. 1940. The Reproductive Efficiency of the Albino Rat under
Different Breeding Conditions, Jour. Agr. Res., 60 :847-854.
BAKER, A. L., and QUESENBERRY, J. R. 1944. Fertility of Range Beef Cattle,
BOGART, R., et al., 1940. The Influence of Reproduction upon Growth in the Female
Rat, Amer. Jour. Physiol., 128 :355-371.
BRANTON, C., BRATTON, R. W., and SALISBURY, G. W. 1947. Total Digestible
Nutrients and Protein Levels for Dairy Bulls Used in Artificial Breeding, Jour.
Dairy Sci., 30 :1003-1013.
BRIGGS, H. M. 1936. Some Effects of Breeding Ewe Lambs, N. Dak. Agr. Expt.
Sta. Bul. 285.
COLE, H. H., and HUGHES, E. H. 1946. Induction of Estrus in Lactating Sows
with Equine Gonadotropin, Jour. Anim. Sci., 5 :25-29.
HAMMOND, J. 1921. Further Observations on the Factors Controlling Fertility
and Foetal Atrophy, Jour. Agr. Sci., 11:337-366.
HUTCHINGS, L. M. 1948. Sterility in Swine, Jour. Amer. Vet. IVIed. ASSaf., 112:
114-119.
LASLEY, J. F., and BOGAR'l', R. 1943. Some Factors Influencing Reproductive
Efficiency of Range Cattle under Artificial and Natural Breeding Condition,
Mo. Agr. Expt. Sta. Res. Bul. 376.
McKENZIE, F. F. 1928. Growth and Reproduction in Swine, Mo. Agr. Expt. Sta.
Res. Bul. 118.
_ - - and BERLINER, V. R. 1937. The Reproductive Capacity of Rams, Mo.
Agr. Expt. Sta. Res. Bul. 265.
_ - - , MILLER, J. C., and BAUGESS, L. C. 1938. The Reproductive Organs and
Semen of the Boar, Mo. Agr. Expt. Sta. Res. Bul. 279.
_ - - and TERRILL, C. E. 1937. Estrus, Ovulation, and Related Phenomena in
the Ewe, Mo. Agr. Expt. Sta. Res. Bul. 264.
MORGAN, R. F., and DAVIS, H. P. 1938. Influence of Age of Dairy Cattle and
Season of the Year on the Sex Ratio of Calves and Services Required for Con-
ception, Nebr. Agr. Expt. Sta. Res. Bul. 104.
l\ICHOLS, J. E. 1926. Fertility in Sheep, Jour. "iJ,fin. Agr. (Gt. Brit.), 33 :218-225.
PHILLIPS, R. W. 1939. Reproductive Failures in Livestock, U.S. Dept. Agr. Year-
book, pp. 476-482.
- - - and ZELLER, J. H. 1941. Some Factors Affecting Fertility in Swine, Amer.
Jour. Vet. Res., 2 :439-442.
REID, T. J. 1949. Relationship of Nutrition to Fertility in Animals, Jour. Amer.
Vet. M ed. Assoc., 114 :242-250.
R~BESON, W. L. 1918. Mating Sows before Their Litters Are Weaned, Ohio Agr.
Expt. Sta. Monthly Bul., Vol. 3, No.5.
ROBINSON, A. 1921. Prenatal Death, Edinb. Med. Jour., 24 :137-151.
STEWART, H. A. 1945. An Appraisal of Factors Affecting Prolificacy in Swine,
Jour. Anim. Sci., 4 :250-260.
TANABE, T., and SALISBURY, G. \V. 1946. The Influence of Age on Breeding
Efficiency of Dairy Cattle in Artificial Insemination, Jour. Dairy Sci., 29 :337-344.
TRIMBERGER, G. W., and DAVIS, H. P. 1943. Conception Rate in Dairy Cattle by
Artificial Insemination at Various Stages of Estrus, Nebr. Agr. Expt. Sta. Eul. 12fJ.
CHAPTER VII
LOWERED FERTILITY AND STERILITY
For the breeder there is no more important economic problem than the
fertility of his animals. Animals are simply natural machines that man
has domesticated and improved for the purpose of turning feed into
power, meat and/or milk, meat and wool, meat and/or eggs, or just
meat alone; all these qualities except power and wool being dependent
on fertility.
It is commonly believed that fertility and prolificacy are synonymous
and are the exact opposites of sterility. This is not true. Whereas
§lerility, in a technical sense, pleans complete reproductive failure. the
level of fertility may vary from 1 to 100 per cent. Since the labor, feed,
and shelter costs of pregnant animals are but little more than in non-
pregnant females, any reduction in fertility reduces the efficiency of
livestock production. Breeding animals of the meat classes that do not
bear young annually are clearly liabilities during such years. Dairy
cattle that fail to conceive within 3 to 4 months following calving may be.
regarded as deficient not only in terms of offspring but in maximum milk
production as well. The question of fertility is second to none in breed-
ing. An animal's individual merit and genetic prepotency to stamp its
good qualities on its offspring are without value if such animal~ leave few
or no offspring.
Definitions.-..Eez::ijlity denotes the ability of an aniI_Ilj1l to reRr_Qduce
jt&._~d. In animal;cllaract~arieproauction, fertility may
be limited by either sex. The mating of a highly fertile male to an
infertile female has the same result as the mating of two sterile animals.
Any animal which has the ability to cause pregnancy or to become preg-
nant is, technically, a fertile individual. It is of great importance that
the livestock breeder recognize that there are all degrees of fertility. All
too frequently herd sires are purchased with a guarantee of fertility.
Many such animals produce a few offspring, thus leaving the purchaser
no recourse against the seller, but their degree of fertility is so low that
they are economic liabilities as herd sires.
Sterility may be defined as complete reproductive failure. Such ani-
mals are easily recognized. It is frequently stated that a sterile sire is
far less expensive than one of lowered fertility. This is true because the
198
\
\
\
WWERED FERTILITY AND STERILITY 199
failure to cause any conceptions is soon apparent and must be dealt with
at once. The sire which settles an occasional female is often continued
in service in the hope that he will soon improve. Such optimism may
disrupt the entire breeding program and cause considerable financial loss.
Fecundity, meaning fertility or prolificacy in a general sense, comes
from the same root as fetus, and for this reason its use is more or less
restricted to females.
Prolificacy implies especially frequent or especially large numbers of
offspring and is also more or less restricted in its application to the female.
In all the higher species the production of young is contingent upon the
FIG. 62.- Jersey cow, Silken Lady's Ruby of F., 919141, owned by c tate of J. W. Coppini,
Ferndale, Calif. In September of 1949 this cow was nineteen years and four months old
and had produced 181,977 lb. of milk, 10,048 lb. of butterfat, and 18 calves. (COltrtesyof
American Jersey Cattle Club, photo(Jraph by Phil Palmer.)
union of viable germ cells from both parents. In those species, however,
which normally pro~than_ Qi e ~ring at a time, prolificacy is
measured through the f~ndity of the female, for the reason that a
fertile male produces enough spermatozoa at ~ervice to fertilize billions
Qf_Qya. In swine the number of pigs per litter depends not primarily on
the boar, provided he is producing viable spermatozoa, but upon the
number of ova that have matured in the sow's ovaries. Likewise a ram,
granted. normal breeding health, cannot affect the number of twin or
triplet births, for this depends on the number of ova produced by the
ovaries of the ewe. Nevertheless, the inherited power to produce large
litters in hogs or twins in sheep can be transmitted by the boar or ram to
.his offspring, and its effects make their appearance in the second gener-
ation. For this reason, it is wise to select boars from large litters and
rams from twins, other things being equal or being given due consider-
ation. In any case, males should be selected from naturally fertile

strains, which can be determined, of course, only by complete records
of the ancestors and collateral relatives.
Reproductive e:fficiency connotes the optimum effective use of the repro-
ductive powers of all the animals in a herd or flock. In the female it is
the regular production of offspring over a period of years; in the male it
1. Cystic ovaries.
2. Hydrosalpinx.
FIG. 63.-Female genital abnormalities in swine. Compare the size of the follicles with (2) .
Compare the enlarged Fallopian tubes with (1). (Courtesy of Drs. Warnick, Grummer, and
Casida, University of Wisconsin.)
is the successful fertilization of the brg;est number of eggs with the fewest
possible services. This problem will be dealt with in a separate chapter.
Causes of Reproductive Failure.-There is the common erroneous
belief that sterility is a single problem, that it is a specific disease which
can be corrected by the administration of certain drugs, the injection of
specific hormones, or the addition of certain vitamins and minerals to
the ration. No concept could be further from the truth. In some
LOWERED FERTILITY AND STERILITY 201
instances a specific condition or a particular disease can be identified as

the cause of reproductive failure. In other cases no apparent reason
for sterility or lowered fertility can be ascertained. However, the causes
of reproductive failure can ~t frequently be found among the f~owing:
-.AI) anatomical defects, (2) mechanical injury of the genitalia, (3) genital
FIG. 64.-Female genital abnormalities in swine. Missing uterine segment. (Courtesy oj

Drs. Warnick, Gntmmer, and Casida, University of Wisconsin.)
diseases, (~ nutritional deficiencies, (5(genetic make-u2. (6)_physiologj-

cal disturbances, and (7) miscellaneous environmental factort;l.
The r~productive processes may be interrupted in a wide variety of 1
ways'~nd at any stage. J:'he causes range from the lack of sex drive and
tne unwillingness of animals to mate, the death of the gametes or the
fertilized ovum, abortion, or the birth of a mature but dead fetus. ./
Anatomical Defects of the Genitalia.-Literally hundreds of types of
anatomical defects of the genital organs have been reported. Some of
these are of sufficient severity as to cause sterility and others affect the
degree of fertility. No attempt will be made to describe all the anomalies
which have been reported in the literature, but to give a brief account of
the more general type,<;. _
Probably the most common m~ genital defect ~p~hism. This
conclition has been described in a previous chapter. e results of
cryptorchism depend upon the exact location of the testes and vary from
normal fertility in the unilateral cryptorchid to complete sterility when
both testes are fully within the abdominal cavity. It js established that
FIG. 65.-Complete absence of the cervix in the sow. (Courtesy of Dr. A. V. Nolband<ro,
Department of Animal Science, University of Illinois.)
the condition may be heritable and it should not be condoned in a

breeding animal.
The occurrence of scrotal hernia is not uncommon. The descent of
the viscera through~nto the scrotum may interfere
with the circulation of the testes and result in their atrophy. The
development of a hernia in any part of the abdominal ~ugh the
testes may themselves be unaffected, may interfere with service and
indirectly affect fertility.
The penis is sometimes malformed or may be prevented from full
extension b~eath. Paralysi~tor
muscles may allow the penis to protrude from the sheath and be subjeet
to injury. Such defects may have no influence on the actual production
LOWERED FERl'ILl1'Y AND STERILITY 203
of sperm by the testes but may prevent effective service. The complete
absence of portions of the genital system is rare, but does occur.
-------------- ------
The general anatomic soundne~l'l_of males, especially their feet and legs,
should not be Ignored. It is well ~--..
known that broken limbs, sore~ver-
grm\~;'lrormed feet, and arthritic __igigts mayprevent -males of
normal fertility from mating.
Probably the best known anatomic abnormality in the female is the
freemartin condition. Of all the farm animals, cattle app-eartobe~~ique
--------------
in the occurrence of this abnormality. The freemartin, at birth, appears
to be a normal female calf. However, the genital organs are very small
and fail to develop as the animal grows. The vagina is usually smaller
than normal and cannot be penetrated to the normal depth. The uterus
does not develop, the ovaries are abnormal, and the general appearance
of the animal is more like that of a steer than a female. Such animals are
sterile. This condition appears mabout 90 per cent of the females which
are born twin to a bull cal£. The most acceptable explanation for the
condition is that it results from a hormouaLimbaJ::l,Ilce iI!_Jhe female,
presumably caused by the secretion of the male hormone by the testes
of the twin male fetus. If the placentae of the two calves are fused and
have a common blood supply, the female develops abnormally. If there
is no fusion of the blood vessels, both animals develop normally.
The term whz'le heifer disease has been widely used to describe abnormal
genital deve~tln;~hite or nearly white cattle. This condItIon-has
beenreport~d to cOllsist ofa-perniStent oan;'rof tissue which occludes the
vaginal opening just anterior to the urethra. Such an obstruction
obviously interferes with copulation. It has been shown that conception
may follow surgical removal of the tissue. However, some clinicians use
the term to describe abnormal genital structures in other than white
cattle and abnormalities not limited to vaginal occlusion.
Many different anatomical abnormalities have been described in the
female. Unilateral or bilateral absence of the ovaries, Fallopian tubes,
01' uterine horns occur in all species of farm animals. 1n cattle such
abnormalities can be readily detected by a rectal examination. Fortu-
nately, in both sexes, such abnormalities are not the most common cause
of reproductive failure.
Mechanical Injury of the Genitalia.-From an anatomical standpoint
the testes appear to be extremely vulnerable to injury; however, instances
of bruisir;w inflammation, and laceratioll~ of the scrotum and testes are
~on. The consequences of such injuries vary from temporary
reduction infertility to complete sterility, and all injuries should receive
prompt, expert treatment. The testes of the stallion are more likely to
be injured during service than in the other farm animals. Mares which
(
"''4 ...
are improperly handled or incorrectly restrained may permanently

damage the testes or penis of the stallion.
The penis
...--may become bruised or lacerated during service, thus result-
ing in temporary or permanent loss of the male as a sire. Most frequently
such injuries could be avoided by the use of approved breeding practices
or restraining devices.
A wide variety of injuries may occur in the female. The two critical
times, as far as injuries are concerned, are parturition and service. As
explained in an earlier chapter, complicate~ion may result in
the loss oj the yffspring, perm~nitalia of the female,
l ·lG. 66.-University of Massachusetts Guernsey cow with her 1941 triplets, which followed
sets of twins in 1938 and 1939. Two of these calves are bulls, one a heifer. What is the
chance that this heifer will prove to be fertile?
or even death of the dam herself. Perforation of the uterine or vaginal

wallyaceration of the cervix, eversion of the ~ix,ang_\lt~rus
or of the rectum~ sequelae of e~licated birth. n some cases
adhesionsorse-condary infections of the genitalia cause tissue damage
which prevents further reproduction. Service by a large or vigorous
male may cause permanent genital injury. Mechanical damage of the
female genitalia due to other causes is rare.
Genital Diseases.-TheoreticaJly, any disease which affects the well-
being of an animal may have either temporary or permanent influence
on fertility. In the male, for example, pneumonia, or any other disease
accompanied by fever, may affect the rat~~atogenesis. Severe
systemic infections may affect fertility or result in abortion in pregnant

females. Conditions of this type are outside the scope of this discussion.
Of the farm animals, cattle are most likely to be affected by specific
transmissible genital diseases. The economic losses suffered by cattle
as the result of such diseases are probably many times greater than for
all other farm animals combined. The authors will make no attempt
to estimate the magnitude of such losses. The official U.S. Department
of Agriculture estimate of losses attributable to brucellosis in cattle in
1942 was 30 million dollars (Mohler et al.). It has been estimated that in
N e'" York State alone the losses in cattle directly or indirectly due to
breeding difficulties are 20 million dollars per year and seem to be increa::;-
ing (Asdell, 1948).
Brucellosis.-Brucellosis of cattle (]'3ang's disease) is one of the major
causes of loss in cattle production. This disease is caused by a specific
organism, Brucella abortus. The brucella organisms ordinarily live in
the pregnant ~us,- bUt they may localize in other tissues such as the
udder or in the testes. One of the most frequent, but not the only,
evidence oflnlec-tlon is abortion. It is estimated that 85 per cent of
abortions.
in cattle are due
-'.-
to brucellosis. Abortion ma"YOcCul'
.....
........__-.-~.--~~
at any
stage of pregnancy but it is most likely to occur between the fifth and
~g~onths.Aborled calves and the "fetal membranes of brucella-
infectecrcattle usually contain millions of the brucella organisms and
the infection of other animals or the human may result from exposure to
such material. Infected animals do not always abort, and their mem-
branes may likewise contain the disease organism. The bacteria may
be eliminated in the genital discharges for several weeks and have been
known to be present in the milk for long periods of time.
Brucella infection may be introduced into or spread through a herd in
several ways. Most frequently an outbreak follows the purchase of a
supposedly healthy cow or the exposure of healthy cows to animals of
unknown disease status at public sales and exhibitions or in community
pastures. Brucellosis can be spread through contaminated feed, water,
and bedding or directly from the body discharges of an infected animal
of either sex.
One of the unfortunate misconceptions in regard to brucellosis is that
ahortion is the only sympton and the only cause of loss. Brucella
infection does not result in complete immunity, but most cows abort
only once following infection. The animals seem to develop a tolerance
for the organism, although a few do abort twice and rarely three or more
time::;. It has been estimated that 25 to 30 per cent of the cows which
contract brucellosis may manifest breeding difficulties. This may be due
206 BREEDING AND IMPROVEMENT OF' FARM ANIMALS
to various types of tissue damage in the genital organs. The fetal mem-
branes are frequently retained following abortion.
Although many "treatments" for brucellosis have been proposed,
research workers in this field know of no treatment which could be
described as a "cure." The present (1949) method of control is based
upon the detection of infected animals by testing the blood serum for the
presence of agglutinins. This is accomplished by the addition of an
antigen, prepared from Brucella abortus organisms, to measured amounts
of cattle serum. Calves, even those born to infected dams, usually
remain free of the disease if they are separated from infected animals and
are not fed infected milk. It is believed that the vaccination of calves,
when four to eight months of age, with a live vaccine prepared from a
strain of Brucella abortus organisms of low virulence (strain 19) will confer
sufficient immunity on such animals to protect them against subsequent
exposure to brucellosis. Official plans for the elimination and prevention·
of brucellosis in cattle have been developed by most state livestock
sanitary associations or the offices of the various state veterinarians.
Recommendations are likewise available through the United States Live-
stock Sanitary Association and the Bureau of Animal Industry of the
U.S. Department of Agriculture. Since no one plan is adaptable for all
situations or each specific locality, it is recommended that the official
agency responsible for the control of brucellosis in the area be contacted
w hen the presence of this disease is suspected.
Brucellosis in swine is most often caused by Brucella suis. Swine are
sometimes infected by Brucella abortus, and cattle may be affected by both
the abortus and 8uis Brucella. Man may be affected by three types of
Brucella, abortus, 8uis, and melitensis. The latter organism causes
brucellosis in goats but rarely in cattle and swine and is not of much
economic importance in the United States. Swine brucellosis is often
a serious economic problem to the swine producer and may occasionally
prevent the profitable operation of a swine enterprise (Hutchings, 1944).
In cattle the presence of brucellosis in a herd is soon recognized by the
occurrence of abortions. The incidence of abortions may vary greatly
between swine herds, ranging from an occasional abortion to the majority
of the pregnant females. In the boar the organisms may localize in the
testis and produce a severe orchitis. Hutchings and Andrews (1946)
have shown that brucella may be eliminated in large numbers in the
semen, and it is well known that the boar is frequently a spreader of the
disease. These bacteria may localize in the joints, causing arthritis; in
the vertebral column, causing posterior paralysis; and abscess formation
is not uncommon. •
Abortion may occur at any time during gestation. This symptom has
been observed in experimental herds as early as 23 days after breeding.

Under ordinary herd conditions such early abortions would probably
not be detected, and this may account for the fact that the abortion
rate sometimes appears to be very low. Swine usually abort only once,
but as in cattle, such animals may be dangerous spreaders of the disease
and their subsequent fertility may be affected.
The means of diagnosis of swine brucellosis is very similar to that of
cattle. Satisfactory vaccination procedures for the prevention of the
disease have not as yet been developed, and there is no known treatment.
Hmvever, Hutchings (1947) has reported considerable success in the
elimination of the disease from badly infected herds by the repeated
blood testing of all swine and the isolation of negative pigs on clean
premises as far removed as possible from the infected parent herd. The
testing is begun when the pigs are weaned at eight weeks of age and is
continued up to and during the first pregnancy. All reactors should be
removed as they occur. The infected parent stock should be disposed of
as soon as it is apparent that the younger animals are remaining free of
the disease.
Bovine Venereal Tr-_iclwmoniasis.-This disease is caused by a protozoon,
Trich()_rriiiiii!:sfe[us~~ Under natural conditions it is believed to be trans-
mitted only by copulation, but it can be transmitted by infected instru-
ments or in infected semen at the time of artificial insemination. The
incidence of this disease and the losses which result from it are not so wen
known as in the case of brucellosis. Morgan (cited by Bartlett) reports
that 61 infected herds were detected by the University of Wisconsin
animal disease diagnostic laboratory between 1941 and 1946. Bartlett
(1949) estimated that in nine trichomad-infected herds (more than 800
cattle) in the Beltsville, Md., area, the losses due to the infection \\'ere
at least $200,000.
The damage ,vhich occurs following infection is confined primarily to
the female. The organisms inhabit the uterus and bring about the early
destruction of the embryo, usually within 3 to 5 weeks after conception.
The animals then return in heat and may remain infected for several
months. In some cases the uterus may become filled with pus and
greatly enlarged. If such animals are not bred, they usually recover
spontaneously, since the protozoa cannot persist permanently in the non-
pregnant uterus. If noninfected bulls are used, the possibilities of
eliminating the disease are good, but care must be exercised that the
bulls do not become infected. The cattle must not be reb red until it is
known that the organisms are no longer present.
In the bull the trichomonads inhabit the preputial membranes and
glans penis. There is no outward sign of infection and the sperm are
unaffected. Unlike the female, the male remains permanently infected

and may transmit the organisms at nearly every service. The detection
of this disease in the bull requires the services of an expert, since the
habits of Trichomonas fetus can best be described as "tricky." It is
the opinion of most specialists in this field that infected bulls should be
regarded as permanently infected and should be disposed of. Numerous
types of treatment have been tried, and there is some experimental evi-
dence that a satisfactory treatment may eventually be developed (Bart-
lett, 1949).
,
Vibrio fetus.-Vibrio
'- -, ,-----~
fetus is recognized as one of the occasional causes
of abortion in both cattle and sheep. In cattle, abortion due to this
cause may occur at' any stage of pregnancy but most frequently between
the fourth and seventh months. The incidence of the disease, the per
cent of infected animals which abort, and the exact manner of trans-
mission of the disease are poorly understood (Bartlett, 1949). Abortion
in any species should be regarded as a potentially serious problem. An
attempt to determine the cause of each abortion should be made. If
brucellosis, the most probable cause, can be eliminated, the possibilities
that either Trichomonas fetus or Vibrio fetus might be involved should
be investigated.
It should be recognized that abortion, like sterility, is only a symptom
and that it can have many causes. Injury, failure of the corpus luteum
function, nutritional deficiencies, embryological abnormalities, systemic
infection, or any general physiologic imbalance may be followed by
abortion.
Vaginitis.-This term, which refers to inflammation of the vagina, is in
verygeneral use, but there is a divergence of opinion as to the C&,use or
causes and to the significance of the condition. It is believed that one
type of vener(:)al disease which occurs in cattle is due to a virus. This
disease is called v~uhr, \,~nereal disease and is supposedly rare in the
United States but common in Europe. It is recognized at the beginning
by small papules which form ulcers which in turn heal in about 2 weeks.
Both males and females may be affected (Bartlett, 1949).
-----
A second venereal disease which appears to be widespread is nodular
venereal disease (granular or nodular vaginitis). Some lesions or sup-
posed evidences of this disease can be found in nearly every cattle herd.
In some cases the tissue changes are confined to the vulva and in other
instances may involve the entire vagina. Varying degrees of inflamma-
tion of the mucous membrane may be present and small nodules may
occur. Lesions may be present in virgin heifers and either open or
pregnant cows (Bartlett, 1949). In some cases there is no apparent
effect on fertility and in others natural service may not be followed by
pregnancy. In obviously infected herds where the conception rate was

low following natural service, the use of artificial insemination has been
reported to improve breeding efficiency. The explanation for this is
apparently that the sperm are deposited in the cervix beyond the infected
area. There is evidence that the infection (or some infections) may be
spread by the bull, yet there is other evidence that the condition may
not be primarily infectious.
Infectious Abortion in Mares.-Because of the decline in brood-mare
n~';;'bers there is little current research on infectious genital diseases of
the horse. Dimock and Edwards (1928) report that, in their experience,
streptococcic abortion is most common in mares but that Salmonella
abortivo-equinus and abortion due to a virus do occur. ~
--~N~triti;;na.l Deficiencies.-Few topics in livestock production have
received so much attention as the il1fiuence of nutrition on reproduction.
Since grO\\-th and reproduction are the two most fundamental character-
istics of animals and since Doth are easily measured, it is not surprising
that both are included in the design of the majority of nutritional investi-
gations. There is hardly a nutritional substance or element that has
not at one time been thought to be essential for normal fertility. It is
self-evident that maximum reproductive efficiency depends upon the
maintenance of a reasonable state of body health. It is likewise under-
standable that the absence of an essential nutrient may indirectly affect
reproduction. In the case of a nutritional anemia, for example, the
severity of the anemia may cause general debilitation and weaken the
individual to the point where mating does not occur. It is entirely
possible that fertile ova or sperm might be produced but that reproduction
might be physically limited.
It is well known that the general plane of nutrition may affect fertility.
In the rat it has been shown that a decrease of about 30 per cent in energy
intake will affect fertility and that a marked retardation in the growth of
the immature rat or a loss of weight in the mature animal may reduce or
prevent gametogenesis in both sexes (Friedman and Turner, 1939). It
has been observed, although not corroborated by experimentation, that
both human and animal fertility is lowered following prolonged drought,
famine, or ,val'.
There is the general belief among livestock men that both extreme
undernutrition or a very high plane of nutrition are equally harmful. It
is commonly said that a particular animal is sterile because it is extremely
fat. This problem should be considered from several points of view.
From the physiological standpoint, it is well known that a reduction in
the production of the sex hormones is followed by increased fat deposition.
This is the basis for the practice of castration. It is logical, therefore. to
210 BREEDING AND IMPROVEMENT OF FAR1II ANIMALS
consider the possibilty that some animals which become excessively fat
on the same ration which produces satisfactory results in the rest of the
herd are fat because they are sterile, not sterile because they are fat.
From the nutritional standpoint, a ration which is conducive to fat deposi-
tion may be quite-unsatisfactory from the reproductive standpoint. The
requirements for fattening are more easily met than are those of growth,
reproduction, and lactation. A ration high in carbohydrates may pro-
duce excellent fattening but may not meet the requirements for optimum
fertility. For this reason the selection of breeding animals from the
fattening pens is subject to criticism.
Probably the best known of all feeding practices in relation to breeding
management is flushing. It has long been an accepted practice to
increase the plane of nutrition of breeding ewes prior to actual mating.
This can be accomplished in a variety of ways such as the use of lush
pastures and/or protein, carbohydrate, or vitamin supplementation. It
seems to be accepted as the result of experiments in several countries,
that flushing results in a larger lamb crop, although estrus is not induced
earlier than usual (Friedman and Turner, 1939). Yet, when ewes have
previously been maintained on a high plane of nutrition, flushing appears
to have no beneficial reproductive effects. An 8-year study at the
Oklahoma Station by Briggs e:t al. (1942) showed that, under the condi-
tions of the experiment, flushing did not increase the number of lambs
dropped and that the cost of the increased feed used during the breeding
season was not justified by an increase in lamb crop.
When a flock of purebred range ewes which had produced an 80 to 100
per cent lamb crop under range conditions was transferred to the Cali-
fornia Experiment Station, the lamb crop during the next 6 years aver-
aged 135 per cent and reached a high of 160 per cent. It appears that
improved feeding and management can markedly affect fertility under 1
some conditions (Hart and Miller, 1937).
At the Utah Station Esplin et al. (1940) found that differences in the
fertility of sheep were related to plane of nutrition during the lamb's
first winter. Ewe lambs fed supplemental feeds gained about 25 lb. from
October to April, whereas comparable ewe lambs on range gained 10 lb.
Both groups were on the same range from April until the following Octo-
ber at which time they were placed in the breeding flock. At breeding
time the lambs which received supplementary feed during the first winter
weighed only 2 to 3 lb. more than the others. The lambing results of
the two groups were quite different. Only 45 per cent of the "range"
lambs produced offspring, whereas 65 per cent of the" fed" lambs repro-
duced. It was concluded that the nutritional regime during the growth
period may subsequently influence reproductive capacity.
LOWERED FERTILITY AND STERILI,]'!' 211
Studies at the Massachusetts Station clearly showed that the testicular

development of boars was retarded when the plane of nutrition was
sufficiently reduced to limit growth rate.
A comprehensive review of nutrition and sterility in dairy cattle by
Asdell (1949) cites over 50 references on cattle alone.
early every vitamin or special dietary factor has been suspected of
limiting reproduction. In the farm animals, with the exception of the
chicken, vitamin supplementation has been chiefly confined to vitamins
A and D. While vitamin D can limit skeletal growth and general health,
FIG. 67.-Effects of vitamin-A deficiency in the bull. 1. Testis of vitamin-A-deficient bull;

observe absence of sperm and scarcity of all epithelial cells incl uding spermatogonia. 2.
Same bull after 4 months of vitamin-A therapy, extensive but not complete repair.
it has no direct effect on the genitalia. Vitamin A, however, is essential

for normal reproduction. Under natural conditions yitamin A deficiency
is likely to occur in animals which have been on dry feed, poor roughage,
or drought-stricken range for extended periods of time. Under such
conditions systemic evidences of vitamin A deficiency are usually recog-
nizable. Night blindness, ophthalmia, diarrhea, and difficulties in loco-
moti~n and coordination may occur. In the male the rate of spermato-
genesis and quality of the semen declines. In advanced stages of the
deficiency the male may be unwilling or unable to copulate, even though
motile sperm may be present. The cause of impaired fertility in the
male is usually attributable to testicular damage. In the female con-
ception may occur but is frequently followed by abortion, the birth of
212 BREEDING AND 111fPROVEMEN1.' OF FARM ANIMALS
dead or weak calves, and retention of the fetal membranes. Some work-
ers report that vitamin-A deficient cows do not conceive so readily as
normal animals (Asdell, 1949).
Several recent studies have been made of the effects of vitamin-A
deficiency in bulls. In the young male the absence of vitamin A prevents
both normal growth and normal testis development. Fluid-filled cysts
may occur in the pituitary gland and complete reproductive failure can
be expected. In sexually mature bulls vitamin-A deficiency produces
varying degrees of testis degeneration and general debility, depending
on the degree of the deficiency. In older animals the damage can usually
FIG. 68.-Vitamin-A deficient bull. Vitamin deficiencies are not always easily recognized
by the untrained observer. This bull exhibited impaired spermatogenesis, night blindness,
muscular incoordination and diarrhea, yet his general appearance is not unlike that of many
farm bulls.
be repaired following proper therapy (Sutton et al., 1940; Hodgson et al. ,

1946; Erb et al., 1947).
Although all of the B vitamins are probably not yet known (at least
13 had been named by 1949), it is thought that this important group
has little direct effect on reproduction in the larger farm animals, espe-
cially the ruminants. In the chicken the B vitamins are likely to be
absent or present in SUboptimum amounts in some rations, e.g., those
containing primarily corn and soybean oil meal. Proper B supplementa-
tion of such rations does improve hatchability.
Vitamin C is often present in suboptimal quantities in the diets of man,
LOWERED FER1'ILl1'Y AND STERILITY 213
monkey, and the guinea pig. The deficiency of this vitamin causes severe
debility and ~ven death, yet it is reported that there are no direct effects
on fertility in these species. It has been reported that the injection of
vitamin C (ascorbic acid) in both bulls and cows which had been classified
as "poor breeders" was followed by improved fertility in both sexes.
There seems to be current (1949) agreement that the therapeutic benefits
of ascorbic acid therapy have not been submitted to a critical test and
that judgment must be withheld until more clear-cut evidence is available.
The role of vitamin E in the pre-
vention or treatment of reproductive
disorders of the farm animals is still
a controversial subject. In the rat
and mouse there is no doubt that
vitamin E is essential for reproduc-
tion. A deficiency of this vitamin
prior to puberty resul ts in the absence
of sp~rmatogenesis and in testicular
atrophy following puberty. In the
female conception may occur but
abortion follows in the vitamin-E
deficient rat or mouse.
Numerous reports on the value of
wheat-germ oil (a rich source of
vitamin E) in the correction of cattle-
breeding problems appear in the
scientific journals and literally hun-
dreds of testimonials of satisfied users
have been published. Attempts to
reduce fertility in farm animals by
the use of vitamin-E free or deficient
rations have not accomplished this FIG. 69.-A severe case of hypothyroid-
purpose. It has been concluded in ing ism. Observe the thickening and crack-
of the skin.
a report by the National Research
Council that, "There is no convincing evidence that the addition of extra
vitamin E to normal rations has a direct or specific effect upon reproduc-
tion in practical herd and flock management" (Phillips, 1942).
The minerals which are most lilwly to limit animal growth and health
are phosphorous, calcium, and iodine and, in some regions at least, iron,
copper, cobalt, and manganese. Of these, phosphorous is most apt to
be deficient. Under natural conditions phosphorous deficiency is often
complicated by both protein and vitamin-A deficiencies. Under such
circumstances reproductive problems are almost certain to occur. In
214 BREEDIXG AND IMPROVEMENT OF FAll,)! AXIJfttLS
cattle phosphorous-deficiency symptoms, such as unthrjftjness ancl

depraved appetite, usually precede breeding difficulties.
The importance of iodine in the prevention of goiter, the necessity of
copper and iron in the prevention of anemia, and the role of calcium in
skeletal growth and lactation are such that the diet should ahvays contain
them. The relationships of manganese and cobalt to reproduction in
farm animals are not known. In areas where the state experiment
stations have shown these elements to be deficient, local recommendations
should be followed.
FIG. 70.- 1. Normal la mb thyroid. 2. Abnormal lamb th~'Toid due to iodine deficiency.
I
Genetic Causes of Sterility.-Certain forms of sterility apparently
have a genetjc basis. The work of Bridges with Drosophila has shown
that a male of sex-chromosome constitution XO (due to nondisjunction)
is sterile. Evidently the absence of a Y chromosome in this species
renders an animal sterile.
An instance from practical breeding history which appears to belong to this
category is that of barrenness in Bates' famous Duchess i:1mily of Shorthorns.
This family was noted for superior individual excellence, consequently breeders,
naturally desirous of maintaining this excellence, followed a practice of breeding
within the famiJy, an example of which is given in Fig. 71. But the family was'
tainted from the beginning with the curse of barrenness. . . . Shortsighted
breeders at the time considered it a fortunate circumstance that Duchess cows I
were so often barren, for it kept down the number of individuals of this favorite .·
strain and resulted in prices correspondingly high. But as a result of barrenness

the strain eventually ran out completely. In Fig. 72 an attempt has been made
to show diagrammatically how barrenness was inherited in this family. The
diagram is not complete, for it includes only the females in the family. Never-
theless it brings out very forcibly how barrenness occurred very early in the
family history, and how it reappeared in about the same proportion of the total
population throughout its history. Far from showing an intensification of the
defect as a result of inbreeding, this diagram merely illustrates the heredity of a
defective family trait. l
Probably most breeders of some years' experience can recall certain
females in their herds of 10 or 20 years ago that now have no descendants
in the herd. If a breeding chart of the herd has been kept or can be made,
Short Tail (2621) 1Belvedere (1706)
4th Duke of N orthum-
I 1Duchess 32nd
!
berland (3649) Duchess 34th 5Belvedere (1706)
Duchess 55th 1Duchess 29th
. lKorfolk (2377) 52nd Hub.back (1423)
Duchess 38th I Nonpareil
Duchess 33rd j Belvedere (1706)
IDuchess 19th
}'IG. 7l.-Pedigree of Olle of the late~t Duchess cows, illustrating ~ystem of Jinebreeding
followed in maintaining the family, Duchess 55th produced two calves.
such lines are practically sure of being evident. Some of these lines
perhaps disappeared because of poor type or poor production or both.
Others, no doubt, have disappeared because of lowered fertility or
sterility. Whether this has been due to poor management or to poor
genetic make-up is perhaps hard to determine, so to be conservative we
will say that some of it has been due to one, some to the other, some to
both.
( In species characterized by multiple births, fertility and prolificacy are
\ of extreme economic importance. This problem has concerned livestock
breeders probably since the domestication of animals and has received
the attention of research ,rorkers during the last half century. The fact
that there are significant differences in the prolificacy of different breeds
of swine and bet",reen lines within breeds is definite evidence that this
trait is heritable. The exact degree of heritability of prolificacy is not
known. In some instances where attempts to select for litter size in
swine have been made, there has been little increase in prolificacy.
Morris and Johnson (1932) found, that in 1,035 litters of Poland-China
swine, litter 8ize was increased only one pig per litter during a 20-year
period. In a more recent study at the Minnesota Station, Stewart (1945)
1 BABCOCK, E. n., and CLAUSEN, R. E., "Genetics in Relation to Agriculture,"
2d, ed, pp. 5.5.5-556. McGraw-Hill Book Company, rne., New York .. 1027.
estimated that the heritability of the number of pigs farrowed is about

17 per cent. In actual practice increasing fertility in an outbred stock
of low fertility would be a slow process if simultaneous selection for other
characteristics is made. However, even though a heritability of 17 per
FIG. 72.-Illustrating inheritance of barrenness through the female line of the Duchess
family of Shorthorns; barren cows represented by solid black circles. (From Babcock and
Clausen, Genetics in Relation to Agriculture.)
cent seems low, it does tend to show that breeders have not selected so
closely for fertility as is possible. In a companion study of other factors
which affect prolificacy of swine, Stewart (1945) found that age and
weight of gilts at mating account for 4 per cent of the variance in size of
first litters and were of considerable importance in the Helection for
WWERED FERTILITY AND STERILITY 217
fertility. An increase of 10 pel' cent in the inbreeding of dams of the

flame age resulted in a decrease of about 0.6 pigs pel' litter.
The inheritance of functional regulation of reproductive efficiency has
beenreviewed by Gilmore (1949). This review, containing 142 refer-
ences, cites data clearly illustrating the fact that the degree of fertility
can have a genetic basis. Because of the nature of the data and the
complexity of the problem, the mechanisms of genetic regulation remain
obscure. Of the various factors which may influence fertility or result
in sterihty, the follmving have been shown to be heritable: ,yhite heifer
disease, the freemartin condition, underdevelopment ~of the gonads,
abscence of the gonads or various portions of the genitalia, and inter-
sexuality. A wide variety of lethal factors resulting in fetal or early
postnatal death do i:[;fluence fertility and may be regarded as causes of
reproductive failure. Such lethal factors occur in all species."__'-
Systems of Breeding and Sterility.-The fact that too wide crosses,
e.g., bison and cattle, or horse and ass, often result in sterile -or relatively
infertile offspring, coupled with the fact that too close inbreeding often
produces similar results as far as sterility is concerned, is further proof
that some sterility is caused by genes.
In selecting breeding stock, therefore, one should be careful to make
his selections only in strains or families that by their actual records of
reproduction have proved themselves to be at least relatively free from
genes which lower fertility. It is often observed that animals of a strain
which has been closely inbred for a number of generations are sterile
when mated among themselves but fertile with other nonrelated strains.
In fact, the above condition sometimes applies as well to animals that
are not closely related. In a great many of the experiments on inbreed-
ing, there have gradually developed a loss of vigor and a lowering of
fertility. For instance, Weismann's and Von Guaita's work with micp.
gave the following results.
TABLE 14.-DECREASING FERTILITY IN IlI<'IlRED MICE
Average
Generation Per Litter
1-10 .................. 6.1
11-20 .................. 5.6
21-29....... . . . . . . . . 4.2
29 and 30...... .... . .. 3.5
31 and 32 .............. 3.6
33 and 34 .............. 2.9
Similar work by Ritzema Bos on rats gave the following results:

TABLE 15.-DECREASED FERTILITY DUE TO INBREEDING
1887 1888 1889 1890 1891 1892
7>2 777 717"17 62~6 4Yia 3>t
This effect of inbreeding is not universal, however, as King reports

on two series of rats (A and B) inbred, litter brother and sister, for 25
generations:!
Data given for the A series of inbreds comprise 1,752 litters containing 13,116
individuals, or an average of 7.5 young per litter; records for the B series of
inbreds include 1,656 litters, having a total of 12,336 members, or an average of
7.4 young per litter. The two series combined comprise a total of 3,408 litters,
which contained 25,452 individuals. For the entire strain the average size of
the litter was 7.5 young. . . .
A comparison of the data for the inbred strain with data for litter size obtained
from a series of stock Albinos reared under the same environmental conditions
as the inbred strain shows that each litter of the stock series was relatively smaller
than the corresponding litter in the inbred group. For the entire series of 424
stock litters the average size was 6.7 young per litter. This average is 0.8 less
than the average for litter size in the inbred strain.
Crossbreeding, on the other hand, where the resulting individuals are

not completely sterile, as is the mule, often produces plants or animals
1
that are more vigorous in all respects than their pure parents and are
often more fertile. This increased vigor is termed heterosis or hybrid
vigor. The exact cause of these phenomena in inbreeding and in cross-
breeding is not fully understood. It has been suggested, however, that l
the germ plasm of all species probably carries some inhibiting factors,:
which, if true, explains why inbreeding often does result in lowered
fertility, because these inhibiting factors are intensified, an~ why cross-:'
breeding often results in increased fertility, because each diverse parent'.
supplies to the offspring factors that tend to dominate and offset the)
expression of the weak characters supplied by the other parent.
Among species hybrids, there is wide range of ability to reproduce, viz.,
from a point of increased fertility over the parents down to a point of
absolute sterility. The mule and the hinny, the former a jack-mare
cross and the latter a stallion-jennet cross, are both sterile, as are also
the horse-zebra crosses. Babcock and Clausen report ~ne reference to
fertile male hybrids between the zebra and the ass. The researches of l
Wodsedalek on the reproductive cells of the mule show abnormal reduc- (
J
tion divisions during spermatogenesis, which should logically result in
imperfectly formed and functioning germ cells. If the hybrids in any
cross have characteristically a different number of chromosomes, it is
difficult to see how the reduction divisions could be normal and result in
the production of functional germ cells. Even if the chromosome num-
bers were the same in the two species, physiological incompatibility
might be present. 2
1 KING, H. D., "Studies in Inbreeding," Wistar Institute, Philadelphia, 1919.
• BABCOCK and CJ,AUSEN Op. cit., p. 591.
LOWERED FER1'ILITY AND STERILITY 219
In many crosses between species of Bovidae, as common cattle with bison,
gaur, gayal, and yak, the female is fertile and the male sterile. Among domesti-
cated birds, as in the sterile hybrids between various species of pheasants, the
se1.'llal organs are imperfectly formed, and there is a strong suspicion that sterility
may be due in part to imperfect sex determination resulting from hybridization.
Even aside from this difficulty which is zygotic, there is the further one that the
chromosome condition in these hybrids may be unbalanced and the differences in
genetic content of the chromosomes may be so great as to prevent functioning of
the gametes which are formed. The sterility of species hybrids is in quite a
different category from that of the sterility occurring within species. l
Davenport observes in regard to hybrids that" it is safe to assume that

about all the possible fertile ' hybrids were long ago produced in nature,
and either went down under natural
selection or became good species
before they came into our hands."
Physiological Disturbances.-When
the cause or causes of lowered fertility
are not readily apparent, they are
frequently attributed to physiologic
disturbances. Such diagnoses are
sometimes made because of a general
lack of information regarding the
particular condition and sometimes
because of the limitations of the
diagnostic techniques used. For ex-
ample, a highly skilled gynecologist FlO, 73,-Orchitis in It raID. Examina-
tion showed that the affected testis had
might find no gross evidences of genital increased in size to the extent that the
abnormalities as the result of rectal normal testis had been forced up into
the inguinal canal.
and vaginal examination, and there-
fore conclude that there were no anatomical or pathological lim-
itations for conception. There might, ho'wever, be tissue changes in
the Fallopian tubes which would prevent the passage of sperm or ova
but which could not be detected except by microscopic post-mortem
examination. Likewise, there may be transmissible genital infections
which are as yet unrecognized, or vitamin or trace-mineral deficiencies
which are still unknown. There are, however, several known physio-
logical disturbances which may prevent normal reproduction.
The failure of the anterior pituitary gland to produce the gonadotropic
hormones at the proper time and in the proper amounts will most cer-
, tainly affect fertility. The absence of ovarian or testicular development
and consequent failure of the animal to become sexually mature can
• t BABCOCK and CLAUSEN Op. cit., p. 591.

often be explained on the basis of pituitary dysfunction. Likewise, the

cessation of the estrual cycle or the loss of sex drive and sperm production
in the mature animal may follow pituitary failure. Such animals can
sometimes be restored to a breeding condition by the use of the gonado-
tropic hormones, but the fact that such tendencies may be heritable
always raises the question as to the desirability of this approach. Theo-
retically, at least, abnormal function of nearly any endocrine gland may
indirectly affect the reproductive process. Extreme under- or over-
activity of the thyroid gland may alter fertility. Abnormal adrenal
cortical function may affect general body health and thus secondarily
limit reproduction. Tumors of the adrenal gland may produce abnormal
quantities of the sex hormones and even bring about sex reversal changes.
Tumors or other abnormal growths of the gonads and accessory genital
structures may produce abnormal quantities of hormones, especially the
steroid sex hormones and gonadotropins, and influence the entire body
as well as the genital system.
The above conditions may be difficult to recognize in the farm animals,
and accurate data as to their incidence are not available. There are,
however, physiologic disturbances which nearly every livestock owner has .
observed. Among the more common of these is the irregular estrual
-21.91e. The heat periods may occur at very irregular inte~vals without
apparent pattern. In some cases hefJ.t does not occur at all, or if it does
is of such low intensity that it is difficult to detect. In other instances
the heat periods may be abnormally long or short. Even though ovula-
tion may take place and the ova may be viable, the time irregularities
make it difficult to have ovulation and copulation coincide. Thus, poten-
tially fertile animals may remain barren. The cause of these irregular-
ities is undoubtedly endocrine and probably involves both the anterior
pituitary gland and the ovaries. Seasonal and nutritional factors may
be involved.
Abnormal corpus luteum function can affect fertility in at least several
distinct ways. As previously described, the corpus luteum may persist
in the ovary and continue to secrete progesterone. If progesterone is
being produced in large amounts, new Graafian follicles are not formed
and estrus and ovulation do not take place. This situation is typical
of the lactating sow and the dairy cow during the first part of lactation.
In the sow the corpora lutea undergo regression very shortly after weaning
the pigs, and in high-producing dairy cows the estrual cycles should be
naturally resumed within 40 to 60 days following parturition. The
corpus luteum can be removed by ovarian manipulation in cattle, and
heat can be induced in sows by the injection of equine gonadotropin.
The failur~ of the corpus luteum to function properly during pregnancy ....·
may prevent implantation of the fertilized ovum or may result in the

abortion of the embryo or fetus. Sufficient data are not yet available
to justify definite conclusions, but it seems logical to explain some of the
reproductive failures which occur in farm animals as due to deficient
progesterone secretion.
In cattle, particularly, some females remain in more or less constant 1
heat, will accept the bull, but fail to become pregnant. If this condition 1
persists, the tail head frequently becomes elevated, the animal may bellow
a great deal, and behavior may be somewhat more masculine than
feminine. The behavior is frequently referred to as nYIllphomania,
meaning unusual sexual de_sire. The general condition is often called
~i~ oV_3::r:Y_oecause of the development of large, fluid-filled, thick-walled
cY§t~in the ovaries. These cysts, in contrast to normal Graafian follicles,
do not rupture spontaneously, and ovulation does not occur. The con-
diti;n has been recognized for many years and was formerly treated by
the mechanical rupture of the cysts.? Following the discovery of the
gonadotropic hormones, it was found that the secretion or injection of the
follicle-stimulating hormone alone was often followed by the development
of very large follicles ·which did not rupture. The injection of FSH and
LH in proper amounts and sequence resulted in both follicular develop-
ment and ovulation. In light of this knowledge it appears that cystic
ovary is probably the result of an endocrine imbalance. Research
workers at the University of Wisconsin have found that the injection of
special pituitary gonadotropic hormone preparations will cause the rup-
ture of such cysts and will restore a large percentage of nymphomaniac
cows to breeding usefulness (Casida et al., 1944).
One of the most perplexing problems in artificial-breeding associations
is the" repeat-breeding" cow which shows no detectable genital abnor-
malities. Some facetiously refer to this problem as "sterility in normal
cows." A carefully selected group of 104 such cows was studied at the
University of Wisconsin by Tanabe and Casida (1949). These animals
were selected by veterinarians experienced in cattle-breeding problems.
All animals had been bred at least four times without apparent con-
ception.. all had previously calved at least once, were less than ten years
old, and were free of genital abnormalities of all types as nearly as could
be determined by gross examination, and exhibited estrual cycles of
normal length. In order that management might not bias the results,
no more than two cows were selected from anyone herd. The cows
welie artificially inseminated and slaughtered on either the third or thirty-
fourth day following breeding. At 3 days 66.1 per cent of the cows
examined had fertilized ova, but at 34 days only 23.1 per cent had normal
embryos. Visible genital abnormalities not detected by clinical exami-
222 BREEDING AND IMPROVEMENT OF PARM ANIMALS
nation were found in 10.6 per cent of the animals and 8 per cent were
physical barriers to fertilization. It was concluded that the cows could
be divided into three groups, (1) failure of fertilization, 39.7 per cent;
(2) embryonic abnormalities and mortality before 34 days, 39.2 per cent;
and (3) embryos still normal at 34 days, 21.1 per cent. Exactly why
fertilization fails to occur and why early embryonic mortality frequently
follows when fertilization does take place have not been satisfactorily
answered.
Theoretically, at least, any marked deviation from normal in secretion\
of estrogen, androgen, or the gonadotropins may affect fertility. When
one considers that the entire genital tract of each sex, as well as the
gametes themselves, are dependent upon these hormones, this does not
seem unusual. Although these hormones have been available for clinical
use for at least a decade, thus far their practical applications in both
human and animal medicine leave much to be desired in the treatment
of sterility. Rapid and encouraging advances have been made, but as
previously pointed out, sterility is not a single problem and no one treat-
ment will correct it.
Environment and Reproduction.-Sufficient evidence has already been
presented to demonstrate the importance of light and temperature in the
regulation of reproduction. The reproductive patterns of the seasonal
breeders and the seasonal fluctuations in fertility of the continuous
breeders should be thoroughly understood for each species and class of
animal if a breeding program is to be carried out most efficiently. The
importance of relative humidity and altitude are largely unknown, but
it seems safe to conclude that when these factors affect the general well-
being of an animal, fertility is likely to be affected.
The hazards of the atomic age to man and animal have received much
attention in both the popular and scientific press. Many of the con-
clusions which have been drawn or hinted at have no basis of fact, but
others of equally startling nature have been thoroughly substantiated.
It is indeed interesting to speculate as to the factors which regulate
genetic make-up and induce mutations. Whatever these are, they have
been going on in a reasonably orderly way for thousands of years. It is\
well known that proper dosages of X rays and other types of radiation
do affect cells, chromosomes, and genes and may alter the rate of muta":
tions. But as to whether whole populations exposed to atomic radiations
will ,become altered or even extinct within a few generations is strictly
within the realm of speculation in 1950.
The deleterious effects of X rays upon the testes and ovaries were
known as early as 1903. The germinal epithelium is especially sensitive
to X rays, and complete and permanent sterility can be produced In
,
either sex by proper dosage. The degree of damage is proportional to

the time and dosage and may vary from temporary lowered fertility to
permanent sterility. Exposure of the genital organs to X rays, even for
short periods of time, is hazardous and should not be permitted except
for excellent reasons and under expert supervision.
During the past 5 years many studies of the effects of externally applied
beta and gamma rays and fast neutrons on various body tissues have
been made. The data are far from complete and only a few species have
been studied. At this writing it appears that the effects of these radia-
tions on the gonads are qualitatively similar to X rays. Numerous
experiments of the effects of internally administered radioactive isotopes
have been performed. In a general way it appears that the effects of
such isotopes on reproduction will depend upon the active life of the
isotope, the amount and type of its radiation, the rapidity of elimination
from the body, and the types of tissue in which the isotope is concentrated
if it is retained in the body (Bloom, 1948).
Summary.-In this chapter we have stressed the difference between
sterility and degree of fertility and have e~phasizeathe point that sterility
and fertility can be influenced by a wide variety of known factors and
perhaps an equal or greater number of unknown factors. Anatomical
abnormalities of the genitalia can be readily detected and should be
regarded as serious and disqualifying defects in breeding stock. Mechan-
ical injuries of the reproductive organs are often the result of careless
management and should be guarded against. When injuries do occur,
they should be promptly and expertly treated. Transmissible genital
diseases, especially brucellosis. are an important cause of economic loss
to livestock producers. Since effective diagnostic methods are available
for the detection of several of these diseases, the introduction of animals
of unknown disease status into healthy herds and flocks, at least without
a period of isolation, is risky and inexcusable. The importance of feeding
well-balanced rations supplemented with the nutritional factors now
known has been discussed. That certain genital abnormalities and level
of fertility have a genetic basis is clearly a matter of fact. The fre-
quency and mode of inheritance of some of these conditions are obscure,
but the possibilities that the conditions will be transmitted are sufficiently
great that rigid selection must be practiced. Some of the more common
physiological disturbances and their causes have been discussed.
The breeder's task would seem to be that of selecting his breeding
animals from naturally fertile strains; of feeding and managing mature
males and females so that viable healthy germ cells may be produced;
of growing out his young stock in a sanitary environment and on suitable,
complete rations; of attempting to strike an intelligent balance between
224 BREEDIiliG AND 11IlPROVEMEj'IiT OF FARM ANIMALS
production and reproduction; of setting all the barriers possible between

his animals and injuries or organisms that might lower their breeding
efficiency; of keeping accurate records of all matters pertaining to breed-
ing health and efficiency so that he may know at all times where he
stands and in what direction he may be heading.
Finally, it hardly seems necessary to warn against putting our faith in
some cure-all of doubtful efficacy and thereby delaying the real job of
securing professional service, thus allowing some malady to get a firm
foothold. Such a policy is the best possible example of being "penny-
wise and pound-foolish."
References
Books
BLOOM, W. 1948. "Histopathology of Irradiation from External and Internal
Sources," McGraw-Hill Book Company, Inc., New York.
"Diseases of Cattle." 1942. U.S. Dept. Agr. Bur. Anim. Indus.
"Diseases of the Horse." 1942. U.S. Dept. Agr. Bur. Anim. Indus.
MORGAN, B. B. 1944. "Bovine Trichomoniasis," Burgess Publishing Co., Minne-
apolis.
WILLIAMS, W. L. 1939. "Diseases of the Genital Organs of Domestic Animals,"
W. L. Williams, Ithaca.
ASDELL, S. A. 1949. Nutrition and the Treatment of Sterility in Dairy Cattle:
A Review, Jour. Dairy Sci., 32 :60-70.
1948. Sterility and Delayed Breeding in Dairy Cattle, N. Y. Agr. Col.
(Cornell) Ext. Bul. 737.
BARTLETT, D. E. 1949. Infectious Disease as a Cause of Infertility: A Review,
Jour. Dairy Sci., 32 :36-44.
BRIGGS, H. M., et al., 1942. The Influence of Nutrition on the Reproduction of
Ewes, Okla. Agr. Expt. Sta. Bul. B-255. .
CASIDA, L. E., MCSHAN, 'V. H., lind MEYER, R. K. 1944. Effects of an Unfrac-
tionated Pituitary Extract upon Cystic Ovaries and Nymphomania in Cows,
Jour. Anim. Sci., 3 :273-282.
ERB, R. E., et aZ. 1947. Observations on Vitamin A Deficiency in Young Dairy
Bulls, Jour. Dairy Sci., 30:687-702.
ESPLIN, A. C., MADSEN, M. A., and PHILLIPS, R. W. 1940. Effects'of Feeding
Ewe Lambs during Their First Winter, Utah Agr. Expt. Sta. Bul. 292.
FRIEDMAN, M. H., and TURNER, ·W. A. 1939. Nutrition and Reproduction, U.S.
Dept. Agr. Yearbook, pp. 482-491.
GILMORE, L. O. 1949. The Inheritance of Functional Causes of Reproductive
Inefficiency: A Review, Jour. Dairy Sci., 34:71-91.
HART, G. H., and MILLER, R. F. 1937. Relation of Certain Dietary Essentials to
Fertility in Sheep, Jour. Agr. Res., 55:47-58.
HODGSON, R. E., et aZ., 1946. The Effect of Vitamin A Defi('.ieney on Reproduction in
Dairy Bulls, Jour. Dairy Sci., 29 :669-687.
HUTCHINGS, L. M. 1947. Field Control Experiments with Brucellosis in ~winc,
U.S. Livestock Sanitary Assoc. Proc., pp. 124-136.
1944. Swine Brucellosis, Mich. State College Veterinarian, Winter No.,1944,
\
\
\
- - - and ANDREWS, F. N. 1946. Studies on Brucellosis of Swine: III. Brucella

Infection in the Boar, Amer. Jour. Vet. Res., 7 :379-384.
J']NNINGS, W. E. 1941. Some Common Problems in Horse Breeding, Cornell Vet.,
31 :197-216.
MOHLER, J. R., et al. 1942. Losses Caused by Animal Diseases and Parasites,
U.S. Dept. Agr. Yearbook, pp. 109-116.
MORRIS, H. P., and JOHNSON, D. W. 1932. Effects of Nutrition and Heredity upon
Litter Size in Swine and Rats, Jour. Agr. Res., 44:511.
PHILLIPS, P. H. 1942. Nutrition and Reproduction of Farm Animals, Nall. Res.
Council Reprint and Cir. Ser. 112.
STEWART, H. A. 1945. An Appraisal of Factors Affecting Prolificacy in Swine,
Jour. Anim. Sci., 4 :250-260.
1945. The Inheritance of Prolificacy in Swine, Jour. Anim. Sci., 4 :359-366.
SUTTON, T. S., KRAUSS, W. E., and HANSARD, S. L. 1940. The Effect of Vitamin A
Deficiency on the Young Male Bovine, Jour. Dairy Sci., 23 :574.
TANABE, T. Y., imd CASIDA, L. E. 1949. The Nature of Reproductive Failures ip
Cows of Low Fertility, Jour. Dairy Sci., 32 :237-246.
CHAPTER VIII
PREGNANCY AND PARTURITION
The factors concerned with the initiation of pregnancy have been

discussed in a previous section. The factors concerned with the main-
tenance of pregnancy and with the birth of the new individual at the time
of parturition are of equal importance with conception, since the objective
of the livestock breeder is the production of normal, healthy animals
capable of efficient milk, meat, wool, egg, or work production. During
pregnancy the uterus must undergo considerable modification in size apd
structure to accommodate the rapidly growing fetus. The uterine mus-
cles must increase in size and numbers but must not be activated too soon
lest premature expulsion of the fetus occur. At the time of parturition
there must be sufficient relaxation of the birth canal to provide a passage-
way for the offspring, yet there must be sufficient contractility of the
uterine muscles to expel the young quickly and at the same time without
damage. In mammals, as described in a separate chapter, the develop-
ment and function of the mammary glands is an integral part of the
reproductive process. The development of the mammary glands is
begun at puberty and is completed during pregnancy. The initiation
of active milk secretion is ordinarily dependent upon normal parturition,
although if the mammary glands are sufficiently developed, copious milk
secretion may follow abortion or manipulation of the teats.
Ovarian Hormones during Gestation.-As previously outlined, both
estrogen and progesterone are essential for the preparation of the uterine
endometrium for implantation of the embryo. Ovariectomy before the
uterine lining is prepared will prevent implantation. There are great
differences between animals in the dependence of pregnancy on the
ovaries. In the mare and woman the ovaries can be removed in early
pregnancy without harmful effects, but in the cow, rabbit, and rat the
ovaries are essential at least until very late pregnancy. In the pregnant
human relatively large amounts of both estrogen and progesterone are
excreted in the urine. Urinary estrogens are detectable as early as the
first week of pregnancy and reach a peak just before parturition. Most
of the urinary estrogen is in the conjugated form as estriol glycuronide
and has little physiological activity as such. Progesterone is eliminated
in the urine as pregnandiol glycuronide. It is present in small quantities
226
\
\
\
PREGNANCY AND PARTURITION 227
during the first third of pregnancy and then increases gradually until
parturition. Both estrogen and progesterone practically disappear from
the urine after parturition.
Endocrine Functions of the Placenta.-There is little doubt that the
placenta is an endocrine organ of considerable significance. The hor-
monal functions of this organ are best understood in woman and the mare.
Why these two species should have so much in common and why the
other farm animals should differ so much from the mare is as yet unknown.
As already discussed, there is little doubt that the placenta can produce
estrogens and progesterone. The importance of these hormones in preg-
nancy is not clear.
Perhaps the most remarkable activity of the human and horse placentae
is the secretion of gonadotropic hormones which are very similar to those
produced by the anterior pituitary gland. These gonadotropins are
secreted by the chorion in the human and the endometrial cups in the
mare and are referred to as chorionic gonadotropins or as anterior
pituitarylike hormones. In the human the hormone appears in relatively
high concentration in the urine during the second "'eek of pregnancy. In
the mare the hormone is not present in the urine in easily detectable
amounts but is abundant in the blood serum or plasma from the forty-
fifth until about the one-hundredth day of pregnancy. This substance
is frequently referred to as equine gonadotropin. The physiological
response of test animals to the human and horse gonadotropins suggests
that the human substance is most apt to cause luteinization and rupture
of ovarian follicles and that the pregnant mares' serum produces both
follicular growth and ovulation. The purpose of these placental gonado-
tropins is not clear. In the mare during the period of maximum gonado-
tropin secretion the ovaries become extremely active and produce numer-
ous Graafian follicles which may rupture and pe followed by corpora
lutea. Why the mare, a species in which more than one fetus cannot
safely develop at anyone time, should produce many follicles when
pregnant is a biological mystery. Estrus and ovulation are not uncom-
mon in pregnant females of other species but superfetation, the establish-
ment of distinctly separate pregnancies, is a rare phenomenon.
Pregnancy Diagnosis.-The importance of knowing whether pregnancy
has been established cannot be emphasized too much. The cessation of
the heat. periods is nature's first indication that conception has occurred.
Pregnant animals fail to come in estrus because of the persistence of the
corpus luteum of pregnancy and its inhibition of follicular development.
In most instances the failure of estrus to reappear is a reasonably reliable
sign of pregnancy. However there are sufficient violations of this rule
to make it unreliable for maximum livestock-breeding efficiency. In
228 BREEDING AND IMPROVEMENT OF' FARM ANIMALS
seasonal breeders the absence of heat may indicate that the ovaries have
become quiescent. This is the rule in sheep and occurs in other farm
animals, particularly the mare.
If the corpus luteum persists in the nonpregnant female, estrus will not
follow. The corpus luteum of lactation, which forms shortly after
parturition in cattle and swine, may prevent the reestablishment of the
estrual cycle. The implications of this will be discussed later. It is
clear, therefore, that more exact methods of detecting pregnancy are
desirable.
In species in which manual examination or palpation of the uterus can
be made, a trained person can diagnose pregnancy with a high degrae of
accuracy by the second or third month of gestation. This method is
limited to the cow and mare. It is an invaluable aid in the diagnosis
und correction of infertility, but unfortunately the supply of qualified
diagnosticians cannot meet the need. Other indications of pregn[',ncy,
such as the cervical seal, the ascultation of the fetal heart, fetal ·move-
ments, increased size of the female, or the tendency of pregnant animals
to become quiet and to fatten, are either unreliable or require the passage
of considerable time before they are useful.
Numerous laboratory tests have been devised for the determination
of pregnancy. These are based on the different levels of hormones found
in the blood and urine of pregnant females. In the human the Friedman
modification of the Ascheim-Zondek test is the one most generally used.
It is based on the greatly increased amount of anterior pituitarylike
substance found in the urine of pregnant women. The procedure consists
of injecting 7 to 10 cc. of morning urine into a mature female rabbit's ear
veins at two 12- to 15-hour intervals. The rabbit is then killed at the
end of 48 hours and the ovaries removed for examination. The presence
of subserous hemorrhagic areas or corpora lutea constitutes a positive
reaction, whereas clear retention follicles with no hemorrhagic areas are
signs of a negative reaction. This test is positive, in cases of pregnancy,
3 or 4 days after the expected date of the first missed menstrual period.
Since estrogen is also produced in increased amounts during pregnancy, its
presence in urine, as determined by estrual changes in the vaginal smears
from rats, is also used to determine pregnancy.
In the mare, pregnancy may be detected from the fortieth to one-
hundredth days of gestation by injecting the mare's serum into immature
female rats. A positive reaction is manifest by increased size and activity
of the rat ovaries and uterus because of the increased gonadotropic
hormone in the blood of pregnant mares. From the one-hundredth day
of pregnancy untH term, sufficiently large amounts of estrogens are
present in mare's urine to cause castrate female rats to exhibit estrus or
spayed mice to show certain vaginal changes when injected with urine
of the pregnant mare.
Many modifications of these basic tests have been made. It is possible
to diagnose human pregnancy within 5 or 6 hours following the injection
of urine into proper test animals. Various chemical tests have been
developed for the detection of urinary estrogens in the diagnosis of
pregnancy. Unfortunately, no practical laboratory tests for the detec-
tion of pregnancy are available for farm animals other than the mare.
This is because the hormonal patterns of the other animals differ from
the horse.
Other Endocrine Changes.-During pregnancy the pituitary gland,
particularly the anterior lobe, increases in 'weight and activity. Since
this gland regulates the activity of the gonads, thyroid. adrenal cortex,
and mammary glands, and is of vital importance in growth, protein,
carbohydrate, and fat metabolism, it is not surprising that it should be
modified during pregnancy.
In the human, and probably other species as well, the thyroid gland
becomes markedly enlarged in 50 to 90 per cent of all pregnancies.
Increased thyroid activity is necessary in order that the pregnant female
may satisfactorily meet the needs of both her own and the tissues of the
developing fetus. Insufficient thyroid activity may result in abortion
or, in the case of farm animals, hairlessness, weakness, and high mortality
during the first few dl1Ys of life. Thyroid deficiencies are particularly
common in the iodine-free areas of the United States, and livestock men
in such regions supply pregnant females with iodized salt or potassium
iodide to prevent the difficulties described.
The parathyroid glands, which are located on the surface of or adjacent
to the thyroid, are directly concerned with calcium metabolism and
indirectly with the regulation of phosphorus metabolism. The increased
needs of calcium and phosphorus during pregnancy and lactation result
in increased parathyroid activity. If calcium is not provided in sufficient
amounts, because of dietary inadequacy or underactivity of the para-
thyroid glands, calcium will be withdrawn from the bones and increased
nervousness or severe muscular tetany may result. Milk fever, which is
in reality temporary calcium deficiency due to the sudden need of calcium
for milk production, is most satisfactorily treated by the intravenous
injection of calcium gluconate, which supplies the body needs until the
parathyroids become adjusted to regulating calcium metabolism at a
new level.
Pseudopregnancy.-Among the most characteristic changes which
occur during pregnancy are cessation of the estrus cycle, proliferation of
the ute~ine endometrium, and development of the mammary glands.
These phenomena are observed in several species in the absence of preg-

nancy. In the rabbit, cat, and ferret, ovulation is normally dependent
on copulation. In these species, following mating ,vith a sterile male
or mechanical stimulation of the cervix, pseudopregnancy is character-
istic. Ovulation, development of the corpora lutea, proliferation of the
endometrium, and mammary growth occur. In other words the repro-
ductive and mammary systems assume that pregnancy has been estab-
lished. In these three species pseudopregnancy lasts for about half the
normal gestation period, and estrus cycles are then resumed. In the
bitch estrus is followed by a long period of time during which corpora
lutea persist. This occurs whether copulation takes place or not. There
is extensive mammary development and from external appearances the
animal seems to be pregnant. Pseudopregnancy in the dog extends for
a period of time equal to normal gestation and is frequently follo"lYed by
lactation. Milk production does not persist, however, unless suckling
is allowed to occur.
Pseudopregnancy of the types described does not occur in the farm
animals. Corpora lutea sometimes persist for long periods of time in
nonpregnant animals, particularly high-producing cows and lactating
swine, but the other characteristics of pseudopregnancy do not occur.
Physiological Regulation of Parturition.-Why parturition occurs at
the moment it does remains one of nature's yet unexplained mysteries.
Dozens of theories have been developed, but as yet it is impossible to
induce parturition at will. It is obvious that the survival of the newborn
does not depend upon birth at a partic111ar moment. In the human,
for example, the normal gestation period extends from 220 to 330 days,
nor does the gross size of the fetus determine its intrauterine life.
The relative distention of the uterus, pressure of the fetus on the
cervix, stimulation of the cervical ganglia, and changes in endocrine·
balance may all contribute to the termination of pregnancy. In recent
years there has been increasing evidence that the hormonal relationships
are the determining factor, but even endocrinologists will agree that much
is yet to be learned of this process. Estrogen, progesterone, relaxin" and
the posterior pituitary oxytocic factor appear to be the chief hormones'
involved. During early pregnancy estrogen is concerned with the
increase in size and numbers of cells in the uterine lining and muscle
layers. Since the uterus must be relatively quiescent during pregnancy .
lest premature emptying take place, it seems logical to conclude that
progesterone is necessary for endometrial development and the jnhibition
of uterine contractions. The fact that pregnancy is usually terminated
in the cow following removal of the corpus luteum lends suppo~t to this
view. The uterine muscle-stimulating factor of the posterior pituitary
gland (oxytocic principle) readily increases uterine contractility under

certain conditions, although there seems to be some doubt as to whether
it is responsible for the initiation of uterine contractions at parturition.
Since it is known that estrogen sensitizes the uterine muscles to the
oxytocic factor, it is not surprising that the follo,ying theory has devel-
oped: The increasing secretion of estrogen during the latter part of preg-
nancy, followed by a decline in progesterone secretion, culminates in
vigorous uterine contractions induced by the posterior pituitary hormone.
This theory has much to commend it, but since its application will as yet
not induce parturition at will, it can be accepted only with reservations.
Relaxin.-In 1926 Hisaw found that the blood serum of the pregnant
rabbit would cause relaxation of the pelvic ligaments of virgin guinea
pigs. The principle responsible for this reaction was later found in the
blood of pregnant women, mares, and sows, and was extracted from the
placenta and corpus luteum. This substance was called relaxin. Since
relaxation of the pelvic region is characteristic of parturition, the possi-
bility that this hormone has a vital role in the birth process should not be
overlooked. Because of the ,,~ide variation in the source of relaxin, there
was considerable controversy about its existence and function. Recent
work of Hisaw et al. (1944-1946) indicates that relaxin is a separate
hormone and is not a steroid as are the estrogens and progesterone. The
uterus appears to be concerned in the production of relaxin and seems to
be dependent upon progesterone in carrying out this function.
Significance of Parturition to the Livestock Breeder.-The period of
parturition is one of the most critical stages in the life of any animal.
All the advantages gained in selecting genetically desirable and healthy
parent stock and providing the very best of environmental and nutri-
tional conditions throughout pregnancy can be dissipated at parturition
through carelessness or ignorance. Death takes a large toll during
parturition. Some animals are too weak to survive the birth process,
others die because of mechanical injury or suffocation, and others are
lost due to the rigors of the environment. In still other instances the
young and dam alike may subsequently suffer from damage and infection
incurred during or following parturition. It is not the intent of the
writers to train -yeterinary obstetricians but to describe the normal process
of birth and the recognition of complicated or difficult parturition.
Place of Parturition.-Because of the ever-increasing complexity of
the delivery room in the modern hospital, we tend to consider the devel-
opment of more and more elaborate maternity stalls as desirable. There
is little expe~imental evidence to justify this view in practical livestock
production. If environmental conditions are favorable and there are
no known complications of delivery anticipated, by far the best place for
mares and CO'YS to drop their young is in a clean, nutritious pasture. In

order that there may be no doubt as to meaning, a crowded exercise lot
adjacent to the barn is not considered a clean pasture nor is a heavily
wooded lot with growthy underbrush considered desirable. In the West-
ern range states cattle and horses are expected to remain away from the
barns while giving birth, and it has long been recognized that the infection
of the young is most apt to occur when parturition takes place in or
around ranch buildings. The obvious disadvantages of this practice are
uncertainty of ~\Yeather conditions and the fact that difficult birth may
not be known until it is too late to render assistance.
During parturition ewes should be kept under fairly close surveillance
in a lambing pen, and sows should be shut up by themselves in a house or
pen. With proper thought and care the advantages gained, in all classes
of livestock, from giving birth in the open can be duplicated in the
maternity stall or pen with the distinct added advantage that assistance
can be given if it is required. Maternity stalls or pens should be planned
so as to give the highest degree of safety to the dam and her offspring
and the greatest degree of convenience to the caretaker. The stall or
pen should be roomy, and there should be no protruding constructions
that might lead to injuries. Mares sometimes lie dalm at parturition-
with their buttocks against a wall and some breeders obviate this possi-
bility by building false walls that can be quickly moved. Doors into
maternity stalls should be as wide as possible, in order to prevent injury
to pregnant females. Maternity barns or pens should be well ventilated,
easily cleanable, and located where the animals contained can see and
hear other animals with which they have been kept. Under no circum-
stances should parturition take place in a stanchion.
The heavy death losses which occur in swine at or following parturition
have focused the attention of swine producers and research workers on
this problem. Surveys of these losses on corn-belt farms have shown
that at least 20 to 30 per cent of all pigs which are farrowed are dead o~
die within the first few days of life. Some of these losses are due to
nutritional deficiencies during gestation, others may have genetic, physio-
logic, or pathologic causes, and still others may be due to injury or
chilling of the pigs. As a result of these losses more innovations have
been introduced to assist parturition in the_sow than in all other farm
animals combined. Most farrowing pens have guard rails about 8 in.
from the wall and 8 in. from the floor to prevent the sow from crushing
her pigs against the wall. Concrete floors, unless ,yell insulated or
heated, are usually best covered with planks or a plank platform provided
for the sow. One of the most effective methods of reducmg losses due to
chilling and crushing is to provide heated brooders or hovers in one
corner of the pen. This not only prevents chilling but provides a safe
place for the pigs to lie 'when they are not suckling. The heat lamps,
which are produced by several electrical manufacturers, provide a cheap
but satisfactory source of heat for such brooders. The observation that
sows usually lie with their backs uphill has led to the use of pens with
sloping floors. Swine producers report reduced losses ,vhen such arrange-
ments are provided. One of the newest developments is the use of the
farrowing crate. Based on the theory that a sow which cannot move
about freely or turn around would be unlikely to injure the newborn pigs,
<;everal types of crates which restrain the activity of the sow have been
~. roduced. .tI:merous swine producers in the Midwest used such con-
l ivances during 1949, and only time ,,,illprove their merit. That man
is ingenious is apparent; whether he will be able to improve on nature's
best conditions remains to be seen.
Sanitation.-All maternity stalls or pens should be built or arranged
so that maximum advantage can be taken of the most efficient germicide
and .also the most efficient growth promoter known, l'iz., sunshine.
Before putting any female into a maternit.y stall or pen, both the animal
and the pen should be thoroughly washed ,\'ith a disinfectant solution.
The coal-tar disinfectants, such as cresol, are good for this purpose, a 2
per cent solution to be used on the animal and a 5 per cent solution on
the stall. The young are quite likely to suck or lick various parts of the
mother's body; therefore, her whole body, more particularly the teats,
should be kept clean. The floor and ,yalls should be scrubbed thoroughly,
care being taken to work the disinfectant well into the cracks and corners.
Fresh, clean bedding should be used, preferably straw or hay rather than
salvdust or shavings, as either of the latter is quite likely to get into
the stomach or lungs of the newborn and cause inflammation. For sows,
the hay or straw must be short enough so that the pigs cannot burrow
under it, thus becoming hidden from view and running the risk of being
trampled to death.
If brucellosis, contagious abortion, is known to exist in the cattle or
swine herd, special care must be given all maternity stalls. The genital
discharges and fetal membranes of infected animals are one of the most
. potent soqrces of spread of the disease to other animals and to the human.
Any aborted material should be regarded with suspicion and the cause
of the abortion should be investigated by a veterinarian.
Time of Parturition.-A very necessary condition for successful par-
turition is that the breeder know when to expect it, and this predicates
the keeping of a record of the breeding of all females. The record book,
coupled with a knowledge of the normal gestation period and a sharp
lookout for the unexpected, will enable the breeder to have things in
TABLE 16.-GESTATION TABLE
Date bred Mare, Cow, Sow, Ewe,

336 days
I 281 days 114 days 147 days
Jan. 1 Dec. 3 Oct. 9 Apr. 25 May 28

Jan. 6 Dec. 8 Oct. 14 Apr. 30 June 2
Jan. 11 Dec. 13 Oct. 19 May 5 June 7
Jan. 26 Dec. 28 Nov. 3 May 20 June 22
Jan. 31 Jan. 2 Nov. 8 May 25 June 27
Feb. 5 Jan. 7 Nov. 13 May 30 July 2
Feb. 10 Jan. 12 Nov. 18 June 4 July 7
Feb. 15 Jan. 17 Nov. 23 June 9 July 12
Feb. 20 Jan. 22 Nov. 28 June 14 Jull' 17
Feb. 25 Jan. 27 Dec. 3 June 19 July 23
Mar. 2 Feb. 1 Dec. 8 June 24 July 27
Mar. 7 Feb. 6 Dec. 13 June 29 Aug. 1
Mar. 12 }'eb. n Dec. 18 July 4 Aug. \)
Mar. 17 Feb. 16 Dec. 23 July 9 Aug. 11
Mar. 22 Feb. 21 Dec. 28 july 14 Aug. 16
Mar. 27 Feb. 26 Jan. 2 July 19 Aug. 21
Apr. 1 Mar. 3 Jan. 7 July 24 Aug. 26
Apr. 6 Mar. S Jan. 12 July 29 Aug. 31
Apr. 11 Mar. 13 Jan. 17 Aug. 3 Sept. 5
Apr. lI) Mar. 18 Jan. 22 Aug. 8 Sept. 10
Apr. 21 Mar. 23 Jan. 27 Aug. I3 Sept. 15
Apr. 26 Mar. 28 , Feb. 1 Aug. 17 Sept. 20
May 1 Apr. 2 Feb. 6 Aug. 23 Sept. 25
May 6 Apr. 7 Feb. 11 Aug. 28 Sept. 30
May 11 Apr. 12 Feb. 16 Sept. 2 Oct. 5
May Ie. Apr. 17 Feb. 21 Sept. 7 Oct. 10
May 21 Apr. 22 Feb. 26 Sept. 12 Oct. 15
May 26 Apr. 27 Mar. 3 Sept. 17 Oct. ZO
May 31 May 2 Mar. 8 Sept. 22 Oct. 25
June 5 May 7 Mar. 13 Sept. 27 Oct. 30
June 10 May 12 Mar. 18 Oct. 2 Nov. 4
June 25 May 27 Apr. 2 Oct. 17 Nov. 19
June 30 June 1 Apr. 7 Oct. 22 Nov. 24
July 5 June I) Apr. 12 Oct. 27 Nov. 29
July 10 June 11 Apr. 17 Nov. 1 Dec. 4
July 15 June 16 Apr. 22 Nov. 0 Dec. \)
July 20 June 21 Apr. 27 Nov. 11 Dec. 14
July 25 June 20 May 2 Nov. 16 Dec. 19
July 30 July 1 May 7 Nov. 21 Dec. 24
Aug. 4 July 0 May 12 Nov. 20 Dec. 29
Aug, 9 July 11 May 17 No>'. 31 Jan. 3
Aug. 14 July 10 May 22 Dec. 0 Jan. 8
Aug. 19 July 21 May 27 Dec. II Jan. 13
Aug. 24 July 26 June 1 Dec. 16 Jan. 18
Aug. 29 July 31 June \) Dec. 21 Jan. 23
Sept. 3 Aug. 5 June 11 Dec. 26 Jan.· 28
Sept. 8 Aug. 10 June 16 Dec. 31 Feb. 2
Sept. 13 Aug. 15 June 21 Jan. 5 Feb. 7.
Sept. 18 Aug. 20 June 20 Jan. 10 Feb. 12
Sept. 23 Aug. 25 July 1 Jan. 15 F'eb. 17
Sept. 28 Aug. 30 July 6 Jan. 20 Feb. 22
Oct. 3 Sept. 4 July 11 Jan. 25 Feb. 27
Oct. 8 Sept. 9 July 10 Jan. 30 Mar. 4
Oct. 13 Sept. 14 July 21 Feb. 4 Mar. 9
Oct. 28 Sept. 29 Aug. 5 Feb. 19 Mar. 24·
Nov. 2 Oct. 4 Aug. 10 Feb. 24 Mar. 29
Nov. 7 Oct. 9 Aug. 15 Mar. 1 Apr. 3
Nov. 12 Oct. 14 Aug. 20 IMar. 6 Apr. 8
Nov. 27 Oct. 29 Sept. 4 Mar. 21 Apr. 2.3
Dec. 2 Nov. 3 Sept. \! Mar.2G Apr. 28
Dec. 7 Nov. 8 Sept. 14 Mar. 31 May 3
Dec. 12 Nov. 13 Sept. 19 Apr. 5 May 8
Dec, 17 Nov. 18 Sept. 24 Apr. 10 May 13
Dec. 22 Nov. 23 Sept. 29 Apr. 15 May 18
Dec. 27 Nov. 28 Oct. 4 Apr. 20 May 23
\.
readiness for the reception of the newborn. In some instances a knowl-
edge of breeding date and of gestation period is not enough to ensure
that everything shall be in readiness, but without this information the
case is well-nigh hopeless.
If a stall or pen is to be used, the females should be watched with
even more diligent care than usual as the time for parturition approaches.
They should not at this time be subjected to sudden changes of any kind
in feed, exercise, or care. Moderate exercise is necessary to keep the
muscles in good tone and to prepare them for the strain that lies ahead.
The feed should be of a somewhat laxative nature--this point cannot be
overstressed. The droppings must be watched closely, and if, even in
spite of proper selection of feeds, constipation develops, it must, for the
safety and well-being of both mother and offspring, be quickly remedied.
Feeds indicated at this time are pasturage, silage, roots, legume hays,
bran, linseed-oil meal, and ground oats. In case of emergency, linseed
TABLE 17.-EsTRUS-GESTATION RELATIONS
Estrual cycle Gestation, Ratio estrual

in days days cycle to gestation
Mare .......................... . 21 336 1:16

Cow ..................... . 20 280 1: 14
Ewe ................. . 16-17 148-153 1: 9
Sow. 19 112-114 1: 6
oil or epsom salts may be administered. An abundance of good drinking

water is also essential.
When bred, animals should be in fairly thin condition, but they should
be gaining and should continue to gain slowly all through the gestation
period, coming to the time of parturition in good condition regarding both
quality and amount of flesh. Extremes in thinness as well as fatness
should be sedulously avoided.
The female should be separated from the herd at least 1 to 3 weeks
before parturition, so that danger of violent contact may be avoided.
This will also enable the breeder to watch more closely and will provide
opportunity for the animal to become accustomed to the stall or pen
where parturition is to occur. Animals are naturally more or less nervous
at this time, and it is necessary that everything possible be done to allay
their fears, which conduces to ease of parturitjon as well as to the safety
of the newborn.
Preliminaries to Parturition.-Animals generally evidence a loosening
of the ligaments and muscles in the pelvic region just previous to parturi-
-
tion, this being more noticeable in mares and cows. The pelvic region
relaxes, the tail rises, the flanks and croup become hollow, and the
enlarged abdomen drops. At this time, animals spring in their middles
and start to make bag, with the udder becoming hard and tender, espe-
cially in primiparae. If the congestion in the udder becomes too great,
it should be relieved by milking before parturition. Most animals give
evidence of approaching parturition by their general nervousness and
fickle appetites. The mare whisks her tail, the cow bellows, the ewe
bleats, and the sow grunts plaintively. Movements of the fetuses are
often observable through the body wall or recognizable to touch. If not
restrained, the female at this time will seek seclusion from its fellows, and
the sow often starts to pick up straw, sticks, etc., with which to build a
nest. Individuals, of course. vary in these manifestations, sometimes
beginning to react a considerable time before parturition is due. How-
ever, these patterns of behavior and the breeder's record book of dates of
matings taken together do give ample and accurate warning.
Act of Parturition.-If everything is normal and the animal has been
properly fed and cared for, the act of parturition can usually be' accom-
plished without assistance. Under these conditions the best procedure
for the manager or herdsman, after checking up and seeing that all details
have been properly carried out, is to go on "'ith his work and let nature
take its course. Here again the happy medium between too much atten-
tion and too little should be the aim. Meddling is often dangerous alike
to parent, offspring, and owner. On the other hand, the day of parturi-
tion is rather a poor one to choose for a holiday. Mere human presence
at this time is often objected to by the animal, especially by mares, 'which
are particularly sensitive to human presence and will delay foaling as long
as possible if such persists. Other animals, though not so sensitive, also
prefer seclusion at this time. An occasional visit to note progress of
events is always desirable and often profitable, and at this time an invalu-
able asset is the animal's confidence in her caretaker.
Parturition may be divided into several phases. The first phase con-
sists of the dilation of the cervix and the loosening of the pelvic region.
The second phase consists of the gradually accelerating contractions of
the uterine and abdominal muscles causing the actual expulsion of the'
fetus. The third stage is a continuation of the second. the muscular
contractions beginning to abate but resulting in the expulsion of the after-
birth. A convalescent phase covers several days during which the uterus,
cervix, and pelvis gradually return to the normal nonpregnant state.
Labor pains set in from 1 to a fmy hours before final expUlsion of the
fetus. They are mild at first, lasting from a few seconds to 1 or 2 minutes
with intervals of 15 minutes between. Their tempo, duration, and
gth are gradually accelerated. The animal becomes increasingly
,s 'and nervous, shows evident signs of pain, and the skin becomes
"ld dry. The pains are caused by rhythmic contractions of the
longitudinal muscles in the wall of the uterus, which in this way begin the
dilation of the cervix to allow passage of the fetus. The dilation of the
os is soon followed by the contraction of all the uterine muscles, as well
as the abdominal muscles and the diaphragm, the contractions being fol-
lowed by a short period of calm for rest. The combined contractions of
the uterine and abdominal muscles soon force out the fetal membranes,
which act as an elastic wedge to open the passages gradually. After
protrusion, the water bag soon breaks, releasing part of the fluid; and the
FIG. 74.- No.rm al position of calf in utero . (From U.S. Department of Agriculture, DiBeases
of Cattle.)
young, which in the meantime has turned first onto its side and then all
the way over onto its stomach, soon makes its appearance.
The duration of parturition varies in different species of animals, being
in the mare normally from 5 to 30 minutes; in the cow, 1 to 2 hours; in
the ewe, 15 to 30 minutes for each lamb born; and in the sow, 10 to 30
minutes for each pig, with sometimes an interval up to an hour between
the pigs.
The last stage of parturition consists of the expulsion of the fetal mem-
branes, or afterbirth, and the man in charge should keep a careful watch
to see whether or not this occurs normally. This may follow soon after
birth, but, if it is delayed more than 24 hours, or 48 at the outside, steps
should be taken to ascertain and obviate the difficulty. When expelled.
the afterbirth should be removed at once and buried or burned. If this
is not done, many sows, cows, and even some mares will eat thls material.
If, as sometimes happens, the young are born in the intact membranes,
they should be liberated immediately in order to prevent asphyxiation.
The mare. cow, and ewe give birth either standing or lying, the sow
reclining.
Types of Presentation.-The two normal presentations of a fetus are
front feet first with the head resting between or on the knees, or hind feet
first, the first-named being the most usual and showing the least compli-
cations. The pelvic passage is generally large enough to permit fairly
easy passage of the forefeet and head. It is none too large, however, for
the largest portion of the fetus, viz., the thorax and shoulders. After
the head has passed the vulva, there is often a short pause before the
much greater effort now required. The dorsal vertebral spines over the
shoulders are compressed backward, the thoracic cavity contracted and
elongated, which diminishes its vertical diameter and permits passage
through the pelvis. The posterior portion of the fetus generally passes
through the pelvis without difficulty. ..
The normal hind-feet presentation, with the fetus lying on its belly and
its back uppermost, is not quite so favorable as the anterior presentation.
Birth usually takes longer with this presentation, and the pinching of the
umbilical cord may result in asphyxiation.
There may be various types of malpresentations, such as a buttocks
presentation, back presentation, all four feet presented, and many others,
some of which are shown in Fig. 75. Any malpresentation requires early
recognition. together with a lot of skill and patience on the part of an
experienced person. in order to prevent the death of the fetus and/or its
dam.
Assistance at Parturition.-If, after reasonable time and effort have
been expended, a female appears to be making no progress in parturition,
it is advisable that an examination be made so that proper assistanbe may
be rendered before the animal has completely exhausted her strength in
futile efforts at expulsion. For such an examination the external parts
of the vulva should be well cleaned, the hand and arm should be thor-
oughly cleansed and lubricated, and the fingernails cut short. After
proper precaution has been taken against injury, the hand, fingers held
in a cone shape, is inserted into the vagina. There may be various types
of malpresentations, such as only one front leg with the other turned
back. one hind leg and the other stretched forward, abdominal or sacral
presentation, head turned back on the side. head protruding with fore-
legs retained, all four feet presented, etc.
a
The cardinal features in the work of correcting the position of fetus
are cleanlint;\ss, quietness, gentleness, and perseverence, backed up, of
PREGNANCY liND PARTURITION 239
course, by knowledge, skill, and experience. If the presentation is normal.

strength for expulsion merely being lacking, the feet can be grasped and a
pull exerted downward and backward in conformity with the shape of
the pel vis and the position of the young. The pull should only be exerted
at the time the dam is in labor, never in the rest periods, except to main-
tain the gain that has been made. In correcting an abnormal presenta-
tion, it is often necessary to push the young well forward into the uterus,
and this should be done gently and during the rest period, not during the
labor period. A small clean rope can be used to slip around the feet of a
colt or calf, and a piece of No. 9 wire with one end bent into a hook to
catch under the lower jaw is often useful as a last resort with Iambs or pigs.
When parturition is unduly delayed or retarded, the fetus often dies.
This may be due to twists or knots in the umbilical cord or to too long a
stay in the passage. leading either to the stoppage of fetal circulation, or
the depriving of the fetus of a supply of oxygen, or both. The same
result is likely to follow premature breaking of the water bag with the
consequent uterine pressure on the fetus or the too early disunion of the
fetal membranes and the uterus of the mother. \Vhen labor has once
started, it should be expedited in every manner that is not inimical to the
safety of the fetus. The experienced caretaker can often cheat death
through his own efforts, but there is no other time at which a skilled
veterinarian can be of greater value to the progressive breeder.
Care of the Young at Parturition.-If birth is normal and the young are
normal and vigorous, interference is not advisable. A colt or calf s'lould
be on its feet and sucking within an hour after birth, but, if not, some
assistance may be given so that the young animal can begin to get
nourishment. The same applies also to lambs, where, in case of twins,
the shepherd generally removes the first lamb until parturition is com-
plete, when both Iambs are allowed to nurse. In the case of sows, it is
advisable for the herdsman to be uresent at parturition in order to remove
the pigs when born, dry them, and put them in a warm safe place until
parturition is complete, when all are allowed to nurse.
Losses from various forms of sterility are severe enough without adding
to them by negligence at the time of parturition. Sometimes it is neces-
sary to remove the young from the membranes so that they can start
breathing; whereas, on other occasions, it is necessary for the same reason
to remove mucus from the mouth and nostrils. Many animals apparently
stillborn can be induced to start breathing by slapping the sides, working
the front legs and chest, as in restoring a man who has been submerged
in water, blowing into the nostrils, tickling nostrils with a feather, etc.
1£ an animal is stillborn, vigorous action on the part of the attendant will
PREGNAl{CY A.ND PA.RTURITION 241
often cheat death, as will protection from the elements for those which
have safely arrived.
Cleanliness of the stan, the dam, the owners' hands, and any instru-
ments which might be used are essential. As previously mentioned, the
mother should be thoroughly cleaned before parturition. The mammary
glands should be clean. Many livestock men dip the umbilical cord
in iodine soon after birth to assist in preventing navel infection, or
scours, but the most effecti ve means, as stressed, are cleanliness of the
surroundings.
Many swine producers clip or remove the sharp" wolf" teeth of new-
born pigs. Although these teeth appear to serve no useful purpose, their
removal sometimes results in injury or infection of the jaws and they
should not be removed unless they injure the teats of the sow.
Care of the Mother at Parturition.-The general care and feeding of
the parturient animal wiII of necessity vary with the species and with the
local conditions. It is obvious that the requirements of a cow calving on
pasture in June are quite different than a sow farrowing in January.
Without attempting a treatise on management and nutrition, a few basic
principles will be summarize. After safely delivering her young, the
things a female needs most ar~ '\st and quiet, together ""ith a moderate
, amount of temperate water. .8, should not be fed for 12 to 24 hours
. after parturition and should be started on feed very gradually. Gruels
of bran and ground oats with perhaps a little linseed-oil meal are good at
this time, together with a small amount of good-quality leguminous hay
and some pasturage. The feed must be moderate in amount for two
reasons: (1) A return to the normal condition of the dam's system will be
hindered by overfeeding, and (2) the presence of too much milk in the
udder as a result of overfeeding is sure to cause scouring in the young,
which is always a serious drawback to normal health and growth. The
wise herdsman never feeds a nursing female until he has observed the
character of the droppings of the young, and at the first sign of scouring,
the dam's feed should be reduced.
One of the most common disturbances of high-producing cattle is milk
fever. This condition is actually not accompanied by fever but is recog-
nized at first by general depression of the cow, followed by nervous
excitation. Twitching of the muscles occurs and is followed by collapse
and eventual loss of consciousness. These symptoms are caused by a
subnormal amount of calcium in the blood which is believed due to the
sudden drain of milk secretion on the stores of body calcium. It can be
successfully treated by the intravenous administration of calcium glu-
conate. For this reason it is unwise to completely milk cows following
calving. It is necessary, however, to see that some milk is drawn from

all quarters in order that excessive pressure does not injure the udder.
There is still controversy as to whether cows should be milked prior to
calving. If there is evidence of extreme mammary congestion, the relief
of the pressure by partial milking should be considered. In general" pre-
partum milking is to be discouraged because of the great value of colos-
trum to the calf.
In general about 2 weeks should be all owed to get the dam back on
full feed. Mares can go back to light work in 2 weeks. Care must be
exercised that the foal left in the stall cannot harm itself, because it will
be restless without its dam. When mares are at work, the foal should be
allowed to nurse at midmorning and midafternoon after the mare has been
cooled off, if warm, and after her udder has been ,vashed.
If genital discharges persist following parturition, particularly if these
are bloody or foul smelling, the possibility of infection should not be over-
looked, With valuable breeding animals, routine genital examinations
by a veterinarian specialized in reproductive. problems can be considered
a wise investment. As the proverbs" stitch in time" and" ounce of pre-
vention" have proved true, so too can the prevention or early correction
of genital disorders be expected to increase breeding efficiency.
Feeding and Care of the Newborn.-As knowledge of the nutritional
requirements of man and animals is gained, it becomes increasingly dif-
ficult to know when the effects of nutrition first make themselves felt on
embryonic and fetal development. It seems likely that the pregestational
care and feeding of the female can affect the young. It is now established
that the anterior pituitary gland of the newborn calf can be permanently
affected by vitamin-A deficiency of the pregnant dam. The damage
cannot be repaired by subsequent administration of vitamin A. The
possibility exists, therefore, that malfunctions which might manifest
themselves in a pregnant or lactating animal might have been caused
during the fetal development of the animal itself.
The importance of the proper nutrition of all animals which are to be
used for breeding purposes cannot be overstressed. K ot. all the nutri- .
tional requirements are yet known, but it goes without saying that the
animals should be fed according to the recognized standards of the day.
Assuming that the young are normal and vigorous at birth, the breeder
can do much to keep them that way. For a time following parturition
newborn mammals are dependent almost entirely upon milk. The first
milk produced following parturition is quite different than ordinary milk.
In cattle colostrum is about five times as high in protein and twice as h~b.
in fat and minerals as ordinary milk. It is especially high in globulin and
contains antibodies which tend to protect the young from various micf{)-
r·.
, t ;
organisms. Cattle colostrum contains ten times as much vitamin A and

three times as much vitamin D as ordinary milk and over ten times as
much iron. Although some species, such as the guinea pig, are apparently
benefited little by colostrum, it is regarded as highly desirable that the
farm animals receive it. If for any reason colostrum is not available, as
in the case of animals milked prepartum or in the case of death of the dam,
the young should be fed colostrum from another source. Many of the
larger dairy farms preserve extra colostrum by quick freezing in order
that it ·will be available for other animals. When colostrum is not avail-
able, the young animal is frequently injected with whole blood from the
dam. The purpose of this is to attempt to increase the antibody reserve
of the newborn.
It has long been recognized that in general the young of one species thrive
better on milk of their own kind than on that of another species. Furthermore,
the sooner after birth the milk of another species is substituted for milk of the
same species, the more unfavorable is the outlook for the young animal. The
results of some studies published by Bilek are cited as indicating that some of the
unfavorable effects so often obtained from feeding milk of another species may
well be attributed to qualitative rather than quantitative differences. In these
experiments, cow's milk and goat's milk were fed to the young of several species.
The different species did not react alike. Kids used both milks equally well,
while calves reacted to goat's milk with severe gastrointestinal disturbances
which lasted as long as goat's milk feeding continued. Pigs showed no disturb-
ance on cow's milk but reacted to goat's milk in a fashion similar to that observed
in the calves. Foals showed a stubborn diarrhea when fed undiluted cow's
milk but not the slightest disturbance on goat's milk. Since the two milks are
very similar in percentage composition, the results indicate the presence in the
milk of substances incompatible or toxic to the animals that reacted unfavorably.
It is suggested that the incompatible factors are proteins.
It is known that the presence of a foreign protein undigested in the circulating
blood is more or less toxic to an animal and may produce a sensitization to further
additions of the foreign protein. It has been established that the newborn
animal normally. absorbs considerable protein undigested through the gastro-
intestinal wall. Although this apparent permeability decreased rapidly after
birth, it probably does not cease abruptly, since as Ratner has shown, it still
occurs to some slight extent in about 50 per cent of adults. The possibility
of the absorption in the very young suckling of appreciable amounts of undigested
and sensitizing foreign protein might be a major factor in the incompatibility of
the milk of one species for the young of another. 1
The young of all species will be better off if they are allowed to suckle
folat least 3 or 4 days. Mter this, they can be hand-fed if desired,
although this is seldom practiced except with dairy calves. Many sys-
1 EARLE, I. P., Food and Life, U.S. Dept. Agr. Yearbook, 1939, p. 508.
tems of feeding dairy calves have been devised and are available III
various books and bulletins.
Golts, calves, lambs, kids, and pigs will begin to nibble at the dam's
feed in from 2 to 4 weeks after birth. They should then be supplied with
suitable fine roughage and with suitable grain mixtures, preferably in a
creep to which the dam does not have access. For colts, a mixture of 2
parts cracked corn, 4 parts crushed oats, 2 parts bran, and 1 part linseed-
oil meal by weight has been found to give good results. For calves,
many starters and meals have been developed and are available under
trade names. They can also be home-mixed from such ingredients as
corn meal, ground oats, bran, middlings, and linseed-oil meal. Similar
starters or home-mixed rations will also serve for lambs and kids. Pigs
may also be fed home-mixed rations of corn, middlings, and protein
supplements such as fish meal or tankage. Pigs should also be supplied
a mineral mixture such as bone meal, 38 parts; cale1um carbonate, 39;
s;)dium chloride, 20; ferrous sulphate, 2.5; copper sulphate, 0.25; and
potassium iodide, 0.03 parts by weight. In order to prevent anemia in
pigs, many breeders paint the sow's udder daily with a solution of iron
and copper or place fresh clean sod or dirt where the little pigs can get it.
For maximum growth and well-being the young must be comfortably
housed in sanitary, safe quarters and must be properly fed both by
their mothers and through their own efforts. To keep young animals on
feed and gaining steadily from the day of birth is an art, gained only
through experience. For maximum eventual returns, proper care and
nutrition are indispensable. Any young animals do best if fed regularly,
often, and moderately. They should clean up their feed with a relish
and look for more, the secret of success being to keep them a little hungry.
Misguided kindness in the form of too much feed must be avoided.
If the young are ever to achieve the maximum of their inherent capa-
bilities, they must get started early along the proper path and must be
kept moving. Normal healthy parents are the first prerequisite, clean
maternity stalls the second, proper feeding and management of the
Lactating female the third. The young themselves ·will soon (1 to 3
weeks) evidence a desire to eat something besides milk. This desire
should be satisfied not ,vith shavings and sawdust but with good pastur-
age, fine-quality hay, and a suitable grain mixture. Reme~ber that
whole milk, and more especially skim milk, is very high in protein.
Along with a suitable quantity and quality of milk from the dam and
such other feeds as just mentioned should go a sufficient amount of
exercise, preferably at pasture and in the sunshine.
Constructive breeding implies that records are to be kept. With the
young animal this starts at birth. Among other things, the record should
.
\\,
show the animal's sire and dam, the length of gestation, the character of
parturition (normal or otherwise), single or multiple births, ,veight of the
offspring, their general condition, number born weak or dead, and any
other pertinent data which will be of service when the breeder is deciding
whether or not to breed the dam again or whether or not to save her
offspring for breeding purposes. Parturition is a busy time for the
herdsman, so that if he does not get his records then and there, they are
likely never to be secured.
If the animals are to be marked in any way, it should be done early
and before there is any chance of making mistakes. Pigs are usually
marked by notching the ears; e.g., a notch in the outer edge of right ear
stands for 1, outer edge of left ear for 3, inner edge of right ear for 10,
and inner edge of left ear for 30. Lambs and calves are best marked by
metal tags, or calves by tattooing. If the tattoo fades, it can be made
permanent in the other ear at about two years of age.
Complications before Parturition.-Many complications can arise
during gestation that 'will interfere with normal parturition. Extra-
uterine pregnancies have been known to occur in the ovaries, abdominal
cavity, Fallopian tubes, the cervix, or the vagina. In human medicine
these are usually recognized and are then removed by surgery. In farm
animals they are not readily detected in their early stages and may
terminate fatally or eventually become aborted or mummified.
Superfetation, the establishment of a second pregnancy in an animal
already pregnant, sometimes occurs. This is not so common as is
thought. The fact that twins or Jitter mates may be of unequal size
does not indicate superfetation but rather that differences in physical
development occur. Other instances of true superfetation may not be
recognized because the second fetus conceived may be expelled at the
time of parturition of the first fetus without arousing suspicion. How-
ever, there are authentic cases in which animals have given birth to
normal young conceived and born at distinctly different times and not at
two normally spaced periods.
Numerous developmental abnormalities have been observed. These
are of many types' and probably have many causes. Many of these
abnormalities are confined to certain organs, e.g., cleft palate, imperforate
anus, etc., and do not interfere with parturition. In other cases fetal
monsters (teratoma) in the form of unorganized tissue containing skin,
hair, teeth, nervous tissue may develop.
In the human there are numerous diseases and functional disturbances
which have either a direct or indirect effect on the course of pregnancy.
,The farm animals are not subject to many of these. The common
genital diseases have been discussed in another section and will be only
24G BREEDING AND IMPROVEMENT OF FARiv[ ANIMALS
mentioned here. Brucellosis may result in abortion in cattle, swine, and

goats. In cows, a microorganism, Trichomonasfetus, may cause abortion.
Abortions due to Salmonella and virus infections have been reported in
mares, and in sheep and cattle Vibrio fetus has been associated with
abortion.
Gross nutritional deficiencies as a cause of gestational difficulties are
not common in the major livestock-producing areas of the United States.
Of the specific nutritional factors which affect embryonic development,
vitamin-A and iodine deficiency are the most common. In the female
gross vitamin-A deficiency may be associated with abnormal embryonic
development and abortion. Clinical iodine deficiency is more likely to
occur in ewes and sows than in cows or mares. It manifests itself by
the birth of ·weak or dead young with enlarged thyroid glands and con-
siderable edema (watery infiltration) of the skin. Hairlessness 'or wool-
lessness may occur.
One of the most spectacular complications of gestation is pregnancy
disease of sheep. This is not a transmissible disease but a metabolic
disturbance. Affected ewes first appear listless, refuse feed, and become
progressively weaker. Within 1 or 2 days they become unable to stand
and for this reason the term parturient paralysis is used. The condition
usually appears in the last month of pregnancy, and ewes carrying twins
or triplets are most apt to be affected. About 90 per cent of affected
animals die. The condition is thought to be due to poor feeding and lack
of exercise.
Accidents caused by slipping, crowding through narrow doors or gates,
fighting, or general mismanagement may cause death or premature expul-
sion of the fetus.
Complications at Parturition.-Parturition may be complicated by
anatomical abnormalities of the dam or fetus, malpresentation, muscular
exhaustion, or extreme excitation of the dam. The proper physical con-
dition of the female as parturition approaches and her confidence in the
herdsman will do much to facilitate the process.
Difficult labor is called dystocia and its correction calls for experience ~
and skill. The interference of an inexperienced person in a truly abnor-
mal delivery is very apt to result in death of the fetus and serious injury
or death of the dam. Perforation of the uterine wall by the feet is
almost sure to be followed by peritonitis. Pulling when pushing is called
for is apt to tear the genital organs or cause prolapse of the uterus or
rectum. The skilled obstetrician seldom resorts to strength alone. If
the problem is one of malpresentation, it is frequently most successfully
met by first administering a local or general anesthetic. If a Caesarean
operation is to be performed, it should be done before the animal is
..
exhausted or injured. In those rare instances when the fetus must be

dismembered in utero, only the most experienced should be allowed to
perform the task.
The breeder of valuable animals is concerned not only with the loss of
one offspring, or litter, but with the lifetime breeding performance of the
dam. Parturient injury followed by sterility is an expensive price to
pay for carelessness or ignorance. We urge, therefore that no time be
lost in securing a skilled veterinary obstetrician when conditions appear
to be abnormal.
Complications Following Parturition.-The strain involved in parturi-
tion plus the avenues of infection that are automatically opened by it
render the dam particularly susceptible to complications following par-
turition. They may take many forms, from retained placenta, hemor-
rhage, eversion of the vagina, uterus, or bladder, to lesions in various
parts of the genital tract or the invasion of the tract by various organisms,
causing inflammation of the uterus, cervix, or vagina, often accompanied
by the formation of pus. In addition, milk fever, septicemia, paralysis,
convulsions, and many other abnormal or diseased conditions may follow
parturition. Surely the lot of the dam at this time is not an enviable
one. The least the livestock breeder can do is to use all the precautionary
measures regarding feed, care, and sanitation he can lay hold of and not
begrudge or delay too long in getting the best skilled attention he can
secure whenever it becomes necessary to have it.
Summary.-In this chapter we have discussed the role of hormones in
the establishment and maintenance of pregnancy, and the causes and
processes of parturition. Understanding something of the complicated
process should make us, as livestock breeders, a bit more appreciative of
the dam's situation during this critical time. With intelligent care and
forethought, most of our animals can be brought through parturition
unscathed and their offspring gotten off to a good start on the road to
their eventual service in production or reproduction. We have seen that
the process of parturition is often beset with a multitude of complications.
If we are going tb succeed in livestock breeding, we must learn to handle
our animals intelligently at this time-the time and process on which the
whole structure of animal breeding ultimately rests, for it has to do with
the safe induction into the world of the animals that we hope our genetic
planning and selection have given a good chance of being superior to
their forebears.
References
Books
. BENESCH, F., and WRIGHT, J. G.. 1938. "Veterinary Obstetrics," William Wood &
Company, Baltimore.
CRAIG, J. F. 1930. "Fleming's Veterinary Obstetrics," 4th ed., Alex Eger, Chicago.
'WILLIAMS, W. L. 1940. "Veterinary Obstetrics," W. L. "'illiams, Ithaca.

ABRAMOWITZ, A. A., et al. 1944. Preparation, Biological Assay and Properties of
Relaxin, Endocrinology, 34:103-114.
ALBERT, A., and MONEY, W. L. 1946. Methods for Concentration of Relaxin from
Blood Serum and Urine, Endocrinology, 38 :56-57.
HISAW, F. L. 1926. Experimental Relaxation of the Pubic Ligament of the Guinea
Pig, Soc. Expt. Bioi. and M ed. Proc., 23 :661-663.
- - - et al. 1944. Importance of the Female Reproductive Tract in the Formation
of Relaxin, Endocrinology, 34; 122-134.
CHAPTER IX
MAMMARY DEVELOPMENT AND THE INITIATION OF

LACTATION
The mammary glands are, to the zoologist, the distinguishing charac-

teristic of all mammals; to the physiologist, an i~l part of the female
reproductive system; and to the dairyman, practically a way of life.
There are wide differences between mammals in the anatomy and physi-
ology of the mammary system. These range from the egg-laying mam-
mals, such as the duckbill platypus in which the young scoop up the milk
as it exudes from nippleless mammary areas, to the highly developed cow
producing in excess of 30,000 lb. of milk annually without ever seeing her
calf or its sire (if artificial insemination is practiced).
Embryological and Fetal Develo:r;ment.-The mammary glands are
generally regarded as modified sweat glands. They originate as integu-
mentary ingrowths in both sexes, but as fetal development progresses,
the rate of mammary differentiation is more rapid in the female. In
cattle the first evidence of mammary development is the appearance of a
band 'of cells on the ventral surface of the embryo posterior to the umbili-
cus. The single band of cells soon divides into two parallel bands, and on
each band two budlike areas of cells appear. Each mammary bud under-
goes'rapid growth and soon differentiates into a teat. In cattle there are
ordinarily 4 mammary glands and 4 functional teats. In other species
the number of teats and glands characteristic of the animal are pro-
liferated along the mammary bands. The outstanding features of fetal
development are the development of a canal in the teat (the teat cistern),
a larger cavity at the base of the teat (the gland cistern), and a few ducts
leading away from the gland cistern. These structures will later give
rise to the duct system of the udder.
Prepubertal Development.-The functional mammary gland will con-
tain connective tissue, fat, a duct system, a lobule-alveolar system, and
large numbers of blood and lymph vessels, nerves, and smooth muscle
cells. The period from birth to puberty is chiefly a period of increased
connective tissue and fat deposition. The udder does increase somewhat
in size, and there is a suggestion as to its future shape, but the amount of
duct growth is very limited. In dairy cattle Swett and Matthews (1935
and 1942) reported that there is greater duct growth in calves which later
249
prove to be high producers than that in low producers. However, the

limitations of a gross examination for the estimation of future micro-
scopic changes make it difficult to estimate the probable eventual output
of the mammary glands.
Postpubertal Development.-There are marked species differences in
the extent of mammary development of the sexually mature but non-
pregnant female. Such factors as length of the estrual cycle, function of
the corpus luteum, and seasonal breeding patterns all contribute to the
differences. In the mouse, which has a cycle of 4 to 6 days, there is
marked duct growth until about 70 days of age but little further duct
growth from 70 to 100 days unless pregnancy is established (Gardner and
Strong, 1935). In the cow, which has an estrual cycle of 18 to 21 days,
development of the duct system begins at puberty and, continues more or
less rhythmically prior to conception (C. W. Turner, 1939). There is
more complete mammary development in the human prior to pregnancy
than in other mammals. Duct growth is very extensive, and there is
some evidence that alveolar proliferation may occur (C. D. Turner, 1948).
Mammary Changes during Pregnancy.-In most mammals the first
half of pregnancy is devoted to the completion of the mammary duct sys-
tem and the proliferation, at the extremities of the ducts, of the lobule-
alveolar system. The chief purpose of the ducts is to transport the milk.
The true secretory function of the udder is carried on by the secretory
. epithelium of the alveoli. Thus, anatomically at least. the mammary
system is prepared for lactation by the mid-point of pregnancy.
During the last half of gestation there is an enlargement of the alveolar
epithelium, and the lumina and ducts become distended with secretion.
As parturition approaches, the degree of distension becomes greater and
the udder may be extremely congested. The fact that the udder does
not increase in size noticeably until the latter part of pregnancy led to the
belief that maximum mammary growth occurs at this time. It is now
clear that cellular proliferation characterizes the first part of gestation
and cellular hypertrophy and secretion the latter part.
As previously discussed, abortion or proper mechanical stimulation is
usually followed by milk secretion after the middle of pregnancy. The
failure of animals to reach high levels of production in such cases is prob-
ably attributable to insufficient hormonal stimulation for secretion m.th~r
than incomplete anatomical development of the gland.
Mammary Involution.-Although the udder is regarded as a highly
efficient organ, it does not function at maximum capacity for long. In
fact, it requires from 2 to 4 weeks for a cow to reach maximum productiv- "
ity, and no sooner is the peak reached than decline begins. This would
be explained in nature on the basis of the needs of the calf. But by selec-
MAMMARY DEVELOPMENT AND LACTATION 251
tion, the cow (or man) has achieved a level of secretion far in excess of the
calf's requirements. The decline in secretion is the result of wear and
tear on the mammary system (involution) and changes in hormonal and
nutritional relationships. In fact, anything which affects the well-being
of the cow will be reflected in mammary performance. As lactation
progresses, the numbers of actively secreting epithelial cells and their
degree of activity decline. The microscopic appearance of the udder of
the dry, nonpregnant cow is very similar to that of the virgin. The
reestablishment of pregnancy is necessary to stimulate both mammary
grov,th and secretion in such an individual.
Endocrine Regulation of Mammary Development.-There is rather
general agreement as to the cellular changes which occur as the mammary
glands develop. It is true that there are wide species differences in the
rate and type of development, but competent anatomists agree as to the
structures which are present at particular times in the various species.
However, there are few body systems whose functions have been explained
by as many different theories as the mammary glands.
During the nineteenth century, when the nervous system was recog-
nized as the chief coordinating agency of the body', it was believed that
both the development and function of the mammary glands were under
~ervous control aud that :there we!:~Ile:r:':9:':l_s J>_:1tE~:::_ay~_bet'~~_Il the u_j;erus
and the mammary glands. This was disproved by such procedures as
~;~erance of the 'cenEi-ai or peripheral nerves and finally by the graft-
ing of mammary tissue to other parts of the body. The development and
function of such grafts clearly showed that the nervous system was not a
primary regulator of the mammae.
It had long been known that sexual maturity was followed by some
mammary growth in most species and that final development and milk
secretion were dependent upon pregnancy and parturition. The djs-
covery that the ovary was ,both an endo_cIjne an9-_gametggenic..Qrgan SQQIl
~to the conclusion that the mammary glands ,yere .lln_<le_r endocrine
.£Q!!.t_~. IhfHx;act nature of the endocrine mechanismsinvolved is not
~agreed upon.
As cited by Trentin and Turner (1948), Allen et al" Laqueur et al., and
Turner and Frank showed that the estrogenic hormone prepared from
follic~ia~ -fluid would cause cl~~pment of the mammary duct ~~'yste~.
M_",~s then found that corpus luteum extracts in combination wj,tn_estr,!?-
gens would induce alveolar growth. As a result of these studies, it was
concluded that estrogen was resp~nsible for mammary duct growth and
that progesterone, preceded by estrogen, resulted in the proliferation of
the alveolar system. It was recognized, however, that complete mam-
mary development could be induced in the guinea pig with estrogen alone.
Simultaneously with the studies being carried out on the physiological

effects and chemical nature of the ovarian hormones, it was reported that
anterior pituitary .extracts or preparations would induce mammary
growth in castrate and hyp-o h sectomized ratll (Asdell, 1931; Nelson
and Pfiffner, 1931, and others). In 1937, Gomez, Turner, and Reece
found that the anterior pituitary glands of estrogen-injected rats would
stimulate mammary growth in hypophysectomized guinea pigs, lending
support to the theory that the anterior pituitary secretes specific mam-
liw. 76.-Mammary stimulation of a virgin gilt after 3 weeks treatment with stilbestrol
resembles that of 1nid-pregnancy.
mary-growth-stimulating hormones. A new theory was developed in

which duct growth was accounted for by the secretion of the mammogenic
duct growth factor, and lobule-alveolar development was attributed to a
second anterior pituitary mammogenic hormone (Gomez and Turner,
1937, 1938).
Agreement has not yet been reached as to whether estrogen and
progesterone act directly upon the mammary gland, causing (1) duct
growth, and (2) lobule-alveolar differentiation, or whether these hormones
stimulate the anterior pituitary gland' which in turn causes mammary
/ development.
In the normal guinea pig and goat, and to some extent in the human
and monkey, estrogen alone induces duoi and lobule-alveolar growth
MAMMARY DEVELOPMENT AND LACTATIOX 253
(C. D. Turner, 1948). In cattle the natural estrogens illoduce only duct
~h, but stilbestrqL ,3. synthetic est:J;:ogen which ha~Qt ,b...e~!0.g~
nature, induces comp~t~ ml:l,ffiill&ry g.ll.velQ12!!le!!ta~d sQIlleti~~s_cop~R
~tio.n (Espe, 1946).
The present status of the mammogen theory was summarized by
Trentin and Turner in 1948. It was concluded that the active pituitary
factor is associated with the protein fraction rather than with the lipid
soluble fraction as previously reported. It ,vas further concluded that
the present evidence does not warrant the conclusion that there are two
separate mammogens, but that they should be considered as the same
pituitary factor. In an extensive series of studies of mammary develop-
ment of the male mouse, these authors found that continted estrogen
injection caused marked duct growth and some alveolar development in
normal male mice. Combined estrogen and progesterone_:t.re.at.l:t:wut_:u:as
much more effective than either hormone alone at the same dos~e.
Ho~~'~~~~ the pituitary gl~s of male mice were remov~-;_C e~tr~gen
produced little mammary growth and a combination of estrogen and
progesterone resulted in slight duct growth (Trentin and Turner).
Thus, in 1949, there are at least two distinct theories for mammary
development. One of the most serious criticisms of the mammogen
theory is the fact that percutaneously applied estrogens have been shown
to have limited local effects \vhen applied to single mammary glands in
mice, rabbits, monkeys, and other animals. In other words, the applica-
tion of estrogen to one gland does not affect neighboring glands. __lLj;he
effects of estr~n were thr~:llgh the pituitary, all glands would be expected
to respond (C. D. Turner, 1948). .
A wide variety of substances has l;leen shO\vn to either directly or
indirectly affect mammary grO\yth. Extensive stimulation has been
'obtained in normal and castrate male ~ale rats following andro~n
administration. In the human male enlargement and ten~rness of the
mamm~ are not uncommon at puberty. Extensive mammary develop-
ment and lactation can be induced in males by appropriate hormone
treatment, and it is not uncommon for considerable milk secretion to
occur spontaneously in male goats. Abnormal mammlk.;5' growth, includ-
ing cancer, is one of the serious problems in human medjcine. The exact
cause remains as yet obscure, but several chemical substances which will
induce experimental carcinogenesis are known. _.
. THE INITIATION AND MAINTENANCE OF LACTATION

Ahe Lactogenic Hormone.-The hormone responsible for the actual
initiation of milk secretion following the anatomical development of the
mammary gland ',vas discovered by Stricker and Grueter in 1928.
. .
These French physiologists found that anterior pituitary extracts would

induce milk secretion in pseudopregnant rabbits and later showed that
such extracts would bring about milk secre~ dogB..t. cows, and swine
if the mammae were properly developed. The specific anterior pituitary
hormone responsible for the initiation of lactation has been called galactin..
l!rolqfiin, ~n_<!. the lactogenic hormone. The last two names are in wide
general use.
The lactogenic hormone has been isolated from the anterior pituitary
gland in rather pure form. It is a protein with a molecular weight
between 22,000 and 32,000 and because of this fact is not likely to be
synthesized in the near future.
This hormone has a rather wide variety of effects in different species.
In pigeons it stimulates the secretion of crop milk, depresses ovarian and
testicular activity, will induce broody behavior, increases metabolic rate,
and even stimulates growth of the intestinal tract. In mammals the
lactogellic h()_tII!-0ne is primarily concerneg ,vtth the initiation and main-
.1ep.ance of milk secretion, but it may likewise play some part in parental
_instinct. .-
It is well known that the lactogenic hormone exerts its greatest effect'
shortly after parturition and that the ~~n_ secretion of the~iQr
pituitary declines as lactation progresses. It has been shown that dairy-
cattfepituit~riescontain more· of t;h~hormone than those of beef cattle
and that pregnant cows have more than do calves (Reece and Turner,
1937). The factors which are involved in the actual initiation of lacta-
tion and its maintenance at optimum levels have recently been reinvesti- .
gated by Meites and Turner (1948).
Determinations of the lactogenic hormone content of the anterior
pituitary glands of various species have shown that it is present in only
small amounts during pregnancy, increases slightly shortly before par-
turition, and is secreted in large amounts immediately following parturi-
~ It~-shown: that the inJection of e~trog'en Idollmyg_d by_a
m~rked increase 'In pituitary lactogen but that when progesterone is
-ru:iministered simultaneously the pituitary response is inhibited.. Aipre-
viously described, b~th estrogen and progesterone are secreted during
pregnancy. It is believed that progesterone is dominant over the
estrogen during most of pregnancy but that at the approximate time of
parturition estrogen becomes dominant, the lactogenic hormone is
secreted by the anterior pituitary in increased amounts, and lactation is
initiated (Meites and Turner, 1948).
The maintenance of a high rate of milk secretion following parturition
is of great concern to dairymen, and, for that matter, to the producers of
all farm mammals. The most outstanding dairy cows are persistent milk
producers. The rate of decline of lactation is slow by comparison with

the average or inferior cow. It appears logical to assume that the rate of
decline is in some way related to the level of secretion of the lactogenic
hormone. The correctness of this assumption has been borne out by
experimental evidence, but the exact nature of the controlling mechanisms
is not wholly understood.
It is well known that the removal of milk from the mammary glands is
quickly followed by the secretion of more milk. If milk is not with-
drawn, involution of the udder~. ~been shown that the suck-
1i_ng stif!l~J~l§jI?:_c.:r:ease~_the secretion and discharge of the lactogenic hor-
mone l2.Y- t_he pituitary and assis~s in the maintenance of the mammary
gI;,~d}n a fun2tiQn~( conditio!). (lVIeites and Turner, 1943).
- Other Hormones.-Milk secretion is a complex function requiring the
cooperation and coordination of all the body organs and systems. High
milk production cannot be attributed to anyone factor, such as mammary
size, level of lactogen secretion, feed capacity, nutritional status, or a
single gene.
Of the hormones other than estrogen, progest{).!:~!lez It!!..d_!~{)J!l:c0genW
hormone, those of the thyroid and adren~l glands .are probably next in
importance.
The thyroid gland, through the mediation of its hormone thyroxine, is
chiefly concerned 'with the regulation of general body metabolism. The
removal of the thyrQid i11_ the lactating ani.maU~followed by a _-r:apid dr_2p
in milk_.,productiQIl,. In contrast, the administration of thyroxine du""i-ing
the declining phase of lactation is frequently follo'ived by increased milk
production and an increase in the fat content of the milk. The discovery
of a relatively simple method of producing thyroxine by the iodination of
casein has led to extensive studies regarding the possibility of increasing
milk production by the feeding of such thyroidally active compounds
(Reineke and Turner, 1942). Iodinated casein is usually referred to as a
..!hyroprotein and has been shown t~ contain about 3 per cent thyroxine.
The place of thyroprotein in practical milk production is not yet clear.
Experimentally, it has been shown that the addition of thyroprotein to
the rations of dairy cattle during the declining phase of lactation may
increase milk £roduction as much as20 per cep.t and butterfat £roduction
as much as ~5 per cent for_shortpeJ:'iods of time. _ Many factOrs m~y affect
the ~~~sults. Some cows respond more than others, and some animals
may not respond at all or even adversely. It must be recognized that
thyroxine has a marked effect on the processes of metabolism and the
circulatory and respiratory systems. Excessive sti,mulation is usually
followed by considerable weight loss and eventually by decreased milk
production. In cattle in which suboptimum thyroid activity is limiting
milk secretion, the judicious use of thyroprotein would seem to offer

possibilities for increased production.
The adrenal cortex (outer portion of the adrenal gland) is ordinarily
regarded as essential for the maintenance of life itself. It is not sur-
prising, theref_9_r(lL~_llll.t_ th~_rgrp._Qy_ai-O[_ the _a_dr.enala.in a._lac.tating_animal
IS foIIo\vea--hY__theu cessati..Qn pf__la.ctation. The hormone, or hormones,
~ed by the cortical cells are conc~rned with normal carbohydrate
metabolism, the distribution of electrolytes, and normal kidney function.
The interference with any of these functions could be expected to influence
lactation.
l'he 12arathyroid glands are concerned with calcium and -e..h~us
metabolism and skeletal de\T_el()I>!!l~!!i__!!...I&Jh_!!s indirectly with ~@l<1al
muscle irritability and milk secretion.. One of the most dramatic dis-
orders of farm animals is !!illkle'yerjn__l_h.~ .c_my. This condition is, in
reality, not a fever, but !!ypocalc~~ia. It occurs almost exclusively in
high-producing dairy cattle shortly after calving and is apparently the
result of the greatly increased requirements of the lactating animal for
calcium. It is a very serious metabolic disturbance and may result in
death unless prompt treatment is given. The original treatment for the
condition, the inflation 0f the udder with sterile air, succeeded because
the air pressure thus created inhibited milk secretion and gave the ani-
mal time to adjust its calcium metabolism. The presently recommended
treatment is the intravenous administrati<?Il of bQth~~hlll.!__ap:<l_g!1}.Q®e.
Although t~e basj_Q_Q§_1}.f3!l_is l!9.t. ~()~pJl.)~ely known, it is tl).Ought__t.h.lLJi..,the
parathyroid glands are i_~'yoJy_ed. -
-The Ejecti.on .or r'Let D.own" .of Milk.- During the nineteenth century
it was generally believed that the greater part of the milk was secreted
during the milking process as the result of the nervous stimulus produced
by manipulating the udder. This view is still widely held by dairymen.
As the result of improved techniques for studying the factors affecting
milk secretion, it is now definitely known that nearly all of the milk is
present in the udder prior to milking. It has likewise been established
that milk is secreted most rapidly immediately after milking when the
udder is empty. As the udder fills \yith milk; the pressure within the
ducts and alveoli increases and results in decreased secretion. As pre-
viously explained, the pressure which is developed in the udder by the
introduction of air in the treatment of milk fever effectively stops the
secretion of additional milk. Every dairyman knows that the most effec-
tive way to dry off a cow is to stop milking her, although he sometimes
hesitates to do so for fear of injuring the udder.
Occasionally a cow does not produce the expected amount of milk at a
particular milking, although the udder appears t~ contain more, Under
these conditions it is frequently said that the animal has "held lIP" her
milk. This is very annoying to the practical dairyman, and he has been
known to resort to various drastic measures to make the cow" let down"
her milk. Most of these measures are ineffective because they are in
opposition to the physiological principles involved.
We know now that a cow cannot "hold up" her milk but that she can
fail to "let it down." .
The greater part of the milk secreted is stored in the millions of alveoli
\\"hich make up the secretory system. Small amounts of smooth muscle
are associated with each alveolus, and it is necessary for these muscles to
contract if milk is to be forced out into the large ducts.
This problem has been extensively investigated by Ely and Peterson
. (1941) and others, and it is now accepted that the "let down" of milk is a
reflex act involving sensory nerves which stimulate the posterior pituitary
"iiand which in turn secretes a hormone (the oxytocic factQr) which
stimulates the smooth muscles around the alveoli and results in the ejec-
tion of the milk.
The production of maximum quantities of milk requires that the milk
be removed from the udder as soon as possible after the hormonal "let-
down" stimulus has been given. These principles are being widely
employed by dairymen and are the basis for the various techniques of
"fast milking" which are being advocated. The natural "let-down"
stimulus is, without much doubt, the crying of the young and the suck-
ling act. Dairy cattle have come to associate many other stimuli with
milking. Such factors as bringing the animals from pasture, the rattling
of milking machines, movements of feed trucks, washing the udder, or
.even radio programs which the milkers tune ilJ.~_§_!,im\lla_te. th~_j)_Q§t.erior
,Eituitary and cause lP_il~~je~ion. The normal stimulus may be easily
interfered with. Changes in the animals' routine, strange noises or ani-'
mals in the barn, mistreatment of the cattle, etc., may prevent the normal
pituitary response. Ely and Peterson (1941) reported that events \vhich
frighten or disturb cattle probably cause the central portion of 'i,he a:<:!!:.e:n111
gland (medullaftosecrete adrenaIIll, which in -turn-lnhiblt~ the--nor~al
actiOn of o~i.Q~in on the muscles sl.lrroUlldlng fhe alveoli. There IS
~n;iderable evidence av:;;ilable In -a-'~id~'-;ariet:y -~fspecies-that fright
or emotional stress results in the almost instantaneous release of epi-
nephrine (adrenalin) into the blood stream.
From the practical standpoint it appears that the hormonal stimulus
for "let down" sometimes is lost before milking is completed. If. for
example, a cow associates the washing of the udder with immediate
milking, delay in milking may result in decreased production. Experi-
ments in which the oxytocic hormone has been injected showed that its
258 BREEDING AND IMPROVEMENT OF FAl12IJ ANIMALS
effects disappear in about 10 minutes. The understanding of the proc-

esses of milk secretion and ejection are of considerable importance in
dairying.
Summary.-We have considered the mammary glands as an integral
part of the female reproductive system. Relatively little mammary
development occurs prior to puberty. After the establishment of ovarian
function there is, in most species, growth of the mammary duct system.
Although there are some species differences, the development of the
lobule-alveolar system, the secretory portion of the gland, awaits the
establishment of pregnancy. During the first half of pregnancy the
alveolar tissue is formed and the gland is anatomically ready for secretion.
There is a difference of opinion as to the exact means of hormonal regula-
tion of mammary development, but it is clear that estrogens and pro-
gesterone are both involW. Whether these substances act directly on
the mammary gland or thr:9u_gll~theanterior pituit~IYis n<?i..a_gr~ed_llP2.n.
The udder appears to grow rapidly during the latter part of pregnancy,
but we now know that this is the result of the distention of the gland with
secretion. The active secretion_QLthe maJIlrnae.a.waita.the production of
t.p.e_anteri.QJ: pit!litary ho~~~~e, la&togl..lIl or prQlaQj:W. This hor~o~~-is
usually secreted in maximum amounts shortly after parturition. Even
though the mammary glands may be fully developed and in active
secretion, the animal must be in a proper physiologic or psychologic
state for the ejection, or "let down," of the milk. This process is now
~r.9.~d as a reflex act. involving sensory neryeJ! whiQh..cause ~~sterior
pij;uitary to produce oxytoci!_l, ,,:hich in turn brings about the contraction
of smooth muscle cells in association 'with the alveoli and forces the milk
into the duct system. where it can be easily removed by the young or by
the milking act.
References
Books
ESPE, D. 1946. "Secretion of Milk," The Iowa State College Press, Ames.
TURNER, C. D. 1948. "General Endocrinology," \V. B. Saunders Company,
Philadelphia.
TURNER, C. W. 1939. "The Comparative Anatomy of the Mammary Glands,"
Univ. Cooperative Store, Columbia, Mo.
ASDELL, S. A. 1931. Recent Developments in the Field of Sex Hormones, Cornell
Vet., 21 :147-152.
ELY, F., and PETERSEN, \V. E. 1941. Factors Involved in the Ejection of Milk,
Jour. Dairy Sci., 24:211-223.
GARDNER, W. U., and STRONG, L. C. H)35. The Normal Development of the
Mammary Glands of Virgin Female Mice of Ten Strains Varying in Suscepta-
bility to Spontaneous Neoplasm, Amer. Jour. Cancer, 25:282.
GOMEZ, E. T., and TURNER, C. W. 1938. Further Evidence for a Mammogenic
Hormone in the Anterior Pituitary, Soc. Expt. Bioi. Qnd Med. Proc., 37 :607.
- - - and - - - . 1936. Non-effect of Estrogenic Hormone on Mammary Gland
of Hypophysectomized Guinea Pig, Soc. Expt. Biol. and M ed. Proc., 34 :320-322.
- - - , - - - , and REECE, R. P. 1937. Growth of Mammary Gland of Hypo-
physectomized Guinea Pig, Soc. Expt. BioI. and AIed. Proc., 36 :286.
MEITES, J., and TURNER, C. W. 1948. Studies Concerning the Induction and
Maintenance of Lactation. I. The Mechanism Controlling the Initiation of
Lactation at Parturition, Mo. Agr. Expt. Sta. Res. Bul. 415.
- - - and - - - . 1948. Studies Concerning the Induction and Maintenance of
Lactation. II. The Normal Maintenance and Experimental Inhibition and
Augmentation of Lactation, Mo. Agr. Expt. Sta. Rell. Bul. 416.
NELSON, W.O., and PFIFFNER, J. J. 1931. Studies on the Physiology of Lactation.
I. The Relation of Lactation to the Ovarian and Hypophyseal Hormones,
Anat. Rec., 51 :51.
REECE, R. P., and TURNER, C. W. 1937. The Lactogenic and Thyrotropic Content
of the Anterior Lobe of the Pituitary Gland, Mo. Agr. Expt. Sta. Res. Bul. 266.
REINEKE, E. P., and TURNER, C. W. 1942. Formation in Vitro of Highly Active
Thyroproteins, Their Biologic Assay, and Practical Use, Mo. Agr. Expt. Sta. Res.
Bul.355.
STRICKER, P., and GRUETER, F. 1928. Action du lobe anterieur de l'hypophyse
sur la montee laiteuse, Soc. de Biol. (Paris) Compt. Rend., 99 :1978.
SWETT, W. W., and MATTHEWS, C. A. 1942. Mammary Gland and Udder Studies,
Report of the Chief of the Bureau of Dairying, U.S. D€pt. Agr., pp. 5-6.
- - - and - - - 1935. Variations in Mammary Gland Development in Dairy
Calves, Amer. Soc. Anim. Prod. Proc., pp. 54-59.
TRENTIN, J. J., and TURNER, C. W. 1948. The Experimental Development of the
Mammary Gland with Special Reference to the Interaction of the Pituitary and
Ovarian Hormones, Mo. Agr. Expt. Sta. Res. Bul. 418.
TURNER, C. W. 1939. The Mammary Glands. In ALLEN, E., DANFORTH, C. 11.,
and DOISY, E. A., "Sex and Internal Secretions," The Williams & Wilkins Com-
pany, Baltimore.
CHAPTER X
ARTIFICIAL INSEMINATION
Few advances in methods of livestock production have been so rapidly

and enthusiastically accepted as the artificial insemination of dairy cattle.
Not that the technique is limited only to dairy cattle, but economic and
management conditions have made it especially adaptable to this class of
animals in the United States. There are certain situations which make
it equally valuable in sheep, swine, horses, chickens, fur bearers, and even
the human.
~nitiQn, artificial insemination is. the deposition of spermatozoa
in the~_~!e genitalia llY instruments rather than by natural service.
The term artificial breeding, which is now widely used to describe the
process, has in the past suggested something unnatural or even repugnant
to livestock owners. To some it has meant the solution of all breeding
difficulties, the possibility of impregnating the entire herd at one time, or
the regulation of sex. To others it has suggested that the offspring
might be weak or otherwise abnormal, and many farmers traveled for
miles to observe the first calves born as the result of artificial insemina-
tion. When properly employed, artificial insemination has little effect
on normal breeding efficiency. There are rare instances of animals "'ith
abnormalities preventing normal breeding which have conceived follow-
ing artificial insemination, and there are numerous instances in which the
improper use of the technique has failed to bring about pregnancy. It·
can be accepted as a fundamental breeding principle that females which·
cannot conceive following service to a normal male will not perform any
better when artificially inseminated.
History of Artificial Insemination.-Artificial insemination is not new.
Loessl (1934) relates a supposedly authentic account of an Arab chief in
1322 who mated a prized mare with the stallion of an enemy chieftain by
stealthily collecting semen from the stallion's sheath and successfully
artificially impregnating his mare. It is definitely known that §pal~an
zani was successful in the artificial insemination of dogs in 1780, and
lIunt~ (1799) produced a pregnancy in the human by this method.
jwanoff, a Russian Rhysiologist, began a study of artificial insemjnation
in farm animal§ as early as 1899, establlshed a labGl"atory fQr_iur.tbgr
j~stigation in 1909, and initiated large-scale investigations and the
I
practical applicati9n of the findings in 1922. -

260
ARTIFICIAL INSEMINATION 261
One hesitates to trace the development of artificial insemination in

the United States, for this technique has undoubtedly been discovered
and rediscovered numerous times. In 1906 Lewis, at the Oklahoma
Station, described the use of the impregnator for the insemination of
mares and in 1911 wrote "the use of the impregnator, or other artificial
means of introducing the semen, in horse breeding has passed beyond
the experimental stage." Yet, in 1949, we are still careful to assure
cattlemen that this technique has now passed beyond the experimental
stage.
The first large-scale artificial-insemination association in the United
States was organized in Clinton, N.J., in May, 1938, by Prof. E. J. Perry
of the Dairy Extension Service, and similar organizations ,,,ere established
in several other states in 1938 and 1939. There is probably not a single
state which does not have at least limited artificial insemination available.
Statistics released by the Bureau of Dairy Industry as of Jan. 1, 1949,
showed that in the United States there are 1,263 artificial-breeding
c'ooperatives with a membership of 316,177 and 2,412,160 cows listed for
insemination. When the third edition of this text was published in 1942,
it was estimated that about 100,000 dairy cattle had been inseminated
the previous year.
ADVANTAGES OF ARTIFICIAL INSEMINATION
Increased Use of Proved Sires.-The greatest advantage of this
method of breeding is that it has increased the usefulness of superior
sires. In previous chapters we have made recommendations regarding
the frequency of use of males. It is a rare bull that is naturally mated to
as many as 50 or 60 cows a year and even more rare for one to sire as
many as 200 calves during a lifetime. Milkdale Aristocrat Rag Apple,
one cif the great Holstein bulls of his time, sired more than 10,000 calves
in the New York Artificial Breeders' Cooperative. It would have taken
him about 200 years to accomplish this in natural service. But the real
value of such a bull is in his ability to transmit high milk production to his
offspring. The first 10 tested daughters of this bull produced 664 lb.
of milk and 46 lb. of butterfat more than their dams, and the dams
themselves averaged more than 500 lb. of butterfat.
Adequate records are now available to prove what animal geneticists
have predicted. For the United States as a whole, the cmvs born as the
result of artificial insemination have produced 859 lb. of milk and 41
lb. of butterfat more than their dams (Lawritson and Nibler, 1946).
These increases are similar to those reported by states such as New York
and New Jersey which have been in the program the longest.
In areas where the levels of production of the cows are not so high as
262 BREEDING AND IMPROVEMENT OF FARIll ANIMALS
in the intensive dairy states, the improvement of the daughters should

be even more striking. In Indiana, for example, the average milk pro-
duction is about 4,200 lb. and fat yield about 190 lb. The use of proved
sires with average indexes of 500 lb. of fat should have a profound effect
on the productivity of the daughters in this case. We cannot assume
that production will fall halfway between the index of the sire and the
dam, for we do not yet know whether the owners of cows "\vhich produce
200 lb. of butterfat will make the adjustments in feeding and manage-
ment which 350-lb. cows require. If butterfat production should be only
half of that anticipated, the increase would still average 75 lb. There
can be little question that the availability of the services of such proved
sires at average fees of $7 per cow provides the average dairyman with
germ plasm seldom if ever available to him before.
Allows More Sires to Be Prove_d.-There is no one answer to the
q~stion of whether only proved sires should be used for artificial insemin-
ation. If all other factors were equal, there is little doubt that ,Ye would
select only those males 'which had been progeny-tested. The supply of
proved sires has never been sufficient to meet the demand. The rapid
increase in the use of artificial insemination has resulted in very keen
competition for the good individuals which are available. As has been
true in the past, many of the best males have gone to the slaughterhouse
before their worth became known, and others are old and debilitated.
Many of these proved sires have fairly satisfactory fertility when used
in limited natural service but have not produced semen of the quality
essential for artificial insemination.
It seems logical to expect that in the near future artificial-insemination
associations must undertake some type of a proved-bull program. The
chief risk to livestock owners in proving bulls is that they might end up
with a group of replacement animals inferior to the parent stock. This
risk could be distributed so widely by artificial insemination that even
disappointing bulls would cause little loss to individual herd owners.
If all members of local artificial-insemination associations bred a few cows
to promising but untried young bulls, the risk would be negligible and
the method of proof would be better than that now used. Instead of
having 5 or 10 tested daughters, we could expect a sire to have perhaps
100 or more, and in addition the results would not be unduly influenced
by conditions in a single herd.
Eliminates Need of Bull in Small Herds.-The owner of a small herd
is faced with a great many problems in regard to the purchase and main-
tenance of a sire. He cannot afford to buy the quality of animal now
available for artificial insemination. He likewise hesitates to build the
\ ., \\
'",
\ \.
kind of quarters necessary for safety in keeping the bull, to say nothing
of conditions which will be optimum for the maintenance of the male in
good breeding condition. It has been shown that the actual cost of
artificial insemination in herds of 10 cows or less is not so great as when
a bull is kept. It is therefore safer and cheaper, and far superior from
the breeding standpoint, for the smaller dairyman to adopt artificial
insemination.
Lessens Disease and Improves Breeding Efficiency.-These advan-
tages can likewise be possible disadvantages if artificial insemination is
not properly used. The selection and care of the bulls used for artificial
insemination must be in the hands of people who are specially trained
for such work. When proper precautions are taken, the spread of
genital diseases by the male should be virtually eliminated. The problem
of reproductive disturbances in the female still is an important one. It
appears that the interest which is being developed in the owners of
artificially inseminated cows is already having its effect in the early
detection and treatment of breeding difficulties. Improved systems of
record keeping, plus the fact that every cow is closely observed dur-
ing insemination, has been helpful in the early diagnosis of genital
disturbances.
Miscellaneous Advantages.-Artificial insemination is often of value
in overcoming certain nonheritable physical difficulties.;' It facilitates
the mating of animals of greatly different size;·it extends the usefulness
of sires that for~ some reason may have become incapable of performing
natural service;lt increases the use of males of I_!!Q!lOgamous species such
as the fox; and it promises to be a useful technique in hybridization
experiments.
Qt. It usually results in the keeping of better calving and breeding records,
and as previously mentioned, this is one of the prerequisites to better
breeding efficiency.
t>Artificial insemination should result in the development of animals of
more uniform type and production and thus attract livestock buyers to
a community.
Perhaps one of the outstanding advantages of artificial insemination is
that it has stimulated greater interest in better livestock-breeding and
man~ement p_r:actices. There is little pride of accomplIshment III We
producti;~ ()f mediocre and inefficient animals. Dairy-extension men
and county agents have found the introduction of artificial insemination
to be followed by increased interest in methods of raising calves, the
feeding and care of pregnant cows, and in dairy-herd improvement
associations.
264 BREEDING AND IMPROVEMENt' OP,P4RM',A-NIMALS
DISADVANTAGES OF ARTIFICIAL INSEMINATION
There are no disadvantages of artificial insemination which cannot be

overcome by the application of knowledge which is at present available.
,. Artificial insemination will not overcome sterility or result in higher
--
ferliiity than can be attained 'when normal healthy cows and bulls are
maj&d,]i!J'he"careless use of artificial insemination will result in lowered
'b~~eding efficiency,'" and failure to follow strict sanitary practices may
result in the spread of disease. These should not occur in a properly
managed association.
The conception rate immediately following the establishment of ani
artificial-insemination association in a particular community is frequently
lower than the previous natural-breeding efficiency. It appears that this
is most frequently due to inexperience both on the part of inseminators
and owners. The rapid acceptance of this method of breeding has
resulted in a great shortage of trained inse'minators. It should be
recognized that even the best conducted insemination schools cannot
turn out a finished product in 10 days or 2 weeks. The basic principles
of insemination can be learned, but the development of an automatically
smooth routine of insemination requires much practical experience.
Most inseminators require experience with about 200 cows before they
become highly proficient. The herd owners likewise need experience.
The detection of heat, observations in regard to the exact onset and
duration of heat, the necessity of calling insemination headquarters at
certain agreed-upon times of day, having cows ,vhich are to be bred
readily available for the inseminator, and cooperating in every 'way
possible with the inseminator will do much to guarantee the success of
the practice.
Some of the disadvantages or dangers which seemed important a
decade ago have not materialized. As far as is known, un~crupulous
operators have not misrepresented the source of semen used. The use
of artificial insemination has not" ruined" the purebred breeder. If any-
thing, the demand for good sires has increased. Some associations have
failed, but the failures have usually been human ones rather than due to
a lack of basic knowledge of how to make artificial insemination work.
METHODS OF COLLECTING SEMEN

The ideal method of collecting semen. is, theoretically, one which is
completely satisfactory to the male, which is easy to use, which requires
a minimum of equipment, and which permits the collection of a sample
of normally ejaculated semen free from contamination with dirt, bacteria,
or secretions from the female genitalia. Many techniques have been
developed. Some methods are applicable only to certain species; others

are too complicated in principle and practice; some do not result in the
normal ejaculation of the male and the production of completely normal
semen; and others yield semen that is diluted with vaginal and ceryjcal
secretions and often heavily laden with bacteria.
FIG. 77.- Diagrammatic artificial vagina for the collection of bull semell. 1. Outer rubber
jacket. 2. Entrance for penis. 3. Warm water. 4. Heavy rubber band. 5. Water
inlet. 6. Rubber cone. 7. Graduated test tube.
The Artificial Vagina.-Durlng the past few years, Russian, British,

and American laboratories ha,>;e developed a type of apparatus called
the artificial vagina. Its principle and c')nstruction, whether for bulls,
boars, rams, or stallions, is simple, and it meets all the requirements of
the ideal method. The apparatus is made up of an outer tube or casing,
FIG. 78.-Bulls can be trained to mount a simply constructed dummy.
usually constructed of heavy rubber, metal, or one of the plastics, and an

inner tube, or lining, of thin rubber. The space between the two tubes
is usually filled with warm water or air or both; one end of the apparatus
is open to allow the entrance of the penis and to the other end is attached
It glass tube, or beaker, to catch the semen after ejaculation ,
The success with which this method is used depends upon the skill of
the operator and the training of the male, but most workers feel that it is
the most satisfactory yet devised. The equipment should be clean,
ster~e. Males are very particular that the apparatus
be of the correct temperature, pressure, and degree of lubrication, and
conditions that suit one male may be unsatisfactory for a second. The
proper temperature is attained by introducing warm water (40 to 45°C.
for most males) between the outer and inner casings; the correct pressure
FIG. 79.-Artificial vagina for stallion.
depends largely upon the size of the penis of the male in question and is
usually regulated by introducing or releasing air between the casings.
A small amount of special lubricant. which is commercially available,
should be applied to the first few inches of the inner rubber tube, but
the amount used should be kept ·to a minimum, for it is undesirable to
have any excess lubricant carried through the artificial vagina and mixed
with the semen.
Males usuaUy respond to this method without previous training.
Best results are attained when the male is allowed to mount an estrus
female and the penis quickly guided into the open end of the artificial
AR1'IFICIAL INSEMINAl'I01'i 267
vagina. If the proper temperature, pressure, and degree of lubrication

have been maintained, ejaculation is instantaneous in the bull and ram
and is very shortly initiated in the stallion and boar. Active males may
easily be trained to mount nonestrus cows, other males, or dummy
fe~les, and highly satisfactory samples of semen can quickly be collected.
"The Massage Method.-Miller and Evans (1934) developed a tech-
nique of collecting semen from bulls by the massage of the ampullae of
the__yasa d~ferentia. The successful applicatioUof-this-method, ,,;hich
-~~nsists of th-e insertion ..of
-
---,~
-- .
----~-.,
-
- -----
__ the arm into the reQ._t,_um, the location... and
manipulation of the ampullae, seminar-vesicles, and prostate region, and
--------- . ,,---. "--- ---".- _". -
~,,-
the collection of semen as it drips from the glans penis, requires a high
degree of ~klrr:-lt is most valuable for the collection of semen from bulls
with injuries to their legs, backs, or ~oductive organs, but for ge;enc
reasons should probably not be applied to bulls suffering from physio-
logical disturbances preventing service that might be transmissible.
This method does not compare in usefulness with the artificial vagina.
There are marked differences in the response of individual bulls; the
semen is often contaminated as it drips from the glans penis; the volume
and concentration of semen obtained by this method is often less than
by service into an artificial vagina; and the mixture of urine with the
semen at the time of collection often kills the spermatozoa.
Burrows and Quinn (1935) perfected a massage method for the col.,.
lection of semen from the domestic fowl and turkey that is widely and
successfully used. The copulatory organ of the male can be protruded
by manipulating the cloacal region, and the semen stored in the bulbous
d~~s can be pressed out by the thumb and first finger.
~ecovery of Semen from the Vagina.-One of the older methods of
semen collection is the recovery of semen from the anterior vagina fol-
lowing natural service. This can be accomplished by introducing a glass
or rubber catheter into the vagina, locating the pool of semen, and
aspirating it into a syringe. This method can be used satisfactorily for
the recovery of both ram and stallion semen if the genital health of the
ewe,or mare is known to be satisfactory. However, this method is not
the one of choice in any species. Breeders who do not have other collect-
ing equipment can make use of the method for the recovery of a semen
sample for a fertility examination. The simplicity of the method is
overbalanced by several disadvantages. The semen may be excessively
diluted by female secretions a~d may have poor storage qualities. If an
infectious genital' disease is present, it is very apt to be spread to the
im;~inated animals.
"the Breeder's Bag.-This technique, which involves placing a rubber
sheath over the glans penis, has-been satisfactorily employed by some
Rta.llion and jack owners for many years and has been discarded by
others as unsatisfactory. The chief criticism of the method has been
that the bag often slipped from the penis as the male dismounted. This
has now been overcome by the development of a longer type of sheath.
The method has the advantages of simplicity and the recovery of a normal
ejaculate, but is limited to equines.
Electrical Ejaculation.-This technique was developed by Gunn in
1936 as a method for the collection of ram semen. Various modifications
have been made for small animals, and it has been of value from a
research standpoint.
Two electrodes are used. In some species one is placed in the rectum
and the other in the muscle abO\Te the loin, in other species one electrode
is placed against the roof of the mouth and one at the base of the skull.
An alternating' current of varying intensity, depending on the species,
is applied. The equipment is expensive, requires a high degree of skill,
and gives variable results, even in experienced hands. It is unlikely that
it will ever be widely used in ,practical semen collection.
THE EVALUATION OF SEMEN
The successful application of artificial insemination is. to a large extent,
dependent upon the quality of semen used. Those responsible for the
collection, processing, and shipment of semen must be able to evaluate its
quality and to predict its potentialities when used under practical condi-
tions. Those concerned with the insemination of cows must likewise be
able to evaluate semen, lest females be inseminated with nonviable sperm.
Nearly every sperm physiologist is hopeful of being able to develop a
single method of evaluating semen quality. The list of techniques which
have been used is a long one and varies from the visual over-all score of
semen following a brief microscopic view, to the chemical determination
of enzymes, vitamins, or specific metabolic processes. Unfortunately no
one method will yield an accurate estimate of the actual per cent of
inseminations which will result in pregnancy when a particular semen
sample is used. However, the fact that the conception rate does not
fluctuate greatly between days or months or between well-managed
insemination associations indicates that a fairly good job of evaluation is
being done.
Volume of Semen.-Volume is usually determined in a graduated
centrifuge tube in the bull, with a 1-ml. tuberculin syringe in the ram and
rooster, and with a 100- to 200-ml. graduate in the stallion and boar.
The volume is not necessarily related to the fertilizing capacity of the
spermatozoa, since sterile animals may produce large ejaculates. Volume
is C?f gr!~r!tc~ic_al}rr~portancei~__artificial inseminatipn. If -aTfOther
11RTIFICIAL INSEMINATION 269
factors in regard to quality are equal, the male with a large ejacqlate is
most desirable. The rate of semen dilutIOn which will be made is deter-
~i~~d largely on the basis of semen volume and numbers of sperm per
unit volume of semen.
Number of Spermatozoa.-The number of spermatozoa per unit
volume of semen varies with species and indi vidual, and breeding males
should be expected to meet the requirements for their species as previously
elaborated. The number of sperm in semen is usually expressed in
terms of sperm per cubic millimeter. The standard method of counting
sperm is the same as that -empI;;yed for counting red-blood cells. This
technique is described in detail in nearly every physiology text. In
brief, it involves the dilution of the semen in a red-blood cell diluting
pipette, the thorough shaking of the pipette to bring about an even dis-
persion of the sperm, and the discharge of a small quantity of the diluted
semen into the measured chamber of a hemacytometer. The number of
sperm ,vithin certain ruled areas of the hemacytometer is counted
heneath a microscope, this number is multiplied by a standard factor,
and the number of sperm is thus determined.
Methods for the estimation of sperm numbers by the centrifugation
of capillary tubes containing whole semen have been described. By
centrifugation at a certain speed in relation to gravity for a standard
time, the sperm are packed into one end of a capillary tube and the
volume of the packed cells is measured. This method has the advantage
that several replications of each sample can be made simultaneously, but
it has not been used in practical insemination work (Shaffner and
Andrews, 1943).
The ideal method of determining sperm numbers in bull studs must
be simple, rapid, and accurate. It had been shown that the dilution of
semen at a standard rate and then measuring the relative light trans-
mission in a photoelectric colorimeter was rapid and accurate, but the
equipment required was rather expensive (Comstock and Green, 1939).
This method was adapted for field use by the development of opacity
standards for the simple visual comparison of diluted semen. Complete
directions for the prep~lation of standardized tubes of different opacity
have been described by Salisbury et al. (1943). Such tubes are now
available commercially. The 'equipment is called the comparator, is
inexpensive and rapid, and is nearly as accurate as any other method.
This is the standard technique of determining sperm concentration in
~ most bull studs.
Motility of Spermatozoa.-The estimation of sperm motility by micro-
scopic examination is one of the oldest of all clinical examinations involv-
ing the microscope. In fact, Leeuwenhoek, the father of microscopy,
270 BREEDING AND IMPROVEMENT OP FARM ANIMALS
and Dr. Ham described the appearance of spermatozoa in 1677. This

estimation of semen quality has many limitations, but is probably the
one most used by those engaged in practical artificial insemination.
As discussed in an earlier chapter, the presence of motility does not
guarantee fertility, but the complete absence of m<;ltiiTti i~ ~ very g~~d
--·iffdicati;~ that fertiIlzmiCcapacity -·has be~n lost,) There are various
mi~or differences in the techniques which are used in different laboratories,
but the one which we will describe is fairly typical.
Prior to the widespread dilution of semen for artificial insemination,
motility referred almost exclusively to the degree of movement of
spermatozoa in whole semen. It is now necessary to distinguish between
the motility of "v\ihole and diluted samples. There is little similarity
between the appearance of fresh whole semen and the same sample
immediately after dilution. The appearance of good-quality fresh semen
from the bull, ram, or cock, species with high sperm concentration, is
that of millions of sperm in vigorous motion, and the semen itself has the
effect of a body of water with fierce eddies and currents and a constant
undulating movement. It is difficult to distinguish the behavior of
individual cells in such samples, and even dead sperm appear to move
because of the motion set up by the live sperm. Because of the dilution
practiced in artificial insem.illa.tion,. there is littre-m;;;;~~nt of the
-J:lqUTcI ltse.!f,-;~fi~~~,:~~~al sperm stand out clearly, It has been our
experience in the conduct of artificial-insemination schools that if fresh
whole semen is first used for demonstration purposes it is subsequently
difficult to convince the trainees that diluted samples have any motility
at all, to say nothing of the differences between good and poor samples.
The initial motility of fresh semen should be determined as soon as
possible after collectio~'l of the semen. This should be done before the
semen temperature has dropped much below body temperature. It is
accomplished by placing a small drop of semen on a hanging drop slide
in a warm room, or temperature can be maintained by the use of a special
slide warmer. Various types of motility have been described, but the
clinician usually distinguishes rapid, progressive, and in place, or undula-
tory, motility and scores semen samples on a basis of 0 to 5 in which:
5 = 80-100 % progressive motility, swirling motion of the drop as a,
whole; " "
4 = 60-80% progressive motility, swirling motion indefinite or absent;
3 = 40-60% progressive motility, movement limited to individual
spermatozoa;
2 = 20-40% of spermatozoa showing undulatory movement;
1 = 1-20% of spermatozoa exhibiting undulatory movement;
o = no motility.
ARTIFICIAL INSEMINATIOiV 271
Semen which has been diluted with egg-yolk buffer is somewhat hard
to see through under the microscope and the presence of large numbers of
varying-sized yolk globules is confusing to the beginner. The standard
for fresh semen ,vhich is described above is the one in general use with the
exception that there is no swirling motion of the semen itself. The type
of motility of the individual sperm is important. Samples which score 4
or 5 should contain sperm showing definite progressive motility, and the
movement of the sperm should be sufficiently vigorous to agitate the
yolk globules.
There are several factors which commonly affect motility and which
will lead to incorrect estimates if not guarded against. There is a pro-
gressive decline in motility as temperature falls. The sample should be
warmed, vreferably t6 a standard of 80 to 900'F., if uniform results are to
be obtained. The determination should be completed rapidly. Semen
dries quickly in a warm room, and exposure to air inactivates the sperm.
If an estimate is not made within a few seconds, a second sample should
be taken. Dirty slides or traces of water or alcohol will affect the degree
of motility. Although it cannot be said with certainty that a diluted
sample scoring 1 will never produce fertilization, it can be said that the
degree of fertility will be highest when samples scoring 4 or 4 plus can be
used.
Duration of Sperm Motiiity.-It has been known for many years that
there is a high correlation between the maintenance of sperm motility
and the fertilizing capacity of sperm when either whole or diluted bull
semen is stored at 5°C. If the samples are properly handled, motility
persists fot 2 weeks or more if the bull is of normal fertility. While this
method is one of the better ones for the measurement of semen quality,
it has the disadvantage of requiring a relatively great length of time and
the evaluation is not completed until the original semen sample has long
since ceased to be usable.
Weisman (1941) developed standards for the survival of human sperm
stored at different temperatures. He reported that the survival times of
normal human sperm at 20 to 23°C., 37.5°C., and 45°C. should be at
least 24, 10 to 12, and 1 to 2 hours, respectively. Studies at Cornell by
Beck and Salisbury (1943) showed that there were significant correlations
between the decrease in motility of sperm in diluted bull semen stored at
5°C. and similar samples stored at from 46.5 to 47.5°C. The correlation
coefficient of storage for 10 days at 5°C. and storage at 46.5°C. for 1 hour
was 0.9088, indicating that by increasing storage temperature much
valuable information in regard to the maintenance of motility can be
gained in a short time. This method has many laboratory applications
but is not widely used in the field.
272 'BREEDING AND IMPROVEMENT OF FARM ANIMALS
The Methylene-blue Reduction Test.-The dye methylene blue loses

its deep blue color when tv.o atoms of hydrogen are added. This
phenomenon of methylene-blue reduction has been used for many years
as a technique in the measurement of certain types of cellular metabolism,
especially in bacteria.
A method has been developed at Cornell for the evaluation of semen
quality by determining the time required for semen of varying qualities
to reduce a standard amount of methylene blue. Two-tenths ml. of
buH semen is diluted with 0.8 ml. of yolk-citrate diluent in a smaIl test
tube. To this is added 0.1 ml. of a methylene-blue solution composed
of 50 mg. of methylene blue dissolved in 100 ml. of sodium citrate buffer.
When the tubes~re incubated in a \~rater bath at 45°C., the best samples
reduce the blue color in 3.5 to 6 minutes. Poor samples require longer,
up to 40 minutes (Beck and Salisbury). It was shown that there were
high correlations between concentration of spermatozoa, sperm motility,
and the amount of ascorbic acid in the semen and the time required to
reduce the blue color. This method has been used in numerous bull
studs as a....measure of semen quality. Semen which requires more than
9 minutes to lose the blue color is regarded as subnormal.
This technique has been adapted to the semen of other species, but it
is the writers' opinion that it seems to be better adapted for bull semen
than for boar or chicken semen.
Measurement of Sperm Metabolism.-As indicated in the chapter on
the male, the chemical nature of semen and the metabolism of sperma- I
I
tozoa have been favorite subjects for laboratory investigation. The
glycolysis of semen and the oxygen consumption and carbon dioxide
production of sperm can be measured. In a very general way there is a
relationship between sperm numbers, motility and survival, and the type
and degree of metabolism. However, these techniques are not easily
adaptable to field conditions.
Differentiation of Live and Dead Spermatozoa.-The relative numbers
of live and dead spermatozoa can be estimated by the.~~~c~_~crosQQP.i.£..,
examination of fresh semen. This method is obviously subject to con-
siderable- hunianerror:"··W~rkers at the University of Missouri have I
reported a staining method for the differentiation of live and dead
spermatozoa. The technique was first worked out for ram sperm by
Lasley et al. (1942) and was subsequently modified for other species by
Mayer et al. (1947). A stain containing fast green FCF and either
eosin B or erythrocin B is made up in a phosphate buffer at a pH of 7.25,
and a small sample of semen is stained according to very specific direc-
tions. Dead sperm are stained by such a mixture and live sperm remain
unstained. This procedure appears to 'have been very reliable in the
\\
\
'I"
\
hands of its originators, but others have obtained variable results. The
method has been shown to be affected by minor changes in pH, osmotic
pressure, and stain concentration, and the technique must be very
carefully carried out (Mayer, 1948).
This method, or modifications of it, will have practical application if it
can be controlled.
Number of Abnormal Spermatozoa.-Although Ham and Leeuwen-
hoek observed the spermatozoa of the human as early as 1677, it is only
during the twentieth century that detailed studies of spermatozoa have
been made. Williams and Savage (1927) made extensive studies of the
morphology of bull spermatozoa and reported that in bulls with poor
breeding records an average of 50.1 per cent of the spermatozoa were
abnormal and that bulls with good breeding records averaged less than
16.6 per cent abnormal spermatozoa. These findings have been con-
firmed by Moench and Holt (1931), \vho found that men producing more
than 25 per cent abnormal spermatozoa were of lowered fertility or sterile,
and by McKenzie and Phillips (1934), who reported that normal rams
produced not more than 15 per cent abnormal spermatozoa. Several
slides of each semen sample to be examined should be prepared at the
time of collection. This can be accomplished by placing a very small
drop of thoroughly mixed semen at one end of a glass slide and drawing
it out in a very thin film in the same way that blood smears are made.
Considerable practice is required to make satisfactory slides of highly
concentrated semen. The problem is one of obtaining a smear thin
enough for study without mechanical damage to the spermatozoa. The
slides may be allowed to air-dry and then may be placed in a saturated
chlorazene solution for 5 to 10 minutes to remove mucus. They should
then be rinsed in distilled water, washed in 95 per cent alcohol, allowed
to dry, then stained with Ziehl's carbol-fuchsin for 1 to 2 minutes, rinsed
in tap ,Yater, and allowed to dry. The stained slides should be studied
under high-dry or oil-immersion objectives and classified by some
definite scheme.
Nearly as many systems of classification as there have been investi-
gators have been devised, but the abnormal sperm types most usually
encountered are head abnormalities such as tapering, shrunken, large,
or small heads; enlarged, beaded, broken, or filiform middle pieces; and
coiled, broken, or missing tails. A study of a single smear of semen is
sometimes misleading, for seasonal effects, excessive use, or long periods
of sexual rest are among the factors that may temporarily increase the
number of abnormal spermatozoa. McKenzie and Berliner (1937) in an
{'xtensive study of the reproductive capacity of rams found that, under
Missouri conditions, the number of abnormal sperm forms increased
during June, July, and August and were at a minimum during November,
December, and January. Shropshire rams were more noticeably
affected than Hampshire rams, the former having an average of 733 and
the latter an average of 90 abnormal sperms per 1,000 during August.
Semen samples collected after prolonged periods of sexual rest should
not be used for the determination of fertility. Spermatozoa that are not
ejaculated tend to become senescent and are eventually broken down and
resorbed. This fact often accounts for the failure of males to settle the
first females to which they are bred at the beginning of the breeding
season and must be taken into account lin the estimation of fertility. "
A few years ago this measurement was considered essential in the . \
evaluation. of semen. More recently the emphasis has been on the j
physiological measurement of semen quality. The determination of

abnormal sperm forms 'will frequently distinguish between superior,
questionable, and inferior semen samples, but it has never been possible
to predict relative fertility very closely.
pH of the Semeno-The development of the glass electrode pH meter,
as well as a variety of special test papers for different pH ranges, makes
possible the accurate measurement of hydrogen-ion concentration of
semen. Unfortunately, this measurement is of little value in practical
semen evaluation since the pH range of semen does not ordinarily
fluctuate widely. '
Purity of Semeno-Semen should be collected and maintained as free
of microorganisms, cellular debris, dirt, blood cells, urine, or other
foreign matter as is humanly possible. There would be no quicker way j
to spread genital diseases than by the use of semen containing such
organisms. Many substances such as red- or white-blood cells, urine,
water, soap, disinfectants, or secretions of the female genitalia are either
toxic to spermatozoa or adversely affect their viability and should not
contaminate semen.
PROCESSING AND SHIPMENT OF SEMEN

Cooling Semeno-Many substances should be cooled as rapidly as
possible to prevent degenerative changes, e,g., milk, meat, and foodstuffs
which are to be preserved by quick freezing. Semen should be cooled
to about 4 or 5°0. (40°F.) in order to maintain sperm motility for maxi-
mum periods of time, but it is now known that too rapid cooling will
permanently reduce sperm viability. This is called temperature shock,
and it has been reported in several species.
Temperature shock must be prevented, especially during the collection
of semen and before the semen has been diluted. The glass receptacle
on the artificial vagina which is to receive the semen should be warmed
before use and protected by insulation during the winter months. An

artificial vagina in which the receptacle is surrounded by the water
jacket was developed at Cornell and has proved practical.
Immediately after collection, semen should be taken into a warm, dust-
free room and motility score and sperm numbers determined. If the
sample is clearly of good quality, it is usually diluted at once \vith fresh
dilutor which has been warmed to about 80°F. It is important that the
semen and dilutor be approximately the same temperature, and since
fresh collected semen cools to about 80°F. rather quickly, this temper-
ature is a practical one.
The diluted semen should then be cooled at the rate of about 1°F.
per minute until the storage temperature of 40°F. has been reached.
This can be accomplished in the laboratory by electrically operated
devices, but under practical conditions it is usually done by placing the
semen container in several changes of water, each a little cooler than the
preceding one.
By a little practical experimentation the semen container can be
put in a jar containing water at 80°F. and then placed in a refriger-
ator. By adjusting the respective volumes of semen and water, the
semen can be made to cool at the required rate.
Semen Diluters.-Many substances have been used for the dilution
of semen, but most of them have been lacking in one or more of the
qualities of a practical diluter. In order to be suitable for use in practical
artificial insemination a diluter must maintain the fertilizing capacity
of the sperm for maximum periods of time, it must be easily prepared, of
readily available substances, be low in cost, and nonirritating to the
recipient of the semen. The history of the diluter problem is a long one
and has not been confined to the bovine species. Among the materials
tried have been physiologic saline solution to which has been added
numerous organic and inorganic substances including glucose, amino
acids, vitamins, hormones, drugs, and antibiotics. Nearly all the body
fluids and secretions have been tried, blood plasma, blood serum, spinal
fluid, milk, reproductive accessory-gland secretions, and semen plasma
itself.
In 1940 Phillips and Lardy developed a diluter made up of egg yolk
and phosphate buffer for cattle semen. This diluter was widely adopted
and is still in general use. The phosphate buffer contains 0.2 g. of
KH 2P0 4 and 2.0 g. of N aHPOd2H 20 in 100 ml. of sterile redistilled
wate~. To facilitate its preparation, capsules containing the proper
amount of chemicals for addition to the water are commercially available.
Fresh egg yolk is prepared by the complete separation of the yolk from
the albumen, puncturing the membrane surrounding the yolk, and allow-
276 BREEDING AND I1J,fPROVEMENT OF FARiVI ANIM",1LS
ing it to run into a clean sterile graduate. Equal parts of phosphate

buffer are mixed with the egg yolk and the yolk buffer is ready for use in
semen dilution.
At this point it should be emphasized that only cheIDicaIIy pure
ingredients should be used, that water redistilled in a glass still is required,
thatTreSheggs fromhens fed a balanced ration should be used, and that
substances or conditions which might be toxic to sperm must be avoided.
A second practical diluter was developed by Salisbury et al. at Cornell
in 1941. It ~was made up with egg yolk and a sodium citrate buffer and
has since been modified by Salisbury and Bratton (1948) to contain
sulfanilamide. The citrate buffer is made up of 3.6 g. of sodium citrate
dihydrate (Na3C6H507.2H20) per 100 ml. of glass distilled water and the
buffer is then autoclaved for 20 minutes at 15-lb. pressure. Equal parts
of buffer and fresh egg yolk are mixed. Sulfanilamide is added at the
rate of 300 mg. per 100 m}. of diluter.
Several entirely satisfactory diluters are readily available for cattle
semen, the yolk-phosphate diluter, the yolk-citrate diluter with or ~\\"ith
out sulfanilamide, a commercial synthetic pabulum, and a commercial,
completely prepared, modified egg yolk-citrate diluter ready for immedi-
ate use.
There has been relatively little use of diluters for ram, boar, and
chicken semen in the United States. Numerous diluters have been used,
in Europe and Asia, and their composition has been discussed by Ander-
son (1945).
Several diluters suitable for stallion and jack semen have been described
by Berliner (1945). These contain egg yolk, glucose, and K-Xa tartrate
and gelatin in varying a.mounts, depending on the exact purpose for
which the diluter is to be use-d.
Dilution Rate.-Theoretically, only one sperm is necessary for the
fertilization of an ovum and only one specific microorganism for the
establishment of a disease. In practice, however, it is unlikely that only
one cell will produce such results. This has been explained in numerous
ways, the most plausible of which may be that large numbers are neces"
sary, since many of the cells are lost on the way to their destination.
Many of the earlier estimates of the number of sperm required for
fertilIzation have been nullified because of advances in knowledge and
techniques in semen physiology. In rabbits, for example, in 1927 it was
thought that 1 million sperm were necessary for maximum fertility and
that sterility was reached if there ,yere 100,000 or fewer sperm. In 1946
it was concluded that maximum fertility could be attained with 330,000
to 420,000 sperm and minimum fertility ,yith 40,000. In 1948 it appears
that 16,000 sperm will give partial and 90,000 spermatozoa maximum
fertility (Cheng and Casida, 1948).
The first work in artificial insemination involved the use of fresh
undiluted semen. In practical cattle insemination semen was diluted
three to four times in 1940 and about 1: 10 to 1: 16 in 1945. Salisbury
(1948) reported that there were no differences in fertility when bull semen
was diluted with egg yolk-citrate diluter at a rate of 1: 100 in comparison
·with rates as low as 1: 40. One ml. of diluted semen containing about
12.8 million sperm was used for each insemination. It has since been
reported that there were no differences in conception rate when yolk-
citrate-sulfanilamide diluter was used at rates of up to 1: 400. It was
pointed out that "with present techniques the minimum number of sperm
for optimum fertility is between 5 and 10 million if bulls of known fertility
are used (Salisbury and Bratton, 1948).
Much work on dilution rate and composition of diluters is in progress,
and it should be recognized that these results may need further modifi-
cation. As previously explained, the diluter problem is of less practical
importance in the other classes of farm animals, and data similar to those
given for cattle are not available.
Storage and Shipment of Semen.-In artificial-insemination practice
semen is rarely stored at its point of origin but by the local inseminator.
The chief problem is one of the maintenance of optimum conditions
during shipment and while the semen is being transported by the insemi-
nator on his rounds.
The type of shipping container depends on the volume of semen which
is to be sent and the distance involved.
When single semen samples are to be shipped, as is usually the case
when semen from a particular sire is purchased by a breeder at a distance,
an ordinary thermos bottle can be used. It is desirable that the temper-
ature of the semen be maintained at a uniform 4 to 5°C. during shipment.
This can be accomplished by placing a full tube of semen within a second
larger container, taking care to separate the two with cotton. Both
containers should be watertight. The larger container is then surrounded
by cracked ice, the thermos bottle placed in a sturdy cardboard carton
and shipped.
When the time of shipment is 2 or 3 days, the thermos bottle ~lf
should be packed in a refrigerated container. A small, jacketed, insulated
ice-cream shipper has been successfully employed at the University of
MissourL The thermos bottle is placed in a metal can and surrounded
by cracked ice and a can or block of fllozen br~ is placed on top of the
can containing the cracked ice (Herman and R~sdale, 1946).
In most states semen can be shipped from the bull stud to all local
associations within 12 hours. The equipment required for this is sur-
prisingly simple.
The refrigerant is provided by one or two milk cans approximately 2.5
by 4 in. The cans are filled to within about an inch of the top with water,
sealed, and frozen solid. Two layers of heavy wrapping paper are placed
about the cans, and the semen vials are then placed against the wrapped
cans and firmly wrapped with several layers of paper. This package is
placed within one insulated paper ice-cream bag during the cooler months
and within two insulated bags during the summer. This in turn is
placed in a heavy corrugated cardboard box and shipped by common
carrier. This equipment is light in weight, inexpensive, and readily
available commercially.
The vials containing semen should be as nearly full as possible. If
only a small quantity of semen is to be sent, a small vial should be used.
It is known that the mixing of semen with air by agitation during ship-
ment -is undesirable and results in decrel1sed sperm motility (Prince and
Almquist, 1948). .
The experienced inseminator can usually tell if semen shipments are
arriving at the proper temperature. If the ice in the cans has melted,
it is well to check the temperature of the semen with a thermometer and
report abnormalities to the bull stud.
All semen samples 'w:hich the inseminator plans to use should be
examined microscropically each day. There are times when a 48-hour-
old sample may be superior to a fresh one. Knowledge of the actual
fertility of the bulls and experience in semen evaluation will soon enable
the inseminatorto select the samples most likely to produce pregnancy.
The inseminator usually receives fresh semen six or seven times weekly_
It is obvious that he cannot guarantee that the semen of a particular bull
will be used to inseminate a particular cow. However, by adhering to
recommended practices, he usually has the semen of several bulls of each
breed available, even though the semen is of different ages.
In the past there has been much wishful thinking in regard to the
storage possibilities of semen. With rare exceptions bull semen should
be used within 2 to 4 days following collection if good results are to be
expected. Ram semen should be used within 2 days; and boar, stallion,
and chicken semen should be used on the day of collection. To the
authors' knowledge the maximum survival of sperm for conception is 10
days, 6 days, 56 hours, and 48 hours for bulls, ram, boars, and stallions,
respectively. These results have been obtained with stored semen and
represent maximum rather than average values.
The maintenance of maximum fertility in stored semen requires special
ARTIFICIAL I NSEM I NATION 279
handling by the inseminator. The semen should be maintained at a
uniform 4 to 5°C. while in his care. It is recommended that inseminators
be equipped with portable refrigerators of the type commonly used by
veterinarians. The semen vials are transported in small thermos bottles,
and the thermos bottles in turn are kept in the portable refrigerator.
Care should be exercised that the semen is not allmved to remain out of
the refrigerator except for the time required to fill the inseminating
pipette.
TECHNIQUE OF INSEMINATION
The method of insemination will depend both on the species and the
experience of the operator. The methods commonly used for cattle will
be described in detail and but briefly for the other farm animals.
FIG. BO.-Equipment for the artificial insemination of cattle. 1. Artificial vagina. 2.

Lucite or glass speculum. 3. Inseminating tube and syringe. 4. Graduated test tube.
6. Head lamp.
If economy of semen is no object, the deposition of several milliliters

of whole or diluted semen in the vagina of the cow will give satisfactory
results. Since the chief advantage of artificial insemination is to extend
the usefumess of superior sires, this method has little merit. The method
in common use prior to 1940 involved ·the partial penetration of the
'280 BREEDING AND IMPROVEj}IENT OF FARM ANIMAL))
cervix with an inseminating tube while it was visualized through a vaginal

speculum. This technique seemed to have the advantage that the oper-
ator could 8ee what he was doing, but we now recommend that he feel
what he is doing. The muscular tensions which are set up following the
introduction of the speculum make it difficult if not impossible to penc-
trate the cervix further than the outer region. Under practical conditions
it is probable that the semen is frequently not deposited in the cervi:;:
at all. It has the disadvantage of requiring a large supply of sterile
specula, one for each cow, and the chore of cleaning them can be a labo-
rious one. The method does give
satisfactory results and is still used.
The technique in most general use
at present is the grasping of the cervix
"with the left hand in the rectum and
the introduction of the inseminat-
ing tube through the vagina and in-
to the cervix with the right hand.
By the rectal manipulation of the
cervix, the tube can be guided
~\1 through the cervix and even into
the uterine horns. This method is
}'IG. 8l.-Insemination of the cow involv-
rapid, requires a minimum amount
ing the use of a vaginal speculum. The
cervix is visualized and the inseminating of equipment, and is less likely than
tube is passed into the first part of the the speculum to introduce foreign
cervical canal.
substances into the vagina.
The step-by-step routine recommended for inseminators is as follows:
(1) Don clean work clothing, boots, and rubber sleeve as soon as arriving
on farm, (2) positively identify cow, (3) locate equipment case and pail
of warm water near cow, (4) attach clean inseminating tube to syringe
and fill with 1 ml. of desired semen, (5) place inseminating equipment on
special hooks on lid of case, (6) lubricate left hand and arm with soap or
other lubricant, (7) enter rectum carefu~'~and remove feces, (8) wipe
manure from rear quarters 'with fresh pa r towels or cotton, (9) wipe
lips of vulva with small quantity of fresh co on, (lO) grasp cervix firmly,
(11) insert inseminating tube into vagina and approach cervix carefully,
(12) introduce tube into cervical opening, (13) "work cervix back and
forth and up and down, working tube into body of uterus, (14) deposit
semen.
The inseminator must have a basic knowledge of bovine anatomy and
physiology and above all must exercjse good judgment. CarelessneE'E'
may result in perforation of the vaginal wall or injury of the cervix or
1ltems. Inseminators are frequently called to service cows which are
ARTIFICIAL I N SEMINA T101Y 28 1
already pregnant. The penetration of the cervix of a pregnant animal

is almost certain to result in abortion. It may be impossible to prevent
such occurrences in the first few weeks of pregnancy. If the cervix is
tightly closed and dry, li~tle is gained by trying to force penetration.
The cow may not be in heat or she may be pregnant.
A very routine inspection of the cow may reveal an anatomical defect
of the genitalia, inflammation, or infection. Arti.ficial inseminators are
FIG. 82.-The most common technique of inseminating the cow. The cervix is grasped
per rectum, and the inseminating tube is carefully worked into and through the cervical
canal. -
neither trained nor expected to diagnose reproductive disturbances.

When it is obvious that some type of an abnormality is present, it should
be called to the attention of the owner in order that he may secure
veterinary assistance.
The mare is the easiest of the farm animals to inseminate. Mares
should be hobbled in order that the inseminator may work in safety.
The tail of the female should be wrapped with a clean bandage of the
type used on the legs of running horses. The region around the vulva
can be cleaned with a damp sponge and the lips of the vulva wiped with
a piece of clean cotton. The operator should wear clean, dry rubber
gloves which have been previously washed in alcohol or some similar
disinfectant. Fresh semen is poured into Yz-oz. gelatin capsules and

quickly inserted as far forward in the cervix as possible. Twenty ml~
of semen per insemination is sufficient in small mares and 30 to 40 m!.,
in large multiparous females. The semen can also be introduced through
the cervix with a catheter attached to a syringe or a flask equipped with
a pressure bulb. As previously mentioned, mares should be inseminated
more than once during estrus if maximum fertility is to be attained.
Sows are usually inseminated with a rubber catheter attachtd to a
syringe. The cervix can be penetrated by a fairly stiff catheter and the
semen slowly deposited in the uterus. If the cervix is not entered, much
of the semen will be lost from the
100 vagina. There is relatively little
'E
<l)
known about the volume of semen
0 80 and numbers of sperm required
;r Ovulation
for swine. Under practical' con-
I
,.,
0
c 60 ditions 20 to 50 ml. of whole
Q)
~
W 40
I semen can be expected to give
satisfactory results. There is evi-
0'
C dence that much smaller quan-
'0
Q)
Q)
20 After tities of semen, when properly
cD Heal diluted, may be satisfactory, but
0 this has not been verified under
0 4 8 12 18 24 herd conditions.
Time of Inseminotion - Hours Ewes are usually inseminated
FIG. 83.-Relationship between time of with the same type of pipette or
insemination and breeding efficiency in dairy
cattle. (Adapted from the data of Trimber(Jer tube that is used for cattle. The
and Davis (1943), Neb. A(Jr. Expt. Sta. ReB. pipette is filled and emptied with
Bul. 129.)
a syringe or small rubber bulb.
The cervix is visualized with a speculum and a small flashlight or
headlamp and the pipette introduced into the outer part of the cervix-
probably not more than X to Yz in. A volume of 0.2 ml. is usually used
both for whole or diluted semen. The egg-yolk diluters used for cattle
have been successfully employed for sheep. There is considerable
difference of opinion in regard to sperm numbers required, ranging from
5 to 50 million.
Artificial insemination has not been widely employed i.I!..-l2_o}lJtry.
It is very useful when hens are being maintained in individual laying
cages, as in many nutrition and breeding experiments, and it has a place
-:>n certain practical poultry-breeding farms. Hens are usually insemi-
lated with fresh semen once weekly. The oviduct of the laying hen
~an be easily protruded by applying pressure in the cloacal region. A
[-ml. tuberculin syringe, without a needle, is inserted into the oviduct
to a depth of about 1 in. and 0.1 ml. of semen deposited. A single

weekly insemination can be expected to yield about 90 per cent fertility.
A limited amount of work with diluters has been done, but thus far
most inseminations are made with fresh whole semen.
Organization of Artificial Insemination.-The advances in artificial-
insemination techniques which have occurred during the last decade
have resulted in a complete revision of our ideas with respect to the organi-
zation and operation of artificial-breeding associations. Since nearly
every state has extension specialists who concern themselves almost
exclusively with a~ insemi~-ation;-werecommend that such special-
ists be consulted in regard to local conditions.
In a very general way artificial insemination is being used by three
groups in the United States, the pURbred breeder, the local association,
and the central breeding unit. The purebred breeder has long employed
artificial insemination on a more or less private basis. It has been used
to extend the services of valuable sires within and between herds and for
the control of certain genit~~es:-- The fees for such services depend
upon the relative value and popularity of the sires, and there is a growing
demand for this type of service.
A few years ago we thought that the local artificial-breeding associ-
ation was most likely to be successful. Organizations of this type may be
cooperatively or privately owned and usually serve an area with a radius
of about 20 mi. Many such associations were established between 1938
and 1942. Some have been completely disbanded. some have become a
part of a larger organization, and some are still in successful operation.
The chief reasons for modification have been economic. The difficulty
and cost of obtaining suitable bulls has been a major problem. The
number of bulls necessary to serve 1,200 cows of three different breeds,
and still provide a safety factor in case some of the bulls become sterile,
is nearly as great as is needed to serve 10,000 cows or more. In short,
the cost of semen per cow served in a small local organization is greater
than when many local associations are served by a central bull stud.
The general trend in artificial insemination is toward the central
breeding unit. In this type of organization the bulls are maintained in a
centrally located stud. Well-equipped laboratories and technicians
specially trained in the care of bulls and the processing and shipping of
semen are a definite requirement for the success of this type of unit.
Semen production is thus concentrated in one place, and the actual
insemination of the cows is carried out by local organizations. The
problem of locating new sires is of sufficient importance to require perma-
nent personnel specially trained in selection. The need of keeping
abreast of the latest developments in semen physiology is essential.
)
It appears that these problems can best be met when some of the workers
devote themselves exclusively to bull-stud work, and others to the insemi-
nation of cows.
In some of the larger states there are several bull studs located in
strategic areas, each serving numerous local insemination associations.
The type of organization in Indiana is described briefly not only because
of the authors' familiarity with it but because of certain unique features.
Local artificial-breeding cooperatives obtain semen from bull studs in both
Indiana and North Carolina. On alternate days semen from the North
Carolina stud is transported by airplane to the Indiana stud and thence
to local associations. And every other day semen from Indiana is
shipped to North Carolina where it is distributed to local associations in
1'\ orth Carolina, Georgia, Florida and Virgina. From the economic
standpoint semen which is produced but not used represents waste, and
it has long been apparent that we have not been getting the maximum
use from our sires. The safety factor cannot be overlooked, however,
and it is regarded as essential that there be some reserve of semen pro-
duction. The regular exchange of semen between states does make for a
wider use of outstanding sires, and it offers some protection to the live-
stock owners of both states in case of serious trouble at one of the bull
studs, such as fire or disease.
Artificial Insemination Problen:s.-It is obvious that artificial insemi-
nation is established as a practical method of livestock breeding but it is
equally obvious that its applications and techniques are being constantly
changed and improved.
The method of processing and shipping bull semen can be regarded as
satisfactory. That constant improvement is being made is evidenced by
the fact that recommended dilution rates have increased from 1: 10 to
1; 100 within the past 5 years and that greater dilution is still possible.
Only when it becomes possible to maintain the fertility of bull semen for
weeks, instead of 3 or 4 days, or even after the sire which produced the
seman is dead, will physiologists be satisfied with their progress.
The proving and selection of sires for artificial-insemination use is a
major problem. A suggested method for proving sires at a young age
has been proposed. The tendency for fertility level to decline in bulls
more than two or three years of age, and the fact that many proved sires
are not of suitable fertility, are limiting factors. We need more knowl-
edge about the basic reproductive processes before these problems can be
overcome.
The rapid adoption of artificial insemination has resulted in a shortage
of trained technicians. In states where several artificial-breeding asso-
ciations ll,ave been established simultaneously, there has usually been
an unsatisfactory conception rate during the first few months of opera-

ti.on. This is apparently due to the inexperience of both technicians and
herd owners. There is no substitute for insemination experience and
good livestock sense. Many owners had no idea that their animals had
not previously conceived on the first natural service and ,\"ere openly
critical of artificial-insemination conception rates of 50 per cent. Some
owners knowingly submitted shy breeders for insemination in the expec-
tation that it would overcome the difficulties. Others do not yet under-
stand even the elementary principles of reproduction and have not appre-
ciated the necessity of keeping breeding records and studying the estrual
behavior of their cows. These problems are being overcome by the
improvement of courses for training inseminators and providing them
with an opportunity to serve an apprenticeship with an experienced
inseminator. Efforts to increase the knowledge and interest of owners in
breeding management are bearing fruit.
One of the major unsolved problems is that of infertility) This is not
peculiar to artificial insemination but concerns the whole livestock
industry. As discussed in another chapter, the failure of animals to
conceive promptly and produce strong healthy young is of great economic
importance. This is not one disease or condition but a complex including
the whole field of biology. Only by much basic research will the solu-
tions be obtained.
The production of physically and genetically superior offspring is of no
value unless the environment is suitable for the expression of each ani-
mal's capabilities. Cows genetically capable of producing 400 lb. of
butterfat will not do so if managed in the same way as their 200-lb. dams.
One of the most encouraging results of artificial insemination has been
the development of new interests of livestock owners in better manage-
ment, but these interests need much further development. :
Transplantation of Fertilized Ova.-The outstanding s~lCcess of the
transfer of sperm to the female genitalia by instruments rather than by
natural service has led to speculation in regard to the transfer of ova
from one female to another. This idea, although not so old as artificial
insemination, was rather well developed in the latter part of the nine-
teenth century. Between 1880 and 1897, European and British biologists
succeeded not only in the recovery of rabbit and guinea-pig ova but in
their fertilization outside the body and in the transfer of fertilized ova
from one female rabbit to another.
As described in the chapter ou the female, the techniqu~s for the
fertilizati.on of rabbit ova outside the body and their subsequent transfer
to other females where normal embryonic development occurs have been
highly developed. In addition, Pincus has reported the development of
286 BREE DING AND IMPRO YEMENT OF FA R M ANIMALS
the rabbit ovum without stimulation by sperm. As also mentioned,

techniques for superovulation, the production of large numbers of ova,
were described for rats and mice over 20 years ago and have more recently
been developed for cattle and sheep at the University of Wisconsin.
The groundwork has thus been laid for the transfer of fertilized ova
from genetically superior females to ordinary healthy females which would
carryon embryonic and fetal development. In this way, superior females
FIG. 84.-Twin fetuses recovered 46 days after the transfer of fertilized sheep ova to a
second ewe. (Courtesy of Drs. Casida, Warwick, and jl lurphree, University of Wisconsin,
Febmu.r y, 1942.)
would be spared the burden of pregnancy and ·would be free to produce

ova at regular intervals. That this can be done in the laboratory has
been established for nearly 75 years; whether it can be done repeatedly
with the same animals on a practical basis remains to be seen.
Several problems are immediately apparent, and there are undoubtedly
others which are not yet known. One of these involves the number of
ova. Should the ovaries be allowed to produce at a normal rate and
the ova be collected at each heat period? If surgical procedures are
necessary, the risk involved for the recovery of one or two ova per oper-
ation may be excessive. Operative losses, as well as effects on the future
production of ova, must be considered. The production of great numbers
of ova at one time (30 to 40) by superovulation techniques has the advan-
tage of balancing the risks with the possibilities of securing more ova
than would ordinarily be produced in the lifetime of a cow. Two other
problems immediately arise. How frequently can superovulation be
produced in the same animal and will future ovarian function be interfered
,dth? Are superovulated ova capable of normal embryonic and fetal
development? These questions cannot yet be answered in cattle. In
some species the gonads become refractory to repeated treatment with
gonadotropic hormones. The work available thus far indicates that
there is a very high mortality rate in superovulated fertilized ova (Casida
ct al. 1943-1944).
The transfer of recovered fertilized ova to other females may be rela-
tively simple, once sufficient basic information is available, or it may be
very complex. It is possible that the ova may be transferred using the
same techniques that are employed for artificial insemination, or it may
be necessary. to place them directly in the Fallopian tubes. In either
case, it should be possible of accomplishment. The endometrium of the
uterus undergoes almost constant change throughout the estrual cycle.
The preparation of the uterus for implantation and the development of
the fetal and placental membranes require a high degree of coordination
between the endocrine glands, the uterus, and the zygote. It would
seem desirable that the recipient female be in almost the same stage of
the estrual cycle as the donor. This would require the maintenance of
a large number of prospective recipient females. It might be possible
to overcome some of these timing problems by the regulation of the time
of superovulation, by the storage of fertilized ova, and by the hormonal
regulation of the cycles of the recipients.
Summary.-vVe have defined artificial insemination as the deposition
of spermatozoa in the female genitalia by instruments rather than by
natural service. We have tried to emphasize the fact that this method
of breeding is the most valuable tool yet developed for extending the
services of outstanding sires at a cost which even the smallest herd owner
can afford. It is apparent that artificial insemination is here to stay
and that there are no disadvantages which cannot be overcome if rec-
ommended practices are strictly adhered to. It is likewise apparent that,
before artificial insemination can be better than it is, we must expect
to do those things which we think at the present to be best but ready to
adopt n~w practices when they are proved to be better.
References
Books
PERRY, E. J. 1947. "The Artificial Insemination of Farm Animals," Rutgers
University Press, New Brunswick, N.J.
WALTON, A. 1933. "The Technique of Artificial Insemination," Oliver & Boyd,
Ltd., Edinburgh and London.
WEISMAN, A. 1. 1941. "Spermatozoa and Sterility," Paul B. lioeber, Inc., New
York.
Imper. Bur. Anim. Genet. Edinb.
BECK, G. H., and SALISBT.:RY, G. W. 1943. Rapid Mcthods for Estimating the
Quality of Bull Semen, Jour. Dail'y Sci., 26 :483-494.
BERLINER, V. R. 1947. Horses and Jackstock. In PERRY, E. J., "The Artificial
Insemination of Farm Animals," Rutgers University Press, New Brunswick,
N.J.
BURROWS, W. H., and QUINN, J. P. 1935. A Method of Obtaining Spermatozoa
from the Domestic Fowl, Poultry Sci., 14 :251.
CASIDA, L. E., et al. 1943. Effects of Pituitary Gonadotropins on the Ovaries and
the Induction of Superfecundity in Cattle, Amer. Jour. Vet. Res., 4 :76-94.
- - - , WARWICK, E. J., and MEYER, R. K. 1944. Survival of Multiple Pregnan-
cies Induced in the Ewe Following Treatment with Pituitary Gonadotropins,
Jour. Anim. Sci., 3 :22-28.
CHENG, P., and CASIDA, L. E. 1948. Fertility in Rabbit as Affected by the Dilution
of Semen and the Number of Spermatozoa, Soc. Expt. Biol. and IIled. Proe.,
69:36-39.
COMSTOCK, R. E., and GREEN, W. W. 1939. Methods for Semen Evaluation.
1. Density, Respiration, Glycolysis of Semen, Amer. Soc. Anim. Prod. Proc.,
pp. 213-216.
GUNN, R. M. C. 1936. Fertility in Sheep. Artificial Production of Seminal
Ejaculation and the Characters of the Spermatozoa Contained Therein, Australia
Council Sci. & Indus. Res. Bul. 94.
HERMAN, H. A., and RAGSDALE, A. C. 1946. Artificial Insemination of Dairy
Cattle, Mo. Agr. Expt. Sta. Bul. 494.
LASLEY, J. F., EASLEY, G. T., and McKENZIE, F. F. 1942. A Staining Method
for the Differentiation of Live and Dead Spermatozoa. 1. Applicability to the
Staining of Ram Spermatozoa, Anat. Ree., 82 :167-174.
LAWRITSON, M. N., and NIBLER, C. W. 1946. Artificial Insemination of Dairy
Cattle, Nebr. Agr. Col. Ext. Cir. 628.
LEWIS, L. L. 1911. Artificial Insemination, Okla. Agr. Expt. Sta. Bul. 93.
MAYER, D. T. 1948. Personal Communication.
- - - , SqUIRES, D., and BOGART, R. 1947. An Investigation of the Staining
Principle and the Background Stain in the Differentiation of Live from Dead
Spermatozoa, Jour. Anim. Sci., 6:499.
McKENZIE, F. F., and BERLIN~;R, V. R. 1937. The Rcproductive Capacity of
Rams, ""t[o. Agr. Expt. Sta. Rc.s. Bul. 265.
- - - and PHILLIPS, R. W. 1934. Measuring Fertility in the Ram, .Jour. Am. Vet.
Med. 048801' .• 84:189.
MILLER, F. W., and EVANS, E. 1. 1934. Technique for Obtaining Spermatozoa for
Physiological Dairy Studies and Artificial Insemination, Jour. Agr. Res., 48 :941-
MOENCH, G. L., and HOLT, H. 1931. Sperm Morphology in Relation to Fertility,
A mer. Jour. Obst. and Gynec., 22 :199.
PHILLIPS, P. H., and LARDY, H. A. 1940. A Yolk-buffer Pabulum for the Preserva-
tion of Bull Semen, Jour. Dairy Sci., 23 :399-404.
PRINCE, P. W., and ALMQUIST, J. O. 1948. The Effect of Agitation upon the
Livability of Bovine Spermatozoa, Jour. Dairy Sci., 31 :839-844.
SALISBURY, G. W. 1946. Fertility of Bull Semen Diluted at 1:100, Jour. Dairy Sci.,
29 :695-697.
_ and BRATTON, R. W. 1948. Fertility Level of Bull Semen Diluted at 1 :400
with and without Sulfanilamide, Jour. Dairy Sci., 31 :817-822.
_ - - et al., 1943. Rapid Methods for Estimating the Number of Spermatozoa
in Bull Semen, Jour. Dairy Sci., 26 :69-78.
SHAFFNER, C. S., and ANDREWS, F. N. 1943. The Determination of the Con-
centration of Spermatozoa in Fowl and Bull Semen, Anat. Rec., 86:99-107.
WILLIAMS, W. W., and SAVAGE, A. 1927. Methods of Determining the Reproduc-
tive Health and Fertility of Bulls: A Review with Additional Notes, Cornell
Vet., 17 :374.
SECTION III
. : Mechanisms of Heredity·, ~
CHAPTER XI
HISTORY AND PROBLEMS OF GENETICS
Genetics is the science dealing ,yith the similarities and differences

exhibited by related organisms. The science of genetics is one of the
most recent of man's accomplishments, for it came into being with the
rediscovery of the Mendelian principles of inheritance in the year 190Q.
A long series of events, which might be thought of as extending back to
the beginnings of agriculture and the artificial selection of breeding ani-
mals following their domestication, preceded its birth. There are, how-
ever, three very important and definite milestones that mark the progress
of events leading up to genetics. The first was the promulgation of the
cell theory by .SQQ~isl~n and Schwann in the year 1838. Hooke had
noticed and discussed the celluIar structure of cork in 1665, and, though
this in itself led to nothing of moment, it is the historical introduction
of the cell theory. The work of Schleiden with plants and Schwann with
animals indicated that all living forms were built up of smaller units
called cells. The discovery of the physical make-up of living things was
significant. It founded the science of cytology, \vhich after a rather slow
beginning has yielded abundant returns, especially since 1870. The next
milestone was that of the establishment of the evolutionary theory.
As we saw in Chap. III, many men had theorized about an evolutionary
process b'efore Charles Darwin in the nineteenth century marshaled the
evidence that made it seem the most plausible explanation of the pres-
ence of the manifold forms of life which have inhabited or do inhabit the
planet Earth. Likewise, the work of many men since Darwin's time has
added to the plausibility of the theory. This theory ties all living forms
into one organic related whole, with the whole of organic life subject
to the same basic physiological laws and principles of conception, birth,
metabolism, reproduction, inheritance, and death. Evolution postulates
that all forms of life have arisen from other preexisting forms back to
the time when life had its origin from nonliving materials, and that the
rise of new species has been due to variation and selection. Evolution,
therefore, is racial "becoming," whereas genetics is individual "becom-
290
H1S1'ORY AND PROBLEMS OF GENETICS 291
ing," the same general principles underlying both processes. The third
milestone was Mendel's epochal work on .the behavior of characteristics
in transmission, concerning which the next several chapters will deal.
Content of Genetics.-Bateson in 1906 coined the term genetics, which
from its derivation from the -Greek root gen, to become, literally means
, the beginning, or coming into being, of organic life. Genetics, however,
does not stop with parturition or birth, but rather it expresses the idea
of the full becoming of the individuality of the organism. This point
should be borne in mind constantly by the student. • Broadly speaking,
genetics is the study of the why and wherefore of individuality. It is
obvious that the individual will be either like its parents or different from
them. What genetics attempts to do is to systematize knowledge regard-
ing these similarities and differences and to furnish a rational explanation
for them.
There is an old deep-rooted belief to the effect that "like begets like."
Experience tea~hes that sows bring forth pigs not calves, and in this sense
"like begets like," but, when the attempt is made to apply the idea in
, minute details to parent and offspring, difficulties are encountered.
Although there may be, and generally is, a greater or less degree of
similarity between the two, the expectation that parent and offspring
may be exactly alike causes many a disappointment. The fact of the
continuity of the germ plasm would lead one to expect that offspring
would resemble their parents to a greater or lesser degree, and this, in
truth, is just what observation teaches. Any organism receives the
determiners for all its potentialities from its ancestors through its immedi-
ate parents. It is inconceivable that they could come from any other
source, and there is no evidence that any animal can exceed in its accom-
plishments the upper limit set by its inheritance.
The sum total of all the possibilities and accomplishments of an organ-
ism is its individuality. This may appear as the expression of character-
istics exactly similar to those exhibited by any of its ancestors, a blend or
a mosaic of ancestral characteristics, or one or many very different
characteristics due to the interaction of ancestral genes or to mutation.
<Strictly speaking then, it may be seen that an organism's individuality
7is the sum total of all its characteristics which result from the union of
{ germ cells produced by its particular parents plus the influence of the
\., environment in which the individual develops. Genetics seeks the cause
and the pattern of transmission of ancestral traits down through the
generations, to find rational explanations for the occurrence of new traits,
and to differentiate the relat!ve influence of inheritance and environment.
Environment and Training.-Two organisms may have different genes
but be similar in appearance or behavior because of the environment.
292 BREEDING AND IMPROVEMEN7' OF FARM ANI MALS
Likewise two organisms may have identical genes but be different because
of the environment. 'Ve shall lay particular stress upon the complement
of ancestral determiners, or upon an animal's inheritance. The student,
however, must not forget that two other influences aid in shaping individ-
uality, viz., environment and training. Any individuality is a product
of the interactions of these three influences and, although res~ing o!! the
complement of ancestral determiners as its base, this does not in any way
minimize the influence of the other two. Many val'iations are due to
environmental causes, The complement of ancestral determiners, how-
ever, does set the upper limit that an organism can reach. It is a
breeder's province, therefore, to select breeding animals that will set a
high upper limit for their offspring and then so to modify environment
and training that that limit may at least be approximated.
FIG. 85.-A scrub cow before and after being subjected to a good environment. Besides
the increase in condition evidenced in the pictures, she increased in production 2,814.6 lb.
of milk and 113.75 lb. of fat. (From Iowa Expt. Sta. Bul. 188.)
Pre-Mendelian Animal Breeding.-Man has been selecting plants and

animals for an unknown period of time, perhaps 10,000, 25,000, or
50,000 years. His method has always been the very simple one of
mating what he considered desirable animals and basing his hopes on
the general principle that "fubegets like." From an early date his
selections were probably influenced to some extent by a consideration
of pedigree and also to some extent by demonstrated ability to beget
desirable offspring. Up until the very recent past, the mechanisms of
reproduction, hereditary transmission, and variation were totally
unknown. Just by trial and error man made wonderful progress in
creating more beautiful and efficient animal types. The scientific basis
of inheritance has unfolded rapidly since 1900, so that the modern breeder
need no longer be handicapped by a lack of the knowledge of basic
principles. •
It is sometimes disparagingly remarked that man has made no progress
in breeding better livestock for the past 100 years. Since the breeder is
unable, up to the present at least, to create new genes foc desirable
HISTORY AND PROBLEMS OF GENETICS 293
characters, such a statement is tantamount to saying that breeders have

not discarded any undesirable genes or increased the number of desirable
ones, or put them together into any new and better combinations. In
other words, such a remark means that the proportion of bad genes and
good genes is the same in our present-day livestock as it was 100 years
ago. Such a statement can be neither proved nor disproved in terms of
the actual genes, and, because records of average accomplishment are
totally lacking for the old animals and not very numerous for the present
ones, and because nutritional and other environmental conditions have
undergone such revolutionary changes, the statement can neither be
proved nor disproved in terms of records. Crude as may have been our
tools of selection, the writer is of the opinion that considerable progress
has been made. And he is also of the opinion that the average merit
of our animals can now be fairly speedily enhanced, because the basic
principles of hereditary transmission are now known.
In the first chapter we sketched briefly the known history of animal
breeding. Tablets and monuments from early Egyptian and Babylonian
civilizations of 5,000 years ago depict very good types of cattle, sheep,
and dogs, and it is known that these peoples used the ass as a beast of
burden. The ancient Greek and Roman civilizations also had well-
developed arts of agriculture and animal husbandry. During the Middle
Ages there were undoubtedly local varieties of animals of good individual
merit, but this period is noteworthy from our standpoint mainly as the
time when the Arabian horse and the Merino sheep were developed.
When Robert Bakewell started farming at Dishley in 1760, the real
fou~ns of our present-day breeds began to bei"alU,ancrdur~e
next 100 years practically all the breeds of livestock now found in America
were developed. Then, roughly from 1860 to 1900, the breed associations
for pedigreeing, protecting, and promoting the various breeds were
established, the whole of the above process taking place before the prin-
ciples underlying reproductive physiology and hereditary transmission
were known, although_!>~tozot;J.J).ad beeen discovered_by L~.el.rwenhQeJ.<:.
eand Ham in 1677 and ~he ovum by von Baer in 1820. During all this
perlod-me;-had to breed by rule-of-thumb methods, ~d it is likely that
magic and superstition played a large part in breeding operations.
Pre-Mendelian Plant Breeding.I-Conditions were very similar in
the domain of plant breeding, good varieties of plants having existed and
having been continuously improved since the remotest time. ~an,
C~merarius, discovel~~ex~a]_Qatur~_oi plants in 169~, and iD __lIH _
an Englishman, Fairchilcl_, roporto_d_1he fiI:st.l1rtificial.plant hyprid._Jhis
-------
See COOK, R., A Chronology of Genetics, U.S. Dept. Agr. Yearbook, 1937, pp.
1
1457-1477.
was followed by extensive work in crossing plants during the next 50

years, climaxed by Kolreuter's work in Germany in the 1760's. During
the next 100 years, plant hybridizers discovered and reported the occur-
rence of dominance, recessiveness, segregation, hybrid vigor, and unit
characters, but these is~lated facts were not synthesized into a pattern
of principles of inheritance, and the statistical relations indicating the
uniformity of the behavior of the hereditary units were not recognized.
The cell theory was advanced by Schleiden and Schwann in 1838. The
decade 1840-1850 saw the beginning of the progeny testing of plants by
Vilmoriu,-the figuring of the chromosomes by Hofmeister, the enunciation
of the "general principle of the continuity of the germ plasm by Owen, and
the observation of actual fertilization of the egg by the sperm in sea\veed
by Thuret. In 1858, Virchow finally disposed of the old conception of
spontaneous generation by means of his pronouncement, omnia cellula e
cellula (every cell from a cell).
Then within a period of 7 years two of the most famous works of all
time came off the press: Darwin's "Origin of Species" in 1859 and a
paper by Mendel, "Plant Hybridizations," in 1866. The former created
an immediate sensation, the latter, although it contained a clear and
concise treatment of the basic principles of inheritance with practically
perfect statistical results to buttress them, lay unnoticed for 34 years.
During these years great advances were made in the field of cytology.
Hertwig proved in 1875 that fertilization consisted of the union of male
and female pronuclei, which established the equal importance of both
parents in hereditary transmission of potentialities; Flemming reported
the splitting of the chromosomes in 1879; van Beneden the reduction of
the number of chromosomes to one-half in the germ cells in 1883; Hertwig,
Strasburger, Kolliker, and A. Weismann in 1884-1885 independently
and practically simultaneously identified the cell nucleus as the basis of
inheritance; and in 1888 the chromosomes were named by Waldeyer.
In the field of the pr,actical, the progeny test in plant breeding was
further developed, trap-nesting of poultry begun, and the Babcock test
for determining the butterfat percentage of milk discovered. In 1891,
Hays proposed the centgener method of progeny testing as the only
adequate method for determining the genetic make-up and breeding worth
of an individual. Statistical treatment of variation by means of correla-
tion was devised by Galton in 1886, and the standard-deviation technique
for determining the significance of deviations from theoretical perfect~on
was developed by Pearson in 1898.
Breeding Advance since 1900.-With the independent rediscovery of
the basic principles of inheritance by De Vries, Correns, and von Tscher-
mak in 1900 , and the bringing to light of Mendel's original work, a new
impetus was given to both the scientific and practical aspects of plant
and animal breeding. These principles were demonstrated in animals
by Cuenot in 1902 and in man by Davenport in 1904. The interrelations
between the behavior of the chromosomes cytologically and the trans-
mission of characters genetically was shown by Sutton in 1903, and the
_ mechanism of sex determination suggested by McClung in 1902 was
confirmed by Stevens and Wilson in 1905. The term genetics was coined
by Bateson in 1906, and the thegry of the gene was developed by Morgan
and his collaborators at Columbia University in 1910, based upon experi-
m~ts with"the pomace fly, Drosophila melanogaster, which has a genera-
tion in 10 days numbering up to 400. Following the lead of Shull and
East, many inbreeding and crossbreeding experiments were begun both
to test and amplify the genetic theory and to try to develop true-breeding
strains of both plants and animals. A host of discoveries of chromosome
behavior in inheritance have developed since 1910, linkage, crossing
over, the nature and production of polyploids, linear order of the genes,
interference, limitation of the linkage groups, position effects, the function
of inbreeding in reducing heterozygosity, the lethal action of certain
genes, etc., the details of which we will meet in later chapters. Finally
in 1927 came the artificial induction of mutations by X rays (Muller)
and in 1933 the complete unification of cytology and genetics through
Painter's work with the giant chromosomes found in the salivary glands of
larval Drosophila.
The possibilities inherent in the application of genetic principles to
the improvement of both plants and animals have become especially
recognized since about 1920. In plants, many new varieties have been
made to order, and" hybrid corn" is now used extensively throughout the
corn belt. In animals we have again centered our attention on the basic
fundamentals of proving sires by means of their progeny, as Bakewell
did in the eighteenth century and as Varro suggested to Roman farmers
in the first century B.C.
Genetics, or the science of breeding, has had a phenomenal develop-
ment since 1900. The earliest characters studied were the most obvious
external ones, such as color, horns, etc. This led some to expect that all
animal characters would follow some such simple pattern of inheritance,
and as a consequence many unjustifiable claims were made relative to
the easy and rapid revolution in breeding practices soon to be achieved
by means of genetics. The basic principles of inheritance have been
revealed by means of the experimental breeding of small inexpensive
and rapidly breeding forms. The detailed mode of inheritance of most
of the commercially valuable characters of our larger forms of livestock
are still unknown. Our lack of knowledge regarding the mode of inherit-
296 BREEDING AND IMPROVEMENT OF FARM ANI1VIALS
ance of commercially valuable characters doubtless rests in part upon

the fact that each parent produces only a small number of progeny, but
probably of more importance is the fact that such characters are of a
quantitative nature with many pairs of genes being involved in their
inheritance. The specific nature of the inheritance of quantitative
characters is still far from understood even in rapidly reproducing forms
such as Drosophila, mice, and many plants where they have been studied
extensively.
Fortunately, we do not need to wait for the details to be discovered
any more than we needed to wait until the exact nature of electricity
should become kno>vn before putting it to use. Knowledge of the broad I
basic principles, which we now have, ·will expedite breeding progress
tremendously whenever we 4evelop the ingenuity and desire to really
put it to work.
The student should keep in mind during the reading of the next fe,,"
chapters that the facts revealed by means of breeding experiments
furnished the materials on which various portions of the theory of
inheritance are based rather than the reverse of this process. Breeding
facts supply the materials from which a working hypothesis can be
formed. Further experiments are then devised to test the validity of the
hypothesis. Such is the modus operandi of science.
Problems of Genetics.-The principal problem of genetics is of a
threefold nature. Obviously the geneticist is primarily concerned with
the mode of transmission of potentialities from parent or other ancestor to
offspring. The solution of this problem is of scientific interest, but much
more so is it of practical concern. On its solution hinges our ability to'
create plants and animals of a more useful type, more useful in terms of '
efficiency, productivity, longevity, as ·well as those more able to withstand
the onslaughts of drought, disease, and insect pests. With this knowl-
edge, man can, if he wishes, gradually eliminate various sorts of defectives
from his own ranks.
In addition there are two other problems facing geneticists. One of
these has to do with the physicochemical nature of the gene itself, which,
when solved, may throw considerable light on the origin and development
of life in an evolutionary sense. The third problem has to do \vith the
'aevelopmental and functional processes of the organism, the forces and
interactions between the genes, the cytoplasm and the environment
leading to the organization of the material into a specific pattern.
Problem of Physicochemical Cause.-An elaboration upon the physico-
chemical entities that probably function in the transmission of similarities
or the rise of differences belongs in the fields of chemistry and cytology.
Much progress has already been made here, and it is through the cytologi-
12 II 10 9 8 7 6 5 4 3 2
Ib t J : tt tacf,.. •• !
, f. J' f ' .... ~ .. ft
e) It.',. I,_tt.r
.. , •• , , • ,.++1
.... [.,I.tA•• f
h 4 •• f f +I ~.+t
•
; Ie t J , rt ~ tAtt1
FIG. 86.-Eight chromosome groups of 12 chromosomes each of Trimerotrop'is.
Morgan after Carothers, Physical Basis of Heredity, J. B. Lippincott Company.)
(From
cal behavior of chromosomes that the foundation for present-day genetics

has been laid. Research has established the fact that the behavior of
the chromosomes in gametogenesis exactly fits the observed appearance
and distribution of characters in the developed individual. This is an
accomplishment of the greatest moment. There is no doubt but that
the chromosomes constitute the main "bridge of inheritance." The
function of the cytoplasm of the cell is not to be overlooked, but it seems
to he no more than secondary to the nuclear material contained in the
chromosomes. In the case of certain plants, there is genetic evidence

that seems to indicate that certain chlorophyll characteristics of these
plants are controlled by plastids which are distributed in a more or less
haphazard fashion to the daughter cells, but this is not a contradiction
in any sense of Mendelian inheritance, but only an additional type of
inheritance. During the past few years there has been much interest
in the possibility that genes or genelike substances may be able to
duplicate themselves in the cytoplasm and be passed on at cell divi-
sion. Positive proof for the existence of true" cytogenes" has not been
advanced, however.
Many investigations are now in progress in an attempt to explain the
nature of the gene. Genes are probably complex nuclear proteins which
contain several types of nucleic acid. The genes themselves are arranged
in a definite order in morphological structures which we call chromosomes.
The physicochemical processes which are involved when the chromosomes
and genes duplicate themselves prior to and during mitosis are incom-
pletely understood. For our purposes in this text, we can regard these
changes as complex biochemical processes which are carried on by nature
with rather remarkable uniformity. Whereas relatively complex steroid
and carbohydrate compounds have -been synthesized in the laboratory
for many years, even the simplest proteins remain to be produced outside
living organisms.
During th~ast decade, many studies with lower organisms, principally
the mold Neurospora and certain other microorganisms, have revealed
much information in regard to genetic make-up of the organism and
cellular function. For example, mutant strains of Neurospora each failing
to produce a specific enzyme are unable to synthesize certain specific
vitamins of the B-complex or amino acids. Other strains synthesize
these materials readily. It is now known that these functions are
dependent upon the presence of specific genes. The selection of strains
of organisms for the maximum production of important antibiotics such
as penicillin or aureomycin and even the newest of the vitamins, the
animal protein factor, are an outstanding example of the role of genetics
. in human welfare. In each case there is good evidence that the bio-
chemical principles involved are related to specific genes.
A complete delineation of the chemical make-up of genes and chromo-
somes may not aid immediately in any practical manner in creating better
organisms, but the knowledge for its own sake will be very gratifying when
it can be achieved, and future generations may put it to practical use.
Problem of Developmental Organization.-Another great problem is
that of hereditary specificity and developmental organization of the
individual., In other words, what is the mechanism resident in the
\
\, \
fer! ilized ovum "which impels it to follow in each case a definite line of
development"? This problem, unlike that of the nature of the gene
and transmission, which deal with the germ plasm, is concerned with
the somatoplasm. The method of attack here is the embryological one-
not the old descriptive embryology with its manifest limitations in this
respect, but the new experimental embryology, "using the term in its
widest sense to include the study of such phenomena in regeneration,
regulation, growth, etc." This is primarily a study of ontogeny, and it
will eventually fill the void now existing between the original zygote
with its content of hereditary determiners and the fully developed adult
exhibiting the characteristics for which determiners were present in the
zygote.
A great deal of intricate experimental embryological research is
.t .,
beginning to shed light on this problem. Transplantation and regenera-
Jt. ~) &, Jl )) ~'- II CC: "

"'.e,,., Ie cc .t" eccc ,en
FIG. 87.-The 24 pairs of chromosomes in human male.
'1
(P. Popenoe, The Child's Heredity,
Williams & Wilkins Company.)
tion techniques have shown that various parts of the body act as organiza-
tion centers, elaborating hormonallike products which prOflide the pat-
terns for further development. Belly tissue of the young frog embryo,
which, if left in place, would have produced ~kin, will, if transplanted
early enough to the region of the jaw, produce a normal frog jaw. If,
however, it is transplanted to the jaw region of a salamander, it will as
formerly produce jaw tissue, but it will be the jaw of a frog because
of the frog genes that it contained. Development, therefore, is due
primarily to the genes and perhaps secondarily to the interaction of the
developing tissues with other nearby tissues, which, of course, have been
gene-determined. Also the amputation of half a limb of a salamander
will be followed by a complete regeneration of a normal limb, bone as
well as muscle and nerves, which indicates a substance in the cells of the
stump capable of directing the normal development of the tissues that
spring from it.
Problem of Mode of Transmission.-There are two methods of study-
ing the manner in which characteristics are transmitted from one genera-
tion to the next. They are (1) by studying the ancestors of certain
individuals or (2) by studying the progeny of certain individuals. The
former, or ancestor study, may be carried on by means of biometry or
by means of data recorded in purebred herd books. The progeny study
is the experimental or Mendelian method, in which the matings are

planned, and the study may have a statistical basis. Inquiry will now
be made into these methods of genetic research, bearing in mind that
in this field, particularly, environmental conditions must be carefully
controlled so that the observed results may justifiably be attributed to
the make-up of the germ plasm.
The statistical method of attacking the problem was originated by
Sir Francis Galton, later developed by Karl Pearson, and is known today
as biometry.
It is a most useful tool in the hands of the analytical geneticist. Its
function is to measure the degree of resemblance between related indi-
viduals and to express the result in numbers. As pointed out by Pearl,
the weakness of the method as a means of research lies in two facts:
(1) it is largely a method of description and does not attempt to give the
cause of the events, and (2) because biometry gives the degree of resem-
blance between parent and offspring, and because heredity is a cause
of the resemblance, the biometrical method (ilorrelation) measures the
degree or intensity of inheritance. As Pearl points out, heredity is not
the -sole cause that can lead statistically to a significant correlation
between parent and offspring. "Anything whatsoever which tends to
bring about local group differentiation within the sample included in
the table will tend to produce the same result, altogether independent of
any genetic relationship or the absence of it."
With its limitations and dangers kept in mind, biometry, properly
used, is a valuable agent in the study of genetics.
Herd books and production records of livestock may be made to
yield valuable information on the problem of hereditary transmission.
With the aid of such records, the manner of transmission of superficial
characters, such as color markings, set of ears, and so on, as well as the
deeper lying functional activities, such as speed, milk production, etc.,
may be investigated. This method can be criticized as being descriptive
rather than experimental, and in it there is also the danger that an
investigator may not have access to all the relevant facts. These records
are now beginning to yield information that should prove of great value
to practical breeders by explaining how certain traits are inherited and
possibly also by showing the falsity of timeworn beliefs.
The two preceding methods have dealt with the ancestors. The fol-
lowing will be concerned with progeny. This is the experimental-
breeding or pedigree-culture method, which Mendel used so successfully.
He succeeded in discovering the basic plan of hereditary transmission,
because he reduced his problem to its simplest form and studied one pair
of contrasted characters at a time. Like the Galtonian, this method is
also statistical. In general, it consists of producing pure races, crossing

them, mating the hybrids, and ascertaining the ratios in the F2 and later
generations. This method has also resulted in the establishment of
Johannsen's pure-line theory and De Vries's mutation theory. The
former of these has been defined as "all the progeny of a single self-
fertilized individual." Such individuals eventually give rise to inbredk
strains ~hat ar~ homozygous ~en~tically but with fluctuating varia~ions
in each mbred hne. These vanatIOns are probably due largely to envIron-
mental conditions and not to germinal variations. In other words, this
points to the static character of the germ plasm, or its conservative
nature.
The mutation theory suggested by De Vries, on the other hand, postu-
lates sudden heritable changes in the genes. This provides one means
of heritable variation and points to the dynamic character of the germ
plasm, or its progressive nature.
Finally the organism itself will often provide a partial index to its
genetic make-up. If a character behaves as a recessive, we know the
genetic make-up of an animal evidencing this character (for this char-
acter). This is true also in a case where dominance is lacking, for the
homozygous dominant (AA), the heterozygote (Aa) , and the homozygous
recessive (aa) will all evidence different expressions of the character.
If the character shows complete dominance, however, we will be unable.
to distinguish between AA and Aa individuals.
In animal breeding, our aim is the creation of desirable and efficient
animals that will of necessity transmit their own desirable qualities to a
high percentage of their offspring. We want all the evidence that we
can get regarding the likelihood that our desirable animals will be good
transmitters. Summarizing the above discussion we can get this informa-
tion from
the individual itself
its progeny its ancestors
Genetics and the Practice of Breeding.-The practice of breeding is

older than history. Its method is selection, and its accomplishments
legion. It has produced fast, fancy, and farm horses; increased the
yield of milk from a few hundred to 41,000 lb. annually; and has produced
the food that has enabled man literally to conquer his world. Genetics
acknowledges gladly the achievements of empiricism and seeks the reason
"why." The method of practical breeders has always been that of
trial and error, and of necessity there has always been a goodly share of
the latter. The greatest service that genetics can render to the practice
---- -
of breeding lies in its ability to analyze and to provide logical explana-
.. .~- .. ,~-- ._"'-
_E.Q._ns. This will inevitably result in the sifting of the true from the false,
a service of no mean value and one which is being fulfilled rapidly. If
genetics did no more than unearth the principles underlying breeding
and post" no-passing" signs at the head of all the blind alleys, it would
expedite successful breeding to a marked degree.
However, knmvledge gained through genetic studies has already
provided new methods that will lead more quickly and surely toward
the goal of more beautiful form and more efficient function. Detailed
knowledge of the mode of inheritance of most of the characteristics
exhibited by higher animals is still unknown and from the very nature
of the problem is difficult, though genetics bids fair to furnish the solu-
tion. When this has come about and the principles are thoroughly
understood, genetics will have proved itself merely the handmaid of
animal breeding, to which it will have rendered a valuable service in
successfully and firmly fitting into place the keystone of all breeding
operations, "rational selection."
To be successful in breeding better livestock, a person should first of
all love to work with animals. Next comes the necessity for a thorough
knowledge of the history, accomplishments, and shortcomings of one's
particular breed. Practice in livestock judging bulwarked by a knowl-
edge of anatomy and physiology and by experience gained through study
of finished carcasses in meats courses should also prove to be very useful.
Finally, courses in veterinary hygiene and science, together with the
materials presented in the books dealing with animal breeding, should
complete one's initial training for animal-breeding work with actual
experience in the field serving as a continuing source of enlightenment
and inspiration.
Summary.-This chapter serves as a general introduction for our
consideration of genetics, which, as we have seen, is concerned with
similarities and differences exhibited by related organisms. We have
laid particular emphasis on the chromosomes as "the bridge of inherit-
ance," but we should be careful not to overlook the large part played
by both environment and training in bringing any inherited character
to its full fruition. We traced very hurriedly the long history of both
plant and animal breeding and saw how the occasionally observed facts
of inheritance down through the ages were finally fitted together into a
complete and complex whole by Mendel and his successors. The three
principal types of problems that genetics and related sciences are trying
to solve were indicated as well as a preview of the methods now being
used to effect solutions. Finally, we listed practical prerequisites for
the scientific animal breeder and indicated the role of genetics in the
practical &,eld of animal husbandry.
References
DEMEREC, M. 1947. "Advances in Genetics," Academic Press, Inc., New York.
MORGAN, T. H. 1934. "Embryology and Genetics," Columbia University Press,
New York.
NEEDHAM, J., and PAGEL, W. 1938. "Background to Modern Science," Chaps. 9-
10, The Macmillan Company, New York.
PEARL, R. 1915. "Modes of Research in Genetics," The Macmillan Company,
New York.
RILEY, H. P. 1948. "Genetics and Cytogenetics," John Wiley & Sons, Inc.,
New York.
SCHWESINGER, G. C. 1933. "Heredity and Environment," The Macmillan Com-
pany, New York.
SHARP, LESTER W. 1934. "Introduction to Cytology," 3d ed., The McGraw-Hill
Book Company, Inc., New York. .
U.S. Dept. Agr. Yearbook. 1936, 1937. Especially last chapter, 1937, by R. Cook.
WADDINGTON, C. H. 1941. "Organizers and Genes," Cambridge University Press,
New York.
WEISS, P. 1939. "Principles of Development," Henry Holt and Company, Inc.,
New York.
CHAPTER XII
THE PRINCIPLES OF HEREDITY
Men have been breeding plants and animals for many thousands of
years. In the earliest stages of breeding, the main purpose was probably
that of seizing upon favorable variations in order to try to improve both
plant and animal species. By allowing these favorable variates to have
many offspring and less desirable ones to have few or none, a slow, steady
improvement ensued. The question was, "how" can we secure better
plants and animals to serve our human needs?
Early man, like children of today, was probably randomly curious.
He wondered in a vague way about the stars, his own whence and
whither, etc. The early Egyptians and, later, the Greeks began to
develop a system of experimental observation and inquiry, but later
events put a stop to these beginnings. Up until about the year 1500,
authoritarianism (believing what others had said) and the fear of religious
persecution succeeded in quite thoroughly throttling any intelligent
curiosity. Gradually these shackles were unloosed, and man began to
add experimentation to speculation. This was the beginning of the age
of enlightenment. Copernicus, Brahe, Kepler, and Galileo soon proved
through mathematical calculations and observation that the earth, far
from being the center of the universe, was in fact but one of several
relatively tiny specks which revolved around a rather puny star, our sun.
Newton soon combined the mathematical physics of Descartes and
Galileo's law of falling bodies into the gravitational theory of an orderly
and immense universe in which nothing is left to chance or caprice.
These discoveries, deductions, and syntheses finally proved that old
authorities were not infallible, that observation and common sense had
their limitations, because things were not always what they seemed to be
and, most important of all, man's own reason backed by observation,
calculation, and experiment was the most trustworthy guide available.
Eventually this intellectual curiosity was directed toward the world of
living things. Man began to experiment to try to find out "why" the
results of certain matings yielded certain results. This was especially
true in the plant kingdom, in which many men carried on plant hybridiza-
tion experiments in attempts to fathom nature's secrets in the trans-
mission of characters from parent to offspring. In 1694, Camerarius
304
THE PRINCIPLES OF HEREDITY 305
published a " 'ork which showed that plants were sexual organisms. In
1717 occurred the production of the first artificial plant hybrid by
Thomas Fairchild. The work of Spallanzani in the eighteenth and
Pasteur in the nineteenth century proved the old idea of spontaneous
generation to be a fallacy; Schleiden and Schwann discovered the cellular
nature of plants and animals; and Darwin established the validity of the
process of organic evolution.
Finally, breeding experiments with peas carried on from 1857 to 1865
by an Austrian monk, Johann Gregor Mendel, revealed some of the basic
FIG. 88.- Gregor Mendel, the discoverer of the laws u nderlying inheritancQ.
principles of hereditary transmission. There had been many other plant

hybridizers before Mendel, but he was the first to attack the problem
on a large enough scale and in such a thoroughgoing way that it became
possible (1) to determine the number of different forms under which thf'
offspring of hybrids appeared, (2) to arrange these forms with certainty
according to their separate generations, and (3) to ascertain definitely
their statistical relations. In 1866, he published a report of 8 years'
hybridization experiments, though 34 years elapsed before the validity
of his pioneer work was established. Mendel, like many other men who
have ~hought ahead of their time, died before the results of his splendid
pioneering were appreciated.
Like many other natural phenomena, this matter of inheritance yielded
to a fairly simple interpretation, and Mendel's success was due in large
measure to the fact that he reduced his problem to its very simplest form
by studying one pair of contrasted characters at a time, coupled with the
fact that he happened to work with characters which were determined by
genes carried in different chromosomes and which therefore assorted
independently. Mendel's discovery was duplicated in 1900 by three
botanists, DeVries, Correns, and von Tschermak, and this led eventually
to the discovery of Mendel's original paper.
Mendel's work laid the basis for the later establishment of two of the
general laws of inheritance, viz., those of segregation and of independent
assor,tment. To these have been added, since 1900, the following:
linkage, crossing over, linear order, interference, and limitation of the
linkage groups. Some of these later principles are in a sense antithetical
to the original Mendelian principle of independent assortment, but for\.
the sake of clarity and unity all the phenomena of inheritance, based on
the reac~~s_.2f_~, are generally known under the collect~-te~
Mendelism.
MENDELIAN TERMS
The following are some of the terms now in general use in referring to
Mendelian phenomena. - - - .
Unit Characters. Any characters of any organism that behave as a
unit in inheritance-plant size in peas, horns in cattle, etc.
Genes. The units of inheritance that probably react together and with
the cytoplasm, arid thesetwo with the environment to make patent the
organism's latent potentialities.
Allele*. Mendelian characters are inherited in alternative pairs (or
series). These alterItative forms ()La gene, which are located at the
same point on each one of a pair of chromosomes, are called alleles, e.g.,
albinism (recessive), normal pigmentation (dominant); horns (recessive),
hornlessness (dominant). (Allelomorph, allelic, adjective-variants.)
Dominant*. A character, possessed by one of the parents of a hybrid,
which is manifested in the hybrid to the apparent exclusion of the con-
trasted character from the other parent (the recessiveV-Thus in a cross
of green- and yellow-seeded peas the first generation has yellow seeds.
Yellow is dominant and green is recessive, being transmitted bur'not
appearing in the presence of the factor for yellow.
Hybrid. The individual that arises from crossing parents which are
pure for certain different characteristics. Crossing tall (TT) on dwarf
(tt) peas produces a hybrid (Tt).
kJ onohybrid. A hybrid that is heterozygous for one pair of allelic
genes.
* Definitions marked with an asterisk taken from "Glossary of Genetic Terms,"
,-
U.S. Dept. AgT. Yearbook, 1936, pp. 153-164.
• \
Dihybrid. A hybrid that is heterozygous for two pairs of allelic genes.
Trihybrid. A hybrid that is heterozy-gous for three pairs of allelic
genes.
Phenotype. The expressed character (or sum of all the characters) of
an organism; e.g., if, in peas, genes for both the tall and the dwarf char-
acteristics (T and t) are present, the pea grows tall, and therefore belongs
to the tall phenotype.
Genotype. What an organislll--.a,_ctllally is as determined by its germ
lliasm; e._.g., TT and Tt are hoth tall p~as, therefore, the same phenotype,
but because of their different genotypic constitutions will transmit
differently and therefore belong to different genotypes. 'v/
Heterozygote*. An organism to which its two parents have contributed
unlike genes with rei;pect to any given allelic pair gove;iling contrasted
characters, and which in turn produces two kinds of germ cells with
respect to the character. (Heterozygous, adjective.)
H omozygote*. An organism whose parents contributed to it a similar
member of any given pair of genes, and whose germ cells are therefore
all alike with respect to the genes for that character. (Homozygous,
adjective.)
Race*. A group of individuals having certain characteristics in com-
mon because of common ancestry-generally a subdivision of a species.
Gamete. Jr reproductive cell of either sex; a sperm or ovum.
"" ';/[{eredity*. The resemblance, derived from the ancestry, among
organisms related by descent.
P. Parental generation.
Fl. First hybrid generation-made by crossing P X P.
F 2 • Second hybrid generation-made by crossing two FI hybrids.
Variation*. In biology, the occurrence of diffeyences among the
individuals of the same species or variety.
Haploid. Single; referring to the reduced number of chromosomes in
the mature germ cells of unisexual organisms (n).
Diploid*. Having two sets of chromosomes. Body tissues of higher
plants and animals are ordinarily diploid in chromosome constitution (2n).
Polyploid*. Normal body cells of the higher plants and animals have
I two sets of chromosomes. Polyploids are forms having three or more
of these basic chromosome sets (3 or more n). ,>.- ' ' I JI/ "'" I ,;',... -0
Heteroploid*. An organism characterized by a chromosome nuinber

that is not a multiple of the basic haploid number.
~hysical Basis of Inheritance.-In Chap. III we presented a short
discussion of chromosomes and genes, and in Chap IV and V we outlined
the mechanisms involved in the production of sperm and ova and the
behavior of the chromosomes during these processes:
The chromosomes are sometimes called the "bridge of inheritance"

since they serve as the sole connecting link betweeneacht\vo~"succeeding
generations. Th~ ultimate indJ~isible units of inheritance li,!"e thj;l_gen~~
While genes areillilts,-they-ao-not"benave as such because "many of them
are joined together in a linear series and this group of genes is called a
chromosome. The chromosomes consist of a definite number for each
species of plant and animal.
Any individual is duplex in its chromosome make-up (except for the
sex chromosome in the male in most higher organisms). The chromo-
somes are paired in each body cell, one me er of each air havin n
supplied by each parent. When an organ' m produces germ cells, how-
ever, only one member of each pair is found in each resulting gamete.
In this fashion, the number of chromosomes is reduced to one-half in
each gamete, and the normal sp~ies number of chromosomes is again
restored when an egg and sperm unite at fertilization.
The Essence of Mendelism.-One of Mendel's experiments involved
the crossing of tall and dwarf peas, and he found that all the offspring
grew tall, because in this species tall growth is dominant over dwarf.
When these FI hybrids were crossed, the resulting F2 generation con-
sisted of tall peas and dwarf peas in the ratio of 3: 1. The following
diagram should make this clear:
TT tt
pure tall parent pure dwarf parent
T t "'
pollen after reduction
Tt F,
hybrid (tall)
ovules after reduction /
5either ovule fertilizedt '
TxtI I
pollen of hybrid after reduction 1by either pollen S/ :
'I
T
TT Tt t
---------~,,~----------
tall
Tt
ovules of hybrid after reduction
tt
dwarf
3 1
The TT peas are pure for tallness and will produce all tall offspring.
The same holds true for the tt peas in respect to dwarfness. Tho two
Tt peas are phenotypically tall, but carry the factor for dwarfness t as a
recessive. Two of these crossed would again give:
Txt
I I
T t
TT, Tt, Tt, tt, etc., ad infinitum
This 3; 1 ratio is phenotypic, the genotypic ratio being 1; 2; 1. This

was the gist of Mendel's work; viz., starting with parents pure for some
contrasted character, hybrids would be produced in the F 1 that looked
like the dominant parent; and, in the F 2, )i of the population would be
pure for one grandparental characteristic, )i pure for the other grand-
parental characteristic, and Y2 would carry both factors and resemble
the dominant parent.
Mendel's Hybridization Experiments.-It is of importance to know
how far Mendel's results and later experiments meet mathematical
expectations. For this reason it will be well to consider both Mendel's
original work and some later experiments. The point at issue is whether1
Mendel discovered an interesting restricted phenomenon in peas or a -
universal law of inheritance.
Mendel secured some 34 more or less distinct varieties of peas and,
before beginning his own experiments, subjected them to a 2-year trial to
ascertain whether or not they would breed true to type. Knowing that
some characters of plants are transmitted unchanged to the hybrids and
their progeny and that one of each pair of contrasting characters of two
plants appeared in the'hybrid as the dominant character, whereas the
oth"er (recessive) reappeared in the progeny of the hybrid, Mendel planned
his' experiment to observe and record the proportion of each pair of
differentiating characters in the succeeding generations of hybrids and
to deduce the law according to which they appear in the successive
generations.
Mendel selected the following seven pairs of contrasted characters for
study;l
1. The difference in the form of the ripe seeds. These are either round or
roundish, 2 the depressions, if any, occur on the surface, being always only shallow;
or they are irregularly angular and deeply wrinkled.'
2. The difference in the colour of the cotyledons. The albumen of the ripe
seeds is either pale yellow, bright yellow, and orange coloured, or it possesses 11
more or less intense green tint.
3. The difference in the colour of the seed coat. This is either white, with
which character white flowers are constantly correlated; or it is grey, grey brown,
leather brown, with or without violet spotting, in which case the colour of the
standards is violet, that of the wings purple, and the stem of the axils of the
leaves is of a reddish tint. The grey seed coats become dark brown in boiling
water.
4. The difference in the form of the ripe pods. These are either simply inflated,
1 CASTLE, W. E., "Genetics and Eugenics," p. 317, Harvard University Press,
Cambridge, Mass., 1916.

2 In each case the dominant character is italicized.
310 BREEDING AND IMPROVEjyIENT OF FARM ANIMALS
not contracted in places; or they are deeply constricted between the seeds and
more or less wrinkled.
5. The difference in the colour of the unripe pods. They are either light to
dark green or vividly yellow, in which colouring the stalks, leaf veins and calyx
participate.
6. The difference in the position of .the flowers. They are either axial, that
is, distributed along the main stem; or they are terminal, that is, bunched at the
top of the stem and arranged almost in a false umbel; in this case the upper part
of the stem is more or less widened in section.
7. The difference in the length of the stem. The length of the stem is very
various in some forms; it is, however, a constant character for each, in so far
that ·healthy plants, grown in the same soil, are only subject to unimportant
variations in this character.
In experiments with this character, in order to be able to discriminate with
certainty, the long axis of 6 to 7 feet is always crossed with the short one of % to
IH feet.
The results of segregation in the F 2 in those seven series of experiments
by Mendel are shown in Table 18.
Mendel was dealing with strictly contrasting characters and got no
intermediate or blending forms. The matter of personal bias was there-
fore obviated. The phenotypic ratio that Mendel secured in the F2
'with each pair of contrasted characters was in no case significantly
different from 3: 1. Many other investigations of single contrasting
characters have yielded a similar 3: 1 phenotypic ratio in the F 2.
Mendel carried some of the descendants of the above hybrids along
through four to seven generations and reported that" no departure from
the rule has been perceptible."
TABLE 18.-SUMMARY OF MENDEL'S EXPERIMENTS WITH PEAS
" ,:
Numberi Character contrast
1 Form of seed ........... 7,324 5,474 1,850 2.96:1.00

2 Color of cotyledons ...... 8,023 6,022 2,001 3.01:1. 00
3 Color of seed coats ...... 929 705 224 3.15:1.00
4 Form of pod ............ 1,181 882 299 2.95:1.00
5 Color of pod ............ 580 428 152 2.82:1.00
6 Position of flowers ....... 858 651 207 3.14:1.00
7 Length of stem .......... 1,064 787 277 2.84:1.00
Totals ........................ 19,959 14,949 I 5,010 12.99:1.00

He also made crosses of individuals differing simultaneously in two
and in three characteristics and found the same underlying principle
THE PRL'iCIPLES UF HEREDITY 311
active; viz., the FI hybrids looked like the sum of the dominant char-
acters put in by both parents, and in the F 2 all the characteristics were ~
recombined into all the possible combinations.
The Mendelian Theory of Inheritance.-The essence of Mendel's
work is summed up in this statement l of Mendel's.
The results of the previously described experiments led to further experi-
ments, the results of which appear fitted to afford some conclusions as regards
the composition of the egg and pollen cells of hybrids. An important clue is
afforded in Pisum by the circumstance that among the progeny of the hybrids
constant forms appear, and that this occurs, too, in respect of all combinations
of the associated characters. So far as experience goes, we find it in every case
confirmed that constant progeny can only be formed when the egg cells and the
fertilizing pollen are of like character, so that both are provided with the material
for creating quite similar individuals, as is the case with the normal fertilization
of pure species. We must, therefore, regard it as certain that exactly similar
factors must be at work also in the production of the constant forms in the hybrid
plants. Since the various constant forms are produced in one plant, or even in
one flower of a plant, the conclusion appears logical that in the ovaries of the
hybrids there are formed as many sorts of egg cells, and in the anthers as many
sorts of pollen cells as there are possible constant combination forms, and that
these egg and pollen cells agree in their internal composition with those of the
separate forms.
In point of fact, it is possible to demonstrate theoretically that this hypothesis
would fully suffice to account for the development of the hybrids in the separate
generations, if we might at the same time assume that the various kinds of egg
and pollen cells were formed in the hybrids on the average in equal numbers.
Mendel's results were published in 1866, and his paper was discovered
by De Vries, a Dutch botanist, in running down the literature on plant
hybridization following his own experiments with the evening primrose,
from which work he formulated his theory of mutations and his law of
"the splitting of hybrids." Apparently Correns in Germany and von
Tschermak in Austria discovered the latter principle at about the same
time.
The principles evidenced by the work of :Mendel, De Vries, Correns,
and von Tschermak have since been named segregation and independent
assortment.
l. Segregation. A new individual arises from the union of a male
and a female gamete that contain a sample half of the complete genetic
material of each parent, i.e., one or the other member of each pair of
chromosomes. Each parent contributes a complete ha loid set of chro-
~:.::.:.os:.:o:m:::::e.:::.s_t:::.o:_:e::::a~ch.:::....:::o::..__::l.::;s:::....;o=s;:J.p:.:.r,",m:..:.gb!-' When t IS new individual matures
1 CASTLE, op. cit., p. 332.
312 BREEDING AND IMPROVEMENT OF FARM ANIJ"ALS
and begins to produce germ cells, the latter go through the process of
reduction, at which time the chromosomes put in by the parents segregate
or separate out from each other, one member of each pair going to each
of the secondary germ cellR, in all the possible combinations. This
principle is illustrated in the tall- and dwarf-pea cross, the pollen or
FIG. 89.-Diagram showing consequences of independent segregation of chromosomes in

Drosophila ampelophila. (From Babcock and Clausen, Genetics in Relation to Aoriculture.)
ovules produced in the F 1 hybrid containing a chromosome in which is

located either the gene for tallness or the gene for dwarfness.
2. Independent Assortment. The chromosomes of the hybrid assort
themselves at reduction into all possible combinations, each germ cell. J
necessarily having one member of each pair of chromosomes. With·
one pair of chromosomes, two genetically different germ cells are produced
THE PRINCIPLES OF HEREDI1T 313
(21) ; with two pairs of chromosomes, four genetically different germ cells
are produced (22); with three pairs, eight geneticaIly different germ cells
are produced (2 3); etc. Chromosomes and genes were unknown in
Mendel's day. He did not speak of them but rather in terms of the
characters of his plants: Singe, however, all of the characters in peas
with which Mendel ~worked were apparently determined by factors
(genes) located in different chromosome pairs and no two of them by
genes in the same pair of chromosomes, Mendel was able to postUlate
the most important principle of inheritance, independent assortment,
which actually applies to chromosomes and not, as Mendel perhaps
thought, to characters.
Segregation and independent assortment constitute two of the basic
principles of inheritance. For independent assortment or distribution of
characters, the pairs of contrasting characters considered must be deter-
mined by genes carried in different chromosomes. If the two or more
pairs of contrasted characters were determined by' genes carried in the
same chromosome, they would not be expected to assort independently
but rather to remain linked together. Linkage relations wiIl be discussed
later.
Monohybrid Crosses.-A monohybrid is a hybrid that is heterozygous
for one pair of allelic gen~s. When we cross a homozygous tall parent ~
(TT) \vith a homozygous dwarf (tt) we get a monohybrid (Tt). If t\yO
of these monohybrids are crossed, we get a ratio of 3 taIl: 1 d,mrf off-
spring. For an explanation of these breeding facts we assume that one
parent has the genes for tall growth and the other parent the genes for
dwarf growth at similar loci in homologous chromosomes, and that these
genes are both present in the monohybrid but separate or segregate when
the monohybrid produces germ cells with the haploid number of chromo-
somes. Remember, we have never seen these genes, and, if we could see
them, they would have no labels. We get these certain results from
certain crosses, and, knowing the general behavior of the chromosomes
at gametogenesis, we use the simplest theory possible to explain the
breeding results, in this case, one pair of genes.
Polled and horned cattle might be substituted for tall and dwarf
peas and the problem worked out in the same way, the polled being
dominant to the horned condition.
~ce between ~ is not a universal phenomenon; e.g.,
in a cross between a red and a white Shorthorn the Flare all roan.
red RR x rr white
germ cell R r germ cell
F ,\--------. Rr+---I a roan pattern (mixture of red
and white hairs)
314 BREEDING AND IMPROVEMEl,,'T OF FAKlf ANIMALS
If we mate roans, we will get
Sperm
Eggs
R r
--
R RRred Rr roan
I-----
r Rr roan rr white
or a phenotypic ratio of 1 red: 2 roans: 1 white, instead of the 3: 1 pheno-

typic ratio that is evidenced when dominance between the genes is
present. This 1: 2: 1 phenotypic ratio is also seen to be the genotypic
ratio: in other words, we can tell by the appearance of these animals
what their genotypes for color are. Lack of dominance is, therefore!l,
seen to be an advantage, and dominance a disadvantage from a practical;
breeding standpoint, because the latter permits two or more genotypes to\
be masked under one phenotype. ... •
Dihybrid Crosses.-A dihybrid is one that is heteroz ous for two ,
pairs of allelic genes. As an examp e, we can take the case of a cross
between a homozygous smooth, black guinea pig (rr BB) 1 and a homozy-
gous rough, white (RR bb), which would give the dihybrid heterozygous
rough, black (rough and black behaving as dominants).
smooth, black rr BB X RR bbrough, white
germ cells rB Rb germ cells
F, dihybri~r B~eterozygouS rough black /
Mating two of these dihybrids together would mean that any egg
could be fertilized by any sperm. These possibilities are best shown by
means of the following checkerboard arrangement (Fig. 90).
When two of these F 1 dihybrids produce germ cells, spermatozoa
or ova as the case may be, each germ cell will have the chromosome
with the gene for either roughness or smoothness, R or f, and the chro-
mosome with the gene for either black coat or white coat, B or b, because,
being located in different chromosomes, the two pairs of genes assort
independently. All the possible combinations of the four genes repre-
sented in the hybrid would be expected, therefore, both in the spermatozoa
produced by a male as well as the ova produced by a female.
1 The notation used in this book is based on using the capitalized first letter of the
dominant character to represent the dominant gene and the corresponding small letter
to represent the recessive gene. In this case rough and black are dominant, there-
{ore RR (or Rr) = rough; rr = smooth; BB (or Bb) = black; bb = white.
THE PRINCIPLES OF HEREDl7'Y 315
RrBb X RrBb
Ova
Sperms
RB Rb rB rb
RB RRBB RRBb RrBB RrBb

Rb RRBb RRbb RrBb Rr bb
rB RrBB RrBb TrEE rr Bb
rb RrBb Rr bb rr Bb rr bb
FrG. 90.-Checkerboard of the F2 of a dihybrid cross of smooth black X rough white

guinea pig.
In the above checkerboard the presence of at least one ll:lcrge R deter-

mines roughness, at least one large B determines blackness, otherwise
the animal will be smooth and white. _Segregation is an extremely
important feature of inheritance, because through it something entirely
new may be created. Starting with smooth, black and rough, white
guinea pigs, something new, a smooth, white individual was created.
Figure 90 indicates that the following results have been secured.
9 rough blacks-at least one R and one B.
3 rough whites-at least one R and two b's.
3 smooth blacks-two r's and at least one B. q: 3: 3 '.1
1 smooth white-two r's and two b's.
This is the usual dihybrid ratio, although there are several variations
of it due to lack of dominance in one or both pairs of genes or to various
forms of epistasis.
Similarly, if we mated a homozygous polled, black bovine to a horned,
red one, we would get the results shown below.
polled, black PP BB X pp bb horned, red
germ cells PB pb
Fl dihybrid ""Pp Bb./ heterozygous polled, black
Experience has shown that_t_lt~~.20n_9.ition_is dominant over the

_lt2Illild_l<ondition, ang._plack ~Qlor QY£l.!,_L~<i insattle.
The genes for polled or horned and for black or red are apparently
located in different chromosomes, because, when we mate dihybrids
Pp Bb, we get a ratio of
9 polled blacks.
3 polled reds.
3 horned blacks.
1 horned red.
316 BREEDING AND IMPROVEMENT OF FL1ll~}f ANIMALS
indicating that the germ cells containing genes PB, Pb, pB, and pb were
produced in equal numbers by all these dihybrids.
The nine polled, black individuals are of the same phenotype but the
nine are made up of
IPP BB
2PPBb
2PpBB
4PpBb
or four different genotypes in the same phenotype, and the only way to
ascertain the different members of these four genotypes is through breed-
ing trials. This brings to our attention a very important practical con-
sideration, viz., that we can't always tell by the appearance or perform-
ance of an animal, its phenotype, how it will transmit. The PP BB
animal is polled and black and all of its offspring will be polled and black,
because it must transmit both gene P and gene B to all its get. But the
Pp Bb animals are also polled and black, but these animals may transmit
the combinations of PB, Pb, pB, or pb genes to their offspring. One of
the latter animals (Pp Bb) if mated to horned, red animals will get off-
spring as indicated below;
PB I Pb I
I pB pb
pb Pp Bb polled, Pp bb polled, pp Bb horned, pp bb horned,

black red black red
whereas a homozygous polled, black animal, PP BB (germ cells all P B),

if mated to horned, red animals will get all polled, black offspring, Pp Bb.
The way in which an animal will transmit depends on its genotype,
which mayor may not corre~ond with its phenotype. The later chap-
ters of this book on selection in horses, cattle, sheep, and swine will be
devoted quite largely to a consideration of the procedures necessary to
learn the genotypic ma}).e-up of our breeding animals.
A situation similar to. the two dihybrid examples described above pre-
vails in the inheritance Jlif comb shape in poultry, but here the two pairs
of factors influence only one characteristic instead of two. In this case
RR pp birds are rose-combed, rr PP birds are pea-combed, and, when
these are crossed, giving Rr Pp, the combs are walnut-shaped. Crossing
two walnuts, Rr Pp, gives 9 walnuts, 3 rose, 3 p.".!te./.~ and I single comb,
rr pp. -It
Trihybrid Crosses.-A trihybrid is one that ~ heterozygous for three
pairs of alleli~ genes. The reader can now figure out for himself the
THB PRINCIPLBS OF HBRBDIT}- 317
~
Rose
RR pp
•
\, , ~
\7.\
Walnut
F. Rr Pp
e
~ /r!{6
w
6 e.+ s d'
'~-'® ® ® B
Iii' ~'------r----r--~-----.
@~ RR PP
~ ~ ~
RR Pp Rr PP Rr Pp
Walnut Walnut Walnut Walnut
@ ~ ~ ~ ~
RR Pp RR££ RrPp Rrpp
Walnut Rose Walnut Rose
F2
® ~ Rr pp
~ ~ ~Rr Pp rr PP rr Pp
Walnut Wolnut P~a -Pea
G) ir\ ~ ~ ~
Rr Pp Rr~ !.r Pp rr pp
Walnut Rose Pea Single
FIG. 91.-Diagram showing interaction of factors for comb form in fowls. The cross of a
pure rose-comb bird with a pure pea-comb one gives all walnut-combed offspring. The 16
possible combinations of the FI gametes, with their genotypes and the phenotypes resulting
from factor interaction, are shown in the F2 checkerboard. (From Sinnott and Dunn.)
318 BREEDING AJVD IMPROVEME1VT OF FARM ANIAfALS
results to be expected from crossing the trihybrids secured from mating a

rough, black, short-haired guinea pig = RR BB SS
to a
smooth, white, long-haired guinea pig = rr bb 88
giving the trihybrid rough, black, short Rr Bb Ss
remembering that these trihybrids can secrete the following sorts of
germ cells (for these genes are in different chromosomes): RBS, RBs,
RbS, Rb8, rBS, rBs, rbS, rbs.
,.. ~
~,..
,.,.,.,,.,,.,,.,,.,,.,,.,
'7 II !!, ,
~
II 3 J 3 I
FIG. 92.-Diagram showing inheritance of polled. white face. and black body characteristics
and their allelomorphs in an Angus-Hereford cross through F2.
A similar problem can be worked out in cattle involving the crossing

of Angus and Herefords.
Angus (polled,l black body, black
face) = PP BB ww
Hereford (horned, red body, white
face) = ppbb WW
The monohybrid cross = 3: 1 ratio _ f
The dihybrid cross = (3: 1) 2 ratio, or g: 3: 3: 1
The trihybrid cross = (3:1)3 ratio, or 27:9:9:9:3:3:3:1
Multifactor Crosses.-Little and Phillips report an experiment involv-
ing four pairs of independently Mendelizing factors in mice. They
crossed a wild male of the genetic constitution AA BB DD PP (A for
1 In crosses between polled and horned breeds of cattle the offspring are not by
any means all polled, even though it is true that the latter condition is due to a domi-
nant gene and in spite of the fact that the polled parent is homozygous. Some scurred
and horned animals usually appear. This can be explained by postulating a gene
for scurs (Sc), which is epistatic to polled (P) in either heterozygous (Sc 8C) or homo-
zygous (Sc Sc) males, but which is epistatic only in females that are homozygous
(Sc Se). A similar epistatic relation apparently holds for a special horn gene (Ha),
distinct from the gene (H), carried by all horned breeds, and found principally in
certain horned breeds ofAfrica. (Kans. Agr. Expt. Sta. Bien. Rept. p. 87, 1936.)
agouti pattern, B for black coat, D for intensity factor, and P a factor
for eye co!oration) on a pink-eyed, dilute, brown
female of the genetic constitution aa bb dd pp. ~~~:8~:8~:2
The observed ratios from such a cross in the F 2 ilil""ilil""
~ l:i: " "i:l: " "
were in all cases veW close to expectations. The 1>.1>.1>."'-""""
chromosomal interpretation of the segregation and A.""'' ' ' ' ' """,- "'-
recombination of factors in this cross is the same
1'1:1.01'1:1.01'1:1.01'1:1.0
as was found in the mono-, di-, and trihybrids. tq 1'1:1"1"1"1"1><\"1"1
In the case of a monohybrid, 4 individuals are Ii.
" ;l";l;l"ililil
i:i: il "i:i: il "
11."",,,,,,,,,,,,,,-,,,,,,,,,-
""'-"'-"'-"'-"'-"'-"'-
secured in the F2 checkerboard; a dihybrid gives
16; a trihybrid, 64. One involving four pairs of
contrasted characters W 0 u I d fur n ish 256 ~:g~:2~:2~::&
individuals in the F2; and, if ten characters were 1:1: " ili:i: il ;l
considered simultaneously, 1,048,576 individuals ~ l:i:l:i:l:l:l:l:
"'-"'-"'-"'-"'-"'''''''-
Y\'Quld have to be considered in the F2 generation. ""'' ' ' ' ' ' ' '1>.'' -' '-' '-
It can readily be seen that the problem is becom-
"1.0 'GoO 'G.oCl:).o
ing unwieldy. For this reason, very few crosses 'G'G'G'G'G""l'G"'<
involving as many as four pairs of contrasted ~ i:1: il "i:!l:: il il
1:1: I: l:l:!l::l:l:
"'-
charac~ers have been worked out. ~R.~~';l..~~~
Most of the commercially important char- ""'' ' ' ' ' ' ' ' "''''''''''-
acters of our farm mammals do not behave in
inheritan'ce as though they were determined ~::8~~~:gi=Q;g
by only one, two, or three pairs of genes, but

.0
~~§§~~§§
rather as though they were determined by the "'"
" ll..tt.o..Cl,,,,,,,-,,,-,,,-
combined interaction of dozens or scores of pairs ""''''''''''''''''''''''''''''''''''''

of genes, probably scattered through all the I'I:1.oCl:).o""l.oI'l:1.o
tq "'<><\tq"'<'G><\><\"'<
chromosomes. A different technique must, there- il;lil,,;J;JOlS
fore, be employed when dealing with characters
tt. " l:!l:: il ,,!l::!l::;l S
determined by many genes. Even though a ~~~~~~~~

character was determined by as few as three genes,
we have seen that in the F2 we could get eight ~:g~~~;8~;8
different genotypes. ~ l:!l:: s sl:!l:: il il
l:!l::!l::!l::i:::!l::!l::!l::
Il.,
~~~~~~~~
27 with at least one A, one B, and one C.
9 with at least one A, one B, and c's. "'<",,""l.oI'l:1.o~.o
"'<"1><\"1"1><\><\"'<
9 with at least one A, b's, and one C.
9 with a's, one* B, and one* C. ~ i:::!l::" ;Ji:!l::;J S
i:l:l:l:!l::l:l:l:
A.
3 with at least one A, b's, and c's. 2:2:2:2:.t.t.t~
3 with a's, one* B, and c's.
3 with a's, b's, and one* C. ;:t
o ~~~~~~~.o
1l,1l,1l.,<l, "''''-'''-~
1 with a's, b's, and c's.
* At lea~t.
/
This is a total of 64 individuals in the F2 and represents 27 different

genotypes. Only one of these genotypes (the last one, involving all small
letters), can be recognized on sight, so that breeding tests would have to
be instituted to reveal the genotypic differences between AA BB CC and
Aa BB Cc individuals, etc. So the fact that there apparently are so few
unit characters (determined by one or a few pairs of genes only) in our
farm animals is no great loss or hardship, for it is not practical to deal
with more than one or two pairs of genes at a time, especially in species
which have relatively as few offspring as our farm animals and where the
TABLE 19.-RELATION BETWEEN THE NUMBER OF GENE PAIRS INVOLVED IN A CROSS
AND THE NUMBER 0]<' PHENOTYPIC AND GENOTYPIC CLASSES IN THE F2
Number of visi- ~umber of dif- Number of pos-

Number of Number of geno-
bly different F2 ferent kinds of sible combina-
gene pairs typically differ-
classes if domi- gametes formed tions of FI
involved ent combinations
nance is complete by FI gametes
~ -
1 2 2 3 4
2 4 4 9 16
3 8 8 I 27 64
4 16 16 81 256
n 2" 2" 3" 4"
proper combinations of several genes might by chance never be secured

in the small numbers available.
The above table should help to make this point clear.
It must be remembered that, in order to secure independent segrega-?
tion in all the types of hybrids studied above, it is necessary that the'
genes be located in different chromosomes. This of necessity limits, in-
any species, the number of pairs of contrasted characters that can show I
independent assortment to the number of pairs of chromosomes found in

that species. Drosophila melanogaster, for example, with four pairs of
chromosomes could not manifest more than four pairs of independently
segregating characters at anyone time. No cases have been discovered
where there are more independently segregating pairs of contrasted char-
acters than there are pairs of chromosomes. If, in Drosophila, five pairs
of contrasted characters were considered at one time, it would, of neces-
sity, mean that at least two of these pairs of characters were determined
by factors carried in the same chromosome. Such a condition is known
as linkage and will be discussed later.
MODIFIED TWO-GENE RATIOS-DOMINANCE LACKING
We have seen above that, if two genes which determine different char-
acters are located in different chromm;omes and one of the members of
each pair of genes is dominant over its allele, we will secure a 9: 3 : 3 : 1

ratio when we cross two Fl dihybrids to get the F2 generation, and that
the same ratio holds if the two genes control only one character (comb
pattern in poultry).
However, if one pair of genes lacks dominance, we would not get a
9: 3: 3: 1 but a 3: 6: 3: 1 : 2 : 1 ratio, because the first phenotype containing
9 individuals breaks up into phenotypes containing 3 and 6 individuals
each; whereas the third phenotype containing 3 individuals breaks up
into two phenotypes containing 1 and 2 individuals each. The following
figure will be useful in explaining all of the various modifications of the
9:3:3:1 ratio.
Aa Bb X Aa Bb
Sperm
Eggs _.
AB Ab aB ab
AB AA BB(l) AA Bb(5) Aa BB(9) Aa Bb(13)

Ab AA Bb(2) AA bb(G) Aa Bb(lO) Aa bb(14)
aB AaBB(3) Aa Bb(7) aa BB(ll) aa Bb(15)
ab Aa Bb(4) Aa bb(8) aa Bb(12) aa bb(16)
:FIG. 94.-Generalized checkerboard of a two-gene cross.
If we crossed a polled,l red animal, AA BB, with a horned, white one,

. aa bb, the Fl would be polled and roan,2 Aa Bb, assuming gene A for
polled to be dominant over gene a for horns, but gene B for red not to be
dominant over gene b for white, the presence of both Band b giving the
roal! color. Then we would have in Fig. 94:
Squares (1), (3), and (9) ....................... '. 3 polled, red

Squares (2), (4), (5), (7), (10), and (13) ......... " 6 polled, roan
Squares (6), (8), and (14) ...................... " 3 polled, white
Square (11) .................................... 1 horned, red
Squares (12) and (15). . . .................. " 2 horned, roan
Square (16)..... . . . . . .................. " 1 horned, white
If dominance were lacking in both pairs of genes, as it is in a cross

between a red-flowered broad-leaved (AA BB) and a white-flowered
narrow-leaved (aa bb) snapdragon, for the Fl is pink-flowered with leaves
of intermediate width, then the F2 would comprise AA red, Aa pink,
1 We are departing from our usual notation in order to be able to use one figure of
reference for all the types of variation due to lack of dominance and epistasis.
2 Dominance between red and white color lacking.
322 BREEDING AND IMPROVEMENT OF FAR21{ ANIMALS
aa white, BB broad, Bb intermediate, bb narrow (see :Fig. 94), and we

would have in the F2
1 red, broad (square 1)

2 pink, broad (squares 3 and 9)
2 red, intermediate (squares 2 and 5)
4 pink, intermediate (squares 4, 7, 10, 13)
1 red, narrow (square 6)
2 pink, narrow (squares 8 and 14)
1 white, broad (square 11)
2 white, intermediate (squares 12, 15) -
1 white, narrow (square 16)
or a 1 :2:2:4: 1 :2: 1 :2: 1 ratio.

Breeding results with both plants and animals have revealed several
characters thatbehave in inheritance as though they were controlled by
two pairs of genes but do not follow the usual pattern as evidenced in
comb type in poultry described earlier in this chapter where the mating
of rose-combed birds (RR pp) with pea-combed birds (rr PP) gave an Fl
of all walnut combs (Rr Pp) and an F 2 of 9 walnuts, 3 rose, 3 pea, and
1 single comb. In order to explain these deviating results, the principle
of epistasis has been invoked. Again the student is reminded that the
modus operandi of genetics is to get the results first by breeding plants
and animals, and, after the results are secured, to devise a theory
which will provide a rational and logical interpretation for the breeding
results.
Epistasis is the same relation betw~_e_n pairs of different genes that
cio]llinance IS bet\v_e~p.- different single genes of one pair of alleles, i.e.,
gen';A-~ay -be-d~minant over its allele a,or, in an epistatic sense, gene A
may be dominant over genes B or b (dominant epistasis), or genes aa
may be dominant over genes B or b (recessive epistasis), or both these
conditions may hold at the same time, dominant and recessive epistasis.
We may further find cases where duplicate dominant, duplicate recessive,
or incompletely duplicate epistasis may prevail.
We will presently describe six relatively simple forms of epistatic
behavior. The student should realize, however, that there is an almost
limitless number of possible interactions between two, three, or more pairs
of the several thousand genes which are the hereditary capital of our farl1l
animals. Careful studies by a host of experimenters have delineated the
broad, general principles of inheritance. One of these general principles
is call~tasis, but the number of possible epistatic effects is legion.
\\
\
Some would even go so far as to call all nonadditive factor interaction

epistatic.
Recessive Epistasis.-If we mate a black rat, AA BB (gene A for color,
genes aa diluted color, gene B for expression of aey-~olor, genes bb mask-
ing all color, i.e., epistatic to A) to an albino rat, aa bb, all of the Fl are
black, Aa Bb, whereas the F2 will appear as 9 blacks, 3 creams, and 4
whites. This is due to the fact that the presence of at least one A and
one B produces black (squares 1, 2, 3, 4, 5, 7, 9, 10, and 13, Fig. 94), a's
and at least one B produce cream (squares 11, 12, and 15, Fig. 94),
whereas either AA, Aa, or aa, together with b's (squares 6, 8, 14, and 16,
Fig. 94), produce albino. This is because the b's mask the expression
of A or a; i.e., b is epistatic to A or a, so that two portions of the usual
9: 3: 3: 1 ratio are thrown together phenotypically.
Incompletely Duplicate Epistasis.-Another manifestation of epistasis
is one that throws the two middle portions of the 9: 3: 3: 1 ratio together
phenotypically. An example of this occurs in I:?~E2.c-}~rsey swine, where
one sandy-colored race is AA bb and another aa BB. When crossed the
Fl are all red, Aa Bb, and in the F2 all the offspring, which get at least
one A and one B (squares 1, 2, 3, 4, 5, 7, 9, 10, and 13, Fig. 94), are red,
those getting neither A nor B (square 16) are white, and all the others,
getting at least one A with b's or at least one B with a's, are sandy
(squares 11, 12, 15, 6, 8, and 14, Fig. 94). These are sometimes called
mutually supplementary genes. A similar case in plants has been found
in the fruit shape of squash, where genes A and B together give disk
shape, genes AA or Aa with bb and genes aa with BB or Bb give spherical
shape, and the double recessive genes give elongated shape.
Dominant Epistasis.-It has been found that a certain type of white
dogs, AA BB (with color in the eyes), if mated to brown dogs, aa bb,
produce all white offspring, and that, when these Flare mated, the pecu-
liar ratio of 12: 3: 1 appears in F 2. This apparently is due to the whit(.\
dog carrying color factors (A) that are inhibited from forming color in
the hair by another gene (B). The brown dogs lack the gene for black
color A, being therefore brown in color, for they also lack the inhibitor B.
The F 1 cross is Aa Bb and is white because of the dominant epistatic
effect of gene B over gene A or a, whereas in the F2 all those squares that
get at least one B (1, 2,3, 4, 5, 7, 9, 10, 11, 12, 13, and 15, Fig. 94), are
,vhite, those getting b's and at least one A (squares 6, 8, and 14, Fig. 94)
being black, and the remaining (square 16), aa bb, being brown. Genes
that behave in this fashion are often called inhibitors.
Other Types of Epistasis.-Further types of epistasis that the student
can work out by means of Fig. 94 are:
Duplicate Recessive Epistasis. aa epistatic to BB, Bb, or bb.

bb epistatic to AA, Aa, or aa.
Sweet peas: white pea AA bb X white pea aa BB.
F 1 Aa Bb, all purple.
F 2 9 purple: 7 ,,,hite.
Homo sapiens: deaf person AA bb X deaf person aa BB.
F 1 Aa Bb, aU normal hearing.
F2 9 normal hearing: 7 deaf.
Dupliyate Dominant Epistasis. A epistatic to BB, Bb, or bb.
B epistatic to AA, Aa or aa.
Poultry: feathered shanks AA BB X clean shanks aa bb.
FJ Aa Bb, all feathered shanks.
F2 15 feathered shanks: 1 clean shanks.
Shepherd's-purse (Bursa): triangular seed pod AA BB X top-shaped seed
pod aa bb.
F 1 Aa Bb, all triangular seed pods.
F2 15 triangular seed pod: 1 top-shaped seed pod
Dominant and Recessive Epistasis. aa epistatic to BB, Bb, or bb.
B epistatic to AA, Aa, or aa.
Poultry: White Leghorn AA BB X White Wyandotte aa bb.
FI Aa Bb, all white.
F 2 13 white: 3 colored birds.
Summary on Epistasis.-We ,have ~eeJJ. fro!l!.the al;love e_)(amples tha~

the usual two-gene phenotypic ratio of 9: 3:3:1 may be modified in., _!!.~
variety of ,vays by dom.inance between individual genes in an allelic pai(j
arid l:iY·vari~us epistatic manifestations b~t~e~npairs ~Fgenes.!
- ThelolTo';'~inidi~gram brings all the 'above facts into one complete ·whole.
The complications introduced into breeding practice are illustrated in
the following figure. In column I, 'we see that all the 9 genotypes found
in an F 2 of a cross involving two pairs of genes stand out clearly from each
other (AA BB, AA Bb, AA bb, Aa BB, Aa Bb, Aa, bb, aa BB, aa Bb, and
aa bb). Because dominance is lacking, AA, Aa, and aa combinations are
different, and the same is true of BB, Bb, and bb combinations. In other
words, the phenotypes and the genotypes are the same, or the breeder
knows the genotype when he sees the phenotype. Dominance makes its
appearance in one pair of genes (A) in column II with the result that
AA BB and Aa BB can no longer be separated by phenotypic inspection
but must be subjected to breeding trials to ascertain their genetic make-up.
Similarly, AA Bb and Aa Bb are now indistinguishable phenotypically,
etc. The 9 genotypes of column I have been reduced to 6 phenotypes in
column II and are further reduced to 4 in column III. Likewise further
reduction of the 9 original phenotypes of eolumn I to various combina~

tions of 3 and 2 genotypes in columns IV to IX are indicated. Domi-
nance and epistasis complicate breeding practice, because they hide tW6
or several genotypes in one phenotype.
It should be remembered also that we are here dealing with a compara-
tively simple thing, because it involves only two pairs of genes. If three
pairs of genes were acting on the same character and there was no domi-
nance, we would get a phenotypic ratio of 27:9:9:9:3:3:3:1 in the F 2 •
It is not difficult to imagine the complications that dominance and epis-
tasis would introduce into this situation, and it appears likely that the
Gena-
\;/pes II ID. v IX
AA8B '____ Classical
________ 3~ t~ene
AaBB 2 ~ioF2
2 _ 6 ____- - - - - - 9 - 9 - 9 " ' " 9~
AA6b
AoBb 4~ "",2 15
AAbb
Aobb
/J--3--~~~~-.
GlaBB 1---1----3~ 3
aaBb 2 - - 2 ----- ------4
1---1---1""/-----1---1'--------
Dom. Dam. Dom. Ree. Inc. Dam. Dup. Dup. Dom.Glnd
one
neither bath Epis. Oup. [pis. Rec. Dam. Rec.
pair pooir pooir [pis. [pis. Epis. Epis.
genes qenes genes
FIG. 95.-Phenotypic classes in two-gene F" due to dominance and epistasis.
commercially valuable characters of our larger farm animals were influ-

enced by dozens or hundreds of genes.
Two things are apparent: (1) the scientific livestock breeders need a
thorough knowledge of the basic principles of hereditary transmission as
worked out with small laboratory animals; and (2) the breeder of the
larger animals finds little opportunity to apply this detailed knowledge
to his own immediate problems but is forced to devise his own methods of
selection and improvement on the basis of the broad principles that have
been ascertained through the breeding of laboratory animals.
".,.-
/' Analysis of Coat Color in Mice. 1
. A thorough study of the variations in a group of related characters in any
organism will usually reveal an intricate series of interactions between the com-
1 SINNO'rT, EDMUND W., DUNN, L. C., and DOBZHANSKY, TH., "Principles of
Genetics," 4th ed., pp. 11'1-115, McGraw-Hill Book Company, Inc., ::'{ew York, 1950.
,") \-
ponent genes, As an example we shall choose the house mouse, for in this
animal a large number of spontaneous variations have provided the opportunity
for a genetic analysis of .t_h~.._geI!efl~ affecting coat.. color. Many such genes
'have been Stuilled~ and-their interrelationshlpsuinade out:-· C is the fundamental
<;9Jm gene, necessary, for the production of any pigment in the coat.·· Another
gene, A or gray; determines the development of the agouti patt{lrn. Its reces-
sive allcle,-a -is present in the nonagouti mice, such as blacks or browns. Still
another, B;-governs the development6rblack·pigment and is dominant over its
allelic condition of brown or chocolate, b. Many varieties are spotted with white
in a blotched or _piebald pattern, and such mice contain a gene, 8, which is
recessive to self or solid color, S. Another gene, d, brings about a clumping
of the black and brown pigment granules in the hairs and makes these colors
appear faded or dilute, as opposed to the normal fully pigmented form, D.
Another gene reduces the amount of black and brown pigment in the fur, giving
it a pale and washed-out appearance, and also reduces the pigment in the iris,
making the eyes appear reddish or pink like the eyes of albinos. The gene,
which is called pink eye (p) from its most noticeable effect, is recessive to the
normal dark-eyed intense-colored condition, P. These genes all segregate
sharply and may occur in any combination. There are also several other genes
affecting coat color which will be omitted for the sake of simplicity. Some of
these combinations result in characters which are distinctive and have been
given names of their own. Thus the nonblack agoutis are called" cinnamon"
or brown agouti; the dilute blacks, "blue" j the dilute browns, "silver fawn" j
and so on. Table 20 lists these various gene combinations, together with the
type of coat color produced by each.
All of these types are recessive to the wild coat and appear to have arisen
from it by mutation of one or more genes. Thus at any time the wild type may
be reconstituted by bringing into combination all of the alleles of the genes
which are responsible for these new types. In fact, the wild coat color itself is
found to depend on the presence and interaction of all of the genes named.
Thus, in order to produce the agouti pattern, there must be present the genes for
color (C), agouti (A), black (B), dark eye (P), dense color (D), and solid color (S).
With regard only to these genes the genotype of the wild mouse may be written
AA BB CC DD PP SS. These genes all show essentially complete dominance,
so that their heterozygous condition will give the same result as is produced by'
the homozygous form here given. Thus an animal with the genotype Aa Bb Cc
Dd Pp S8 would also be agouti in appearance. The genes named do not include
all that are known, nor is it believable that more than a small sample of the
genes affeeting coat color in mice have been studied. Were knowledge eom-
plete, it is probable that the list of genes neeessary for the production of the
agouti pattern would be much longer and that the letters of the alphabet would
be exhausted in attempting to write the genotype of the wild mouse. Here,
then, is a clear and -convincing example of gene interaction. In order that the
apparently simple pattern characteristic of wild house mice may be developed,
there must be present at least six genes (probably many more) each of which has
TABLE 20.-INTERACTION OF GENES ~'OR COAT COLOR IN MICE
Gametic
Genes Phenotype
formula
P CABDPS Wild-type agouti

D
-Ss CABDPs Spotted agouti
- -
p CABDpS Pink-eyed agouti
B
-Ss CABDps Pink-eyed, spotted agouti
- P S CABdPS Dilute agouti
-
I d
-p 8 CABdPs Spotted, dilute agouti
S CABdpS Pink-eyed, dilute agouti
-
A s CABdps Pink-eyed, spotted, dilute agouti
- - -
p S CAbDPS Cinnamon
D
- 8 CAbDPs Spotted cinnamon
- -
p S CAbDpS Pink-eyed cinnamon
-
b s CAbDps Pink-eyed, spotted cinnamon
-
P S CAbdPS Dilute cinnamon
d
-p
-
-
S
s CAbdPs
CAbdpS
Spotted, dilute cinnamon
Pink-eyed, dilute cinnamon
.
-
s CAbdps Pink-eyed, spotted, dilute cinn.am~n
C ._- - - -
P S CaBDPS Black
-
D s CaBDPs Spotted black
-
p S CaBDpS Pink-eyed black I
- Pink-eyed, spotted black !
B s CaBDps .;
- - - .i
P S CaBdPS Dilute black
-
d s CaBdPs Spotted, dilute black
-p S GaBdpS Pink-eyed, dilute black
,.
-
a s CaBdps Pink-eyed, spotted, dilute bla.ek
- - - - i
P S CabDPS Brown
D
-s CabDPs Spotted brown
- -
p S CabDpS Pink-eyed brown
-
b s CabDps Pink-eyed, spotted brown
- -
P S CaMPS Dilute brown
-
d 8 GabdPs Spotted, dilute brown
-p S CabdpS Pink-eyed, dilute brown
-
8 Cabdps Pink-eyed, spotted, dilute brown
c with any other c Albino, may be of any of 32 genotypes above
genes
328 BREEDING AND IMPROVEMENT OF PARM AVIMALS
a definite effect on coat color. If any single gene is missing or changed, a cout
pattern differing more or less widely from the wild type results. ., '
This type of gene interaction is not exceptional but is found whenever numerous
variations in- a single aspect of the organism--are carefully analyzed. Such
analyses have been made for several groups of characters in maize. More than a
dozen series of genes affecting chlorophyll development in this plant are known.
The normal color is green, which results from the combined action of all of the
normal genes. If one of these genes such as TV mutates to a recessive allele w, the
seedling is albinotic, virtually without any chlorophyll, and being unable to carry
on photosynthesis it soon dies. There are at least 15 different genes susceptible
to this type of change, which means that there are at least 15 genetically different
types of albinos. In another group of recessive mutants known as lutescents only
yellow pigment develops; in another group are more than 20 recessive genes each
of which when homozygous produces virescent seedlings which are albinotic but
eventually develop enough chlorophyll to keep them alive. Other recessive
genes are responsible for pale green color (10 known), zebra-striping (4 known),
piebald spotting (4 known), golden color (4 known), yellow-green color (3 known),
yellow striping (2 known), fine white striping (3 known), and many other modifi-
cations of chlorophyll composition and arrangement. It must be true therefore
that the normal development of chlorophyll in maize depends on interaction
among at least 75 different genes. If anyone of these changes by mutation from
the normal to the recessive allele, some essential step in the interaction fails and i
the chlorophyll is absent or deficient in some way. - Similarly each of another .
large group of genes conditions some step in the development of anthocyanin
pigments, while the normal starchy endosperm of the kernel of field corn
(dent or flint) depends on the interaction of more than 30 different· genes. The
results of the analysis of such cases as those described above suggest that many
of the characters of organisms are the end products of long chainlike series of
related steps, a - t b - t c _ d - n. Separate genes seem to affect separate steps
so that if the gene affecting step b does not perform its task, then c and all later
steps which depend upon it cannot take place and the character, such as normal
chlorophyll, cannot appear. Other evidence for this view of the mechanism of
gene interaction will be presented in a later chapter. Whatever the means .by
,ichJhe_gjlli~in_t(ll~~be accept(l~_ as a ~~n._«:!al rule that the h~~editary
~arac~ers.Qt~ pla?_~Ul!!ll.~~epend~alan~d co~p.eLati(m of a large
~IllbllI_~~ -_- --~-
Ibsen has made an extended study of the inheritance of coat colors in

cattle. 1 It was found necessary to postulate about 25 genes to account
for the novv known color patterns in our domestic cattle. Thi_xample
serves to show the extreme complexity of the problem of inheritance in
our larger animals. It is undoubtedly even more complex in regard to
their commercially valuable qualities. Who would dare even to guess
1 IJlSEN, A. H. L., Cattle Inheritance (Color), Genettcs, 18 :441-480, September, 1933.
, .
how many genes may be involved in the final expression of the amount of
milk a cow may yield to say nothing of the environmental influences
involved in such a matter? We can be fairly sure of the fact that many,
many genes are involved in determining the characteristics of our live-
stock, and we can be very sure that they behave in ways comparable to
those used as illustrations of different types of inheritance in this and the
following few chapters of this book. It may be that we will never know
the exact number and interrelations of the genes involved in milk pro-
duction. Nevertheless, it should be possible to improve our breeding
practices on the basis of the general principles involved in the hereditary
trarismission of characteristics.
Aspects of Dominance.-In the early part of the twentieth century,
soon after the rediscovery of Mendel's laws of inheritance, there was a
tendency toward oversimplification of the theory. That is to say, there
was a tendency to think of each potentiality as determined by one gene
only. We are now a,vare of the fact that, rather than one gene's deter-
mining a potentiality, many or all the genes (together with the environ-
ment) interact to bring the potentiality to full fruition, though, of course,
one gene may still determine the appearance or nonappearance of a
definite potentiality.
In like manner the idea of dominance held full sway in the early days
of genetics. We now recognize that dominance is a relatiye m:ther than
an absolute term. For example, brown eyes in 'man are dominant over
blue, but th~ 'blue-eyed condition is in turn dominant over red or pink -----_
~,-"-,-
eyes (albinism). Whether blue eyes are dominant or recessive depends

therefore on what the characteristic is to which they are being contrasted.
There are varying degrees of dominance, from complete dominance to an
entire lack of dominance~--The possibility also exists that the phenom-
enon of "overdominance" in which the heterozygote is superior to either
homozygotemayoccur. , /
The results from crossing tall and dwarf peas; smooth, black and rough,
white guinea pigs i or polled, black-bodied black-faced and horned, red-
bodied white-faced cattle were clear-cut because in every instance one
character ,was wholly dominant over its allele. If, however, a cross is
made between a red Shorthorn and a white one, the result is a roan
(mixture of red and white hairs) color pattern. Likewise a cross between
~'black fowl and a white one gives the typical Blue Andalusian pattern, I
and a cross between red- and ,,,hite-flowered four-o'clocks gives a pink

hybrid. In these cases, dominance is lacl0ng, and hybrids of this sort
never breed true to their own phenotypic appearance because their
phenotypic character is due to the interaction of t,,·o genes, whe!eas they
can tl'am;mi~ 9nly one geneor the other'loany one offsprirt~
The hyt>.D~~_C?!_er is differeIltf.r()l.!l~g,p.~r of the pa~ents;_j.e., red

crossed' with white Shorthorn yields roan, and apparently the final
expression 'o£this characteristic hinges onOlle pair of genes. Similarly in
poultry, p~a-shaped comb and rose-shaped comb are each individually
dominant'..over a single comb, w'hereas a cross between homozygous
rose- and pea-combed birds yields a new comb character, viz., \~
To explain these facts t.yo sets ot' alleles must be postulated and the
genetic constitution of a single-combed bIrd becomes rr pp, whereas a
pure rose is RR pp and a pure pea rr PP. These genes (R and P) are
evidently located on different chromosomes, are therefore inherited ~E)p_a
rately,. or. according to the law of independent assortment, but act
together in producing the final expression of the comb characteristic. If
we cross pea-combed and rose-combed birds, the F 1 will all be walnut-
combed and the F2 will give the usual 9:3 :3: 1 hybrid ratio, (walnut, rose,
pea, single).
, Dominance may at times be quite variable, depending upon the pres-
\ ence or absence of ce.~enes, and, finally, a character-
istic may act as a dominant in one sex (baldness in man) and as a recessive
in the ot.her. It is seen, therefore,llmt·the idea of dominance is not th~
simple thing it was at one time supposed to be.
Occurrences such as the roan pattern in Shorthorn cattle was formerly
often spoken of as blending'7nheritance, though the term went out of use
as the discrete, ato~ic nature of the genes became probable. The fact
that, when large races of animals are crossed with small ones, offspring
result which are intermediate in size induces some still to speak of these
offspring as being a blend of the two parent races. In terms of chromo-
somes and genes, of course, they are, but it is now fairly apparent that
most of the commercially valuable characters of our animals -are d~ter
mined by the interaction of scores or hundreds of genes rather than by i
one or two pairs. In this sense, such crossing does result from a blending ,
of many genes, although the genes themselves remain discrete bodies_
It is fortunate that inheritance is not blending in nature, for if each
individual were an average of its two parents, the entire population would
be almost completely uniform in a few generations. Do~inance gen-
erally behaves in the same manner in closely related species, and in most
mammals the albino (recessive) gene is the same. However,' a charac-
teristic _that.. is. domillant in one species.msy be recessive in another;-e.g.;
~l:ui9minant over white in guinea pigs, but ~-pigs we have
learned by~ience that whIte IS doIDinant over black. -"'TIle'expres-
sion of dominance may also be i~flW3IiCed'fJYtIle- enVIronment, age, sex,
and by other genes in the organism. In short, domin:a.nce is variable
with many gradations. _-
,..-. ...,,"
THE PRINCIPLHS OF HEREDITY 331
._)q
The nature of the recessive condition has also been a much-debated

question. Some authorities claimed that the recessive gene was a definite
entity, others that the recessives indicated losses of genes. T. H. Morgan
has adduced evidence from several sources which indicates that recessives
are not due'tp losses of genes; e.g., the fact that albino rats always have
a few colored. hairs on the feet, that the wings of fruit flies attain con-
siderable length if the temperature is held at 31°C. even though the flies
be genetically pure recessive vestigials, and that small eyes appear in old
cultures of Drosophila that are pure eyeless recessives. He also cites the
fact that certain recessive mutations have been known to revert to the
original dominant condition and finally that there are more than a dozen
known mutants of eye color in Drosophila at the same locus. From this
evidence, it would seem that recessives are caused by the action of definite
entities (genes) just as are dominant characteristics and that the idea of
considering recessives as losses of genes is an untenable one.
The student should try to appreciate fully that the facts of breeding
came first, theories to explain them second. Genetics is but a systematic
arrangement of breeding facts together with theories that offer a logical
explanation. Any theory must do two things: (1) explain past facts and
(2) allow us to prognosticate. All the theories of genetics to be discussed
later are backed by a multitude of fact. As soon as a new fact is discov-
ered which cannot be fitted into the presently accepted theory, that theory
will have to be supplanted by one which wiTIaccommodate all the old
plus the new facts. This revision of theory has been necessary many
times in the past short life of genetics.
Summary,-In this chapter we have become acquainted with some of
the simpler but basic principles of heredity. We have learned that
heredity is referable to paired, particulate bodies called genes, located in
the chromosomes. We have seen that these genes retain their identity
at all times whether expressed phenotypically or not. Weare now
familiar with two of the basic laws of heredity, viz., segregation and
independent assortment of the chromosomes. The latter exist in pairs,
and one complete set (haploid, n) is given to each offspring by each of its
parents through an egg or a sperm, which makes each offspring'diploid
or 2n. This offspring in turn passes on a random sample of its chromo-
somes, one of each pair, however, to each of its offspring. We traced tlf'
procedure where one, two, or three pairs of genes in separate chromosomes
were involved. We learned the F 2 monohybrid ratio to be 3: 1 phenotypi-
cally (or 1:2:1 genotypically) and the F2 dihybrid ratio to be 9:3:3:1
phenotypically, Finally, we studied various modifications of the two-
gene ratio. due to lack of dominance between alleles or between pairs of
\
genes, the latter condition being known as epistasis, and attempted to

clarify our minds regarding several aspects of dominance and recessiveness.
References
General
BABCOCK, E. B., and CLAUSEN, R. E. 1927. "Genetics in Relation to Agriculture,"
2d ed., McGraw-Hill Book Company, Inc., New York.
CASTLE, 'V. E. 1930. "Genetics and Eugenics," Harvard University Press, Cam-
bridge, Mass .
..J CREW, F. A. E. 1925. "Animal Genetics," Oliver & Boyd, Ltd., Edinburgh and
London.
GLASS, B. 1943. "Genes and the Man," Bureau of Publications, Teachers (,ollegf',
Columbia University, New York.
GOLDSCHMIDT, RICHARD. 1938. "Physiological Genetics," McGraw-Hill Book
JONES, D. F. 1925. "Genetics in Plant and Animal Improvement," John 'Wiley &
Sons, Inc., New York.
Mo:EIR, O. L. 1934. "Heredity and Disease," W. 'V. Norton & Company, New
York.
MORGAN, T. H. 1926. "The Theory of the Gene," Yale University Press, New
Haven.
- - - . 1919. "The Physical Basis of Heredity," J. B. Lippincott Company,
Philadelphia.
SHULL, A. FRANKLIN. 1948. "Heredity," 4th ed., McGraw-Hill Book Company,
Inc., New York.
J SINNOTT, EDMUND W., DUNN, L. C. and DOBZHANSKY, TH. 1950. "Prindpk~Lof
Genetics," 4th ed., McGraw-Hill Book Company, Inc., New York.
_/ SNYD~,L-:II. 1946. "The Principles of Heredity," D. C. Heath and Company,
Boston.
STURTEVANT, A. H., and BEADLE, G. W. 1939. "An Introduction to Genetics,"
W. B. Saunders Company, Philadelphia.
WADDINGTON, C. H. 1939. "An Introduction to Modern Genctics," The Macmillan
Company, New York.
WALTER, H. E. 1938. "Gcnetics," The Macmillan Company, New York.
WRIEDT, C. 1930. "Heredity in Livestock," The Macmillan Company, New York.
Heredity in Man
BURLINGAME, L. L. 1940. "Heredity and Social Problems," lVlcGraw-Hill Book
CHARLES, E. 1934. "The Twilight of Parenthood," "T.
,Yo Norton & Company,
New York.
CREW, F. A. E. 1927. "Organic Inheritance in :!\Ian," Oliver & Boyd, Ltd., Edin-
burgh and London.
GATES, R. R. 1929. "Heredity in Man," The Macmillan Company, Kew York.
-~~- 1946. "Human Genetics," The l\!facmillan Company, New York.
---1949. "Pedigrees of Kegro Families," The Blakiston Company, Philadelphia.
GR1uBARD, M. 1936. "Biology and Human Behavior," Tomorrow Publishers,
New York.
334 BREEDING AND IMPROVEMENT OF F'ARM ANIMALS
HOLMES, S. J. 1936. "Human Genetics and Its Social Import," McGraw-Hill

Book Company, Inc., New York.
HUNTINGTON, E. 1935. "Tomorrow's Children," John Wiley & Sons, Inc., New
York.
JENNINGS, H. S. 1930. "The Biological Basis of Human Nature," W. W. Norton &
Company, New York.
LORIMER, F., and OSBORN, F. 1934. "Dynamics of Population," The Macmillan
Company, New York.
OSBORN, F. 1940. "Preface to Eugenics," Harper & Brothers, New York.
POPENOE, P. 1929. "The Child's Heredity," The Williams & Wilkins Company,
Baltimore.
SCHEINFELD, A. 1939. "You and Heredity," J. B. Lippincott Company, Phila-
delphia.
Problems'
The following problems are designed to promote facility in correct genetic thinking.
Not until you can work these and similar problems quickly and correctly will you
have a comprehensive grasp of the genetic principles that they illustrate.
1A.. If a homozygous rough-coated animal is crossed with a smooth one, what will
be the appearance of the F, ? Of the F 2 ? Of the offspring of a cross of the F 1 back
on its rough parent? On its smooth parent?
"I 2. A certain rough-coated guinea pig bred to a smooth one gives 8 rough and
7 smooth offspring. What are the genotypes of parents and offspring?
3. If one of the rough F, animals in the preceding question is mated to its rough
parent, what offspring may be expected?
4. Two rough-coated guinea pigs when bred together produce 18 rough and 1
smooth offspring. What are the genotypes of the parents?
6. Cross a homozygous rough, black animal with a smooth, white one. iYhat
will be the appearance of the F,? Of the F2? Of the offspring of a cross of the F,
black with the rough, black parent? With the smooth, white one?
6. In the F 2 generation in the preceding question, what proportion of the rough,
black individuals may be expected to be homozygous for both characters?
\., 7. N oTE.-Polled or hornless condition in cattle (P) is dominant over horned (p).
A certain polled bull is bred to three cows. With cow A, which is horned, a horned
calf is produced; with cow B, also horned, a polled calf is produced; and with cow C,
which is polled, a horned calf is produced. What are the genotypes of all these four
animals and what offspring would you expect from these three matings?
8. N OTE.-In Shorthorn cattle the heterozygous condition of red coat cclor (R)
and white (r) is roan.
If two roan Shorthorn cattle are mated, what chance will their offspring have Qf
resembling their parents in coat color?
9. Two black female mice are crossed with a brown male. In several litters
female No.1 produced 9 blacks and 7 browns; female No.2 produced 17 blacks.
What deductions can you make concerning inheritance of black and brown coat
color in mice? What are the genotypes of the parents in this case?
10. A breeder has a group of animals all of which are heterozygous for a given
dominant character, A, and thus have th~genotype
.. ..
Aa. He wants to establish a
two
~-
, These problems and those at the ends of the next chapters are taken for the
most part from lists in Sinnott and Dunn, "Principles of Genetics," 3d ed., McGraw-
Hill Book Company, Inc., New York, 1939, and are reproduced here through the kind
permi~sion of the authors.
race homozygous for A. To accomplish this he allows these animals to interbreed
freely and then practices mass selection among their offspring, saving for breeding
all individuals that show the character A and discarding only the aa animals. Mem-
bers of one generation do not cross with any other. In the third generation, what
proportion of the animals which have the character A will be homozygous for it?
How could the breeder have established a homozygous strain more easily?
11. If the breeder mentioned in the previous question is practicing mass selection
for two characters, A and B, and his original animals have the genotpye Aa Bb, would
selection be effective more rapidly or more slowly than when only one character was
concerned? Explain.
12. It has long been known that Blue Andalusian fowls do not breed true to the
blue color of their plumage. How do you explain this?
13. What offspring will a Blue Andalusian fowl have if bred to birds of the following
plumage colors: (1) black, (2) blue, (3) white?
14. NOTE.-In man, brown eyes (B) are dominant over blue (b). A brown-eyed
man marries a blue-eyed woman and they have 8 children, all brown-eyed. "What
are the genotypes of all the individuals in the family?
16. A blue-eyed man both of whose parents were brown-eyed marries a brown-
eyed woman whose father was brown-eyed and whose mother was blue-eyed. They
have one child, who is blue-eyed. What are the genotypes of all the individuals
mentioned?
16. Assume that in man, right-handedness (R) is dominant over left-handedness
(r). A right-handed man whose mother was left-handed marries a right-handed
woman who has three .brothers and sisters, two of whom are left-handed. \Vhat
chance will the children of this marriage have of being left-handed?
,/17. \Vhat are the chances that the first child from a marriage of two heterozygous
brown-eyed parents will be blue-eyed? If the first child is brown-eyed, what are the
chances that the second child will be blue-eyed?
~ At the time of synapsis preceding the reduction division, the homologous
chromosomes align themselves in pairs and one member of each pair passes to each of
the daughter nuclei. Assume that in an animal with four pairs of chromosomes,
chromosomes A, B, C, and D have come from the father and A', B', C', and D' have
come from the mother. In what proportion of the germ cells of this animal will all
of t? maternal chromosomes be present together? All of the paternal?
;_A9. If one individual is homozygous for four dominant factors and another for
their four recessive allelomorphs, and if these two individuals are crossed, what
proportion of the F2 from this cross will resemble each parent, respectively, in
appearance?
~O. A rough, black guinea pig bred with a rough, white one gives 28 rough, black;
31 rough, white; 11 smooth, black; and 9 smooth, white. What are the genotypes of
t~parents?
/21. Two rough, black guinea pigs when bred together have 2 offspring, one of
them rough, white and the other smooth, black. If these same parents were to be
~
d together further, what offspring would you expect from thcm?
22. If a homozygous polled, white animal is bred to a horned, red one, what will be
e appearance of the FJ? Of the F2? Of the offspring of a cross of the FI back with
~the polled, wh~te parent?
" 23. A polled, roan bull bred to a horned, white cow produces a horned, roan
daughter. If this daughter is bred back to her father, what offspring may be expected
as to horn::; and coat color?
24. In swine, white coat is dominant over black and the "mule-footed" condition
over that with normal feet. A white, mule-footed boar, A, always produces white,
mule-footed offspring, no matter to what sow he is bred. Another boar, B, however,
also white and mule-footed, when bred to black so'ws produces about ~-2 white and
H black offspring, and when bred to normal-footed sows, about 7-2 mule-footed and
Y2 normal offspring. Explain this difference between these two animals by com-
paring their genotypes for these two traits.
25. In poultry, feathered legs (F) are dominant over clean legs (f); and pea comb
(P) over single comb (p).
Two cocks, A and B, are bred to two hens, C and D. All four birds are feathered-
legged and pea-combed. Cock A with both hens produces offspring that are all
feathered and pea. Cock B with hen C produces both feathered and clean, but all
pea-combed, but with hen D produces all feathered but part pea-combed and part
single. What are the genotypes of these four birds?
26. In poultry, the factors for rose comb (R) and pea comb (P), if present together,
produce walnut comb. The recessive allelomorphs of both, when present together in a
homozygous condition, produce single comb.
What will be the comb character of the offspring of the following crosses, in which
the genotypes of the parents are given?
RrPp X RrPp RrPp X Rr pp

RRPp X rrPp Rr pp X rrPp
rrPP X RrPp Hr pp X Rr pp
27. In the following five questions, all of which concern comb form in poultry,
determine the genotypes of the parents:
A rose crossed with a walnut produces offspring % of which are walnut, % rose,
% pea, and % single.
28. A walnut crossed with a single produces offspring 7.4 of which are walnut,
74 rose, 7.4 pea, and :-4 single.
29. A rose erossed with a pea produces 6 walnut and 5 rose offspring.
30. A walnut erossed with a single produces 1 single-eombed offspring.
31. A walnut crossed with a walnut produces 1 rose, 2 walnut, and 1 single offspring.
32. If one of the walnut parents in the preceding question were crossed with one
of its single-combed offspring, what would their offspring be like?
33. A rose crossed with a walnut produces offspring % of which are walnut, %
rose, Va pea, and Va single. What are the genotypes of the parents?
34. A feather-shanked rose-combed crossed with a clean-shanked pea-combed one
produces 25 feather, pea, offspring; 24 feather, walnut; 26 feather, rose; and 22 feather,
single. What are the genotypes of the parents?
35. The offspring of a feather-legged pea-combed cock bred to a clean-legged pea-
combed hen are all feather-legged. lVIost of them are pea-combed, but some singles
appear among them. What are the genotypes of the parents? What would be the
offspring expected from a cross of this hen with one of her feather-legged single-
combed sons?
36. A right-handed blue-eyed man whose father was left-handed marries a left-
handed brown-eyed woman from a family in which all the members have been brown.!'lt
eyed for several generations. What offspring may be expected from this marriage,
as to the two traits mentioned?
37. A" brown-eyed right-handed man marries a blue-eyed right-handed woman.
Their first child is blue-eyed and left-handed. If other children are born to this
couple, what will probably be their appearance as to these two traits?
38. A right-handed blue-eyed man marries a right-handed brown-eyed woman.
They have two children, one lcft-handed and brown-eyed and the otner right-handed
and blue-eyed. By a later marriage with another woman who is also right-handed
and brown-eyed, this man has 9 children all of whom are right-handed and brown-
eyed. What are the genotypes of this man and his two wives?
39. A brown-eyed normal-minded man marries a brown-eyed normal-minded
woman. Their first child is blue-eyed and feeble-minded. What is the chance that
the next child will also show these same characteristics? (Normal mind, dominant.)
40. A breeder has a homozygous race of feather-legged, black, rose-combed birds
and another of clean-legged, white, pea-combed ones. He wants a race of black birds
that have clean legs and walnut combs. Wnat proportion of the F2 raised from a
cross between these two races will be what he desires in appearance?
41. What are the phenotypes of the parents and offspring of the following cross:
ff Rr Pp X Ff Rr pp.
42. In rabbits there are several factors that determine coat color:
C colored c no color, albino

A ticked, agouti a solid color, nonagouti
B black b brown
S solid color 8 spotted
(AB, agouti; Ab, cinnamon; aB, black; ab, brown)
An albino rabbit mated with a black produces agouti and black, spotted young.
What factors are present in the albino parent and not in the black?
43. An agouti animal crossed with another agouti produces offspring of which
J16 are agouti, 716 black, 716 cinnamon, and 716 brown. What are the genotypes of
the parents?
44. Wbat are the offspring of the following crosses? What are the appearances of
the parents?
(a) Cc Aa bb X cc aa Bb
(b) cc Aa Bb X CC aa Bb
45. Two agouti rabbits produce 3 agouti and 1 black offspring. What are their
genotypes? (Use only the A,a and B,b factors.)
46. In what two ways could you develop a pure-breeding strain of polled, black-
bodied white-faced cattle?
47. NOTE.-In cattle the polled condition (P) is dominant over the horned (P)i
and in Shorthorns the heterozygous condition of red coat (R) and white coat (r) is
roan.
If a homozygous polled, white animal is bred to a horned, red one, what will be
the appearance of the F I? Of the F 2? Of the offspring of a cross of the F, with the
polled, white parent? With the horned, red parent?
48. If a group of 100 polled, roan Shorthorn cows by horned bulls were mated to a
roan bull that was heterozygous for the polled condition, what sort of offspring would
~u expect and in what numbers?
49. If a red-flowered broad-leaved plant is crossed with a white-flowered narrow-
leaved one, what will be the appearance of the F, and the F2?
50. If in mice you secured a 9 agouti: 3 black: 4 albino ratio in an F 2 , what were
the gellOtypes of the parents and the Fl llsing the letter C to represent the gene
4
necessary for the expression of any color and A to represent the gene for banding of
the black hairs with yellow to give the agouti pattern. What type of inheritance is
this?
51. In cultivated stocks (a flower) gene C in the absence of gene R, produces
cream-colored flowers; gene C with gene R, red ones; and genes cc with RR, Rr, or rr
white flowers. If you crossed a red-flowered plant with a white and got an Fl that
was all red, what proportion of red, creams, and white flowers would you get in; the F 2?
52. If you cross two spherical-shaped squash and secure a disk-shaped Fl a~d then
cross one of these Fl'S to an elongated-fruit plant, what will be the fruit shape of
their offspring?
K oTE.-In summer squashes the factor for white fruit color, W, is epistatic to
that for yellow, Y; WYand Wy plants are white; wY plants yellow, and wy plants
green.
53. What is the color of the fruit in the offspring of the following crosses, the
genotypes of the parents being given?
Ww Yy X Wwyy ww l'Y X Wwyy Wwyy X ww Yy
NOTE.-In the following three questions, which deal with fruit color in squashes.
find the genotypes of the parents.
54. A white plant crossed with a yellow one produces offspring of which Y2 are
white, % yellow, and Ys green.
55. A white plant crossed with a green one produces offspring of "'hich Y2 are
white and Y2 yellow.
56. A white plant crossed with another white one produces offspring of which
% are white, ~16 yellow, and 716 green.
57. What will be the flower color of the offspring of the following crosses, in which
the genotypes of the parents are given?
CcPp X ccPp ccPp X CC pp
CcPp X CcPP Cc pp X ccPp
58. A white-flowered plant crossed with a purple produces offspring of which
% are purple and % white.
59. A purple-flowered plant crossed with a white one produces offspring of which
Y2 are purple and Y2 white.
60. A white-flowered plant crossed with another white produces offspring of which
% are white and X purple.
61. In wheat, determine the genotypes of the parents in the following crosses:
(a) Red X white giving % red and X white.
(b) Red X red giving Ys red and Ys white.
62. How would you proeeed in developing a feathered-shanked strain of poultry-
from clean-shanked birds?
63. If an F2 yields a 15: 1 ratio, what could you conclude as to the type of inher-
itance involved?
64. Write the genetic color formulas for two breeds of poultry both of which are
white and the F 2 resulting from their crossing gives a 13 white: 3 colored ratio.
65. Starting with two white breeds of poultry, how would you proceed to develop
a true-breeding colored strain?
\
,\
CHAPTER XIII
THE PRINCIPLES OF HEREDITY (Continued)
In the preceding chapter we have dealt with some of the more simple
and clean-cut types of inheritance. During the past 30 years the inherit-
ance of a great many traits in plants and animals has been studied and the
methods of transmission from parent to offspring ascertained. The fol-
lowing might be called different types of inheritance, although basically
there is but one type of inheritance, viz., by means of genes. This method
of inheritance is typical of all sexually reproducing forms, so that what
may appear as different types of inheritance are actually different sorts of
expressions of gene reactions. The procedure has been to cross plants or
animals possessing different characteristics, tabulate the results by sepa-
rate generations, and, finally, to construct a theory that would fit the
observed facts. It should be noted that the facts come first and the
theory to explain the facts comes later. This is the scientific approach.
In animal breeding, we sometimes find people making the directly oppo-
site approach, i.e., starting out with a theory and trying to find facts that
fit their theory. They generally do find some such facts, but unless one
takes all the known facts into account, his contribution is apt to be nil.
If one "selects" his facts carefully enough he can "prove" almost any-
thing. Several typical forms of inheritance will be presented in the fol-
lowing sections.
Multiple Alleles.-Thus far we have spoken of only two alternatives
as far as one pair of genes is concerned. We have talked about gene A
or its allele, gene a. We have said, too, that the genes are probably com-
plex, organic molecules made up of hundreds of thousands of atoms
arranged spatially in some definite pattern. We assume too that the con-
tent and spatial arrangement of the atoms in gene A are different from
those in gene a. It would seem logical, therefore, to expect that there
might be more than two alternative arrangements. Breeding experience
seems to im'icate that there are.
If a colored rabbit, CC, is bred to an albino, cac a , all the Fl are colored,
and in the F 2 there are 3 colored: 1 albino. Likewise another cross,
between colored, CC, and Himalayan, ChC h (white with black nose, ears,
tail, and feet), gives an Fl all of which are colored and the familiar 3
colored: 1 Himalayan in the F 2. Now, if the cross is made between
339
340 Hilj<JfiJDfSG AND IMPROl'EMENT OF PARM AN IMAl,.';
FIG. 97.-Three alleles of a gene for coat color in rabbits. Top, colored; center, Himalayan
albinism; bottom, complete albinism. (From Ca.sile, in Journal oj H eredity.)
Himalayan, c"c", and albino, caca, all the Flare Himalayan and in the F 2
there are 3 Himalayan: 1 albino, which indicates but one pair of genes
being involved. From the above breeding results, which show genes C
and ea to be allelic, genes C and cit to be allelic, and genes c" and ca to be
allelic, 'e conclude that all three genes C, elL, and ca form one multiple
allelic series. StilI an~ther gene, cc\ has been f;u~dln this-series causing
t he chinc.hilla color or silvery gray with no yellow color in the fur, and this
-----_
has been found, as far as dominance is concerned, to fit in between genes

C and ch • Thus we could have the following genotypes.
colored rabbitt CC, Cc'\ Cch, Cc
chinchilla rabbit cchcch, cChch, cene
'...,. Himalayan rabbit eheh, eke
albino rabbit cc
Many other multiple allelic series are known in both plants and animals,1
~J,lld
in most of them do~m~rp.~e~the....lvild-type .
domin~~ing. In mice a multiple allelic series of four genes iSI
known, gene C producing full color; gene cch dilute color or chinchilla;
gene ch extreme dilution; and gene c albinism, or lack of color. Here gene
C is dominant over the other members of the series, but crosses between
the other members give results intermediate between the two used, e.g.,
crossing homozygous chinchilla with albino gives the intermediate,
extreme dilution. Allelic series are known in many plants, and in Dro-
sophiLa there are 14 aIIeles of the gene for red eye grading all the "way down
..~ to white eyes. Series of multiple alleles are also known in guinea pigs,
rats, mice, rabbits, and snapdragons.
In man, blood types show a multiple allelic series of thr~e genes. Blood
itself is not inherited, but type of blood is. The red-blood cells in man
contain an antigen A or B or both. The blood serum contains antibodies
a or b or both. A person of blood group A has antigen A in his red-blood
cells and antibody b in his serum; of blood group B has antigen B in his
red cells and antibody a in his serum; of blood group AB has antigens A
and B in his red cells and neither antibody in his serum; and, finally, a
person of blood group a has neither antigen A nor B in his red cells but
both antibodies a and b in his blood serum. If, therefore, red cells frbin
a person with the A antigen were introduced by blood transfusion into a
person of type B with antibody a in his serum, the reaction between thc
serum antibody a and the red-cell antigen A would cause these cells to
clump or agglutinate with possibly fatal results. In making blood trans-
fusions, blood types must be ascertained in order to prevent red-cell
agglutination.
Genetic make- Antigen in Antibody in Can give Can take

Type
up cells serum blood to blood from
A AA or Aa A b A or AB A or 0
B ABAB or ABa B a B or AB B or 0
AB AAB AB 0 AB All
0 aa 0 ab All 0
342 BRElmI.YG AND IMPROVEMENT OF FARM A,VIJfALS
Knowledge of the genetic make-up of individuals regarding blood type

(Table 21) is sometimes useful in establishing possible, though not actual,
paternity. If, for instance, a mother is of type A (genetic make-up AA
or Aa) and her child is of type 0 (genetic make-up aa thus making mother
Aa) and the assumed father turns out to be type AB (genetic make-up
AA B), this fact removes the possibility of his having actually been the
child's father, because the child received two a's and no man of type AB
has an a to transmit to his offspring. If the tests revealed that the
assumed father was of type A, B, or 0, then the most that can be said is
that he might have been the father as, of course, might also many other
men.
Although many of the genes involved are not known to occur in multiple
allelic series, the work of Irwin and his coworkersl demonstrating the
presence of heritable antigens in the blood of cattle and several species of
birds is of considerable interest. To date more than 40 such antigens
have been detected in cattle blood. In addition to their interest from a
basic genetic standpoint, they are useful in cases of disputed plirentage in
cattle.
Lethal Action of Genes. 2-Any organism is a delicately balanced sys-
tem of actions and reactions and can survive only a more or less limited
range of change. This is nicely illustrated in the blood stream, which
must maintain a proper balance of sugar, minerals, vitamins, hormones,
and other elements if the organism is to function normally. All bodily
characteristics are referable in the final analysis to the genes, which pro-
duced the organism, manifesting themselves of course in our internal and
external environment.
It is not surprising, therefore, that gene changes sometimes result in
the sort of unbalance that destroys the organism. Such manifestations
are known as lethals. In the fowl, for instance, there sometimes occurs a
condition known as creeper (short wings and legs). When two creepers
are mated, some normal offspring are produced, the normal condition
apparently being recessive to creeper, but the ratio of such matings is 2
creeper: 1 normal. Investigation has shown that some of the embryos
die at about the fourth day of incubation, these presumably being
homozygous creepers. Creeper, therefore, has the genetic constitution
Cc, normal cc, this being a dominant lethal, according to Snyder, and the
homozygous and heterozygous conditions are recognizably different.
Several recessive lethals are known in cattle, viz., parrot-beaked (abnor-
1 See OWEN, R. D., et al., An Immunogenetic Analysis of Racial Differences in
Dairy Cattle, Genetics, 32:64-74,1947.
2 See EATON, O. N., A Summary of Lethal Characters in Animals and Man, Jour.
Hered., 28(9) :320-326, September, 1937.
mal lower jaw; calves live only a few hours), amputated (calves born dead
with legs and lower jaw missing), short-spined (fused and compacted
vertebrae), hairless. Bulldog calves with greatly dished faces and very
short legs are a dominant lethal manifestation similar to creeper in fowl,
the heterozygotes being the short-legged Dexter cattle.
Other abnormal conditions that may be due to lethal genes include
mummified calves, resorbed fetuses, ossified joints, and various types of
congenital dropsy. Other lethals in farm animals are known, e.g., para-
lyzed hind limbs in sheep and swine, thick forelimbs, closed anus and cleft
palate in swine, closed colon in horses. If an abnormal embryo is born
the lethal is made obvious. Such happenings are generally due to failure
to reach full development; e.g., hair or limbs do not grow; palate, skull, or
lips do not fuse in the mid-line; joints fail of full development with fluid
and pads between bones. Lethals that destroy the organism in early
embryonic life with its subsequent resorption may pass unrecognized as
sterility.
Like most other mutations, the bulk of lethals are recessive and may
remain hidden in a stock for many generations. For their total removal
from a strain, test matings and rigid selection must be practiced with the
knowledge that in the case of recessives both parents are carrying the
gene.
Many lethals of various sorts are known also in plants and in man.
Lethals may kill the organism at any time, i.e., during embryonic, fetal,
or postnatal times; and, if the lethal gene is located on the sex chromo-
some, peculiar sex ratios will of course result.
The following list of lethals contains those listed by Eaton l in 1937 and
a few others reported in the Journal ?f Heredity since that time.
Horses
Atresia coli-closure of intestine. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. R *
Abnormal sex ratio-55 males: 90 females .............. Sex-linked R
Lethal white-low fertility-lethal or sterility factors or both. . . . .. ?
Stiff forelegs ........................................ Probably R
Cattle
Achondroplasia,-(Bulldog) short legs and head, hernia, die and
aborted fourth month.. .. . .. . ............................. Dt
Achondroplasia2-(Bulldog) short head, cleft palate, deformed jaws,
die soon .................................................. R
Acroteriasis congenita-amputated-appendages short or absent.. It
".,-' Agnathia-very short lower jaw (sex limited to male?) ......... " R
Ankylosis-ossification of joints ............................... R
*R = recessive.
tD = dominant.
1 Loc. dt.
344 BREEDIXG ASD IMPROVEMENT OF FARM ANIMALS
Congenital dropsy-water in tissues and cavities. . . . . . . . . . . . . . .. R

Congenital ichthyosis-scaly, cracked skin. . . . . . . . . . . . . . . . . . . . .. R
Epitheliogenesis imperfecta-imperfect skin, partly hp,irless-
septicemia. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. It
Fetal resorption. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ?
Hernia cerebri-failure of frontal bones to fuse. . . . . . . . . . . . ?
Impacted molars-short jaw, defective teeth-die within week. . ?
Lameness in hind limbs-calves unable to stand. . . . . . . . . . . . . . . .. R
Muscle contracture-head and legs drawn up-joints stiff. . . . . . .. R
Mummification---short, stiff neck, prominent joints-usually die as
8-months fetus. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. R
Short limbs-limb short, hoofs undeveloped ..................... R
Short spine-ribs and vertebrae fused, back bent down. . . . . . . . . .. R
Sheep
Amputated-no claws on feet. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. ?
Earless and cleft palate. . . . . . . . . . . . . . . . . . . . . . . . . . .. .......... It
Lethal gray-in Turkanas and Karakuls. . . . . . . . . . . . . .. ........ R
Muscle contracture-usually stillborn. . . . . . . . . . . . . . . . . . . . . . . . .. R
Paralysis-hind parts paralyzed-live few days. . . . . . . . . . . . . . . . .. R
Skeletal defects-large head, short upper jaw, rigid fetlocks ....... R
Swine
Atresia ani-closed colon. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. ?
Catlin mark-parietal or frontal bones not fused. . . . . . . . . . . . . . .. R
Cleft palate-young unable to nurse. . . . . . . . . . . . . . . . . . . . . . . . . .. R
Excessive fatness-young die at 40 to 80 kg. . . . . . . . . . . . . . . . . . . .. R
Fetal mortality. . . . . . . . . . . . . . . . . . . . . . . . . . . . R
Hydrocephalus-water outside brain in subarachnoid spaces. . R
Hypotrichosis-hairless (lack of iodine).... . . . . . . ?
Legless-(shoulder blades and pelvic bones but no limb buds or kg
bones) ..................... " ......... , .................. R
Lobed ear-usually also cleft palate and deformed hind legs ...... .
................................................ . Probably R
Muscle contracture-thick, stiff legs. . . . . . . . . . . . . . . . . . . . . . . . . .. R
Paralysis-hind parts paralyzed. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. R
Man
Brachyphalangy-short middle phalanx second finger and toe, live
1 year .................................................... D
Congenital ichthyosis-cracked, imperfect skin, live 3 days ....... R
Glioma retinae-malignant tumor of retina. . . . . . . . . . . . . . . . . . . .. ?
Icterus-jaundice, live 4 days to 37~ years ...................... ?
Infantile amaurotic idiocy-degeneration of cerebrospinal nerves,
live 2 or 3 years. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. R
Progressive spinal muscular atrophy-live a few years. . . . . . . . . . .. ?
Xeroderma pigmentosum-sensitive skin, scars, and carcinoma,
live 12 years. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. R
Multiple Factors.-Thus far our discussion has been confined to quali-

tatively differing characters such as color of hair, preRence or ahsence of
THE PIUNCIPLES OF HEREDITY 345
horns or of feathers, types of blood, or the ability or inability to survive,
etc. There are many other characters, in fact most of the important ones,
that do not exhibit this alternative type of expression but rather show all
manner of gradation between rather wide limits. Such characters are
quantitative and, as' examples, size, yield, power, stamina, rapidity of
fattening, quality of carcass, degrees of fertility, and probably most of the
other commercially important characters of our livestock might bE"
mentioned.
It seems probable that the final expression of a great many character-
istics of higher animals may be due to the interaction of many sets of
genes. This point of view has dawned gradually, as more detailed infor-
mation about inheritance in some of the lower forms of life has accuma~
lated, and is in direct contrast to the views held soon after the rediscovery
of Mendel's work. Attempts have been made to analyze complex physi-
ological processes such as milk production and the percentage of fat in
milk on the basis of the interaction of a few pairs of genes, but they have
not been successful. FortulllLtely, it should be possible to make progress
in breeding on the basis of a sound knowledge of the basic principles even
in the absence of specific information concerning the number and inter-
actions of the genes responsible for the appearance of commercially valu-
able qualities. When many factors or genes are concerned, we know that
the offspring arising from the mating of two extremes tend to fall on the
average about midway between two parental levels. This principle if<
involved in indexing dairy bulls, as described in detail in Chap. XXI.
The multiple-factor hypothesis grew out of the work of Nilsson-Ehle
with wheat and East with corn. The former crossed several strains of
red and white wheats. In general the red color was only partially domi-
nant over white, for the F 1 '.vas not as dark red as the red parent. In the
F2 there were 3 reds (1 dark red, 2 lighter red) to 1 white. This indicates
a one-gene pair reaction. Another cross of red and white wheats gave a
similar Fl but an F2 of 15 reds (of varying shades) to 1 white thus indicat"
ing a two-gene pair reaction. Still a third cross gave the usual F 1 but ap
F2 of 63 reds (again of varying shades) to 1 white, which indicates a three-
gene pair reaction. This uniformity of the Fl and variability of the F2 it'
generally true of quantitative characters controlled by many genes.
Breeding tests revealed that there were one, two, and three pairs of genes
operating in these cases. In the two-gene cross it was shown that a wheat I
with four red genes was redder than one with three, this latter one redder:
than one with two, and this in turn redder than a wheat with only one red
gene. In these cases, there is not complete dominance, but the genes act
in a cumulative fashion.
This method of inheritance is now thought to be operative (along with
dominance, epistasis, and other types) in the inheritance of most of the

commercially important animal characters. In other words, instead of
color of wheat, we might substitute size, weight, yield, or many other
quantitative traits. This· will explain the difficulties encountered in
attempting to unravel the exact mode of inheritance of many characters
in the larger animals.
It is possible to determine with a fair degree of accuracy how many
genes are operating to produce a given characteristic. If two extremes
are mated, the Fl will fall about midway between the parents if the genes
are acting cumulatively. If lout of the 4 of the F2 is as extreme as
each of the 2 extreme grandparents, then we are justified in assuming that
1 pair of genes is involved. If lout of 16 of the F2 is as extreme as each
grandparent, then 2 genes must be operative. This interpretation fits
very well the observed reactions of color inheritance in man, black skin
being represented by AA BB and white skin by aa bb. The Fl will be
mulattoes, Aa Bb, and ,Yill yield five grades of color depending on the
number of dominant genes present, AA BB being black, AA Bb, etc.
being dark brown, Aa Bb, etc. being mulatto, Aa bb, etc., being light
brown, and aa bb white.
A cross between long-eared rabbits (220 mm.) and short-eared (100
mm.) gives offspring with ears about 160 mm. in length. These Fl rabbits
in turn produce a graded scale of ear lengths in F 2 ranging from 220 down
to 100 mm., each dominant gene accounting for an increased ear length of
20 mm. and acting cumulatively. The student should make up a table
showing the number of operating genes on the basis of the number of
extreme individuals in the F 2 and will find such a table useful in inter-
preting problems involving cumulative multiple-factor inheritance; lout
of 4 as extreme as each grandparent = 1 pair of genes; lout of 16 = 2
pairs, lout of 64 = 3 pairs, etc.
On the basis of multiple-factor inheritance it is possible to make crosses
, that will yield some F2 members more extreme than either grandparent.
This is probably due to each parent contributing some effective alleles not
supplied by the other parent and is known as transgressive variation.
Thus one parent might be AA BB cc and the other aa BB CC, whereupon
the F 1 would be Aa BB Cc, but giving the possibility of getting AA BB ~C
in the F 2, which would be more extreme than either grandparent.
Most quantitative characters such as size, growth rates, milk produc-
tion, egg production, and prolificacy depend for their expression upon the
interaction among a large but unknown number of heritable factors and
environmental influences. Thus the evaluation of differences between
breeds, lines, or strains of animals does not depend upon determining the
actual effects of specific genes, but rather upon determining the average
THE PRINCIPLES OF HEREDI'l'Y 347
effects of large numbers of genes. Because of the complexity of these

interactions, standard statistical techniques are ordinarily necessary for
the correct interpretation of genetic studies. A knowledge of such pro-
cedures as correlation, regression, and analysis of variance and covariance
is necessary. The detailed statistical techniques are given in such stand-
ard texts as Snedecor (1946).
A question as to the scientific validity of the additive, nondominant
action of multiple genes has recently been raised by MacArthur. I Using
diallel crossing between four varieties of tomatoes with fruit weights
ranging from 0.8 g. through 3.6,65.9, and 106.8 g. per variety, this investi-
gator found that the various F I hybrids were not an arithmetic average
between the pure parent stocks but closely approached the geometric
means of the parental varieties. For example, the cross between the
0.8-g. parent and the 3.6-g. parent did not yield an FI of 2.2 g. (the arith-
metic mean) but the FI averaged 1.7 g. (VO.8 X 3.6). Likewise the
cross between the 3.6-g. and the 106.8-g. parents gave an FI of 19.4 g.
(very close to V3.6 X 106.8 = 19.6) rather than the arithmetic mean of
55.2 g. The whole series of crosses was orderly and folluwed the geo-
metric rather than the arithmetic pattern. The size of the F 1 fruits was
quite evidently controlled to a large extent by the genes provided by the
smaller parent or, stated differently, the amount of additional growth
brought about by the genes of the larger parent was limited in a propor-
tional manner by the substrate upon which the activating substances
reacted. More recent work by MacArthur (1944 and 1949)2 with mice
appears to bear out his earlier conclusions based on tomato work.
Because so many of the commercially important characters of our live-
stock are dependent on cell size, cell number, or cell activity and appear
to be controlled by many genes, this new theory, if it proves to be of
general validity, will have an important bearing on our reasoning and
procedures in livestock breeding. If, for example, 4,000-Ib. producing
cows were mated to a bull with a hereditary complex for 24,000 lb. of
milk, we would expect the resulting daughters to average not halfway
between 4,000 and 24,000 or 14,000 lb. (the arithmetic mean) but rather
at 9,800 lb. (V 4,000 X 24,000). In other words, the genes from this
very high-heredity sire are limited proportionally by the low-heredity
genes of the dams. Where the differences in the hereditary levels are
I MACARTHUR, J. W., Size Inheritance in Tomato Fruits, Jour. Hered., 32(9):
291-295.
2 MACARTHUR, J. M., Genetics of Body Size and Related Characters. 1. Selecting
Small and Large Races of the Laboratory Mouse, Amer. Nat., 78:142-157, 1944;
II. Satellite Characters Associated with Body Size in Mice, Amer. Nat., 78:224-237,
1914; Selection for Small and Large Body Size in the House Mouse, Genetics, 34 :194-
209,1949.
348 BREEDING AND Ill1PROVEMENT OF FARM ANIMALS
small, the arithmetic and geometric means are very similar. For example,
if the sire's level was 14,000 lb. and the dams 12,000 lb., the arithmetic
mean would be 13,000 and the geometric mean 12,961 lb.
Sex-linked Inheritance.-In the examples used thus far to illustrate
various types of inheritance, it has made no difference as to sex where
the various characteristics 'were placed; i.e., this set could have been in
the male, that in the female, or vice versa. In the type now to be dis-
cussed, however, it will make a difference, because the genes determining .
the characteristics will be located in the chromosomes that determine the
sex of the individual.
Drosophila melanogaster, the common fruit fiy, has provided more
material assistance than any other species in the solution of the problems
involved in hereditary transmission. Cytological studies of the cells of
this organism reveal the fact that they have four pairs of chromosomes.
In the female, these are all evenly paired, two of the pairs being large,
bent rods; one pair very small and oval-shaped; and one pair fairly large,
straight, and rod-shaped. The first three pairs are called auto somes, and
the last pair sex or X chromosomes (see Fig. 89). The chromosomal
make-up of males is similar to that of females regarding the autosomes
but the male has an odd pair, one of the pair being an X chromosome,
like that found in duplicate in the female, the other member of the pair
being somewhat smaller, with a hook or bent portion at one end, and
called the Y chromosome.
XY Type of Sex Inheritance.-In the breeding work with Drosophila,
which normally have red eyes, a white-eyed mutant male was found by
Morgan. This male was saved and, when mated to red-eyed females,
yielded an Fl all of which were red-eyed. The F2 gave the expected
3 red: 1 white phenotypic ratio, but it was noted that the white-eyed
individuals were all males. This exception to usual Mendelian expect-
ancy was finally explained by locating the gene for white eyes in the
X chromosome. The Fl female from a red-female white-male cross would
therefore have two chromosomes, in one of which would be found the
gene for red eyes, with the gene for white eyes in the other sex chromo-
some. Mated to a red-eyed male, she in turn would yield an F 2 with
both of the females red-eyed, one male red-eyed, and the other white-
eyed. The diagram on page 349 illustrates this cross.
This is an example of sex-linked inheritance. It is seen that the female
produces eggs, following reduction, all of which are alike as far as the sex
chromosome is concerned, whereas the male produces two sorts of sperm
cells in equal numbers, one containing an X chromosome and the other a
Y chrom~some. The sex of the offspring depends, therefore, on the type
of spermatozoon that happens to fertilize the egg, an X egg plus an
\
x-containing sperm yielding a female and an X egg plus a Y-containinp;

sperm yielding a male. All characteristics that are determined by genes
carried in the X chromosomes will, therefore, show this peculiar type of
sex-linked inheritance. Many such characteristics have been discovered
and catalogued.
White-eyed
Red-eyed female eggs
male sperms
RX RX
rX RXrX RXrX FI red-eyed females, heterozygous
Y RXY RXY F I red-eyed males
and the F2 produced by mating two Fl'S
Red-eyed
Red-eyed female (heterozygous) eggs
male sperms
RX' rX
RX RXRX RXrX F2 red-eyed females, 1 heterozygous
Y RXY rXY F2 males, 1 red-eyed, 1 white-eyed
FiG. 98.-Checkerboard diagram of cross of red-eyed female and white-eyed male through
F 2•
Sex-linked Factors in Drosophila.-There is abundant substantiating

proof of this theory from the behavior of sex-linked factors. The sex
chromosomes (chromosome I in the chart, Fig. 108) carry factors that
determine characteristics just as do the other chromosomes. About 200
sex-linked characters have been discovered in the fruit fly. The factor
for \vhite eyes in Drosophila, for instance, is sex-linked. When a white-
eyed male rXY is mated to a red-eyed female RXRX, the F 1 offspring all
have red eyes, because red eye is dominant to white eye in Drosophila.
In the F 2, red- and white-eyed flies are produced in the proportion of
3 red: 1 white. All the females of this generation are red-eyed, but of
the males 7~ have red eyes and Yz have white eyes. When the reciprocal
cross is made, i.e., when a white-eyed female is mated to a red-eyed male,
the results are different. In the Fl of such a mating, all the female flies
have red eyes and all the male flies have white eyes. In the F 2 of this
cross, Yz the females have red eyes and Yz have white eyes, and likewise
among the males Yz have red eyes and Yz have white eyes.
-
----
rX Y ,~ RX _. Y
{
RX RXrX RXY rX RXrX rXY
---
RX RXrX RXY rX RXrX rXY
Fl cross, white-eyed male and red-eyed F 1 cross, red-eyed male and white-eyed
female female
1 1
RX Y rX Y
----- .
RX RXRX RXY _.
RX RXrX RXY
rX RXrX rXY rX rXrX rXY
F 2 cross, white-eyed male and red-eyed F2 cross, red-eyed male and white-eyed
female female
F,G. 99.-Diagram of cross of white-eyed male X red-eyed female and the reciprocal.
In the foregoing diagrams the presence of two X chromosomes deter-

mines a female, an X and a Y chromosome determine a male. The
presence of at least one R determines red eyes, because red eye is dominant
to white. It has been pointed out previously that, aside from the pair
of sex chromosomes, the pairs of chromosomes in both the male and the
female of Drosophila are alike and carry corresponding factors or deter-
miners of characteristics. In this type of sex inheritance, the female is
homogametic for the sex chromosome. A variation of the XY type of
sex determination occurs in some species that lack the Y chromosome.
Here the females are XX but the males are XO (Y chromosome lacking).
It is evident, however, that the male 'will still produce two types of
spermatozoa, one containing an X chromosome, the other lacking an X
chromosome.
Sex-linked Characters in Man.-In the human family a sex-linked
characteristic is that of color blindness, or the inability, through an
inherent defect in the eye, to distinguish between various colors. It is
well known that there are many more color-blind men than women, the
reason for which should soon be evident. Representing normal vision
with C and color blindness with c, the following is the genetic make-up
of a normal-visioned woman CXCX and of a color-blind man cXY.
Such mating would give children in the F 1 all of whom would be normal-
visioned, though all the daughters would be carriers of color blindness.
In the F 2 , there would be 3 normal-visioned children (2 daughters and
1 son) to 1 color-blind child (son). The reciprocal cross is also diag'ramed
as shown in Fig. 100.
In the foregoing, two X chromosomes determine a female, an X and
a Y chromosome determine a male, and at least one C determines the
ability to\differentiate colors. The occurrence of color-blind men is to
, \.
\
\
/ ,
\
be expected more frequently than color-blind women, for a double dose

of the color-blind gene is necessary to produce a color-blind woman (e.g.,
if 10 per cent of males were color-blind, then we could expect 0.1 X 0.1
or approximately 1 per cent of women to be color-blind).
Recent work has shown that, at least in man and possibly in one or
two rodents, portions of the Y chromosomes are homologous with por-
tions of the X chromosomes. These regions carry allelic genes and
synapse during the maturation processes. Chiasmata form and crossing
over takes place between them. Genes located in these regions are called
"partially sex-linked."
eX I y ex Y
ex excx exy eX excx cXY
ex excx exy eX eXeX cXY
I J
FI cross, color-blind man and normal- FI cross, normal-visioned man and color-
visioned-woman blind woman
ex y eX Y
ex ex ex eXy ex eXeX eXY
eX excx eXY eX eXeX cXY
F 2 cross, color-blind man and normal- F2 cross, normal-visioned man and color-
visioned woman blind woman
FIG. lOO.~Diagram of cross of color-blind man and normal-visioned woman and its
reciprocal.
WZ Type of Sex Inheritance.-Another type of sex inheritance, which,

to distinguish it from the XY type, is called the WZ type, is found in
poultry. Here the distinguishing feature is the fact that the male is
homogametic for the sex chromosome, whereas the female is hetero-
gametic. According to Castle, this condition is found in moths and in
certain fish and birds; e.g., in domestic fowls, pigeons, ducks, and canaries.
A well-known example of sex-linked inheritance in poultry is that of
coat pattern. For instance, if a Plymouth Rock (barred) cock is mated
to a black (orany other nonbarred breed), the Fl offspring, both male
and female, are barred. When these Fl hybrids are mated, producing
the F 2 generation, there results '3 barred individuals (2 males and 1
female) to 1 nonbarred or black (female). When the reciprocal cross is
made, viz., black cock mated to barred hens, there are produced in the
F 1 barred males and nonbarred, or black, females. When these F 1
hybrids are mated, producing the F 2 generation, Yz the number of males
so produced are barred and Yz are black, and of the females ~~ are barred
and Yz are black. Diagrammatically, we have B, barred; b, black;
ZZ, male; ZW, female.
BREEDD'{G AND IMPROVEMENT OF FARM ANI1'>fALS
BZ BZ bZ bZ
BZbZ BZbZ BZ BZbZ BZbZ

I--~B~Z=W=---I---B-Z-W---
W bZW bZW
Fl cross, barred cock and black hen F 1 cross, black cock and barred hen
BZ bZ BZ bZ
BZ BZBZ BZbZ bZ BZbZ bZbZ

W BZW bZW W BZW bZW
F 2 cross, barred cock and black hen F 2 cross, black cock and barred hen
FIG. lOl.-Diagram of cross of barred cock and black hen and its reciprocal in F,.
In the foregoing, two Z's determine a male, one Z and one \V a female,
and at least one B a barred individual.
In some species the Y or W chromosome is lacking, giving XO or ZO
types of spermatozoa or ova.
Sex-influenced Inheritance.-This type of inheritance is not to be
confused with sex-linked inheritance, which has been discussed in the
previous sections. In sex-linked inheritance the genes determining the
characteristic are located in the X chromosome. Such is not the case in
sex-influenced characters, the genes for such characteristics being located
in the autosomes, not in the sex chromosomes.
Basically all characteristics are dependent on genes, or the interaction
of genes plus, of course, a suitable environment, both external and
internal. The environment, be it noted, is just as important as the
genes. A cow, for instance, might have inherited the genes that would
allow her to produce 10,000 lb. of milk under good environmental con-
ditions, but, if she calves out in the woods and runs wild therein, it is
doubtful whether she would give more than a few hundred pounds of milk
at the most.
Animals also have an internal environment, their own bodies, and it
js apparently this internal environment that, together with genes, is the
determining factor in the expression of sex-influenced characteristics.
Both the male and the female are supplied with several sets of glands
that secrete chemical substances which are picked up by the blood stream
and carried to all parts of the body, where they make their presence
known through influencing the activities of cells in various parts of the
body. These are known as the endocrine, or ductless, glands, and their
secretions as hormones. The thyroid gland, which lies athwart the
Adam's apple, secretes thyroxin, which plays a very important part in
both physical and mental growth. The suprarenal capsules, situated

just above the kidneys, secrete adrenalin, which helps to regulate the
amount of sugar in the blood. The most important of these duct-
less glands is probably the pituitary gland, which lies embedded in a
little bony socket at the base of the brain and is known to secrete a
dozen or more hormones, among others those which incite and regulate,
sexual manifestations as well as lactation. The gross picture, as far'
as the above endocrines are concerned, is the same for both male and
female.
However, the female has ovaries and the male testes, and, as was noted
earlier, these glands have an endocrine as well as a spermatogenetic or
ovogenetic function. In other words, the internal environment of males
and females is basically different, and herein lies the explanation of the
expression of these so-called" sex-influenced" characteristics. In sheep,
for instance, some breeds have horns in both sexes, some horns in the
males only, and still other breeds are hornless in both sexes. If ~,e crossed
a horned breed (hh) with a hornless one (H H), the F I males will be horned
(Hh) and the FI females (Hh) hornless. If ,ye go on and get the F2
from this cross, we'find that among the males there is a ratio of 3 horned: 1
hornless, among the females a ratio of 3 hornless: 1 horned. Clearly the
internal environment is influencing the expression of this characteristic.
Warwick and Dunkle I have shown that genes H (hornlessness), H'
(Dorset horns, i.e., in both sexes) and h (Merino and Rambouillet horns,
i.e., horns in male, knobs in female) form a series of multiple alleles.
A somewhat similar situation is found regarding the inheritance of the
very dark red (mahogany) color in Ayrshire cattle, the Mm genotype
producing mahogany and white pattern in males, but red and white in
females. In man, a certain type of baldness shows the same sort of
inheritance.
Male Female
mahogany and white MM mahogany and white
mahogany and white Mm red and white
red and white mm red and white
Still another sort of sex-influenced inheritance is that relating to cock

and hen feathering in poultry. As is well-known, the differences in the ~:.
plumage patterns between males and females are due to testicular and
ovarian hormones. If the ovary of a bird belonging to a race that nor-
mally exhibits differences in male and female plumage is removed and
a piece of testes transplanted into her, she will then proceed to develop
1 See Jour. Hered., 30(8):325-329.

354 BREEDING fiND IMPROVEMENT OF FARM ANIMALS
the male feather pattern. A dominant gene (H) for hen feathering is
known in poultry. Hens of whatever genetic constitution (HH) , (Hh) ,
or (hh) will be hen-feathered because they lack the proper internal
environment, specifically the male sex hormone, to produce cock feather-
ing. In males, the gene H prevents the development of cock feathering·
so that HH and Hh males are hen-feathered, whereas hh males are
cock-feathered. The characteristic is therefore seen to be sex-limited to i
the male and to be due to the interactions of genes with the internal
environment of the animal.
Probability.-Mathematics is coming to playa greater and greater
part in genetic research owing to the fact that inheritance is traceable to
particulate elements, genes, which, although functioning through a com-
plex system of reproductive physiology, yet behave in an orderly, specific,
predictable manner. Mendel's early success in discovering the basic laws
of inheritance was due largely to the fact that he kept accurate records of
the appearance of characteristics by separate generations and found that
they followed a certain statistical regularity.
If we toss a penny, it has an even chance of falling heads or tails, the
chance of its falling either being H. If we toss two pennies, the chance
that one of them will fall heads is H, the chance that the other will fall
heads is 72, and the chance that both will fall heads is the product of the
likelihood of each separate event or H X Yz = 74:. The same situation
prevails for the likelihood of both falling tails. The chance that one will
fall heads and the other tails is Yz X Yz or n, or that the former will fall
tails and the latter heads is 71 X 71 or 74:, so that the chance of getting
one head and one tail simultaneously is 74: + 74: or Yz. Likewise if we
toss three pennies we may get all heads, in fact, the chance that we will
is definitely known to be Yz X Yz X 7~ or 1 chance in 8, and the chance
that we will get two heads and one tail is again the product of the sepa-
rate probabilities; i.e., two heads = Yz X Yz or 7:1 X one tail = Yz or
Ys X 3 = %, because with three coins being tossed the two-heads-an9-
one-tail combination can appear from any of three combinations. Like-
wise, we would have a % chance of getting one head and two tails and a
Ys chance of getting three tails.
This is seen to be simply an expansion of the bionomial (a + b)n,
where a = Hand b = 71 and n is any number. We can let a represent
heads (chance 72) and b represent tails (chance 71). Therefore, we can
quickly find the answer to the question, "How many times in tosses of
three coins simultaneously will we get two heads and one tail? Expand-
ing the bionomial we have a 3 + 3a 2b + 3ab 2 - b 3 • Since ,ye are dealing
with the two heads and one tail combination and a stands for heads and
b for tails, we select the second term of the expansion 3a 2b and substitute
THE PRLVCIPLES OF HEREDITY 355
the values for a, (Yz) and for b, (Yz) and we have 3 X (Yz)2 X Yz or
3>< 74 X 7~= %.
The procedure is th~ same for any problem involving probabilities;
viz., first find the probability of each separate event and then substitute
these values in the proper term of the expanded bionomial. If, for
instance, we wanted to know how many times in clutches of 8 eggs from
birds that are heterozygous for both A and B, i.e., of the genetic consti-
tution Aa Bb, we would get 6 feather-shanked and 2 clean-shanked birds,
first, we must remember that these are duplicate factors and that either
A or B or both will produce feathered shanks, while aa bb is clean-shanked.
The ratio from these parents is then 15 feathered: 1 clean. We will let a
stand for feathered and b for clean. Now our term in the bionomial
expanded to the eighth pO'wer is a 8 + 8a 7 b + 28a6 b2 , and we now sub-
"
stitute the values for a an d b, glVmg 28 (15)6(
- 1
-1 )2 =318,937,500 " or
16 6 4,294,966,296
roughly 31'3. In other words, once out of 13 times \ve can expect to get 6
feathered and 2 clean-shanked birds from clutches of 8 eggs when the
parents are double heterozygotes for the two pairs of genes that act in a
duplicate fashion in producing this characteristic.
In cases where there are three possibilities, e.g., in a cross between two
red hogs that are double heterozygotes for the two pairs of genes either
dominant of which by itself produces sandy-colored hogs in a 9: 6 : 1 ratio,
we would have to expand the trinomial (a + b + c)n.
In all cases the chromosomes assort and recombine at random accord-
ing to the laws of probability.
Gene Manifestations.-Genes are the genetic representatives of specific
characteristics. The exact nature of the physicochemical make-up of the
genes is unknown. At present, they are known simply by their actions.
It is the purpose here to point out some of the ways in which factors mani-
fest themselves.
The simplest gene manifestation is of the sort where apparently one
gene conditions the development of a certain characteristic, e.g., the fac-
tor P, which brings about polledness in cattle. As a contrast to this may
be considered the function of milk production, which is no doubt depend-
ent on a great many genes, for milk production is a complex physiological
phenomenon dependent among other things on the capacity, tempera-
ment, and vitality of the cow. One gene may have a very pronounced ....,.
effect on the individual; e.g., in peas, the gene T, which brings about tall-
ness, whereas genes tt bring about dwarfness; or the gene may cause the
early death of the individual (lethal genes). On the other hand, the gene
may condition the development of a minor characteristic, such as eye
color. The white eye in Drosophila, however, is only one of the several
characters that such mutants exhibit, some of the others being a lowered
fertility and vitality; thus, one gene apparently affects many character-
istics. The individuals of certain races may exhibit a good deal of vari-
ability, but the evidence would seem to point not to the unstable nature
of the gene but rather to environmental effects as the source of this
variability.
Entirely different genes may also bring about the identical character-
istic in the organism. 1
We find, in experience, that we cannot safely infer from the appearance of
the character what gene is producing it. There are at least three \vhite races of
fowls, produced by different genes. We can synthesize white-eyed flies that are
somatically indistinguishable from the ordinary white-eyed race, yet they are the
combined product of several known color-producing genes. The purple eye
color of Drosophila is practically indistinguishable from the eye colors maroon
and garnet.
It should be emphasized that determiners are not characteristics. An

animal does not inherit characteristics from its ancestors, but rather it
does inherit potentialities. Whether the potentiality develops to its
fullest extent in the animal depends upon the environment, and the term
environment is here used in its widest sense. The new organism has an
environment from the minute the egg is fertilized by the sperm; or \ye
might go even further back and consider that the egg and sperm each
has an environment before fertilization has been accomplished. The
point is that the environment, speaking generally, may inhibit the full
expression of potentialities from a time preceding fertilization until
physiological maturity has been attained. Developmental aberrations in
utero may affect the embryo deleteriously from either an anatomical or a
physiological standpoint; and the same is true of a multitude of environ-
mental conditions following birth. For example, even though an animal
inherited the potentiality for large body size or high milk production, the
end results actually obtained are capable of extensive modification
through feeding and management. Likewise many people inherit the
potentiality for having considerable pigment in their skin, but this is
conditioned by the action of sunlight. In other words, we tan in summer
and bleach in winter.
The Gene.-Up to the present, genes are known only indirectly; i.e., we
know them by their actions. That they are chemical structures of some
sort is, of course, a generally accepted opinion. Mendel postulated some
sort of entities within the germ cells that were held responsible for the
appearance of certain somatic characters. Later genetic studies sub-
1 MORGAN; T. H., "Physical Basis of Heredity," pp. 239-240, J. B. Lippincott
Company, Philadelphia, '1919.

stantiated this idea, and the genes for many characters have been defi-
nitely located in certain chromosomes in relation to other genes in the
same chromosome on the basis of their behavior in hereditary transmis-
sion. In brief, the gene is thought of as an organic unit, located in a
definite place in the chromosome, which is in some manner capable of
reproducing itself, and the genes in turn are held to be the responsible
agents in making possible the appearance of all the characteristics of a
new organism.
Various estimates have been made regarding the size of the gene, rang-
ing from 10 to 70 mjl as the upper limit. In any event they must be
exceedingly minute. Because of the fixed wave length of visible light,
the smallest objects that can be seen distinctly with the highest powers
of the microscope! are of the order of 250 mjl, i.e., about 1/100,000 in.
(1/100,000 in. compares with 1 in. as 1 in. compares ,vith 1.58 mi.) By
use of the ultraviolet light with its shorter wave length, objects of the
order of 100 mjl have been photographed. Ordinary bacteria, without
distinguishable nuclei or chromosomes but presumably containing genes,
are of the order of 500 to 750 mjl, and the filterable viruses are thought to
be of the order of about 10 to 250 mjl.
Summary.-In this chapter we have considered several additional
types of inheritance. We have learned that there may be more than two
alternative forms of a gene making manifest the principle of multiple
alleles. We also have seen that many of the commercially valuable
characters of our animals are determined by the interactions of a great
many pairs of genes. It is this feature, plus the fact that our larger
animals have so few offspring, that makes animal breeding so difficult.
We have also become acquainted ,vith the action of lethal genes as well
as the various types of inheritance associated with or dependent upon the
sex of the individuals. Finally, we have seen that the transmission of
potentialities from ancestors to offspring behaves in an orderly fashion
which is predictable on the basis of the laws of chance or probability, that
the final result of gene action may follow a variety of patterns, and that
cytology is working ever closer to a satisfactory explanation of the nature
of the basis of all living matter, the gene.
References 2
Problems
1. If two chinchilla rabbits produce both chinchilla and albino offspring, what
are the genotypes of all these animals?
2. Could two chinchilla rabbits produce both Himalayan and albino offspring?
1 The current development of the electron microscope may soon permit much more
detailed physical knowledge of the chromosomes and genes.

2 See lists at end of Chaps. XI and XII.
3. What are the genotypes of the parents in the following rabbit crosses:
(a) Colored X colored give 3 colored and 1 albino.
(b) Colored X chinchilla give 72 colored; 7:i chinchilla; 7:i Himalayan.
(c) Chinchilla X Himalayan give 72 chinchilla; 72 Himalayan.
(d) Colored X Himalayan give 72 colored; Y2 Himalayan.
(e) Chinchilla X Himalayan give 72 chinchilla; 7:i Himalayan; 7:i albino.
4. What will be the phenotype, as to blood groups, of offspring of parents of the
following genotypes for blood groups:
Aa X aba
Aab X aba
aba X aba
5. If a person of blood group AB marries one belonging to group 0, what will be
the blood groups of their children?
6. N OTE.-In the three following problems on blood groups, determine the geno-
types of the parents.
One parent is group A and the other group B, but all four groups are represented
among the children.
7. Both parents are group A, but % of the children belong to group A and % to
group O.
8. One parent is AB and the other B, but of the children >4 are A, X AB, and
>'2 B.
9. In the two following cases of disputed paternity, determine the true father of
the child: Ca) The mother belongs to group B, the child to 0, one possible father to A
and the other to AB. (b) the mother belongs to group B, the child to AB, one possi-
ble father to A and the other to B.
10. In the choice of donors for blood transfusion, a patient's brother or sister is
often selected. Would these be more likely to be successful donors if both parents
belonged to blood group AB or if both belonged to group O? Explain.
11. If both parents belong to blood group AB, what proportion of their children
would be expected to be of such a type as to be able to give blood to their parents?
12. Consult the chromosome map for Drosophila (p. 35i) and determine how many
dominant genes must be present in each chromosome to give the normal red eye color.
13. If a rose-combed creeper male is mated to a rose-combed creeper female and
produces a number of single-combed normal birds, what were the genotypes of the
parents?
14. N oTE.-In poultry, creeper is a semidominant lethal, barring a sex-linked
dominant.
A cross of creeper, barred female and normal, nonbarred male would give what
offspring?
15. A cross of creeper, heterozygous barred male and creeper, nonbarred female
would give what offspring?
16. A cross of creeper, double-heterozygous walnut-combed, barred female and
normal, rose-combed, heterozygous barred male would give what offspring?
17.* A sex-linked recessive lethal factor is known in poultry. What would be the
sex ratio of the offspring of a male, heterozygous for the lethal, crossed with normal
females?
* NOTE.-In this list the problems that are starred are taken from Snyder, "Princi-
ples of Heredity," D. C. Heath and Company, Boston, 1940, and are reproduced here
through the kind permission of the author and publishers.
18. * In a certain series of matings between normal pigs, 38 offspring were born.
Of these 29 were normal, and 9 had greatly swollen forelegs. The latter lived only
a few hours. How could these results be explained genetically?
19. * In the Tzouracana sheep of Romania, gray individuals and black individuals
are found. The black animals when bred together give all black offspring. Black
mated to gray results in approximately equal ra,tios of black and gray lambs. When
grays are mated together, approximately two-thirds of the offspring are gray, and
one-third black. What kind of inheritance appears to be involved?
20. * In a certain strain of Oldenburger horses, tracing back to the mare, Jelka, the
females of the line regularly produce only half as many male offspring as female
offspring. What genetic explanation can you suggest for this?
21. * In cattle, the polled condition is dependent upon a dominant factor (P), the
horned condition upon its recessive allele (p). In the Dexter-Kerry cattle the short-
legged (Dexter) condition is dependent upon a dominant lethal factor (D), the long-
legged (Kerry) condition upon its recessive allele (d). Suppose a Dexter bull homo-
zygous for polled were mated to a horned Dexter cow. 'What kinds of calves could
they produce, and in what proportions?
22. * Suppose two polled Dexter animals were mated and produced a horned Kerry
calf. What would be the genotypes of the parents?
2S. * What kinds of offspring would be expected from the mating of two horned
Kerry animals? Two horned Dexter animals?
24. NOTE.-Assume that in man the difference in skin color between Negro and
white is due to two pairs of factors; that AA BB is "black" and aa bb "white"; and
that any three of these factors produce "dark" skin; any two, ,imedium"; and any
one "light."
What will be the skin color of the offspring from a mating of white with black?
From a mating of two individuals genotypically like these Fl offspring?
25. What are the genotypes of the parents in the two following matings of Kegroes:
medium X light, giving % dark, % medium, % light, P-S white; median X light,
giving 72: medium and 72: light?
26. * A variety of squash bearing 6-lb. fruits is crossed with one bearing 3 lb. fruits.
The F 1 plants bear 472: lb. fruits. Among the F ,fruits there is considerable variation,
but out of 200 such fruits there are three which weigh as little as 3Ib., and three which
wl'igh 6 lb. How many pairs of factors are responsible for the difference in weight
between the two parent lines, and how much does each effective allele contribute to
this difference?
27. * How many different homozygous strains bearing 4-lb. fruits could be
developed?
28. * A race of oats yielding 10 g. per plant was crossed with a race yielding 4 g.
per plant. The F 1 plants yielded 7 g. per plant. Out of 253 F 2 plants, four yielded
as much as 10 g. per plant and four yielded as little as 4 g. per plant. How is yield
inherited in these oats?
29. * In the preceding problem, how much does each effective allele contribute to
yield?
SO. * Two different races of corn each averaging 68 in. in height were crossed.
The Fl also averaged 68 in. in height. In the F 2 , however, there was considerable
variation, ranging from 36 to 100 in. Out of 1,942 F2 plants, eight reached a height
of 100 in., and seven were as short as 36 in. Explain these results in terms of genetic
factors, assuming that the environment was held similar for these plants.
N oTE.-Color blindness and hemophilia in man are sex-linked recessives.
31. A girl of normal vision whose father was color-blind marries a man of normal
vision, whose father was also color-blind. 'What types of vision will be expected
in their offspring?
32. A color-blind man marries a woman of normal vision. They have sons and
daughters, all of normal vision and all of whom marry normal persons. Where among
the grandchildren may color blindness be expected to appear? If there are cousin
marriages among these grandchildren, where among their offspring may color blind-
ness be expected to appear?
33. A man and woman, both of normal vision have (a) a color-blind son who has 1
daughter of normal vision: (b) a daughter of normal vision who has 1 color-blind and 1
normal son; and (c) another daughter of normal vision who has 5 sons, all normal.
What are the probable genotypes of grandparents, children, and grandchildren?
34. A brown-eyed woman with normal vision whose father was color-blind and
blue-eyed marries a man who is blue-eyed and of normal vision. W"hat offspring may
this couple expect as to eye color and vision?
35. A man (A) with hemophilia marries a normal woman (B). Their daughter
(0) marries a man (D) with hemophilia and produces 2 sons, 1 (E) with hemophilia.
The normal son (F) marries a woman (G) whose paternal grandfather (II) has hemo-
philia but whose parents (I and J) were apparently normal. 'Yhat are the genotypes
of all the individuals concerned?
36. A man's maternal grandmother had normel vision; his maternal grandfather
was color-blind; his mother is color-blind; his father is of normal vision. What are the
genotypes, as to vision, of the two parents and grandparents mentioned? What type
of vision has this man himself? V\nat type have his sisters? If he should marry a
woman genotypically like one of his sisters, what type of vision would be expected
in the offspring?
37. The mother of a right-handed brown-eyed woman of normal vision is right-
handed, blue-eyed, and of normal vision, and her father is left-handed, brown-eyed,
and color-blind. This woman marries a man who is left-handed, brown-eyed, and of
normal vision whose father was blue-eyed. ''{hat chance will the sons of this couple
have of resembling their father phenotypically?
38. What effect on the sex ratio would a recessive sex-linked lethal factor have in
man?
NOTE:-Barred plumage in poultry is a sex-linked dominant.
39. In poultry, if a nonbarred cock is crossed with a barred hen, and an F, female
from this cross is mated with her father and an F, male with his mother, what will be
the appearance of the offpsring of these last two crosses, as to barring?
40. A single-combed barred cock crossed with a walnut-combed barred hen pro-
duces the following offspring:
4 rose, barred males.

5 walnut, barred males.
2 rose, barred females.
3 rose, nonbarred females.
2 walnut, barred females.
2 walnut, nonbarred females.
What are the genotypes of the parents?

41. In poultry, assume high egg laying to be determined by two factors, one of
which, M, is sex-linked and the other, L, not sex-linked. A breeder has two high-
producing hens and five males of unknown genotypes. Each hen is bred to each male
with results as follows:
Hen 1 X male 1 gives % high, Yz medium, and Ys low layers.

Hen 1 X male 2 gives 74 high, Yz medium, and 74 low.
Hen 1 X male 3 gives Yz high, Yz medium.
Hen 1 X male 4 gives % medium, 74 low.
Hen 1 X male 5 gives 7-2 medium, Yz low.
Hen 2 X male 4 gives all medium.
Hen 2 X male 5 gives all medium.
wnat are the probable genotypes for fecundity of these seven birds?
From which of these 10 crosses would you be most likely to get the best males for
producing high-laying hens?
42. N OTE.-- Vi'hite eye in Drosophila is a sex-linked recessive.
In Drosophila, if a white-eyed female is crossed with a red-eyed male and thc
F2 allowed to interbreed freely, what will be the appearance of the F, as to eye color?
43. In Drosophila, if a homozygous red-eyed female is crossed with a white-eyed
male and the F2 allowed to interbreed freely, what will be the appearance of the F3
as to eye color?
44. In Drosophila, vestigial wings (v) are recessive to the normal long wings (V),
and the gene for this trait is not in the sex chromosome. If a homozygous white,
long female is crossed with a homozygous red, vestigial male, what will be the appear-
ance of the Fl? Of the F2? Of the offspring of a cross of the FI with each parent
type?
45. In Drosophila, two red-eyed long-winged flies when bred together produce the
following offspring: females, % red, long; >4
red, vestigial; males, % red, long; %
white, long; Ys red, vestigial; Ys white, vestigial.
Wfiat are the genotypes of the parents?
46. In Drosophila, a cross between bar-eyed female and wild-type (round-eyed)
produces only bar-eyed male and female in the F1. Wild-type female crossed with
bar-eyed male produces bar female and wild-type male. Explain the inheritance of
bar eyes and predict the appearance of the F2 from each of these crosses.
47. If a white-eyed nondisjunctional female Drosophila, (Xr) (Xr)Y, is mated to
a red-eyed male, what kinds of offspring may be expected as to sex and eye color?
48. A factor 1 in Drosophila is recessive, lethal, and sex-linked. If female Ll is
crossed with a normal male, what should be the sex ratio of the progeny?
49. In Drosophila vermilion eye color is recessive and sex-linked. In exceptional
cases vermilion female crossed with normal male produces, in addition to the usual
vermilion male and red-eyed female, a few vermilion female and red male. Explain
this result and predict what classes of offspring should appear when the vermilion FI
females from above are crossed with red-eyed males.
60. In Drosophila yellow body color (y) is sex-linked and recessive to the gray (Y)
body of the wild fly.
If a yellow female is crossed with gray male, and (a) if an FI female from this
cross is mated with her father and (b) an FI male mated with his mother, what will
be the appearance of the offspring of these last two crosses as to hody color?
51. If a hen that undergoes sex reversal, and thus becomes a functional male,
produces gametes of the same chromosomal constitution as before (although they are
now sperms instead of eggs), what will be the sex of her offspring when she is mated
with a normal hen?
52. If such a sex-reversed hen were barred, what would be the appearance of her
offspring when bred to a nonbarred hen?
53. * A bald man whose father was not bald marries a nonbald woman whose mother
was bald. What are the genotypes of these two people in regard to the factors for
baldness and nonbaldness? What kinds of children can they have in regard to these
characters?
54. * A nonbald, normal-visioned man marries a non bald, normal-visioned woman
whose father was color-blind and whose mother was bald. What kinds of offspring
may they have, and in what proportions?
55. * A nonbald man marries a nonbald woman. They have a son and a daughter.
At the age of thirty-five the son becomes bald. What are the chances that the
daughter will also become bald because of her genetic constitution?
56. * A red-and-white Ayrshire cow whose mother was mahogany-and-white is
bred to a red-and-white Ayrshire bull. If she produces a male calf, what are the
chances that it will be mahogany-and-white? If the calf is a female, what are the
chances that it will be mahogany-and-white? This is a sex-influenced character.
57. * Outline thc breeding procedure necessary to establish a homozygous mahog-
any-and-white Ayrshire herd.
58. * Outline the breeding procedure necessary to establish a homozygous red-and-
white Ayrshire herd.
59. NOTE.-In sheep, white fleece (W) is dominant over black (w); and the horned
condition (h) is dominant over the hornless (H) in males but recessive in females.
If a homozygous horned, white ram is bred to a homozygous hornless, black ewe,
what will be the appearance of the F 1 and F 2 as to color and horns?
60. A horned, black ram bred to a hornless, white ewe has the following offspring:
Of the males, X are horned, white; X horned, black; X hornless, white; and X
hornless, black. Of the females, Y2 are hornless, black and Y2 hornless, white. What
are the genotypes of the parents?
61. * How many times in families of five would you expect four boys and a girl?
Four girls and a boy? Five girls?
62. * In a certain family there are six girls. What are the chances of the next child
being a boy?
63. * In human beings, deaf-mutism may be the result of the homozygous state of
either or both of two recessive factors d and e. Normal hearing results only when both
dominants D and E are present. 'When both parents are Dd Ee, how many times in
families of two would you expect two deaf children?
64. * ""here both parents are DdEe, how many times in families of four would you
expect three normal children and one deaf child?
65. * Two normal parents produce an albino son. What are the chances that their
next child will be a normal girl?
66. * In Shorthorn cattle, the factors for red and white are alleles but neither is
dominant, the heterozygous condition resulting in roan. Twins are sometimes born
in cattle. If twins are born in crosses between red bulls and roan cows, how often
should both twins be roan?
67. * In poultry, feathered shanks are the result of either or both of the dominant
factors F ltl\d S. Unfeathered shanks are the result of the double recessive condition.
'.
\
},
! \
In crosses where both parents are Ff S8, how many times in families of three off~pring
would all be expected to have feathered shanks? All three unfeathered shanks?
68. * If one parent were Ff S8 and the other.ff 88, how many times in families of four
offspring would you expect one feathered-shank bird and three unfeathered-shanked
birds?
69. * In poultry, the factors for black and for splashed-white are alleles but neither
is dominant, the heterozygous condition resulting in blue (the Blue Andalusian).
A pair of Blue Andalusians are mated, and the female lays two eggs. \Vhat are the
chances that both eggs will hatch out Blue Andalusians?
70. * If three eggs are laid by the bird mentioned in Prob. 69, what are the chances
that all three eggs will hatch out Blue Andalusians?
CHAPTER XIV
THE PRINCIPLES OF HEREDITY (Continued)
Thus far we have studied various types of inheritance that behave as

if certain characters ·were dependent on one pair of genes. Sometimes
these genes were in an autosomal chromosome, sometimes in a sex
chromosome, sometimes the gene was a dominant or a recessive lethal,
sometimes the manifestation of the character was influenced by the sex
of the individual. We also discovered that the gene may have more than
just two alternative forms, that there may be three or many more genes
in a series of multiple alleles.
In addition, we have studied the behavior of two genes simultaneously,
both in cases where each pair of genes was responsible for the production
of a separate character and those in which the two pairs of genes acted
on only one character. In the latter cases, we met many exceptions to
the general dihybrid 9: 3: 3: 1 ratio due to various types of epistasis,
and we also learned that lack of dominance in one or both pairs of genes •
can change this ratio. If, however, each of two pairs of genes causes the
development of a certain separate character, or in some cases (e.g., comb
pattern in poultry) both act on the same character, if there is dominance
between the members of each pair of genes, and if the pairs of genes are
located in different chromosomes, then the F2 will give the usual 9:3:3: 1
ratio. If instead of mating two F l'S (Aa Bb) to get the F 2 we made what
is termed the backcross by mating an Fl (Aa Bb) to the double recessive
(aa bb), then we would get equal numbers of four different phenotypic
types according to the following diagram.
Fl (Aa Bb) germ cells .................... AB Ab aB ab
Double-recessive (aa bb) germ cells ab . ..... AaBb Aa bb aa Bb aa bb
Because each chromosome consists of scores or hundreds of genes, it is

logical to suppose that we might sometimes be dealing with two pairs of
genes that are located in the same chromosome. In such a case the genes
would not assort independently, although the chromosomes would.
Genes located in the same chromosome show varying degrees of linkage.
364
Linkage.-In working with sweet peas in 1906, Bateson and Punnett

made a cross involving purple flowers and long pollen grain in one parent
and red flowers and round pollen grains in the other. Because crosses of
purple and red had given a 3: 1 ratio in the F2 (purple being dominant)
and crosses of long and round pollen grains had likewise given a 3: 1 ratio
in the F2 (long pollen grains being dominant), it was expected that the
F2 of the purple, long crossed with red, round would thus give a ratio of
9 purple, long: 3 purple, round: 3 red, long: 1 red, round. Such, how-
ever, was not the case, for out of 6,952 F2 flowers 4,831 (instead of the
expected 3,910.5) were purple, long; 390 (instead of 1,303.5) were purple,
round; 393 (instead of 1,303.5) were red, long; and 1,338 (instead of
434.5) ,vere red, round. Combinations of the two genes that had gone
into the hybrid together (purple and long from one parent and red and
round from the other) were much too numerous in the F 2 • Actually
there was a total of 6,169 of these two classes where only 4,345 had been
expected. Bateson and Punnett called this feature coupling, and its
opposite, "where a dominant and a recessive were supplied by each parent
(purple and round and red and long) and still too few of the new types
appeared in the F 2, repulsion, these of course being different aspects of the
same problem.
Facts like these stood out as exceptions to Mendelian inheritance until
1910, when Morgan and his coworkers at Columbia University discovered
the basis for these phenomena from their work with Drosophila and
coined the term that describes them, linkage. If two genes with recog-
nizable end products are in the same chromosome, they will, of course,
have more or less tendency to be inherited together.
In Drosophila, if a cross between a wild-type (gray) body color with
long wings and a black body color with vestigial wings is made, the F I
all have gray bodies and long wings, for these patterns are dominant to
black and vestigial. If a male of this FI hybrid generation is mated to a
black, vestigial female, we would expect to get some gray, long; gray,
vestigial; black, long; and black, vestigial offspring. If the genes for
these characters were in different chromosomes, we would expect equal
numbers of the above classes. If they were in the same chromosome,
we would expect to get more than 50 per cent of the grandpa rental types
and less than 50 per cent of the new combinations. Actually, in this
case, we get 100 per cent of the grandparental types and 0 per cent of new
combinations. In other words, these genes remain completely linked
when tested by this backcross.
The backcross to the double recessive is used to determine the sort of
germ cells that the F 1 dihybrid is producing, because there is nothing in
the double recessive to hide what is going on in the dihybrid.
366 BREEDING AND IMPROVEMENT OF FARirf ANIMALS
The usual system of notation when dealing with linkage problems is

to use the plus sign for wild-type genes (usually dominant) and a letter
for the mutated gene. In the above case, the notation applicable with
independently assorting genes, e.g., GG LL for a gray-bodied long-winged
fly cannot be used because we know from the genetic behavior that the
genes for gray body and long wings are in the same chromosome and,
therefore, belong together. We could write the genetic make-up of such
a fly (GL) (GL) , but it is more convenient to write it ++1++ (plus
sign standing for the wild-type genes).
Thus a diagram of the above cross would be written
++ gl
P l ++ X {ji
gray, long black, vestigial
Fl ++
gl
gray, lo.ag male
and the backcross of an F 1 male to the double recessive
++ X f!!:. = ++ and f!!:.

gl gl gl gl
gray, black,
long vestigial
50% 50%
Crossing Over.-We have just seen that the backcross of a dihybrid .

gray-bodied long-winged male, ++ I gl, to a double-recessive black,
vestigial female, gllgl, gives only two sorts of offspring, 72 of them being
gray, long, + + I gl, and Y2 black, vestigial, gIl gl, because the linkage ~f
the G and the L genes and the g and I genes in the dihybrid male was
complete.
However, if we take a dihybrid female with gray body and long wings,
++ I gl, and mate her to a double-recessive male, gIlgl, we will secure four
kinds of offspring instead of two, viz., gray, long; black, vestigial; black,
long; and gray, vestigial. The reason for this evidently is that there has
been crossing over between the loci for body color and length of wings,
as illustrated in Fig. 102. The end result is that the genes G and Land g
and 1 in this female dihybrid usually stay together but occasionally
through crossing over get into the new combinations of G and I and g
and L, or +l and g+.
The following schematic diagram shows the manner in which the new
gene combinations are formed. We have learned earlier that the pairs of
homologous chromosomes come to lie side by side during the early stages
of germ-cell production in the process called synapsis. In the early stages
of synapsis, each member of each homologous pair of chromosomes repro-

duces itself so that the pair of homologous chromosomes becomes two
pairs of closely associated sister chromatids. Crossing over, or the equal
interchange of blocks of genes in sections, apparently takes place between
two nonsister chromatids after the two chromosomes have become bundles
of four chromatids. In other words, crossing over involves two chrom-
atids rather than two chromosomes.
In mitosis, or somatic cell division, the chromosomes shorten and
thicken and arrange themselves in the equatorial plate (prophase). They
next split or reproduce themselves
G~ ~G 9~ ~q
(metaphase) so that each chromo-
some is now double. The mem- x
bers of each exactly similar
doublet now start to move- t9
opposite poles of -the cell (an8,~ L~ ~ L ~ ~
I 1
.~~Y ~ vestigi~1
phase). Finally the migration is
complete (telophase), and the cell s long Black
wall constricts, forming two cells
09 G~q Go 0'3
--
G
exactly like the parent cell. In
other words, in mitosis the
chromosomes appear in the diploid
number of paired threads. In L
-
IlL I L
meiosis, or germ-cell production,
however, they appear as the F, femG\le dihybricl
'd d FIG. l02.-Schematic crossing over between
hap101 number of paire threads. loci for body color and wing length.
This latter feature is due to
synapsis, or conjugation of the homologous chromosomes, which is specific
in that all the allelic genes are attracted to each other and come to lie
side by side in a regionally specific fashion-gene A with A or a, gene B
with B or b, etc.
In the earliest prophase stage of meiosis, the chromosomes are strung
out into long unpaired threads in the diploid number (leptotene stage).
They next are attracted into homologous pairs (zygotene stage or synap-
sis). They next split or reproduce into bundles of four chromatids, two
from each chromosome (pachytene stage). Finally in the last stages of
prophase (diplotene, diakinesis) they show chiasma as the paired chrom-
atids begin to open out. During these four stages the chromosomes have'
been shortening and thickening.
Each group of two sister chromatids is held together by an unstaining
portion of the chromosome called a centromere. These centromeres
appear to repel each other, and, as the bundle of four chromatids begins
to be drawn apart by the retreating centromeres pulling each pair of
two sister chromatids (formed from one chromosome) to opposite poles

along the plane of conjugation, two of the nonsister chromatids often
appear to be crossed, these crosses being chiasmata (singular, chiasma).
These crosses are apparently due to one of a pair of sister chromatids
breaking at the original point of the chiasma and becoming attached to
the remaining portion of one of the members of the other pair of sister
chromatids and the remaining portions of the two broken chromatids
themselves uniting into a whole chromatid as diagramed in Fig. 103.
lt is obvious that, if crossing over took place in every cell containing
the chromosomes diagramed in Fig. 103, the result would be equal num-
bers of GL, Gl, gL, and gl germ cells and would amount to 50 per cent of
crossing over. Such a situation could not be differentiated genetically
from independent assortment. Actually such a situation is not known.
The greatest amount of crossing over yet found amounts to about 48 per\'---·
cent, the least~r cent as illustrated earlier in the Drosop~ss
':j~ L\lt
N
~ ~l,fi,
II
G G 9/19
}
rr 91/'
(, \ I
~\I\l L L't L
c \ I
\\\ \ \ I
\ I \ I
h\ \\
\
\ \ I
\\ • \ I
\\ '\
';{ \\
\\ \\ \, I
FIG. l03.-Diagram of chiasma at x with resulting crossing over between the genes G and
L, thus resulting in the production of germ cells containing GL, Gl, ilL, and III genes.
involving the use of a GL/gl male on gl/gl females, from which only
GL/gl and gl/gl offspring resulted.
lt has been found that the number of chiasmata at any given region
is about twice the amount of crossing over. Again referring to Fig. 103,
if we were considering 100 sets of these chromosomes and found no
chiasma in 80 of them, all these resulting g€rm cells would contain the
original combinations GL and gl. The other 20 per cent would show
chiasma, but 25 per cent of the resulting germ cells ·would contain GL
and 25 per cent gl, whereas the other 50 per cent (of the 20) would be Gl
and gL (25 per cent of each). In other words, there were 20 per cent
of chiasma to but 10 per cent of crossovers. Crossing over takes place
between only two of the four chromatids, in something less than 100 per
cent of the possible cases and always between two nonsister chromatids.
Crossover percentage, of course, is obtained by dividing the total
number of crossovers by the total population. Morgan gives,l among
others, the following instances of crossing over in Drosophila:
I MORGAN, T. H., "The Physical Basis of Heredity," pp. 87-88, J. B. Lippincott
Company, Philadelphia, 1919.

THE PRIA'CIPLES OF HEREDITY 3(i9
When a black fly with vestigial wings is crossed to a wild-type (" gray") fly
with long wings the offspring are, as we have already seen, gray, long. If one
of the F 1 females is back-crossed to a black vestigial male there are four kinds of
offspring produced; viz., the two original combinations, black vestigial, and
gray long; and in addition two recombinations of these; viz., black long, and gray
vestigial. The two latter classes are called the crossover classes, or, more briefly,
crossovers. The percentage of crossovers is definite for a given stock, of a given.
age, and under given environmental conditions. In this case the percentages
are as follows:
l\on-crossovers Crossovers
Black vestigial Gray long Black long Gray vestigial
41.5 per cent 41.5 per cent 8.5 per cent 8.5 per cent
83 per cent 17 per cent
If a pair of chromosomes in the Fl fly is represented as carrying the genes of

the characters here involved, one member of such a pair carries both a gene for
black and a gene for vestigial; the homologous member of the pair of chromo-
somes carries both of the normal alleles; viz., a gene for gray and a gene for long
wings. When crossing over takes place so that a gene for black goes over into
the other chromosome, the converse phenomenon takes place, a gene for gray
goes over into the chromosome that gave up its black gene. It is the constancy
of this interchange that makes the phenomenon reducible. to exact mechanical
treatment. "
Using the usual notation, we would have, in this last cross of an F~

female from a gray, long crossed with black, vestigial mated back to a
double-recessive black, vestigial, the results indicated in the following
diagram.
++ gl
++ X (it
gray,,,,,- /black,
long "" vestigial
++ female X gzgl = backcross
F l , '{jl
Fl germ cells plus double-recessive germ cell, ~-:- ~ gt +l

gr
gray, blaek, black, gray,
long vestigial long vestigial
numbers, 41.5% 41.5% 8.5% 8.5%
nonerossovers crossovers
Strength of linkage = 83% Crossover percentage = 17%
The reduction in the number of chromosomes in the germ cells is

accomplished by the two meiotic divisions. In the first, the chromosomes
aiter synapsing become double through reproducing or splitting, each
chromosome forming two chromatids closely applied to each other. This
bundle of four chromatids then separates along the plane of conjugation
370 B REEDING AND I M PROVEMENT OF FA R M ANI MALS
(with possible crossing over), two closely applied sister chromatids passing
to each secondary spermatocyte or oocyte, this division being spoken of
as the heterotypic division. These secondary spermatocytes then quickly
go through another cell division, at which time one member of each pair of
sister chromatids goes to each resulting cell. This is known as the
}wmotypic division, and through these two rapid cell divisions the number
of chromosomes is both reduced to one-half and the possibility of recom-
bination of blocks of genes from the two parents provided.
It is readily seen in Fig. 103 that if another pair of genes C and c
were located near the bottom of the chromosomes, and if another chiasma
appeared at point y, we might have a repetition of the above-described
1 2 3 4: 5
FIG. 104.- Diagrammatic representation of single crossing over between homologous
chromosomes 1- 3; and double crossing over 4 and 5.
process with the result that genes GlC and gLc might be located in two
chromatids. This would be a case of double crossing over, and, although
two crossings over had occurred between G and C, they would still be in
the same chromatid, which makes it appear genetically that no crossing
over had occurred between them. Thus each case of double crossing
over hides two actual crossovers, and allowance must be made for this
fact in ascertaining crossover percentages involving situations of this
sort.
If, in any cross involving genes located in the same chromosome and
thus showing various degrees of linkage, the crossover percentage is
found to be 20 per cent, the same fact can be stated by saying that the
strength of the linkage is 80 per cent.
Under standard conditions the percentage of crossing over remains con-
stant between any two pail's of genes. However, experimental work has
shown that a variety of internal and external conditions may affect this
rate. Among these influences are sex, no crossing over in the male of
Drosophila normally, although this can be induced in the autosomes by
X-ray treatment; chromosomal aberrations such as the inversion of a sec-
tion of the chromosome, translocations, etc., as well as gene mutations;
and, in addition, age and temperatures.
Much of the study of crossing over is made by means of the back-cross.
However, the F2 is sometimes used, and mathematical formulas have been
devised for the interpretation of the results.
If ,ye were dealing with a case that showed 20 per cent· of crossing over
between two genes AB and ab, then the F2 ratios could be found by t.he
usual checkerboard arrangements, as illustrated in Fig. 105.
4AB 1 Ab 1 aB 4 ab
(40%) (10%) (10%) (40%)
4AB 16 (AB) (AB) 4 (Ab) (AB) 4 (aB) (AB) 16 (ab) (A B)

(40%)
1 Ab 4 (AB) (Ab) 1 (Ab) (Ab) 1 (aB) (Ab) 4 (ab) (Ab)
(10%)
1 aB 4 (AB) (aB) 1 (Ab) (aB) 1 (aB) (aB) 4 (ab) (aB)
(10%)
4 ab 16 (AB) (ab) 4 (Ab) (ab) 4 (aB) (ab) 16 Cab) (ab)
(40%)
FIG. 105.-Checkerboard showing the expected composition of the F2 from a cross between
individuals differing in two linked genes which show 20 per cent of crossing over. The F2
ratio is 66 AB: 9 Ab: 9aB: 16 abo
Cytological proof~e~ ----_.-

Crossing Over.-We have 100,ked at the
_evidence that crossing over takes plac~.:! We saw that a cross involving
a gray-bodied long-winged female Drosophila of genetic constitution
++/gl and a double recessive male gl/gl gave four types of offspring,
++/gl, gl/gl, +l/gl, and g+/gl, in the proportion of 41.5 per cent gray,
long; 41.5 per cent black, vestigial; 8.5 per cent gray, vestigial; and 8.5
per cent black, long owing to the supposed fact that crossing over between
these points occurred Itn 17 per cent of the germ cells of the female. To
clinch the matter or to establish this supposed fact, we need cytological
evidence that it actually occurs.
This is difficult to get because the chromosomes and their various parts
generally lack distinguishing marks. However, Stern, working with
Drosophila, and Creighton and McClintock, working with maize, have
furnished this proof. Stern found a strain of Drosophila in which a por-
tion of the Y chromosome had been broken off and become attached to an
X chromosome. He found another strain in which a portion of an X

chromosome had been broken off and become attached to the IV chromo-
some. A cross between these two races gave him a race in which both the
X chromosomes had distinguishing marks and could be differentiated
"if
B
I
l
#fL
i
p-
Parents
j-
crt
!L
. fi-
Females
leIG. l06.-Diagram of Stern's experiment demonstrating crossing over cystologically.

See text for details. (From Stern, after Sinnott and Dunn.)
microscopically from all ordinary X chromosomes. Stern now made up

a stock of flies that had the genes for .£a~natLoJ"l ~ye colQr.. and the bar e~
..pattern
'
in the upper end of the broken X chromosome and the gene for
.....•. ---,-
red eyes 'and round eye pattern in the upper end of the X chromosome
\.
THE PRINCIPLES 01/ HEREDI'PY 373
with the attached piece of the Y chromosome, the red eye and the bar
pattern being dominants.
Red-eyed and bar-eyed females of the genetic constitution crB* / + +
were then mated to males of the genetic constitution cr + i.e., round eyes
of carnation color in their only X chromosome (round' being the wild-
type eye shape but recessive to the mutant bar pattern). From this
croSS we would expect four types of females if crossing over between the
two pairs of genes for eye color and eye shape had occurred. Stern
secured the expected four classes of females, viz., carnation, bar; red,
round; carnation, round; and red, bar. The first class should have had
the broken X chromosome plus a normal X from the male. The second
should have had one X chromosome with the attached piece of the Y
chromosome plus a normal X from the male. The third should have
had two normal X chromosomes, one from the male and the other as the,
result of crossing over between the two marked chromosomes of the
female at a point between the two genes being considered, whereas the
fourth should have had one normal X from the male plus a broken X
chromosome with the piece of the Y chromosome attached to one of the
pieces. l
These are the results which should obtain if there had been an interchange
between the two X chromosomes of the mother at a point near the upper end
of the X and between the locations of the cr and B genes. Of the FI female
flies, 364 were tested by Stern and in all but 5 (and these presumably the result
of experimental error) there was a complete correspondence between the genetic
and the cytological facts. In other words, genetic crossing over was proved to be
accompanied by cytological crossihg over, an actual exchange of material between
homologous chromosomes.
"j Linear Order of Genes.-We have seen in the two preceding sections
that blocks of gene~~_Qr.§~"(?f hQmQ.logou"~_,£h!2mos<?!lles s~~eti!lle_a
. in~erch~!!i~h:.:2.9siti~p..~_~r: the_chr~o~omes sothaQ~w combinatioll~
of genes can be formed, even though-i!fie genes controlling these characters
are not in separ~t~ Cl:lromosome pairs. We learned too that th~"perceIlt::
age of the crossing over can be easily ascertained by dividing the total
crossovefsOytne Total' population.
Now, if we assu~e that the chromatids are as likely to form chiasma,
with resulting interchange, at one poi~t as ~t a~~ther, then the"percentagE'
of crossing over can be' taken as the measure of the distance between the
genes being studied.-' For, if cr~~~i-;:;_-g~~e~ isasapt to occui-at one poinf'
as at another then there will be less likelihood of a break occurring
* The bar-eyed mutation behaves as a dominant.
SINNOT'l" EDMUND N., and DUNN, C., "Principles of Genetics," 3d ed., p. 222,
1
McGraw-Hill Book Company, Inc., New York, 1939.

-r
IOe)
:yelloweS)
" O.± Hoiry win¥ (W)
'. 0.+ scute (H
,"0.3IethOlI-T
\ \0.6 broad (W)
\ 'I. prune eE)
\ 1.5 white(E)
n
O.
'2.
tel~raph
SToor (0
3.± ori 5+00 less (8)
6.! expoonded (W)
fYI)
m
O. l"Oughoid (E)
I
N
••• <Wl B)
bshooven
eyeless ~E)
rOTooTed(B)
Minute-I'll' (H)
"\ \{3.3.± fOlcet(E~
Notch (E
12.t Gull (W)
13. Tru nCOIte (W)
\ 4.5 Abnormal (B) 14.± dOichsous (8)
, '. 5.5 echinu$ (E) 16. Streak (B)
\ \6.9 bifid (W)
" '7.5 ruby (E)
\ 13] crossveinlE'ss (W)
\15.± club 011) 20. divergent CW>
17.:!: dE'ltex(W)
20. cut (W)
21. singed (H)
27.5 TOin (B) 31. elaeha (8) 26. sepia (E)
27.7 lozenge (E) 26.5 hOllry(B)
35. Ski-U 011)
33. vermillion(E) y
35.1 miniOlture(W) 41. Joommed (W) 35. rose (E)
36.2 dusky (W) 36.2 creOlm-m (E)
38.± furrowed eE) 46.± Min ute-e (H) 40.1 Minute-h (H)
4aS blOick (8) 40.2 Tilt (W)
43. sooble (8) 48.1 jOlunty (w) 40.4 DichOiete (H)
44A (:lO!rneT eE) 42.2 threOid (6) moole fertiliTY
54,5 purple (E) 44. sCOJrlet (E)
5'1.5 cinnOibOlr(E)- 48. pink (E)
54.2 small wing 49.1 maroon eE)
54.5 rudiment"ry(W) 60.! safrOinin (E)
\15o.t dwarf (8)
56.6 forked (H) 50. curled (W)
57. BOir (E) 54.8 HOIiry Win& supr
58.5 smOillMe 64.± pink-wing(EW)
58.2 Stubble )
59. fusecl ) 58.5 spineless(H)
59.6 BeOidex
62. Minute-n H)
(Wd 67.
58.±
vestigiOiI (w)
telescope CW) _ 58.7 b ith orax (B)
-'59.5 bithorOlx-b
65. cleft ('N) _62. stripe (8)
72. Lobe eE) \63.1 glOiss (E) Long bris'I'leol
14.± gap(W) 66.2 eltOi ew)
10. bobbedCH) 75.5 curved (W) 69.5 hOJirless (H)
10J ebony (B)
72. banol (8)
'75.7 cardinal eE)
83.5 frInged (W) 76.2 white ocelli (E)
90. humpy (6)
91.1 rough(E)
99.5 arc eN) 93. crumplecl (W) mGlle fer-titi'l'y
100.5 plexus (W) 93.B BeOidec! (W)
102.± lethOiI-UOI 94.1 PrAinted(W)
(105. brown (E)
·\105.± blistered (W) 100.1 clooret eE)
106. purpleoic! eE) '1101. Minute nil
07.± morula (E)
01. speck (B)
107.5 brAlloon 0N) 106.2 Minute-g (H)
FIG. l08.-Linkage map for Drosophila melanogaster, showing relative position~ of many of
the known genes in the chromosomes as determined genetically. The letters in parentheses
indicate the portion of the fly in which the characters appear: B, body, E, eye; H, burs;
W, wings. The arrows indicate positions of spindle-attachment regions. In the Y-chro-
1Il0some, "Long bristled," which is the normal allelomorph of "bobbed," and the two
factors for, male sterility have not been precisely located. In chromosome IV the genes
are all very closely linked. [From Sharp, adapted from Morgan, Sturtevant, and Bridge<
(1925) and Stern (1929).1
376
Interference and Coincidence.-We have seen that crossing over may

take place between two nO]J_sister chro~l1tids a~ synapsis, which gives rise
to recombinations of genes, although they are located in the same chromo-
some, and that these crossover percentages are taken as the measure of
the distance between the genes in constructing chromosome maps based
on the assumption that the crossing over is as apt to take place at one spot
as at another. If two genes are very close together in the chromosome,
they might show complete linkage, i.e., the chromosome might never
break between them. Likewise, if genes are very far apart, there might
be two breaks, or crossovers, between them, so that in spite of this they
would still be in the same chromosome and appear as if no crossing over
had occurred between them. Therefore, the maps are built up on the
basis of crossing over between genes that lie relatively close together, up
to 10 units.
With three pairs of genes A, B, and C that were located in the same
chromosome we might get the sorts of combinations shown in Fig. 109.
Type I Type II Typem TypeN

I 2 I 2 I 2
loIIb
I 2
AI :0 AlIa AI !0
B
6 IbC
I b! 16 bl IB
C Ie
I
e I
I c i C Cl ie
Non· Two types of Double
crossovers single Crossovers crossovers
FIG. l09.-Diagram showing single and double crossing over.
~ow, if we were studying the linkage relations between genes A and

C, types I and IV above would appear similar genetically because genes
A and C and a and c occur together. It is evident, therefore, that, if
genes are relatively far apart in the chromosome, there will be more
actual cytological crossing over than is apparent genetically, because
of double crossing over, and that allowance must be made for this fact
mathematically.
If breeding experiments had shown genes A and B (Fig. 109) to have a
crossover value of 20 per cent and genes Band C 10 per cent, then we would
expect that genes A and C would show a crossover value of 30 per cent if no
double crossing over took place. However, genes A and C are far enough
apart so that double crossing over is possible and would infrequently
Occur. Such double crossing over would stiIlleave genes A and C and
a and c in the same chromosomes, the first marked by containing gene
b, the second by containing gene B, so that each double crossing over
hides or masks two actual crossovers. If, then, in the backc~oss -:;e
378 BRREDING AND IMPROVEMENT OF FARM ANIMALS
~ecured 71 per cent of type I, noncrossovers; 9 per cent of type II and

19 per cent of type III, single crossovers; between A and C; and 1 per cent
of type IV, double crossovers, between A and C, we would have to add
two times the 1 per cent, or 2 per cent, to the 28 per cent of single crossing
over between Band C and A and B to get the actual amount of crossing
over, or 30 per cent. Thus, when genes lie relatively far apart in the
chromosomes, the apparent amount o(cr;;'sing over will be less than the
ac-tual amo~nt because of double crossing over.. ' -
. Now, if the break between A and B occurs in 20 per cent of the cases
and between Band C in 10 per cent of the cases, and if these breaks are
/ entirely independent of each other, then they should both occur 10 X 2U,
or in 2 per cent of the cases, for the likelihood of two independent events
occurring at the same time is the product of the likelihood of each
J occurring separately. We saw, however, th_a~J:he brel1~ occurred in both
~ces in only 1 per cent of the c~~es.
Varying degrees of less than expected double crossing over are usually
secured. That is to say, these breakings and crossings over in different
regions of the chromosomes are not independent events but are influenced
by each other so that if a break occurs between any two given genes,
e.g., A and B, the likelihood of a break occurring between Band C is I
somewhat reduced. A!:rrea,k at a given point in a chromosome protects ~
~i.gl!Q..~_I"(:)gi_()Q_from allybre!l'k~~~ This principle is known!
as interference. If three genes lie very close together in a chromo- \
some, we may get breaks between the first and second or between the \ /
second and third pairs but never between both the first and second and)
the second and third simultaneously, i.e., no double crossing over. Such'
a situation holds, for example, between the mutant genes yellow (body),
white (eyes), and bifid (wings) in the X chromosome of Drosophila,
which show 1.5 per cent of crossing over between yellow and white and
5.5 per cent of crossing over between white and bifid. Here, if the
events were independent, we would expect both to happen in 1.5 X 5.5
or 0.08 per cent of the cases. J)1terference is sometimes expressed as a
coefficient of coincidence, which -Is secured by dividing the actual pro-
portion of double crossovers by the expected proportion. In the above
case we would have
o
0.0008 = 0, the coincidence value
A coincidence value of 0 means that interference is complete. Like-

wise a coincidence value of 1 would mean no interference or complete
independence of crossing over between the two regions of the chromo-
some. In the other example cited earlier in this section, the actual pro-
portion of double crossing over was 1 per cent, the expected 2 per cent
thus giving a coincidence value of 0.5.
Limitation of the Linkage Groups.-We have now discussed six
Mendelian principles, viz., segregation, independent assortment, linkage,
crossing over, linear order of the genes, and interference. There is one
final principle to be added to this list known as limitation of the linkage
groups. In Drosophila melanogaster, four linkage groups are known;
viz., a large group of sex-linked genes located in the I chromosome, two
large groups of autosomal genes in the large II and III chromosome pairs,
and a small group of genes located in the small IV chromosome pair.
TABLE 22.~LINKAGE-GROUP RELATIONS
Chromosome Known linkage

Species
pairs groups
Drosophila melanogaster . .......... . 4 4

D. willistoni. . . . . . . . . . .......... . 3 3
D. pseudobscuria . ................. . 5 5
D. virilis .. ....................... . 6 6
Corn ............................. . 10 10
Pea .............................. . 7 7
Tomato .......................... . 12 10
Morning glory. . . . . .. . ........... . 15 12
Mouse........... . ............. . 20 15
Rabbit ........................... . 22 11
Each new mutant in Drosophila has proved to be linked to other genes

in one or another of these four pairs of chromosomes. Such findings are
of course basic to the validity of the fundamental theory of chromosomal
inheritance. Obviously, if a fifth group of genes ,vere ever discovered in
D, melanogaster, the whole theory of chromosomes and genes would have
to be entirely revamped or discarded.
Table 22 shows the linkage-group relations in several species.
Application to Farm Animals.-The foregoing principles of inheritance
have been worked out from breeding experiments with plants, insects, or
small laboratory animals. These species reproduce rapidly, have large
numbers of offspring, and are inexpensive to grow. None of these three
features holds with our larger farm animals.
The 2!:,incipLes_ have been and are being used, however, to explain
various ty:p'~s of i~heritance in the larger farm mammals. Up to the
present, no contradictions to. established principles have been found,
although in many instances the principles must be applied in general
ruther than in ~pecific ways. Unfortunately in the higher animals we
have l~_rge numbers of chromosome~pair,ll and small numbers of offspring.

This, together with the fact that the environment IS so diverse and its
influence in many cases so great, leads"to inany difficulties in applying
the known principles in any detailed manner.
A knowledge of basic principles and an assurance that they hold for
all living forms gives the animal breeder a firm foundation on which to
base his breeding operations. At least the main routes toward his goal
of beautiful form and efficient function are now more clearly delineated
and should prevent his choosing an entirely wrong road or taking too
many expensive and time-v,asting detours.
Genes and Development.-It will be recalled that in Chap. XI we
listed three problems of genetics. One of these had to do with the actual
nature of the genes, about which very little is known at the present.
Another had to do with the modes of inheritance between ancestors and
offspring, with which the past three chapters have dealt and with which
the three following chapters will be concerned. The third problem listed
was that of the manner in which the characters latent in the zygote
became patent in the resulting offspring.
Thus far we have talked about genes and about their final end products,
characters, without attempting to say how the genes finally come to
express themselves as they do, or, in other words, what the causal relation
is between genes and chromosomes and expressed characters. We havei
pointed out that development results from the actions and interactions of
genes, cytoplasm, and environment, and that, although certain genes are
known by some final character, there are often many other conG...omitant
developments, as though this gene had one major and many minor effects,
and, finally, that most characters depend on the combined interactions of
many genes rather than on the specific action of one individual pair of
genes.
We know that the genes determine both the general characteristics of
an organism, i.e., to what species it will belong, as well as its individual
characteristics. We know too that all the nucleated somatic cells have
identical sets of complete chromosomes and genes. Although the genes
might function at different rates at different times, it seems obvious that
genes alone do not condition differentiation, but that we must invoke
gene interactions; gene and cytoplasmic interactions; gene, cytoplasmic,
and environmental interactions; as well as gene and,,,cytoplasmic produc-
tions to explain developmental differentiation and development.
Sinnott and Dunn have pointed out that size differences are only
differences in amount of growth and that these may be due to differences
in size of the early embryo, to differences in rate of growth, or to differ-
ences in duration of growth, all of them under gene control. In animals,
growth depends on hormonal stimulation from the anterior pituitary,

which has been shown in mice to have'a genetic basis. In several plants
it has been shown that there is a direct correlation between the size of
a very young ovary primordium "embryonic capital" and the size of the
eventual fruit, and in animals Gregory and Castle have shown that as
early as 40 hours after fertilization the embryos of a large race of rabbits
are significantly larger than those of a small race. The duration of
growth accounts for the differences in size between dwarfs and giants
of the same species, for the size of cells is approximately the same in both ..
Likewise differences in shape or size of organs or parts are knmm to
have a genetic basis, e.g., comb shapes in fowl (rose, pea, walnut, single),
fruit shape in squash (club, spherical, disk), length of legs in sheep
(Ancon), and length of rabbit ears. Here the genes influence the amount
or differentiation of growth in parts or in one direction at the expense of
another. The reduced growth rates of legs and wings of creeper fowl
have been shown by Landauer to be in evidence at the seventh day of
incubation.
Kuhn and his associates found that color of various parts of the
European flour moth was due to a dominant gene A, and Caspari showed
that testes of the AA or Aa genotype transplanted to aa host not only
retained their color but caused a certain amount of color to develop in
the normally colorless aa host, and that colorless aa ovaries or brain
transplanted to AA or Aa hosts took on color, due presumably to a
hormone secreted by A tissues. Similarly, through transplantation
procedure in Drosophila, Beadle and Ephrussi have demonstrated a
chain of two hormonal substances to be necessary for the production of
the wild-type red eye color. Wright has offered a theory invol ving
oxidizing enzymes to explain the color patterns in guinea pigs. Melanin
pigments are known to be formed from colorless chromogens in the cyto-
plasm. Wright assumes two enzymes in the nuclei of cells capable of
oxidizing chromogens to varying degrees. One of these is assumed to be
necessary for the production of any color, and, when present alone,
produces red. The other, nonfunctional in the absence of the former
and producing no effect by itself, when present in the full amount together
with the first, produces black color, and in lesser amount produces brown.
Genes evidently control the presence and amount or absence of these
enzymes as well as their potency. Albinism is presumably due to a lack
of the first enzyme, and various modifying genes affect the expression of
the second enzyme. Somewhat comparable theories of color in plant
have been proposed by Scott-Moncrieff, Lawrence, and Wheldale.
The work of Beadle and others! ,Yith the mold Neurospora is con-
I BEADLE, G. W., Biochemical Genetjcs, Chem. Rev., 37 :15-96, 1945.
sidered classic in this field. A number of mutant strains have been

produced as a result of X rays. It has been found that the growth of
some of these mutants is limited in the absence of a specific vitamin of
the B-complex. When the vitamin in question is added, normal growth
occurs. Some strains require pantothinic acid, some strains require
niacin, and others are limited by thiamin, riboflavin, or other members of
this group. It has been concluded that the inability of the mutant strains
to grow in the absence of these specific vitamins is the result of the
modification of specific genes which limits or inhibits the production of a
specific enzyme needed for the synthesis of the vitamin in question.
Other mutants require various amino acids.
The amounts of various genes have been shown in both plants and
animals to have a distinctive effect on certain characters. Since the
endosperm of corn is normally triploid, races can be made up containing
none, one, two, or three of the dominant Y genes for yellow color. Man-
gelsdorf and Fraps have shown that these varying amounts of the Y
gene are directly correlated with the relative amount of vitamin A present,
yyy endosperm having 0.05 rat units, Yyy 2.25, YYy 5.00, and YYY 7.5.
Dosage data are also available in Drosophila, where the gene sv, shaven
(bristles and hairs removed), in the IV chromosome can exist in the
haplo-IV form (only one of the IV chromosomes present), as well as the
normal diplo IV (two IV chromosomes), triplo, and tetra forms. In a
fly with only one IV chromosome containing the recessive sv gene, the
shaven condition exists in extreme form, less extreme in a diplo form,
still less in a triplo form, and where four of the IV chromosomes are
present (tetra form), each containing the recessive gene for shaven, the
bristles and hairs are about normal. In other words four recessive genes
do about the same job as two dominants.
In the eggs of some of the lower forms, the general pattern of develop-
ment is already laid down before fertilization; i.e., it is determined by the
genes of the mother. In snails, right- or left-handed coiling is deter-
mined by the cytoplasm of the mother. This in turn is affected directly
by the genetic constitution of the mother in which the egg was formed.
Right-handed coiling is dominant, so that a cross between a homozygous
right DD and left dd shows right-handed coiling Dd. The Fl DD, Dd,
and dd are all coiled to the right (because the mother did), but the last
class, dd, give all left-handed coiling offspring.
In some instances the genes make their presence felt right from the
start. In corn, for example, ordinary pollen (starchy) stains blue with
iodine, whereas pollen from mutant waxy plants stains reddish-brown
owing to the production of a different enzyme. When a cross is made
between starchy and waxy, however, the hybrid produces two sorts of
poll/n, the one with the starchy gene staining blue, the other with the
waxy gene reddish-brown.
In some cases the genes delay their action for a short while. In the
higher species the anlagen of both male and female genitalia are present
after the first few weeks of embryonic life. Then apparently the different
gene complexes come into play, and the primordia of one sex develop to
maturity, whereas those of the other atrophy. There are other character-
BAR - REVERTED
BAR
NORMAL BAR-DOUBLE
FIG. llO.-The bar region in salivary chromosome I, showing duplication of bands in bar
and double-bar. (From Bridoes. after Sinnott and Dunn. )
i tics that may remain latent for a considerable time after birth, and some
do not appear until middle life or later.
Sturtevant and Morgan first reported a case in which(the position of a
gene in a chromosome apparently affects the expression of a character)
This is known as e_osition effect, and an example is the bar gene in Droso-
phila. Bar eye in Drosophila (facets in a bar pattern rather than round)
I
has long been known as a dominant at point 57 in the X chromosome.
Zelany discovered that homozygous bars very infrequently produced a
normal (1 in 1,500), and, still less frequently, a still more restricted bar,
or ultra or double bar. By using bar females with two other mutant
genes near bar, Sturtevant and Morgan showed that the production of
normals from bar was always due to a peculiar unequal cro sing over.
Bridges finally, through a study of salivary-gland chromosomes, showed

normal round eyes to be due to one band in the X chromosome, bar to a
duplication of this band, and double bar to this band being in triplicate.
In leaving this entrancing but difficult field of the basis of genic,
cytoplasmic, environmental, enzymatic, hormonal, and organizational
influences in the development and organization of new characters into a
completed unique ,vhole (epigenesis), we would remind the student that
any individual begins its life as a tiny fertilized egg, but that very soon
the complete hody pattern emerges in a very minute form, but it is
definitely organized. The month-old embryo is only a' few centimeters
long but is definitely organized as a miniature of the adult form. All of
its parts and organs are at this time closely related in space, were even
more so earlier, so that the influence of one portion or organ of the body
on contiguous ones is not difficult to imagine. Spemann's classic experi-
ment in transplanting a tiny portion from near the dorsal lip of the
blastopore of an early embryonic amphibian to a totally different location
in another embryo with the resulting formation of a new embryo out of
tissues that would normally have produced totally different structures
was the first indication of definite organizing centers in. the developing
embryo.
It must be admitted that but little is yet definitely known of the
causes bringing about differentiation and specific organization in a devel-
oping embryo, but much research is now in progress to bridge the void
between the genes in the zygote and the fully developed character in the
resulting organism.
Evidence Supporting Gene Theory of Inheritance.-There are many
lines of evidence lending support to the chromosomal, or gene; theory of
inheritance!J' First and foremost is the fact that almost all organisms
arise from the union of an egg from the female parent and a spermatozoon
from the male parent ....€.This is all the living material the offspring gets
from its parents, though, of course, it derives its materials for growth
(proteins, fats, minerals, etc.) from its mother's blood.) Results of
breeding experiments show that, speaking generally, both parents are
jointly and equally responsible -for the potentialities of their offspring.
~ Because this is so and because the two germ cells, egg and sperm, are)
alike only as to their nuclei, it seems logical to conclude that the deter- (
miners of potentialities are located in the two nuclei which con~ain the
chromosomes and which in turn are made up of genes.
(j Additional support for the theory is to be found in the phenomenon
known as parthenogenesis; e,g., an organism arising from an egg that has
been stimulated to cell division artificially; e.g., ,vith a hot needle. These
organisms evidence maternal characteristics only. In like manner an
THE PRINCIPLES OF HEREDITY 38.5
egg, the nucleus of which has been removed or inactivated by X rays,

may be fertilized by a sperm, and in this case the organism will evidence
v' paternal characteristics only.
It. Further evidence is available from cytological studies, which have
demonstrated that, if one of the small (IV) chromosomes in Drosophila
is lost from an egg or sperm, the resulting offspring is different from a
normal fly in many parts of the body; i.e., specific effects are produced
by this IV chromosome. For example, if a female with normal eyes
but having only one IV chromosome is mated with a mutant eyeless male,
half the offspring are eyeless, and they are also weak individuals, many of
them not getting through the pupal stage successfully. This proves that
the gene for the eyeless condition is located in the small IV chromosome.
S~ The behavior in inheritance of sex-linked characters, i.e., characters
determined by genes located in the sex chromosomes, is additional strong
evidence that the genes located in the chromosomes of the nuclei of the
sex cells, ova and spermatozoa, are the actual determiners of -the poten-
tialities of any organism.
t Finally the work of Stern, with Drosophila, and Creighton and
McClintock, with maize, which demonstrates a complete correspondence
between genetic and cytological crossing over, seems to furnish the final
bit of positive proof for the complete substantiation of the chromosomal,
or gene, theory of inheritance.
Summary of Modem Genetic Theory.-The modern theory of inherit-
ance is ba_sed 1lpon the behavior in gametogenesis and ontogeny of more
or less hypothetical entities called genes. However, the theory has now
been tested by means of thousands0rplanned breeding experiments and
has been corroborated by numerical data, as -indicated in the definite
numerical ratios for the mono-, di-, tri-, and polyhybrids. In other
words, the theory is no longer a theory but an established fact. It should
be recalled that the facts of breeding came first and the theory to explain
the facts came second. We demand two things of any theory: (1) that
it logically explain events that have transpired in the past, and (2) that
its principles enable us successfully to predict future happenings. These
criteria of fitness have been successfully met by the chromosomal, or
gene, theory of inheritance.
In brief, the gene theory assumes that the characteristics of an organ-
ism are conditioned by paired genes located at similar loci in homologous
chromosomes; that one member of a gene pair may dominate the situa-
tion more or less fully in the developed organism, so that the other
member of the pair is not externally recognizable; that the genes are
organized into groups, each group forming a chromosome; that at the
reduction division in gametogenesis one member of each homologous pair
~
386 BREEDING AND ]J[PROVEMENT OF FARM ANIMALS
of chromosomes is found in each daughter cell; that the distribution

of the homologous chromosomes to the daughter cells is at random; that
previous to reduction, i.e., during synapsis, equal interchanges of chrom-
atin material may take place between the non sister chromatids; and that
the chromosomes are always organized in a definite manner, each gene in
its inviolable place and the genes arranged in a linear series.
Later chapters will indicate how the genes may change, thus giving
rise to new variates. Once again the writer would like to caution against
any tendency to oversimplify the matter of the genes and their behavior
in inheritance. '''The genes themselves are probably very complex in their
.11 structure. They are very closely associated with the cytoplasm of the
cells, and full development of the individual should be looked upon as a
result of the interactions between the genes themselves, between the genes
and the cytoplasm of the cells, and between the cells and their environ-
ment. It is the interactions and interrelations of all these forces that
function in the development of a new individual rather than the individual
actions of separate genes.
It is known, for instance, that in Drosophila at least 50 pairs of genes
function in bringing about the development of the normal red eye. It is
also known that a change at locus 44.4 in the I chromosome, a change at
locus 54.5 in the II chromosome, or a change at locus 49.7 in the III
. chromosome will each result in purplish rather than red eyes. These
genes have been named garnet, purple, and maroon, respectively. It
would be possible, therefore, to mate two purple-eyed individuals and to
get from them all normal red-eyed offspring, for each parent might supply
the missing normal gene that the other lacks. It is also readily seen that
one type of purplish eyes (garnet) would show sex-linked inheritance, for
this gene is located in the sex chromosome, whereas the other two sorts of.
purplish eyes (purple and maroon) would show ordinary Mendelian
inheritance.
As has been shown earlier, dominance is a relative term, and a char-
acteristic that is dominant in one type of mating may be recessive in
another. Inheritance, therefore, should be looked upon as the inter-
action between genes, cytoplasm, and environment rather than thought
of in an absolute sense.
Summary.-In this final chapter dealing with the principles of inherit-
ance, we have considered the complications introduced into breeding
data by the fact that many genes may be located in one'chromosome, so
that these genes do not assort independently, although the chromosomes
always do so. W;'haVe seen that varying degre~'i)f linkage up to 100
per cent ma'y be encountered. If linkage is not complete, various
amounts of crossing over up to nearly 50 per cent may be encountered,
,
and crossover values between any two genes remain constant under
uniform conditions. The use of crossover values in constructing the very
ingenious chromosome mlill_s was indicated on the basis of the genes being_
arranged i~ linear series aiong the chromosomes. We comPleted our
discussion of th~ seve~ Mendelian principles with references to inter-
ference and to limitation of the linkage groups to the number of pairs of
chromosomes present in any species. Finally, we discussed the as yet
rather incompletely known role of the genes in development, indicated
the various lines of evidence that substantiate the chromosomal and gene
theory of inheritance, and presented a summary of the modern genetic
theory.
References l
Problems
1. Assume that genes a and b are linked and show 40 per cent of crossing over.
If a ++/++ individual is crossed with one that is ab/ab, what will be the genotype
of the F l? What gametes will the F 1 produce and in what proportions? If the F 1
is crossed with a double recessive, ab/ab, what will be the appearance and genotypes of
the offspring?
2. If the original cross is +b / +b X a+ /a +, what will be the genotype of the F I?
\Vhat gametes will it produce? If the F 1 is crossed back with a double recessive, what
will be the appearance of the offspring?
3. An individual homozygous for genes cd is crossed with wild type and the F 1
crossed back to the double recessive. The appearance of the offspring is as follows:
903 ++, 898 cd, 98 +d, 102 c+. Explain this result, giving the strength of the
linkage between c and d. If assortment between c and d were independent, what
would be the result of this cross?
4. If the cross in the preceding question had been between a homozygous +d
individual and a homozygous c+ one, what would be the result of the cross between
Fl and the double recessive?
5. How far apart are A and B if in the F2 you derive 706 AB; 291 Ab; 304 aB; 697
ab?
6. Results of a cross of Aa Bb Cc X aa bb cc give:
600 ABC 140 aBC 15 aBc
580 abc 120 Abc 5 AbC
130 abC
120 ABc
\Vbat are the percentages of crossing over between A and Band Band C? How
far apart are A and C?
7. NOTE.-In Drosophila the mutant known as black, b, has a black body in con-
trast to the wild type, which has a gray body; and the mutant arc, a, has wings which
are somewhat curved and bent downward, in contrast to the straight wings of the wild
type.
In the two following crosses the parents are given (homozygous in each case)
together with the counts of the offspring of Fl females bred to black, arc males (data
from Bridges and Morgan).
388 BREEDING AND IMPROVEMENT OJ.? FARM ANIMALS
a. Black, arc X wild type (gray, straight)
Fi females X black, arc males give

1,641 gray, straight.
1,251 gray, arc.
1,180 black, straight.
1,532 black, arc.
b. Black, straight X gray, are
FI females X black, arc males givc
281 gray, straight.

335 gray, arc.
335 black, straight.
239 black, arc.
From these data calculate the crossover value between black and arc.
S. NOTE.-In Drosophila the mutant known as vestigial, V, has wings that are
very much reduced as compared with the long wings of the wild type.
In the two following crosses the parents are given, as in the previous question,
together with the counts of offspring of FI females crossed with black, vestigial males
(data from Bridges and J\lorgan).
a. Black, vestigial X wild type (gray, long)
J.? I females X black, vestigial males give

822 gray, long.
130 gray, vestigial.
161 black, long.
652 black, vestigial.
b. Black, long X gray, vestigial

F I females X black vestigial males give
283 gray, long.
1,294 gray, vestigial.
1,418 black, long.
241 black, vestigial.
From these data calculate the linkage strength between black and vestigial.
9. NOTE.-In tomatoes, Jones has found that tall vine is dominant over dwarf,
and spherical fruit shape over pear. Vine height and fruit shape are linked, with a
crossover percentage of 20 per cent.
If a homozygous tall, pear-fruited tomato is crossed with a homozygous dwarf,
spherical-fruited one, what will be the appearance of the F I? Of the F I crossed with a
dwarf, pear? Of the F2?
10. A certain tall, spherical-fruited tomato plant crossed with a dwarf, pear-
fruited one produces 81 tall, spherical; 79 dwarf, pear; 22 tall, pear; and 17 dwarf,
spherical. Another tall, spherical plant crossed with a dwarf, pear produces 21 tall,
pear; 18 dwarf, spherical; 5 tall, spherical; and 4 dwarf, pear. What are the genotype;;
of these two tall, spherical plants? If they were crossed, what would their offspring
be?
11. NOTE.-In rats dark eyes are due to the interaction of two genes Rand P, the
recessive allele of either producing light eyes. These genes are in the same chromo-
some.
When homozygous dark-eyed rats, + + / + +, were crossed with double-recessive
ones, rp/rp, and the F, crossed back with the double recessive, the following offspring
were obtained (data from Castle): 1,255 dark-eyed; 1,777 light-eyed.
When +p / +p animals were crossed with r + /r + ones and the F, crossed back with
the double recessive, the following offspring were obtained: 174 dark-eyed; 1,540
light-eyed.
Calculate the linkage between rand p.
12. In the fowl, assume that e (early feathering) and B (barring) are sex-linked and
show 20 per cent of crossing over (in the male only). If a male from a cross of late-
feathered, barred male with early, black female is mated with an early, black female,
what will be the appearance of their offspring, as to feathering and barring?
13. Assume that genes a and b are linked, with a crossover percentage of 20 per
cent; and that c and d are also linked, with a crossover percentage of 10 per cent, but
are in another chromosome. Cross a plant homozygous for ABCD with one which is
abed and cross the F, back on abed. "'nat will be the appearance of the offspring of
this cross?
14. In Drosophila, yellow body is sex-linked and recessive to the gray body of the
wild fly. Vermilion eye is also sex-linked and recessive to the wild red eye. The
genes for yellow and vermilion show about 28 per cent of crossing over. The gene for
vestigial is in one of the autosomes. If a homozygous yellow-bodied red-eyed long-
winged female is crossed with a homozygous gray-bodied vermillion-eyed vestigial-
winged male, and if an F, female is crossed with a yellow, vermilion, vestigial male,
what will be the appearance of the offspring of this last cross?
CHAPTER XV
THE PRINCIPLES OF VARIATION
Genetics, as was said earlier, is concerned with heredity and variation.

Its task is to explain why an organism does or -does -not fe'semble its
ancestors. The three previous chapters have dealt with hereditary
likenesses between related organisms. Heredity is the great conserving
force. The fact that the genes and their combinations are relatively
stable has acted as 3.. ratchet mechanism to hold whatever gains have
been made from the standpoint of evolllt.!()!1_,_ F;r-progress, however,
\ve need-stJffietIlillg mo~~ than this, viz., -variation. For many centuries,
species and varieties were believed to be immutable. The rise of many
varieties of dogs from one or two remote ancestors by means of the
principles of variation and selection and the same phenomenon in horses,
cattle, sheep, and swine was pointed out in an earlier chapter.
Henry van Dyke is reported to have said that there is one thing in
which all men are exactly alike, and that is that they are all different.
This statement also holds true for farm animals. To the observing it is
quite evident that" the most invariable thing in nature is va_riation."
When we consider that-pr~bably no '£' '0
blades of grass, U-o t\~;o-calves,
and no two human beings have ever been just exactly alike (though
identical twins are essentially so), we get a fleeting glimpse of the resource:-
fulness of nature. It would seem that the new plans and specifications
must sometime be depleted, yet life flows along in its thousands of
varieties and millions of unique individuals, each different from all the
rest of its kind. Fortunate, too, that this is so, because without this
principle progress would be impossible. As has been said, "Variation
is at once the hope and the despair of the breeder," the hope because.
through it may be produced offspring better than their parents, the
despair because after animals have been greatly improved, they may
and very often do vary again toward mediocrity.
That animals vary in size, type, rate of growth, efficiency of feed
conversion, quality of carcass, proportion and distribution of fat and
lean, wool quality and quantity, fertility, longevity, resistance to disease,
milk production, butterfat percentage, color of milk, persistency, speed,
stamina, style, tractability, color._ and in a hundred other ways is too
evident to need much discussion .
•
;\ 390
THE PHINCII'LES OF VARIA'l'I01Y 391
Here are two steers, one of them gains 22.1 lb. for each 100 lb. of
nutrients fed; the other only gains 14.3 lb. from a like amount of feed.
Here are the litters of two sO\\"s, one of them gains 100 lb. from 504 lb.
of feed; the other litter gains 100 lb. from 360 lb . of feed .
Here are t,yO dairy cows, one of them returns 36 lb. of total digestible
llutl'jents for each 100 lb . fed; the other returns only 18 lb. from a like
amount of feed.
Here is a Guernsey bull bred to 8 cows that range from 12,500 to
13,500 lb. of milk per year; the 8 daughters range from 8,600 to 15,600 lb.
Here is an Ayrshire bull bred to 13 cows that range from 10,500 to
11,500 lb . of milk per year; the 13 daughters range from 7,800 to 15,300 lb.
F IG. 11 2.- Beth of Amherst 3rd, a Grand Champion at Eastern States Exposition (picture
taken when dry). H er twin sister was nearly as poor a dairy type as this one is good .
Variation is tremendous, partly o'\'ing to the mechanism of inherit-

ance, partly to the effects of environment. Most of the qualities of
oUT animals are elastic. Depending on feed, care, etc., a cow may produce
8,000 lb. of milk in a year or 16,000 lb. A sow may farrow ten 3-lb. pigt'
and raise them all, or seven 2-lb. pigs and raise four of them. The
breeder's task is to provide the best possible environment at all times
and to make the environment as uniform as possible in order that his
variations will be known to have a genetic basis.
Causes and Types of Variation.- There can conceivably be but two
causes of genetic variation, It is caused either by some outside, super-
natural force, or else it is inherent in the nature of living things, their
natural response to stimuli,
V . t' {Supernatural
arIa lons Natural
392 BREEDL1>v'G AND IMPROVEMENT OF FARM ANIJ1ALS
General opinion leans toward the latter explanation. There are two
types of stimuli that might conceivably bring about variation, one inter-
nal, the other external to the organism. Any fully developed organism
is the result of the combined action of these internal and external stimuli.
V . t' {AutOgenetic
arIa lOns Exogenetic
The organism starts as a single-celled zygote, with its full complement

of genes derived from both paternal and maternal sources. This genetic
heritage has a tendency to develop in a certain definite way, owing to
its chemical and physical make-up. The norm for this new individual
may vary from that of one or both of its parents because of its different
chromosomal content. During intrauterine development, any organism
is subject to various influences from its mother, with whom it lives in
very close association. Chemical products in the blood of the mother
may be transfused into the circulation of the embryo and may react in
various ways on the developing embryo or cause it to react in a variety
of ways. The mother's blood, or the fluid surrounding each cell, may
contain nutritive material that will assist in the proper growth and full
development of the embryo. On the other hand, they may contain
waste or toxic products, secretions or hormones from various glands,
etc., any of which may cause the embryo to vary more or less from its
original normal tendencies. Effects due to the parent's habits or to
contingencies in the developing embryo itself cause many variations to
appear in the fully developing organism, all the way from a doubling of
parts, through various types of changes, including blemishes and mal-
formations to an entire pinching off or absence of the part. '
. l{Recombinations
.'
" S ermma
VarlatlOns {G t' Gene and chromosome mutatlOns
.
." , oma IC { Ch anges m
. soma, b eh '
aYlOr, e t c.
In this chapter, we shall consider only those variations that have their
origin in the germinal material, the genes and chromosomes in the germ
cells, leaving the somatic variations to be treated in the next chapter,
Variation as a Too1.-Variation is one of the most useful tools that
th~p'oSSeSSeS:-- It provides him the raw material fro~hich
he can fashIon b~i1er animals. The breeder is constantly on the lookout
for desirable types of variation. He sees variation as it manifests ,"
itself in his animals and seizes upon it in the hope that it will prove to
be hereditary. Much variation does have a genetic basis, much of it
too is environmental, whereas some perhaps is both. It is no simple
\
THE PRINCIPLES OF VARIATION 393
matter to separate these three categories. In addition to these com-
plications is the further one that the hereditary variations may be of
various sorts. Some of the allelic pairs of genes may show varying
degrees of dominance, from complete to an entire lack. This same feature
of dominance bet~\Yeen different pairs of genes known as epistasis may be
operating, whereas in still other cases the various genes may behave
in a cumulative or additive fashion. To secure desirable variation of a
hereditary sort and to guide it toward some preconceived ideal of perfec-
tion is the problem that the breeder constantly struggles with.
Fortunately there is generally plenty of variability available. In
horses, cattle, sheep, and swine there are literally all sorts. The animals
100
I
90
80
/
. . .y '" :'\.
10
.!!!
] 60 / \
II • I\.
1350
~40
..0
V \
§ 30 / I'"1\
z
20
v
10
/ \
0 l-
V .........
6 q 10 II 1'2 13 14 15 16 11 18 Iq 20 21 22 23
1 8
1houscmcls of pounds 3.5'/, milk or its eq.uivQlent in 4'1, f.c.m.
FIG. 113.-Graph of Holstein Indexes, Vol. 11, Holstein-Friesian Red Book.
of a breed, as far as their commercially valuable qualities are concerned,

probably simulate a normal curve. Some of them are poor, a few more
mediocre, most of them just average, a few very good, and a very few
excellent, the excellence probably being due in mos.t cases to good heredity
plus ~oo_1._~}iYI~~~me.nt. '. Tll.i~..g~:neral feature of variability is shown in
Fig. 113, a graph of 703 Holstein bulls listed with their indexes. 1
The above graph shows that 5 of these bulls had indexes of 6,000 lb.
of 3.5 per cent milk, 7 had indexes of 7,000 lb. of 3.5 per cent milk, 11
indexes of 8,000 lb., etc., up to 4 bulls with indexes of 22,000 lb. and
4 with indexes of 23,000 lb. of 3.5 per cent milk. Because the index is
computed from a comparison of dams' and daughters' records, it is
obvious that the females would also show this sort of normal curve of
production. The breeder is, of course, trying to make his selections
from animals in the upper half or top quarter of these curves, and he
1 From Vol, 11 of the Holstein-Friesian Red Bonk,
hopes that a goodly part of this variation is hereditary and will be passed
along to the offspring.
FIG. 114.-Two Ayrshire cows and their twin calves showing mark"ld variation in amount
of red color in the coat.
Freeman 1 found the mean number of body vertebrae of 52 Duroc

pigs to be 20.52 ± 0.067, and for 79 Yorkshire pigs to be 21.27 ± 0.046.
The increased number of vertebrae were associated "I"I'ith greater body
length, which would add to carcass
value and might possibly be corre-
lated with greater fertility.
Variations from Recombina-
tions.-We have already discussed
pretty thoroughly the general
proposition of recombinations of
FIG. 115.-Showing the different types of
germ c_ells possi ble in a species with two pairs chromosomes during meiosis. In
of chromosomes-chromosomes remaining an organism with two pairs of
intact.
chromosomes, there are four possi-
ble ways of recombining these intact chromosomes as indicated in
Fig. 115.
In each resulting germ cell there must necessarily be one member of
each pair of chromosomes. The possible numbers of different gametes
1 FREEMAN, V. A., Variations in the Number of Vertebrae in Swine, J our. Herecl. ,
30 (2):61- 64, February, 1939.
with some varying numbers of pairs of chromosomes are shown in

Table 23.
TABLE 23.-NuMBER OF DU'FERENT KINDS OF GERM CELLS POSSIBLE FROM VARYING
NU~IBERS OF PAIRS OF CHROMOSOMES
Number of differ-
Number of pairs
ent kinds of germ
of chromosomes
cells
Ascaris ............................... . 1 21 2
A. megalocephala . ....................... . 2 2' 4
Drosophila melanogaster . ................. . 4 24 16
Corn .................................. . 10 210 1,024
Swine .............................. ' ... ' 19 219 524,288
Human ................................ . 24 224 16,777,216
Cattle, goat, horse ..................... . 30 230 1,073,741,824
It is evident from the above table that the simple recombination of

~
intact chromosomes in the :g.jgher species can aC~~Unt for a ti~mendous
_ _~.-~ -- • . _ .~ ___" _ _ _ v~ ... _ .v"..... _.
amount of variation. From a perusalof the above table one migbt

conctude tna::Cbreeding is a pretty haphazard and impossible task any-
. way. The writer would not try to maintain that it is easy, but neither
is the job as impossible as might appear at first glance. What we are
trying to do from a practical breeding standpoint is to make our animals
pure for genes that lead to the development of desirable characters.
At present they are rather heterozygous and therefore do not breed
true. Two homologous chromosomes may bear the following sorts of
genes.
If, 'by selecting, we bring it about that these two chromosomes become
homozygous,
then it will make no difference which one of these two homologous

chromosomes go to any egg or sperm, for they are now exactly similar.
It would still and always be true that the 30 pairs of chromosomes in
cattle could and would be capable of producing over a billion different
recombinations, but, if our animals were entirely homozygous, all of

these billion combinations would be identical. It seems unlikely that
we can ever reach this goal, but the fact remains that the more hetero-
zygosity can be reduced, the less variation we will have.
If a male Aa is mated to a female Aa, there can be produced two
different types of gametes (A or a) by each parent and offspring of three
different genotypes (AA, Aa, or aa).
If a male of the genetic constitution Aa Bb (with genes A and B in
different chromosomes) is mated to a female of the same genetic make-up,
there can be produced four different types of gametes by each parent and
offspring of nine different genotypes.
With three pairs of heterozygous nonlinked genes in each parent, we
would have eight different types of gametes and offspring of 27 different
genotypes.
TABLE 24.-INDEPENDENTLY SEGREGATING GENE RECOMBINATIONS
Number of pairs of Number of different Number of different

heterozygous genes gametes genotypes
1 2 3
2 4 9
3 8 27
n 2" or (10 0 •s0b ) 3" or (100 . 477 ,,)
If two animals were of the genetic constitution shown below (genes in

different chromosomes), '
sire, Aa bb CC Dd
nam, Aa Bb cc DD
then ~ere would be the following possible genotypes of offspring
3 X 2 X 1 X 2 = 12
Thus we could have 12 different sorts of full brothers or sisters, as to

genotype, from mating the above tViO parents. All of these animals
would have identical pedigrees but their genotypes might all be different
ranging from
AA BbCc DD to aa bb Cc Dd
If the capital letters represented desirable traits and the small letters
undesirable ones, we can see that there would be a vast difference between
these full sisters or brothers. Identity of pedigree, then, does not meall
identity of inheritance except when the parents are homozygous for each
pair of genes.
Our example has dealt with four pairs of genes, or eight in all. We do
not know how many genes are present in the chromosomes of our farm
animals. There are 60 chromosomes in some of these animals, and, if
there were an average of 100 genes in each chromosome, there ,,,,ould be a
total of 6,000 genes. Each parent would then be transmitting not 4
genes, as in our simple example, but 3,000. If two animals were hetero-
zygous for all of these 3,000 pairs of genes, then there could be 3 3• 000
different genotypes, assuming that all combinations were possible by
means of independent assortment and crossing over. Such a number
is, ofcou~~e, beyond human comprehension. It could be written 10 0. 477n
or 10 1,431, and it has been estimated by physicists that the whole universe
contains 1080 electrons. Three to the nineteenth power is a number
twice as large as the number of cattle at present in the world. These
3,000 genes are in bundles (chromosomes) of perhaps 100 each and, of
course, will show linkage in each bundle. Even if linkage is complete in
all chromosomes, we could still get 2 30 or 1,073,741,824 different recom-
binations of complete chromosomes, whereas there are only about 650
million cattle in the world.
In our simple example where we were dealing with only eight genes,
if we assume all these genes to be in the same chromosome the sire might
be
and the dam
which would gIve AbeD/ABeD, AbeD/abeD, abed/ABeD, abed/abeD

combinations.
Thus the very mechanism of inheritance in unisexual animals whereby
a parent gives a sample half of his or her inheritance to each offspring
is a very potent source of variation.
Variation from Crossing Over.-We learned in the previous chapter
that blocks of genes in two homologous but nonsister chromatids may
change places with each otheLlo The working of this principle increa~
the possibility of variati~n t~ an infinite amount, so that the billion
possible recombinations of intact chromosomes in farm animals fades
into insignificance in comparison. Again, however, it will not be so bad
as it seems if we lay our plans to render the genetic material of our animals
somewhat more homozygous that it is at present.
These principles of independent assortment and crossing over make
the possibilities of variations tmly infinite. The workings of these
principles are, no doubt, the greatest cause of variation in farm animals.
They explain why a cow (scrub in this case) can be bred to a certain bull
and a heifer produced that yields 3,600 lb. of milk and later the same
cow be bred to the same bull and a heifer produced that yields 6,100 lb.
of milk. The first heifer yielded 31 per cent less milk than her dam,
the second one yielded 17 per cent more than her dam. Or, to cite a
case from the purebreds, viz., a purebred cow bred twice to the same
purebred bull, in the first case a heifer resulted that yielded 6,200 lb. of
milk, 14 per cent less than her dam, and in the second case a heifer
resulted that produced 9,800 lb. of milk, or 35 per cent more than the
dam. As explained above, the chromosomes, which carry the deter-
miners for all characters, e.g., milk yield, tendency to fatten, color, speed,
type, etc., go through the process of reduction, the chromosomes separat-
ing or segregating, which enables them to reunite into new combinations
of various sorts. The chromosome mechanism, in heterozygous organ-
isms, is one of the most potent causes of variation, including as it does
both the recombination of intact chromosomes and the creation of
different ones by means of crossing over. It must be recognized, how-
ever, that such characters as ~ilk production may be greatly influenced
by environmental factors. The occurence of such metabolic disturbances
as ketosis, the presence of inflammation of the mammary gland, improper
feeding, or general management practices frequently overshadow the
actual genetic potentialities of dairy cattle.
Variation from Gene Mutations.-In addition to the different recom-
binations that might be inherited, there are mutations that are also
inherited. Gene mutations are germinal or genetic variations involving
single genes. They are probably due to chemical or spatial.alie;ations
in the genes for certain characteristics in the germ cells. vTh! c;he~ic9:.l
rearrangem~l}~_~O~e..l1t.9~s..i!,l._th~ that may bring about mutations
migh-roe compared t-othe interchanging of a certain number of hydroxyl
ancfIlYdroge'u ions from one place to another in the structural formula
for dextrose which ~onverts dextrose into le~ulose. Or different atoms
may be substituted for some already present, just as one may by sub-
stitution change methane (CR 4 ) to monochloromethane (CRaCI). Both
the new arrangement of atoms and the substitution of one or more
different atoms for others already present would supposedly~ive rise t.2,.
y_ariatiopA:..J
Since 1927, when Muller first discovered that treating Drosophila
with varying dosages of X rays led to increased rates of mutation, much
experimental work of thiEl'nature has been performed both with X rays
and with radium. The evidence seems irrefutable that the bombard-
ment of the genes by active rays does induce changes in the molecular
structure of the gene, which in turn is evidenced by changes in external
or physiolqgical characteristics. These gene changes ocg_ur naturl.tlly
I, "-'
1\
THE PRINCIPLES OF VARIATION 3\)9
but at very infrequent intervals. The rate of mutational changes has

been speeded up-as ffiuch~aSI50 times by ray treatments. Some genes
have proved to be much more stable than others. Stadler studied the
effects of X and radium rays on seven genes in maize and found a very
~ wide range of susceptibility to change among them. Stadler's results
are shown in Table 25.
Attempts to produce mutations by chemical means go back to the
early days of modern genetical research. Although some earlier work
was suggestive, the first positive results indicating the production of
mutations by chemicals were obtained by Auerbach 1 and Robson during
the Second World War (see Auerbach, 1949, for review) with mustard gas.
Since then various other chemicals have been shown to have definite
mutagenic effects. A new tool is thus available to the research worker.
TABLE 25.-STADLER'S DATA ON THE FREQUENCY OF CHANGES IN SEVEN GENES OF
.': • :\LUZEI
Average per
Gametes Number of
Gene one million
tested mutations
gametes
R ......... ................... . 554,786 273 492

I ... . ' ....................... . 265,391 28 106
Pr ....................... . 647,102 7 11
Su ........................... . 1,678,736 4 2.4
y ................ ............ . 1,745,280 4 2.2
Sh ........................... . 2,469,285 3 1.2
Wx ... ....................... . 1,503,744 0 0
1 STADLER, Jour. Hered., 24, (10):372, 1933.
We do not know exactly what causes gene mutations in nature. We

do know that they occur with greater or lesser degrees o{frequency.
About 600 gene mutations have been recorded in Drosophila melanogaster.
One of the earliest discovered was that which changes the eye color
from red to white. A white-eyed male fly suddenly appeared, and, when
mated with a normal red-eyed female fly, the F 1 offspring were all red-
eyed. In the F 2, however, the white-eyed condition again appeared,
but only in male flies. This is, therefore, a sex-linked, recessive mutation.
Other changes in this same gene have resulted in a series of 14 allelic
characteristics as to eye color in Drosophila, ranging from white through
ecru, tinged, buff, ivory, eosin, apricot, cherry, blood, coral, and wine to
red. Reverse mutations also occur; i.e., red eyes may mutate to white,
and white may mutate back to red.
1 AUERBACH, CHARLOTTE, Chemical Mutagenesis, Bioi. Rev., 24:355-391, 1949.
Some gene mutations, such as lobe 2 in Drosophila, are dominants in

inheritance. A great many of the mutations are l~thal, the offspring
never reaching matunty, wnereas III others, e. g., the white-eyed mutant,
many other characteristics besides eye color are affected, the white-eyed
individuals being less robust, more sluggish, and shorter lived.
The best known mutation among farm animals is, no doubt, the
characteristic of 12o11edness, or hornle~ This mutation probably
-----
a rose because of a chemical change of some sort in the chromoso!Ues.
-
FTG. 116.-A sheep with very short and crooked legs after the style of the Ancon . .
No doubt, they have arisen in all the breeds at times, and in most breeds
they have been noticed and valued and a polled strain developed from
them. The reason for the change in the chromosomes that prevents
'-. ~-
the horns from developing is unknQl\Ql. -
Another mutation, now erlinct, ;;Ccurred many years ago in sheep.
Seth Wright, a Massachusetts farmer, had dropped in his flock a ram
lamb that was a mutant, being low-backed, long-bodied, and having
short crooked legs, something after the fashion of a dachshund. Farmer
Wright recognized its value-its type of build, although not interfering
seriously with wool production, being a decided deterrent to the sheepish
pastime of fence jumping. The ram, a mutant, was used for breeding,
and from this start the Ancon breed of sheep, with the characteristics
listed above, was produced, though the breed has long since been sup-
planted by the Merino.
\
THE PRINCIPLES Of!' VARIATION 401
The Mauchamp sheep of France, with a superior sort of fleece, are

sometimes considered a mutant race, but, inasmuch as the original ram
was a hybrid between a Mauchamp ewe and a Dishley ram, this case,
as Xathusius points out, must be removed from the records of actual
mutations in domestic animals. Another character in sheep reported
from Norway, viz., short ears, though of no practical importance, is
perhaps, a mutant.
Regarding mutations in domestic animals Babcock and Clausen state
that:
Any system of herd improvement founded on the search for and utilization
of mutants is doomed from the beginning to failure, for mutants of a beneficial
character appear so rarely as to have almost no practical significance. If by
some fortunate chance a breeder should find himself in possession of a favorable
mutant individual, however, it is a simple problem in Mendelism to establish
its characters in a constant race.
For instance, if a mutant sheep were produced with no tail, a tailless
breed could soon be developed and the trouble of docking saved.! K 0
system of breeding has yet been advocated to produce mutations, perhaps
none will ever be devised. Tendency to mutate, however, is very likely
an inheritable characteristic, just as are all other features and traits
exhibited by organisms.
GENERAL STATEMENTS ABOUT GENES AND MUTATIONS
1. Genes are ultramicroscopic, chemical entities-possibly giant
molecules-which are fairly sta~nd self-.E!0E~g'1JiD_g.. DN It r P'
2. Gene mutations are sudden variations resulting from chemicalL
structurl11-position, or volume change in genes.--
1£ they occur in the
germ cells or their primordia, they become a part of the hereditary mate-
rial; if they occur in the soma, they may affect only one cell or whole
groups of cells, but these somatic mutations do not become part of the
hereditary material.
3. Mutation generally occurs in one gene at a time.
4. Mutation rates differ in different genes, vary from very low fre~
quency in some genes to relatively great frequency in others, and thE:'
rates may be inherent and/or may be influenced by the environment.
5. Mutations are more likely to occur in some parts of the chro-
mosomes than in others.
6. Mutations may involve a series of changes in one gene (multiple
alleles).
7. Mutation-wild type to new allele is called direct.
8. Mutation-mutant allele back to wild type is called reverse.
I See S. Dak. Agr. Expt. Sta. Cir. 28, H140.
402 BREEDIiVG AeVD HIPROVEMENT OF FARM ANIMALS
9. Mutations have varying effects from almost no effect on up to a

lethal effect.
10. Mutations are usually recessiy_e and generally harl_Dful___o.r_kthlJ,l.
11. Mutations may occur arimy time-probably most frequently at
mt}oturation.
J12. Mutations that are dominant show up at once .. / I
"'13. Mutations that are sex~linked recessives show up at once in males.
14. Mutations occur with ~pp;~~i~bl~'f~eq'uency in nature. I
/15. Mutations can be induced by ra.<!iation, temperature changes,
and chemical agents and are proportional to the severity, within viable
limits, of the treatment received.
JL~~fem("'e~th~!L~~
bOlr-eyeci
x
X-rays
16. Mutations must be proved by breeding experiments and analyses

-white eye, probably a chemical or structural change; bar eye, gene
duplication; notch, a deficiency; double bar, a position effect.
17. Several hundred gene mutations have been found in Drosophila,
corn, peas (Mendel dealt with seven of them), rats, rabbits, and some,
too, in the higher animals including man. Most of the problems at the
end of Chaps. XII, XIII, and XIV involved gene mutations.
Lethals are generally gene mutations, and they have proved very
useful in the study of mutations induced by X and radium rays. The
ClB method was devised by Muller to study the rate of production of
mutations by artificial means. Because most mutations are lethal, it .
was necessary to for~ plan that would reveal their presence even
though the organism was killed before hatching. To do this Muller
bred Drosophila, which had the genes for crossover suppression, a known
recessive lethal, and bar eyes in one X chromosome and their alleles
(nonsuppressor, normal, and round eyes) in the other X chromosome
THE PRINCIPLES OF V ARI ATION 403
(ClB and cLb). These flies were bred to a male with nonsuppressor,
normal, and round-eye genes in its X chromosome (cLb). Before breed-
ing, these males were subjected to X rays. From this cross, bar-eyed
females (ClB and cLb) were selected and bred to normal, untreated
(cLb) males. From this cross, half the males will die because their X
chromosomes contain the original lethal gene (ClB). If the other males
also die (cLb), it shows that a~.was produced by the
X-ray tre~tmeI!t of the original mak"The complete lack ~grand~s,
therefore, proves the artificial induction of a new lethal gene, probably
a cheD1_icaL chlk!},g~hat destroys the balance necessary for
the survival of an organism getting it~~-,_----____,
alanced lethals were first discovered by Muller. His stock of
eaded (deformed wings) flies when mated usually gave % beaded and
H normal offspring, as though beaded were due to a single dominant
gene, lethal when homozygous. Muller was able to select out a stock
that bred practically true for beaded (producing an occasional normal)
but when crossed back to normal gave 50 per cent beaded and 50 per
cent normal, showing it to be heterozygous for beaded. Full analysis
by Muller finally showed that the two stocks of flies differed genetically,
the first having a dominant mutation, beaded, in the III chromosome
which was lethal when homozygous, the second or true-breeding strain
have another lethal closely linked to beaded, so that their genetic con-
stitution is Bd+ / +l. With a crossover suppressor (probably an inver-
sion) they produce only Bd + or +l gametes and only Bd+ / +l offspring
(with an occasional + + normal if crossing over occurs), because the
Bd+ / Bd+ offspring and the +l/ +l offspring both die, which gives
an apparently true-breeding strain because of a system of balanced
lethals.
Variation from Chromosomal Aberrations.-In addition to gene or
point mutations a considerable variety of chromosomal aberrations
involving various-sized blocks of genes, whole chromosomes, or whole
sets of chromosomes have been discovered in recent years. Changes
of this nature are generally brought to light by discrepancies in breeding
behavior resulting in disturbances of the normally expected Mendelian
ratios, in the creation of new linkage relations, or in the appearance of
variable offspring. When such things occur, cytological study has
generally revealed certain changes in the number or arrangement of the
loci within the whole chromosome complex or a change' in the actual
number of chromosomes. This close correspondence between the visible
changes in the chromatin and changes in ratios or in individuals is, of
course the strongest possible evidence supporting the chromosomal
theory of inheritance. \\ Since visible changes in the chromosomes result
404 BREEDING AND Bll J 1WVEJlEN1 OF F"lHM ANIMALS
in recognizable variations in the fully developed organisms, it seems

more than justifiable to assume that the chromosomes and genes arethe_
actuaTaeterminers of potentialities.
CHANGES IN BLOCKS OF GENES
Translocation.-It sometimes happens that a chromosome becomes
broken and one of the pieces attached to some other chr~r;o~ome:
-Bridges discovered that a group-of genes that normally belonged in the
II _ghromosome suddenly showed linkage relations with genes in the III
-chromosome, and this was termed translocation. If the Riece beco;nes
.-attached to a DouhomQ)ogQus chromosome,- a new linkage rel;:iion -is-'

'set up. If, for instance, a small piece of the X chromosome in Drosophila
became attached to one of the two large pairs or to the small IV chro-
mosome, then the genes that were contained in the broken piece of X
chromosome and that had formerly shown sex-linked inheritance would
no longer do so. Likewise, if a piece of one of the autosomes was broken
off and became attached to an X chromosome, these autosomal genes
that had formerly shown ordinary Mendelian inheritance would now
show sex-linked inheritance. ~ I
Such a case involving a block of autosomal genes in the II chromosome

was discovered by Altenburg. This block of genes suddenly began to
exhibit sex-linked inheritance as if it had become attached to an X chro-
mosome. Cytological examination revealed the fact that one of the
members of the II pair of large autosomal chromosomes was very much
reduced in size, whereas one of the X chromosomes had been increased
in size by just about that amount. Many cases of translocation have
been studied in recent years. They are generally first evidenced by
peculiar genetic behavior and then checked cytologically, though the
reverse of this process has been true in a few cases.
Reciprocal Translocations.-W e learned earlier that during meiosis
there might be an equa~_~terchall&: of various-sized blocks of genes
between nonsister, homologous cbromilfids. ThIs process :"vas called
c;:ossing O-ver. S~o, may there be interchanges of blocks of genes
between two nonsister, nonhomologous chromatids or perhaps between
nonhomologous chromosomes (illegitimate crossing over). This is called
recriprocal translocation, or segmental interchange. In ordinary trans-
location there is no trading between the chromosomes or chromatids, a
section of one chromosome simply going over to another chromosome
without any reciprocity on the part of the latter. In reciprocal tra!1s-
location, there is reciprocity, and it is between nonhomologous chromatids
or chromosomes. This type of translocation happens more frequently
than does simple translocation.
\.
l'HE PRINCiPLES OF VL1RIAl'ION 40;j
If ihere were reciprocal translocation between one end of the II and

III chromosomes in Drosophila, then at synapsis, when like genes
apparently attract each other to form the synapti~:nQt, a ring of chro-
mosomes would be formed as indicated in Fig. 118.
These ring formations are known as catenations.. Many peculiar
synaptic configurations can be explained by means of reciprocal trans-
locations, for the homologous parts of chromosomes apparently attract
each other at synapsis. Chains and other queer formations of the
chromatin material also result from the dtraction of homologous parts
of chromosomes that through reciprocal translocation have come to
occupy abnormal places.
)~~f .)~~(2
3 3 3 3
FIG. 118.-Diagram showing reciprocal translocation and catenation.
Since these abnormal formations of the chromosomes at synapsis

are found in nature, many having been discovered by Blakeslee, Belling,
and others in the .Timson weed (Datura), as well as being possible of
creation by experimental treatment of plants through irradiation, it is
assumed that they, along \vith all the other types of gene and chromo-
somal changes, are responsible for certain evolutionary changes. It has
been shown for instance that the Fl hybrid between a "North and a South
American species of Datura shows two circular formations of four
chromosomes each, together with eight synapsing pairs. Painter's
discovery of a similar amount of chromatin material in the rat and the
mouse, although differently arranged, has furnished the basis for the
belief that these two species have arisen from a common ancestor through
translocations and gene mutations. l
The present wide complexity of the gene contents of the chromosomes must
also be due in large measure to the translocations which have occurred in the
past. For example, in Drosophila the numerous experiments have shown that
genes producing reactions which control the same part of the body (such as the
eye) may be located in different chromosomes. This may be due to ancient
translocations which duplicated the genes for the particular character concerned.
Later these would become differentiated from one another by changes in the
genes, and we now find these similar genes in the different chromosomes all
1 HURST, C. C., "The Mechanism of Creative Evolution," p. 81, The Macmillan
Company, Kew York, 1932.

40(:\ BREEDING AND I21,fPROVEjvIENT OF FARM AjVIMAJ,8
producing a reaction on the same part of the body. In this way the complexity
of organisms must have been immensely increased, and it is evident that trans-
locations giving rise to transmutations of chromosomes have played a great part
in the general progress from lower to higher organisms which has taken place and
they may be regarded as one of the prime factors in Creative Evolution.
Inversion.-Another type of chromosomal mutation ·is that called

inversion. In this situation a portion of a chromospme has in some
manner become inveretd. This might happen because a chromosome
became looped upon itself and then failed to uncoil in the normal manner.
This possibility is i.ndicated in Fig. 119.
ABC D A D A C B D
cOB
l'·IG. 119.-Diagram of process of inversion.
An instance of inversion is found in Drosophila. There are two

species of Drosophila (melanogaster and simulans) that are interfertile,
although the hybrids are sterile. Chromosome maps for both species
have been constructed from the data of breeding experiments and these
show that the genes are very similarly located in the two species in
chromosomes I and IV. In chromosome II the genes so far studied
appear to be in the same order although there are variations in the
distances between genes in the two species. In chromosome III not only
are there differences in the distances between genes' but in about the
middle of the chromosome in D. simulans a section has become com-
pletely inverted in comparison with D. melanogaster.
Deficiency.-Still another type of mutation is that known as deficiency.
This arises from the loss or inactivation of a portion of chromosome,
which is not lethal provided the deficiency is not too large and the
organism is not homozygous for the deficiency. One of the best known
examples of deficiency is that leading to the production of notched wings
_-
iR.prosgphi~~. Notch behaves as a dominant, is lethal whel!_homozy-
gous, and is located at about point 3 in the X chromosome.!
A Notch female from a red-eyed stock crossed with a white male gives Notch
and normal offspring as expected, but all the Notch flies show the recessive white
eye (which is located at 1.5), as though the Notch female contained no wild-type
allele of white. The normal-winged flies are all red-eyed. Similarly, the gene
facet, at 3.0, showed this peculiar pseudodominance when crossed with Notch,
and the gene Abnormal at 4.5 was also affected. Bridges and Mohr, who studied
this condition, assumed that Notch was due to the absence of a piece of chtomo-
1 SINNO'IT, EDMUND W., and DUNN, L. C., "Principles of Genetics," 3d ed., pp.
233-234" McGraw-Hill Book Company, Inc., New York, 1939.
some in this case bearing the loci from white through Abnormal. This was called
a deficiency and in conformity with the assumption, the map of the X in Notch
is abou t four units shorter than normal. Much later the actual absence of this
section of the chromosome in Notch females was proved by examination of the
salivary gland chromosome.
Notch ~ (Red Eyed) While ~
)I_l~ --....--\-::----------- ---_

_~ .t
.. --_::.::.::::-
"l
//
-~--~':::. :.:>-<==:=-;-:.-;.--<:.:~--------//
r/ r:---- ~ :/- ~--1
/'
( ----:
'",
Notch White ~ Normol Red ~ JDies Normal Red ~

(Deficiency Lethal}
FIG. 120.-The inheritance of notch deficiency in Drosophila melanogaster. (Sinnott and
Dunn, Principles of Genetics.)
Duplication.-We have seen in the previous section that a chromosome

may become deficient or lacking in certain genes by a portion of a chro-
mosome becoming lost. This might occur through unequal crossing
over between two homologous chromosomes. If this--happened, the
unequal-interchange, besides producing a deficient chromosome, would
also give rise to a chromosome with some genes duplicated. This is
illustrated in Fig. 121.
I~
One of the above chromosomes is A
deficient for genes Band C, whereas the C C
B B AHA B
C
other has a double dose of these genes, i.e., D 0 B
E £ C
they are duplicated. o
A chromosome fragment, however it E
FIG. 121.-Diagram illustrating
arises, is known as a duplication, and it both deficiency and duplication due
may exist by itself or become attached to unequal crossing over.
to some other chromosome.
One of the best known duplications is that of the bar eye pattern in
Drosophila, long known as a dominant at point 57 in the X chromosome.
Zeleny found that hOplOzygOUS bar-eyed flies very occasionally produced
normal round-eyed offspring and even less frequently the ultra- or
double-bar pattern. By the use of other mutant genes close -to bar,
Sturtevant and Morgan showed these exceptions to be due. to uDf'qllal
408 BREEDING AND IMPROVEJ1fENT OF FARM ANIMALS
crossing over at the bar region, so that the normals showed a slight
deficiency compared with bar and the double bar a duplication. Since
the normal flies showed no actual deficiency, it was reasoned that bar was
originally a duplication. These facts have been substantiated by
Bridges's cytological study of the salivary-gland chromosomes, which
shows bar to be a duplication. The further fact that the duplication in
separate _c~romosomes pr()_r1_11Q!")_~J:)ar, and that, when_the duplication
occurs in one-chromoson:te, it produces double bar led Sturtevant to the
conclusion that the position of the gene in the chromosome, as well as its
nature, was influential in the final effect produced. This is the position-
effect theory.
Duplications may also arise from crossing over between inverted
portions of two chromosomes as illustrated in Fig. 122.
1 2 3 4 1 2 3 1
~ '-------_/
",------- .....
/4231'
S;jI1apsis- inverted Non cross-over Cross-over
chromatids chromatids chromCllticls
FIG. 122.-Diagram of duplications due to inversions.
Heteroploidy.-A new organism generally arises from the union of an

egg and a sperm in each of which the number of chromosomes has been
reduced to the half or haploid number (n). The coming together of
these two half numbers at fertilization again restores the chromosomes
to the diploid number (2n). Sometimes a new organism gets one (rarely
two) too many or one too few chromosomes and is therefore 2n + 1 or
2n - 1. Many of these are viable and have been studied both genetically
and cytologically. These individuals differ slightly f.rom the wild type
in appearance and are often less viable, less fertile, and do not breed true.
In Drosophila, forms are knm~n th~t lack ~~e member of the small IV
" pair of chromosomes. Such a fly with only one IV chromosome is
called a haplo-IV fly, and one with an extra IV chromosome is called a
tl'iplo-IV fly. These individuals, with any greater or smaller number of
chromosomes than the normal for the species, vary from the normal to a
greater or less extent. In other words, the normal number of chromo-
r-;omes is necessary for the production of a "normal" individual, though
this does not mean that those individuals not having the normal number
of chromosomes will necessarily be unable to survive and leave offspring.
Triplo-IV Drosophila have narrow, pointed wings, smooth eyes,
course bristles. When crossed with eyeless~··tIiey give-_Yz triplos: Yz wil~
type. These triplo F 1 flies crossed back to eyeless give 5 wild type: 1 eye-
less. N ondisjunction (~.Ohe_tWQ_~_2hrom_()s()I!l~S, discovered by Bridges,

prodyc{}S....aDJ.>_Hi~r: l.grm of}~~>t.~!Qpl_Q0y. White-eyed females crossed
With red-eyed males should yield red-eyed females and ,vhite-eyed males
but occasionally do just the opposite owing to nondisjunction of the X
chromosomes following synapsis.
In the Jimson weed, which has 12 pairs of chromosomes, Blakeslee,
Belling, and their coworkers have discovered many types of heteroploids.
Anyone chromosome present in a triple dose gives ,,-hat is termed a
primary heteroploid. All the possible 12 primaries have been studied,
and each of them has a characteristic mutant appearance. Another
type with the extra chromosome made up of parts of two homologous
chromosomes rather than being just an additional whole chromosome
is called a secondary heteroploid, and still another type with the extra
chromosome made up of parts of two heterologous chromosomes is called
a tertiary heteroploid. These latter two types give many peculiar ring
formations of the chromosomes at synapsis.
In the evening primrose (Oenothera), in which the diploid number of
chromosomes is 14, the following number of chromosomes, according to
Gates, have been reported: 15, 20, 21, 23, 27, 28, 29, 30. Those with
other than 14 chromosomes have definitely distinguishing somatic char-
acters, which indicates again that the characters are under the control
of the genes.
Haploidy.-The haploid condition is normal in some species that
reproduce parthenogenetically, e.g., ~es, wasps, some moths, and rotifers.
A few other species can be made to reproduce parthenogenetically by
artificial means, e.g., certain amphi~i~ns. This has recently been
accomplished by Pincus in the rabW,;:ith the probable early restoration
of the diploid condition in the embryo by means of division of the chromo-
somes without subsequent division of the cell. Artificial haploids are
generally small and lacking in vigor and very rnrely produce viable
gametes.) Wher;viable germ cells are produced, they are undoubtedly
tIle ~esUlt of meiosis without reduction. Haploids can be induced by
various environmental means such as cold and irradiation.
_, Polyploidy.-Forms with some multiple of the haploid number other
than bvo are called polyploids. The most frequent ones found are
triploids (3n) and tetraploids (4n), with occasionally pentaploids,
hexaploids, and octoploids appearing. PolZ2!gXda_fl,[~_~~own in many
plants, e.g., :qatllr~ Oenothera, ~s, and some mosse~arurln a
few of the lower animal forms, mcluding Drosophila. Th~robably
~is~ naturally th~ough fa~lure()fthe germ cells to divide after the-ii"Uclear
~eri~~divided. / They can also be induced artificially by various
environmentaT agencies, and recently a method for ir..ducing polyploids
410 BREEDING AND IMPROVEMENT OF FARM ANIkIALS
in plants through the use of colchicine has been developed. Polyploids

have a commercial as well as scientific interest, because the plants and
fruits are generally larger, heavier bearing, and more vigorous.
Triploids generally result from the crossing of diploid and tetraploid
parents. Reproduction by these forms is, as expected, very irregular,~
for the three sets of chromosomes assort at random into all possible'
combinations, thus giving very few functional germ cells ..
Tetraploids are of two sorts. Those which arise within one species
are called autotetraploids. These reproduce approximately normally,
for the four members of each set of chromosomes come together at
synapsis forming a quadrivalent from which two of the homologues
pass at random to each gamete. A cross of a purple-flowere<!~etr~J2!.oid
Datura (PP PP) and a white-flowered (pp pp) gives a purple Fl (PP pp)
and an F 2 ratio of 35 purple: 1 white, for the gametes forming at random
will be 1 PP, 4 Pp, and 1 pp.
Tetraploids formed by crosses between two species or genera, with
subsequent doubling of each parental set of chromosomes to form a
tetraploid, are called allotetraploids. One of these was reported by
Karpechenko in 1928. It resulted from a successful cross between the
radish and cabbage. The haploid number in each of these species is 9,
so that the F 1 hybrid had 18 chromosomes. These hybrids produced
a few seed, enabling an F 2 to be produced. Some of these F 2 plants
resembled the hybrid and on examination were found to contain 36
chromosomes. Study of the meiotic divisions of the F 1 showed 18
univalents, indicating that no pairing had taken place between the radish
and cabbage chromosomes. It is assumed that the F2 resulted from the
union of gametes with the combined chromosome make-up. The F2
wer~_~v~L~etraploids with_I8 ra~h~d ~ag_~chxomosome~:
These plants were fertile and breatrue, each transmitting 9 radish and
9 cabbage chromosomes through each germ cell. This intergeneric
hybrid is a new, fertile species that cannot be crossed back on either
parent. L_!t d~~onstrate~_that new species can arise.l£()!:Il._:;p~ej~l?_,or
~1!~rl1 cr~ It is, in effect, evolution at work.
If any change in the gene complex or in the distribution of chromo-
somes arises in the egg or sperm that are to give rise to the new individ-
ual, in the developing embryo previous to the laying down of the germ
plasm in its testicles or ovaries, or in the viable germ cells produced by
the individual after sexual maturity, variations will result in the off-
spring which may be heritable. Aberrations in the distribution of
chromosomes may occur at any of the millions of cell divisions that
must intervene between the zygote and the completely developed indi-
vidual. Many such aberrations are known. This means, of course,
that any animal may, under certain conditions, transmit characteristics

which he (or she) does not exhibit and may not transmit characteristics
which he (or she) does exhibit.
--------
:Mutations and freaks should not be confused. A mutation is a
sudden germinaL variation that breec!~_true. A freak is probably a
~suddensomahc vanation, s~wo-headed calf, a six-legged pig,
~and many others that may be seen at the circus but which will not breed
true. The freak has no genetic value, and, although the mutation may
not have value, if often does, and, because of the fact that it breeds
true, the characteristic may be perpetuated.
Breeding Systems and Variation.-It is obvious from the foregoing
that genetic variation is dependent on what transpires in the genes
, and chromosomes. In addition to these gene or chromosome changes,
however, there is the further consideration of the sorting out of the whole
genetic material into different sorts of combinations. Self-fertilizing
plants tend to separate themselves into homozygous forms. If one of
these plants were Aa, its offspring would be 7~ AA, .% Aa, 7~ aa. Con-
tinued selting would constantly change the proportion of the genotypes
according to the formula 2n -1 AA: 2 Aa: 2n -1 aa, where n is the
number of generations of self-fertilization. In other words, about one-
half the heterozygous gene combinations become homozygous each
generation the inbreeding is continued. This is true for every pair of
heterozygous genes that the original parent might have possessed.
Even with 15 pairs of heterozygous genes, practically all of them will
have been separated out into homozygous forms by 10 generations of
selting. There would, of course, be many combinations of these 15 pairs
of genes, but they would all be sorting out into homozygous forms.
With two pairs of genes, Aa Bb, we would be getting AA BB, AA bb,
aa BB, and aa bb forms, or 22. With 15 pairs of genes, we would get
2 15 different forms. Selting i§__of _col!rse th_f;_closest sort of inbr~~ding
possible. It should be:iJ.~ted that, gr~tea eq-UarvrabilitY;---t11;~u~b-er
~r~cessive and dominant genes in the population remains the same.
Inbreeding, therefore, is seen to have a direct effect on variation.
When it is practiced, the population tends to increase in variability by
becoming broken up into many separate, though highly uniform, families
or strains. This, of course, offers greater opportunities for selection
within desirable uniform lines, and, when this sort of selection is prac-
ticed, the population becomes more uniform, owing to the selection,
however, rather than the inbreeding.
In the higher animals, however, selfing is not possible because the
union of an egg and a sperm produced by two separate individuals is
necessary to 'produce a new individual. If we start with the animals
Aa and Aa and their offspring interbreed freely, the population will

become X AA, Y2 Aa, and ~;! aa. The recessive gene is again, granted
equal viability, just as numerous as the dominant and does not tend to
become obliterated unless it has a lower survival value or unless selection
is practiced against it. Whatever the original proportion of the two
genes is, the mechanism of inheritance tends, in a freely breeding large
population, to maintain this proportion. If we let q represent the
original proportion of A, and 1 - q the original proportion of a, then
in any succeeding generation the proportions of the homo- and hetero-
zygotes will be q2 AA: 2q(1 - q) Aa: (1 - q) 2 aa.
If we knew that 1 per cent of Holstein calves were the recessive red
(aa), then we would also know that 81 per cent were homozygous black
(AA) and 18 per cent heterozygous (Aa), because, if (1 - q)2 = 0.01,
then (1 - q) = 0.1, and, since the whole population is unity, or 1, then q
must equal 0.9 and q2 0.81 and 2q(1 - q) = 0.18. If inbreeding were
practiced in this population, the proportion of homozygotes would of
course be increased at the expense of the heterozygotes.
The above discussion of genetic equilibrium applies only to very large
populations. In small populations, however, the mechanism of inherit-
ance tends to reduce variation by eliminating certain genes at the same
time that it is fixing others in a homozygous state.
Nicking.-If offspring of certain matings are of a high order of excel-
lence and in general better than their parents, then breeders say they
"nicked" well. It sometimes happens that the mating of two average
animals results in the production of an outstanding individual from the
standpoint of type, performance, or both. This effect may, of course, be
largely due to environment. It may, on the other hand, be due to
inheritance. Possibly each parent was lacking in certain genes that
would make for excellence although possessing a great many of the
necessary genes. If the two parents lacked in different genes, the off-
spring might get a full complement, some from each parent. Crosses of
white-eyed and eyeless Drosophila result in normal red-eyed offspring.
One parent is recessive for a pair of genes making the eyes white, though
it has the normal genes for presence of eyes. The other parent is reces-
sive for the genes for presence of eyes but has the normal genes for red
eyes. When mated, each parent supplies what the other lacked, thus
giving normal red eyes.
It must be remembered, however, that these offspring are heterozy-
gous and cannot breed true. Sometimes the argument that geniuses
do not always beget geniuses is used against the chromosomal theory of
inheritance. If, however, the genius arose as did the red-eyed flies just
mentioned, we would not necessarily expect him to beget geniuses, so
7'HE PRINCIPLES OF VARIATION 413
that in place of refuting the theory of inheritance, facts such as these

tend to substantiate it. When favorable "nicks" arise, their breeding
ability must receive careful scrunity before they are approved, because
it is possible that their diverse origins will be manifest in their offspring.
Prepotency.-The term prepotency is often used in rather a loose
fashion and is applied generally to males. It is usually thought to be
associated with the generalcllaractenstic of vigor. That is, it is applied
very often to an ou~~_~,jgorous male. The_ ter~_ ~~ll,l~_~~ans
the ability of an animal, male or feffiaTe;tOstamp its own cliaract~_ristics
~n its off~pring:-F(if instance; The H~~;;fo~d bulls us~d ~n the grade
;;ttle~f the range generally put white faces on all the offspring of such
matings. If a Hereford-Angus cross is made, however, a white-faced
black-bodied 1 polled individual is produced, and the question arises as
to which animal is prepotent. Some will say the Hereford because of
the white face on the offspring. Others will say the Angus because of the
polled characteristic and black body. The truth is, however, that one
is not justified in speaking of either animal as being prepotent. The
animal is not prepotent; the various characteristics that the animal may
exhibit mayor may not be prepotent. In this case, polledness of the
Angus, white face of the Hereford, and black body color of the Angus is
each prepotent. The same results would be secured no matter which
one of the breeds in the above cross was selected as the male and which
one as the female and also regardless of vigor. Graphically the results
in regard to these three characters and with the dominant characters
italicized might be represented as follows:
PP palled!
ww black face Angus
BB black body
pp horned !
WW white face Hereford
bb red body
~PWB pwbl PpWwBb

polled, white-faced, black-bodied crossbred
It is not enough, however, to say that certain characters are pre-

potent. When the above crossbred produces spermatozoa or ova, as
the case may be, the factors will separate and recombine into all the
possible combinations, reduction taking place at the same time, so that
ova or spermatozoa of the following make-up are produced: PWB,
PWb, PwB, Pwb, pWB, pwB, pWb, pwb. If one of the above crossbreds
is mated to a Hereford pure for horns, white face, and red body, or
1 The Angus might be heterozygous for color and therefore give rise to a red-bodied
calf.
414 BREE DING AND JU PRO YKM1!JNT OF FARM ANIMA LS
pp WW bb, and hence capable of producing only p lVb spermatozoa or

ova, calves will result 50 per cent of which will be horned and 50 per cent
of which will be red-bodied. Thus, the characters which are fully pre-
potent in the purebred, viz., polledness and black body, cease to be fully
prepotent in the crossbred, because the crossbred produces more than one
kind of reproductive cell, actually producing eight different kinds as to
these three characteristics of horns, face color, and body color. The
above characters are simple in their natmc, being controlled specifically
by one factor, although many other factors may be involved. According
to Wright:
Most characteristics probably depend on a large number of hereditary units,
but nevertheless the nature of prepotency is believed to be essentially the same.
So far as there is prepotency, it is a property of characteristics (or really of the
hereditary units back of the characteristics), not of individuals, breeds, or sexes,
and whatever the characteristics, there can be no prepotency unless the individual
produces only one kind of reproductive cell so far as it is concerned.
FIG. 123.-Inheritance of a recessive pattern of white spotting seen in "hooded" rats.

The parents (le/t) are a homozygous hooded mother and a heterozygous "Irish" fathet:
(black with white belly). An entire litter of their young is shown at the right. Four are
homozygous hooded like the mother, five are heterozygotes like the father . Note fluctua-
tion in both classes. Such fluctuations are found to be in part heritable. (From easUe,
Genetics and Euaenics, Har1)ard University Press.)
Technically, prepotency depends upon the character being dominant
and the individual being homozygous; i .e., the individual must carry AA
in its chromosomes, not Aa or aa. The more prepotent an_individual i~
1J.le le~he probability _of varia.:tiQn:good .or _b~d. \ .
A prepotent animal will have offspring that closely resemble it, but
whether or not these offspring are themselves prepotent will depend on
what genes they got from their other parent. If they have many homo-
zygous dominant genes, they will be prepotent. An animal is seen,
therefore, to be unable to transmit its own prepotency because it con-
tributes but one-half of its total inheritance to any of its offspring. .
Variation Guided by Selection .-Although it may be questioned
whether selecting animals with a definite end in view ran specifically
be said to cause variations in that direction, it is, nevertheless, true

that one variation in a given direction seems to make other variations
in that direction more probable. Thus it is well known that by selecting
in Holsteins or Ayrshires the proportion of pigmented to nonpigmented
area may be shifted from one of almost entire pigmentation to one
almost white. Likewise, in Herefords, the white of the face, which
behaves as a unit in inheritance, may be increased or decreased in area.
Functional variations, such as speed in horses, milk production in cattle,
wool production in sheep, etc., can also be guided by selection-in fact,
it has been selection, both natural and artificial, that has given rise to
most varieties and perhaps species. Castle states 1 as an interpretation
of factorial behavior in such selection:
Many students of genetics at present regard unit-characters as unchangeable.
They consider them as impossible of modification as are the atoms. To recall
Bateson's comparison, the carbon and oxygen of carbon monoxide, CO, are eaeh
FIG. 124.-A series of grades for classifying the plus and minus variations of the white
spotting pattern of hooded rats. (From Castle, Genetics and Eugenics, Harvard University
Press.)
unchangeable. Adding another atom of oxygen does not alter them, though it
changes radically the compound formed which becomes carbon dioxide, CO 2,
possessed of very different properties. But the carbon and the oxygen are still
there unaltered and recoverable. This question is one of great practical impor-
tance-are unit-characters as constant as atoms, so that we can merely recom-
bine them, or are they different in nature from atoms so that we can modify as
well as recombine them. Much careful work has been devoted to the solution
°
of this question. It was at first assumed from chemical analogy that characters
which behave as units in heredity must, like C and in the case of carbon dioxide,
emerge from combinations unmodified. But presently case after case came to
light in which this was not true. Albinism emerged from crosses tainted with
color; clear yellows emerged from crosses intensified to red, or diluted to cream,
or sooty with minute quantities of black; patterns such as are seen in Dutch or in
English rabbits, or in hooded rats, emerged considerably altered in appearance.
1 CAS'fLE, W. E., "Genetics and Eugenics," pp. 237-240, Harvard University Press,
Cambridge, Mass., 1930. (Heprinted by permission of the President and Fellows of
Harvard College.)
Facts such as these were interpreted in two different ways. It was assumed by
;5ome that the actual unit-character, factor, or gene involved was subject to
quantitative and possibly to qualitative chapge. By others it was assumed that
the observed character changes were not due to changes in single genes but to the
supplemental or modifying action of the other genes.
Hooded Rats and Selection.-As a test of these rival interpretations, we may
discuss a typical case. The hooded pattern of rats clearly behaves as a simple
unit-character allelomorphic to Irish pattern or to self in crosses. But the hooded
pattern as seen either in pure-bred or in cross-bred litters of young varies slightly,
and such variations have a genetic basis since by selecting either the whitest or
the blackest individuals, one can either whiten or blacken the average racial
condition. Races corresponding with the extremes of the series shown in Fig. 124
were thus produced. The question now arose whether the observed changes had
occurred as a result of change in the single unit-character or gene clearly con-
cerned in the case, or whether this was due to other agencies. To test the matter
the selected races, now modified genetically in opposite directions, were crossed
repeatedly with a non-hooded (wild) race. The recessive hooded character
disappeared in Fl but was recovered again in F2 in the expected 25 per cent of
this generation. These extracted hooded individuals, following each cross, were
less divergent than their hooded grandparents from the ordinary hooded pattern.
After three successive crosses (six generations) the whitest individuals extracted
from the dark hooded race were no darker than the darkest individuals extracted
from the white hooded race. In other words repeated crossing with the non-
hooded (wild) race had caused the changes in the hooded character, which had
been secured by selection, largely to disappear. The conclusion was drawn that
the hooded allelomorph itself had remained unchanged throughout the selection
experiment, and that the phenotype had been altered by associating with the
hooded gene a different assortment of other genes in each of the selected races,
these serving as genetic modifiers. In the course of the selection experiment a
mutation was observed to occur in the plus series to practically the Irish stage.
This is not included in the summary but is mentioned to show how it is possible
for selection to be aided in its progress by the occurrence of contemporaneous
genetic changes, no less than by the sorting out of variations originally present
in the foundation stock. Apart from the mutation mentioned the results of
selection in this case show conclusively that the changes obtained had not occurred
in the gene for the hooded pattern, but in the residual heredity. Other cases of
apparent gradual change in unit-characters under the action of selection may
be explained in a similar way. Accordingly, we are led to conclude that unit-
characters or genes are remarkably constant and that when they seem to change
as a result of hybridization or of selection unattended by hybridization, the
changes are rather in the total complex of factors concerned in heredity than in
single genes.
Types and Frequency of Variation.-All animals will produce varia-
tions to a greater or less degree, depending upon the interaction of a
great many factors. So~etimes e~~ely vaJ.ll_I1_1;>l~_a._nimm§__~:r:~~~om
the mating of mediocre parents. Such variates should be studied

-carefully as to their ability totransmit their own desirable features,
and nothing along this Ime shoilJd be taken for granted. Of decidedly
more value than the animal that produces one such outstanding indi-
vidual and a host of inferior ones is the animal that has a tendency to
beget a uniform lot of offspring better than the preceding generation,
even if the increase in excellence is small in each individual.
Innumerable instances could be cited of full brothers that are evidently
genetically unlike, because their offspring are very unlike. The same
is true of full sisters; e.g., in cattle, one sister is often vastly superior to
another in milk production. It is quite evident, therefore, that variation
is a very potent and ever-present force molding new forms both ana-
,tomically and physiologically, and that its mechanism is to be sought
in the genes and chromosomes, their interactions with each other, and
their potentiality for individual change.
Summary.-We have seen in this chapter that the genes and chro-
mosomes may undergo a great variety of chan~s. As WeWOufoexpect,
such-geneo~-chromosomal changes te-n:d to be hereditary, for the genes
and chromosomes are the actual mechanism of inheritance. The
breeders' raw material for fashioning finer animals is variation. He
must determlne-wh~t the cause of the variation is and so pla~ hls breeding
operations to conserve the good and weed out the bad variations. He
does this largely by selection and systems of breeding. Genetic variation
is a complex mechanism and is made more difficult by the effects of
various types of dominance and epistasis as well as by the fact that
most characters are more or less elastic under different environmental
rnffuerces.I An ~nderstanding of the underlying mechanism of variation
makes us more cognizant of the inherent difficulties involved as well
as more confident of ultimate success when our understanding of prin-
ciples is fully mature.
References'
CHAPTER XVI
THE PRINCIPLES OF VARIATION (Continued)
As stated in the previous chapter, there are two sorts of variation,

genetic and somatic. We have seen that any changes in the genes or
chromosomes may be inherited, because the genes and chromosomes
from the parents form the actual "bridge of inheritance" between any
individual and all of its millions of ancestors. We also learned that
variations might be due to internal or to external stimuli. In this
chapter we shall be concerned primarily with somatic variations that
are generally caused by external stimuli, or, in other words, by the
environment.
It seems almost certain that early in the evolutionary process the
environment must have wielded a marked influence on the simple forms
of life and caused their germ plasms to vary in different ways. Experi-
mentally, this is possible in plants and lower animal forms today. In
the higher animals, however, with their various adaptations for pro-
tecting the delicate germ cells, the possibility or extent of such external
influences is much lessened. Parallel induction (same influence affecting
both soma- and germ plasm) in the cases of alcohol or lead poisoning
must be admitted, but here the effect appears to be temporary.
Environment and Somatic Variation.-Among the effective external
stimuli are light, temperaturez fo 0.9. , moistuIe, and other influences
tending to make the organism deviate from Its normal as specified by
the make-up of the germ plasm that gave rise to it. As pointed out by
Babcock and Clausen, external stimuli may affect the development of
characters in three ways.
1. They may modify the development of inherited characters; e.g.,
red, white, or bhieTIght causes marked ;;'riations in the showy Sedum,
even "though the -;Weral plants be of the same genetic make-up. Tem-
perature affects the degree of pigmentation in moths and butterflies-:-
Food supPly affects the relative SIZf) of"tmay _parts of Hyalodophina, a
crustacean. Moisture affects the plumage color of pigeons.
2. They actually condition the production of characters whose hered-
itary determiners are present in the germ plasm. For example, light
conditions normal development in plants. Temperature, when low,
causes red flowers in the Chinese primrose, when high, induces white
\
418
THE PRINCIPLES OF VillUA'1'lON 419
flowers. The kind of food supplliid to the larvae of bees determines

whether the f~s shall be fertile (queen bees) or infertile (workerS).
Moisture concentratio;];rings~ut development of abnormalities in
black banding on the abdomen of Drosophila, a fruit fly.
3. They may cause germinal variations that result in the appearance
of new heritable characters.
:Modifications may be impressed on the soma from the outside; gene~c
variations-;;;-expressed from the inside, specifically from the germ
plasm. The latter are heritable, whereas modifications seem to be
nonheritable. If it is possible to modify the function of any orgamsm
in a '~ay, it must be because the organism inherited the ability
to react in such a way. This does not necessarily mean that offspring
'from such an individual would necessarily show the same aptitude to
modification. That would depend on the genetic make-up of the
individual in question. The possible effect of amphimixis, or the bringing
together of two diverse bits of germ plasm, must also be taken into
consideration. Animals of a good strain that are themselves stunted
and not developed to the limit of their inherent capacities may beget
offspring of a type capable of great development. Daughters of scrub
cows, developed und~r an environment tending toward full development
of capabilities, produced 13 per oent more milk and 12 per cent more
butterfat than their scrub dams of the same type of breeding, which had
been developed under less favorable conditions.
It is the province of animal breeders to develop fully all the capabiiities
that their animals possess. Not that these so-called "modifications"
will necessarily be inherited, but, if the parents exhibit favorable modifica-
tions, it must be because they carry that particular type of germ plasm,
and, unless they are extremely heterozygous or amphimixis interferes,
the offspring of such parents should in general exhibit like capabilities.
Modification is an aid to selection but like all other aids must subject
itself to the breeding test.
Acquired Characters.-The inheritance of ac9.!!i!ed characters has
b~~ll called" one of the hi;t;riebattle groundS-of biol;gy."" The~nimal
breeder is vitally interested in the solution of the questIon of the inherit-
ance of acquired characters. The successful breeder seeks to develop
for himself and to leave to posterity a better type of animal than the
one with which he was supplied at the beginning. It is perfectly natural
for him to hope, and perhaps believe, that any increase in quantity or
quality of produce he can bring about will be cumulative in his animals.
Before proceeding to a discussion of the evidence pro and con, it is
essential to define specifically what is meant by an acqui~~d c~ha;acter.
Shull defines acquired characters as "th~~l:l.difications of bodil;
structure or habit which are impressed upon the organism during its
individual life." As Thompson h;~ stated-it, -th~ preci~~ question at
issue is this: "Can a structural change in the body, induced by some
change in use or disuse or by a change in surrounding influence, affect
the germ cells in such a specific or representative way that the offspring
will through its inheritance exhibit, even in a slight degree, the modifica-
'lions which the parent acquired?"
Anyone of the higher organisms is the result of the union of a certain
spermatozoon with a certain ovum. The chromosomal content of the
particular spermatozoon and of the ovum that have united to produce
any given individual determines the normal characteristics for' this
individual. With normal nutriment and a normal environmel).t the
individual will approximate its own norm as specified through its inherit-
ance. Now both the anatomy and physiology of the organism are
,,
anil.so on
FIG. 125.-Schematic representation of the supposed method of transmission of acquired
characters.
perfedly capable of modification, and the precise question at issue is

whether or not such bodily modifications will be impressed upon the
germ cells and thus become a part of inheritance.
In an evolutionary sense all the characteristics that present-day
organisms exhibit have been acquired, for evolution teaches that life
originated in a very simple form and that it has gradually grown more
complex-in other words, has ever continued to evidence new char-
acteristics. Even this, however, is far from proof that changes in the
somatic cells caused changes in the germ cells. May not the reverse
of the last statement be the truth of the matter? Aligning cause and
effect into their true relationship is not always an easy thing to do.
Lamarckism vs. Weismannism.-Lamarck, a French scientist, and
Darwin, the famous English author of "The Origin of Species," believed
that acquired characters could be inherited, and for a mechanism it was
postulated that the blood stream which bathes all parts of the body
would have something (gemmules) put into it by the part affected or
by the character acquired and that the blood, later flowing to testicles
and ovaries, would deposit this something in the ova or spermatozoa
and thus bring about the development of the character in the offspring.
This is the theory of pangenesis.
Weismann has been the chief antagonist of the theory of the inherit-
ance of acquired characters, and he also developed the idea of the con-
tinuity of the germ plasm. Under this conception the germ cells that
any individual produces are thought of as arising, as do all the other
cells of this individual, from the original zygote. Under this view the
type of germ cell produced is dependent on the nature of the original
zygote ~nd is not affected by the soma or body that has also arisen from
the original zygote. The accompanying diagrams should make clear
the differing features of these two systems.
The new Thenew

} individual } individual
~~--------------~~--------~andsoan .
Ovum
FIG. 126.-Schematic representation of the continuity of the germ plasm. The drawing
also illustrates the fact that germ cells arise from previous germ cells rather than from
present somatoplasm.
Requirements for Proof of Inheritance of Acquired Characters.-

Babcock and Clausen say:
To satisfy the rigid requirements of an experimental proof any evidence of
the inheritance of acquired characters must fulfill the following conditions:
First, a specific character or modification in the body or soma must be impressed
upon the organism by a known factor in its environment or in its exercise of
bodily function.
Second, the character or modification must be new. There must be no ques-
tion of the reappearance of ancestral traits or characters, or of the specific relation
of the determining factor to the character or modification in question.
Finally, the induced change in the organism must reappear to some, though
not necessarily to its full degree in succeeding generations in the absence of the
original factor which determined its production. Other conditions in the life
of an offspring must remain unchanged.
The problem of the inheritance of acquired characters can be con-

veniently divided into the five following categories: (1) anatomical
modifications, (2) environmental modifications, (3) functional modifica-
/tions, (4) pathological modifications, and (5) psychological modifications.
Anatomical Modifications.--There are abundant data relating to
mutilations. When the tail of a sheep is docked the sheep has acquired a
character during its individual development, viz., taillessness. It is also
recognized at once that this characteristic is not passed on to descendants
422 BREEDING AND IMP1WVEMENT OF FARM ANIMALS
of this sheep, but in succeeding generations it will be necessary to dock

the lambs notwithstanding the fact that docking has been practiced for
hundreds or thousands of generations. In this category come also the
clipping of the horns of cattle, docking of horses'· tails, clipping of dogs'
ears and tails, etc. In none of these cases is the acquired character con-
tinued in the offspring. In the human family, there might be mentioned
circumcision, which has been practiced for thousands of years, and also
the binding of feet to prevent their full development practiced by Chinese
women. Here, no more than in the cases cited of farm animal$l, is there.
found evidence that these acquired characters are being inherited in even
the slightest degree.
Weismann in order to test this hypothesis cut off the tails of mice for
19 generations in succession and secured no shortening of the tails or
absence of tails in any of the descendants. As Walter remarks: "It i~
good thing that children of warriors do not inherit their parents' honorable
'SCa:rs'-bToattle-elsewe would have long- since -been-;;;~~e -~f cripples."
'The same author also remarks in this regard that" evidently wooden legs
are not inherited but wooden heads are."
Environmental Modifications.-The second category is that of environ-
mental effects. Examples from both the plant and animal kingdoms can
.
be cited. Capsella, a wayside weei.\, has graduaUy climbed to a moun-
tainous habitat. As it climbed it "acquired" a dwarf character-at least
it grows luxuriantly in the valley and is small, compact, ~nd dwarfed in
its mountain home, and, when taken back to the valley, continues in its
dwarf characteristics. This can just as well, if not more logically, be
explained by saying that those plants germinally constituted so that they
grew somewhat dwarf would, for that reason, have a better chan-ce of
surviving under the more rigorous conditions and decreased food supply
that prevail in an ever-increasing degree as the mountains are ascended.
In other words, the force bringing about the change was not the inherit-
ance of acquired characters, but "natural selection," which no doubt
always has and always will, provided man does not interfere, weed out
those types and individuals poorly fitted for their environment and
preserve and prosper those better fitted to survive under the existing
conditions.
Another example is the wind-lashed trees of our sea coasts, with all
the branches growing out on the land side and pointing landward, but.'
it is not known that seed from such trees would produce trees with that
peculiar habit of growth if the trees were grown where the almost steady
!:lea breezes did not blow. The case of the persi!:ltent !:lunburn of whi_tel!
men long resident in India might also be mentioned, but this" acquired
characteristic"
\
fails to reappear in their"_ children
- -~ ,_
born in England. ----,
.-- --____.----_
Ponies inhabiting the Shetland Islands are dwarfed because of the

h~r~E ~!~matic c.?_nditions and ~esultant sc~~!y f~l::. If brou~ht
to Amenca?s corn belt,) sllch ammals wo]J.kl nodoubt give nse to offspnng
that woU1cI gre~tly excel them in size. vN ow, if these larger offspring, after
reaching maturity, a~e Shetland Islands, will they con-
tinue to produce offspring that will attain large size? It is very unlikely
that they will, and, even though they did, natural selection would give as
logical and perhaps a more reasonable explanation than the inheritance
of acquired characters could muster.
• Functional Modifications.-The third category relates to i~itance
of functional modifications due to use and disuse of parts. From a
practical standpoint, along with the inheritance of disease, this probably
furnishes the most interesting phase of the subject. It will be well at
this point to examine critically just what is meant by acquired characters.
Is the term more or less a misnomeTif so, would it not account for
much of the loose thinking and arguing over the much-debated question?
As Davenport suggests, the differences between animals in a given species
are differences of degree and not of kind-they arequa.ll_tti~_tiY~La_lliLnot
qualitative. Horses show horse characteristic sand not rabbit character-
iStics~e~~pt as both have certain general characteristics in common;
cattle show characteristics of the cattle family and not those of the swine
family, except as they have certain ones in common. Horses have draft
power or speed, cattle give milk, hogs lay on fat-some, of course, to a
greater degree than others-but they all exhibit the characteristics of their
species and not the particular characteristics of other species. In other
words,,_Jhey inherit in the germ plasm all the potentialitJ~s -~f their
species in greater or less degree, depending upon the strain to which they
~elong. In addition,. the environment in which they develop and live
determines how far t~ese potentialities are developed. Therefore, if a
horse is able to trot fast because of his having inherited the tendency and
because the tendency has been given play for full development by a fast
track, careful trainer, proper harness, etc., or if a cow inherited the
tendency to produce large quantities of milk and this proclivity is subse-
quently fully developed by proper c'are, feeding, and management, is the
assumption at all justified that these ammals have acqUIred any new,
characteristics? The fair answer is that no new characteristics have been:
acquired by either the horse or tlieco\\. Inhetent C'llaf"acters have been)
developed, perhaps exceeoingly well developed, but no new ones have
been acquired.
The point at issue is whether or not any increased fu~y 1
will be registered in the germ cells and thus influence future offspring.
Urider this heading Redfield's theory of "dynamic" evolution can be
discussed. This is the belief that the exercise of any organ or function
results in a corresponding stora:ge olenergy in the germ cells, such that
the effects are transmitted to~;'-t· i~~~rntiOn-in other w0rds, that
these '<aCqUlreacnaracters j; aietransmTtte'd. This theory would postu-
late that, if a cow were developed from an 8,000-lb. two-year-old into a
16,000-lb. ten-year-old, the calf produced in her tenth year would be a
much higher milk producer than the one produced when she was two years
old, owing to the cumulative effect of the increased production she had
attained, the same bull having been used to sire both heifers. This has a
distinct practical bearing in all classes of livestock. '
In his work to try to prove the existence of the inheritance of this
acquired character or development in fast horses, Redfield first computed
the average age of the immediate sires of the first 1,000 stallions taken
alphabetically from the American Trotting Register, which he found to be
9.43 years. He then computed the average age of all the sires, four
generations back, of the entire class of 2.10 trotters, a faster class than
that dealt with previously, the upper limit for admission to the register
being 2.30, and this average he found to be 13 years. From this material,
he concluded that the 2.10 class was faster than the group first considered •
because their sires had gradually grown more speedy as they increased in
age up to thirteen and that they had transmitted this dynamic develop-
ment to their offspring. ,
Marshall critically points, out the following shortcomings in Red-
field's work. In the first place, he took in one case the immediate sire
and, in the second case, all the sires included in the four-generation
'/ pedigrees. Marshall reminds us also that these four generations carry
us back over the formative period of the trotting horse to the time of
Hambletonian 10, who ,vas himself a great sire of speed and left many sons
and grandsons which, bred to faster mares, were even better sires of speed
than he. Taking only the immediate sires of the 2.10 list, as was done
with the first list of sires, Marshall finds the average to be 9.41 yea;rs,
nearly the same figure Redfield secured for the first group, 9.43, and
evidencing no advantage of the old sire over the young one.
In this matter of increased speed in succeeding generations of horses,
it has not been proved that functional modifications have not been con-
tinued in the next generation, though it is a biased mind which would •
attribute all increase in speed to this fact()r and fail to recognize possi-
bilities in amphimixis, possible mutations',' better environmental influ- .
ences, better harness and training methods, faster tracks, more skilled
drivers, etc. Also, as has been suggested previously, is it not more likely
that the germ cells caused the somatic changes rather than that the soma
caused the germinal changes?
There is no conclusive scientific evidence in support of the inheritance

of "acquired characters" regarding functional modifications or use and
disuse of parts. This fact, however, should not influence-l)-reeaers-to
decreasetheliefforts to provide the most favorable environment and to
attain the fullest development of their breeding stock. Whether or not
functional modifications are inherited, breeders should still seek the
germ plasm carrying the greatest possibilities for high-class production
and should seek, when offspring come from such germ plasm, to place
it in the best environment and supply the best possible feeding and
management in order that its inherent capabilities may be developed to
the fullest extent. Not that one must believe that this acquired develop-
ment will be transmitted, because there is no sure ground for any such
belief, but it is desirable to know the horse of greatest speed, the cow of ,
highest production, the meat animals with the greatest tendency to )
grow, and the sheep capable of the greatest production of high-quality
wool and mutton, because, if in general "like does produce like," it is
such germ plasm that should be used for increasing herds and flocks.
Pathological Modifications.-The fourth category, viz., the inheritance
of disease is also a consideration of vast practical importance. It should
be borne in mind that, from a scientific standpoint, the only thing that
iE; inherited is the germ plasm, more particularly, the vano~;- faCt()rs
carried in the chromosomes.· From- this standpoint it will readily be
see~that there is no more true inheritance in the case of disease than
there was in the case of functional develop~e_n_!.? The problem, however,
like the mtter, has a very practical bearing, and although it may be
said that in neither case has the affirmative been proved, yet this lack
of proof does not necessarily preclude the possibility of its existence.
One of the commonly cited cases under this heading is that of tuber-
culosis. This, however, is clearly a case of reinfection in the offspring,
abundant opportunity for which is furnished in the intimate relatio;;
between parent and offspring, though it is very possible and no doubt
true that the tendency or susceptibility to the disease is inherited in the
defective germ plasm from which have come both parent and offspring.
It is well known that it is possible to produce healthy calves from tuber-
/,-{cular cows and bulls, provided the calves are at once removed upon
being born and taken to a clean barn-in other words, that the oppor-
tunity for the parents to reinfect the offspring is removed.
In this regard, it may be well to mention the factor of immunity to
certain diseases that exists in different species and varieties. If, for
instance, a husky Irish policeman and an anemic Jew employed in a
sweat shop were both exposed to the same kind and intensity of tuber-
culosis, the husky policeman would be more liable to succumb to the
disease, because, in the Jewish race, selection has established at least

a partial immunity to this disease. The zebu is being crossed on the
native cattle of Texas because of the former's relatively greater resistance
to tick fever. The mulefoot hog is often claimed to be immune to cholera,
though the claim has never been scientifically tested and substantiated.
The microscopic organisms causing various diseases, if present in
the blood of a parent, may be carried to and deposited within the ovum
before it is fertilized. Such organisms, although not part of the chro-
mosomal material that alone is heritable, are carried in the same cell
with it and will thus cause the disease to reappear in the offspring; in
other words, the reinfection may come either previous or subsequent to
birth. The organisms causing pebrine in silk moths, Texas fever in
cattle, and syphilis in man are of this type, but, as Walter says, the
disease is no more inherited than a grain of sand placed within the ovum
would be inherited, though, of course, for all practical purposes, it
amounts to the same thing, as the offspring will suffer from the disease in
question.
A very interesting study of disease susceptibility and resistance has
been reported by Dr. L. C. Strong of St. Stephens College. The following
quotation 1 is taken from his paper:
My first experiment dealt with the inoculation of a malignant tissue (adeno-

carcinoma) subcutaneously into a series of hybrid mice of known relationship.
Two original stocks of mice were employed. Every individual of the first sto(;k
(dBr) grew the transplant progressively; and every individual of the second stock
(albino) proved to be resistant to the same tissue. By crossing these two dis- •
tinct races of mice, a hybrid generation, known as the first filial generation, was
produced, all the individuals of which reacted to the transplant as did the first
original stock, that is, all the individuals grew the transplant progressively. By
mating two individuals of this hybrid first filial generation together, a second
filial generation was obtained that contained two classes of individuals as far as
their physiological reaction to the same transplanted tissue was concerned.
Three individuals out of every four obtained were like the original susceptible
race and grew the tumor progressively; one out of every four individuals was.
negative like the second original albino non-susceptible race; a typical 3: 1 ratio
was obtained. This experiment demonstrates, therefore, that susceptibility to a
transplantable tumor is dependent upon the same ultimate mechanism that the
inheritance of eye pigmentation apparently is dependent on. A particular
individual grows a transplanted tumor because there is present in its ultimate"
make-up at least a single germinal determiner for a special physiological function
that was obtained from the zygote that gave rise to the individual.
1 STRONG, L. C., Inheritance of Cancer, Jour. Hered., 15(8) :355-360, 1934.

THE PRINCIPLES OJ;' VAlUATION 427
Roberts and Card have reported 1 on a lO-year study of the inheritance

of resistance to bacterial infection by Salmonella pullorum in chickens.
Their summary and conclusions state:
From this experiment, ten years in duration and involving more than 29,000
hirds, evidence has been obtained which clearly indicates that heredity is an
important factor in resistance and susceptibility to infection with Salmonella
pullorum. The existence of hereditary factors for resistance and susceptibility
to pullorum disease is shown by the following results:
1. Selection was effective in producing strains of the domestic fowl more
resistant than were unselected stocks in respect to infection by Salmonella
pullorum.
2. The selected stocks were consistent in maintaining resistance through
successive generations.
3. The F 1 generation produced by crossing resistant and susceptible stock
was as resistant as the resistant parents.
4. Progeny of the F 1 individuals mated to resistant were significantly more
resistant than were the progeny of the back-cross to susceptible.
5. In the F 2 generation susceptible and resistant strains were recovered by
selection.
6. A susceptible male mated to susceptible females produced progeny which
were much less resistant than were progeny of the same male mated to resistant
females.
7. No significant difference was found between the progeny of susceptible
and resistant females mated to the same resistant male.
S. Acquired immunity was not present in the experimental birds, the progeny
of infected hens exhibiting no greater resistance to disease than the progeny of
noninfected hens, infection and freedom from infection being determined by
the agglutination test.
9. Resistance is dominant to susceptibility, but probably more than one
gene is involved.
10. In an examination of the blood of noninoculated young chicks the number
of erythrocytes was found to be greater in the resistant (6 out of 7 cases) than in
the susceptible strain. The number of leucocytes was greater in the susceptible
strain. The percentage of neutrophiles was lower in the resistant individuals
(6 out of 7 cases). In inoculated chicks, the percentage of neutrophiles was
much higher among the susceptibles at 6, 7, and 9 days than among inoculated
resistant individuals of the same ages.
Psychological Modifications.-:VIany people believe in the ordinary

principles of inheritance as far as physical traits are concerned but
balk at applying these same principles to mental traits. On the other
1 ROBERTS, E., and CARD, L. E., Inheritance 'of Resistance to Bacterial Infection
in Animals, Ill. Agr. Expt. Sta. Bul. 419,1935. (Contains an excellent bibliography.)
hand, the belief that mental or artistic accomplishments are transmitted

to one's offspring is particularly prevalent. True, if two parents develop
artistic qualities their children are likely to be artistically inclined owing,
however, to the fact that the parents had sets of genes which tended in
this direction plus the very important additional fact that their children
had a very favorable environment as far as the development of artistic
qualities was concerned. Some experimental work with white rats
(McDougall) has been interpreted as the inheritance of acquired learning
of parent rats in choosing paths of escape after having been dropped into
a tank of water. There appears to have been a more or less steady
improvement in the succeeding generations of trained rats. In the
fourteenth generation the best rat made 42 errors, the worst rat 102, with
an average for all of 80. In the twenty-third generation the best rat
made 3 errors, the worst 71, with an average of 25 errors for all. Even
when reverse selection was practiced for two generations, improvement
was evidenced. On its face, this looks like the inheritance of acquired
learning. However, two important questions remain unanswered.
Was the stock pure (homozygous) to start with? Was no selection,
conscious or otherwise, practiced? Although the stock had been long
inbred, McDougall states that there were "very large individual differ-
ences between the animals in respect of the rate at which they master
their tasks."
Tryon has shmvn that the assortive mating of bright with bright and
dull with dull rats can lead to the separation into bright and dull strains
of rats as far as running a maze is concerned. 1
Griffith's work indicating the transmission of acquired disequilibration
in rats has not been substantiated, whereas Pavlov himself withdrew
his earlier claims for the transmission of acquired learning through
conditioned reflexes in mice.
There is as yet no clean-cut proof that acquired psychological modifica-
tions are inherited by the offspring. Because grammar schools have
been in existence for several hundred years, one is tempted to ask, some-
what facetiously, why our present children do not enter high school or
college rather than grammar school at the age of six, if the learning of
their ancestors actually became part of the hereditary material.
Arguments against Inheritance of Acquired Characters.-The reasons
for the general unwillingness to subscribe to the belief in the inheritance'
of acquired characters may be grouped as follo,ys:
1. There is no known mechanism in the body capable of impressing
soma or body changes on the germ cells.
1 See SNYDER, L. II., "The Principles of Heredity," pp. 395-398, D. C. Heath and _
Company, Boston, 1940.

2. The evidence, although partial in some cases, IS by no means
conclusive.
3. The theories of the continuity of the germ plasm and of germinal
variation can account sufficiently well for all observed facts of heredity.
In order to study the effect of the soma on the germ cells, Dr. Castle
at Harvard removed the ovaries of a white guinea pig and substituted
the ovaries of a black guinea pig. The operation was entirely successful,
and in due time the animal was bred to a white male and produced in all
three litters totaling 6 individuals. This female was white (with sub-
stituted black ovaries) and was bred to a white male. Albino bred to
albino in guinea pig invariably produces white offspring, but in this
case all the offspring were black. Some of these in later breeding trials
behaved as normal black guinea pigs would be expected to behave.
This critical experiment shows that the ova produced by those two
black ovaries which had been transferred to a white body were, after a
year's location in their new environment, still producing black ova,
and it is indeed inconceivable that they could ever have produced any
other kind. This experiment indicates that there is no mechanism in
the body capable of impressing soma or body changes on the germ cells,
t.hough it, of course, does not prove this to be absolutely so, it being
impossible to prove a universal negative.
Summary of Inheritance of Acquired Characters.-Castle gives the
following summaryl on the inheritance of acquired characters:
The problem of acquired characters, after all, concerns only the higher ani-
mals. In the lower animals and in plants 110 such sharp distinction exists between
body and germ-cells as we find in the higher animals. We may reproduce the
entire plant from a cutting of root, stem, or even a leaf in some cases. Hence
there is more chance in such cases of direct modification of the cells capable of
reproduction, for most of the cells of the plant retain this capacity. In the lowest
organisms (protozoa, bacteria) there is no distinction whatever between body
and germ-cells. Every cell is capable of reproduction; and modifications pro-
duced in a cell by the environment are handed on directly to the next generation.
If in the lower organisms the potentialities of living substanre can thus be
altered, it seems reasonable to suppose that the same possibility may exist in the
higher animals and plants, provided agencies capable of producing change are
allowed to act on the germinal substance. It is the sheltered position of the
germ-cells which seems ordinarily to exempt them from direct modification, but
we cannot safely assume that they are in all cases free from such modification.
Experiments of Stockard show that in guinea-pigs repeatedly intoxicated with
alcohol, the germ-cells are enfeebled so that offspring of such<~rents, whether
t"~~
1CASTLE, W. E., "Genetics and Eugenics," pp. 87-90, Harvard University Press,
Cambridge, Mass., 1930. (Reprinted by permission of the President and Fellows of
Harvard College.) .
male or female, are more likely to be feeble and sickly, and so to die. Experi-
ments of Hertwig show that similarly the germ-cells of frogs are capable of being
injured by emanations of radium in consequence of which enfeebled or abnormal
offspring may be produced.
Guyer and Smith have produred probably the best existing evidence for the
artificial production of heritable defects. Injecting into the bodies of fowls the
pulped lenses of rabbit eyes, they induced the production in the fowls' blood of
antibodies which would neutralize harmful effects of the foreign material. Blood
serum was then obtained from these immunized fowls and this serum was injected
into the circulation of pregnant rabbits and made its way, it is thought, through
the placenta into the circulation of the developing fetuses, where it interfered
with the proper development of the eye in a certain number of embryos. Eye-
defects such as opaque lenses, under-sized eyes, and eyes rotated out of their
proper position, were thus obtained in two different and unrelated races of
rabbits. These defects reappeared sporadically among the inbred descendants of
the abnormal rabbits, and so may be regarded as having become hereditary.
Guyer and Smith consider the mode of inheritance that of a recessive Mendelian
character, though fewer recessives are reported than theory demanqs; but this
may be explained as due to the authors' failure to detect cases of slightly defective
eyes, or to the failure of the character to find somatic expression in all cases.
In a third unrelated series of animals, Guyer and Smith injected pulped rabbit
lens directly into the circulation of pregnant rabbits, and in this case also they
obtained from one treated mother one young with defective eyes in a litter other-
wise normal. The defect was transmitted in this case also.
The eye-defects in the experimental rabbits of Guyer and Smith were trans-
mitted through male as well as through female parents. The authors believe
that the antibodies artificially produced by the injected foreign bodies affected
the constitution of germinal determiners at the same time that they affected
corresponding somatic structures in the same individuals. In other words they
favor" parallel induction" as an explanation of their results.
Finlay (1924) has repeated on mice the experiments of Guyer and Smith on
rabbits, but with wholly negative results. He used rat lens, sheep lens, and ox
lens to produce antibodies in pregnant mice, but without observing any eye-
defects in the young of either the Fl or the F2 generations.
On the other hand, Little and Bagg, by X-raying pregnant mice, obtained
young with pronounced eye-defects which were apparently inherited in the next
generation as recessive characters. Hansen, too, X-rayed rats in utero and
obtained eye and other defects but without studying their inheritance. Stockard
had previously obtained similar eye-defects in guinea-pigs by exposing the
mothers to alcohol fumes. Hansen and Stockard are agreed that a variety of
harmful agencies acting during the embryonic development of a vertebrate may so
interfere with the complicated process of development of the eye as to lead to
defective end results. They regard the defective eyes of the rabbits of Guyer and
Smith not as specific effects of lens antibodies, as those authors supposed, but as '
due to disturbed development, maintaining that like results can be produced by
any other disturbing agency.
If the germ-cells are capable either directly or indirectly of modification by

outside agencies, evolution guided by the environment must be in some measure
at least a reality. The truth then lies neither in the extreme Lamarckian view
that all acquired characters are inherited nor in the extreme Weismannian view,
that no extraneous influences modify the germ-plasm, but somewhere in between.
Practical Fhase of Inheritance of Acquired Characters.-Breeders
should continue to develop their animals to the utmost. If this develop-
ment is cumulative from generation to generation, breeders will profit
thereby. If not, they will at least be able to see which animals are
inherently capable of responding in the fullest degree to a favorable
environment, and then, through selection of these animals as breeding
stock and through wise matings, they will be able to bring about a gradual
improvement in the stock under their care.
Telegony.:::---Telegony may be illustrated by the more or less common
belief that, after a female has borne young by a certain male, her subse-
quent offspring ·will show characteristics derived from the previous sire.
The classic example is that of a mare ,,,hich bore offspring by a quagga
and later produced horse colts that showed some striping. Numerous
attempts have been made to confirm this by crossing mares and zebras,
but in all cases they have failed. The basic elements that determine the
characters of any individual are the ovum and the spermatozoon uniting
to produce it. Spermatozoa from this first service could not possibly
live over a gestation period in the female organs of any higher species to
fertilize the next egg, as they live for a very few days at the most. If this
first offspring had any influence on other undeveloped eggs in the ovaries,
it would come under the heading of the inheritance of acquired char-
acters, for which there is no conclusive proof. Mumford and Hutchinson
made investigations in the mule-breeding district of Missouri, where
mares often bear mule and later horse colts, but could find nothing to
substantiate telegony. In the light of present knowledge of inheritance,
telegony has neither experimental nor theoretical basis.
Maternal Impressions.-Belief in maternal impressions assumes that
what a pregnari.t ~othermay see, hear, or experience may influence her
offspring. In general, this old belief can be repudiated also, because, as
Marshall points out, if it worked, all calves born in the spring up north
would tend to be white because the mothers have viewed a white land-
scape all winter, and, similarly, calves born in the fall would be green.
There is no direct nerve connection between parent and offspring to afford
means for transporting ideas. Animal experiments have all given nega-
tive results in this field. It is, indeed, a fortunate provision of nature
that in the higher animals the embryo is so well protected from all
external influences. If maternal impressions were actually registered
432 BREEDING AND IMPROVEMENT OF FARM ANIJIALS
on the offspring, all types and races would have long ago become a
hideous, conglomerate mess.
Atavism or Reversion.-These terms, meaning the r~appearaIl_()~ ~L
some ancestral trait or character after a skip of one or several generations,
ate-often -encountered in the older literature on animal breeding. Such
, reappearances were more or less mysterious before the physical basis of
inheritance was understood. The birth of a red Angus__when the past
four generations have been bl!1ck is known to be due to each parent
supplying a gwe for 'red. Re()~§Bives may be carried along, hidden by
dgminants, -for any Qumber..(;f g~nerations:-- Wlieiiever two recessives-
~ome together, or, in other words, wheneverthe dominant gene l~-i~ck~g;
the "atavistic" character will be evident. Atavism or reversion is the
sudd~!LL~allJ.>earance of some ancestral trait, but there is nothing mys-
teri~-us about it,for it is- oneoTthenOrmai"manifestations of the hereditary
mechanism.
Summary.-We have tried to indicate in this chapter that the environ-
ment does playa large part in variation. Any organism inherits a certain
"norm" based upon the nature of the genetic content of egg and sperm
that unite to produce it. This "norm" however, may be influenced by
many internal or external factors. It is like any elastic material, dough
or newly mixed concrete. Within limits it can be molded into different
forms by the environment. However susceptible the somatic tissue may
be to molding, there is no conclusive evidence that molding of the soma
itself can have any influence on the germ cells. By regulat1ng food
intake, any parent could be one-third lighter or heavier in weight, but his
children's actual weight would not thereby be directly influenced. If we
want heavier children, we will naturally select as the parents those who
most readily put on weight. Variation therefore can be a very useful tool
in directing selection. The task of differentiating between environmental
and hereditary variation and between the various types of the latter is
not easy but on its accomplishment depends the breeder's ultimate sue..
cess. Variation can indicate new and desirable gene combinations.
Seeing, seizing, perpetuating, and amplifying these lie at the base of any
breeder's progress.
References l
DAVENPORT, K 1907. "Principles of Breeding," Ginn & Company, Boston.
KAMMERER, P. 1924. "The Inheritance of Acquired Characteristics," Liveright
Publishing Corp., New York.
\
CHAPTER XVII
SEX DETERMINATION
Sex in the higher animals is a relative matter in all except one respect,
viz., the kinds of cells produced. And even in this respect the different
types of cells, ova and spermatozoa, are essentially alike regarding
chromosome content and differ only regarding cytoplasm and motility.
For obvious reasons, man has always been keenly interested in the
factors determining sex, with the hope that he might ultimately find··
ways and ,means of controlling them to his own benefit. Because of
the ,vide interest and enormous value of such knowledge, there have
been conceived in men's minds a great number of hypotheses regarding
sex determination and its control. It is said that over 500 of these
theories have been recorded. One theory states that the right testicle
produces male-producing spermatozoa and the left testicle female-pro-
ducing spermatozoa. The same idea is often applied to the ovaries of the
female. l\Iany breeders believe that alternate heat periods are produc-
tive of opposite sexes. The same idea is sometimes applied to service at
early and late portions of the heat period. Such theories as these have
easily been proved false. One of the most weird superstitions states that
"if a man has a sty on his eye, he concludes his aunt is pregnant. If the
sty is on the upper eyelid the offspring will be a boy;on the lower a girl."
Any theory for sex determination must, from the very nature of the case,
be correct half the time, the offspring, of necessity, being either male or
female, in most species these occurring in equal numbers.
Sex Ratios!-:::ln most of the higher species the two sexes are born
in approxi~ately equal numbers, whereas in some of the lower species
there are great differences lii-r;:~mbers of the two sexes. Since in some
of the multiparou~pecies many embryos die and are absorbed during
gestation, tl1e"actual number born is neither a correct record of fecundity
nor is the sex ratio at birth necessarily the same as at fertilization if
there is a differential survival rate between the sexes. Both Hammond
and Parkes have shown that in swine there is a preponderance of males
soon after fertilization, but in this species the males are more liable to
die during embryonic deveIQP.meEt, so that at time of bi!th they exist
in approxImately equal numbers, and Parkes has reported the same
phenomenon in mice. Other evidence indicates that in mammals,
433
generally, there is a heavier fetal de~ toll on the male than on the
female embryos. For these reasons tJ}Mex ratio at fertilization is some-
( times referred to as the primary sex ratio, that at bIrth as the secon4;;Y~
h;thi;discussion, sex-;:atio means ttlat eXIsting atl)irth.
Sex ratlblnlle expressed as the number of males per 100 females or as
the percentage of males of the whole number born. For statistical
reasons the latter figure is preferred.
Table 26 gives some sex ratios that have been found in some of the
higher animals.
"-
TABLE 26.-SEX RA'rIOS IN FARM MAMMALS AND MAN*
Number of Percentage of males Number of

Animal
births studied in all births studies
Horse ........................ . 1,249,337 48.9-49.9 6

Mule ......................... . 1,416 44.3 1
Cattle ........................ . 155,579 49.4-51.5 3
Sheep ........................ . 91,640 49.5 1
Swine ........................ . 74,639 50.6-52.3 3
Goat ......................... . 3,000 50.1 1
Man ......................... . 50.7-51. 7 Several-
* Adapted from LUSH, J. L., "Animal Breeding Plans," p. 315, Collegiate Press, IIlC., of Iowa State
College, Ames, Iowa, 1937. (By permission of author and publisher.)
,
If sex is determined by the chromosomes and genes, as seems most
: probable, and the heteJ:ogametic sperinat~ (in rtmmnTals) are produced
' in approximately. equal .numb~r~, then differehces in fertilization rate
II must be due to dIfferentIal actlvl~y of the two types of spermatozo~or
i to differential selectivity by the ova. If there is a differential fertilizing
rate between the male- and f~male-'Conditioning sperm but approximately
equal numbers of the sexes at birth, then there must also be a differential
mortality rate between the two types of embryos, with the male embryo
in mammals apparently having a higher rate of mortality.
Chromosomes and Sex Determ~la melanogaster,
the common fruit fly, has been more thoroughly studied, from the stand-
point of inheritance, than any other species, and the study has yielded a
rich harvest. Along with the knowledge gained regarding behavior of
genes and chromosomes in transmission from parent to offspring has come
definite knowledge regarding sex and what determines it. There are two
sources of such information, one cytological and the other the behavior of
sex-linked factors. Cytology has established the fact that the normal
number of pairs of chromosomes in D. melanogaster is four. Of these,
three pairs are autosomes and one pair sex chromosomes, and although
'\.
\
SEX DETERMINATION 435
hoth sexes have autosomes that are morphologically identical, there is a

characteristic difference between male and female regarding the sex
chromosomes. The female has two so-called X chromosomes, whereas
the male has one X chromosome and one Y chromosome. Figure 127
should make this clear.
After reduction has taken place, as shown in the diagram, there can
be but one sort of ovum l"roduced; i.e., every ovum will contain one
member of each pair of chromosomes, and each one will, therefore, con-
tain one X chromosome. In the male, after reduction, there will be
two kinds of spermatozoa, both of which are similar as to autosomes
but different as to sex chromosomes, one having an X chromosome and
FEMALE MALE
~ Wfu
\:lD \Jj/
. I
; C!}j0fro,m&ctiMl!fj @
Germ cells
FIG. 127.-Diagram of nuclei of male and female germ cells in Drosophila, showing the
kind of ova and kinds of spermatozoa produced.
the other a Y chromosome. Presumably, these kinds of spermatozoa

are produced in equal numbers. Starting with an ovum containing an
X chro..."llosome and adding a spermatozoon also containing an X chro-
mosome will give an individual with two X chromosomes, or a female.
Starting with an ovum containing an X chromosome and adding a
spermatozoon containing a Y chromosome will give an individual con-
taining both an X and a Y chromosome, or a male. Cytological evidence,
therefore, proves beyond a reasonable doubt that, in Drosophila, sex
is determined at time of fertilization by the kind of spermatozoon that
fertilizes the ovum, for the sexes invariably show the differences just
outlined.
Nondisjunction in Drosophila. n aberrant typ~__~___re_Q.1lQ.t~
Drosophila, occurring very rarely, len s support to the chromosomal
interpretation of sex determinatio!!,) This is the feature of nondis-
junction, which has been worked out by Bridges. This investigator
has shown that occasionally reduction in the germ cells takes place in
f;uch a way that after synapsis the two sex chromosomes of a female
do not disjoin but both of them pass to one pole of the cell, giving rise
436 BRBEDI},;G AND IlvIPIWVEMENl' OF FARJl ANIMALS
°
to egg cells of the constitution XX and instead of two normal egg cells,
each containing one X chromosome. The relations holding may be
made more clear by means of introducing a sex-linked character. If,
for instance, a white-eyed female (rXrX) is mated to a red-eyed male
(RXY), all the normal daughters will be red-eyed and the normal males
white-eyed.
I Eggs
Sperms I
rX rX
RX RXrX RXrX
---
Y 'rXY rXY
If, however, in a white-eyed female, nondisjunction occurs, individuals

of the four types indicated in the diagram belov{ should be produced.
I Eggs
Sperms I
RX
-~------
Y
rXrX
Il~'
I rXrXY
x,t I
!
0
RX 0
YO
As a matter of fact, the RXr Xr X individuals usually die, as do alsb

the YO individuals, the former, as Morgan suggests, because of too
many X's and the latter presumably because they lack an X chromosome.
In other words, if the genetic constitution is thrown too far out of bal-
ance, the animal cannot survive. The other two individuals rXrXY,
a white-eyed daughter, and RXO, a red-eyed male, are the exceptions
to the normal expectancy.
Breeding results from such individuals in F2 (which the student can
diagram) give results that would be expected from individuals of such
chromosomal make-up, and the point to be stressed here is that, in either
F 1 or F 2 individuals, two X chromosomes result in producin~female
and one X chromosome a male 1 ----
These results not only furnish very strong proof of the chromosome theory of
sex, but serve also to show how a knowledge of the actual mechanism involved
leads to the discovery of how a change in the mechanism gives a new output.
1 MORGAN, T. H., "The Physical Basis of Heredity," p. 203, J. B. Lippincott Com-
pany, Philadelphia, 1919.
SEX DETERll;IlNATIOX 437
The conclusion that females behaving in this manner must contain a Y chromo~
some was confirmed by the cytological demonstration that showed in them two
X's and a Y.
The distinguishing feature of the XY type of sex inheritance, which

prevails in mammals, is that the female is homogametic and the male
heterogametic for the sex chromosome. It was thought at one time
that the Y chromosome was simply a neutral and more or less useless
mate for the X chromosome. This, however, has been disproved. Only
one dominant gene has been demonstrated in the Y chromosome of
Drosophila, but the fact that all males of the genetic constitution XO,
i.e., lacking a Y chromosome, are sterile leads one to the conclusion that
something in the Y chromosome is essenti&! for fertility-in other words.
:w r normal maleness. }t is known that in sQme species certain char-
acteristics are passed along always and onl from father to son, iwiica.tjug
that their eterminers must be in the Y chromosollle.
The Theory of Genic Balance.-In several species of animals, there
have been found individuals that partook partly of the characteristics
of both sexes, i.e., were in fact partly male and partly female. This
was difficult to explain on the basis cited above: that one X results in
producihg a male and two X's a female. Goldschmidt first studied
these so-called "intersexes" in the gypsy moth Lymantria dispar. He
found that in the F2 of a cross between Japanese males and European
females half the daughters were intersexes. Goldschmidt after an
extended genetic study concluded that sex was due to a balance between
male- and female-producing genes in the sex chromosomes. 1
These and other crosses leave no doubt that the tendency to form intersexes
is inherited in a regular manner. In moths, in general, the males are XX, the
females XY; and within pure races this sex chromosome constitution is decisive.
In crosses between races which produce intersexes, however, two further variables
have to be taken into account; one is a factor for maleness in the sex chromosome
which has different strengths, that is, presumably occurs in different allelic forms
in different races; the other is an influence tending in the opposite direction toward
• femaleness; also varying in different races, and this seems to be inherited through
. the cytoplasm of the eggs, that is, all the eggs of an individual are alike with
respect to the strength of this tendency. The sexual condition of the offspring
of crosses is decided by the relative strengths or balance existing between these
two opposed tendencies. This interpretation is supported by extensive evidence
from many interracial crosses.
Goldschmidt and his coworkers have also proposed a mechanism by which
these sex genes influence the development of the sexual characters. The essential
1 SINNOTT, EDMUND W., and DUNN, L. C., "Principles of Genetics," 3d ed., p. 2tH.
:\[cGraw-Hill Book Company, Inc., New York, 1939.
assumptions here, as found first in the worm Bonellia, established by a careful

study of the embryological development of Lymantria and later extended to
Drosophila, are that intersexes in these forms begin their development as males
(or females) and develop as such up to a certain critical point, the so-called
"turning point," after which their development is of the opposite sexual type.
The mh:ture of male and female parts is thus explained as due to the supersession
of one type of sex tendency by the other. Sex development is conceived as
essentially a competition between opposed tendencies in which the race is eventu-
ally won by that type of process (either male or female) which proceeds most
rapidly at the critical period of determination of each organ or part involved.
The discovery of certain individuals in Drosophila cultures, which in

their sex characteristics were between (intersexes) or beyond (supersexes)
the normal male or female limits, has yielded further evidence that
sex manifestations are due to a certain balance between the genetic
content of the automosomes and that of the sex chromosomes. Bridges
has shown that the normal balance between two sets of autosomes (2A)
and two sex chromosomes (2X) produces a female, and the balance
between 2A and IX produces a male. That is to say, the autosomes
have genes tending to produce maleness, which they do when working
in conjunction with the genes found in one X chromosome. When,
however, there are 2X chromosomes present working in conjunction with
2A, the result is a female; i.e., the genes in the X chromosomes are
preponderately female-producing. A graded series of sexes have been
found or produced experimentally as indicated in Table 27.
TABLE 27.-RELATION OF SEX TO CHROMOSO:\fES IN Drosophila rnelanogaster
Sex type Formula X = 100 A = 80 I Sex index

Superfemale .................. 2N. 3X:2A 300 160 1.88
4N 4X:4A 400 320 1.25
Female ....................... 3N. 3X:3A 300 240 1.25
2N .2X:2A 200 160 1.25
IN . lX:IA 100 80 1.25
Intersex ...................... 4N 3X:4A 300 320 0.94
3N. 2X:3A 200 240 0.83
Male ......................... 2N. 1X:2A 100 160 0.63
4N. 2X:4A 200 320 0.63
Supermale .................... 3N. 1X:3A 100 240 0.42
It will be seen from this table that a normal balance between auto-
somes and sex chromosomes 3A and 3X or 2A and 2X results in a female,
whereas a preponderance of the X ingredient results in a superfemale
and a preponderance of the autosomal ingredient results m 'varying
degrees of manifestations of maleness. 1
If we represent the net effectiveness of the female tendency genes in the X by

100, then we should represent the net male effectiveness of a set of autosomes by
some lower number, approximately 80. In the 2X2A individual the ratio of
female effectiveness to male effectiveness is 200: 160, or 1.25: 1. On this formula-
tion an index of 1.25 corresponds to a normal female. As is indicated in Table 27,
this same index of 1.25 holds for 3N individuals, for the ratio is 300: 240, or 1.25: 1.
The identity of sex indices for the 3N and 2N forms corresponds to the observation
Male. female Super- female
5upermale Intersex Triploid (Female)

FIG. 128.-Effect on sex of the balance between X chromosomes (solid) and autosomes
(outlined) in Drosophila. (Rearranged from Sinnott and Dunn, Principles of Genetics.)
that there seems to be no sexual difference between them. In the X2A individual
the ratio is 100: 160, or the sex-index is 0.63, that of the standard male. The
sex-index of the 2X3A. intersex is 0.83, intermediate between the 1.25 of female-
ness and the 0.63 of maleness.
New tests of the general validity of the genic balance formulations as a ratio
between opposed sets of genes came from several sources. First, both myself
[Bridges, 1925] and L. V. Morgan found tetraploid individuals. We observed
that these 4X4A individuals were completely normal and fertile females, practi-
cally indistinguishable somatically from 3N females. That the sex was not
1 ALLEN, E., DANFORTH, C. H., and DOISY, E. A., "Sex and Internal Secretions,"
2d ed., pp. 40.-:42, The Williams & Wilkins Company, Baltimore, 1939. (Paper by
C. B. Bridges.)
changeAl when the chromosomes were doubled is in the best conformity with the
ratio formulation and is difficult to account for on the arithmetical system
(Schrader and Sturtevant 1923). '
A second and more striking test came from the work with the haploid or one
N individual in Drosophila. It was predicted that this type should be female in
sex since it must have the same ratio of X to A as the 2N, 3N, and 4"Y females.
No such haploids were then known for Drosophila, and this expectation of female-
ness was in direct opposition to the fact that all of the many haploid animals
known up to that time were males without exception. By a special technique a
few mosaic individuals, one in several thousands of normals, were found in a
special strain of flies. These rare mosaics have parts of their bodies haploid as
shown by autosomal genes and by the smaller cell size. In some of these mosaics
the haploid regions included body parts which enabled the sex to be diagnosed.
They were actually female as expected (Bridges 1925). Since then at least three
other workers have found Drosophila mosaics in which haploidy and femalenef's
seem to be associated.
In unpublished results Sturtevant has found that the 2X4A is a male and that
3X4A individuals are intersexes, both fitting into place in the series.
The observation of IN = ~ in Drosophila no longer stands alone. Frank-
hauser (1937) secured haploid embryos of the newt, one of which lived nearly
through metamorphosis and showed definite female-type gonads, conforming to
expectation from the 2X2A formula for the normal female.
In applying the concept of genic balance to sex Bridges (1922) had made the
postulate that female producers predominate in the X and are scattered through-
out it at random in the same sense as are the producers of eye color or any other
character. Experimental proof of these assumptions were obtained by Dob-
zhansky and Schultz (1934) through the use of fragments of X (duplicationst of
various lengths and taken from various regions of the X, and also by deficiencies
or losses of sections of the X.
For a sensitive test of the sex effect of these extra pieces or losses of sections
of X the grade of the intersexes was used, marked off into six classes from extreme
male-like to extreme female-like types. Thus, a 3N female all of whose X's
carried the recessive yellow was crossed to a male whose X was also yellow, but
which was hyperploid for a non-yellow fragment extending from the left end of
the map to beyond the locus prune. Half the intersexes would be without this
duplication and would be yellow while the other half would carry it and would
have the yellow" covered" by the wild-type allel in the fragment.
The experimental findings were that each of the dozen duplicatio_ns tried
causes a shift in the female direction, none in the male direction, as compared
with the siblings without it. Moreover, the longer the piece of X used the greater
was the shift toward femaleness-Bven resulting in intersexes fertile as females!
On the other hand, one deficiency shifted the intersexes toward maleness, while
another, involving mostly the "inert" region at bobbed, was neutral in effect.
Since these duplications covered all of the X chromosome, the conclusion follows
tha t no part except the "inert" region is free of female producers in excess of
male producers.
Sex-producers and Sex-differentiators.-From the numerous examples of

tIlutant sex-characters in Drosophila, Zea, and other forms, it is clear that a fairly
large proportion of the genes must be concerned with the production of tho
llormal sex organs in their normal grade of development. All these genes whose
action is part of the normal sex-development, which would have another outcome
if mutant alleles were present instead of these normal genes, one may call" sex-
producers." This includes the so-called "sex-modifiers," where the degree of
effect is perhaps less than some of the other genes or the effect is observable
only or predominantly in one or the other sex.
Any of the above producers of sex may become a "differentiator" of sex in
tIlutant form as contrasted to the standard form in the same cross. Some may
require special circumstances such as the sensitive condition in 2X3A intersexes or
the cooperation of other genes to bring out their differential action. Some allelo-
morphic differences are so slight in effectiveness as only to change the individual
from female to male or from male to female as shown in fishes, maize, and
Habrobracon.
The difference between a producer and differentiator of sex may be illustrated
by an analogy. As equal weights are added to the pan of a balance the beam
finally tips. Let us say ten weights left it un tipped and after the eleventh was
added it tipped. But this eleventh weight has no more intrinsic significance in
the tipping than each of the weights added before. If anyone of them had been
omitted the beam would not have tipped when this one was added. All together
produce the result. The last one is the differentia tor of the position of the beam.
_/:..(~ ..,..J>;.;2"_.J / l X /~ )
/ Genes and Sex Determination.-We have seen that sex is not deter-
mined by any specific whole chromosome, but rather that it appears
to be due to certain balances between genes located in both the auto-
.somes and the sex chromosomes. Recent work by Patterson and his
students at the University of Texas has seemed to indicate the localization
of the genes in the X chromosomes, which must be present in duplicate
to produce a female or in a single dose to produce a male. This \vas
accomplished by treating flies with X rays, with the result that the X
chromosome was broken into various-sized fragments. Thus, stocks of
flies could be made up containing one normal whole X chromosome and
various-sized fragments of another X chromosome. This procedure
indicated that a certain small region of the X chromosome between the
genes for pleat~d and garn~t was the essential portion oper.1tive in throw-
ing the balance of all the genes toward the male or the female side. This
is shown diagrammatically in Fig. 129.
It thus appears that the genes or gene in the X chromosome which
reacts with autosomal genes in determining sex jl,re located in this sma!!
region of the X chromosome. 1
1 Ibid., pp. 25-28.

442 BREEDING AND IMPROVEMENT OF FARlv! ANIMALS
~x-mOSaiCS and G~androI!!,Q!Fhs.-An interesting and important phase of

sex-determination is involved in the production of mosaic individuals which are
of both sexes at once, but in different body-regions, and with a sharp line of
dem'3,rK"ation which at the sameti~rresponds"1o a difference in genetic or
chromosomal constitution. A comprehensive survey of gynandromorphs in all
types of animals has been given by Morgan and Bridges (1919) and a special
review of the Drosophila cases by L. V. Morgan (1929).
Doncaster (1914) in studying sections of fertilized eggs of the moth Abraxas
observed an occasional egg with two separate maturation spindles and two female
pronuclei each about to be fertilized by a separate sperm. He suggested that
these pronuclei could differ as to whether they carried a sex-chromosome or not'
(female heterogametic) and consequently the double fertilization process could
give rise to gynandromorphs. Definite proof of the occurrence of such dizygotic
gynandromorphs came in Drosophila with the discovery of sex-mosaics which
were at the same time mosaics for autosomal marker characters. Occasionally
o
~:L+~+~.O+ + +
o
Normal X Normalll
o o
NormalY frogmentafX FrogmentofX Fr.gmenlsofX Fr.gmen!ofX Fr~9mentafX
Chromosome Chromosome Chromosome Chromosome Chromosome Chromosome Chromosome Chromosome
Femo:le MOlle Male Male Male female female
FIG. 129.-Diagram of section of X chromosome determining sex. (After Paterson.)
such dizygotics were mosaics for autosomal characters in two different autosomes,
though it is rare to make a cross that would show this. The first of the clear
dizygotic gynandromorphs was one found by Bridges (Morgan, Bridges, and
Sturtevant, 1925) in a backcross involving two second-chromosome recessives,
vestigial and speck. The right side of the body was mostly female and showed
speck; the left side was mostly male and showed vestigial. The explanation is
that the right side had come from an X-egg carrying speck, fertilized by an
X-sperm carrying vestigial and speck. The left side came from an X-egg carrying
vestigial, fertilized by a Y-sperm carrying vestigial and speck. Some 20 dizygotic
mosaics have been found in Drosophila, though the majority were alike in sex
on the two sides.
In Drosophila there is another type of gynandromorph, of which several hundred
have been found. These start as XX zygotes and give rise to XO male tissues
on one side or part of the body. The male tissue results from the loss ~a
tion" of one of the two original X chromosomes, the autosomes all remaining
diploid. If the elimination takes place at the first cleavage, Mit of the body
becomes male; if at the second, only about a quarter. _
A ~triking example of such bilateral gynandromorph was one found by A. M.
Bro,vn and included in the general account of gynandromorphs by Morgan and
Bridges (1919). This gynandromorph occurred in a cross between an eosin
miniature male and a female heterozygous for eosin and miniature. The male
right side showed miniature and eosin, while the left side was female and wild-
type. Presumably the gynandromorph started from an eosin miniature/wild

zygote and at the first cleavage division one of the daughter chromosomes of the
maternal wild-type X chromosome lagged upon the spindle and was lost. Hence
one cell retained the original constitution and developed as female while the other
was XO in constitution and developed as eosin miniature male.
In case the male parts of an elimination gynandromorph include the abdomen,
such an individual can be mated as a male. But he is invariably sterile, as he
should be from the XO constitution of the male parts.
That an X-chromosome only, and not the autosomes, is involved in the elimi-
nation was shown by a test proposed by Muller, namely that a cross be made
involving the sex-linked characters yellow and white, as markers for the X, and
ebony or some other recessive as a marker for the autosomes. Gynandromorphs
in such crosses showed sex-linked characters in the male parts but never the
autosomal ones-except those due to double fertilization, as noted above.
Morgan and Bridges (1919) studied carefully about 100 gynandromorphs and
found that the maternal X was eliminated about as frequently as the paternal X.
In experiments in which all fiies were counted 40 gynandromorphs occurred in
88,000 fiies, or 1 in 2,200. The line of separation was usually sharp with respect
to both sex and whatever sex-linked characters showed in the different areas. Of
the dozens of sex-linked characters seen in the gynandromorphs all were autono-
mous in development locally except vermilion, and, to a slight extent Bar (Sturte-
vant 1920).
A third type of sex-mosaic was found by Bridges (1925) in crosses in which
the mother was heterozygous for the sex-linked dominant Minute-n and the
father showed one or more sex-linked recessives, such as yellow, eosin and singed.
The Minute-n daughters frequently (10 to 40 per cent) showed one or more spots
that were not Minute, but were yellow eosin and singed. Some of these spots
could be classified as male by including eosin (lighter in (') ) or the fore-leg with its
sex comb. The explanation offered by Bridges was that Minute-n induced fre-
quent elimination of the Mn-bearing X in the somatic calls. This occurred rela-
tively late in imaginal disc development, thus accounting for the small size of the
spots. But Stern (1936) has studied their origin thoroughly and shows that they
are due to crossing-over in the imaginal cells. This crossing-over leads auto-
matically to somatic segregation, since the equational separation of spindle-
attachments draws the crossover terminal regions into separate cells. Stern
finds that the homozygous Mn spots then die, leaving the homozygous y sn spots
which were observed by Bridges. Stern finds that most spots are XX and female.
Male spots, such as those detected by Bridges, do occur, principally when the
Mn mother carries an extra Y chromosome. Then somatic segregation gives XY
male spots, instead of XO spots as Bridges supposed.
Endocrine Function in Sex Differentiation.-In the earliest forms of

life, there probably were no distinct sexes. These organisms reproduced
only by fission, or budding, where the division of the parent was very
unequal. This, of necessity, meant that the offspring would be very simi-
lar to their parents, for a considerable portion of the parent went to form
each new offspring, although an occasional mutation might happen.
Eventually the division of labor among the cells making up organisms
meant that certain cells were developed for the particular purpose of
reproduction. At length, there apparently came to be formed different
types of reproductive cells, which had to unite to produce a new organism.
At first these cells were produced by one parent (hermaphrodite), but
eventually there came a time when the union of dissimilar cells, particu-
larly as far as cytoplasm is concerned, produced by two separate indi-
viduals was necessary to form the beginning of a new individual, and
finally the two sexes themselves came to be quite different, in part because
of the constitution that allowed one to produce eggs and the other
spermatozoa. Sexuality has therefore come about through a long line
of evolutionary development, and the advent of unisexuality probably
greatly speeded up the evolutionary process because of the gene differ-
ences brought together in the new organism, and the fact that the new
organism in turn was destined to give a random, sample half of its
diversely acquired genetic material to each of its offspring.
Many plants and animals are still of one sex in one portion of the body
and the other sex in another portion. In some the production of male.
or female gametes alternates or varies in time owing probably to environ-
mental influences. In higher mammals, of course, the two sexes are
recognizably different both in regard to their sexual products (eggs
and spermatozoa) and their somatic differentiation. These differences
are basically genetic, i.e., owing to differences in gene complements,
but superimposed on the genic differences are those lately discovered to
be of an endocrine nature.
It is probable that in Drosophila and many other species of lower
animals the same mechanism, i.e., the genes, underlies both these phe-
nomena. If, at the first cell division in a female Drosophila zygote,
one of the resulting daughter cells gets only one X chromosome, a
gynandromorph half male and half female results. If the discrepancy
occurs at the second cell division, approximately one-quarter of the
resulting organism will show male characteristics.
In the higher vertebrates, however, the evidence indicates that,
although the genic balance usually determines the sex of the individual,
this is subject to another, hormonal, control. Lillie has shown that the
freemartin (abnormal sexual female) in cattle arises where the choria
of unlike sexed twins become fused and the blood vessels anastomosed
so that the blood of the two embryos becomes interchangeable. Lillie
has shown that this is due to the influence of a hormone secreted by the
male gonad which, when carried by the blood into the female calf,
replaces the development of normal ovaries and, to some extent, the

female accessory organs of reproduction with testes and male ducts.
, The female is ge:p.etically a female and appears phenotypically as a
female, but the male sex hormone in its blood during early embryonie
life prevents the development of a normal set of female genital organs
and substitutes an imperfect male set. Hughes has reported the same
phenomenon in the pig. Specifically, the medullary portion of the
gonad forms sex cords instead of the normal medullary cords of an
ovary.
Experimentally, this has been accomplished in several of the lower
animals by surgically joining embryos, by gonad transplantations, and
by the injection of sex hormones into embryos. The female sex hormone
injected into an embryonic genetic male chick causes the left testis to
become a flattened, ovarylike, ovotestis, and the right testis is sometimes
affected and the oviducts persistent. Injection of the male sex hormone
into an embryonic, genetic female causes the cortex of the left ovary
to thin out and the medulla to hypertrophy and transform into testicular
tissue at the hilus.
The first set of sex cords that pass into the gonad form the medulla,
and this later forms the sex cords of the male. The medulla is present in
the ovary, and later there is added the cortex. The ovary, therefore,
consists of two tissues of different physiological capacities. In the early
stages, the male gonad also has a cortex, but this does not persist. The
seminal cords of the medulla are already established at the time of mor-
phological sexual differentiation, and the male gonad, therefore, starts
to function as an endocrine gland before the female gonad to which the
cortical layer is added somewhat later.
Whether male or female sex organs begin to develop in the early
embryonic stages is conditioned by the chromosomal make-up. There
is apparently no real sex differentiation, however, until after the primor-
dial sex tissue has been laid down in the embryo. The male organs of
reproduction have their homologues (similarity of origin, not function)
in the female and vice versa (penis and clitoris or uterus and uterus
masculinus, etc.); in other words, the primordia for both sexes are
present, the make-up of the chromosomes being the determining factor
as to which will begin to develop, the full development in turn being
conditioned by hormones secreted by testicles and ovaries.
Sex is, therefore, determined by the genic balance, and actual differ-
entiation is brought about by the sex hormones. That the chromosomel:l
or genes are not of themselves able to bring about sex differentiation
in the higher animals is well illustrated by the results following castration
(or spaying). If castration of mammals is performed soon after birth,
the two differently sex-determined classes, male and female, tend to

remain the same; i.e., sex differentiation does not ensue. The castrates
have the sex genes XX or XY in all the cells of their bodies, but male or
female secondary sex characters do not develop in the absence of the
gonad hormones (see Chaps. IV and V).
Sex Reversal.-Sex reversal has been produced experimentally in
several lower forms of animals. In the toad, for instance, there is a
rudimentary organ, composed of rounded cells that resemble egg cells,
located at the anterior end of the testicle. This is known as Bidder' 8
organ. If the testicles of a young male toad are removed, the organ of
Bidder will develop after 2 or 3 years into an ovary that secretes viable
eggs. In other words, castration in this species throws the sex clear
over to its opposite. Experimental work has also demonstrated that,
in poultry, removal of the functional left ovary from a young chick
causes the bird to develop the sex plumage and characters of the male,
though it eventually returns to its original character. This is apparently
due to the fact that the rudimentary right" ovary," in the absence of
the suppressing effect of the ovarian hormone, develops into a testislike
organ with seminiferous tubules in which some investigators report
normal spermatogenesis. Occasionally, reports are circulated concerning
hens that have changed into roosters and actually sired offspring, and
the case reported by Crew is apparently an authentic case of sex reversal
in the fowl.
In the higher mammals, sex is established at time of fertilization,
and sexual differentiation begins fairly early in embryonic development.
The young embryo has the primordia for either sex, and its genetic make-
up determines which one will develop. The mechanism, genic, hormonal,
environmental, sometimes fails to function perfectly and pseudoher-
maphrodites are produced. These are of various sorts; i.e., both the
male and the female embryonic primordia have developed to varying
degrees from slight variations from one sex or the other to practically
equal development of both sexes, but perfect in neither. Pseudo-
hermaphrodites are more numerous in goats and pigs than other farm
animals, and a few cases are known in man where one gonad developed
into a testis found in the scrotum, the other into an abdominally located
ovary. The external genitalia are also abnormal. Whether these
aberrations are genetic or developmental is generally difficult to deter-
mine, but many no doubt do have a genetic basis.
In higher mammals once normal sexual development has occurred,
complete sex reversal would be extremely unlikely. In the human,
however, the secondary sexual characters of either sex sometimes appear
in members of the opposite sex. Adrenal tumors have been shown to be
responsible for excessive growth of facial hair and the appearance of

other masculine characters in the human female. Males subjected to
certain estrogenic substances or males suffering from certain types of
carcinoma may exhibit marked mammary development.
In the chicken following normal sexual development and egg produc-
tion in genetic females, the atrophy of the functional ovary and the
development of the ordinarily rudimentary right gonad into a sperm-
producing organ is not uncommon. When this phenomenon occurs, the
secondary sexual characters of the female regress and there is develop-
ment of the comb and spurs, and crowing, treading, and other evidences
of male behavior are common. A few cases have been reported in which
such individuals have apparently sired chicks.
Numerous attempts to alter the sex of developing embryos by hormone
treatments have been made. It has been possible to alter the morphology
of the sexual organs, but complete experimental sex reversal is yet to be
achieved.
Conditions Influencing Sex Ratio.-With sex evidently determined by
the interaction of genes in the sex chromo~.9Ig~s and others in the auto-
somes and\~ith'-olle- se-;ZOfl,ne other-'(varying in different species) being
hete~ogamet'i~e would expe~t generally to find an equality of numbers
between the sexes provided there is equality of numbers and viability
of the heteroganietic sex -cells, no selectivity between germ cells of the
two types, and IlO differential ~embryonic mortality between the sexes.
K evertheless, the; sex ratio is not always 1: 1. --Variations from the
---~---------, - - -.'--'----- --~
numerical equality. of the sexes may conceivably be due either to genetic
or environmental causes. ./' _,
Gen~ti~ -;~~tioris have been found in Drosophila. One of these
behaves a·sa: sex~EnI~ed and sex-limited-to-male gene that, when present
in a male, either prevents "the Y-bearing spermatozoa from being formed
i.n numerical equality ~r IQwers their ability to reach and fertilize the(i
egg, for the offspring. bf such ,males have 96 female: 4 male offspring.
Even in the homozygous form, it apparently produces no ill effect in the
female. l..Various lethal and semilethal sex-linked mutants are known
in Drosophila, some 'of which kill only the females, because the Y chro-
mosome carries the normal allele. £ethals in the autosomes sometimes
give r~s,~Jg peculiar sex ratios . ~riteiSpeC1nc- crosses, no mares
.:.:..fn
or'a greatly reducediiUinl:)erls born; in others, the reduction comes in
the females. '
Environmental conditions are also known that have an influence on
the sex ratio. Seiler has shown that temperature affects the extrusion or
retention of the X chromosome in a moth (Tael~p;;'i~ tubulosa) in which
the female is hete"rogametic. Season of breeding h~s been reported by
with ovarian activity. The idea that one ovary produces ova which give
rise to males and the other to females has been proposed. The obvious
fallacy that mammalian sperm rather than ova control sex negates this
belief. The possible role of the female in the determination of sex
persists because of the not uncommon observation that some individual
females have offspring of rwedominantly one sex.
It is difficult to trace the exact-origin of the hypothesis that the reaction'
of the female genitalia, usually expressed in pH, influences sex. There
are numerous variations of the general idea that an alkaline reaction
prod..w!ed by ..douchiog ;will ~t in a prepondernace--2Cii1a~s- aria an
aci .on in the production of females. Although some data nave
apparently in I, sex ratios have deviated some-
what from the normal, the consensus of opinion seems to be that sex
cannot yet be controlled by such techniques. It is well known that the
secretions in the various parts of the female genital organs are well
buffered and that douching the vagina, for example, will have little or no
effect on the remainder of the tract. For this reason, some investigations
have been made on the alteration of the pH of the semen. To the
authors' kno,yledge, the exposure or storage of sperm at different pH
levels has no significant effect on the sex ratio.
Because of the presence or absence of the X or Y chromosomes, it
seems logical that there are physical or chemical differences in the sperm.
It has been proposed that ;'s-ep~~onOfSperm on tIle basisuoCSizemight
result in a separation of those of different sex potential. This idea has
some merit, but it is obvious that a separatIon of the sperm cells on this
basis has many practical limitations. It should be possible to accomplish
this by centrifugation, and several mathematical calculations for such a
separation have been made. It is entirely possible that centrifugation
under highly controlled conditions might be effective. What effect this
would have upon fertility and whether sex could be effectively controlled
is hypothetical.
Still another, but not the last, approach is based upon the electrical
charge of the sperm of different genetic make-up. It has bemnuggested -
tliatsPe~ different charges could be separated by passing an electric
current through a suspension of sperm. It has been reported that by
sampling a charged suspension at the anode or c~regions that the
sex ratio has subsequently been altered. ---:Although this idea is a relatively
old one, it has not yet been sufficiently explored to warrant the con-
clusion that sex can be controlled by such procedure. ,f
While effective sex control is far from realization in 1950, it seems

possible that advances in the physical and biological sciences will even-
tually make possible the pr6duction of males or females at will, or will at
'.,"'.
least alter the normal sex ratio. From the standpoint of pracQcal live-
stock production, this would constitute a real advance in technology.
Summary.-The general role of the chromosomes in the determination
of sex was substantiated in 1905. The two principal additions to the
chromosomal theory since that time have consisted of the theory of
"gen~£J:l~~' between the autosomes and sex chromosomes and the
more recent localization of the effective materials in the sex (X) chromo",
some to a known small regioll of this chromosome. We have latelY come
to understa:rrdll:lore fully the part played by the endocrine secretions
in sex differentiation among the higher animals as \yell as som~-;;tthe
conditions that affect the sex ratio and the possibilities of affecting the
latter by selection. In spite of many attempts to solve it, the problem
of controlling sex remains unsolved up to the present.
References
Books!
ALI,EN, E., DANFORTH, C. H., and DOISY, E. A. 1939. "Sex and Internal Secre-
tions," The Williams & Wilkins Company, Baltimore.
CREW, F. E. A. 1927. "The Genetics of Sexuality in Animals," The Macmillan
Company, New York.
GOLDSCHMIDT, R. 1923. "The Mechanism and Physiology of Sex Determination,"
Doubleday & Company, Inc., New York.
SCHEINFELD, A. 1943. "Women and :Men," Harcourt, Brace and Company, Inc.,
New York.
CRAFT, W. A. 1938. The Sex Ratio in Mules and Other Hybrid Mammals, Quart.
Rev. Bioi., 13 :1940.
HENNING, W. L. 1939. Prenatal and Postnatal Sex Ratio in Sheep, reprint from
Jour. Agr: Res., Vol. 58, No.8.
JULL, M. A. 1924. The Relation of Antecedent Egg Production to the Sex Ratio
of the Domestic Fowl, Jour. Agr. Res., Vol. 28, No.3.
LILLIE, F. R. 1917. The Free-martin: A study of the Action of Sex Hormones in
, The Fetal Life of Cattle, Jour. Expt. Zool., 23 :371-452.
MCPHEE, H. C. 1927. The Swine Herdbook as a Source of Data for the Investiga-
tion of the Sex Ratio, etc., reprint from Jour. Agr. Res., Vol. 34, No.8.
SWETT, W. W., MATHEWS, C. A., and GRAVES, R. B. 1940. Early Recognition of
the Free-martin Condition, etc., Jour. Agr. Res., 61(8) :587-623.

SECTION IV
The Art of Breeding
CHAPTER XVIII
SYSTEMS OF BREEDING-UNRELATED ANIMALS
We have now completed three sections of this text on animal breeding.

The first section was designed to give the student an over-all view of the
history and status of breeding at the present time as well as an intro-
duction to the origin and relations of all living forms in terms of the proba-
ble physical basis of life and inheritance--the genes and chromosomes.
The second section dealt with the scientific and practical phases of
reproductive physiology in order to acquaint the student with the funda-
mental procedures in the production of a new generation and to indicate
the practical procedures that will best ensure a high degree of repro-
ductive efficiency.
The third section dealt with the workings of the genes and chromo-
somes in the perpetuation of similarities and the creation of new types
among related forms. Perhaps the student sometimes wondered why
he should learn all the details of mono- and dihybrid crossing, of dom-
inance and epistasis, of lethals, multiple alleles and multiple factors, of
sex-linked and -limited inheritance, of linkage, crossings over, chromo-
some mapping, interference and limitation of the linkage groups, and of
the various and sundry types of gene mutations and chromosomal
aberrations, for it is so difficult to find examples of the actual working of
these principles in the farm mammals. Knowledge of the basic principles,
nevertheless, should be valuable in acquainting the student or breeder
with the complexity of the problem with which he is dealing. He can
rest assured that all these principles are at work in the germ cells of the
animals in his herd or flock, although in most instances so many genes
are involved, in such complex interactions, and so few offspring are
produced that the result of the operation of anyone principle is generally
difficult, if not impossible, to measure.
Such knowledge should make the breeder approach his practical prob-
lem of creating better animal types in a less cocksure manner and with
a greater realization that he is dealing with a complicated mechanism
452
.,
SYS'1'E;HS OF BREEDING-UNRELATED ANIMALS 453
which demands as full a knowledge of details as it is possible to achieve.

In rebuttal, it might be argued that old-time breeders often achieved
great success in total ignorance of the basic principles. This must be
admitted, but we should be aware of the fact that less pUblicity has
probably been given to those old breeders' failures. Their task was to
create a few outstanding individuals, on which breeds could be built;
ours is to raise the average merit of all our animals, which can be accom-
plished only if we fully understand the natural principles underlying
reproduction and the mechanics involved in hereditary transmission.
This, the fourth and concluding section, will deal with the practical
matters involved in livestock improvement-selection of animals for
specific matings. Systems of breeding. and selection constitute the only
tools available to the breeder for animal improvement, since we cannot
-makeany -n~';- genes but only, starting from where we are, arrange what
we have or can buy into more desirable groupings.
Each breed of livestock, as now constituted, has as its genetic capital
the complete reservoir of genes as they exist in the living members of the
breed. Each breed has a certain number of good genes and a certain
number of bad ones. If the number of good genes can be increased at the
expense of the bad ones, the breed will progress. If the opposite result
eventuates, the breed will retrogress.
N early all breeds are debarred by their own rules from going outside
of their own limits for new genes. It cannot hope for any speedy progress
from the natural occurrence of beneficial gene mutations, and there is at
present no kno,vn way of creating helpful mutations through artificial
means. Breed progress depends, therefore, on the relative gains in
frequency of its desirable genes and possibly their recombination into
more desirable groupings. New breeds can also be created, if and when
that seems desirable or necessary, by combining characters, as they now
exist in separate breeds, through systems of crossbreeding and selection
followed by inbreeding.
An individual breeder, likewise, has as his genetic capital the genes
existing in his living animals. He, of course, can augment this supply
by purchase from other breeders. He usually does this through the
purchase of males.
Before the breeds were formed as such and their books closed, a group
of breeders in a region could and did go outside the not-too-distinct
confines of their own type to get some genes that they desired. In getting
these good genes, they also had to take some that they did not want.
Their task was, therefore, that of so matjng and selecting their animab
that they might retain the good and discard the bad genes. It was in
this fashion that our purebreds were formed.
454 BREEDING AND [llIPROVEMENT OF FARM ANIMALS
The pool of genes in any breed has, in general, seemed to be elastic

enough to permit the ideal to range between rather wide limits. At the
beginning of this century the commonly accepted type of hog was short-
legged, deep-bodied, and extremely rapid-fattening. Differential growth
rates exist between the various parts as well as tissues of the body. In
this" chuffy" type of hog the bone growth and muscle growth proceeded
rapidly to their rather low upper limits, whereas the third phase, that of
fat deposition, came into play early and had a high limit. The result
was an early-maturing, fine-boned animal with a high proportion of fat to
lean. This type of growth was highly favored also by the prevalent
practice of feeding largely on corn. The consumer demand gradually
shifted to a leaner type of animal. To meet this changed demand,
breeders changed their selection ideal as well as their system of feeding.
By the late twenties or early thirties, the selection ideal in hogs had
shifted to the other extreme-that of an upstanding, rangy, relatively
shallow-bodied, and lean type of animal. In these animals the bone-
and muscle-development phases continue for a long time and have a
high upper limit; the fat-deposition phase arrives late and has a low limit
as compared with hogs of the early 1900's.. Probably if a group of
breeders of II chuffy" hogs in 1900 had been shown a picture of the I ' race-
horse" type of 1930 'and told that their job was to create such a type,
their first reaction would have been, it can't be done. Due, however, to
the facts that the market demanded a change and that the kind of genes
necessary to produce such a change existed in the gene pool of all the
breeds, the change was brought about in a remarkably short time and,
in fact, the change went too far. Some physiological changes are very
complex in their nature; they undoubtedly depend on the actions and
interactions of a great many genes that probably act in large measure .. i
through the mechanism of the endocrine glands; and for their full expres-
sion they are dependent upon the environment, especially its nutritional
phases. Similarly, the pool of genes in our present-type hogs undoubt-
edly contains enough of the genes for" chuffy" type to permit the reverse
of the above-described change from "chuffy" to big type to be brought
about should it ever be desirable to do so.
Another familiar example is the smoothing up of the type of our dairy
breeds during the past 30 years. While average production has been
improved, a far greater change has ensued in type also during this time.
Some might and do argue that the change in production has been due
most of all to better feeding and management. This argument can
hardly be used to explain the change in type. Again it is evident that
the pool of genes in the various breeds contained a sufficient supply of
genes making for smoothness of topline, proper set of legs, body, and
SYSTEMS OF BREEDING-UNRELATED ANIMALS 455
udder capacity, and general quality so that conscious, directed selection

for these things could be successful.
In one of our dairy breeds the byword among breeders 30 to 40 years
ago was" the 30,000-lb.-of-milk cow." In practicing such selection, the
genes making for higher fat test were automatically discarded, owing to
the general negative correlation between high amount of milk and hiih
butterfat test. But again there was seemingly enough of a supply of
genes for higher test left in the general gene pool of the breed to make
possible the raising of the butterfat test of the breed by at least H of 1
per cent, although such a change might cost something in milk production.
One of our breeds of cattle has followed two divergent lines of devel-
opment, one for milk, the other for beef. The genes for beef are still
present in the former and those for milk in the latter, so that these two
strains could be made to mutually replace each other if there were any
point in doing so. Similarly, it would still be possible to make any beef
breed over into a milking breed and vice versa by means of selection
and systems of breeding.
With the actual formation of the breeds and closing of the books,
the door to the admission of new genes, except such as might occur
naturally through gene mutation, was closed. From that point on breed
progress became simply a question of manipulating gene frequencies.
The technique for doing this resides in selection. The breeds or
br~aers try to increase their stock of good genes by permitting those
animals which have the greater number of good genes to leave many off-
spring while restricting the number of offspring of those which have
relatively few good genes. The first task of a breeder becomes, therefore,;
that of devising methods for ascertaining the sort of genes which his I
animals possess. This calls for some sort of measurement and for record
keeping. . \
In addition to selection the breeder has the use of the tool of breeding
systems. 1 Some systems of breeding render the germ plasm more
heterozygous, others make it more homozygous. Both selection and
systems of breeding can be used to influence gene frequencies, and they
are the only means available for doing so. Systems of breeding and
selection are the most important problems with which the breeder must
deal. They constitute the art of breeding.
It should be understood at the very beginning that there is nothing
about any system of breeding which guarantees it to be a success, or,
on the other hand, that foredooms it to failure. Popular articles some-
1 The term systems of breeding is here being used in a somewhat restricted sense.
Constant selection is a breeding system, as are also assortative and disassortative

matings.
times portray the systems of breeding used by some breeder or the type
of matings used to beget some champion with the implication that similar
methods should be adopted by all and sundry. However,. the fact that
some successful breeder used linebreeding has no necessary application
to any other herd or flock.
Similarly, a crossbre~.ding or an inbreeding experiment that turns out
well (or ill) cannot be liken as a rigid standard of what will happen the
next time or in some other group or class of animals.
The success or failure of any system depends entirely on the genes
present in tlie animals concerned, plus of course the breeder's ability to
mate and select intelligently. '
All systems of breeding can be grouped under two categories, , viz.,
outbreedin~ and inbreeding. Outbreeding includes hybridization, grad-
i,ng, crossbreeding, and outcr~srng; inbriegi~].g includes linebreecilrig and
closebreediEg. -qutbreeding is ~e mating of unrelated animals, inbreeg-
ing the mating of related anImals. N ow all animals on earth are related
if there has been an evolutionary process, as the evidence seems to
indicate; i.e., anyone in Tibet or Tombouctou probably has come origin-
ally from the same parent stock from which we came and has a lot of the (
same genes that you and I have, since we all belong to the genus Homo,
These so-called foreign peoples are related to us but not very closely.
Since inheritance is a halving process, any ancestor in the fifth generation
back of us means very little, since his or her inheritance has been halved
five times before it got to us and, therefore, there is very little of it left.
As an analogy, think of halving a dollar five times-first, halving 50 cents; .
second, halving 25 cents; third, 12.5 cents; fourth, 6.25 cents; fifth, 3.125
cents.
To provide ourselves with a practical working limit, we say that if
any two animals of opposite sex have no common ancestor back through /
four or five generations of their pedigrees, they are unrelated and their
offspring will be outbred. On the other hand, if a sire and dam have a ;
common close-up ancestor or ancestors, their issue will be inbred.
This chapter is concerned with the various types of outbreeding.
Heterosis.-For a considerable time it has been realized that the
mating_gf_unrelated animals is likely to yield offspring of increased vigQ.r.
ThiS phenomenon is called heterosis or hybrid v!W:' -':-The gcn~~easont·
off~r~d to explallli£msed on the fact-that genes'favorable to production 1
are generally dominant over their opposites. Species or breeds develop
somewhat different sets of favorable (dominant) genes, and they all
have some unfavorable (recessive) ones. Wh~ species_.or bree~e
crossed, one pare.nt_may ~,::pply.~ favora.hl~ ,domillant gfl~_to off~et ~e
unf:worable
, - recessive
,._."
one supplied by the other
-_.---_._- ~ -_ parent and vice versa.
- .-
The offspring, therefore,. has a larger number of dominants than does

either parent and is likely to be more vigorous. If one pa.r:ent- (male)
was AA bb and the other parent (female) was aa BB, the offspring would
be _Aa Bb.-Not" if these female hybrids or crossbreds were mated to a
male aa BB and those offspring to a male AA bb and this system con-
tinued, the proportion having at least one A and one B would soon level
off to about two-thirds of those born. l!e~C>_8.!~;"is prQ.bably dl!eJ:!!:..~t ~
to 1he heterozygous condition and in part to the presence of more dom~
inant favorable genes, although what part each plays we do not knm~.
Since the hybrids or crossbreds are more heterozygous, hmvever, th~y
cannot be expected to transmit so uniformly as purebred or inbred strains.
The retention of heterosis in various crossbreeding systems depends,
therefore, on the genetic merit of the purebreds used and on how favorable
"nicks" result.
Hybridization.-The widest possible type of outbreeding is the matin_g
of two distinct species, and a good exaJIll>lc of this is the mating of_the
horse and" the ass species to produce the mule or the hinny. It seems
probable that the horse and the mule (also the zebra and 1he quagga)
have come from common parent stock In the dim past natural selection
• working on variation (recombinations, mutations, and chromosomal
aberrations) began to separate them off into distinct species. The
divergencies progressed so that now the horse and the ass are distinct
species. They still have many genes in common, but they also have
_enOjJ..gh di~;~s so that they are recognized as distinct species.
Up to now, howe';er, their genetic divergencies, while enough to set them
apart as distinct species, have not been enough to interfere with the
fertilization of eggs of one by sperms of the other. Fertilization will take
place and offspring (mules or hinnies) will be born, although the males
resulting are apparently always sterile and the females quite generally so.
Sheep and goats are also related (have arisen from common anc~:tOl:sl,.hut.
a ITttrelesscloscly than th~horse an(r~s -~nc~-:-wh"il;;-fertilization between
therrl" will occur, the embryo does not reach full development, but ·i~
resorbed or aborted. Sheep and cattle have also come from commOfl
ancestors, but in them the divergence has progressed so far that even
fertilization is prohibited.
Ther~ can be little question but that nature has used hybridization
down through the eons of time to produce new species. ~r ll_ybri~i
zation nor any other [mm qLIDaiing produc~s .new genes, all that any of
them d~ is "i2.._gr~~p the old genes into I!ew combinations. fu~
genes do occur tii.roug;h·rn.ut~£ions, new c~inations of genes do oce_ul'
through the various systems of bree<rrni, and the ncliesult has been
cyolution.
,158 BREEDING AND IMPROVEMENT OF FARM ANIMALS
Various Commercial Hybrids.-The.. mul_§ j1' on~ of tIle best known

hybrids in the United States:.- The mule is the result of crossing a jack
(male of the ass family) on a mare (female of the horse family).:... This
cross has been valued for many centuries. Fertile mare mules have been
reported at various times, but the reports have not until recently been
capable of absolute verification. In 1920, L. T. Branham of :Montalla,
Tex. reported that a mare mule in his' possession, which had been bred to
a jack, had dropped-;u~~g-f~~. The - ~~ntion was supported by
affidavits of the owner and s o ; neighbors. The mare was loaned to
the Texas' Agricultural and Mechanical College and, after failing to con-
ceive when bred to a jack in 1921, dropped a living stallion colt in 1923
as the result of being mated to a Saddle stallion the year previous.
Another fertile mule has been reported from Indiana. 1 When bred to
'a Percheron, she dropped a stud colt vdth no mule or ass characteristics.
Still another fertile mule has been reported from Arizona. 2 This mule
bred to a jack dropped a mule foal, and motion pictures were taken of
the birth. This mule is said to have dropped one other foal previously .
• In the above, we have two cases of mare mules bred to jacks producing
"mule foals and two mare mules bred to stallions producing horse foals.
Since the chromosome number in the ass family is thought to be "ome- •
what larger than the 60 found in the horse, the inference in each of the
above cases is that these mare mules occasionally produce an egg con-
taining nothing but 30 horse chromosomes, all or essentially all of the
ass chromosomes having been extruded in the polar body. These
results indicate that these fertile mare mules actually function as mares
as f r as the genetics of their e s is concerned. If all the horse chromo=-
somes were extru ed, these mules would function genetically as asses.
No cases of this sort have yet been reported, and the possibility of
securing a fertile hybrid strain of true-breeding mules is as remote as
ever and would seem to be theoretically impossible.
The genetic pedigree of these offspring of mules illustrates the sampling
.-
nature of the hereditary process, for they should evidently be written
Stallion or jack
Stallion or jack { Mare or jennet
Offspring: Stallion
Horse if by stallion
Mule if by jack Fertile mule
(passed on nothing This line of pedigree
but horse chromosomes) obliterated genetically
from this point back
Fig. 130.-Pedigree of offspring of a mule.
1 See Jour. Hered., 1939, p. 549.
2 See Jour. Hered., 1939, p. 548.
--"
The mule approximates more closely to the ass than to the horse in
generaTcliaracteristics.
The hinny is the reciprocal of the mule, being the result of a stallion-
jennet cross:- It~ said to more closely approximate the horse in general
c-ra~acteristics. The hinny has never become popuiar -or numerous.
The zebroid, a zebra-horse hybrid, is fairly popular in the tropics
becauseoTits docility and its resistance to disease and the effect of the
heat.
Several hybrids in the genus Bos are fairly common. Of these one
of the best known and most valuable results from crossing the zebu, or
humped variety of cattle (Brahman), on ordinary cattle. Heterosis is
evidenced and the zebu's resistance to ticks, fever, and various diseases
is at least partially dominant. This cross is quite common in Brazil and
seems to be gaining in popularity in the southern part of the United
States, especially in Texas. More will be said of it later.
Bison and cattle belong to the genus Bos and are interfertile. Bison
bulls are bred to domestic cows to produce this cross. Males are seldom
born alive, owing either to their height back of the shoulders, which
prevents their passing through the cow's pelvis, or to too great a quantity
of amniotic fluid, and seem never to be fertile. Some of the females are
fertile. The first cross, bison bull on domestic cow, is called a hybrid or
half-bred bison. The second cross is made by using a domestic bull on
the foregoing hybrid. The females resulting from this cross are mated to
bison bulls and the offspring are termed cattaloes. Besides more rapid
gruwth and greater vigor, the following advantages over cattle are
attributed to the cattalo, viz., greater resistance to many diseases, includ-
ing blackleg and Texas fever, higher dressing per cent, failure to run from
heel flies and to drift in storms, lower feed requirement, and a greater
ability to withstand lack of water.
Grading.-Breeds of animals arose by slow degrees from the general
population of livestock by a process of selection. Along with better flesh-
ing, milking, or other qualities, certain breed trade-marks, such as color
of hair or wool, shape and set of horns, etc., were secured. By systems
of mating (often involving quite close breeding) and a continuous dis-
carding of animals that did not show the desired combination of char-
acters, animals which would breed relatively true were finally secured.
When this had been accomplished, the obvious best thing for the breeder
without purebreds to do was to secur a urebred sire to his
grade....animals. Thus the breeding system known.in~~~(f-€~~-qtl:m,
naturally int~ bemg.
Grading is the practige_ of using purebred sires of a given pure breed
c
on native or grad"e' re~ales. Its purpose -Ts to develop uniformity' ana
~.-=_:_~". ,- , - .' - - -----------
460 BREEDING AND IMPROVEMENT OF FARJf ANIlj,fALS
to increase productivity and quality in the offspring. The pure bre~ds

are generally capable of producing these results because they have been
selected for many generations for particular points and purposes within
rather narrow limits and specifications."~ This does not mean that a (
purebred will necessarily effect an improvement when bred to grade (
animals, for, unfortunately, as is quite generally recognized, there are'
even at the present time a goodly number of inferior purebreds. The
average purebred, however, will help to bring about uniformity and to
increase productiveness in the average grade herd and can be used for
this purpose. Th~_jirst-generation offspring will carry 50 per c~_nt oj
p~.!"e. "blood." When these iu. turn are mated to another purebre~ire
of the same breed, the resulting offspring will carry 75 per cent of pure
"blood'," etc._ After five or six top crosses of this sort, where one bre~
has been adhered to for the sires, the animals are, to all intents and
purposes, purebred, though they are not eligible to registration. This
is because the purebreds have certain definite distinguishing character-
istics that breeders do not want to see disturbed. Such might occur if
five or six top-crossed grade animals were admitted to registry, it being
possiblE:) an4_ pr~~~ble that_1;_lle grades !y.ould ]:)~..Qgtrryi.l}g al~gsome
.genes in their germ plasm-for unciesirable chara~teristics from th~.stand
point of the
'the'1'~~e~J
pure b~eedwhich woul~l,~'futu;e time, crop out in
_", . . ' .-~,-.- '-
As the amount of pure "blood" approaches closer to 100 per cent, the
grades come to look and behave more and more like purebreds, and they
finally come to be practically indistinguishable from them. A native
female has 100 per cent grade genes. Her offspring by a purebred sire
has had one-half of its grade genes replaced by purebred genes. If one
of these first-generation grade females is in turn mated to a similar pure-
bred sire, its offspring will in turn have had one-half of its remaining
grade genes (50 per cent of ,those it has) replaced by purebred genes, etc.
We, therefore, would expect the greatest improvement in the first cross
and less in each succeeding generation as the number of purebred genes
is increased.
The fact that some grade animals outdo some purebreds has a tendency
to lead some people to the erroneous conclusion that grading is better
than pure breeding. Some purebreds have the genes that give them the
external trade-marks of their breed but few of the genes for efficient
production. Grading with this sort of purebred will not lead to rapid
improvement, whereas grading with purebreds that have not only the
external trade-mark genes but also a goodly supply of those causing
efficient production will lead to rapid improvement. The majority of
our animals are produced by grade breeding and probably will continue
to be, as long as the average merit of the purebreds continues to excel

that of the average commercial herd or flock. Here lies the real challenge
to the purebred breeder to de;:ise ways of ascertaining just what genes
his animals do possess and alBo the means and the courage to rid his
stock of poor genes.
2E-~_()i_ th~_Q_r:i:r!~ipal fUl)s~tjons Qf_il!~lL1!!:{l__~reeds is tht of_§~~Il_g_a~ a
source of seed stock for cQIl}merQ_iaLm:oqW:J;lXs) The commercial producer
depends>u~the purebred breeder to supply him wi~trated
doses of good genes, whlchlntum means that the principal task of the
purebred breeder is that of ridding his stock of poor genes. With such
a double-barreled system in full operation, it would seem logical to advise
the average man starting in the livestock business to start with the
system of grading. It ,vill be admitted at once that this system leaves
something to be desired, but it must also be admitted that it has its
advantages. The most important consideration is, no doubt, that of
cost. With grades, a start can be made for considerably less money
than would be required for purebreds. Breeders of purebreds recognize
that this practice is sound and advantageous to them, for it will furnish
a market for all of a breeder's really meritorious males and for his rela-
tively few excess females, most of which he will need to keep and test.
It will mean more business eventually for breeders of purebreds to have
beginners start with a grade herd and a purebred sire and later, when
they have learned something of the business and have become established,
change over to a purebred basis. If beginners are induced to start
with purebreds, the herd must be a small one because usually the capital
at hand is limited and the opportunity for eventual success and growth
in the business is thereby greatly curtailed. :Many men have started
with purebreds, have soon become discouraged, and either have gone
back to grades or have gone entirely out of the livestock business.
Besides the greater initial cost of starting with purebreds, there is another
influence to consider, viz., that when a man has invested in purebreds, he
is quite likely to think that he ought either to sell all his herd increase
at rather high prices or to retain it in his own herd as breeding stock.
This is an element that hurts the purebred breeding business in many
ways. In the first place, all the increase of these herds is not capable of
improving the breed, and a part of it should go to the packer. The
average man will not hesitate to send surplus off-type or low-producing
grade offspring to the butcher, but it takes more courage than some men
possess to see the offspring of purebreds, for which they have paid con-
siderable money, going that way. For purely commercial purposes, U
herd of good grade animals, headed by a good purebred sire, is just as
efficient and will make the owner as much money as a herd of purebreds.
For the young breeder, it is a serious question whether to start with

purebreds or grades. Many arguments can be advanced on both sides,
and in the final analysis it is a matter to be settled individually. It
would no doubt often be possible to start with somewhat old, timeworn
Scrub cow No. 60. Average

production 3,313.2 lb. of milk and
178.47 lb. of fat.
Half-blood Jersey No. 241 out

of scrub No. 60. Average pro-
duction 6,126.4 lb. of milk and
348.98 lb. of fat.
Three-quarter blood Jersey No.

348 out of half-blood Jersey No.
241.
FIG. 131.-Showing results of grading up. (From Iowa Agr. Expt. Sta. B 'ul. 188.)
purebreds about as cheaply as with grades. A prospective breeder

might find that he could secure individuals that were growing old, cows
with only three good quarters, etc., for the same prices he would have to
pay for grades. If such animals were of good blood lines, had no heredi-
tary defects or unsoundnesses, had a reasonable prospect of producing a
few more offspring, and could still produce enough so that they would
yield a little more than the cost of their keep, they no doubt would prove
to be profitable investments. Young breeders of limited means who plan
eventually to have purebreds may well consider this and somewhat
similar plans for getting good "blood" (good germ cells) at reasonable
prices.
McCandlish, Gillette, and Kildee, at the Iowa Agricultural Experi-
ment Station, graded up a herd of scrub cows with purebred sires of the
FIG. 132.-The grade Clydesdale gelding, Major MacFarlane, champion at the Inter-
national 1921 and 1923. He is only two generations removed from a cayuse mare.
dairy breeds. The first-generation grades produced 55 per cent more

milk and.44 per cent more fat than their dams, and the second-generation
grades produced 116 per cent more fat than their scrub granddams,
owing to the fact that these cows not only produced more milk, but
had longer lactation periods, things for which the purebreds have been
selected for a great many generations. The champion gelding at the
International Livestock Exposition in 1921 and 1923, Major MacFarlane,
was a grade horse two generations removed from a cayuse or Indian
pony mare. The thousands of purebred beef bulls that have been used
on the cattle of the Western ranges during the last half century have
graded up these herds and greatly improved the quality of the carcasses
produced.
464 BREEDING AND IMPROVEMENT OF FARM A.NLMALS
The following 1 is the summary and conclusions following 10 years of

grading up )beef cattle at the Sni-a-Bar ranch in Missouri:
The usli"of purebred sires of acreptable quality results in successive improve-
ment in the quality of the calf crop, as shown by conformation and market price.
The greatest single step toward improved quality, compared with common
stock, occurs in the first cross. Subsequent crosses increase quality and market
value still more, though in less marked degree.
Quality and best market prices are approximately in proportion to the number
of crosses of pure breeding.
After the third or fourth cross the offspring compare very favorably with
purebred stock in conformation and only exceptionally good sires can bring about
further improvement.
Steers sired by purebred bulls at Sni-a-Bar Farms have topped the market
16 out of 20 times and have 4 times been the highest for the year to date of sale
on the Kansas City market for fat yearling beeves.
Early maturity is a conspicuous result of beef-cattle improvement through
the use of purebred sires.
Steers raised at Sni-a-Bar Farms as a part of the demonstration have sold
consistently for about $2 a hundred pounds more than the average of other cattle
on the same market, and during the period of high prices in 1918 for as high as
$5.95 a hundred above the market average.
Show-ring results are in general agreement with market preferences so far as
indicated by the successful showing of Sni-a-Bar stock in market classes and by
sales of similar cattle on the market.
The demonstration shows clearly that breeding is a dominant factor in the
production of high-quality beeves and that good feeding and management will
not return best results unless the element of good breeding is present also.
In grading up farm cattle the quality of calves is approximately in proportion
to the individual excellence of the sires used, hence the importance of selecting
bulls possessing qualities sought for in the offspring.
Attendance at the grading-up demonstrations has increased from about
500 persons in 1917 to 10,000 in recent years, indicating a wide public recognition
of the practicability of raising better cattle.
Purebreds do not always effect improvement when used on grade or
native animals. The reason for this may be genetic-lack of sufficient
I good dominant genes in the purebred, as already mentioned. It might
also be due to environmental or environmental and genetic reasons.
Purebred stocks which give good results in one set of environmental
conditions do not always give favorable results in some different environ-
mental milieu. Purebred dairy cattle from temperate zones often
degenerate when used in tropical areas, and their offspring do not have
the vigor and constitution for high production in the warmer climate.
J U.S. Dept. Agr. Misc. Cir. 74, 1926, pp. 26-27.
."
SYSTEMS OF BREEDING-UNRELATED A.NIiUALS 465
For grading to be successful, it is not enough to use good purebreds,

but these purebreds must have evidenced the ability to perform ,Yeli
under the conditions which their offspring must meet. -.
It often happens that under grading-up systems the offspring receive
better care than did their female parents. In this way merit which
is environmentally caused is likely to be attributed to genetic worth of
the males used. This same feature may be found in artificial breeding
'where the proved sires' offspring may tend to get better care than did
their grade dams, thus falsely giving genetic credit to sires.
Crossbreeding.-Crossbreeding
,\ includes several types of matings.
::\Iost generally the term means th.§ mating of purebred animals belonging
to different breeds." It sometimes is used to describe the mating of
~nimals belonging to different species, e.g., the horse and ass cross result-
ing in mules and hinnies, but this might better be called hybrid breeding.
Again the term crossbreeding i:; used to cover cases of mating purebred
sires to high-grade females, but this actually is grading.
Crossbreeding brings together more or less divergent sorts of germ
plasm and may result in considerable uniformity in the Fi but may give
rise to variations, especially in the F2 and subsequent generations. In
this, as well as in the matters of size, vigor, and fertility, one or more of
which crossbreeding usually increases, it is quite often the opposite of
closebreeding. l 'Crossbreeding.. has both desirable and undesirable
features. " -_-- -
--~~~ossbreediJ!g is used to a considerable extent in the production of
~--;ranimals,-be~a:.rse crossbred animakoften show an increase in growth,
~-;--and-effiClency-of feed conversion over either one of their purebred
p_arentsJiSome of the grand champion7teers at the International Live-
stoCk E~position have been crosses among Herefords, Angus, Shorthorn:;.
and Galloways. California Favorite, the grand champion in 1916, ,vas
a crossbred Shorthorn-Hereford and was reputed to have been one of the
finest steers ever exhibited. A crossbred Shorthorn-Angus won in 1941.
l 'Crossbred animals are ofte_n superior to purebreds in individual merit.
pwing however, to their heterozygous nature, they often lack the ability
'to transmit breed trade-marks in a uniform manner!' A Hereford-Angus
~rossbred will be black-bodied, white-faced, and polled and heterozygous
for all three characteristics so that eight different combinations of these
three pairs of genes will be found in such an animal's germ cells. If,
however, the parental Hereford and Angus had the genes for desirablp
meat type, the crossbred would probably be of good meat type and
would also transmit it.
A great many crossbreeding experiments have been carried out during
the past 30 years. In l\lareh, 1924, a double litter of pigs was horn at
-Uj(j BREED/XG AX[J IJIPR01 'EJIENl' OF F"lHM . LV/MAL::>
the farm of the College of Agriculture of the University of Illinois. It ·

was a double litter in that a Duroc-Jersey sow was mated to both a
Duroc-Jersey and a Poland-China boar at one estrus. The resulting
litter contained 10 pigs, 6 of which were Duroc-Jerseys, whereas the
other 4 were spotted red and black and, therefore, evidently crossbreds
sired by the Poland':China boar. The purebred pigs weighed 3.23 lb.
each at birth, whereas the crossbreds averaged 3.75 lb. The pigs were
all h andled and fed in the same manner, and at six months the 2 remaining
purebreds (females) averaged in weight 185.5 lb., whereas the 4 crossbreds
FIG. 133.-Loyal Alumnus 4th, the Grand Champion Steer at the 1941 International Live-
stock Exposition. This steer was out of a Shorthorn cow and by an Angus bull. (CoUl·tcsy
of Animal H usbandry Department, P urdue University.)
averaged 235.2 and the 2 fems-le crossbreds averaged 228 lb. This is a
striking example of heterosis.
Roberts and Carroll carried on an experiment involving double matings
of Poland-China and Duroc-Jersey sows to both Poland-China and Duroc-
Jersey boars in which 5 crops of pigs and 105 litters were produced. Of
these 24 litters were purebred, 16 crossbred, and 65 were mixed litters.
Litter size for purebred litters was 8.0 pigs, for crossbred litters was 7.4,
and for mixed litters was 9.8. All purebred pigs averaged 2.62 lb. at
birth, all crossbreds 2.64. When paired for sex and in same litters, the
purebred pigs weighed 2.63 lb. at birth, the crossbreds 2.76. There was
little difference in strength at birth between purebreds and crossbreds
judged subjectively. The mortality before vaccination (average age 38
days) was 43.3 for the purebred pigs, 41.1 for the crossbreds. Paired
purebre~s and crosshreds from 20 litters were used in feeding tests with
---
SYSTEivlS OF BREEDING-UNRELATED ANIMALS 467
the following results in pounds: initial weight, purebreds, 65.3; crossbreds,

68.4; average daily gain, purebreds, 1.59; crossbreds, 1.65; last weight at
same age, purebreds, 179.1; crossbreds, 185.6. The following 1 is the
summary of this work:
Double matings were used with Duroc-Jersey and Poland-China swine to
produce litters that contained both purebred and crossbred pigs.
A significantly larger number of pigs were produced in litters sired by two
boars (mixed litters) than in litters sired by a single boar (purebred 01' crossbred
litters).
The birth weights of purebred and crossbred pigs were subjected to three
methods of analysis to determine whether the differences between them were
significant. By the method considered best adapted to the problem, a small but
significant difference in favor in the crossbreds was demonstrated.
Among pigs farrowed alive the strength gradings were slightly in favor of the
crossbreds, but a slightly larger percentage of crossbred pigs were farrowed dead.
Mortality before vaccination was slightly less in crossbreds than in purebreds.
Small differences in favor of crossbreds were found in respect to weight at
beginning of feeding test, daily rate of gain, feed per 100 lb. of gain, and weight
near market age, but these differences were not satistically significant.
TABLE 29.-PUREBRED-CROSSBRED COMPARISON*
'- Feed,
Weight, Number Weight, Daily 100-
Litters Pigs
birth weaned 70 days gain pounds
gain
---
Purebred Yorkshires. 38 10.6 2.40 7.6 38.4 1.21 375
Purebred Chester 0
Whites ........... 36 9.8 2.38 6.6 39.4 l.30 403

York. male X Chester
White female ...... 29 9.9 2.49 7.4 42.7 l.35 371
Chester White male
X York. female ... 29 10.1 2.42 8.0 42.8 l.33 370
* From Iowa Agr. Expt. Sta. Bull. 380.
Crossbreeding between Chester White and Yorkshire swine at the

Bureau of Animal Industry Farm at Miles City, Mont. is summarized
in Table 29.
If we average the above figures for the purebreds and for the crossbreds,
we find that the crossbreds weaned 8 per cent more pigs which weighed
11 per cent more at 70 days and gained 7 per cent more per day up to
market weight on 5 per cent less feed than the purebreds.
1 ROBERTS, E., and CARROLL, W. E., A Study of Hybrid Vigor in a Cross between
P,land-China and Duroe-Jersey Swine, Jour. Agr. Res. Vol. 59, No. 11, 1939.
4G8 HREEDING AND IMPIWVElVIBNl' OF FAKl1 ~JNJ1HALS
The Iowa Experiment Station reports as follows on 108 litters of

crossbred pigs from the Duroc-Jersey, Poland-China, Yorkshire, and
Landrace breeds: (1) fewer stillborn pigs, (2) crossbreds' more vigorous
at birth and more reach weaning age, (3) cr~ssbreds 3 to 4 lb. heavier
at weaning, (4) crossbreds gained 0.09 to 0.12Ib. more per day, (5) cross-
breds saved 10 to 14 days in reaching a weight of 225 lb., (6) crossbreds
saved 25 to 30 lb. of feed in reaching 225-lb. weight. 1
The crossbred sows observed in this study proved to be efficient pig producers,
either when mated back to a boar of one of the parent breeds or to a boa; of a
third breed. When sired by a purebred boar the pigs from the crossbred sows,
either backcross or three-breed cross, compared favorably with the first-cross pigs.
There is some general reason to suppose that breeds differ in their response to
crossing and also that families or strains within breeds differ, but not enough
evidence to determine which breeds can be expected to cross best with each
other, nor whether distinct families which cross better than others exist and can
be identified and maintained within pure breeds.
Crossbreeding can be continued as a steady policy only by going to purebred
herds for the boars needed for replacement. Crossbred animals have a lower
value than purebreds as transmitters of inheritance. Crossbred sows may be
used successfully for breeding if the boar is a purebred. In this way the hybrid
vigor of the crossbred dam in nursing and rearing pigs may express itself enough
to more than compensate for her lower value as a transmitter of inheritance.
No such offset for his lowered transmitting value could exist in the case of a
crossbred boar. Planless and unsystematic crossing may quickly result in a
mongrel herd from which the owner will get neither profit nor pride of ownership.
CarrolJ2 and Roberts have analyzed reports on crossbreeding from the

U.S. Department of Agriculture, nine state agricultural experiment
stations, and six foreign countries involving over 50,000 animals. They
measured hybrid vigor in the aspects shown in Table 30.
Carroll and Roberts took the following as their point of reference,
"For crossbreeding to be judged beneficial, the performance of the cross-
breds must excel the performance of the better of the two parental
strains of purebreds." Only in one item "average daily gain" did this
happen by 0.006 lb. per day . We think it fairer to compare the pure-
breds with the average of the parental breeds as is done in Table 30
right-hand column. In five of the six items the crossbreds are slightly
better than the average of the purebreds.
A triple-crossing experiment with beef cattle has been carried out by
1 LUSH, J. L., SHEARER, P. S., and CULBERTSON, C. C., Crossbreeding Hogs for
Pork Production, Iowa Agr. Exp. Sta. Bull. 380, 1939.

2 CARROLL, W. E., and ROBERTS, E., Crossbreeding in Swine, Ill. Agr. Exp. Sta.
Bull. 489, 1942.

SYSTEMS OF BREEDING-UNRELATED ANHI'ALS 469
TABLE 30.-ANALYSIS OF CROSSBRED SWINE
Adapted from Carroll and Roberts
- % by
which
• cross-
Av. of
Larger Smaller breds
Crossbred purebred
purebred purebred ex.cel
parents
avo of
'I pure-
breds
Size of litter .... 1,538*-10.1 lb. 1,081- 9.4 1,515- 9.5 9.75 -2.5
Av. birth weight. 1,728 - 2.90 lb. 4,176- 2.65 lb. 6,137- 2.79Ib. 2.775 lb. +0.6
Survival ability. 8,288 -80% 15,874-72% 9,935-80% 76% +5.3
Weaning weight. 15.522 -33.4 lb. 8,133-31. 7 lb. 9,519-33.3Ib 32 . .5.5 lb. +2.3
A v. daily gain .. 489 - 1.43 lb. 574- 1.319 lb. 794- 1.436 lb. 1.3745 lb. +4.3
Economy of gain. 346 -366 lb. 274-3821b 591-368 lb. 374 lb. +1.6
* First figure in each column shows number of pigs involycd.
Knapp, Baker, and Clark of the U.S. Range Livestock Experiment

Station in cooperation with Montana Agricultural Experiment Station.
Random-selected Hereford cows were mated to Shorthorn bulls, these
females to Angus bulls, and these females to Hereford bulls, and the
offspring tested against purebreds over a 2-year period for each cross.
Five to eight purebred Hereford steer calves from each Hereford sire
used were fed out against the crossbreds. Table 31 shows some of the
steer data, more fully reported elsewhere. 1
TABLE 31.-TRIPLE CROSSING IN BEEF CATTLE
1st generation 2d generation 3d generation

avo of 2 years' avo of 2 years' avo of 2 years'
results results results
Cross Pure- Cross- Pure- Cross- Pure-

breds breds breds bl'eds breds breds
--- --- --- --- ---
No. of steers .............. , ....... 57 67 24 91 20 161
Birth weight, lb ........... , ....... 84.0 79.3 78.7 81.3 82.1 8l.3
Weaning weight, lb ................ 423 403 440 390 467 388
Final weight, lb ................... 948 879 974 887 1,033 912
Daily gain per feed lb .............. 1.92 1. 75 2.00 1.86 2.32 2.10
Gain per 100 lb TDN, lb ........... 18.3 18.0 17.4 18.3 18.9 19.0
Return above feed and marketing
costs ........................... $52.87 50.31 78.30 66.00 150.13 133.1:;1
.,' KNAPP, B., JR., .BAKER, A. L., and C~ARK, R. T., Crossbred Beef Cattle for the
Northern Great Plams, U.S. Dept. Agr. Ctr. 810, 1949.
-i70 nJO~'HlJl.W} .ANlJ IJJPRUVEME,V'l' OF FA.. RM A.Nn1ALS
In the females, the weaning weight and weaning score of the crossbreds improved ~
with each generation and all generations were superior to the purebreds. The
weights of the heifers at 18 months were also in favor of the crossbreds. The
first-generation crossbreds were the heaviest at maturity, followed in order by
the purebred Herefords and the second-generation crossbreds.
Growth studies indicate that the first-generation females reach the heaviest
mature weight, but up to 18 months of age the second-generation crossbreds
were heavier than the first generation. The purebred Hereford females were the
lightest of the three groups up to 5 years of age when they become intermediate
in weight between the first- and second-generation crossbred females.
The calf-crop percentages as a whole were higher for the crossbreds than for
the purebreds.
We assume that there was no selection in either group; i.e., crossbred
heifers were all kept to produce further crosses and the offspring of
random-selected Hereford cows bred to random-selected Hereford bulls
were saved each year to compare with the crossbreds.
The authors' discussion follows: 1
The data presented indicate that all three generations of crossbreds were better
in nearly every characteristic than the purebreds fed during the same year and
handled as nearly the same as possible. In progeny groups from individual sires
among both the purebreds and the crossbreds, considerable variation was found.
For example, in daily gain in the feed lot, four purebred Hereford sire groups
outgained the crossbreds and four others gained equally as welL Pure-bred-sire
groups were also observed that weighed more at the end of the feeding period,
that sold for as much per hundredweight, and that returned more money over
feed and marketing costs than did the crossbreds. None of these superior bulls
were used to produce crossbreds in the third generation because of their usefulness
in another project.
Of eight bulls whose progeny on feed showed equal or superior gaining ability
to the crossbreds, seven are from one inbred line that has shown superior merit
in rate of gain. However, in the case of the three Hereford bulls used in both
purebred and crossbred herds, the crossbred progeny materially outgained the
purebred progeny from the same sire. The same statement applies to the Short-
horn bulls, based on tests at Beltsville. These results seem to indicate that the
best results from crossbreeding are dependent on the procuring of good bulls, but
indications are that the progeny of all bulls in this experiment performed better
when out of females of a different breed than when out of females of the same
breed as the bull.
For range producers of feeder cattle, crossbreeding has a distinct disadvantage
that should be mentioned. Order buyers of feeder cattle tend to purchase steers
of uniform breeding, type, size, color, and quality. Thus, crossbred cattle with
1 KNAPP, B., JR., BAKER, A. L., and CLARK, R. T., Crossbred Beef Cattle for the
Northern Great Plains, ['.S. Dept. Agr. Cir. 810, 1949.
'.
SYSTEMS OF BREEDiNG-UNRELATED ANIMALS 471
mixed colors are often discriminated against with a resulting lower price per
hundredweight. The feeder buyer may be justified in his antipathy to off colors,
because of the fact that inferior bred animals of poor fattening qualities may
appear similar to ~~e better bred crossbreds espec.ially at the younger ages. Yet
the finishing quahtles of the two are sure to be dIfferent. It would seem, there-
fore that unless the range producer is either selling direct to the feeder or produc-
ing ~rass-fat steers, the advantages of crossbreeding are largely lost on the feeder
market. Such a disadvantage does not exist on the fat-cattle market where
purchase is made on finish, quality, and killing yield.
The crossbred fem[Lles were excellent range cows, high in fertility, and produced
very heavy calves at weaning time. There seems to be a distinct advantage in
crossbreeding from the female standpoint. The second- and third-generation
calves [Lppeared to profit as much from the fact that they were out of crossbred
cows as from being crossbreds themselves. Milk production of the crossbreds
was high yet it was not necessary to milk out any of the crossbred cows to prevent
spoiling of the udders.
The conclusion seems sound that crossbreeding can be carried on most profit-
ably where the range producer is able to crossbreed systematically and where he
either feeds his own steers or sells direct to the feeder.
Commercial meat production can perhaps be carried on most easily
and most profitably by crossbreeding provided good sires are used. For
the purebred breeder to equal or surpass the results of crossbreeding,
performance data must be collected and evaluated and relatively pure
breeding lines be built up through close- and linebreeding.
Crossbreeding work with dairy cattle has been in progress at Beltsville,
Md. by the Bureau of Dairy Industry since 1939, using Holsteins,
Jerseys, Guernseys, and Red Danes. Since production data on the
parental stocks was limited, practically completely lacking for the sires,
it again is difficult to evaluate the degree of benefit from heterosis. Two-
breed and three-breed crosses have produced very well. This could be
anticipated if the parent females had a good hereditary complex to start
with (genotype perhaps exceeding phenotype) and were to be mated to
high-transmitting males down through the generations of crossbreeding.
We would also anticipate good results from straight breeding these cows
to good sires of their own breed, in fact the Bureau of Dairy Industry
has improved both their Holstein and Jersey herds from levels of over
600 lb. of butterfat to that of over 800 lb. by the use of a succession of
proved sires in each breed. Good females (from good families) mated to
good sires either in straight breeding or crossbreeding will yield increas-
ingly better offspring. Perhaps a slight amount in production will be
gained by crossbreeding, but whether a crossbred dairy herd at the 525 lb.
of butterfat level is worth)more than a purebred herd at a 500-lb. level
might be a dehatable point.
Crossbreeding for the Market.-Next to grading, crossbreeding is used

most extensively in the production of meat animals including poultry.
Its greatest use up to the present has been with swine, less with sheep,
and least of all with cattle and horses. In cattle a small amount of
crossbreeding is practiced between Angus or Galloways and white Short-
horns to produce the blue-gray crossbred, between Angus and Herefords
to produce black animals with ·white faces, as well as crosses bet,veen
the beef and dairy breeds. Some beef ranches use a small proportion of
bulls of another breed to utilize somewhat the principle of heterosis,
whereas in some of the Gulf states a small proportion of Brahman blood
is maintained in the cow herds to give added resistance to insects and
disease. Crossbreeding has also been used in cattle for the purpose of
studying the genetic mechanisms involved in the transmission of fattening
and milking propensities.
Crossbreeding is also practiced quite extensively by range sheepmen
by using rams of the Down breeds on grade l\lerino or Rambouillet ewes
for the production of market lambs.
I The most extensive use of crossbreeding has occurred with swine.
One system is that of using purebred boars on purebred or high-grade sows
of another breed. This often results in the production of faster and
cheaper gains. The problems in this system are: (1) the continuing
search for good sires (but this is true in all breeding) and (2) the problem
of female replacements.
The rate of replacements will, of course, vary with different classes of "
livestock. With cattle the useful life of a cow probably averages between
4 and 6 years, which means that 16 to 25 per cent of the cows must be
replaced annually, so that in a cow herd even with 100 per cent breeding
efficiency, which of course is seldom achieved, one-third to one-half of
the herd would have to be bred to a purebred sire of the same breed each
year and all the resulting females raised in order to provide replacements
at home. With ewes the length of the useful life is also about 4 to 6
years, so that a somewhat similar proportion of the flock must be bred
pure each year, though this may be reduced somewhat if the percentage
lamb crop exceeds 100. With sows the useful life is from 4 to 5 years,
and, because they may be managed so as to produce six to eight pigs
twice a year, the replacement problem is greatly simplified, for the
replacements could be provided from one purebred litter each year.
For the production of market animals the crossbred females can he
llsed for breeding purposes, thus making use of any gain in fertility and
milking ability due to their heterozygosity. Workers at the Minnesota
Experiment Station have suggested systems known as crisscrossing and
triple crosKifig. In crisserossing sows of breed A are m~oar-6f ....c.....
SYS'l'EMS OF BREEDING-UNRb'LAl'b'D ANIMALS 473
breed B. Crossbred gilts from these matings are then selected and bred
back to a boar of breed A, selected females from these matings heing
mated to boars of breed B, etc.
It is seen from Fig. 134 that the crossbreds under crisscrossing soon
come to have about two-thirds of their blood from the breed of their
immediate sire and one-third from the breed of their maternal grandsire.
The Minnesota Station reports;!
The crossbred litters averaged from one-third pig to two pigs larger at weaning;
on the average, each pig weighed from 5 to 7 p()unds more at weaning, and the
litters weighed from 39 to 96 pounds more than the purebreds. The crossbred
A
Crossbreds 100 %
A 68.5
B 31.5 B
Crossbreds 100 % A
A 37.5 100%
B 62.5 B
100% .
Crossbreds
A 75
B 25 Crossbreds
A 50
B 50
A
100%
FIG. 134.-Diagram of crisscrossing.
pigs reached a market weight of 220 pounds from 17 to 22 days earlier than
comparable purebreds, and they reached that weight on from 27 to 36 fewer
pounds of grain.
_/'
Another system of continuous crossbreeding involves the_u~. of__ three
pure breeds ann kndwn ItS tnple cro88ing_. As indicated in Fig. 135, the
- ;ci"mals -soonCOnietCi-have about four-sevenths of their blood from the
breed of their immediate sire, two-sevenths from the breed of their
maternal grandsire, and one-seventh from the breed of the sire of their
maternal granddam. Such a system retains heterozygosity in the ani-
mals and, provided the animals and breeds "nick" well, desirable results
may be anticipated.
In both crisscrossing and triplecrossing, desirable boars of the two or
three breeds must be located and purchased. The success of cross-\
breeding for the market (as indeed for any kind of breeding) hinges pri- i
marily on one's ability to secure purebred sires capable of introducing I
into the life stream the desirable hereditary determ~ners for vigor,!
1 Minnesota Expt. Sta. Bull. 320.
474 BREEDING AND IJfl'IWVEJIEA'l' OF FAR21J A1VD"IALS
fertility, rapid growth, and good carc~ss qualities. We like to think of

a
. the bottom line ofa pedIgree as 'stream that flows from female to female
down the generations. If this is a pretty good stream to start with (good
female family as determined by works, not faith) and the series of sires
continually adds desirable genes, then the family, if sound selection is
practiced, can get better as time goes on. This is equally true for cross-
breeding or straightbreeding, be it noted.
There is some tendency to ascribe magic to crossbreeding, but there is
none, nor in inbreeding either. All that any system of breeding can do
is to put the genes now present in our livestock into new combinations.
No system of breeding can make new genes-just new combinations of
old ones. There is apparently some advantage i~:r;:_~osity itself, ___
C
100%
Crossbreds B
A 15.6 100%
B 28,1 A
C 56.2 Crossbreds 100%
A 31.25 C
B 56.25 Crossbreds 100%
I
C 12.5 A 62.5
B 12.5 Crossbreds ~OO%
C 25 A 25
B 25
C 50 Crossbreds
A 50 A
B 50 100%
FIG. 135.-Diagram of triple crossing.
and the use of crossbreeding in commercial.. production of livestock

products_~~ s~li]ii'c'i-~i:-~ng: -rrne growing realization' of the n~dJor.
good purebred sires for use in crossbreeding is both a challenge to the
breeder 'and' an assurance th~t h'is creations will be in high demand and
at good prices. It will be most unfortunate for all concerned if cross-
I
breeding comes to be considered a panacea for commercial meat pro-

duction and the idea adopted that mating anybody's son of some breed
with everybody's daughter of some other breed, or grades, would always
give desirable results. Actually the value of crossbreeding lies in the
genetic merit of the stocks crossed and whether the genetic complexes
complement each other. We will not get any more out of a cross than
\ve put into it, except a little in the way of interest. '
The quick and widespread acceptance of artificial breeding in dairy
cattle should provide the means for steady improvement in this class of
animals. Formerly, we depended on purebred sires to effect improve-
ment. However, all the sires of a breed would probably group themselves
into a close semblance to a normal distribution curve if their transmitting
powers could be measured. This means that 5 to 10 per cent of them are
'.
SYS7'EMS OF BREEDING---UNRELA7'ED ANIMALS 475
very poor, 15 to 20 per cent are mediocre, 40 to 60 per cent are average,
15 to 20 per cent are superior, and 5 to 10 per cent are excellent in trans-
mISSIOn. Under the old scheme of using purebred sires, so many were
demanded that some rather pOOl' ones were bound to be used. Artificial
breeding still depends on purebred sires but makes it much more probable
that fewer poor ones will be used.
It is being suggested that commercial dairymen 'with mixed-grade
herds use semen from the best bull in the inseminator's kit on a given day
when he has a cow in heat, rather than grading up to a certain breed by
always using sires of that breed. We see no objection to this in principle.
It will give the maximum in hybrid vigor. However, the dairyman will
have to use some judgment as between sires in such matters as probable
size of calf, milk yield, and fat test of cow to be bred, etc.
Crossbreeding and New Breeds.-Crossbreeding with its attendant
variation, plus later painstak.m~ection and inbreeding, has been used
for the creation of several new breeds. In sheep, the Oxford Down is the
direct result of crossbreeding, the Cotswold and Hampshire Down furnish-
ing the pure breeds for the cross. This crossing, followed by rigid
selection for specific points and type, has established the Oxford DovVll
as a pure breed. The Corriedale, aNew Zealand breed of sheep, resulted
from the use of Leicester and Lincoln rams on Merino ewes, and the
Columbia breed was created by the specialists of the U.S. Department of
Agriculture, using the Lincoln and Rambouillet breeds.
Two things should be borne in mind so far as the creation of new breeds
is concerned. One is the fact that two men might start to cross breeds
A and ~ to produce a new breed C. Since none of our breeds is entirely
homozygous, they would surely start with different genetic materials.
In addition, chance sampling of the hereditary material plus different
ideals in each breeder's mind might result eventually in two samples of
a new breed, called C, made up of very dissimilar genes and therefore
very dissimilar in type and productive capacity. If they sold stock to
other breeders, the two strains might further diverge. The other point
to be kept in mind in the creating of new breeds is that of the time
involved. Whether it is to take 6 generations or 16 or 60 before -One
would be justified in saying that the result of the original crossing and
later selection now merits being called a new purebred is also an arbitrary
matter. Many of the so-called new breeds still show a great range oj
'variation after 25 or 30 years of careful selecting, and, as we have jus,
stated, even the oldest of our pure breeds are still relatively heterozygous
To the King Ranch of Kingsville, Tex., goes the honor of having
created the first American breed of cattle, the Santa Gertrudis. 1 Estab·
1 RHOAD, A. 0., The Santa Gertrudis Breed, Jour. Hered., Vol. 40, No.5, 1949.
lished in 1851, the King Ranch was stocked originally with "Texas
Longhorns" which had environmental fitness but yielded a poor carcass.
Shorthorns and Herefords were used for top-crossing between 1880 and
1910 and successfully upgraded the carcass but downgraded environ-
mental fitness. To restore the latter, recourse was had by Mr. Robert J.,
Jr., and Mr. Richard Kleberg to crossing with Brahman bulls, Bos indicus,
after an exploratory period (1910-1918) indicated that the cross had
promise. Three-quarter to seven-eighths Brahman bulls were secured
from the Pierce Estate in Texas for mating with purebred Shorthorn cow
herds. Red males and females from this F 1 were selected to beget the F 2
without inbreeding. One of the Fl bulls (Brahman bull X Shorthorn
1916 to Brahman ;'16 cow) was named Monkey and became the real
FIG. 136.-Santa Gertrudis steers in feed lot. (Courtesy of Kino Ranch, Kingsville, Tex.)
founder of the Santa Gertrudis breed. Because of the marked superiority

of his calves, he was used extensively in breeding for a period of about
10 years, and 150 of his sons and later descendants have been used on
the first-cross and double-cross cow herds. Planned inbreeding and
linebreeding to Monkey plus some inbreeding to females has in 30 years
yielded a new breed (% Brahman and % Shorthorn) combining environ-
mental fitness and good carcass quality.
The Lasater Ranch of Falfurias, Tex., has developed the Beefmaster
strain by using Hereford, Shorthorn, and Brahman blood. This pro-
gram was started in 1908. The Lasaters make no claim to producing a
new breed, their sole 8,im being to produce as good beef as possible under
their conditions. These cattle are being tried out in other parts of the
United States.
Turner and Thomas of Welaco, Tex., are developing a new beef breed
ca.lled the Char bray through crossing the French Charollais breed with
SYSTEMS OF BREEDING-UNRELA 'l'ED ANI MALS 477
Brahmans in the proportions of .% and}>i. These cattle are also being

tried out in various parts of the United States.
The Hamprace hog has been developed at the U.S. Range Livestock
Experiment Station, Miles City, Mont. by the Montana Agricultural
Experiment Station and the Bureau of Animal Industry, U.S. Depart-
ment of Agriculture, and is now generally referred to as ]\II ontana No.1.
Vnbelted I-Iampshires and Danish Landrace provided the original stock.
These strains were crossed and the breed developed by selection from
the F2 and backcrosses to each strain. For the 9-year period 1939- 1947
inclusive, this new breed has averaged 10.6 pigs farrowed, 8.1 pigs raised,
FIG. 137.-The grandmother of the Montana No. 1. This black sow was from a second-
generation litter of tbe Landrace X Hampshire cross. In 8 litters she farrowed 96 pigs
and weaned 83 at 56 days of age. The litters averaged 351 lb., or 34 lb. per pig. Selection
has largely elimina ted the depression behind the shoulders but has not increased produc-
tivity. (C01,rtesy of J. R. Qu e8en.berry, u.s. D epartment of Aor1.cuU.ure, Mt7es City, Mont. )
with an average 56-day litter weight of 248.4 lb. Record of performanco

tests on an average of 68 pigs per year over a 5-year period showed an
average daily gain of 1.43 lb. per day requiring 371 lb. of feed per 100
lb. gain, the latter showing a range of from 310 to 444 lb. Carcass tests
on 48 hogs made at 184 days when the hogs reached 221 lb. showed a
carcass length (aitch bone to first rib) of 30.4 in., the five primal cuts to
average 47.1 per cent of the live weight and a backfat thickness of 1.5 in.
The Montana No. 1 is a solid black hog. The back is slightly arched. The
sides are smooth and uniformly deep. The hams are thick and carry well down
to the hocks. ' The legs are medium in length with a neat jowL The ears are
of fine texture, medium in size and vary somewhat in carriage, ranging from
slightly erect to slightly drooping. The disposition is very quiet and docile.
-J78 JJHJJ}·; })}.W' .LYD IJIPIwr ' K 1!H:\ l ' OF F .1RJ[ .L\}.1ULS
The sows are quiet at farro-wing time and suckle well. The breed is a good
forager.
The Minnesota No. 1 and No. 2 are also new breeds-the former
resulting from crosses of Tamworth and Landrace and the latter from
crosses of Yorkshire and Poland-China.
FIG . 138.- Minnesota No. 1, Boar (top) and Minnesota No.2, Boar (boUom) . (Court esy of
L. M. Winters, Univ ersity of Minnesota.)
It~uld be Qossible through cros"mreeding to establish new breeds
I
wit!: any de~'e~ combinatio. ns of characters, unless, of course, the latter
are an agorustic to each other or certain genes are linked in such a way
in the germ plas'"m th~t segregation is impossible. he drawbacks to
this procedure are serious ones, however, 'because of the time and expense
involved a,nd the large number of animals that must be produced in
\
\
\
\
SYSTEMS OF BREEDING-UNRELATED ANIMALS 4i9
order to give ample range for selection. After these difficulties have
been overcome, there would still remain even a greater one, viz., selling
the new breed to the breeders and growers, a matter which would vary
among the different classes of livestock. The breeds as they exist today
seem to fill most of the needs fairly well. Considerable variation still
exists in all of our breeds, in fact, probably enough so that those in each of
the classes could for the most part be made uniform in productive
qualities through suitable matings and selection, although they would
still differ in external character such as color, presence or absence of
horns, set of ears, etc.
This whole matter of the possible creation of superior new breeds in
any and all classes of livestock is an extremely fascinating one and has
intriguing possibilities, though it is not a field in which an ordinary
breeder \yith limited time, capital, and number of animals can hope to
achieve success. He lacks the necessary equipment. If exploited at all,
these fields should generally be left to the state agricultural experiment
stations or the U.S.. Department .._\.
of Agriculture.
Uses of Crossbreeding.-Crossbreeding is useful in making gen~ic
studies of hereditary transmission of characters. It provides the oppor-
tunity for'combining the better genes of two or more breeds and thus
getting enhanced commercial production. It-offers the opportu~ity for
I....
'1))the cr~aMon of new breeds- by s~itable ci"ossing -.
and
-
selection fgllowed by
....••.... -
inbreeding to render the desirable combined genes relatively homozygous.
The practical breeder is generally not suitably equipped for genetic
studies or for the formation of new breeds. These functions can best
be carried on by experiment stations. He is vitally interested in the
best methods of commercial production of various livestock products.
Maintaining two or more separate breeds for commercial crossbreeding is
not generally feasible, while mating crossed females to a series of good
purebred sires of()_l!e ()r_s_El:ver"ilbreeds is~··- But there is no magic in cross--
b~~~di:r{g;a~d promiscuous,- planless crossbreeding is not likely to lead to
beneficial results.
Finally, we should mention the aesthetic side of the problem. We all
know of the pride that the good purebred breeder takes in his uniform,
trade-marked flock or herd, sometimes bordering on biased, prejudiced
partisanship. "Belonging" to a pure-breed association gives many men
a deep feeling of satisfaction, and they can not only "belong" but also
bathe in the reflected glory of the breed as a whole. For most men there
probably is -just more fun and satisfaction in working with a uniform
flock or herd of purebreds than in working with grades or scrubs. As
one farmer put it, "The extra five or ten per cent efficiency of a well-
planned hut heterogeneous group of grade animals is just not sufficient.
compensation to pay me for what I lose in satisfaction by not having typi-

cal representatives of a well-defined, homogeneous group of purebreds."
Outcrossing.-The systems of breeding previously discussed in this
chapter; viz., hybridization, grading, and crossbreeding involve the
mating of unrelated animals that do not belong in anyone breed; in
other words, they involve the uniting of very diverse bits of germ plasm.
Outcrossing, on the oth~rh~n<i, is the mating of animals that are members
of the--~~~~ breed but show no relationship for at least the firs~to
~~rati()n~lJJt.Siiidicated i~-the early chapters of this hook, it s;;;;n;-
probable that all organisms extant and extinct which have ever lived on
this planet are somewhat related. The theory of evolution suggests that
all living things trace back to a single, or perhaps a few, similar origins.
Cattle,sheep, and goats belong to the family Bovidae, i.e., they have
;;om~ through variation from the-~~~e progenitors:---rii the early stages
of this diversification, the germ plasms were still compatible and the
animals interfertile. Continued variation and selection among the genes
finally reached the point ,vhere the germ plasm of cattle was not com-
patible with that of sheep, the animals, therefore, being sterile with each
other.
So in the broad view all cattle are related and, of course, still inter-
fertile. All cattle, in other words, probably have many genes in common.
But for practical purposes, an anirga) that shows no common ancestor
on both sides of its pedigreeOTor-ofour or five generations is called an
outcross. To put it another ,Yay, when we mate two unrelated animals
in th.e same breed, their offspring is called an outcross. The term out- - i
crossing is also used to signify the selection of a-:n ~nrelated male to use
on a group of inbred females with the inference that the breeder will
subsequently return to his inbreeding. It is used, in other words, to
introduce some desirable genes into an inbred stock.
Most of our purebred animals are the result of out crossing. The
mating of unrelated anim-als sometimes results very favorably, to be
sure, but it is also a fact that most breeders ,vho achieve distinction do
so by (1) securing desirable animals of certain strains or families in their
particular breed and (2) intensifying the good characteristics of these -
animals through close- or linebreeding. Nevertheless, the master
breeders at times have recourse to out crossing for its regenerative effect
or for the purpose of introducing new genes. Only ,yhen used in these
latter ways with a specific aim in view is out crossing to be commended.
Crossbreeding Effects within a Breed.-The good effects of cross-
breeding are thought to be due not to the fact that breeds are crossed
but rather to the fact that the genes making one breed good are somewhat
different frbm those making another breed good. These two "goods"
SYS'l'EMS OF BREEDING-UNRELATED ANIMALS 481
when brought together often make something better than either one
Hingly. In thinking of crossbreeding, therefore, we should think of genes
rather than breeds. Since this is so, we could expect to find evidence of
heterosis within a breed if selection and inbreeding had separated out
different complexes of good genes. The Oklahoma Experiment Station
working with the Regional Swine Breeding Laboratory reports such [l,
case. 1
The Oklahoma Station started a project with Duroc Swine in the fall of 1937.
The project has been carried on in cooperation with the Regional Laboratory
since that time. Four inbred lines of Durocs were developed. Line 1 had
already been inbred in another project and was retained to be used in this project.
It has been necessary to discard two lines to date and start two new ones. One
was discarded because of a lack of fertility and the other one was discarded on
account of inverted nipples. Moderate inbreeding is being used in this project.
Most of the matings are half-brother X half-sister.
TABLE 32.-A COMPARISON OF 15 LINE CROSS LITTER8 WITH PURE LINE LITTERS
FROM EACH OF THE PARENT LINES, AND WITH CROSSBRED LITTERS
I
Inbreed- No. of pigs per Av. weight of Av. daily
litter at pigs at gains No. of
No. of ing of weaning litters
litters litters,
to 6 tested *
% Birth 6 months Birth 6 months months
------
.. Pure line 1
1938-1943 .....
Pure line 3
62 39 7.8 3.3 2.3 111 0.72 10
1938-1943 ..... 94 15 9.0 5.6 2.6 169 1.14 32

Line 1 X
Line 3
Fall 1943 ...... 15 0 9.4 6.2 2.4 199 1.38 10
Crossbreds, fall
1941 and
spring 1942 .. 8 0 9.6 7.5 2.7 160 1.11 6
* A litter is tested by taking 4 representative pigs at weaning time and placing them in a small lot
which has a concrete floor. The pigs are fed in this lot until they reach 225 lb. in weight. In case 3
pigs reach 225 lb. and the fourth pig is smaller, it is removed at the same time.
In the fall of 1943 fifteen litters were produced by crossing lines 1 and 3. Seven
line-1 sows whose inbreeding averaged 39 per cent were mated to line-3 boars
" whose inbreeding averaged 28 per cent. Eight line-3 sows whose inbreeding
averaged 25 per cent were mated to line-1 boars whose inbreeding averaged
36 per cent. While the inbreeding within the lines has not progressed very far,
it was considered worth while to check the combining qualities of the lines
especially since some preliminary crosses have shown promise.
1 Okla. A gr. Expt. Sta. Cir. 113, April, 1944.
482 BRFJBDING AND IMPROVEMENT OF FARM ANIMALS
It seemed advisable to check the "line-cross" pigs with crossbreds as well as

the pure line. Crossbred pigs were produced by mating line 1 sows to a highly
inbred Poland-China boar from the Minnesota Experiment Station. The four
comparisons made, "line cross," crossbreds, pure line 1, and pure line 3, are shown
in Table 32. The line-cross litters were superior in about every respect to the
better one of the parent lines. There was no difference in the economy of gains.
The line-cross pigs gained faster from weaning time to 180 days of age. They
were better in this respect than the crossbred pigs. The -crossbred litters were a
trifle larger at birth and a larger percentage of the pigs were raised to 180 days of
age. Hybrid vigor seems to be manifested in the line-cross litters. ~
The average daily gains made by line-cross' pigs during the fattening period
were compared with the gains made by pure-line and outbred pigs. The line-
cross pigs gained a quarter of a pound more per head per day than the outbred pigs.
The Committee on Investigations of the American Society of Animal

Production, with W. V. Lambert as chairman, submitted the following
report in regard to terminology to describe the progeny from various sys-
tems of breeding at the 1940 meeting.!
Crossbred. The progeny resulting from the mating of different breeds or, in
the case of poultry, different breeds or varieties.
Incross. The progeny resulting from the crossing of individuals of inbred
lines within the same breed, or variety.
Inbred Line. The progeny resulting from breeding closely related animals.
To be classed as an inbred strain, the individuals should have an average coeffi-
cient of inbreeding of 37.5% (equivalent to 2 generations of brother-sister
matings).
Incrossbred. The progeny resulting from the crossing of individuals from,'
inbred lines of different breeds, or varieties.
Outcross. A cross of relatively unrelated animals within the same breed, or
variety.
Topcross. The mating of a male of a certain family to females of !$'tOther
family of the same breed, or variety. '
Topincross. The progeny resulting from the cross of inbred sires on noninbred
dams of the same breed, or variety.
Topcrossbred. The progeny resulting from the mating of inbred sires with
non-inbred dams of different breeds, or varieties.
Heterosis (or Hybrid "Vigor). The superiority over the better parent that is
exhibited by the progeny. This applies to the progeny from the crossing of
/ strains, varieties, breeds, or species. The committee suggested that this ter- (
minology is sufficiently inclusive to describe various types of crosses without
the use of the term "hybrid."
No decision could be reached on the term "hybrid" in view or the many
different interpretations of the word.
1 See Amer. Soc. Anim. Prod. Proc., 1940, p. 378; and Jour. Anim. Sci., 1(1):90,
February, 1942.
'.
\
\. \
Summary.-This chapter has dealt with outbreeding systems-all of

which involve the mating of unrelated animals. All animals in truth are
related, and the term unrelated was defined and limited to mean no
common ancestors in the first four to six generations of a pe<iigree. The
widest out crossing is the ~of ani~aJs in different specie~-hybrid
ization followed by grading and crossbreeding. The narrowest out-
breeding system is outcrossing-mating of unrelated animals within one
breed.
Outbreeding helps phenotype and hurts genotype or produces animals
which in themselves are ~ighly desirable but which cannot transmit
uniformly. It _keeps the genes heterozygous so thatYllcl.~~ir:~!=>le reces-
~ C!\,Ilnot sho\vup. Selecfi.OUis-not sO' necessary because there is -no
way of prediCtlng-'what crossbred genotype may" nick" best with the
next purebred sire.
Crossbreeding is a very useful system for the commercial producer-
the animal multiplier:'- Outbr~e_<iingha~:RQ§§ibilities for bringing together
diverse kin.ds of germ plasm in the formation of new breeds. The form
~~utbreeding ~ost useful for the constructive purebred breeder if>
outcrossing, and here the purpose is somewhat similar to that just
mentioned, namely, the introduction of some new and missing genes into
a purebred herd from another strain in the same pure breed.
References
Books
EAST, E. M., and JONES, D. F. 1919. "Inbreeding and Outbreeding," J. B. Lippin-
cott Company, Philadelphia.
KLEBERG, R. J., JR. "The Santa Gertrudis Breed of Beef Cattle," Kingsville, Tex.
LUSH, J. L. 1945. ,I Animal Breeding Plans," Collegiate Press, Inc., of Iowa State
College, Ames, Iowa.
KICHOLS, J. E. 1945. "Livestock Improvement," Oliver & Boyd, Ltd., Edinburgh
and London.
WINTERS, L. l\JL 1948. "Animal Breeding," John Wiley & Sons, Inc., New Ymk.

BAKER, A. L., and QUESENBERRY, J. R. 1()44. "Comparison of Growth of Hereford
and F, Hereford X Shorthorn Heifers," Jour. Amer. Sci., 3(4):322-325.
BARTLETT, J. 'V., REECE, R. P., and l\IrXMER, J. P. 1939. Inbreeding and Out-
breeding Holstein-Friesian Cattle in an Attempt to Establish Genetic Factors
for High Milk Production and High Fat Test, N.J. Agr. Expt. Sta. Bul. \)()7.
BROWN, H. B. 1940. Hybrid Corn, La. Agr. Expt. Sta. Gir. 25.
CARROLL, 'V. E., and ROBERTS, E. 1942. Crossbreeding in Swine, Ill. Agri. Expt.
Sta. Bull. 489. l
COLE, L. J. 1935. Crossbreeding Poultry, American Poultry Industries, Chicago.
1925--1926. Genetic Foundations in Crossbreeding, reprint from A mer.

Soc. Anim. Prod. Proc. (Papers from the Department of Genetics, Wis. Agr.
Expt. Sta., 59.)
1924. The Wisconsin Experiment in Crossbreeding Cattle. (Papers from
the Department of Genetics, Wis. Agr. Expt. Sta., 40.)
GOWEN, J. W., STADLER, J., and JOHNSON, L. E. 1946. On the Mechanism cf
Heterosis, etc., Arner. Nat. 80 :506-531.
HEADLEY, F. B. 1940. Purebred and Crossbred Pigs, Comparison of Rate of
Growth and Economy of Gains, Nev. Agr. Expt. Sta. Bul. 153.
HORLACHER, W. R., SMITH, ROBERT M., and WILEY, \VILLIAM H. 1941. Cross-
breeding for Broiler Production, Ark. Agr. Expt. Sta. Bul. 411.
HUTTON, R. E., et al. 1948. The Hamprace Hog, Mont. Agr. Expt. Sta. Bull. 454.
KILDEE, H. H., and :MCCANDLISH, A. C. 1916. Influence of Environment and
Breeding in Increasing Dairy Production, Iowa Agr. Expt. Sta. Bul. 165.
~-- and GILLETTE, L. S. 1919. Influence of Environment and Breeding in
Increasing Dairy Production II, Iowa Agr. Expt. Sta. Bul. 188.
KNAPP, B., JR., BAKER, A. L., and CLARK, R. T. 1949. Crossbred Beef Cattle
for the Northern Great Plains. U.S. Dept. A gr. Cz"r. 810.
LUSH, J. L., SHEARER, P. S., and CULBERTSON, C. C. 1939. Crossbreeding Hogs
for Pork Production, Iowa Agr. Expt. Sta. Bul. 380.
MANRESA, MIGUEL. 1939. Animal Breeding ]l.lethods Used in the Formation of
Types of Cattles Suitable for Raising in the Tropics, reprint from Philippine
Agr., 28 :479-490.
RICKEY, F. D. 1935. The What and How of Hybrid Corn, U.S. Dept. Agr. Farmers'
Bull., 1744.
ROBERTS, ELMER. 1929. A Zebra-horse Cross, Jour. Hered., Vol. 20, December.
- - - , and CARROLL, W. E. 1939. A study of Hybrid Vigor in a Cross between
Poland-China and Duroc-Jersey Swine, Jour. Agr. Res., 69(11) :847-854.
WILLHAM, O. S. 1944. Hybrid Vigor within a Breed. Okla. Agr. Expt. Sta.
Mimeo. Cir. 113.
WINTERS, L. M., COMSTOCK, R. E., and DAILEY, D. L. 1943. The Development
of an Inbred Line of Swine from a Crossbred Foundation, Jour. Anim. Sci., Vol. 2,
No.2.
- - - , et al. 1935. A Six Years Study of Crossbreeding Swine, Minn. Agr. Expt.
Sta. Bul. 320.
WRIGHT, S. 1921. The Effects of Inbreeding and Crossbreeding on Guinea Pigs,
U.S. Dept. Agr. Bul. 1090.
CHAPTER XIX
SYSTEMS OF BREEDING-RELATED ANIMALS
The last chapter dealt with the systems of breeding involved in the
mating of unrelated animals, which were defined as those having no
common ancestors on both sides of their pedigrees for at least the first
four to six generations. It would be possible to bring to America 20 sows
and 3 or 4 boars of some foreign breed and thus establish the new breed in
America. The animals would become scattered to all parts of the
country, somewhat different ideals of perfection might be set up in the
different sections, and after thirty or forty generations there perhaps
would be several thousands of animals in the breed. Most of the matings
would then probably be between boars and sows that had no common
ancestors in the first few generations of their pedigrees. Such matings
would be called "outcrossing" although all the animals in this breed
would trace back to some of the original 20 sows and 3 or 4 boars and
would, therefore, be more or less related. Mutations, different selection
ideals, and the sampling nature of the hereditary process would have
made some of the sectional strains quite different genetically, although
all the animals in the breed would probably have many genes in common.
In fact, all living organisms are related, to some extent, from an evolu-
t~onar! standpoint, and it is not inconceivable that all may have some)
genes III common.
Any two cattle of the present day trace back eventually to the few
animals that through variation actually gave rise to the class of animals
we now knO\v as cattle. So much mutation, variation, selection, and
recombination has transpired since tha~ote time that thefact. that
all cattle are of the same" blood" means little or nothing so far as their
present genetic make-up is concerned.
It would be very interesting if you and I could throw on a moving-
picture screen a parade of our ancestors, generation by generation,
starting with our two parents, then following with our four grandparents,
then our eight great-grandparents, etc. Occasionally, if all the facts
were known about these people, we might be inclined to blush and would
be reminded of a saying attributed to Mark Twain that "!!ill!! is Hle
only animal that blushes or needsJo." If our parade extended back to
485
486 BREEDING AND IMPIWVEMEN'l' 01/ FARM ANIMALS
the time of Christ, there would probably be lots of duplications in it,

since this would involve about 69 generations, counting 28 years to a
generation, and, since the number of ancestors doubles in each genera-
tion, it would mean a total of 2 69 or about 600,000,000,000,000,000,000
individuals in the single generation of our pedigree in A.D. 1. It is
obvious that this many people were not alive at that time (there are
more people in the world now than ever before, about 2 billion) or, in
other words, that the same individuals occurred many times in that and
other generations of our pedigree. Your pedigree and mine would
probably also show many duplications.
, In the sense here used, therefore, "related" animals are those with one
-
/ or more common ancestors within the first four to six generations and,
~
therefore;" with probably more than the average number of genes in
,
commo~. When two such animals are mated, w~. call this inbreed~
Diverse opinions are held regarding the practice of inbreeding. Speak-
ing generally, it is in bad repute, because it is believed to result ill a more
or less degenerate offspring. There can be no doubt, however, that
many cases of improvement have also resulted from its use.
In the human family inbreeding is generally frowned upon and pro-
scribed. Many instances of both good and bad results from inbreeding
in our species can be cited. In general, however, it is probably a good
thing that we do not allow inbreeding in our own species for the reason
that our species is the only one which goes out of its way to save its bad
genes. In spite of all our efforts to root out bad genes in our livestock,
many still persist and make their presence known following inbreeding.
In addition, we can dispose of the bad results of inbreeding in our live-
stock, but not in our own species. So our laws against close matings
in our species are justified, but it is unfortunate that this idea is so
generally carried over into our thinking about livestock breeding and
thus limits the use of inbreeding, one of the most valuable tools in the
livestock breeder's kit.
, Inbreeding is technically divided into two categories, viz., closebreeding
. and 1ineQr~eding. Closebreeding is the mating of a full broth~rto- sister,
of sire~to his d~ught;;--or-dam toller son. These types of matings give
th~.greatest concentration of similar "blood," and this concentration
appears close up in pedigree. In fact, the concentration is discovered
in the second generation back in all the cases listed as examples of
closebreeding. Linebreeding is the mating of animals of wider degrees
of relationship than those just stipulated for close breeding, .and is . k'
generally directed toward keeping the relationship high to some desirable ..
ancestor OI: ancestors. In outline form, this is as follows:
, \ . i
\ .
\, \
SYSTEJIS OF BREEDING-RELATED ANIMALS 487
Sire to daughter
ClOSebreeding Son to dam
(
Inb d' Full brother and sister
ree mg{ Half brother and sister or matings
Linebreeding (Of animals more distantly related;
e.g., cousin matings, etc.
~(ml.d_j}~rhap§ p_e pointed out that many breeders use the term
tnJ_)reeding jg_ refer t() " ~lose " ma tinge; aI~_(iLfue_breeding to refer to_~-vider
OllBS,_ Nevertheless, two animals which have any close-up (four to six
-generations) ancestors are related, i.e., more related than average animals
of the breed, and if we mate such animals we are practicing some degree
of inbreeding. In this chapter inbreeding means either close- or line-
breeding, and" close" and" line" are illustrated herewith schematically.
It is said that only 21 Brown Swiss bulls and 129 cows were ever
imported to the United States. All the many thousands of Brown Swiss
cattle now existing in this country must trace back to these, so that any
two Brown Swiss picked at random today quite probably would be found
to have some common ancestors if their pedigrees were traced back far
enough. Likewise only 6,000 or 7,000 Holsteins were brought to the
United States, and the millions of Holsteins now trace back to them and
are, therefore, related. So when we say that outbreedin__gj~the mating
of unrelated animals and inbreeding the mating of relai~d ani~~i~~--we-are
~singthese terms in the spe~ial, limited time sense o[four~to SIX gener-
ations back. Since there have been, especially in the formative and
early stages, particularly favored animals or families in all breeds, it is
probable that the animals in any breed are inbred to a small extent. 1
For example, in a study of Holstein-Friesians born in 1909 these were found
to be related to each other about 2.6 per cent relative to the foundation stock of
about 1883. If a present-day Holstein-Friesian is related 40 percent to another,
both pedigrees being traced back only to 1909, we are not apt to be seriously in
error if we assume that the relationship found if both were traced back to 1883
would be about 41.6 per cent (40 per cent plus 2.6 per cent of the remaining
60 per cent). In other words, 40 per cent relative to 1909 is about the same as
41.6 per cent relative to 1883 in this breed.
Two animals that have any close-up ancestors in common are related,
and if we mate such animals, we are practicing some degree of inbreeding.
What inbreeding does genetically is to sort the original heterozygous
genes _():ut into homozygous form. If we started with organisms that
were Aa Bb and practiced inbreeding, we could by testing and selection
1 LUSH, J. L., "Animal Breeding Plans," Collegiate Press, Inc. of Iowa State College,
Ames, Iowa, 1945.
other a, 74 chance, or one get a and the other A, 74 chance). There-

fore, on the average, there is a 50 per cent (Yz) chance that they will
both get the same gene from S. Since the same is true as to genes from
the dam D, then X and Y have on the average Yz X Yz, or 74, of their
genes as duplicates from S and ~f X Yz, or ~4:, of their genes as duplicates
from D, which means that, on the average, Yz of the genes in X and Y
are identical, and they are, therefore, related by 50 per cent, although
they have 100 per cent the same "blood." X and Yare full brothers
but are only 50 per cent related because of the sampling nature of inheri-
x{~ (Aa) 1 pair of genes Y1~ (Aa)

FIG. 141.-Schematic pedigrees of two full brothers.
tance, the fact that a parent passes along only a sample half of its total
genetic material to any offspring.
The formula for relationship between X and Y is, therefore,
Rxy = summation of [(Yz)n+n']
The fraction Yz stands for the halving or sampling process of inheritance
in each generation, n stands for the number of times this halving or
sampling has gone on between S and X, and n 1 stands for the number of
times the halving process has gone on between Sand Y. Therefore,
for S, R = Yz1+1 or ~/;)2 or 25%
and for D, R ~'2Hl or Yz2 or 250/0
and the summation is RXj,
l
B~~
A C 5F
1G
FIG. 142.-Schematic pedigree of half first cousins.
In a similar fashion it can be shown that in half brothers or sisters ~4:

of the genes, on the average, will be identical, and these animals are
therefore related by 25 per cent. Similarly, half nrst cousins, as illus-
trated in Fig. 142, will have 7J'6 of their genes identical and will, there-
fore, be related by 6.25 per cent.
In all cases of relationship, we simply apply the formula Rzy = sum-
mation of all lines of inheritance reaching the related animals from all
the common ancestors that they have [(Yz),,+n'j. For the father-son
relationship (Fig. 139) we would have Ryou-your father = Yz1+0 = Yz = 500/0, '
for thpre is one generation between you and your father, or, in other
words, tlw genet.ic material has heen halved once in getting from your
SYSTEMS OF BREEDIl1iG-RELATED ANIMALS 491
father to you, and there is no generation between your father and himself.
Likewise, in Fig. 142 the relationship between A and X is
Rax = Yz2+2 = Yz4 = 71:6 = 6.25%

The figure 6.25 per cent means that A and X probably have 71: 6, or 6.25
per cent, more identical genes than unrelated animals of their species and
breed.
In the following two pedigrees (Fig. 143) of double first cousins we have
still 100 per cent the same "blood"; i.e., both A and B get all their
i' lC}M
IN
A )0
DIP
FIG. 143.-Schematic 1}edigree showing 25 per cent relationship.
blood from M, N, 0 and P but the sampling process has gone through
one more generation than is the case with full brothers so A and B
(Fig. 143) are only 25 per cent related.
Applying the relationship formula to A and B we have for M R = %2+2
and the same for N, 0 and P so that the summation for all four is
4~4 or 25 per cent.
The term" blood" is being put into quotation marks, because, although
it is the usual way the breeder expresses the idea of inheritance, actually,
FIG. 144.-Simultaneous direct and collateral relationship.
of course, blood as such does not function in inheritance, this mechanism

being provided in the germ cells with their chromosomes and genes.
Each embryo produces its own blood, and the only way any animal can
get any blood from another, even its own mother, is by means of a blood
transfusion. Used in this sense, the term "blood" means inheritance, or
chromosomes and genes.
We might, of course, have both direct and collateral relationship at
the same time, as illustrated in Fig. 144, in which D and E are related
both directly and collaterally, for E is an ancestor (sire) of D (direct),
and B is an ancestor of D (through F) and E (collateral), The direct
relationship of D and E is expressed by
~HO = Yz = 50 %
492 BREEDING AND IMPROVEMENT OF FARM ANIMALh
and the collateral relationship by

Yz2+l = Yz3 = 12.5%
Summating, we have
62.5%
That is, since E is the sire of D, then D got half of his genes from E
and is related to E by 50 per cent. In addition, D and E are related
through B, which is the sire of E and also the maternal grandsire of D.
What we are asking is how many of the genes that E has are also likely
to be found in D, and the answer is 62.5 per cent. If an animal that is a
OrmSb Y Sensation,
!
. 274343
Ormsby SensatIOn .
45th 442551 Shady . Maple Kmg
Newmont Mutual , Pontlac Hockster,
Orrnsby Lad, 219580
493889 Sir Mutual Scotia,
~ewmont Mutual 170631
Name, N.J.E.S. Mutual Scotia, 575310 { Isa Scotia Beets,
Ormsby Jewel Alice 315248
No. 654594 ormSb Y Sensation,
Born Dec. 1, 1931 . 274343
Newmont Ormsby
Jewel Alice,
1239022
Ormsby SensatIOn Sh d M I K
45th 442551
,
~ewmont Ormsby
a Y. ap e
219580 !
Ormsby Sensation
45th,442551
mg
Pontmc Hockster,
Jewel,1094461 { Clothilde Maid

Ormsby Bess
FIG. 145.-Holstein pedigree showing collateral relationship.
common ancestor of two related animals is itself inbred and, therefore,

more homozygous, a correction for this fact must be made in the formula
for relationship; and likewise, if the animal itself is more inbred than the
ancestor, a correction must be made. These corrections will be indicated
after we have discussed the coefficient of inbreeding.
Figure 145 shows the pedigree of the Holstein bull N.J.E.S. Mutual
Ormsby Jewel Alice. This bull carries 62.5 per cent of the "blood" of
Ormsby Sensation 45th getting 25 per cent from each of the latter bull's I
appearances in the second generation of the pedigree and 12.5 per cent
from his appearance in the third generation. Since the amount of
Ormsby Sensation 45th "blood" in the bull N.J.E.S. Mutual Ormsby
Jewel Alice is the average of that found in his parents, we could' find. ,
the amount in N.J.E.S. Mutual Ormsby Jewel Alice by averaging the
amounts in:,Newmont Mutual Ormsby Lad and Newmont Ormsby Jewel
\
\,'\
SYSTEMS OF BREEDING-RELATED ANIMALS 493
Alice. The former has 50 per cent of Ormsby Sensation 45th "blood"
and the latter 75 per cent, the average of the total of the two parents
being 62.5 per cent as noted above. N.J.E.S. Mutual Ormsby Jewel
Alice is 62.5 per cent; Newmont Mutual Ormsby Lad, 50 per cent; and
Newmont Ormsby Jewel Alice, 75 per cent related to Ormsby Sensation
45th.
Since the sire and dam of N.J.E.S. Mutual Ormsby Jewel Alice are
collaterally related, i.e., have one or more common ancestors, this bull
is inbred. He is likely to have gotten identical genes from Ormsby
Sensation 45th through three different lines, viz., through his sire, his dam,
and his maternal granddam. To find the relationship between the sire
and dam, we must brace all these lines and count the generations involved
according to our formula Rxy = ~*[(Y2)n+nll
From sire to Ormsby Sensation 45th = 1

From dam to Ormsby Sensation 45th = 1
From sire to Ormsby Sensation 45th = 1
From dam through maternal grand-
dam to Ormsby Sensation 45th = 2 .. Y21+2 = Y23 = 12.5%
37.5%
That is, Newmont Mutual Ormsby Lad and Newmont Ormsby Jewel
Alice are 37.5, per cent related (small correction needed for inbreeding
involved).
A simpler way of finding this relationship, since these animals are
collaterally related through only one common ancestor (Ormsby Sen-
sation 45th), would have been to multiply the
B
A /~~
percentage of Ormsby Sensation 45th "blood" in
each (50 X 75 = 37.5). This latter method be-
comes complicated, however, where there is more " 'D
than one common ancestor, especially if these are ~,/
related. c~--------o
FIG. 146.-Simplified
It may help to simplify matters by writing t he pedigree of Fig. 145.
pedigree of X.J.E.S. Mutual Ormsby Jewel Alice
in a little different form and using letters of the alphabet instead of the
actual names. This is done in Fig. 146.
The solid lines from D (Ormsby Sensation 45th) to Band C indicate
that both Band C are by D, and that some identical genes may have
reached A from D through these two different paths. The same pos-
sibility holds for some other identical genes reaching A from D through
the paths indicated 'by the broken lines.
* Greek letter sigma is used for summation.
494 BREEDING AND IMPROVEMEN1' OF FARM ANIMALS
Dir{lct relationship is what most breeders refer to' as percentage of

blo-;;ri.
-They sayan animal gets 50 per cent of its" blood" from its sire,
50 per cent from its dam, 25 per cent from each grandparent, 12.5 per
cent from each great-grandparent, etc., as illustrated in Fig. 147, together
,yith the figures of chromosomes below using cattle with 60 chromosomes
as the example.
Each parent in Fig. 147 is shown as having contributed 30 chromo-
somes to animal A. These 30 chromosomes were a random sample half
of the 60 chromosomes that each of these parents had. The figure shows
15 of them to have come from each grandsire and granddam. This
mayor may not have been true in any individual case. It ,vould be
l
Great great grandsire
· {Great grandsire
Gran d SIre 6.25 %
12.5% (3 or 4)
Sire 2~~ (7 or 8)
50% (dd) Great granddam
Gran am
(30)
25%
Animal A
(15)
100%
Grandsire
(60)
25%
Dam
(15)
50%
Granddam . "
(30)
25%
(15)
Total blood each gener-
ation ................ 100% 100% 100% 100%
Total chromosomes each
generation. . . . . . . . . . . (60) (60) (60) (60)
FIG. 147.-Showing most probable distribution of "blood" and chromosomes in a cattle
pedigree.
possible for the 30 chromosomes that the sire passed on to animal A

to have all come from the sire's sire (or from the sire's dam), thus in
genetic terms actually terminating one line of the pedigree at that point.
It will be noted that there are 60 chromosomes in each generation of .the
pedigree. In animal A they are found in their entirety. In the parental
generation, they are divided between two individuals, the sire and dam.
In the grandparental generation, they are divided among four individuals,
etc. This illustrates why it is generally not important to trace or con-
sider pedigrees for more than three or four generations, and also why it
is unnecessary to pay any particular attention to an outstandingly good
or poor individual that occurs farther back than the fourth or fifth
generation.
As far as direct relationships are concerned, "percentage of blood"
and degree of relationship are seen to be the same thing. It is a mistake,
however, to try to use "percentage of common blood," as a measure of

relationship between animals related collaterally. Figure 141 showed
two full brothers X and Y to have 100 per cent the same "blood" and
to be related by 50 per cent. Figure 143, on the other hand, showed the
two double first cousins A and B also to have 100 per cent the same
• "blood" but to be related by only 25 per cent because of t1_,_e additional
intervening generation.
Coefficient of Inbreeding.-As indicated earlier, inbreeding results
whenever related animals are mated. 'rVe have already seen that full
brothers and si~ters are related by 50 per cent; i.e., Y2 of their genes are
~ {~\=iI~
~:""J ~:::
~l~~
.. ~{~'
S~{~
.. '~{~
~~; ~
FIG. 148.-Pictorial pedigree showing relative size of ancestors in terms of their contribution
according to Galton's law.
probably duplicates (~:4: from each parent). Now, if a full brother and
sister are mated, we produce a new generation that is once further
removed from the common ancestors. Since each such r~oval halv~s
the genetic material, we can simply multiply the relationships bet,,'een
the full brother and sister (50 per cent) by Y2 and thus arrive at th;
coefficient of inbreeding f~r the new individua1~-'- -.
In other words, to find the coefficient of inbreeding of any animal, we
can first find the coefficient of relationship between its sire and dam and
then divide this amount~ 149 represents such a mating. \
Another way to do this is to count the number of times the genetic
material has been halved in getting to the ,parents of the inbred animal
from the common ancestor (or ancestors) and then adding 1 to the
exponent of ~~ to take care of the halving in getting from the sire and dam
to the inbred animal in question, This gives us the formula for inbreed-
496 BREEDING AND IMPROVEMENT UF FARM ANIMALS
ing suggested by Wrightl as

F" = ~[(H)n+nl+11*
This formula measures the reduction of heterozygosity in relation to

the foundation stock in the most remote generation of the pedigree being
considered. Figure 150 shows a half brother-sister mating, Band C
being 25 per cent related, because there is one generation from B back.
to the common ancestor D and also one generation from C back to D.
or
FIG. 149.-Full brother-sister mating.
The only common ancestor here is D. From the sire back to the
common ancest?r is 1 (n in formula) ; from the dam hack to the common
ancestor is 1 (nl in formula); and the 1 that must be added to take care
of the halving in getting from Band C to X gives us a total of 3 as the
exponent of 72, and 72 3 = >.g, or 12.5 per cent as the inbreeding coeffi-
cient of X. This figure 12.5 per cent simply means that this system of
mating has reduced the heterozygosity (or increased the homozygosity)
by that amount over what it was in the generation containing D, E, and
F. If the animals D, E, and F were 50 per cent homozygous and 50 per
B 5D
X
l
C~~
1E or
{>
FIG. 150.-S"hematic half brother-sister mating. f.
cent heterozygous, then animal X is 12.5 per cent less heterozygous or

.50% X 12.5 = 6.25% less heterozygous than they were, or only 43.75
per cent heterozygous. Looked at the other way, X is now 56.25 per
cent homozygous.
If a common ancestor is himself or herself inbred, this will, of course,
make a difference in the likelihood of its inbred descendant getting
identical genes from each of its parents. Being inbred means that the
common ancestor (A) is homozygous for more pairs of genes than is an
outbred one. In other words, two offspring of such an inbred animal
have less chance of receiving different genes from this ancestor and
1 WRIGHT, S., Coefficients of Inbreeding, Amer. Nat., 56:330-338, ]922.

* Complete formula will be found below. .
likewise more chance of receiving identical genes. This fact is provided

for by changing the formula for inbreeding from
to
In other words, if ~[(Yz)n+n'+11 was found in a given case to be 25

per cent, but the confmon ancestor was himself or herself inbred by 12.5
per cent, then we would have to multiply 25 per cent by 1.125 per cent,
making the final inbreeding 28.125 per cent.
A case of this sort is shown in Fi!!;. 151 with X 26.56 per cent inbred.
It is obvious in Fig. 151 that X is an inbred animal, for it resulted from
the mating of full brother and sister Band C. In other words, X is
more homozygous because it received some identical genes from D
through its sire and dam and also some from E through its sire and dam.
But D, one of the common ancestors, is also seen to have been an inbred
! FIG. 151.-Inbreeding involving an inbred common ancestor.
animal; in other words, D had more homozygous sets of genes than an

outbred animal because he received some identical genes from H through
both his sire and his dam. This being the case, D would be expected
to have fewer sets of heterozygous genes, one member of which would be
passed to X through its sire B and the other member of which would be
passed to X through its dam C. If D were not inbred, he might be
homozygous for certain genes, but, if he is inbred, he is undoubtedly
homozygous for more sets of genes. Since the coefficient of inbreeding
measures directly the probable increase in homozygosity, the inbreeding
of any animal must be multiplied by 1 plus the percentage inbreeding of
any common ancestors that are themselves inbred, in other words, must
be increased by the added amount of homozygosity due to the increased
homozygosity of its inbred common ancestors. We cannot, of course,
know the actual amount of homo- or heterozygosity for any animal. The
coefficient of inbreeding simply accounts for the added homozygosity
due to the opportunity an animal has of getting identi~al genes from any
ancestors that are common to its sire and dam.
Figure 152 shows the pedigree of an animal that has some repeat:,; ill
its pedigree but is not an inbred animal because it has no common
ancestor on both sides of its pedigree; in other words, its sire and dam are
not related.
It is true that animal B is inbred, since it resulted from a full brother-
sister mating. Animal B is 25 per cent inbred, 25 per cent more homozy-
gous than its immediate ancestors were, but B transmits to A only one-
half of its genetic material, one member of each pair of genes that it
has and there is nothing in the pedigree of A to lead us to suspect that B
and C will pass identical genes to A. Animal A, therefore, is not inbred.
By the same token, "'e might mate a highly inbred Holstein bull to a
highly inbred Guernsey cow. The resulting offspring would, of course,
not be inbred although both of its parents were, but crossbred. The
inbred Holstein and Guernsey would not be likely to pass identical genes
to their offspring, although since they both belong to the species cattle,
or A
{
""-".
I
/ FIG. 152.--Pedigree of an out bred animaL
they probably do have some genes in common and might, therefore, I

pass some identical genes to their offspring.!
A decline in heterozygosity should result automatically under continued self-·

fertilization Or closebreeding. Thus an individual heterozygous for a Bingle
factor pair Aa will produce, if selfed, offspring of which one-fourth are AA, one-
half A.a, and one-fourth aa. Under continued selfing the homozygous types
\
\ \ breed true and are progressively augmented by segregates from the dwindling
\ group of heterozygotes. The proportions of the three groups may be de~rmined

by any generation by the formula deyeloped by Mendel:
\ 2" - 1 A£1:2 £1a:2 n - 1 aa
where n is the number of generations of self-fertilization.

Similar formulas have been developed by which it is possible to compute the'
results of other systems of mating. Thus in brother-sister mating the proportion
of homozygosity pr~gressively increases but at .a slower rate, 6 generations of
1 From EAS1', E. ]\f., and JONES, D. F., "Inbreeding and Outhr('('ding," J. B. Lippill~
cott COIflPll,ny, Philadelphia, 1!)19.

\ .
•
\ ")
{99
!'\elf-fertilization being approximately as effective as 17 generations of brother-

sister mating.
This reduction in heterozygosityl takes place automatically in all other factor
pairs regardless of the number of pairs involved (Fig. 153). Thus continued
self-fertilization leads to a decline in heterozygosity for about seven generations,
about 99 per cent of homozygosity being attained in the seventh generation.
This reselilbies the actual results of inbreeding, for here there is a decline in size
and variability for about seven or eight generations, after which a line becomes
essentially stable.
Per Cent of Heterozygous

ndivlduClls 1M Each Selfed
neratlon when the Number
of Allelomorphs Concerned
Are; 1,5,10)15
;J'
~
~50r-~~-+~~~;-~+-~---~---+--~--i
.
....
, 25
I
"
,
) !~,t
0
0 10
FIG. 153.-Graph showing the reduction in the proportion of heterozygous individuals and
of heterozygous factor pairs b successive generations of self-fertilization. (After East and
Jones.)
All the evidence can best to interpreted, therefore, as indicating that the
reduction in size and vigor is related in some way to the attainment of homozygo-
sity and that the mere process of inbreeding, of itself, does not produce this
result. The automatic reduction of heterozygosity will occur, of course, only
if new mutations in inbred lines are very rare. If the mutation rate is high, this
may defer or prevent the attainment of homozygosity.
Coefficient of Relationship.--Because of the fact that inbreeding

makes an animal more homozygous, and, therefore, if said animal is a. -b
common ancestor or--t,vo -relateo- animals makes it ~re lill~ly that he
lSI~NOTT' E. w.,-a~:~:::0.7~,iiiiP{es of Genetics," 3d ed., McGraw-Hill

Book Company, Inc., New York, 1939.
500 BREEDING AND IMPRUVEMENT UF FARM ANIMALS
will pass identical genes_!? 90th descen~~nts, and also because inbreeding
makes a population more variable by producing separate inbred strains,
two corrections are needed in the formula for relationship to make it
scientifically correct. To take care of the fact that an i.gl>!ed_ common
a~gestor is more like~~_pa~sidentical ge_J1_es~~~o des~da~ec_~:USe
his inbreeding has made him ~oiYgous, tIlenumerator of the
relationship coefficient must have added to iC(C+F a ) as a multiplier,
as we have just seen; and to take care of the fact thatln~ding tends to
=t- create separate families, a denominator must be supplied in the form of
~ . I + F~, -I + F 11 making the complete formula for relationship between
animals X and Y.
R"" = ~[(Yz)n+nl. (1 + Fa)]
vI + F z ' VI + Fy
In the mating for instance, of a son to his own dam, the relationship
between his offspring and his noninbred dam becomes
75
v
% . In
1 + Fo
this case the most probable situation is that in half the pairs of genes in the
offspring both members will have come from the common ancestor. But
in half those cases the two genes will be duplicates of only one gene that
the common ancestor had, so that in one-quarter of the pairs of genes
the offspring will be homozygous, whereas the common ancestor \yas
heterozygous, although both of the offsprings' genes came from the com-
mon ancestor. This is illustrated in Fig. 154.
o{s{~ ROJ) =
75
VI + Fo = V'l
75
+ 0.25
75
= 1.1 = 68.2
D
FIG. 154.-Parent-oifspr;ng mat;ng w;th relationsh;p computed.
The real function of these corrections is to reveal the actual degree in

which the genotype of related animals are similar, rather than leave it in
terms of the percentage of genes from a common source. In the above
parent-offspring mating, it is most likely that 75 per cent of the genes of
the offspring 0 came from its dam and paternal granddam D, but since
the offspring is inbred, whereas the parent D was not, then 0 will have a
certain number of homozygous gene pairs that existed as heterozygous
pairs in the parent D.
Because there has been so little inbreeding practiced in our farm live-
stock, the denominator of the complete relationship coefficient seldom
becomes very much larger than 1.0 and can, therefore, be dispensed with
in gene~l practice.
Sl'STEil;fS OF BREEDlNG-RELATiW ANiMALS 501
The final form ofthe formula for direct or ancestor-desqendant relation-

ship becomes
The term under the square-root sign in this formula is necessary

owing to the fact that, if the ancestor is more highly inbred (more
homozygous) than the descendant, the coefficient of relationship will be
larger than the "percentage of blood," and vice versa. In most pedi-
grees of our farm animals, however, there is not likely to be enough
difference in the amount of inbreeding of ancestor and descendant but
that the portion of the complete formula under the radical can he dis-
pensed with.
The following table \\"ill be found convenient in working out inhreeding
coefficients.
TABLE 33.-ExPONENTIAL VALUES OF 72
W = J4 = 25.0%
W = Js = 12.5%
72 4 = J16 6.25%
72' = %2 3.125%
72" = Jt.4 1.5625 %
72' = H28 0.78125%
72 8 = 72 56 = 0.3\)0625 %
72 9 = %12 = 0.1953125%
In the mating of full ~_roth~_and sister we have seen that the coefficient
of inbreeding is 25 per cent.
In the case of continued full brother-sister matings for two generations

the inbreeding coefficient would be 37.5 per cent.
It should be observed that we trace back from B, the sire, to D, a
common ancestor, then down to D on the dam's ~ide of the pedigree, and
then forward to the dam. We do this also for E, another common ances-
tor in the second generation, and this completes all the ancestors in this
generation. N ow we move to the third generation where we find the
same two animals F and G in each line of the pedigree. When we traced
the paths through D a moment ago, we also took care of the F's and G's
502 BREEDING AND IMPROVEMENT OF' F'ARM ANIMALS
in these lines, since the likelihood of animal A getting identical genes

from D is taken care of in its entirety when we trace the D paths and is
not increased because of the F's and G's behind D. The same thing
applies to the E's in the second generation. However, since the presence
of F and G behind D on the top side of the pedigree and behind E on the
bottom side (and E on the top side and D on the bottom side) does give
additional opportunities for A to get identical genes through these paths,
these opportunities must be evaluated.
I
B{D~~
A{D~~ C
E~~
5F
E(G
or
A:X:X:25% 1272%
FIG. 155.-Brother-sister matings for two generations.
,
We have, therefore, in this two-generation full brother-sister mating
four common ancestors, D, E, F, and G, and they have the following
values.
D = ~~1+1+1 = %3 = 12.5%
E = 721+1+1 = 72 3 12.5%
F = H2+2+1 X 2 = Yz5 X 2 = G.25%
G = 722+2+1 X 2 = Yz5 X 2 = 6.25%
Total 37.5%
With three generations of full brother-sister matings, we would have
or
A:><:X:X:
3174% 12 72% 674:%
FIG. 156.-Brother-sister matings for three generations.
Here we meet the fact that some of the commdn ancestors (D and E)
are themselves inbred, so that our summation of all the different paths,
by means of which A might get identical genes from its common ancestors,
,
is \.
\
\
1\
S Y S T EMS OF BREEDING-R ELA T E D ANI MA LS 503
D = Yz. X 1.25 = 15.625%

E = Yz3 X l.25 = 15. 625 %
F = Yzs X2 6.250%
G = Yz5 X 2 6.250%
H = Yz7 X4 3.125%
I = Yz7 X4 3 . 125 %
Total 50.000%
In Fig. 156 it is evident that the sire B and dam C are connected
through two lines involving two generations, through four lines involving
four generations, and through eight lines involving six generations. In
FIG. 157.-Comet (155). The first bull ever sold for $5,000. He was very strongly inbred .
(From Sanders, Sh01·thorn emtle, The Breeder8 Gazette.r=--
addition, the parents of the sire and dam have one generation of inbreed-
ing behind them, or 25 per cent. Our formula becomes, therefore,
This shows that, assuming a 50 per cent heterozygous condition to

start with, an individual which is the result of three generations of full
brother-sister matings is now 75 per cent homozygous.
In full brother-sister matings one generation gives an inbreeding
coefficient of 25 per cent; two generations, 37.5 per cent; three generations,
50 per cent; ten generations, 88.6 per cent. In parent-offspring matings,
one generation gives an inbreeding coefficient of 25 per cent; two gener-
ations, 37.5 per cent; three generations, 43.8 per cent) approaching 50 per
cent as a limit.
/
504 BIlEEDING AND IMPROVEMENT OF FARM ANIMALS
SYSTEMS OF BREEDINC.--RELATED ANIMALS 505
From a practical standpoint, the obvious use of relationship is to

assess the probable merit of an untested animal by its relationship to
tested ones. If a cow produced 50 lb. of butterfat above the breed
average, her daughter 50 per cent related could be expected to produce
25 lb. above breed average. Other more distant relatives could be
assessed on their degree of relationship. This is practical within limits
but leaves out of account the sire involved and various effects of environ-
ment, dominance, and epistasis. We cannot know the complete genotype
of any animal and must of necessity consider the phenotype to be a more
or less accurate indication of genotype, which it mayor may not be.
As will be pointed out later, we are in great need of broadening our
basis of selection by considering family as well as individuality. In
this, in- or linebreeding to set off separate strains or famities and relation-
ship to help measure probable genetic merit should find an ever-increasing
usefulness.
Examples of Inbreeding.-Figure 158 gives the pedigree of the famous
Shorthorn bull Comet, the first bull ever sold for as much as $5,000.
As will be noted, the bull Favorite was bred back to his own dam
Phoenix and their offspring Young Phoenix bred to her sire Favorite.
Since Favorite is himself inbred, we shall first compute his coefficient of
inbreeding. His sire and dam are related through the bull Foljambe and
the cmy Favorite, our formula being
FA = ~[(~2)3] + [(Yz)4] = 18.75%

,.
As shown in the pedigree, Comet has four common ancestors, viz.,
Favorite (bull), Phoenix, Foljambe, and Favorite (cow), and our formula
becomes
F", = ~[(Yz)2 X 1.1875] + [(Yz)3] + [(Yz5)] + [O~26)]

= 0.2969 + 0.1250 + 0.0312 + 0.0156
= 46.87% coefficient of inbreeding for Comet
We can perhaps see this more clearly if we skeletonize the pedigree of
Comet and use letters instead of names as in Fig. 159.
It is evident from Fig. 159 that one of the common ancestors, B, is
himself inbred and that E, F, and H are also common ancestors.
Figure 160 gives the pedigree of the Jersey bull Sybil's Gamboge. The
sire and dam of this bull are more than half brother and sister, the sire is
the result of mating a three-quarter brother and sister, and the dam of
Sybil's Gamboge traces in all four lines to the bull Champion Flying Fox,
for he is the sire of both Gedney Farm Oxford Lad and of Agatha's
Flying Fox.
We notice that there are five ancestors in his pedigree (Fig. 160) that
are common to both sire and dam, viz., Oxford Majesty, Royal Majesty,
Gedney Farm Oxford Lad, Fontaine's Oxford Pride, and Champion
Flying Fox, and we also notice that Oxford Majesty, one of the common
ancestors, is himself inbred. We have, therefore,
Oxford Majesty. One chain of two generations, together with his
own inbreeding coefficient of 12.5 per cent. Contribution 01)3 X 1.125. ~
Royal Majesty. One chain of four generations (the other .chain having
been accounted for in working with Oxford Majesty). Contribution
(Yz)5.
, , or !
~.
/ , ,~
, .;
FIG. 159.-Skeletoniz:d pedigree of Comet.
:<~; "~I'
("
Gedney Farm Oxford Lad. Seven remaining chains of six generationJ;,;

Contribution 7 X ('YzF. I
Fontaine's Oxford Pride. One chain of six generations. Contribution'f.
(72)1.
\\ Champion Flying Fox. Four remaining chains of eight generations.
Contribution 4 X (%)9.
Total. Fx for Sybil's Gamboge = 24.22%.
Role of Inbreeding.-Now that we understand the basic principles
, underlying inbreeding, it will no longer surprise us that both good and,
harmful results arise from its use. Inbreeding in itself does not_create
any new genes, nor for that matter -does any system of breeding. All that
any system can do is to recombine the genes already present into new
combinations. With outbreeding, we tend to keep our livestockhetero-
zygous. With inbreeding, we make them more homozygous. If we
have a preponderance of good genes and gene combinations in,our stock
to start with, inbreeding plus selection will increase the good qualities
SYS1'EMS OF BREEDING-RELATED ANIMALS 507
" _.- , ~ .,._

508 E1UtEDING AND IMPROVEMENT OF PARM ANIMALS
we already have. If we have a preponderance of poor genes and gene

combinations to start with, inbreeding will increase the poor qualities
\ye already have.
What inbreeding actually ~_chara...ct~rs. These characters
may be either good, bad, or indifferent, and whatever they are, inbreeding
will not change them in the least. Inbreeding is no more potent than any
other system in creating new characters of any sort; in fact, it is less so,
for variation is reduced. All that inbreeding does is to take the genes
th~ present and 8_9rt !_hem out into homQzygous forms. For instance,
if an animal was carrying severa1 genes, or determiners, for characters
in the heterozygous form, i.e., one parent after reduction had put in AbC
and the other one aBc, the hybrid would carry Aa Bb Ce . If such an
animal were bred to another animal of similar genetic constitution, the

genes would tend to combine in all possible combinations from AA BB CC
to aa bb cc. Then if sufficient numbers were available so that rigid
selection could be practiced, the AA BB CC type could be selected ·out.
Supposing that in the foregoing, the capital letters stood for desirable
characteristics and the small letters for undesirable ones, one can see why
inbreeding may give either good or bad results, because AA BB CC
animals with no recessive, undesirable factors would be a decided im·
ment over either parent, whereas aa bb cc animals would be just the
reverse. It will bear repeating that inbreeding is powerless to create
either good or bad. It is, however, the very best method of finding out
what is present in a strain, for inql'eeding will sort out the various genes
into a homozygous condition. [)-'he desirable animals may then be
retained and the undesirable ones discarded.
\
SYSTEMS OF RREEDING-RELilTED ANIMALS 500
Animals might be carrying ~cessivy _genes for letha,UI!Q~Q~or lack

of fertility, for weak constitution, for inefficient production, etc. If ,,"e
outbreed sUCh animals to good sires, these undesirable recessives are
lIkely to TI;"main hidden and unsuspected by being covered up ~y dom-
in~nt gen~~om t~~ other parent. If we inbreed, however, we prOVlde-,
the opportunity for these recessives to combine in the homozygous state
genetically and to_ emerge as ph~!lotypes. The desirable animals may
then be retained and the undesirable ones discarded.·...-1£ a breeder has
an average or below-average herd, their inbreeding will probably result
in intensifying poor traits faster than good ones with inevitable deterio-
ration of his stock. The question is so~etimes askedas to w~at degree
of inbreeding i& pf.r!!1issible. There is, of course, no correct answer to'/'/
stIch a question. All one can say is that 25 per cent of inbreeding in
herd A might be beneficial while 5 per cent in herd B might be harmful,
depending entirely on the genetic complex of the two herds.
If a breeder thinks that he has a particularly good sire and really
wants to learn the truth, he can do so by mating this sire to 20 of his own
daughters,. as suggested by Wriedt. Comparisons between the imme-
diate daughters of this sire and the offspring resulting from breeding this
sire to his own daughters will provide the answer to the sire's real worth
as a breeding male, for it will reveal his genetic make-up by allowing.
both his dominant and recessive genes to appear in homozygous form.
Bakewell's Work.-Robert Bakewell, an Englishman, born in 1725,
was the first man as far as is known who dared to use inbreeding in 11.
constructive way. He worked with Longhorn cattle, Leicester sheep, and
Shire horses. It is said that he secured two heifers and a bull after a
great search for just what he wanted, and that his entire herd of cattle
was developed from these without any outside blood. His success as a
breeder was noteworthy, and is today well known. Culley, writing in
]794, said:
The great obstacle to the improvement of domestic animals seems to have

arisen from a common and prevailing idea amongst breeders that no bull should
be used in the same stock more than three years, and no tup more than two;
because (say they) if used longer, the breed will be too near akin, and liable to
disorders; some have imbibed the prejudice so far as to think it irreligious, and if
they were by chance in possession of the best beast on the island would by no
means put a male and female together that had the same sire, or were out of the
same dam. Mr. Bakewell has not had a cross for upward of 20 years, his best
stock has been bred by the nearest affinities, yet they have not decreased in size,
neither are they less hardy, or more liable to disorder, but, on the contrary, have
kept in a progressive state of improvement.
Since Bakewell's time, his example has been emulated by many

breeders. Breeds are best known by means of individuals and families,
and most of the latter have arisen as the result of intensifying the charac-
teristics of certain individuals thi'ough inbreeding.
Good Results from Inbreeding.-In practically any breed of livestock
it is possible to find animals that have resulted from rather intensive
inbreeding. As an example, Fig. 160 shows the pedigree of Sybil's
Gamboge, a famous Jersey bull.
It will be agreed that this is a rather closely inbred bull. The bull
is a double grandson of Imported Oxford Majesty, a triple great-grandson
of Royal Majesty, and a septuple great-great-grandson of Gedney Farm
Oxford Lad, as well as a triple great-great-grandson of Oxford Ixia 3d,
a double great-great-grandson of Lucy 12th, and a double great-great-
grandson of Fontaine's Oxford Pride. Royal Majesty founded the
Majesty family of Jerseys, which contains many notable show winners
and producers. "Get" of Sybil's Gamboge won 14 ribbons at the 1920
National Dairy Show, including one junior champion, three firsts, three
seconds, two thirds, two fourths, two fifths, and one sixth, as well as
second and fourth in the get of sire.
Sybil's Gamboge has 88 Registry of Merit daughters whose records on
a mature basis average 12,234 lb. of milk testing 5.12 per cent. The
best record to be found among these daughters is one of 19,239 lb. of
milk and 933 lb. of fat made by Sybil's Miss Mayas a junior'six-year-
old cow. Sybil's Gamboge also has 21 proved sons.
The late N. H. Gentry, a successful breeder of Berkshire swine, is
quoted by Mumford as having said:
If it is true that inbreeding intensifies weakness of constitution, lack of vigor,
or too great fineness of bone, as we all believe, is it not as reasonable and as certain
that you can intensify strength of constitution, heavy bones, or vigor, if you have
these traits well developed in the blood of the animals you are inbreeding? I
think I have continued to improve my herd, being now able to produce a larger
percentage of really superior animals than at any ti!lle in the past.
The foregoing examples illustrate the fact that not only during the
foundation period of the breeds was inbreeding a potent factor for
improvement but that it still retains its potency.
Harm from Inbreeding.-In exactly similar manner, one might unearth
opinions and examples, both in the past and in the present, that would
indicate that inbreeding is al! ~e~tLeJ1lely dangerous practice and that
anyone using it is liable to encounter serious difI.i_c_~s. This will not
be done, however, for there are many references in the literature of ani-
mal husbandry ,as to the harll![,:!!.!~~_f_r_~t!ce of inbr~e<:ling.
\ - ~~;N'v
\ v ~
! \
SYSTE.n;IS Of' BREEDING-RELATED ANIlIJALS 511
Experimental Data.-Much experimental work has been done with a

,'iew to illuminating the problem of inbreeding. Castle saysi in regard
GO some inbreeding experiments with Drosophila:
My pupils and I bred brother and sister for fifty-nine generations in succession
without obtaining a diminution in either the vigor or the fecundity of the race,
which could with certainty be attributed to that cause. A slight diminution was
elbserved in some cases, but this was wholly obviated when parents were chosen
from the more vigorous broods in each generation. Nevertheless, the crossing
elf two inbred strains of Drosophila, both of which were doing well under inbreed-
ing, produced offspring superior in productiveness to either inbred strain. Even
in this case, therefore, though inbreeding is tolerated, crossbreeding has
!1dvantages.
Bos and "Weismann inbred rats for about 30 generations and secured a
diminution in both fertility and vigor. King also inbred rats for 25
generations and, according to Hays,2 found that:
Inbred males in seventh to fifteenth generations were 18 per cent heavier than
check males not inbred. Inbred females were about 3.7 per cent heavier than
controls at one year old. The variability in both weight at maturity of the
fifteenth generation of inbred males and females was about 40 per cent less than
among the control animals. King (1918) also studied the effect of inbreeding on
fertility, the time of puberty, and the longevity of albino rats inbred by brother
and sister mating for twenty-five generations. The average size of litter of all
inbred matings was 7.5 rats; and the average litter size for controls 6.7 rats.
Inbred rats seemed to re&ch sexual maturity and to breed slightly earlier than
stock females not inbred. . .. The number of sterile animals is not given, but
King states that inbreeciing did not decrease the productiveness of the animals.
Inbred rats lived longer than control animals not inbred. The value of selection
in inbreeding animals is concisely illustrated in this pain'staking and thorough
study of the question of inbreeding.
Reporting on his extensive investigations, Wright says:3
There has been an average decline in vigor in all characteristics during the
course of 13 years of inbreeding of guinea pigs, brother with sister. The decline
is most marked in the frequency and size of litter, in which it is so great that it
would have to be accounted for even though the decline in other respects was
assumed to be due wholly to a deterioration in the environmental conditions.
The decline is greater in the gains after birth than in the birth weight, and greater
is the percentage raised of the young born alive than in the percentage born alive.
1 CASTLE, 'V. E., "Genetics and Eugenics," p. 221, Harvard University Press,
Cambridge, Mass., 1916.

2 HAYS, F. A., Inbreeding Animals, Del. Univ. Agr. Exp. Bul. 123, p. 15, 1919.
3 WRIGHT, S., The Effects of Inbreeding and Crossbreeding on Guinea Pigs, U.S.
Dept. Agr. Bul. 1090, 1922, pp. 31-32.

512 BllEEDING AND IMPROVEMENT OF FARM ANIMALS
The ability to raise larger litters has fallen off much more than ability to raise
small litters.
A comparison of the inbred guinea pigs with a control stock, raised under
identical conditions without inbreeding, and derived in the main from the same
linebred stock as the inbred families, indicates that the inbreds have suffered a
genetic decline in vigor in all characteristics. The decline in fertility is again
shown to be marked. Experimental inoculation with tuberculosis has shown
that the inbreds were inferior, on the average, to the controls in disease resistance.
A study of sex ratio yields results in marked contrast to those obtained in connec-
tion with the other characters. There are no significant fluctuations from year
to year, no contrast between inbreds and controls, and no indications of change
due to inbreeding.
In addition to the points brought out in this bulletin, which indicates genetic
decline during inbreeding, extensive experiments have been made in which
different inbred families have been crossed together. These are described in
another paper (U.S. Dept. Agr. Bul. 1121) in which it is shown that crossbred
guinea pigs born of unrelated inbred parents are distinctly superior to their
inbred relatives in nearly all elements of vigor. A slightly larger percentage are
born alive, in small litters at least, and a distinctly larger percentage of those born
alive are raised. The young are slightly heavier at birth in a given size of litter
and gain much more between birth and weaning. They mature earlier, produce
larger litters and produce them more regularly than inbreds. Of the young which
they produce, a much larger percentage are born alive, especially in large litters,
and even more of these are raised than in the first generation. Their young
show a further increase in birth weight and in later gains.
It is believed that the results point the way to an important application of
inbreeding in the improvement of livestock. Nearly all of the characteristics.
dealt with here, like most of those of economic importance with livestock, are of
a kind which is determined only to a slight extent by heredity in the individual.
About 70 per cent or the individual variation in resistance to tuberculosis and
over 90 per cent of that in the rate of gain and size of litter are determined by
external conditions. Progress by ordinary selection of individuals would thus be
very slow or nil. A single unfortunate selection of a sire, good as an individual,
but inferior in heredity, is likely at any time to undo all past progress. On the
other hand, by starting a large number of inbred lines, important hereditary
differences in these respects are brought clearly to light and fixed. Crosses
among these lines ought to give a full recovery of whatever vigor has been lost by
inbreeding, and particular crosses may safely be expected to show a combination
of desired characters distinctly superior to the original stock. Thus a crossbred
stock can be developed which can be maintained at a higher level than the original
stock, a level which could not have been reached by selection alone. Further
improvement is to be sought in a repetition of the process-the isolation of new
inbred strains from the improved crossbred stock, followed ultimately by crossing
and selection of the best crosses for the foundation of the new stock.
This Ipethod of improvement has not been unknown in the past. In faet,
most of the rero~nized hreeds of livestock were developed, more or less UllCOIl-
•
;
sciously, in this 'Way. Close inbreeding was practiced by the pioneer breeders--
Bakewell, the Collings, Bates, Cruickshank, Hewer, etc. The relatively few
promising families and the successful nicks between them were the foundation
stock of the breeds. Further development may be expected by the intelligent
application of the same principles.!
Combining the results de~cribed in this and in Bul. 1121 there seems no escape
from the conclusion that a loss of vigor, especially in fertility, took place as a more
or less direct consequence of close inbreeding. The question whether this is an
inevitable result of inbreeding or merely a likely one, as well as other phases of
the subject, will be discussed after the presentation of further data?
Hughes 3 has recently reported the results of an interesting inbreeding

experiment with Berkshire swine. The pedigree of a litter born in 1931
is shown in Fig. 162.
It is evident from the pedigree that full brother-sister matings were
made for four generations, i.e., there are only two animals in each of
these four generations. Enhancer's Leader 2d and Enhancer's Leader
Belle 2d were by the same sire, Enhancer, and out of half sisters by the
same sire. There is enough additional inbreedmg further back in the
pedigree to give these pigs a total inbreeding coefficient (Wright) of 74
per cent.
Hughes giyes the following summary of the results thus far achieyed:
1. The average number of pigs farrowed, in the inbred litters, has been greater
(9.78 pigs per liter) than the average for the Berkshire herd (8.14 pigs per litter for
the years 1919-1926).
2. There has been a slight but gradual decrease in the number of pigs farrowed
in the inbred litters since 1923. The last litters, however, were the first litters of
the sows in question, which tended to be smaller.
3. The types of the pigs in all the inbred litters has been similar.
4. There has been no noticeable color change or structural abnormalities in any
of the inbred pigs.
5. The results obtained thus far seem to agree in part with those of Miss King
(no noticeable loss in size or vigor) and in part with the results of other workers
(there has been a slight decrease in size of the inbred litters).
6. The standard deviation from the mean of the inbred litters is 2.56 with a
probable error of ±0.288. The standard deviation from the mean of all Berk-
shire litters from 1919 to 1926 inclusive is 5.34 with a probable error of ± 0.302.
Woodward and Graves (1933) inbred grade Guernsey and grade
Holstein cattle-females mated back to their O\\"n sire for successive
generations, etc. In the Guernseys the birth weights declined, some
1 Ibid., Bul. 1121, p. 49.
2 Ibid., Bul. 1090, pp. 31-32. .
3 HUGHES, E. H., Inhreeding Berkshire Swine, Jour. Herpd., 24(5) :200, May, 1933.
deformities appeared, mature weight declined, amount of milk production

remained about the same, fat percentage increased. In the Holsteins,
birth weights declined, vigor declined, mature weight declined-amount
of milk production increased, fat percentage decreased.
Double Leader 4th
Sir Double
Leader 2d
Double Leader of U.F.
350282
l 331377
Double Leader Belle

331378
Enhancer's
Leader 2d
318540, Univ.
350285 of Calif.
Double Leader 4th
I
331377
Double Leader Belle
of U.F., 350285
Inbred litter Double Leader Belle
(8 pigs) 331378
Berkshire,
Mar. 24, 1931
Univ. of Calif.
(
Double Leader 4th
331377
Double Leader of U.F.
350282
Double Leader Belle Enhancer's
331378 Leader Belle
Miss Double 2d
Leader Belle 318539, Univ.
550285 of Calif.
Double Leader 4th
331377
Double Leader Belle
of U.F., 350285
Double Leader Belle
331378
FIG. 162.-Pedigree of inbred pigs.
Woodward and Graves (1946) make the following statement in their

summary and conclusions:
Generally speaking farmers should not practice inbreeding . . . . An important
exception to the general rule should be noted. If a farmer has an outstanding
cow or bull, he is justified in inbreeding to the cow or bull in order to obtain a
bull or bulls for breeding that will carry a high percentage of the characteristics
of the outstanding cow or bull. This course is justified because the fertility of·
the inbred bull is not likely to be seriously impaired and any lack of size and vigor
in the inbred bull is not likely to be transmitted to the offspring if he is mated to
unrelated animals. Continued inbreeding of whole herds, however, is almost
certain to be disastrous.
Bartlett, Reece, and Mixner (1939) have inbred Holstein cattle at the
New Jersey Agricultural Experiment Station with lethals and deformities
appearing in some families, very poor type and red color in other families,
good results in some families. They stress the point that rigid selection
must be practiced along with inbreeding.
Bartlett et al. (1942) reported no harmful effect of inbreeding (up to
20 per cent) on the birth weight, rate of growth, and type of dairy cattle,
provided one starts with genetically superior animals and employs rigid
selection.
Margolin and Bartlett (1945) from further studies arrive at the follow-
ing conclusions:
This paper further confirms earlier work demonstrating that Holstein-Friesian
dairy cattle can be inbred without necessarily causing a decrease in body weight
or size at any stage, from birth to maturity, as compared to outbred controls
of the same blood line, provided the Wright coefficient of inbreeding does not
exceed 0.20. Females inbred to a coefficient greater than 0.20 develop normally
to approximately first calving age, but show markedly abnormal development
thereafter. However, the animals in this group with growth records complete
to 72 months of age do not vary significantly in mean weight, height and heart
girth from the Ragsdale standard, although they are considerably smaller than
the outbred controls.
Tyler et al. (1946) found considerable variability in the effects of

inbreeding on growth and production of Holstein-Friesian cattle, and
their final conclusion is that, "The variation in these partial regression
roefficients between sires was large enough that the offspring of some
sires might be inbred as much as 25 per cent without any apparent
decrease in either body size or milk and butterfat production."
Baker et al. (1945) from a study of 88 daughters of the only Holstein-
Friesian bull used in the University of California herd for a period of
13 yeats-24 of his daughters were outbred, F = 0; 8 had an inbreeding
coefficient of 12.50 per cent; 27 are of 25.00 per cent; 10 had a coefficient
of 31.25 per cent; 13 had 37.50 per cent; and 6 had an average F of 41.66.
These investigators' conclusions are as follows: "By means of suitable
mathematical and statistical techniques, we have demonstrated a signifi-
cant, proportionate per cent decrease of size in height, weight and heart
girth among the daughters of Bear Valley Ormsby Esther, 518683, with
increasing coefficient of inbreeding."
51ti BREEDING AND IMPROVEMENT OF FARM ANIMALS
Regan et al. (1947) analyzed calf mortality in the 25-year-old California

dairy-cattle-breeding experiment and conclude as follows:
Class I, the control group, contained all animals whose coefficient of inbreeding
was 0 to 0.1249; Class II from 0.125 to 0.2449; Class III from 0.245 to 0.3749;
and Class IV 0.375 and over. . . . The percentage of abortions in the control
group was just as great as in the highest classes of inbreeding. It is concluded,
therefore, that the degree of inbreeding has little or no effect on the number of
abortions . . . .
Both genetic and environmental factors were found to influence mortality.
Total mortality, which includes abortions, stillbirths, and deaths after birth
up to 4 months of age, increases with the degree of inbreeding in Jersey females.
This is also true for Holstein female calves, but, because of fewer numbers and •
certain influences of the environment upon mortality, the effect .is less marked.
The mortality to 4 months of age for calves born alive increases with an increase
in the degree of inbreeding. . . . Individual bulls sire progeny of a characteristic
birth weight-light, medium, or heavy-but there is no evidence of a correlation
between characteristic birth weight and mortality.
Within the same breed, sex and class of inbreeding, when the progeny of all
the sires are combined, there is no significant difference in the mean birth weight
between calves that lived and calves that died. Calves with extremes in birth
weight, either light or heavy, succumb more readily than calves near the .mean
weight . . . .
The percentage of mortality varied among the inbred progeny of the different
sires. Inbreeding apparently increased the mortality in the progeny of all sires,
but the increase over the controls was not significant in the progeny of three of
the sires. The mortality among the inbred progeny of all the other sires that
had sufficient numbers of offspring was significantly greater than in the controls.
It was tentatively concluded that two different lethal genes, conditioning anoma-
lies of the liver and heart, but with no external morphological effects, may have
been responsible for deaths in the inbred progeny of two of the sires. These two
genes, however, are insufficient to account for all the mortality observed. It
seems that more subtle genetic relationships or interactions may be involved in
causing most of the mortality.
In inbred calves, the proportion of mortality that cannot be attributed to
specific lethal genes, was greatly influenced by management and medication.
The mortality was reduced by keeping the calves on fresh, clean ground that
had been free from Mttle for several months and by the use of sulpha therapy in
the treatment of certain infectious diseases.
Dickerson et al. (1947) studied the effect of inbreeding on performance

in swine and found that for each 10 per cent increase in inbreeding there
was an average decline in litter size of 0.2 pigs at birth, 0.4 pigs at 21
days, and 0.5 pigs at 56 and 154 days; in pig weight no decline to 56
days, b.~t 3.6 lb. at 154 days.
"\
Winters et al. (1948)1 report that performance in inbred swine has not
deteriorated with inbreeding. "The advance of inbreeding has not been
accompanied by any noticeable decline in any of the factors of perform-
ance in the lines as maintained." These workers attribute these results
to (1) rigorous selection on the basis of performance and (2) a flexible
system of mating best to best within lines rather than a rigid pattern
of full brother-sister or one-sire herd scheme mating.
t Use of Inbreeding.-Inbreeding has been used very little in animal
breeding up to the present. 13a,~ewell practiced inbreeding to some
extent in England during the eighteenth century, and, when other
breeders took up the practice, the foundations for our present breeds
,,'ere laid. During the past 100 years, however, little animal inbreeding
has been practiced. The larger and more influential breeders have used
inbreeding to a limited degree, but the belief that there is something
i~~_!!y_p.3:~~ful in inbreedingyll!§. thejustified prohibitions ~~~~st
it i_n the humar;. baye gre~tly restncted_ its us~ and in consequence many
possi1:)ilities of more rapid progress in breeding were lost .•
In 1905 G. H. Shull at the Station for Experimental Evolution of the
Carnegie Institution at Cold Spring Harbor, N.Y., and E. M. East at
the Illinois Experiment Station began the experimental inbreeding and
hybridization of corn" This work has now developed to the point where
the majority of field corn produced is "hybrid" corn. The procedure
involves the inbreeding of corn for seven or eight generations, by which
time the strain is practically pure or homozygous. Rigid selection is
practiced, and the weakest strains discarded. There is always a dec,r'e::.,se
in yield in these inbred strains. After the best surviving strains h:ne
been purified by inbreedil1&....tb.ey_are crossed, hybridized, in an experi-
mental way to determine which inbred strains best complement each
other. Many systems of hybridizing are in use, including the single
cross between two inbred strains, the three-way cross of a third inbred
on the result of crossing two other inbred strains, double crosses between
two single crosses involving four inbred strains, and top crossing of an
inbred strain on a commercial variety. All these systems have certain
advantages and disadvantages, the double-cross system being the one
most generally practiced.
This hybrid corn has been compared with the mule hybrid in animals.
Both of them are excellent in themselves but genetically poor from a
.breeding standpoint, the mule, of course, seldom being fertile anyway.
Only the first generation of hybrid corn should be used for commercial
planting, as the second generation generally drops in yield by from 10
to 25 per cent. Properly produced and tested hybrids that are carefully
1 Minn. Agr. Ext. Spec. Bul. 400.
511f' BREEDING AND IMPROVEMENT OF FARM ANIMALS
selected for their adaptability for certain regions will yield, on the average,
about 8 to 10 per cent more than commercial seed-corn varieties.
The corn breeder is thus analyzing his materials, finding out what they
are genetically through inbreeding and record keeping, and trying to fit
these known strains together into superior strains. He, of course, has
a big advantage over the animal breeder in that self-fertilization can be
practiced, making the inbreeding twice as intense as is possible with ani-
mals, his number of offspring is many times that possible with even the
most prolific animals, and the cost of producing a corn plant is much less
than that involved in producing a colt, calf, lamb, or pig.
The man with just average livestock should, of course, not practice
inbreeding. The fact that his stock is average means that it has a-g-oodly
share of undesirable genes, which inbreeding would make homozygous
and therefore worse. For average livesto~ ~utcrossing would seem
to be a better system, for unrelated animals wouIa have1es~d
of carrymg Identical poor genes.
"'"The man wIth better-than-average livestock should do some inbreed-
ing. The fact that his stock is above the iVerage meritclthe breed
'ineans that a random-selected, unrelated animal will have a tendency
to pull his herd or flock back toward the breed average. Since many
breeders still shy at the word inbreeding, we will say that th~n
average breeder should practice linebreeding, which of course is a form of
inbreeding, but a term whiclid.~ha~e the same fearsome connota-
tions as inbreeding.
Every breeder should try to buy his males in the best available family
in his breed. In the present state of our knowledge, the genetically
best families in any breed are not easy to know, and one must be wary of
advertising and "ballyhoo." Assuming one does buy a sire in a really
good family, then, generally speaking, he should try to stay in that
family, intensify its good qualities through linebreeding, and weed out
the poor genes.
Linebreeding.-Man bre del's fear ci,osebreeQillg (which they often
ut favor line reeding. This is sound practical
judgment. Probably only a tiny per cent of our herds could bep.efit
from closebreeding, a much larger per cent could benefit from linebreed~
ing.- What the breeder is constantly trying (albeit unknowingly in
many instances) to do is tq re lace poor genes with good ones-improve
his stock. When he gets a great SIre proved so by hIS oTfspring) all
would be lovely if that sire could be kept alive indefinitely, or at least
his testicles through transference to a series of young sires.. But even-
tually he will die and direct samples of his inheritance will be no longer
available. Now it is obvious that we can get the closest approach to his
good germ\cells from his offspring. So the intelligent breeder uses his
\
\)
sons and grandsons and thereby practices ljnebr~'H;liRg. In the last
cnapter we warned that there was no J;agic in out breeding systems, as,
for example, crossbreeding. Again let's remind ourselves that there is
no magic in inbreeding systems, as, for example, linebreeding. Some-
times breeders set out deliberately to linebreed to certain animals. They
draw up a gedigreuf an animal to be born 10 years hence by mating
certain present animals in line- or closebreeding systems. Often these
plans are based on present phenotypes rather than genotypes, and the
whole plan is pedigree or paper breeding. These plans seldom work out
satisfactorily.
Such general plans can be made to work out very satisfactorily if
they are soundly based on present animals' genotypes and enough animals
so bred according to plan so that many ca:ii'1ie tested" and those that meet
the test used to continue and build the ~he plan.
Linebreeding should be practiced in only better-than-average herds,'
seldom if ever, in a~ or grade herds.,. but perhaps even here when
such breeders turn up with an outstanding sire. In any case, it must
be done intelligently and in the light of all the records of performance
possible. Some deterioration. is to be expected from all inbreeding
because even the best of our ammals seem to carry some undesIrable
traits and these as well as their good traits will be intensified (become
ho~golLS). It, therefore, b~comes a race between selection and
deterioration. Obviously, if a breeder is to win such a race, he must
start with good stuff and select intelligently. The annual 500-mi.
Memorial Day race at Indianapolis is always won by a good machine
(proved good by rigid test, not necessarily by its appearance) piloted by
a skillful and daring man who selects his course and his varying speeds
intelligently and, to carry our analogy further, who goes to the pits ,,,hen
necessary for an outcross of oil, gas, or new tires.
Linebreeding is not necessarily measuredgy .,!:he inbreeding coefficient
as is illustrated in Fig. Ip3.
Three of the pedigrees in Fig. 163 show the same amount of inbreeding,
but the percentage of "blood" from a certain animal is very different.
If we are ever fortunate enough to own a really good sire, we will want
to keep his inheritance as intact as possible in our herds.
If we have l', good sire, B, his offspring are good presumably, because
they get one-half of their inheritance from B, in other word,B's inherit-
ance has been halved just once. "
Now, if we get another sire A, unrelated to B, then
A's offspring {A {~
W {B
B's granddaughters B's offspring F d t· . 1
oun a Ion anIma s
B's inheritan..c.e has been halved once more; it is only one-half as strong
now, and will continue to be further diluted if we use lines unrelated to B.
When we get a good sire, we should try to keep our herd related to him
as much as it will stand, and this, of course, involves some degree of
x)B 1 E~~
D~H
F1~
C
1jK
GIL
A = 12.5 % inbred A = 12Yz % inbred
25 % "blood" from II 50% "blood" from M
25% "blood" from I 25% "blood" from N
25% "blood" from Ie 25% "blood" from P
25% "blood" from L Line-bred to M
Not line-bred
zl:[ 1:1:
A = 12Yz % inbred A = 37.5% inbred
62.5% "blood" from M 87.5% "blood" from M
25% "blood" from D Very strongly line-bred to M
12.5% "blood" from X
More strongly line-bred to M
FIG. 163.-Pedigrees to illustrate linebreeding.
inbreeding. The following diagram illustrates how a herd or flock might

be made to retain its 50 per cent relationship to a good sire B.
B's double grandda~ghterS{Son of B 50 % related to B {~

12Yz % inbred {B
50 % related to B Daughters of B 50 % related to B F
It is evident that the above scheme involves some inbreeding, actually

some linebreeding. Each animal born is unique as far as genetic make-
up is concerned, i.e., there never has been, nor can there ever be, another
animal with an identical genetic make-up. When a great sire dies, we
can only retain a degree of relationship to him as strong as the relation-
ship evidenced by any of his living descendants. If we have animals
that are 50 per cent related to him and mate them, the herd will remain
50 per cent rel\l.ted to the original good sire, but this relationship to him
SYSTEMS OF BREEDING-RELA'l'ED ANIMALS 521
cannot be increased after he has died. If we have animals that ar~

per cent related to him and mate them, the herd will continue to be 75
per cent related to the good sire. Even five generations of breeding sire
to his daughter will only give us an animal 96.875 per cent related ;Q the
original s i r e . · i./
Linebreeding is most generally pointed to some great sire because'
generally sires through their many offspring become more widely known
than do dams through their few_offspring. Linebreeding can be pointed
to a great female. We think of one well-known and deservedly popular
herd whIch was-· f(1unded quite largely on one female. She was a great
female both phenotypic.?Jly ~g_genDt¥Pically. She had five daughters
as good as herselCand this breeder has used two of her sons as herd sires
and two of 4~r_g£~I_!_~. The herd is becoming strongly linebred to this
old female and is one of the top herds of its breed.
Closebreeding.-This is the final degree of close~es" in breeding
"tarting as we did in the previous chapter with the widest system pos"ible,
hybridization, and working up through grading, crossbreeding, out-
crossing, and in this chapter linebreeding. Closebreeding is the mating,
of parent and offspring or flI11 brother and sister. This is the most
potent toot ~dering the genes homozygous. All the
c~ipulated for linebreooing apply to cIosebreedmg and with
added emphasis. There, no doubt, are a few animals in every breed
which could be closebred, with rigid selection, to advantage. To know
unfailingly which these are is difficult. Certainly it can seldom, if ever,
be achieved through appearance (phenotype) alone. Closebreeding must
be practiced in only the best herds and only on the basis of complete
records of performance, our onll approach to genotype.
O"ne place where closebreeding would find greater advantageous use is
in the production of sires. When a breeder finds himself with a truly
~t sire, he often look;£ar afield for his successor, when a better choice
is in his own hands. He cannot possibly know as much about some
distant herd as he can about his own and he always runs the risk of an
unfavorable "nick." Such a breeder might breed his great sire to a few
of his own daughters which are members of good cow families and get
some males to be tried out as successors to their great sire. This is close-
breeding, and thes90ungsires wg! have 75 per ce_nt of the "blood" ~f
the old sire. Such a bull can be bred to females in the herd which are
not desCend;;,;;ts of the old sire without any inbreeding at all. If he is
used on granddaughters or daughters of the old sire, the offspring will
be 15 to 18 per cent inbred, in other words, linebred, and whether good
or bad depends on the validity of our original judgment as to the genetic
worth of the old sire,
The obverse side of the above picture could be had by breeding the son
of a "great" cow back to her to get a sire.
Conclusion on Inbreeding.-From the foregoing, one is forced to
conclude that inbreeding in itself is powerless to create either good or bad.
If good or bad is present in theg;;rm plasm, inbreeaing' will sort ICout
and bri~g it to light. Closebreeding may be used in the better herds
a~(Cflocks if acc~mpanied by very rigid selection. However, line-
breeding, a less intense form of inbreeding, is decidedly safer. I:his
consists of mating animals of the same general line of descent. Because
-~yC:~'eo(the same line, they have the same general type of germ plasm
and so exhibit the same general set of characteristics. In other words, a
line bred herd or flock is extremely un~f~rrn' a characteristic almost •
invaluable (if of a good sort) in animal breeding. In a general way,
closebreeding has the greatest permanent genetic possibilities both for
good and for harm, whereas outbreeding or mating of animals of no
degree of kinship has perhaps the least permanent genetic possibilities
for good and the least danger of producing harmful phenotypic results,
provided desirable animals form the outcross. It is conceivable, of
course, that two animals showing no pedigree relationship for several
generations back might have identical genes for one or more character-
istics or potentialities, and, that if these two animals were mated, it
would, as far as their like genes are concerned, be tantamount ~o inbreed-
ing. Most people think of inbreeding in terms of the names in a pedigree.
Since we can see the names on a pedigree and cannot see the genes in our
animals (but only their effects and the picture may be distorted through
environmental or genetic influences), this is perhaps the most intelligent
way to think about it. But we should keep in the back of our minds the
fact that inbreeding is concerned primarily with the genes. If the
homozygous state is equal to or superior to the heterozygous state, then
we can have as good or better animals through inbreeding and in addition
animals which ,vill transmit uniformly.
If a breeder is keen enough in mating his animals and severe enough in
\ culling, he may practice closebreeding. Such pioneering requires both
eourage and skill. The majority of breeders ought perhaps to seek the
middle road, the happy medium. Their desire is to progress, but they
eannot afforg to take great chan:ce~ For such breeders, both therr;
hop~-~~d'their ~afety lie in linebreeding, which is an attempt to secure a
greater degree of homozygosity of certain desirable genes without running
too great a risk of intensifying undesirable ones. This result can be
accomplished in a variety of ways.
It is perhaps easier to line breed to an outstanding male rather than
to an outst~nding female because of the greater number of offspring of th£)
\
'\ \
SYSTEMS OF BREEDING-RELATlW ANlMALS 523
former. If one found himself in possession of a really great sire (proved

great by his offspring) with many sons and daughters of this sire, he
could select two or more of the best sons to mate with the daughters,
balancing defects, if present, in the particular matings made. The next
generation could be formed by mating the daughters of one sire to a son
of another, etc. There is, of course, some danger in the first matings of
half brothers and sisters, but, if this stage is passed successfully, there
should be no great future danger. The second generation would be
double grandsons and granddaughters of the original sire, and the third
generation would be quadruple great-great-grandsons and -daughers of
the original sire. The fourth generation and later ones would also show
some inbreeding to the sons, grandsons, etc., of the two original sons used.
These amounts are very small, however. This system should gradually
render the genes from t4e original sire more homozygous, and in general
half of the genes of the future herd would be those from the original sire.
Pearson and Lush 1 have recently pointed out that essentially this system
is being used with marked success by C. C. Good of Ogden, Iowa, in hiB
breeding operations with Belgian horses.
Lush has reported 2 on a herd of beef Shorthorn cattle bred for 20 years
wit~ new blood" with the merits of the early good ancestor reasonably
well conserved, with the whole herd nearly as like each other genetically
as ordinary full sisters, and with the herd still maintaining a reasonably
high average of individual merit." A composite pedigree of this herd is
shown in Fig. 164.
Another system of linebreeding would result from mating the daughters
of a sire to a son of the same sire but out of an unrelated dam, the resulting
daughters to be mated to a son of the sire of the daughters again out of an
unrelated dam. In this way one is always mating half brothers and
sisters and producing. double granddaughters of certain bulls. However,
outside" blood" is being i~troduced by selecting the sires from unrelated
dams. Various other types and combinations of linebreeding systems
are, of course, possible, the aim in all of them being to intensify certain
characteristics by rendering them homozygous, at the same time weeding
out the homozygous undesirable characteristics. The success or failure
of any system of breeding lies first in the genes that the original animals
have and second in the breeder's ability to balance defects and mate
wisely-in other words, his ability to select. From the genetic viewpoint,
I'!ome form of inbreeding would be of greatest advantage in cases where
1 PEARSON, P. B., and LUSH, J. L., A Linebreeding Program for Horse Breeding,
Jour. Hered., 24(5): 185, 1933.

2 LUSH, J. L., A Herd of Cattle Bred for Twenty Years without New Blood, Jour.
Hered., 25(6')".209-218. June, 1934.

524 BRB'EDING AND IMPROVEMENT OF FAR}pI ANIMALS
epi,;tasis is involved. In effect epistasis means that the goodness derives

from a certain combination of genes. If we outbreed in such a situation,
we tend to break up and scatter these desirable combinations of genes.
The only way to tend to hold them together in future offspring is through
some form of inbreeding.
Our herds were probably formed by selective matings (likes or unlikes)
to get desirahle combinations of qualities followed by inbreeding to
17-::============~
20,i>
16~~------------------------
2;~~=======
. I ..;o...~~-,-----c----=
~~ ~";":~k---::::::::::::~~
28
2
6
iii
29
!~.
11:;~~~~f~~r~~:=:::::~~::=:=:::='=~7~~""""-
I~~ r o o/5
:;~~=====::eB:E:EAWUrr?=::::::::vCHAMPION
"i---------CHAMPION BEAIJTY 4TH.
BEAUTY2""~:::U1V_IMPROVER'S BEAIJT/
FIG. 164.-Five generations of inbreeding. Pedigree of the entire young herd late in 1932
complete back to the time of Sultan's Banner, and with coefficients of relationship (r) based
on the ancestors alive in or near 1900. Each animal is shown but once. If it has more than
one son or daughter, an additional arrow indicates that. Selection has not yet llad much
opportunity to thin out the young herd. When that inevitable culling takes place, many
of the lines from Banner will disappear. In the herd 10 or 15 years from Ifow, even though:
the present plan is continued, Banner may not appear much more important than Banner-
view or Banner's Last. In allY case he is not apt to be more important than those two
combined. (From Lush, J. L., A Herd of Cattle Bred for Twenty Years without New Blood,
Jour. Hered., 25 (6) :209-218, June, 1934.)
render the animals somewhat capable of uniform transmitting ability.

The amount of inbreeding being practiced in most breeds at present is
very small. Most breeds probably have some good and some poor'
genetic material. Inbreeding would allow the poor stuff to show itself,
and it could then be discarded rather than being carried along in a hidden
state, as is likely with outbreeding. No doubt there is some good genetic
material in most herds, but it is difficult to know this very definitely
without records of performance. Research and discovery of the best
within the breeds foIIowed by line- or cIosebreeding to render the material
more homozygous is badly in need of doing. Such purified inbred lines
could then be used for intelligent crossing within the breeds, and they
would also be available for crossing between or among the breeds in both
fields, thus hastening the creation of better genetic lines. For progress
in animal breeding, we need first to know what we now have. When that
knowledge is available, it can be used intelligently both to strengthen
existing breeds and to create new ones.
Summary.-We have seen in this chapter that animals which evidence
relationship by having common ancestors within the first four to six
generations of their pedigrees are likely to have received identical genes
from their dose-up common ancestors. When two such related animals
are mated, they may both pass their identical genes to their offspring.
Matings involving re!!l,~~d animals are called inbreeding) and an inbred
aninl::LJi~JikfJlY_tQ biholJlozyg6~us in~rnore s~ts-of genes than is an outbred
:@jQlal.J Inbreeding creates no new gen~~\)Ut allows those already present
to get into homozygous combinations. Inbreeding results in improve-
ment or deterioration, depending entirely on the genes present in the
related animals. <;!DIy those hreeders who,,!,! herds...are fairly Ill.r~d of
more than avera e merit should ractice closebreeding. Linebreeding
shou ave a greater vogue t an it has had in the past.
Inbreeding will be successful (1) if the good genes pretty well out-
number the bad ones to start with,(2fir the breeCIeY lias-more than
average skill 'in pliiimiitg his matrngs, (3) iLthe breeder can and win
practice veryrIgid sclection. Finally, it isn't the system of mating__::_
it's'the genes in the animaTsand the intelligence in the man, and the
latter is perhaps a'!solargelya;'inatter of genes.
Inbreeding will uncover u~<!ysirable trait~ in animals, expose th~ so
that they can be discarded. It will create separate more or less homo-
zygous lines which may in themselves lack something of being completely
desirable but prove to be very potent for good in crossing between lines
or onto heterogeneous material. Good lines may be sorted out from
inbred stocks; but when an outcro~s is resorteato, the good results of
many generations of inbreeding will be scattered and lost, i.e., inbreeding
must again follow the outcross. Inbreeding will succeed if the genetic
complex was good to start with and the breeder's selection is keen enough
and ruthless enough to offset probable deterioration.
In closing this chapter, it seems safe to say that animal breeding has
lost more through fear of inbreeding than it has lost in any other way.
Only occasionally do we get a particularly desirable combination of gene:;
in an animal. Due to the halving and sampling nature of inheritance,
outbreeding soon tends to dissipate this original good combination.
Each animal is unique, there never has been nor can be another just like it
(except identical series). The easiest way to get other animals as near
genetically like the original good one is through some form of inbreeding.
'ow that the principles involved in breeding systems are becoming
f
better known and understood, it seems reasonable to hope that greater
se will be made of inbreeding systems.
Rapid progress in animal breeding will have to wait upon the intelligent
~~~~nIl:br!leding. - ---
References
Books
EAST, E. M., and JONES, D. F. 1919. "Inbreeding and Outbreeding," J. B. Lippin-
cott Company, Philadelphia.
JONES, D. F. 1925. "Genetics in Plant and Animal Improvement," John Wiley &
Sons, Inc., Kew York. .
LUSH, J. L. 1945. "Animal Breeding Plans," Collegiate Press, Inc., of Iowa State.
KICHOLS, J. E. 1945. "Livestock Improvement," Oliver & Boyd, Ltd., Edinburgh
and London.
WINTERS, L. M. 1948. "Animal Breeding," John Wiley & Sons, Inc., New York.
Bulletins and Papers!

ASDELL, S. A. 1945. Breeding Systcms Used to Produce the Highest Yielding
Guernsey and Holstein Cows, Jour. Anim. Sci. 4:146-150.
BAKER, G. A., MEAD, S. W., and REGAN, W. M. 1945. Effect of Inbreeding on the
Growth Curves of Height at Withers, Weight and Heart Girth of Holstein
Females, Jour. Dairy Sci. 28 :607-610.
BARTLETT, J. W., REECE, R. P., and MIXNER, J. P. 1939. "Inbreeding and Out-
breeding Holstein-Friesian Cattle in an Attempt to Establish Genetic Factors for
High Milk Production and High Fat Test," N.J. Agr. Expt. Sta. Bul. 667.
- - - , --"-, and LEPARD, O. L. 1942. The Influence of Inbreeding on Birth
Weight, Rate of Growth and Type of Dairy Cattle, Jour. Anim. Sci. 1 (3) :206-212.
- - - and MARGOLIN, S. 1944. A Comparison of Inbreeding and Outbreeding in
Holstein-Friesian Cattle, N.J. Agr. Expt. Sta. Bul. 712.
DICKERSON, G. E., et al. 1947. Performance of Inbred Lines and Line Crosses in
Swine, Jour. Anim. Sci., Vol. 6, No.4.
HAYS, F. A. 1919. Inbreeding Animals, Del. Agr. Expt. Sta. Bul. 123.
HODSON, R. E. 1935. An Eight-generation Experiment in Inbreeding Swine, Jour.
Hered., 26:209-217.
HUGHES, E. H. 1933. Inbreeding Berkshire Swine, Jour. Hered., 24:199-203.
KING, H. D. 1919. Studies on Inbreeding, Wistar Institute, Philadelphia.
LUSH, J. L. 1937. Linebreeding, Iowa A"{,\. Expt. Sta. Bul. 301. ..
- - - . 1932. The Amount and Kind of Inbreeding Which Has Occurred in the
Development of Breeds of Livestock, 6th Internat'l. Genet. Congo Proc., 2 :123-126.
MARGOLIN, S., and BARTLE=, J. W. 1945. The Influence of Inbreeding upon the
Weight and Size of Dairy Cattle, Jour. Anim. Sci., 4(1) :3-12.
MCPHEE, H. C., RUSSELL, E. Z., and ZELLER, J. 1931. An Inbreeding Experiment
with Poland-China Swine, Jour. H ered., 22 :293.
! See als<\ Annual Reports, Regional Swine Breeding Laboratory, Ames, Iowa.
\
\
O'CALLAGHAN, M. A. 1923. Cattle Breeding and Inbreeding, World Dairy Congo
Froc. pp. 1401-1405.
REGAN, W. ]\1., MEAD, S. W., and GREGORY, P. W. 1947. The Relation of Inbreed-
ing to Calf .Mortality. Growth,11(2):101-131.
RICHEY, F. D. 1935. The What and How of Hybrid Corn, U.S. Dept. Agr. Farmers'
Bul. 1744.
RITZMAN, E. G., and DAVENPORT, C. B. Some Results of Inbreeding on Fecundity
and on Growth in Sheep, N.H. Expt. Sta. Tech. BuZ. 47.
ROBERTSON, A. 1949. Inbreeding Experiments in Dairy Cattle, Anim. Breeding
Abs., 17(1):1-6.
STEELE, D. G. 1944. A Genetic Analysis of Recent Thoroughbreds, Standard-
breds, and American Saddle Horses, Ky. Agr. Expt. Sta. Bul. 462.
TYLER, W. J., DICKERSON, G. E., and CHAPMAN, A. B. 1946. Influence of Inbreed-
ing on Growth and Production of Holstein-Friesian Cattle, Jour. Anim. Sci.,
5(4) :390-391.
----, CHAPMAN, A. B., and DICKERSON, G. E. 1949. Growth and Production of
Inbred and Outbred Holstein-Friesian Cattle, Jour. Dairy Sci., 32(3) :247-255.
'VILLHAM, O. S., and WHATLEY, J. A., JR. 1941. The Improvement of Swine
through the Use of Moderate Inbreeding and Selection, Okla. Agr. Expt. Sta. Cir.
69.
WINTERS, L. M., et aZ. 1948. Experiments with Inbreeding Swine, Minn. Agr.
Expt. Sta. BuZ. 400.
et oZ. 1930. A Survey of Inbreeding Research, Amer. Soc. Anim. Prod.
Proc., pp. 114-134.
WOODWARD, T. E., and GRAVES, R. R. 1946. Results of Inbreeding Grade Holstein-
Friesian Cattle, U.S. Dept. Agr. Tech. Bul. 927.
- - - and - - - . 1933. Some Results of Inbreeding Grade Guernsey and Grade
Holstein-Friesian Cattle, U.S. Dept. Agr. Tech. Bul. 339.
WRIGHT, S. 1933. Inbreeding and Homozygosis, etc., Natl. Acad. Sci. Proc., 19:
411-433.
1923. Mendelian Analysis of Pure Breeds of Livestock. II. The Duchess
Fam~ly of Shorthorns, etc., JZ!:r. Hered., 14 :405-422.
1923. Mendelian Analysis, etc. 1. The Measurement of Inbreeding and
Relationship, Jour. H ered., .14 :339-348.
1922. Coefficients of Inbreeding and Relationship, Amer. Nat., 56 :330-338.
1922. The Effects of Inbreeding and Crossbreeding on Guinea Pigs, U.S.
Dept. Agr. Bul. 1090.
- - - and MCPHEE, H. C. 1925. An Approximate Method of Calculating Coeffi-
cients of Inbreeding and Relationship from Livestock Pedigrees, Jour. Agr. Res.,
31 :377-383.
CHAPTER XX
GENERAL CONSIDERATIONS IN SELECTION
/ Selection is. the keystone of the arch in, anilll!!:.LbrE)eding. The fact
that a breeder's ultimate success depends upon his skill in selecting and
l
mating animals is so obvious that it permits of no argument. It will
be recognized immediately also that sel~ction and systems of breeding
are bound together so closely that there. is no likelihood of being able.
to separate them. They are but different angles of this same basic
problem. We have seen that an animal's ultimate and intrinsic worth
depends in the final analysis on the genetic contents of the egg and the
sperm that united to produce it. We must, therefore, relate all of our
thinking about selection to the ultimate ,determip.ers of IJotentil11ities-
"genes and chromosom~s~and it hardly seems necessary to add that
whether or not ~n animal with a good hereditary complex actually
develops into a desirable individual depends in '. turn -on' the type of
environment, especially feed, care, and management, which is provided.
If our seleCtionlsSuccessful,' itt1le animals we b~ed, in other words, are
superior in the qualities that enable them to perform their respective
functions successfully and profitably, It will be ~.-
because their good genes
.... --
have had an o.Eportunity to express the'msel~n_Ylro.nment.
Selection is nothing new. . I fhas been going on for over a billion years,
or since the first forms of l~fe appeared on this planet. This is called
natural selection and is the idea on which .Charles Darwin formulated his
the~ry ~i ~-;olution. For millions or years before man even appeared
on the scene, nature was, busy putting the gene~together into new
combinations (variations). Some of these new combinations were
successful, giving rise to new types of life that were adapted to the given
environment. They, therefore, prospered and multiplied. The unsuc- ,
cessful combinations were those giving rise to animals that were ill-
adapted to their environment. They could not live and prosper and
therefore left few or no progeny. Nature was ruthless, her one criterion
was fitness, and on the dual bases of variation and fitness she has peopled
the earth with myriad forms of life. A long, long succession of successful
variations finally produced man himself.
One of man's early and most far-reaching accomplishments was the
domestication of plants and animals. When this had been accomplished,
528
---
GENERAL CONSIDERATIONS IN SELECTION 529
nature's passive selection was supplanted by man's active artificial

selection. The latter has been going on for some 5,000 to IO;OOO-years;
affil,ln spite of certain cynical opinions to the contrary, it has had a
truly remarkable degree of success. For proof of this statement, all
we need do is contrast a wild boar and a champion barrow; a Texas
Longhorn and a grand champion steer, a mouflon and a modern sheep,
Przhevalski's horse and a modern Percheron; an early type cow and a
modern dairy queen. Both in looks and utility such a wide gulf separates
these extremes that one marvels it could be bridged in such a short space
of time. Some cri ti cs aver that there has been no progress in livestock
breeding during the past200-years;- that-all our gains in animal form and
function are due to better feeding and management. That the latter I_
have played a large part no intelligent person would wish to deny, but to
give them the whole credit seems"'unjustified. It is tantamount to saying
that all livestock breeders' efforts at selection have been fruitless, that as a
group they have not increased to any degree the relativ.:.e._f@_9.lJgncy_()f.
good genes or been. able to discard any bad ones. -
-If one\vlsnes to- maintain that SelectIOn methods have not been 100 per
cent efficient, he would have ample justification for his position, but to
deny any progress in breeding for the past two centuries is simply not in
keeping with facts, and before being too critical of breeders of the past,
one should, in fairness, recognize the handicaps under which they worked.
The mechanism of reproduction was the last field tackled by the physi-
ologist, and its complett" functioning in terms of the immediate organs
concerned and the remote controls situated in the endocrines is only now
beginning to be understood":in a relatively complete manner. Likewise
the mechanism of hereditary transmission has been understood for only
50 years, and its real use as a tool in creating better plants and animals
was begun only 30 or 40 years ago. ~_phenotypic selection h8:s
brought us a long way from the crude and inefficient types of animals with
which man began. :r;;. can be admitted that "ox-cart-" or "horse-and-
buggy" methods of selection need modernizing without belittling the
accomplishments and tools of those who have passed on.
For a long time, selection in all classes of livestock was based upon
individual appearance, that is to say, ideals of bodily form were set up
after which the breeder tried to mold his own herd or flock. Many
influences helped to shape these ideals, one of the most important of
which was the show-ring standard. It was thought, in other words, that
there ,vas a relation between form and function with horses and dairy
cattle as well as with meat animals. Selection was based largely on
anatomy. Gradually, however, it became apparent that, in dairy cattle
especially, there was no necessary correlation between type and produc-
tion, between external anatomy and internal physiology. Selection then

shifted its base somewhat by including a consideration of mtrformance at
the drawbar, at the milk pail, or in the feed lot. Selection, in oth~
words, came to include physiological functioning as well as anatomy.
Finally, however, it became apparent that an animal's own performance
was also far from a perfect guide to its transmitting ability.
In recent years, therefore, we have heard much of progeny-performance
tests as the only safe measure of an animal's breeding ment. We "stiIIPay
attention to type, anatomy, or external appearance; we still pay attention
to performance, physiology, or internal activity; but to these criteria of
selection we have now added a consideration of how the animal actually
performs iP- ,tr;1nsmitting its ?,vn desirable qualities. We still pay as
much attention as' possible to individu~ance and individual
performance. We want both these things in our breeding Mllmals, bert
1I:'e have'sIo-ivly come to realizJ that having either one or both does not
guarantee that an animal will be a good transmitter. If we can make our
~
animals' pedigrees complete enough in terms of records of accomplish-
_. .-.' ~I}d..~_goodly
in the pedigree "'-- share _
ments (both as individuals and as transmitters) of the direct ancestors
__ of the collateral
'----.. relatives, we can pre-.
........ ~
dict with a fair degree of assurance how the ammals will transmit, How-
ever, the final and only sure test of an animal's breeding worth is now
recognized to reside in the quality of its offspring. Good type, good
performance, good pediwee (in terms of records) are indications of
,breeding. merit., b;-;-i>ihe ~n.lY assuraii~e ~{ breeding meritresldes 1I1Th'e
offspring themselves. # • :
Variation, the Base for Selection.-The breeder's stock in trade is

variation. The milk production, butterfat test, and total butterfat of
co"'ws; the rate of gain and quality of carcass nitatstock; the amolmt of
wool, the ability and willingneSstopUU, the oolor pattern, the fertility
;rte, the type, the length of useful life (we could go on almost inc1efinit~iY)
do show conshlerable -;ariati~n. Selection is the breeder's attempt to
seize and perpetuate such favorable variations as come his way. If a
Holstein is black and white, we know it has at least one gene B-what
the other gene of this pair is, we cannot tell by inspection; i.e" one black
Holstein may be BB, another Bb. Likewise, two Holsteins may each
produce 10,000 lb. of 3.6 per cent milk, but this is no sure ground for
assuming that they have identical sets of genes. One of these Holsteins
may transmit very well to her daughters-the other one very poorly.
We do not know how many genes each species of our farm animals has.
A recent estimate of the number in man by Spuhler (1948) 1 based on
total length of human chromosomes compared to Drosophila and on esti-
1 SPUI'I,LER, J. N. "On the Number of Genes in Man," Science, 108:279-280, 1948.
GENERAL CONSIDERATIONS IN SELECTIOiV 531
mated mutation rates in the sex chromosomes yields the limits of 20,000
to 42,000 genes in the human. Possibly the number of genes in our farm
animals is something fairly comparable. It is unlikely that we can ever
know very accurately what genes any animal has. Selection, therefore
(marshaling favorable genes) must be done indirectly.
Since all living forms are probably the result of an evolutionary process,
they may all still have some genes in common anTI the more clo~ely
related individuals are, the ~ore genes will they have in common. An
amoeba and a man may have some genes in common, though probably
not many. A monkey and a man no doubt have more genes in common
(sometimes, apparently many). A Chinese and an American no doubt
have still more genes in common, two Americans still more, two Smiths
still more, the two "coughdrop" Smiths still more, and identical twins
have all their genes in common, except for the slight possibility of
mutations.
Breed differences are gene differences. Breed A and breed B, no
doubt, have many (or most) genes in common. But they have enough
different genes so that they are set off into two distinct breeds. .If 1 in
200 Angus calves is born red, then the frequency of the b gene must be
about 0.07 (bb or b2 being 0.07 X 0.07 or approximately 0.005 per cent)
and the frequency of the B gene is, therefore, 0.93. Then pure (homo-
zygous) blacks, BB, are 867~ per cent; Bb are 13 per cent and bb are H per
cent, since the. zygotic ratio is the square of the genetic ratio. Short-
horns, however, and Herefords, while no doubt having many of the same
genes as do Angus, are 100 per cent bb, having 'no blacks. By selection
the frequency of the Band b genes in the Angus breed could be changed
in either direction. If the selection was for more red (more b genes),
then it would move up to 1 per cent red calves when the breed would be
81 per cent BB, 18 per cent Bb, and 1 per cent bb; then up to 10 per ce~
red calves when the breed would be 48 per cent BB, 42 per cent Bb and
10 per cent bb; then up to 25 per cent red calves when the breed would be
25 per cent BB, 50 per cent Bb and 25 per cent bb; etc. If there is no
selection for one gene or against another, the original frequencies of the
genes will remain just as they are.
While breeds have many genes in common, they also have many which
are different. Breeds may be homozygous dOl!linant§ (for some genes),
homozygous recessives, or be orany degree of heterozygosity ... The
memoers of a breedin one section of the country (descended largely from
a certain foundation) may be genetically quite different from the mem-
bers of that breed in some other section of the country. The genes in
t.vo herds in the same section of the country may be quite different in the
relative frequency of certain genes. The early success of selection in
molding animal and plant types toward an ideal led to a belief that
selection had practically no upper limit. The discovery of the hereditary
mechanism as resident in the genes finally dispelled this illusion. This
was brought about largely through the work of Johannsen with the
self-fertilizing garden bean. From a mixed population, he selected
plants that bore large seed and others that bore small seed. Attempts
at selection within these "pure lines," however, proved unavailing.
Selectinglarge_seed from the large line gave no larger seed, on the aver-
~-than did small seed fr~~ this line ... Likewise, the selection of small
s~m-the ·smallline gave no smaller seed, on the average, than did
large seed from the small line. Johannsen designated these as "pure
lines," i.e., pure from a genetic standpoint. Small seed from the large
line was small because of environmental effects, but genetically these
small seeds were just like the large seeds in the same line. •
Selection, therefore, did prove to have an upper limit, and selection
within a "pure line" was not effective. Thus, for the first time, it was
fully realized that the effectiveness of selection ·was entirely dependent on
the nature of the heredltaI-Y units.-- -If a quantitative character depends
on two sets of genes A and B and we start with a heterozygous generation
Aa Bb, we can get out a line that is AA BB or aa bb, thus showing the
character to a greater or less degree than did ancestors. 'Vhen we get a
line that is AA BB, thus showing the character to a greater degree, we
cannot go beyond this point. This is a pure line which, under optimum
environmental conditions, will always breed t r u e . - - - - - --
- This general prmcip1e IS apphcablet'Oair1iving forms. In our live-
stock, it is probable that most commercially desirable characters are of
a quantitative sort and probably controlled by scores or hundred of genes.
What we are attempting to do in selecting is to develop our animals into
"pure lines" or as close to this as we can get with unisexual animals.
The task is not an easy one because of the time necessary and the com-
plications involved in dealing with large numbers of genes, but knowing
the underlying scientific basis will at least enable us to chart our course
more intelligently. Dealing with large numbers of genes simply mean"
that it will take a long time to get them sorted out into the desired
homozygous forms, but we can have the comfort of knowing that when
we once get them sorted out they will remain in that condition indefinitely
except for an occasional mutation. There seems to be a growing belief,
however, that the best animals phenotypically may not be the most homo·
zygous-in other words, that an Aa animal may be better phenotypically
than either an AA or an aa animal. If this idea proves to be true, it will,
of course, have a very marked bearing on systems of breeding.
This, brings to mind the importance and gravity of the decision to
GENERAL CONSIDERATIONS IN SELECTIO]'{ 533
introduce a certain male into our breeding plans, When once we bring
a male into our herd or flock, his influence, good or bad, is apt to remain
with us for some time. When once his genes are put into our "hereditary
mix," they remain there more or less indefinitely. Probably most males
will bring in some undesirable genes, and, if there are not too many of
them, we can hope to rid ourselves of them in a reasonable length of time.
If a ma!e brings a host of bad genes, they may plague~rough many
generatIOns. \ ,>(., j}..
A!}_;iv.~ saw earl~~/_th.~._g_~nj_)'§'J!la.y act in a variety of ~ They may
shoW additive or cU!Il.ulat_iye. effect~. If in sheep, for example, genes
A, B, C, and D stood for 1 lb. of wool each and a, b, c, and d for Yz lb.
each and a breeder has a flock that was Aa Bb Cc Dd, he could by inter-
breeding select out lines tlrat were AA BB CC DD (shearing 8 lb.) or
aa bb cc dd (shearing 4 lb.). Our assumption is that these genes act
additively so that when AA replace aa wool production goes up 1 lb.
It seems certain that much inheritance for commercially valuable traits
in our livestock is of the additive type, though far from all of it.
Dominance effects may be misleading. In the above example if A
(1 lb. of wool) were dominant over a 0,2' lb.) then substituting a large A
for one of the small a's in aa animals increases yield by ~2' lb., but not
when substituted for the small a in Aa animals. AA and Aa animals
yield alike but do not breed alike-,ve will make some mistakes in
selecting because we can not readily distinguish between the AA and Aa
individuals. Epistatic effects, dominance between pairs of nonallellic
genes, behave in a siniilarfa15himl and arealiillcIrance'to'srueciTon because
to have the desired effects we must keep ce~~un8"*genes
together, a hard thing to do because of the principle of independent
assortment. A
Genetic variation is then of three sorts-additive, dominance, and
epistatic. The first sort presumably shows rt:;-presence whenever it
~~-ln a phenotype. The offspring will, therefore, average halfway
between the phenotypes of their parents. Dominance and epistatic
variance, while genetic and hereditary in the broad sense, does not behave
in this simple additive way, and the phenotype does not tell so complete
and simple a story of probable genotype as it does when the inheritance
is of the simple additive sort. If a good quality is due to simple additive
inheritance, it can be seized and enriched by selecting and mating parents
which show the quality in increasing amounts. If the quality is influ-
enced by dominant or epistatic genes, it will not necessarily be passed
to all offspring because of the halving and sampling nature of inheritance.
Finally, an animal with an average Of mediocre genotype may, through
the stimulus of excellent environment, perform fairly or very well. But
534 BREEDING AND IAfPROVEMENT OF FARM ANIMALS
t.hese environment.al, dominance, or epist.atic effects are not. necessar-

ily passed on to offspring which may fall far short of t.heir parent.s'
accomplishments.
Technically, variation is called variance, which is the average of t.he
squared deviations of all t.he variat.es. The square root of the variance
is the standard deviation represent.ed by small Greek sigma or u. Total
observed variance or u; is, therefore, made up of four port.ions: additive,
u';; dominance, u~; and epistat.ic, u;; and environmental variance which
is represented by ui .
Breeders t.ry to select offspring from their better animals on the basis \
of how these animals look and/or behave, their phenotype. If the
variance were entirely caused by additive genetic differences (none by
environment, dominance, or epistatic effects) the phenotype of the par-
ents would be their genotype, the offspring would average the same as the
average of their selected parents and the entire selection differential
would be achieved. The selection differential is the difference between
the average in any quality of those selected to be parents and the average
of the whole p09ulation in which they were born. If the average litter
size in a herd of hogs this year was 8 pigs and we saved pigs from litters
of 10 pigs, the selection differential would be 2 pigs. If this were all
due to additively genetic variance, next year's litters, while variable,
would average 10 pigs per litter. Animal breeding would then be too
easy to offer much of a challenge.
Actually much of the variance in the commercially valuable traits of
our animals is from environmental, dominance, and epistatic causes. All
observed variance is u~, and this is made up of genetic u';, dominance u~,
epistatic u;,
and environmental ui fractions. What portion of the
phenotypic superiority of parents we are likely to get in their offspring
consists of the genetic variation u6 divided by the sum of all the -sorts
of variance, plus a small amount of the epistatic variance in the first.
generation.
Methods of estimating heritability all depend in one way or Jnother
on the amount of resemblance between relatives as compared to resem-
blances between animals selected at random from a population. Another
way of saying this is that heritability can be determined by comparing,
the amount of variation among animals selected at random with that
found among animals of known relationship. Where possible, the best
method of ascertaining heritability is to compare the variation in a
random-breeding population where both hereditary and environmental
variation exists with that in pure lines where practically all the variation
is environment.al in origin. Unfortunately, this is not feasible in farm
animals except
, in the case of cattle where identical twins and ordinary
\.
twins can sometimes be so used. Estimates of heritability based on the

resemblances between relatives are fraught with the danger that relatives
may resemble each other both because they have similar genes and
because they have been exposed to similar environmental influences.
Lush discusses! the use of parent offspring resemblances as follows:
The resemblance between parent and offspring is the most widely useful
method, but is likely to include some environmental correlation between parent
and offspring. Also it will include something from the resemblance of the off-
spring to the other parent, if mating were not random. In using this method
the procedure is as follows: (1) observe the correlation between parent and
offspring, (2) subtract from that the environmental contribution, (3) double the
remainder, and (4) divide by one plus the correlation between males. 2 The
second step is always likely to be difficult, and the fourth will be unless the
deviations from random mating are known more exactly than is usually the
case.
A useful dodge which makes steps two and four unnecessary is to divide the
mates of each sire into a high and a low half on their own performance, combine
the data for all sires, divide the difference between the daughters of the high and
the low halves by the difference between the two groups of dams and double the
result. This measures the additive portion and a bit of the epistatic portion of the
differences between such dams as were mated to the same sire. It leaves unan-
alyzed the average differences between the groups of cows which get mated to
different sires. A similar division of the offspring of the high and low sires
mated to the same dam would answer as well in principle but, because of the
usually small number of offspring per dam, is rarely possible with farm animals.
Half-sib correlations are also used in determining heritability. Wright,
using a strongly inbred line and a control, outbred line of guinea pigs,
measured the average likeness between parents, between parents and
offspring, and between litter mates and was able to separate the variance
in amount of white spotting into three categories: (1) that due to heredity,
(2) that due to environment common to litter mates, and (3) that due to
environment not common to litter mates. The results showed that,
while the variance due to environment was nearly the same in the two
groups in actual units, it made up 97.2 per cent of all the variance in
the inbred group and only 57.8 per cent in the control, or outbred, group.
In other words, the hereditary variance in the inbred stock was only
2.8 per cent, while in the outbred stock it amounted to 42.2 per cent.
1 LUSH, J. L., "Animal Breeding Plans," pp. 93-94, Collegiate Press, Inc., of Iowa
State College, Ames, Iowa, 1945.

2 This should be the genetic correlation (coefficient of relationship) if the departures
from random mating were of the inbreeding kind, but should be the actually observed
correlation if the mating choices were based on each animal's own individuality for
the characteristic being ~tudicd. One will rarely be certain about this.
536 BREEDING AND IMPROVE11IENT OF FAR111 ANDI.fALS
Many studies on the heritability of various characters in farm animals

have been made during the past two decades. Most characters which
are important in the production of profitable livestock have been found
to have heritabilities in the range from 15 to 50 per cent, with a few
estimates above and below this range. Summaries of actual heritability
estimates will be given in later chapters in connection with the discussion
on selection in the various types of livestock.
As will be seen later, rather different estimates of the heritability of
the same character have sometimes been 'obtained in different studies.
Possible reasons for this diversity quickly become apparent when we
remember that heritability depends upon the relative amount of variation
which can be ascribed to heredity and environment, respectively. Vari-
ation in the actual amount of variance due to either of these causes will
greatly affect heritability estimates.
The amount of genetic diversity in the populations sampled will ha,'c
a great influence on heritability estimates. Increasing genetic purity
leading to less genetic variation results in lower heritability estimates.
The work of Wright previously referred to in which heritability of degree
of hair-color spotting in guinea pigs was greatly decreased in inbred
strains is an example of this. The more recent work of Hetzer ct al.
(1944) in which apparent heritability of type in swine was found to be
about 90 per cent \vhen three widely different types were considered as
one population, but only 38 per cent within strains is a further illustra-
tion of the same general principle.
Conversely, a decrease in variation of environmental origin will lead
to higher heritability estimates. This points out the necessity for 1
maintaining fairly uniform environmental conditions if breeding opera-
tions are to achieve maximum success. .j
Finally, small numbers of animals have been included in some studies,

and some of the disagreement in the estimates is doubtless due to sampling
errors.
~ -----
A heritability estimate is really nothing more nor less than'"---an
~--
of the average correlation between genotype-aiicrpTleiiotype. A high

__ estimate
heritability indicated a. high correlatiOri;--WhIcli in' turn means' that an
animal is probably similar genotypically and phenotypically. In such
cases progeny tests and family selection are unnecessary. In the majority
of cases, however, heritability is relatively low, and information from
progeny tests and the performance of relatives will be helpful in estimating
the probable genotype of an animal. Picking the better representatives
in a herd or flock to be the parents of the next generations is not enough
to ensure progress. Because so much of the variance is other than simply
additively',genetic in nature, we must broaden the base of selection to.
include as many of the close relatives as possible and select from good
phenotypes which, according to the nature of the ~e and
transmitting abilities of their near relatives, are probably also --g~~a:-
genotypes~--Tf the--maternal great=granddam was a productive femal~-
and hadseveral good offspring among them, the granddam, 'which was
by a good proved sire, and this maternal granddam was in turn a good
producing female and got good offspring among them, the dam, which
was by a good proved sire a,rd the dam herself was a good producing
female and has gotten some good offspring, then her present offspring
which we are considering, if by a good proved sire, has a reasonable
expectation of being a good one. In short, selection is likely to be more
successful if the six animals making up the bottom line of the pedigree for
three generations back are not only good phenotypes, but whose geno-
types have also been tested through their offspring and found to be
good.
Characteristics of a Livestock Population.-The numerical data as to
numbers and values of animals in the various classes of livestock are
given in Tables 1 and 2. Dividing the total value in 1949 by the total
number gives us an average value of $95 per head for all classes of animals
combined. The specific values for each class are listed in Table 1. The
modest value of the average horse, mule, cow, sheep, or pig should provoke
serious thought on the part of livestock breeders. These average values
indicate that billions of dollars worth of product are being produced
annually in the United States with low-grade, inefficient animals. Such
economic waste should be checked as soon as possible.
The average dairy cow is said to produce 5,000 lb. of milk and 200 lb.
of butterfat yearly. In 1947, 775,000 cows in Dairy Herd Improvement
Associations (D.H.LA.) averaged approximately 8,700 lb. of milk and
350 lb. of butterfat. According to the Bureau of Dairy Industry, the
feed cost per pound of butterfat in the latter group of cows was 43 cents,
while in cows producing only 200 lb. of butterfat, the cost per pound was
60 cents. Speaking generally, dairymen who test their cows are a
selected and superior group of men. They pay more attention to breed-
ing, they grow their young stuff better, they farm·and they feed at higher
levels than the average. So part of the increased production of D.H.LA.
cows is, no doubt, due to better environmental conditions, part is due
to better inheritance, and how much, on the average, environment plays
and how much inheritance, no one can say. D.H.LA. figures show
conclusively that" income over feed costs" mounts steadily with increas-
ing production. True, the high-producing cow eats more feed than the
low producer, but a smaller percentage of her feed needs to be used for
maintenance. Or to put it another way, the higher producing cow
538 BREEDING AND IMPROVEJIEST OF FARM ANLlfALS
produces more product per pound of feed consumed and for this reason
is more efficient and more profitable.
It should be remembered that an average is a mid-point. If the
average production of all cows in the United States is 5,000 lb., of milk,
then about half our cows must produce less than 5,000 lb. of milk per
year. To people living in a good dairy area, this is almost unbelievable.
Table 34 shows the distribution of 216,489 cows on which dam-daughter
comparisons were secured in proving the first 25,000 bulls in D.H.LA.
testing.
The average butterfat production of all the dams in Table 34 is 376 lb.
TABLE 34.-DISTRIBUTION OF Cows' RECORDS USED IN PROVING THE FIRST 25,000

DAIRY SIRES BY THE BUREAU OF DAIRY INDUSTRY*
Dam and Number Dams' Daughters' Bull's

daughter of sires butterfat, butterfat, Equal-parent
pairs used lb. lb. Indext
--_._-
94 16 188 243 298
324 52 215 254 293
1,351 182 240 271 302
4,483 579 264 292 320
10,365 1,286 289 310 331
19,317 2,301 313 329 345
30,462 3,507 338 346 354
40,308 4,479 362 365 368
39,536 4,464 387 383 379
30,924 3,466 411 403 395
20,304 2,363 436 421 406
10,888 1.299 460 438 416
5,321 626 485 451 417
1,799 238 510 472 434
544 75 535 474 413
283 40 560 496 432
106 13 584 535 486
56 7 610 436 462
24 4 637 476 315
-----
216,489 25,000 376 Av. 375 Av.1
* Adapted from data in D.H.LA. Letter, Vol. 24, No. 12, December, 1948.
t Indexes will be discussed at length later.The formula for the Equal-parent Index is twice the
daughters' average minus the dams' average. This places daughters just halfway between the parental
levels.
It will be seen that the greatest number of cows (79,844) is in this group
and that the number in the other groups tapers off fairly regularly as we
approach the upper limit and not so regularly as we approach the lower
limit because the poorer cows are disposed of early. In short, most
\
GENEUAL CONSIDERATIONS IN SELECTION 539
animals are average animals-there being fewer very poor or very gOOd~
ones. The breeder's job is to find the upper half of his herd in trans-
mitting ability and to save his replacements from these better-than-
average females which are in better-than-average cow families and see L--
to it that they are by better-than-average males. Only in this way can
improvement be brought about.
It will be seen in Table 34 that the dams are arranged from low to
h~gh by 25-lb. intervals. They start at 188 lb. and proceed upward in
a straight line to 637 lb. Their daughters start at a higher level and end
at a lower level. When their daughters' production is fitted to a straight
line by the method of least squares, the daughters' production is found to
400
II 335 I
!!2 300 I l - 328
::J
OJ . ~50V \ 233
'0 200 Mos! Bulls 1\
~ lore overage
4; \ I
OJ
.D
E 1 \/22
100 _Some Bulls
::J
z A few Bulls-
~6
ore very poor
"""5) I
o Iy
1 9
ore/ops
~/
o 125 225 325 425 525 625 725
Bull Index in Pounds of Butterfal
FIG. 165.-Distribution of Holstein-Friesian bulls. (From Vol. 15, Holstein-Friesian Red
Book.)
start at 247 lb. and to end at 541 lb. The regression of daughters on
dams in an unselected population is known to be 0.5. This means that
these daughters should start at 275 lb. and end at 512 lb. They actually
start at a lower figure and end at a higher one. This would seem to
indicate that there is effective selection now used in some D.H.LA.
herds. The poorer herds seem to use below-average sires and the better
herds to use above-average sires. If the environment were the same for
all herds, we would know this to be the case. Some of this possible
hereditary difference may actually be due to environment, the poorer
herds holding down production of both dams and daughters through
poor feeding and management and the better herds boosting production
by superior feeding and management. For the data in Table 34, the
sires' transmitting levels (Equal-parent Index) start at 305 lb. and end
at 449 lb., instead of running straight across on the 375-lb. level as they
would do if all environmental conditions were similar and there were no
effective selection. The above data shows that, so far as milk pro-
duction is concerned, most cows are about average and that they grow
LfB;~ARY
~,:sityColle,;" 01 Vell'rinilrV S(
& l\nimal I !tlS : ","
540 BREEDING AND IMPIWVEMEN7' OF FARM ANIMALS
smaller in numbers as the extreme of low production or high production

are approached.
A few years ago we studied the transmitting abilities of some 1,563
Holstein bulls. From their daughter-dam comparisons, the apparent
transmitting qualities of these bulls "'ere known. Their distribution is
shown in Fig. 165.
In Table 35, we show the distribution as regards type classification in
our dairy breeds.
TABLE 35.-DISTRIBUTION OF TYPE IN DAIRY CATTLE
Brown Guern- Hol-

Classification Ayrshire Jersey All %
Swiss sey stein
--- ---
Excellent ...... 729 130 290 2,746 1,401 5,296 2.2

Very good ...... 5,676 2,264 3,716 29,995 15,687 57,338 23.4
Good plus. . . . . . 8,259 4,683 8,494 49,664 35,261 106,361 43.4
Good. 0 .....
••• 3,923 2,429 4,672 20,623 30,995 62,642 25.6
Fair .. .... 611* 248 818 2,908 8,306 12,891 5.2
Poor. ...... 3 115 60 382 560 0.2
19,198 9,957 18,105 105,996 92,032 245,088 100.0
* Includes any classified "Poor."
Again we see in Table 35 that the animals distribute themselves in a

fairly regular way around the average.
The main purpose of calling attention to the distribution of any
population around its mean is to portray the principal problem involved
in improving our livestock. As we have already said, our task is to find\
the better females and get them bred to the better sires. Up to the
present, it has been a case of average females being bred to average sires
with the foreordained result of getting average offspring.
Results similar to the above for dairy cattle would no doubt also be
found for our various classes of fat stock were records available. That
animals vary is an observation that none can miss, When the quality
being considered is one determined by the interaction of many sets of
hereditary determiners or when it is one that is easily influenced by
environmental factors or both, the population distributes itself around
the mean in a fashion closely resembling a normal curve, as is illustrated
in Fig. 165.
From the very beginning, animals of any class have probably tended to
vary about their respective mean or average in any quality. The fact
that our animals have been improved down through the ages signifies
that man has been able to select from the upper half of the population.
In this way thf{'"
, rage has been moved upward by almost imperceptible
\! '-
'~.
stages. Now that some of the facts regarding the principles of heredity
are known, a somewhat more rapid shifting of the average toward higher
goals should be possible. It must be remembered, however, that gradual
progress made through the centuries may have limited somewhat the
working range of animal breeders. For example, further improvement
in the egg-laying ability of a 250-egg strain of chickens may be more
difficult than in a wild strain laying one clutch a year. In general, the
means effecting further improvement would seem to consist of two things:
(1) getting more accurate and complete facts about the various qualities
of our animals and using the facts more intelligently by arriving at an
animal's probable genetic value from a consideration of its close relatives
as well as the animal's own qualities; i.e., broadening the base of selection;
and (2) being more careful to separate, in so far as possible, genetic from
environmental variations. We will probably continue to find that our
animals will vary around their mean. How to locate the favorable/~
genetic variations will continue to be our main problem in selection.
Changing Gene Frequencies. 1-Since genes determine characters,
breeders, in selecting for certain things, are actually trying to influence
the proportion of different genes in their stock. Wright, in studying
the coat ~olors of the parents of 3,000 Shorthorn cattle, found that 47.6
per cent were red, 43.8 per cent were roan, and 8.6 per cent ,vere white.
Since this is a character seemingly controlled by one pair of genes lacking
dominance, we know that to the 8.6 per cent of white genes showing as
white and to the 47.6 per cent of red genes showing as red must be added
one-half the roan value 43.8/2 or 21.9 (roan being Rr), in order to get the
frequency of the white and the red genes in the population. Thus,
47.6 + 21.9 = 69.5% red genes and 8.6 + 21.9 = 30.5% white genes in
the population. N ow, if the population had been mating at random,
we would have had
G'J.5R 30.5 r
69.5R 48.3 RR 21.2 Rr
30.5 r 21.2 Rr 9.3 rr

I
or
48 . 3 RR: 42 . 4 Rr: 9 . 3 rr
instead of
47.6 RR:43.8 Rr:8.6 rr
This slight discrepancy shows that mating was not at random but that
there was a preference on the part of breeders for roan.
1 See ibid., pp. 64--71.
•
5.t2 BREEDIXG M·iD IMPROVEMENT OF FARM ANIMM,8
Generally, q is used to represent the frequency of one gene and I - q

its allele, and since in random mating the proportion of different zygotes
will be the square of the gametic ratio, the ratio of zygotes becomes:
q2:2q(1 - q): (1 _q)2
If either the first or the last term is known, the others can be computed.
If 1 per cent of Holstein calves were born red in color, then red, rr, or
(1 - q)2 = 0.01 and (1 - q) = 0.1 and q = 0.9. So under this assump-
tion 81 per cent of Holsteins would be pure for black, homozygous
BB; 18 per cent heterozygous, Bb; and 1 per cent would be red. In
other words, if 1 per cent of Holstein calves are born red, then 1: 5 or 6
black Holsteins are heterozygous for color.
q2 2q(1 - q) (1 - q)2
AA Aa aa
0.25 0.50 0.25
q2
BBO.25 0.0625 0.1250 0.0625
AABB AaBB aaBB
2q(1 - q)
Bb 0.50 0.1250 0.2500 0.1250
AA Bb AaBb aaBb
(1 - q)2 0.0625 0.1250 0.0625

bb 0.25 AA bb Aa bb aa bb
I I
FIG. 166.-Independently segregating genes A and B-frequency 0.5.
This principle applies also to independently segregating, nonallelic'

gene pairs. This can be shown by the usual g: 3: 3: 1 dihybrid ratio.
If genes A and a have a frequency of 0.5, genes Band b also a frequency
of 0.5, then they will recombine as shown in Fig. 166.
This formula could be extended to n genes by serial multiplication of
the squared binomial of each independently segregating gene, Ii'
Thus we can see that ,yith two pairs of genes of equal frequency (0.5)
we would have 1: 16 chances of getting an animal pure for both pairs of
dominant genes (AA BB). This is calculated according to the formula
q2n (n being the number of pairs of genes involved). Thus in Fig. 166,
q was 0.5 and n pairs of genes was 2, so 0.5 4 = 6.25%, or 1: 16 chances.
With qat, 0.4 and five pairs of genes, we would have 0.4 10 = 0.0001, or
I.
\
\
! \
1: 10,000 chances of getting an animal homozygous for all five pairs of

genes whose frequencies were 0.4. If the gene frequencies had been
0.6, our chance would have been 1: 166; if it had been 0.8, our chance
would have been 1 : about 10; whereas, if the gene frequencies had been
0.99, we would have had 1: 1.1 chances. That is, if a herd was almost
all AA with very few Aa or aa, BB, with very few Bb or bb, CC with very
few Cc or cc, etc., it would not be difficult to keep them AA, BB, CC, etc.
It is obvious that increasing the number of gene pairs being selected
for reduces the chances of securing them all in a homozygous state.
If we increase the pairs to 10 and leave the frequency at 0.99, our chance
of success drops to 1: 1.2, whereas increasing gene pairs to 20 reduces our
chance to 1: 1.5, and increasing the gene pairs to 50 reduces our chance
of success to about 1: 3. In practice the more things ,ye try to select for
simultaneously, the more likely we are to fail. In our college dairy
herds, we feel that we must have both good type and good production.
The genes, hmvever, that bring about style, straight tops, tine withers,
well-set legs, etc., do not necessarily have anything to do with the
physiological functioning of the organs and glands that result in a heavy
milk flow. If we could select for production alone, we probably could
get it more quickly, and we might end up with animals which were of
less-than-average merit in type, though this is not necessarily inevitable.
If one-third of the animals in any dairy breed would suit us from a
production standpoint and one-third of them from a type standpoint and
these two things were not closely correlated, then it is obvious that
only one-ninth of the animals of the breed would suit us from a combined
standpoint of type and production. The more uncorrelated things we
select for simultaneously, the more we narrow our range of selection and
the more we reduce our chances of success.
It is apparent that it would be easier to find an animal having at least
one of the desired genes than it would to find one having both, since,
if q2 are AA and 2q(1 _ q) are Aa, then those which have at least one
Ji are q2 + 2q(1 _ q), or 2q _ q2, which may be written q(2 _ q).
This can also be extended to n pairs of genes by applying the formula
[q(2 __ q)]n. Thus, in our last example of a dihybrid with equal gene
frequencies, we would have [q(2 _ q)]n = [0.5(1.5))2 = 0.75 2 = 56.25%,
or 1:2 chances of getting an animal that had at least one A and one B.
With q at 0.40 and five pairs of genes, [q(2 - q)]n = 10 per cent, so we
would have 1: about 10 chances of getting an animal with at least
one dominant gene in all the five pairs involved; whereas, if the gene
frequencies had been 0.6 for each of the five pairs, our chance would
have been 1 :2.5; at 0.8, it would have been 1: 1.2; at 0.99, it would have
been 1: 1.001, or a practical certainty. Lowering the gene frequencies or
544 BREEDING AND I1vIPROVEMEl\lT OF FARZI4 ANIMALS
increasing the number of pairs of genes involved curtails the likelihood

of selection being successful.
From the above discussion and in view of the fact that most of our
1 animals are still relatively heterozygous (gene frequencies around 0.5)
(I for many pairs of genes and that most of the commercially desirable
qualities are probably controlled by many (perhaps 20 or 100) pairs of
genes, the breeder's task is seen to be very difficult and complicated.
With intelligent selection, however, based on recorded data, he can
favorably influence gene frequencies in his herd or flock. He may never
render his animals homozygous for all the desirable genes, but he has the
assurance that the merit and value of his animals will increase with the
increasing frequency of desirable genes.
It is obvious also that selection for a recessive is much more simple
than selection for a dominant owing to the fact that, if dominance
prevails, then AA and Aa animals are indistinguishable phenotypically,
while the aa animals can be recognized genetically from their phenotype.
The Suffolk horse is a good example of this, its color, chestnut or sorrel,
being due to a pair of recessive genes, this being the only color found in
the breed. If early importers and breeders of Holsteins had elected the
red and white instead of the black-and-white color pattern, they would
not have been plagued with off colors that often lead to suspicion of the
purity of breeding of the animals producing ~hem. When a red calf ·is
born in a Holstein herd, the sire and the dam are, of course, jointly and
equally responsible, each having transmitted the recessive gene. With
1 calf in 100 born red, the genetic make-up of the breed is 81 per cent BB,
18 per cent Bb, 1 per cent bb (red).
Just to weed out the reds as they are born will have little effect on
purifying the breed for color, since 90 per cent of the b genes are hidden
in Bb animals, while only 10 per cent are visible in bb animals. If we
started ·with an F2 generation that was 25 per cent BB, 50 per cent Bb,
and 25 per cent bb, it would take eight generations of eliminating bb indi-
viduals to get the proportion down to 1 per cent bb animals, or 1: 100;
whereas to halve this amount, i.e., make it 1: 200, would require four
more generations; to get it to 1 :400 would require six more generations;
and to get it to 1 :800 would require eight more generations. This sort
of mass or phenotypic selection is seen to work very slowly. It can be
greatly hastened, of course, by discarding any animals that have offspring
showing the undesirable recessive character and also those whose sibs,
cousins, or other collateral relatives show the character, in other words, by
adding genotypic selection to the phenotypic.
Most of the problems and examples used to illustrate genetic principleH
involve a Very few sets of alleles. The stuof'nt (lno hreeder needsj
\
\
therefore, to remind himself constantly that this is, in general, an over-

simplification. It is known, for instance, in the pomace fly, Drosophila
melanogaster, that upwards of 40 genes function in producing the wild-
type eye color of red. If 40 of these pairs were the necessary dominant
genes but the forty-first pair had mutated to, say, garnet, then the eyes
would be ga1'net instead of red. Some of these 4Q-odd pairs of genes
are located in each of the 4 pairs of chromosomes.
In our domestic animals, there seem to be 20 to 30 pairs of chromosomes
with perhaps 100 or more genes in each chromosome. What we are
actually dealing with, therefore, from a practical standpoint, would seem
to be a complex gene complex. We talk about individual genes, but
we are actually dealing with blocks or groups of genes. It may be
possible too that in a certain gene complex, the homozygous dominant
AA serves best, while in another perhaps Aa does, and in still another
perhaps the homozygous recessive is best. As we have said previously,
the breeder is attempting by selection to put together the best possible
gene complex, to influence the proportion of desirable genes in his stock
in a favorable manner, and he uses a variety of criteria to measure the
value of the gene complexes that he has or desires to get.
Gene frequencies for various qualities are quite variable both in and
between various breeds. To take an extreme example, the genes making
for the type and temperament associated with speed are very scarce in
the Percher on breed of horses. Nevertheless, there are probably enough
of them present so that many generations of selection for these genes
would make them sufficient in number to produce a horse of considerable
speed. Similarly, the genes for red color are very few in the Angus breed, '
but constant selection for them would result in changing the breed from
black to red. And in Holstein cattle the genes for higher butterfat test
(4 to 5 per cent) are relatively scarce, but enough of them exist so that
selection over many generations could result in an average test for the
breed of over 4 per cent.
If gene A is homozygous in a breed A_~{, then one mating is as good
as another. If gene B is heterozygous, Bb, then best mating is BB X BB;
next best, BB X Bb; and poorest, Bb X Bb. If we start with Bb X Bb,
we have a ratio of IB: Ib gene. N ow, if we cull the double recessives,
in the next generation we have 2B: Ib; in the next, 3B: Ib, etc., according
to the formula (n - 1)2A_.4_:2(n - 1) Aa:1 aa. In the second genera-
tion we cull 25 per cent as aa; in the fifth generation, 4 per ceD,t; in the
tenth generation, 1 per cent under the formula 100jn 2 , but even her<:>
there still will remain 18 per cent as heterozygotes. Selection can and_
does change the relative frequency of genes, but it is_ very slow when
selecting for a dominant gene and unable to fix---theheterozygous con-
._~~." J __ - - - , __ . _ •• __ _ •••• _ -
BREEDING AND IMPROVEltIENT OF FARM ANIMAUI
dition, as, for example, the roan color in Shorthorns or the blue in Anda-
lusian fowl.
Heredity vs. Environment in Selection.-The past has seen many
arguments between the hereditarians and the environmentalists, with
first one and then the other having seemingly the bette~f the argument.
The debate has of late tended to be resolved by tlte simple but far-
reaching expedient of sUbstituting and for VB, In other words it is not
a case of whether heredity or environment is the most important. B.Jill!.
_!!:!e important, on~_ much as t__~.?ther. Neither good genes and a/_
poor environment nor bail genes and a good environment will give
us a profitable and satisfactory animal husbandry. We must have
both.
There is still a considerable problem in this field, not the old one of
attempting to determine which is the most important, but rather that
of perfecting devices that will measure the influence of the two in a given
case. Obviously some characters, like color pattern, are largely heredi-
. tary, owiJ,HL!:_Q__t~~nes.; Bthers, like milk in the pail 01'- fat on -the
carcass, are determined both by the genes and their various interactions
and by enyironmental factors-feed, care, etc. - ------
The br~ed~r'·s-ta:skTs-to d~termine to the best of his ability how much
-of the variance in any given case is due to environment and to make due
allowance for it in practicing selectiop.. This is necessary because, as
'we have already learned, only the hereditary variations are capable of
being passed along to an animal's p~ If a group of heifers in
comparison with their dams ha~ belter feed and care, calved at a more
advantageous season of the year, ,,,ere better fed while milking, were not
bred so soon after freshening, etc., with the result that they considerably
exceeded their dams' production, it would obviously be unfair to attribute
all this excellence to their heredity. Better or worse feed, weather, and
other environmental conditions could make a considerable difference in a
crop of lambs or pigs, even though there were no appreciable genetic
differences.
At present, the breeder has no practical tools for separating the genetic
from the environmental causes of variation except hi~ own judgment
applied to individual cases, and his judgment unsupported by factual
material in the form of carefully kept records is liable to be untrust-
worthy. If he will keep records, however, and appraise them intelligently
and fairly, not letting his judgment be warped by individual likes and
dislikes in regard to his breeding animals, he will be well on the way
to avoiding the all-tao-common mistake of assigning to heredity what
rightfully belongs to environment. His success or failure depends to a
large exte~t on his ability to do just this, for the simple reason that
GENERAL CONSIDERATIONS IN SELECTION 1')
hereditary differences may be transmitted, whereas environmental . s
apparently cannot be transmitted.
In nature the gene complexes which would yield the best result in . e
given environment ,,,ere the ones which eventually won out in the ruthless
battle for survi'Vw and perpetuation of the speeies. The environment
probably did not 6ause the original variations, but crossing, recombina-
tions, crossings over, mutations, and chromosomsl aberrations did. The
germ plasm provided the base for variation and the environment acted
as a screening agency.
In practical animal breeding or animal production, the environment is
of equal importance with heredity. "Half the breeding goes down the
throat." If we are going to measure differences in heredity, the environ-
ment for different animals or groups must be ~tS constant as possible.
Likewise, if we attempt to measure differences in environment, we must
do it with animals which are as alike as possible genetically-inbred
lines. There are many examples in animal-husbitndry literature of pure-
bred animals which have been selected to perform well under certain
environmental conditions failing dismally when taken to some different
environment for crossing with native animals. Khishin,l for example,
reports the failure of dairy Shorthorns to hold their own pr.,duction for
more than oile or two lactations in Egypt and to cross successfully with
native cattle beyond the Fl. Dltiry breeds with high inheritance for
milk production have quite generally failed in crosses in tropical countries.
Various admixtures of Brahman cattle with European beef breeds have,
on the other hand, been successful in establishing more successful lines
or breeds for tropical or semitropical conditions.
Animals should be selected and tested under conditions similar to
those in which they are to be expected to perform. If the regime is to
be three-times-a-day-milking, cattle should be tested under those con-
ditions. Cattle which do best under a two-times regime will not neces-
sarily do best under a three-time system and vice versa.
Optimum conditions for performance should be provided. Two or
more groups of animals may gain at about the same rate under limited
feeding thus making them appear phenotypically the same. When
placed on full feed, however, there may be marked differences in their
ability to consume feed and to utilize it for growth. If we expect a
certain quality to manifest itself, we must provide the maximum oppor-
tunity for this to happen.
The Market and Selection.-The ultimate purpose for ,,,hich animals .........
are bred is that of supplying the needs and desires of consumers. The
1 KHISHIN, A. E. F., Twenty Years of Shorthorn Feeding in Egypt, Empire Jour.
of Expt. Agr. VoL 17, No. 66, 1949.
markft pays more for a low-set, blocky, well-finished, and marbled steer
than it does for an upstanding, narrow animal in thin condition, because
the former satisfies the wants of the consumer to a greater degree. Horses
of accepted draft type and cows of dairy type will command higher
prices than animals lacking these outward appearances, because the
buyer feels that they are better equipped to perform their respective
functions. The breeder, therefore, should be a careful student of market
demands. Coupled with this should be a careful scrutinizing of his
records in order to ascertain the type of animals that actually returns
the greatest profit to the breeder, because the animal that brings the
most money on the market is not necessarily and always the one that
nets the breeder the most profit.
Most breeders must make their living through selling the animals they
breed or the products therefrom. The mere fact that they have produced
efficient and attractive animals will not of itself prove profitable if the
market does not happen to be in the mood to buy that particular sort of
animal or product. The breeder who is to get ahead must be a keen
student of what the market demands and is apt to demand in the next
few years. This should not, in the slightest degree, be construed to
mean that breeders should follow every whim of the market and be
continually attempting to change type. Such practice can lead to
nothing but chaos. The successful breeder is the one who can sense the
difference between a passing whim and a permanent change. The market
is not a fixed institution incapable of change. For substantiation of this,
one need but refer to pictures of accepted animal types of 20 years ago
and to consult market statistics of two decades past. The progressive
breeder senses each change somewhere near its beginning and shapes his
course accordingly. In so doing he begins to practice selection for the
new type one or a few generations before his fellow breeders and generally
reaps a financial reward for his shrewdness and foresight. He also, of
course, runs the risk of guessing wrongly-for which he probably will pay
a penalty.
What type changes in draft horses will be expedient or necessary in
view of the increasing use of tractors? What type changes would be
desirable in dairy cattle owing to the likelihood that we will want to
feed them an increased (or decreased) proportion of roughage? Will a
changing feed situation, new kno\vledge of the functions of the endocrine
glands in growth and fattening, or possible adjustments in our system
of governing ourselves make somewhat different types of meat animals
desirable? "Ty" Cobb was a great baseball player because he had the
imagination to outguess the opposition plus the skill to carry out his
plans. He is a great general who outguesses the enemy, a great lawyer
GEN ERAL CON SIDERATION S I N SELECTION 549
who outguesses his adversary in the courtroom, a great scientist who

outguesses "nature." Outguessing simply means that the person had
curiosity or imagination or vision. For real success the livestock breeder
needs more than a sprinkling of this ingredient. Before turning his
F'IG. 167.- Grand Champion Durocs showing how ideals of type change. Above, Wave-
master Stilts, 1931 Grand Champion; and, below, Top Set, first aged boar, Iowa, 1947,
(Courtesy of Dttroc Record Association. )
hand to produce anything, every breeder and farmer should reflect on

this most important of all questions, what will the market pay me best
to }l oduce? Until that question has been answered, there can be no
consistent selection,
Breeds and Selection.-That there are differences between our pure
breeds of horses, cattle, sheep, and swine needs no particular elaboration
-- 550 BREEDING AND IMPROVEMENT OF FARM ANIMALS
at this point, for they are obvious. They arose because breeders have
followed somewhat different ideals in practicing selection over a period
of time and were able to influence the frequencies of certain genes in a
recognizable fashion.
There are wide differences both between breeds and between individuals
of a given breed. Breed differences may be due to the fact that breed 1
is generally AA, whereas breed 2 is generally aa. Such differences do
actually exist, but they are not now thought to be so important as
formerly. Another way in which breeds differ concerns the relative
frequency of a given gene within the two breeds. Breed 1 may be 90
per cent A. and 10 per cent a, whereas another breed may be just the
reverse.0he genes for high-butterfat test are much more prevalent in
Jerseys and Guernseys than they are in Holsteins. This is due largeIy-
to selection. Holstein breeders for many years based their selection
largely on amount of milk. It is well known that there is a negative
correlation between amount of milk and its butterfat percentage. In
other words, in selecting for high milk, Holstein breeders automatically
rejected the genes for higher test, so that the Holstein breed now produces
a milk with about 3.5 per cent butterfat, whereas Guernsey milk averages
about 5 per cent and Jersey milk about 5.4 per cent butterfat. The
relative scarceness of the genes for higher test in Holsteins makes the
job of developing a 4 per cent herd very difficult. If a breed totally
lacked the genes for some desired character, then there would be no other
recourse but to go outside the breed to get these desired genes.
There is nothing in the mechanism of inheritance itself that will change
the frequency of any gene in a breed or population. If we start with
two animals that are Aa (one-half the genes A and one-half, a) and do
not select for gene A or gene a, then after any number of generations
and with any system of breeding, there will still be as many A genes as
a genes in the population. It must be recognized, however, that chance
may lead to rather large fluctuations in gene frequencies in small
populations.
A stock phrase for many years has been, "There is no best breed,"
though "tlltiStat~ment is only true when there has been added, "for all
conditions." There are breeds better suited to certain conditions than
are others of the same class, because certain breeds have been grown
under, and with a view to fulfilling, similar conditions to the ones now
contemplated. Some of our breeds of dairy cattle produce a milk better
suited to cheese making or the production of evaporated or condensed
milk than do others. Comparable situations exist between breeds of
horses, sheep, and swine. Breed tests of various sorts are likely to be
very misleading, because in them it is more apt to be the particular strain
rather than the breed as a whole that is being tested. If enough so-called
"breed tests" were run, under a wide enough range of environments, no
doubt all the breeds would get a share of the winnings. A breeder will
find it profitable to make a study of the breeds in order to ascertain
which one has been produced under conditions similar to his own and has
given the most general satisfaction therein. On the other hand, caution
must be exercised in accepting the offerings of overenthusiastic breed
advocates. This matter of an unbiased opinion and a thorough knowl-
edge of the breed from its earliest inception to the last sale and shmvyard
winning or loss cannot be overstressed. Such knowledge is essential
before a pedigree in the breed can be properly evaluated.
Each of the breeds has certain exterior trade-marks that make it
possible to differenti~te the breeds without any serious likelihood of
error. In this sense the trade-marks are an advantage. They may
also, of course, be a disadvantage if breeders select for fancy points at
the expense of points of utility. In sheep, one breed calls for a long,
horizontal ear, others for short, horizontal ears, and others for an erect
ear; some breeds demand a very heavy face covering of wool, others a
moderate face covering, and still others a clean face; some demand the
color markings on face, ears, and lower legs to be black, others brown,
others gray, others white; some breeds have no horns, others horns in
the male only, others in both sexes.
In swine, some breeds demand erect ears, others moderately erect ears,
others lop ears; in some the face must be greatly dished, in others moder-
ately dished, in others straight; some are black colored with a slight
toleration for white, others white with a complete intoleration for black,
others varying shades of red, and one requires a white belt to include
the front legs and shoulder. The dairy and beef breeds also vary in
color, horns, face. The most acceptable colors of Percheron horses, grays
and blacks, are not acceptable in Belgians; and the ones the Percherons
frown upon, chestnut, sorrel, bay, and roan, are the desired colors in
Belgians.
The breeds should continually scrutinize these trade-marks. Most
of them in themselves are harmless, although some may have positive
detriment, horns, face coverings, etc. Reasonable trade-marks should
be maintained, but care must be exercised at all times that "the tail
does not get to wagging the dog." The principal advantage of the
trade-marks is not that they have any intrinsic merit in themselves but
that they do enhance uniformity of appearance, and their very existence
and recognition reduces the likelihood of fraudulent or careless regis-
tration and so helps to protect the purity of the breeds. The existence
of separate and distinct breeds has two very important functions; (1) in
encouraging cooperative effort among men in the same breed, and (2) in
stimulating competition among the breeds.
Trade-marks generally mean some compromise in selection. The
breeder has to get a little less of some intrinsically valuable trait because
he has to get more of some breed trade-mark. Breed score cards gener-
ally try to allot point scores on the basis of relative merit of the part or
trait considered. Breeders should use this technique in practicing
selection so as to strike the happy medium between paying no attention
to breed trade-marks and paying them too much attention.
In the choosing of a breed, one must reach a decision as to whether
to select a breed already numerous and well established in the region or
to introduce a new and unknown one. Both of these procedures have
advantages and disadvantages. If a new breed is chosen, one has the
advantage of novelty and monopoly but the verYJerious disadvantage
of standing alone and perhaps being far removed from other sources of
replacement. Conversely, if one chooses a b.reed that_ is prominent
in the locality, he can have the distinct advantage of other breeders'
experience 'with the breed and the opportunity of wide choice near at
hand for his foundation animals and later additions. The good ·and bad
features of other herds in the community are more or less generally
known as well as the integrity and dependability of the men themselves.
With the growing trend toward the proving of sires in all classes of
livestock, the exchange of good sires in a region in order to extend their
use over longer periods of time will probably ensue. Having animals
of the same breed as others in the region would allow a new breeder to take
advantage of opportunities of this sort. General adoption of such a plan
would also tend to create line bred herds in the district, something greatly
to be desired in American livestock breeding.
Likewise, the trend toward artificial insemination in some classes of
livestock 'would indicate the much greater likelihood of one's being able
to have the use of better sires if he were breeding animals in a breed
already numerous in a region.
Finally, the greater likelihood of developing a market that would
attract outside buyers and permit the fostering of consignment sales
is an advantage which can grow out of a community's sticking to one
breed. Choosing a breed already well established in a region seems to
hold more advantages to the newcomer than the alternative scheme of
choosing a breed with few or no breeders in the region .
. / Individuality and Selection.-What an individual is and does is called
~
its phenotype, so that selection based on in.diY~an properly be
~ called phenotypic selection. This has been the most commonly used tool
\\ in selecli0n,-alld to it can b7ascribed most of the progress that has been
\
made in shaping nature's imperfect offerings into more......efficient and

de£il~~b10~. N ow that the physical mechanism of inheritance is
known, we can, of 'course, better appreciate the shortcomings of pheno-
typic selec~i?_ll:_1 Breeders used to think (and sOJne still dof that -they
could tell from the appearance (phenotype) of an animal just hmy it
would breed. The informed breeder now knows better than this. He
knows that he may learn certain things about probable transmitting
worth from appearances, that 'he can add to his knowledge from a
pedigree which is relatively complete with phenotypes of the ancestors
and collateral relatives and their offspring, but that actually to ascertain
the breeding merit of 'any animal, he must study the animal's offspring,
in other words, apply the progeny-performance test. ./
All the~siderations are important, though there can be no ,question
but that, Ideally, the progeny-performance test takes first rank.::X·A good
individual and a go~ pedIgree are obviously considerations of great
importance ...........A deficiency in either individual or edigree is, of course,
to be avoided ~vhereverpoSsibie.- "'Selection based on indivi uality alone
or on pedigree alone or both, however, will involve only temporary loss
and setb~k, provided the progeny-performance test is rigorously applied.
A thorough study of its physical make-up is essential in selecting an
individual to go into the breeding herd. It is here that knowledge of
anatomy and practice in judging livestock will stand the breeder in good
stead. There can be no question but that livestock shows are very
influential factors in establishing type, and breeders generally would
profit by more regular attendance at the leading livestock shows. It
should be remarked however, that show standards for various breeds
and judges' interpretations of them are not necessarily infallible. As an
instance, we need but mention face covering in certain breeds of sheep-
demanded by show-ring standards but now thought to be anything but
advantageous from a practical production standpoint. Like other arts,
livestock judging is more or less a gift, but it too can be improved through
study and practice. The animals that go to make up a herd should be
as nearly perfect as possible, but in all animals it is possible to find faults
of various sorts. This being true, wise selection will depend on the
breeder's ability to balance one fault, or several faults, against others and
to choose the lesser of the evils.
In selecting foundation animals, it is essential that there be _~§_I!nJGh_
uniformity as possible in the band of females. If these are of different
types':Tt \~iii~fficult toget a sire which, when mated with
them, will be able to produce a uniform ty~_Qf offspring. In "isiting
herdS or sales for the purpose of securing foundation animals, one of the
cardinal features to be observed and appraised is this matter of nni-
formity. Uniformity of a good sort is a virtue almost beyond price.

One should hesitate, if not entirely refrain, from buying animals out of
a flock or herd that evidences samples of all conceivable types. The
females, bestdes being uniform in type, should be of good size in order
to facilitate prolificacy and to pr0n:l0te effi~. The desideratum is
as much size as is consistent with quality. Size, however, as well as
anything else, can be overdone. If the females of any class of livestock
are not fairly roomy and stretchy through the middle, the most important
feature of all in breeding livestock, viz., fecundity, is apt to be low.
Ins_§lectiUKthe sire to mate with a uniform_band of good-jSized females,
the c~rdinal point to be borne i~~~d is that he shoUld offset anyfa"lIlts-
that are present in the tamales. If the females have a tendency to
sloping rumps, the male should be particularly strong and level in this
respect, as well as being a niember of a family strong in this regard;
if they show weak pasterns, he must stand well up on his toes, etc. If
size and vigor are present in the females, one can logically accept a little
more compactness and quality'in th~ male, especlaJly if producing meat
animals. In the sire, ,there should be evidence of masculinity, but it
should be /emembered that this is often overdone at the expense of
quality, V13re~QQ~ an!L!2ad r must also be considered.
Particularry:-iSit essential in selecting animals "of any 15reed to aVOId
those individuals exhibiting the common breed faults, such as droopy
rumps, small teats, etc., in some of the dairy breeds, too fine or too
coarse bone in some breeds of horses, undue coarseness in some of the
br~ds of swine, etc., through all our classes of livestock.
/Many breeders base their selection almost ,vholly upon individual-
ity. This procedure, however, is nQt~etical!y 1?2u_~<!. !o.I_the :@.a,sQ_n,
explained earlier in thjs ~hapter, that yariation, or variance, is caused by
_ilomi!lllJllie, epistasis, and the environment, as well as by genes acting
~d(litively. 'Mo;t~iperiments--~how that the latter kind of variation
makes
up only one-tenth to one-third of the total variation. This does
not mean that we should make no effort to capture this type of variation,
Fj!§l_hQrses do not always beget fast offsprin_g, but they. are more likely
tQ,than ar§ slow horses, and s~ on for'other qu~lItie-s'-i~~the:;:~la~of
liy~~togk) Phenotype and genotype are not necessarily identities;but
phenotype is a worth-while indicator of genotype. Phenotype is deter-;-
mined both by genes working in a variety-of ways and by an unimag~~ab!x
large number of environmental effects. Certain, but not al~~c:1~
will be transmItted. No environmental effects will be. We cannot even
\ tell by inspectioIl whether a black bovine is BB or Bb. In either cai:>e
"\! it will be black; i.e., these two genoty~elc311g'to'th;;-black phenotype.
From a br~ding standpoinF,Iiowever, these two animals are very differ~
I
~\
",Ii <l
VV\
\'
," \)B-)",
\\
"C-A I
I
r 11
~"
ti .
~'~
'V
1 M3.16~\""{<-:: L
~~~S !I"'~
(1
GENERAL COSSIDERAl'IONS IN SELECTION
ent, for they belong to two different genotypes. When mated to red
cattle, the first, BB, will give all black offspring, whereas the second will
give 50 per cent black and 50 per cent red off"priug. This is a simple
case,!to he sure, involVing only ~aiI" of ~nes, but the saine reasoning
and the same or more complica e genetic principles apply to the more
complex characteristics;'such as speed, milk production, ability to fatten,
H
plus the much greater range of e~Y!EoIlIIleIltJ11 effects. -l1ls alr:r'lght to

demand high individuaf--excelience in selecting breeding animals, but
it is all wrong to stop ,vith that single consideration.
Dairy-cattle breeders ha,:e, to some extent, aped the breeders of fat
stock and horses and set up arbitrary standards of dairy type. There
is a great difference, however, between the selection of beef cattle, for
example, and dairy cattle. In the one case we deal with tangibles and
in the other largely with intangibles. The beef animal, male and female,
exhibits directly the presence or absence of the desired qualities, e.g.,
degree of ~shing, width of loin, fullness. o_fJolJllil.-etc. With dairy
cattle, on the other hamr,-Tnere seems--to-be little correlation between
t_ype and pr~9.u9~iQ!.l in the feQ1ale, and, of course, the bull gives no
milk whatsoever. On the face of it, it would seem to be a very naIve
assumption that straightness of topline or stylish carriage, for instance,
could have any direct bearing on the amount of milk which a cow could
give.
-./ It is extre~ifficJlJt to tell by looking aLa dairy cow how much
milk she is capable of yielding. Many of these so-called '~lillessillg
c~~ have been staged. lOne was held at the New York State Fair
in 1919 which included nine cows ranging in production from 4,454 to
20,852 lb. of milk a year, with about a 2,000-lb. interval between each
pair. Over 70,000 people guessed, no one got over five of the cO\vs
properly arranged, and only 12 people did this well, one-third of the
guesses failed to have even one cow in the proper place, and many
"guessers" picked the lowest producing cow as the highest producer.
So, if it is hard to estimate individual performance by inspection, ho,,,
futile and hopeless appears the task of estimating transmitting abilities
by inspection.)
JIndivic!.u~b~diutermR of individual characters,
groups of characters, or in terms of general average excellence. In hogs
~to cull all animals with weak pasterns or those with
steeply sloping rumps or those with heavy jowls or perhaps those with
all three deficiencies. In doing so, we might be thrmYing away many
genes for ~i.w. fertility, for ~gh proportion of valuable cuts, for large
milking ability, etc. Likewise in Keepmg the good pasterns, rumps, or
light jowls, we might also be keeping low fertility, poor quality, lack of
ability to raise pigs, etc. If selection were practiced on the basis of..
general average excellence, we would perhaps keep some sows with poor
pasterns, steep rumps, and heavy jowls because they had enough other
good qualities to give them an average score that wai-higher than other
sows which might excel in some specific characters. Similar illustra-
tions can be envisioned in all classes of livestock. It is often noted with
a beginning class in livestock judgingthat as specific animal faults are
pointed out, each one in turn tends to bulk too large in the beginning
student's mind, so that an otherwise excellent animal which exhibits this
particular fault is dropped clear to the bottom of the class. Some
students find it impossible to overcome this tendency. Improvement in
livestock h!.dging on the part of other students is merely the growth in
'~biil'ti'to'balance defects or imperfections and arrive at a sound judgment
Q_ased on the total average excellence of the ~in each judging
rlng:----'I'he gooifJudge-sees attthe good anCfU;;-PQ_oIP.0511ts of his animals
but makes his placing on the basiS-ofgeneral balance of all parts. The
good breeder practicing phenotypic selection does the same thing in his
own "Little National or International" that he holds in his own barnyard.
The shortcomings or selection based on individuality can best be
illustrated by calling to mind excellent individuals in fat-stock classes
or horses that ha<V~' failed to transmit their own good qualities to their
~sprin.K:. There h~--p-;;bably'-been 'many animarstllatstooCl at or
near the top in the show ring and at the same time were mediocre or
worse as breeders. Likewise, some animals that stood lower in their
show classes, or perhaps were never deemed worthy of showing, have
proved to be. excellent transmitters of desirable type.
' .....Alloth~~-·~o;::e 'gen(!ral critique of 6e limitatiori.- of selection based on
individuality is t~~~V~~X,!l!,?.'':.. ,.P.~9~,.0!'..Q:_~.rise in productiQQ__in dairy
catthl, .. For many years dairymen have saved heifers from the highest
Pr~ducing females and sons from their highest producing cow. Yet the
average cow now yields 5,000 lb. of milk and 200 lb. of butterfat.
I For qualities which akthighly hereditary, individual selection is the
.§i!nplest and most useful tOOfWe have. Its main limitation arise~-f~~m
the fact that herItabIlIty of the commercially valuable qualities is rela-
tively low. It is a general principlethfl,_t the longer we can delay selection,
the fewer mistakes we are lii~ely to make. li,ve selected ewe lambs the
day-they ',vere Dorh, many mistakes would be made; if we delay until
they are yearlings, fewer mistakes would result; if we could try them
all out for 4 or 5 years and then select on the basis of individuality and
progeny, relatively few mistakes need be registered. So another limita-
tion of individual selection lies in the fact that it is based on young and
untested JUdgments.
\
Pedigree and Selection.-It might perhaps seem logical to expect the

individuality of an animal to mirror fairly exactly its transmitting ability,
but such an expec~tion comes far from l:JeUlg 100 per cent fulfilled.
Like\yise, it might ~em logical for a pedigree to predict breeding worth
\~ith considerable accuracy. If the expectation is built upon pedigree
phenotypes, it will fail for the same reason that individuality failsasn
prognosticator. Actually, the pedigree is often used not even as a group
of phenotypes, but only as a group of names, the estimator actually
knmving little or nothing about the animals represented by the names.
In such a situation pedigree study probably hurts more than it helps.
Many new breeders; after reading nothing more basic than breed maga-
zines for a few months, consider themselves experts at predicting future ,
breeding worth of young animals after a seance with the names in their
pedigrees. Perhaps this partially explains the rapid rate of turnover in
this class of breeders .
./'Pedigrees have limitations' as sel~d&xes. The first one is
perhaps the fact that a pedigreejs a selected group of animals. In a ./
hog's pedigree, his sire illaybe-~ne of Hio sons of the paternal grandsire
and one of 25 sons of the paternal granddam. In cattle the sire may be
lout of 200 ~ons of the paternal grandsire and lout of 3 or 4 sons of the
paternal grAriddam. Tlie sire in any pedigree has a whole host of possible
paternal and maternal half brothers and sisters, uncles and aunts of
th-;;-~~i~al -\~ho;~-'pedigree we are considering and their offspring cousins
to the pedigreed animal. The sire in any pedigree may have a lot of
other offspring, half brothers and sisters to the animal whose pedigree we
are considering and their offspring nieces and nephews to the pedigree.~
animal. Likewise, for the dam, grandparents, etc. ~ pedigree is a tiny,
selected sample of a whole host of animals, the whole being the immediate
family of J.h.~_~l!!Ip._aLIf you were shown a picture of a first-floor door, -
two second-floor \vindows, and four third-floor clapboards, could you
very accurately visualize \yhat the whole house looked like? That's
about how difficult it is to arrive at a reasonable judgment of the probable
genotype of an animal from its direct ancestors, its pedigree.
Not only are the direct ancestors a selected group, but the printed
catalogue material about them is also often of a selected nature. A bull
has 4D tested daughters and the records of the top 5 are listed. A cow
has six records and only the best one is listed. In many fat-stock
pedigrees nothing is given except the animals' names. So pedigrees are
lame as selection tools because the animals and their accomplishments
me both selected. If one had a barrel of apples of which a few were
sound and the rest in varying stages of decay-some being completely
rotten-one would get a distorted notion of the whole barrel if he based
his judgment on someone's picking out 20 apparently sound apples and

then sorting them down to 10 by throwing out those whose rottenness
was underneath and did not show up when the choice of 20 from the
barrel was made.
,j ,Ail prE;se~~'!y_usE:lQ,J pedigrees _are_ !l9t __ y~ry _:t±s"f)f~l t_O_Qls _lI}__:::;~lec~ion.
They can be improved in a variety of ways. The most obvious way is
to print all the knownfacts'a"hOlit-'-eacIi-animal as an individual, i.e.,
give as complete a _E.!i"f)nq!~pic picture of each al}imal as possible. Next
include as many of the close~sslpte,~ti_hdes~ aunts, cousins,
nieces, nephews. True, an animal gets no inheritance -directly from an
aunt, but an aunt is 12.5 to 25 per cent related to the ani~al in question .
. The more close relatives we can see and whose records we can study,
~niore--reasonabIe-rucfgment we can make regarding probable-genetic
. worth of"i:lriUiiiesteaanimar- Collater"ar-relatives'" and direct ~ncestors'
, accomplishments can be evaluated in terms of the degree of relationship
, existing in each case.
The earliest pedigrees were not written in the present-day bracket
form. Instead, the animal's name was set down, and over to the right
on the same line was listed its sire (and still further to the right the
breeder of the sire). Underneath the animal's name appeared that of
her dam and out to the right of the dam was listed her sire and her sire's
bn;ci~n other words, the left-hand column, as you read d"own~ gave
the ammal, his or her dam, his or her . maternal granddam, his or her
maternal great-granddam, and so on, and the' next column to the right
gave the sire of each of these cows. The first way of writing pedigrees,
therefore, gave just the bottom line of the pedigree. In the schematic
pedigree shown as Fig. 168, animals below the diagonal line would appear
in the old-style pedigree; those above the diagonal line would not appear.
We think that the old-style pedigree, properly used, has several
advantages over the present-day bracket form of pedigree. We do not
expect breeders to go back to the old-style pedigree-there is no need to,
since the old-style pedigree is included in the bracket style anyhow (it is
the bottom line).
What we mean by using the old-style pedigree properly now needs
some discussion. Let's start with sire A in Fig. 168. We are assuming
that this i~. a good proved sire. Since not too much can be told from
the type of' it oull"about howrns daughters will produce or look and
because of the shortcomings of pedigrees as production-prediction tools,
it makes it necessary that sire A be a proved sire if we are really to know
about his transmittin_g__.abilitie.s4 .
If sire A lsa~proved sire, that means that good stuff has come out
of him. Now, if good stuff ~as come out of him, it is very certain that
good stuff must have gone into him from his parents-if it had not gone
into him, it could not have come out. If, on the other hand, sire A is a
poor proved sire-poor stuff has come out of him-then it makes no
difference how good his pedigree seems to look (or even actually is). If
there was good stuff in A's pedigree, but A did not get it in the sperm
from C and the egg from D, then A certainly cannot pass it on; he can
only pass along sample halves of what he actually got from his parents.
So, we repeat, let sire.f. be a proved sire, and whether pro;ed goocror-
proved poor, what is appa'rentfy III back of him in hif3_.pedigr~_cuts
relatively little figure.. .,c- --
ANY
AjIMAL,
DAMB
MATERNAL GREAT-
l
MATERNAL GRA],;DSIRE 1\1
GRANDDAM
F ;\1ATERNAL GREAT-
GRANDDAM N
FIG. 168.-Schematic pedigree showing" old-st:de" pedigree below diagonal line.
(If, on the other hand, sire A is a young, unproved bull, then we have
got to try to guess from his pedigree as to what he got and we would do it
by the method we are describing.)
We would apply the same line of reasoning to maternal grandsire
E (Fig. 168) as we have just applied to sire A. If grandsire E is a good
proved sire, then what's back of him cuts relatively little figure.
And likewise for maternal great-grandsire M.
In short, if sire A-grandsire E-and great-grandsire M, are good
proved-sires, then we can forget about what's back of them-the material
above the diagonal line in Fig. 168-and concentrate all our efforts and
study on the material below the diagonal line.
This may sound a bit revolutionary, but actually it is not. When
.1)60 BREEDING AND IMPROVEMENT OF FARM ANIMALS
we say that if sire A (and grandsire E and great-grandsire M) is a good

proved sire, ",e.ea,n "forget what is bac.!.of him," we are not'""lICtuarry
- forgetting anything. If sire A-iS-a good proved sire, there must have
been good stuff in his ancestors-that is the only place from which he
could have gotten it. But, if sire A is proved good, then the goodness
behind him has been focused or brought together in him. When it is
once brought together and proved in him, we know it is there and so
need not go behind him to his ancestors. That is all we mean by saying
that we can forget what is behind him. We do not actually forget wh&t
is behind him in his pedigree-we use it-but we use it where it is focused
or brought together in sire A, rather than where it was diffused out in his
pedigree.
Now, let's look at the female side of the bottom tine of Fig. 168.
/'We will start with the dam B. As indicated earlier, we can learn
something about how_she will transmit fro~~vn type aiiQj)roducti~n
records-'=-we can learn -Something more perhaps from studying her"-p;ti'"'
gree~hln.-the real proof of how she can transmit can eome only from
h'o,v she does transmii,___..In short, until dam B has had some records,
and some ~offspring ,~ith records, we can't really know what kind of a
producer and transmitter she actually is. We demanded that sire A
be a good provea~nt to take as much of the guess
as possible out of breeding, we must also demand that dam B be a good
proved cow. If dam B has transmitted well to a couple of daughters or
"ons (the more the better), there is a fairly good chance that she will
continue to transmit well to her future offspring (half or full brothers or
sisters to those she has already had).
We, of course, apply the same reasoning to granddam F that we have
just applied to dam B. She is older and can and should }fave more
offspring than dam B. Granddam F's other daughters or sons are, of
course, aunts and uncles to the offspring of dam B, and any offspring of
these aunts and uncles are cousins to the offspring of dam B.
The same line of reasoning applied to dam Band granddam F we will
also apply to great-granddam N. ,
, The trend to'ward printing all the offspring of the females on the
bottom line of the bracket-type ~ee, together with their type and
production accon{I'~ to be highly commended. All these
animals taken together make up the female ~rnily;alld when we have
this data, we have a much wider anamore ~ecure foundation on which to
base our judgment of probable genetic worth than we do when we have to
base it oIljiist'danrnnrrmatemaI granadam ana great-granddam.
We like to think of this femaleootr6m line of a pedigree as a stream
which flmv~\in Fig. 168 from great-grandd~'anddam F to dam
\
B, and finally to the present" Any Animal" being considered in Fig. 168.
If great-grand dam ·N is a great female herself, in ~and production,
and has had sev~ good 0ff_s.p_ring~_among them_~rl1Ili~~!ll F~-and if F
in turn has had several good offspring, among tIiem dam B, and if B
has had one or some good offspring, then we can be sure that the stream
(which is actually the female family) is a good one. j
Now if, in addition, great-~as a good pr~re and
poured into the stream tne-nereditar ,?-eterminersrorg;od type, highl
production, disease resistance, regularity of bree lng, longeVIty, and the \
ot'fier good things which we want III our herds, and If maternal grandsire. . .
E "andSire A did the same thing-then the good maternal stream which
we started with in gre:1t-granddam N will have had the best possible
chance to become en.r;lChed as it flowed down the generations.
Many breeders seem to consider all the l~als in a three-generation,
bracket-type pedigree as about of equaflmportance and to base their
judgment largely on the individuality (type and/o~_production) of those
14: indw~ocecrure-rs-gOoaaS far as it goes, but not so
complete and efficient as it might be.
lt would be much more efficient, we think, to work with only the six
animals concerned in a three-gen~, old-type pedigree-the bottom
line, but to get all the in1umnl:tlOn available about these animals. Study
sire A thoroughly: _ /'
What kind of females was he bred to (production and type)?J'
What kind of offspring did he get? ./ ..
Did the offspring and dams have similar environments (feeding, etc.)?
Has he any sons that are transmitting well? J
Has he any full or half brothers or sisters which have done well? ,.J
Perhaps sire A has 30 daughters, 5 sons, and 30 or 40 paternal and
maternal full or half brothers and sisters. So, instead of just studying
sire A as an individual and learning about a few of his offspring, we should
study carefully and completely not only sire A himself, but some 70 or
80 of his offspring and close relatives.
We would carry out similar studies on the other five animals-B, E,
F, M, N, in Fig. 168. We are going to take the really functional part
of a pedigree and study it as completely and exhaustively as possible.
So, while we advocate discounting to a certain extent about half the
animals in a bracket-type pedigree (those above the diagonal line in Fig.
168), we also advocate doubling or quadrupling the amount of study and
thinking we give to those making up the bottom line of the pedigree.
T.he. bO. ttom. line.of the pedigr.e.e-t¥ old-style pedigree-is really the
"Functiop-a~ :part of the Pedigree." ",hen used intelli e etting
alrti~~ relevant d1~ta p()Hsihle, and in conjunction ~v· selection for goon
type and production in the females involved, we have the presently most
valuable tool for arriving at the most probable transmitting values of
our livestock.
_" Breeders have traditionally used three mesns to try to evaluate the
paRsmittiHg abjlj~ of their animals, and their availability to the
breeder in point of time places them in the following order: (1) pedigree,
(2) individuality, (3) actual breeding performance. When properly used,
lall these methods are lJseful in breeding better livestock, but they all
'have certain limitations. We can summarize by saying that, from the
Istandpoint of hereditary transmission: /'
, The pedigree of an animal tells us what the ~tnimal ought to be. /
V"'
The individuality of an,.animal tells us what the animal seems to be.
The bree..ding peI!!!!!!!!J/nce of an animal tells us "'hat the animal actu-
ally is. .
L njustified merit i's sometimes assigned to any and all animals of
certain strains or families. In common vernttcular, this is known as a
"pedigree craze." They are sometimes propagated for the sake of
res~petition, and the livestock breeder will do well to investi-
,gate thoroughly before shaping his course by them. Utility rather than
'popularity will, in the final analysis, be the criterion by which all classes,
breeds, strains, and individuals will be tried. This should not be con-
strued in any way to mean that certain strains or families are not better
producers of speed, milk, flesh, wool, or ,,'hat not, than are others,
because such an interpretation ,,"ould be entirely removed from the truth.
In planning matings, pedigree study has 11 legitimate place. Hard
and fast rules cannot be laid down.· It is a matter of reflection and
judgment requiring intimate knowledge of the dominant characteristics
of the strains in question. Some strai;us, are too fine, others too coarse,
some .Qlature quickly, others -Slo\vly,' b~tterfat percentage is high in some,
low in ~mate wisery, s~ in all probability ther-e will be
blended in the offspring some of the extremes found in the parents is an
art in the achieving of which pedigree study has its legitimate function.
It will readily be recognized that the number of animals in the most
efficient strains is more or less limited. It is also a fact that because of
their recognized merit there is a large demand for them, and they con-
sequently sell for high prices. To the small breeder, this is discouraging,
but he need not be without hope. These, to be sure, are proved strains,
though, when their origin is studied, it is found that they arose more or
less by chance through the mating of two particular individuals. Each
parent probably supplied some previously missing factor or facton"
which when combined into one individual yielded an animal of much
greater, \~orth than either parent and, moreover, an animal capable of
\
\/ \
GENERAL CONSIDERATIONS IN SELECTION 5()3
transmitting its own good qualities. It would be foolish to believe that

what has occurred cannot be repeated. The small breeder with limited
capital may, through a proper blending orstrains, produce-a-~~~!1!n more
beauiifulaiid more efficient than anything yet known. This does not
mean that any .breeder can be careless in selecting foundation stock or, "
in general, expect to get something from nothing. He should, by all
means, selectfrom proved strains as far as his finances will allow, because
the isolation interpretation explains much more logically than does the
mass-selection interpretation the progress that has been made in the
past. When an outstandingly good animal arises from mating two more
or less mediocre animals, the progeny-performance test must be very
rigidly applied.
The greatest lack in animal breeding today is the.~ of records.
Until this deficiency is remedied, pedigrees will continue to be of rela-
tively little use in enhancing the accuracy of selection. ,_,Attempts at
remedying this situation are being made'oy some of the purebr~ breed
9i
associations by mean~ rec~f performance, but the problem is a very
difficult one and a-complete solution is not to be expected immediately.
Any individual breeder, however, is at liberty to set up his own system
of record keeping at once, which greatly enhances his chance of successful
selection in his own herd. Such systems of record keeping will be indi-
cated for each of the classes of livestock in the next three chapters.
Even when pedigrees can be made relatively complete from the stand-
point of records, we must realize that we can probably never have a
complete knowledge of the genetic make-up of any animal, and, even
if we did, we would still have to face the sampling nature of the hereditary
process, which would make it uncertain which member of each heterozy-
gous pair of genes any individual actually received from its parents.
Whether we gain more than we lose by culling a certain fairly good
individual because of some fault in its pedigree is an unanswerable
question. Again, as ,vas said of judging or estimating individuality, we
must try to strike a proper balance between individuality an~pedigree. .
The br,e~<!~r'~be to sel~ct good individuals. Co~~lete
pedigrees in terms of records of perrormance would be of inestimable
bimetlt' t~im in deciding to keep one or another-of two-lrldivTduals of
approximately equal merit. Complete pedigrees can be of great benefit
also in arriving at an estimate of dlfferenffemale lines--\'~lthiri a breeders
own herd. W1l(m pedigrees can be made relatively complete in terms
of recordS' of performance, one will alsonave considerable data on the
collateral relatives of each individual, and, although it is difficult or
impossible to assign definite numerical values to records of collateral
relatives, they will, nevertheless, be very valuable in a general way.
564 BREEDING AND IMPROVEMENT OF FAR11~f ANIMALS
Progeny Test and Selection.-The idea of progeny testing as an aid in

selection has lately been revived and given mu~h publicity. Like- many
other general ideas, this one is not new, having-'been advocated by a
Roman, Varro, 2,000 years ago and having been used by Robert Bakewell
in the eighteenth century in his practice of bull and ram letting. The
EE.2Ke.!lY t~st findsJts great~sefulness with characteristics that-ea~
be expressed by only one sex, e.g., milk or-egg production, although there
is nC;r-;;~;o~ ~vhyit ~;'nnot be used: to advant~~ \;;th-~ny class of live-
stock. Like any other selection device: this one has both advantages
and disadvantages.
Owing to the relatively few offspring_~f_~ost females, it cannot in
general be applied to animals of this sex~ Even with males their progeny-
-performance ratirig must come relatively late in their lives, after a
sufficient number of their offspring have been born and reached such an
age that their own performance has had time to be measured. Care
must be exercised that excellences or deficiencie~ which are actually
environmentally caused be' not ascribed 'to inlieritan-ce in either the
offspring'or the'·parents. Allowance must be made for the fact that the
females to which a male is bred are generally a somewhat selected group.
It can thus happen that a breeder can be making progress, even though
.-
the total offspring of each of a series of males averages lower than their
dams. We must,not forget that the hereditary process is of a sampling
nature, and we, therefore, will need enough comparisons to make reason-
ably sure that we'have a fair sample of any male.'s inheritance.
Even with the J:tbove limitations, progeny testing is one of the keenest
tools available t~ the breeder. It should be used to supph~ment indi-
vidual and pedigree selection-not to totally displace them, as is some-
times advocated. The simplest type of progeny testing consists of the
average record or m~oCa sire's offsptin!1;. If the females to which
two males are bred are an average of the breed as a whole, then such
progeny records can be used ·to measure the breeding value of the two
males concerned. If, however, the mates of one male were considerably
above br'eed average, those of the other considerably below it, then such
direct comparisons would not be jus~ified.
Among the farm mammals, progeny t~f3ting has been used most with
dairy cattle, a little with swine, and is just begrnning to be used with the
other classes.. In dairy cattle; pr'Ogeny testing has eventuated in several
systems of indexing bulls for their level of transmitting ability for amount
QLmilLand. butterfat percentage~'- The sys-tem generaI~se(f can be
illustrated by tIle ioitowi~rn:pte.""' Suppose a bull is bred to 12 cows
whose average production is 7,000 lb. of 3.8 per cent milk, and their
daughters by this sire average 9,000 lb. of 4.0 per cent mille Presumably
tj
this bull has raised the production of his daughters by 2,000 lb. of milk
and 0:2 per cent of butterfat. ~ecause these qualities are apparently
contron~~L~iple genes, (~e-~ssun{e- that-the-~sl;:e-- and damai:e
jointly and equ~le for the production of ~~ers,
in other words, that the daughters' producti~l!_ falls half~~_l>etween_
the two parental levels. KuO\vrng--the lever of production of the cows
and ili:eii- daughters ena15les us to estimate the transmitting level of the
bull.
Butterfat
Pounds
percentage
Dam's production ........................ . 7,000 3.8
Daughter's production ............. . 9,000 4.0
Bull's index ................... . 11 ,000 4.2
It is obvious that before such figuring is justified, the records of both

dams and daughters must be corrected to some standard basis. This
is accomplished by means of various and sundry conversion factors. It
is tacitly assumed that if the bull 'were sold to another breeder whose
herd production also averaged 7,000 lb. of 3.8 per cent milk, the resulting
daughters would also average 9,000 lb. of 4.0 per cent milk. This might
or might not be the case, depending on many things but especially on
the genetic make-up ~~-in-YQlved. --"Many different gene
combinatio~nceivably result in a-production of 7,000 lb. of
3.8 per cent milk. Whether this proved bull would" nick" equally well,
better, or worse in the second herd could only be ascertaine~trying_
The U.S. Department of Agriculture Bureau of Dairy Industry proves
bulls on the basis of 5 daughter-dam comparisons. Some investigators
urge at least 6 comparisons, others 8 or 10. Obviously the error will be
cut down as more daughter-dam comparisons are added, because the
extreme variates will have a tendency to cancel each other, thus giving
a truer average. If a breeder can successfully standardize his records-
smooth out the environmental differences-and will test the first.6~_8.,.__
or. 1..Q_da~hters as ~_lJ.ey lJ.,ral)Qrn in his herd without any selection, he
will get about as good a genetic picture of his bull as he would with 30
or 40 daughters.
Attempts are now being made to work out comparable and practical
systems of indexing males in the other classes of farm livestock. When
available, such information should aid greatly in taking some of -the
"guess" out of breeding. And, even though most males cannot practi-
~~~:v. ~e ind~xed until after they are d~ the"illforma,m can appear
III _pedIgrees, 5YIi§!i makes t~e~ of much greater value.-~
Probably the gr~est-'a(fvantage tot~~progeny testihg
566 BREb'nB'G Alf_D IJIl']WVb'Ml#Nl' OF FARM ANLMALS
" ~-
will arise from the records themselves rather than from their manipulation
as indexes.- When records are-Kept and intelligently studied, selection
is bound to be benefited. Breeders too often depend on a somewhat
unreliable memory and are .prone to have favorites among their animals.
Even a condemnatory record as a breeding animal at times fails to
outweigh a breeder's fancy for some particular pet. Actually, of course,
if one is not going to use the records, he is simply wasting his time in
getting them./ Progeny testing must be based on records, which, if
carefully compiled and intelligently and, on occasion, ruthlessly used,
can be the surest guide that a breeder can have.
Mass or phenotypic sel~9..n was practiced with poultry at the Maine
Agricultural Experiment Station beginning in the year 1899. From this
time until 1907, only those females were used as breeders that produc,ed
at least 100 eggs in their first year, only males ~whose mothel:s laid at
least 200 eggs in their first year. Under this system of selection, the
yearly egg production decreased steadily. The point to be noted here
is that mass selecLion failed to increase production in the hereditary,
material involved. Whether this result was due entirely to the breeding
system involved or in part due to unsuspected environmental changes,
we do not know.
After 1907, the system of selection was changed, All females were
now selected from high-producing mothers, the !emale ]j:-~ny of ,VhiCIl"--
'were all high producers, and, in c'itse- such a feni~tIe failed to yield high-
proaucmg' progeny The first year, she was not retained in t.he breeding
flock. Males were selected from hj.g.h-vroducing mothers whose female
progeny;Vereatr1ligh-prOducers, and a male was discarded immediately
iNris progeny fatted to be high producers. Pedigrees were also kept and
proper attention was given to the blending of blood lines. With such a
system in vogue, ~he production increased steadily from 1908 to 1920.
Mass selection alone in any given case might or might not yield success-
ful results. In order to reduce the chaupe of failure, progeny testing
needs also to be invoked. Certainly much of Our livestock improvement
has come about through the intensive use, often involving inbreeding,
of animals which have demonstrated their breeding worth through their
progeny.
Family and Selection.-The term family is used to cover ,a wide degree
of relationship. In taxonomic classification it occursifi the series phylum,
clas-S;orcte'f,lamily, genus, species. Such family members need be very
slightly rell1t.~d and only so in relation to very distant common'ancestors.
<7'rhe opposite of this is found in poultry where a famIly Often means a
set of_ full sibs. In animal breeding the term family can mean remote
de~c~n~well-known sires or cowsand having'but slight relation-
,'- - - -_ .. _-------_.- --~------ .
ship. Or again it may mean a fairly closely related group of animals·

liiiebred to a certain anlmru.·· ¥JuaIlY,· the· term l!E!!d!:JL~an mean the
two- or three-generatiOn series of offspring and descendants from a
f~undatJ.on female, cow famIly, sow family, etc. -
--tr?"to iio\v'seTectionYha~-been pracuceupreponderantly on an individual
basis. In order to make selection more effective, its base now badly
needs to be widened. This means that family selection, in the last-l
named sense, needs to be added to individual selection. In doing this,
we must necessarily compromise to some extent. Whether to select an"
excellent individual from a relatively poor family or a relatively poorer
individual from an excellent family poses many problems. Ideally, of
course, ,ye would Ei~.lfct excellent ipdivjduaJ-s from excellent female..
fam~
~If \ye ~o back two generations on the bottom line of a pedigree, other
offspring of this female rna ernal granddam) are aunts and ~cles of
the animal being considered and their offspring are ~ns. Other off-
spring of the fiI;~generation female (dam) are full or half brothers and
sisters, and t!;l.elr offspring are nieces and nephews. The whole group
makes up the female family, and the average merit of the should
be considered in selection, with relative w . emg assessed according
to ......d~g;~;Trcloseness of relationship; half cousins, Y24; full cousins,
uncles and aunts, nieces and nephe,,'s, Yz3; half brothers and sisters,
Yz2; and full brothers and sisters, ~~. .
The breeder's main problem is that of recognizing favorable variations
among his animals and assigning them to their proper source and cause,
so that he may lay plans to captllre, perpetuate, and increaae them.
This is an intricate problem at best, and in small herds or flocks it must
be accomplished with rather inefficient tools. There are many pitfalls
along the way.
In small herds or flocks using one male at a time, the offspring for a
year or two will, all be by this sire. There will be differences in these
offspring which are likely to be attributed to their dams or female families.
Since the males in use are far from homozygous, many of these differences
should probaDly be attributed to them, but "'hich ones or how much of
those observed, it is impossible to say.
The variation between f3mi.l~s is made up of the additively genetic,'
the env~ genetic sorts '\'hich are similar· among members
of tKe same family but different between families and the random
environmental. In large families, where the members are closely related,
the environmental ana nonadditive genetIc vanabons are small betweeii'
families and the' . similarities among family members exceed their
phenotypic similarities, family selectIon c -ost effective.
568 BREEDING AND I11,fPRO VEMEN '1' OF FAR"\! ANIJfALS
We think it advisable for every breeder to chart his herd into its
respective female families. This is not a difficult task and generally
will pay good-sized dividends.
The simplest way to arrange a herd into its female families is probably
to make out a little card for each female, showing the a~imal's name at
the top, her date of birth, her sire and dam. A)Vhen such c~~Js-have been
mane out -for all the remal~ have existed in a herd for the past 5,
10, 15, or 20 years, they should be sorted, putting the oldest animals on
the top of the pile and the young females born last week down on the
bottom of the pile. Next secure some large sheets of paper, at least
2 ft. or more wide and a foot or more in depth. Enter the name of
the oldest female in the middle at the left-hand side of the large sheet
and her daughter just at the right of her name, with the daughter's sire
above her own name. That card can then be dispensed ~with. Next,
look down through the pile of cards and see if there are any other offspring
from this old female. If so, they can go in the column to the right of
the foundation female and a straight line can connect dams and daughters.
In this way, the female family will build up across the page, starting
with the old foundation animal and showing her daughters, grand-
daughters, great-granddaughters, and so on, in columns as we move to
the right across the page. This whole array is a female family. We can
guarantee that no matter how well a breeder knows his herd, he will
learn many things he didn't know by working the herd up into its female
families. flame fm.llQi_es are better producers than others; some families
a,re more regular breeders than others; some families ar! itm__ger live
some families are better type, and so on. Qualities do run in families,
as we all know in a general way. What ~ must learn in a
specific way is the nature of the things running in his female families.
Only in this way can his selection be placed on a sounder basis.
~ Production, Type, and Show Ring in Selection.-Production is generally
thought to be closely associated with type, which in turn is determined
by the market as well as by show-ring standards. With meat animals,
type is ll1easured by the relati~portlons bet\veen hams, legs, rounds,
loiii.~l rIbs, bellies, and shoulders, as well as their proportion to the less
ViiUable portions of the carcass. Type, in other words, in these classes
is--production, and to a certain degree it is evident to the eye before
slaughter and is evidenced by both sexes. In horses, certain types of
build have been found by experience to be better suited for certain
performances than are others, although temperament is also of great
importance. In dairy cattle, there is somewhat less correspondence
between external appearance and production, for the latter is due largely
GENERAL CONSIDERATIONS IN SELECl'IO.\' 569
to the physiological functioning and coordination of organs and glands

located internally.
An investigation by Gowen of the Maine Agricultural Experiment
Station is of interest in this regard. Figure 169 shows the relative
efficiency of even short records as indicators of yearly production as
compared ydth the total score of the animal and the score in several
respects commonly believed to be indicators of productive capacity.!
Pausing to consider briefly these facts, we see that a seven-day test might be
considered as an objective test, whereas the value of the points of conformation
are a subjective test for the cow's producing ability. That is, the seven-day
CONFORMATION IN RELATION TO 365-DAY MILK YIELD

? Day ond 365 Day Test (Some Test) • • • • • • • •
? Day and 365Day Test
Total Score
Milk Veins
Size and Qualily of Udder --
Size of Rear Udder
Body Shope and Paunch Size
General Appearance
Thighs Flat and Well CuI Out
--•
Rump Length
•
FIG. 169.-The length of the black bars indicates the relative value of the different points
as measures of milk-producing capacity. (After Gowen.)
test is simply dependent on the reading of the scales that weigh the milk, whereas
the use of a scale of points to judge a cow for milk yield depends, not on any
external scale, but on the mental processes or mental ability of the judge so to
balance his cuts as to show the true worth of the cow. From this it follows that
the conformation of the cow as a measure of milk yield would in all probability
be subject to the personal bias of the judge.
This is, in fact, shown to be the case in study of the records. There are nine-
teen men, all well-trained dairymen, who have judged enough cattle with milk
yield to make a test of their ability as judges of these cows for milk yield. Nine
of these men clearly could judge dairy cattle by the score card and select the
better milkers. On a scale ranging from 1 to 0 and 0 to -1 (correlation scale)
these men varied from the most accurate judge of milk yield from the conforma-
1 GOWEN, J. W., Report of Progress on Animal Husbandry Investigation, Maine

Agr. Expt. Sta. Bul. 299, 1921, pp. 87-88.
570 BREEDLVG AND IMPROVEMEN1' OF FARM ANIJIALS
tion of 0.614 to the least accurate judge of -0.098. The average ability of these
men is 0.246, or they are, on the average, about 25 percent better judges of dairy
cattle for milk production than the average trained dairymen. Clearly, some
men are good cattle judges. Equally clearly, some men cannot judge cattle for
milk production by the use of the score card. Such being the case, the individual
man will do well before he selects cattle by their conformation alone to make a
sufficient test to convince himselfthat he is one of those gifted men who can
judge dairy cattle.
On the other hand, almost anyone can weigh milk. No personal equation
need be present in recording the weights. Such being the case, when they are
obtainable the dairyman or buyer would do well to consider the milk yield
carefully in selecting dairy cows as indicated by the figures above.
Table 36 shows the productimi by classified groups for several breeds

of dairy cattle on a percentage basis calling" Excellent" 100 per cent.
It is evident from Table 36 that the higher scoring cows in all the dairy
breeds are the more productive ones. However, there are enough good-
type, low-producing cmys and poor-type, high-producing cows in all the
breeds so that the actual mathematical correlation between type and
production is usually rather low-of the order of about 0.2. This means
that selecting for good type would not necessarily give us high production
and selecting for high production would not necessarily give us good type.
If \ye want both, we must select for both and that means we must do
some compromising, taking less of one in order to get more of the other.
TABLE 36.-BuTTERFAT PRODUCTION AVERAGES OF CLASSIFIED DAIRY ANIMALS IN

TER;){S OF PERCEN'l'AGES
Classification Brown Average

groups
I Ayrshire Swiss
Guernsey Holstein .Jersey
of all
I
Excellent ..... . . ... 100.0 100.0 100.0 100.0 100.0 100.0
Very good ......... 91.8 89.6 92.4 92.2 95.2 92.2
Good plus ......... 87.1 81.3 86.3 90.5 92.7 87.5
Good ........... .. 83.2 75.7 84.4 85.4 89.8 83.7
Fair .............. 80.2 ..... 82.5 81.1 86.9 82.6
I
Correlation studies between type and production in classes of live-

stock other than dairy cattle have not been numerous. As would be
expected, a fairly high correlation has been shown between weight and
draft power in horses. In meat animals, type presumably is production,\~
though many factors are concerned in the question as to whether the ~
best type of animal will net the feeder the greatest profit, among the
most hpportant of which are the initial cost, the rate of gain, and the
GENERAL CW·iSIDERATIONS IN SELECTION 571
efficiency of feed conversion. That there are considerable differences

among animals in their ability to convert feed into flesh is well known,
but there seems to be no evidence that this is associated with any specific
type of characteristics.
The show ring has undoubtedly had a major place in shaping the type
of all the classes of livestock. The fi~t American show was held at
Pittsfield, Mass., in 18lO, and since that time many thousands of animals
have passed before the judges' eyes. As now constituted, our livestock
shows have both advantages and disadvantages from the standpoint of
breed improvement.
Among the disadvantages are the follOlying: (1) a possible lack of high
correlation bet\yeen showyard winning and efficiency of lifetime pro-
duction; (2) a lack of opportunity to measure the transmitting abilities
of the winning animals, except in a very limited manner through the
"get of sire" and" produce of dam" classes; (3) the fact that" fitting"
for the show ring often demands so great a departure from their" natural)'
condition by laying on great stores of fat (\yhich often results in temporary
or permanent sterility), removing a goodly portion of the \yool in blocking,
etc.; (4) the practice of remedying defects by surgical means; (5) the
keeping of animals from productive \York or from reproduction in order
to enhance their likelihood of winning in the shmy ring; (6) the fact that
elever fitting and showmanship can cover up weaknesses of various sorts;
(7) the fact that commercialized show herds making a circuit of shows
tends to discourage or prevent smaller local breeders from showing;
(8) the fact that the offspring of top show winners are apt to be sought
as sires more or less regardless of their own or their winning parents'
actual genetic merit,
/Yet in spite of these limitations, the show ring is the best medium
yet discovered for molding breed type:-' This is probably its greatest
advantage, but it has others, among them: (1) It brings breeders together
for exchange of ideas and experiences; (2) it serves as the best advertising
medium for both the breed and the individual breeder; (3) in a limited
way at least it helps in discovering and popularizing the better genetic
materials in the various breeds.
T~ng_:wmld be of greater service as an adjunct to breeding
and selection (1) if it did not demand overfitting; (2) if the judges in
all cases gave orally their reasons for their placings; (3) if in the classes
~f meat animals the animals could be slaughtered for carcass and cutout
values; (4) if in horses a pulling and behavior test could be invoked,
although it is realized that ~\Ye want a steady all-day tractive exertion
of 150 lb. rather than one of 1,500 lb. for 1 or 2 minutes; (5) if more weight
could be given to the older breeding animals' actual accomplishments
as breeding animals; (6) if in dairy cattle a satisfactory combination

based on both type and production could be utilized.!
In general, the livestock producer is interested in the type of animal
that will make him the most money. Whether this is the type that is
currently winning top honors at our livestock shows needs investigation.
Since the show does wield such a great influence on type in the breeder's
mind and since the breeder is interested primarily in profitable life~ime
production, these two angles need constant scrutinizing and for the best
interests of all concerned should be brought and kept in line.
Sometimes men waste time and effort in selecting for some mark that
is supposed to be positively correlated with productivity. The best
example of this was perhaps the old escutcheon theory, through which
some men thought they could foretell the milk yield of dairy cattle by
means of the hair pattern found on the rear and inside of the thighs.
It has never been possible to prove the slightest connection between
these two things, 'which in the light of common sense is not surprising.
Some men think that a slight hump in the middle of a cow's back, a
notch at the tail setting, or a heavy shoulder, etc., are correlated with
high production. Selection based on all such things as these is missing
the mark by a very wide margin.
,r Fertility and Health in Selection.-Earlier chapters of this book were
devoted to fertility and sterility. Since, with most classes of livestock,
usefulness and profitableness depend so largely on the number as well
as quality of offspring produced, this is a matter that should receive a
breeder's close attention in selection. In the purchase of mature breeding
animals, one should ascertain their actual breeding history. If this is
questionable or deficient, one would in general be much better off not to
buy. In the purchase of mature males, semen tests will reveal the
likelihood of fertility, though they cannot guarantee it in any special
instance.
In the purchase of young ammals with no breeding histOrY, one must
rely on th~rformance in this regard of its clo-se ancestors and collateral
relatives. Rate of fertility is, in the final analysis, controlled by the
genes working through the endocrine glands and the genitalia and, of
course, is greatly influenced by environmental conditions. Noone cares
for twin foals, few for twin calves. In sheep a certain proportion of
twins is desirable, and in swine we want moderately good-sized litters.
One should select, therefore, in a naturally fertile strain but avoid
extremes. The size of the immediate litter in which a boar may have
been born is of some consequence, but the fertility of the strain or family
1 Such a proposal has been formulated by W. W. Swett and R. R. Graves of the
Bureau of Dairy Industry and is published as U.S. Dept. Ayr. Misc. Pub. 409.
GENERAL CONSIDERA7'IONS IN SELECTION 573
is of great importance. Males in the other classes of livestock should be

selected from fertile families and from dams that were regular breeders
throughout their lifetime.
Much work has been done in plants in breeding up disease-resistant
strains. In animals, there is little comparable to this as yet, but there is
evidence that some strains of animals are more resistant to certain
diseases and other troubles than are others, and in some instances at
least this appears to have a definite genetic basis. Card and Roberts
have reported experimental evidence that seems to prove that" resistance
and susceptibility to infection by Salmonella pullorum (in poultry) are
due, at least in part, to the existence of dominant hereditary factors."
Lambert, Speelman, and Osborn 1 have reported considerable differences
in resistance to encephalomyelitis among the different breeds of horses
at the U.S. Range Livestock Experiment Station as Miles City, Mont.
In nature, disease susceptibility tends to be self-limiting, for the weak
strains are not apt to leave so numerous a progeny as are the more
resistant strains. In man this weeding out of the unfit and their bad
genes is interfered with through hospitalization and the marvelous
accomplishments of the medical profession. Weare making beginnings
along this line in our animals by means of vaccines and other preventives.
Whether or not this is a wiser scheme than breeding and selecting animals
with a natural immunity because of their genetic make-up, the future
must determine. Some diseases have no doubt been conquered by
genetic means; others, like bovine tuberculosis, we have practically
conquered by destroying animals that reacted to the test; others, like
hog cholera, we have learned to live with through the help of vaccines.
Human and animal disea.ses must be controlled or conquered if man is
to persist on this planet, and perhaps there is no one best method for
use in all cases. Since it is difficult if not impossible to make money
from diseased animals, the intelligent breeder should use all the means
at his disposal for securing naturally fertile, healthy, vigorous, and
resistant animals and avoid at any cost the introduction of disease itself
or susceptibility to any disease into his herd or flock.
In this category also, we might call attention to certain known lethals
in livestock, such as "bulldog" calves, cleft palate in pigs, skeletal defects
1 LAMBERT, W. V., SPEELMAN, S. R., and OSBORN, E. B., Differences in Incidence

of Encephalomyelitis in Horses, Jour. Hered., 30(8) :349-352, 1939.
See also WEBSTER, L. T., Heredity in Infectious Disease, Jour. Hel'ed., 30(9);
365-370, 1939, GOWEN, J. W., Contributions of Genetics to Understanding of Animal
Disease, Jour. Hered., 28(7):233-240, 1937; Murphy, J. M., et al., Comparison of
the Incidence of Udder Infection and Mastitis in Two Cow Families, Cornell Vet.,
Vol. 34, No.3, July, 1944.
in lambs, and closed colon in foals. These abnormalities are due to

specific genes. In selection, some attention should be paid to them,
and every effort made to avoid their introduction into a herd or flock.
Although in no sense a disease, such characteristics as red calves in
Holsteins or Angus, bay color in Percherons, too light a red in Durocs,
black skin spots in Chester Whites, etc., are from the purebred stand-
point undesirable and should be guarded against by means of records
and pedigrees as much as possible, although such things are often fairly
successfully covered up. The same may be said for any character
that detracts from an animal's quality as a member of a certain breed,
Although many of these things are inconsequential from a utilitarian
standpoint, they are of consequence from the standpoint of the pure
breed. Only fertile, healthy, vigorous animals that are typical of their
breed in all respects should be admitted to one's own group of purebreds.
,..,-/ Averages and Selection.-Since the environment may play such an
important role in the actual productivity of any animal, it is important
that the breeder give the matter some attention in making his selection
of breeding animals. In dairy cattle, for instance, age, length of time
pregnant, season of freshening, number of times milked, general health
through the lactation period, etc., may have very important bearings
on the actual amount of milk a cow produces in any given lactation.
In order to compensate for these environmental influences, corre<lti.Q_n
_factors have been devised. The only two that are 'widely used are those
fu;:-~ge atfreshening and number of days or number of times milked.
From the standpoint of genetics or breeding, the only differences with
which the breeder is concerned are those having a genetic basis. He
must attempt, therefore, to smooth out the environmental differences
by correction factors or otherwise in order to make the performance of
his animals comparable. Similarly, the growth rates of beef calves,
the number of pigs that a sow bears and raises, the amount and quality
of wool produced by sheep may be greatly influenced by the environment.
One of the best and easiest ways of compensating for these environ-
mental influences is through the use of production averages. A cow
might make 500 lb. of fat one year, 600 another, and 400 another. It is
likely that her level of inheritance for butterfat production is nearer to
the 500-lb. level than to the 600- or 400-lb. level. In all qualities that
are variable and that may be expressed on more than one occasion, an
average of several measurements will prove safer and more accurate
than anyone measuring. With several opportunities to express a certain
quality, the high and the low expressions are likely to cancel each other,
though they will not necessarily do this. For safety's sake, therefore,
the average'of many records of production or measurement should always
\
be applied to breeding animals in order, in so far as possible, to eliminate

the influence of extremely good or bad environmental conditions. Along
with the records that every breeder should keep must go the ability to
judge and evaluate fairly a host of environmental conditions which play
upon the lives and producing capacities of his animals. Only with this
technique or art well developed will the breeder be able to arrive at
the point \"here he can evaluate actual genetic differences on \yhich, of
course, the entire success of his effort as a livestock breeder depends.
'--Balance and Selection.-We have already discussed a number of
general items that should be borne in mind in selecting breeding animals,
although we have by no means exhausted the list. Additional items will
be presented in the three following chapters as they apply specifically to
certain classes of farm mammals. One further matter of a general nature
should be mentioned at this point, viz., that of always keeping in mind
a general balance sheet bet\"een all the items which it may be the purpose
of the breeder to attempt to incorporate into the genetic pool of his
own herd or flock. It is generally very easy to be a radical, to go to
extremes, and very difficult to stick to the middle of the road or to find
the happy medium.
For example, in selecting the breeding animals, it would be easy to
go to extremes in draft horses and attain great weight and power at the
expense of movability; or, in fast horses, to attain great speed at the
expense of such a high-strung, nervous organization as to make per-
formance either in racing or reproducing difficult of attainment; in dairy
cattle, to attain tremendous producing powers at the expense of regularity
of reproduction over a long lifetime; in meat animals, to attain great
ability to fatten at the expense of reproduction or, conversely, to attain
great reproductive powers (twins in cattle, triplets or quadruplets in
sheep, litters of 20 or 25 young in hogs) at the expense of vigor and
general ability to yield a desirable carcass in a short period of time.
The good judge of livestock is a man who is able to balance success-
fully the defects of one animal against those of another. Likewise, the
good livestock breeder is the one whose selection yields him a proper
balance in his animals of all the various and sundry qualities they must
possess if they are going to be profitable.
We all realize that it would be very much more simple to select for
only one quality at a time and, \\-hen we had that thoroughly incorporated
into our herd or flock, to start on the next item. The drawback to such
a scheme'is the fact that, \"hile we are getting the one desirable quality
we want established in our hcrd or flock, we may also be getting one
or several very undesirable things established also, provided we are
practicing no selection against them. For example, we might select for
nothing but high milk production in dairy cattle, and we could undoubt-
edly secure it, but if it came at the expense of badly pendulous and
broken-away udders or lack of general vigor, disease resistance, or
longevity, it might not be worth the price. Or we might practice selec-
tion for a 4 per cent butterfat test in Holsteins to the exclusion of all
else. Again we could undoubtedly be successful in this aim, but, if we
got it at the expense of a greatly lowered total amount of milk, it probably
would not be worth this price.
From a practical standpoint, therefore, we must reconcile ourselves
to the task of generally selecting for several things and also selecting
against several things at the same time and trying in general to strike
a happy medium, a balance, between the best and the worst in several
respects.
The items that hinder selection most are different in the two sexes.
In the male, it is the inability to test out enough young males so as to be
sure of getting one of the better ones. Probably not more than 10 to 20
per cent of the male offspring of sires are ever used in breeding, and only
an infinitesimal portion of these are properly tested before being put into
heavy service. Most breeders pick males on the basis of type and
pedigree and hope. Probably many of the best males in any class of
animals are never used for breeding.
The main impediment to selection in females is the rapid turnover
among these animals. If a cow is in a herd for 4 years (until she is 6
years old), she will drop 2 heifer calves and we must keep one of them
to replace her-we do not have much choice. The choice may be a little
wider in ewes and sows, but not much so. If we could have our choice
among 10 or 15 females to replace each one that goes out, we could expect
to have more rapid improvement, provided our criteria for selection
were sound. With our present rapid turnover among females little
range of selection is possible. Anything a breeder can do to lower the
rate of turnover in his herd will be of immediate benefit economically
and of ultimately great benefit genetically.
Summary.-We have seen in this chapter that intelligent selection of
breeding~nimal~ presents a great variety of problems. The main purpose
~f selectio;_1~ to increase our stock of desirable-genes and to diminish
our stock of undesirable ones. Since we cannot examine the genes them-
selves, our only recourse is to estimate them by studying individuality,
pedigree, and progeny. In addition to the difficulty of not being able.
to read the labels on the genes directly, there are the further complications
due to dominance, epistasis, and probably other unknown interactions
between the genes, as well as those caused by the environment. To be
effective, o,ur selection must be on the basis of variation caused by
genes, since environmental variations so far as is known are in no sens I

hereditary.
The breeder must first study his probable market and produce what
the market is anxious to buy. He must then select the breed of animals
that will best fulfill his own peculiar market demands. His individual
selections must be guided by individuality, pedigree, and progeny per-
formance. He must select for the type of animal that will most efficiently
serve the production needs of his market over a long life of usefulness,
a type set, it is to be hoped, by the show ring. He must above all else
select fertile, healthy, vigorous individuals from strains or families of
similar qualities, and he must base his selection on high average lifetime
production rather than on the basis of single, exceptional performances.
Selection, the key to animal breeding, is a many-sided problem. To
weigh properly all the factors involved is far from easy, but on the success
of such weighings a breeder's ultimate success depends as on no other
one consideration.
References
Books!
BABCOCK, E. B., and CLAUSEN, R. E. 1927. "Genetics in Relation to Agriculture,"
2d ed., Chaps. 21 and 42, McGraw-Hill Book Company, Inc., New York.
CASTLE, W. E. 1930. "Genetics and Eugenics," Chap. 26, Harvard University
Press, Cambridge, Mass.
DARWIN, C. 1859. "The Origin of Species," Blue Ribbon Books, Inc., Garden
City, N.Y.
FISHER, R. A. 1930. "The Genetical Theory of Natural Selection," Oxford "Cni-
versity Press, New York.
LUSH, J. L. 1945. "Animal Breeding Plans," Collegiate Press, Inc. of Iowa State
MALIN, D. F. 1923. "The Evolution of Breeds," Wallace Publishing Company,
Des Moines, Iowa.
NORDBY, J. E., and BEESON, W. M. 1937. "Livestock Judging Handbook," The
Interstate Printers and Publishers Co., Danville, Ill.
SMITH, W. W. 1941. "Elements of Livestock Judging." J. B. Lippincott Com-
pany, Philadelphia.
Bulletin.~ and Papers

BYWATERS, J. H. 1937. The Hereditary and Environmental Portions of the
Variance in Weaning Weights of Poland-China Pigs, Genetics, 22 :457-468.
COLE, L. J. 1937. The Importance of Threshold Characters in Animal Breeding,
reprint from Amer. Soc. Anim. Prod. Proc., November, pp. 26-28.
COSTILLO, A. E. 1940. Studies on the Correlation between Certain Characters of
Berkjala Sows and the Size of Their Litters, Philippine Agri., Vol. 29, No. 1.
! See also the various breed histories.

GOODALE, H. D. 1940. Cinderella?, Mt. Hope Farm, Williamstown, ::\Iass.

1938. A Study of the Inheritance of Body Weight in the Albino Mouse by
Selection, Jour. Hered., 29(3) :100-112.
HAZEL, L. N., and TERRILL, C. E. 1945. Heritability of Weaning Weight and Staple
Length in Range Rambouillet Ewes, Jour. Anim. Sci., Vol. 4, No.4.
HENKE, L. A. 1935. Is Fecundity in Swine Inherited? Jour. Hered., 26(11):
455-456.
HE'rzER, H. 0., DICKERSON, G. E., and ZELLER, J. H. 1944. Heritability of Type
in Poland-China Swine as Evaluated by Scoring, Jour. Anim. Sci., 3 :390-399.
KISLOVSKY, D. 1927. Types in Animal Breeding and Their Analytical Study, Jour.
Hered., 18(10) :447-455, October.
KNAPP, B., JR., and NORDSKOG, A. W. 1946. Heritability of Growth and Efficiency
in Beef Cattle, Jour. Anim. Sci. Vol. 5, No. 1.
LAMBERT, \Y. V., MURRAY, C., and SHEARER, P. S. 1929. Selection for Resistance
to Hog Cholera, Amer. Soc. Anim. Prod. Proc., pp. 33-38.
' - - - , GOWEN, J. \Y., and COLE, L. J. 1932. Symposium on Genetic Aspects of
Disease, Amer. Soc. Anim. Prod. Proc., pp. 265-272.
LUSH, J. L., HETZER, H. D., and CULBERTSON, C. C. 1934. Factors Affecting Birth
Weights of Swine, Genetics, Vol. 19:329-343.
1940. Intra-sire Correlations or Regressions of Offspring on Dam as a
Method of Estimating Heritability of Characteristics, Amer. Soc. Anim. Prod.
Proc.
1936. Genetic Aspects of the Danish System of Progeny-testing Swine,
Iowa Agr. Expt. Sta. Res. Bul. 204.
- - - and MOLLN, A. E. 1942. Litter Size and Weight as Permanent Charac-
teristics of Sows, U.S. Dept. Agr. Tech. Bul. 836.
- - - , NORTON, H. W. III, and ARNOLD, F. 1941. Effects Which Selection of
Dams May Have on Sire Indexes, Jour. Dairy Sci., 24(8) :695-72l.
- - - and STRAUS, F. S. 1942. The Heritability of Butterfat Production in
Dairy Cattle, Jour. Dairy Sci., 25:(11):975-982.
MANllESA, M. 1927. A Study of the Inheritance of Resistance and Susceptibility
to Infectious Abortion, Amer. Soc. Anim. Prod. Proc., pp. 44-47.
NOllKSKOG, A. W., COMSTOCK, R. E., and \VINTERS, L. M. 1944. Hereditary and
Environmental Factors Affecting Growth Rate in Swine, Jour. Anim. Sci.,
Vol. 3, No.3.
PHILLIPS, R. W., et al. 1941. Standards for Classifying Livestock, Utah Agr. Col.
Ext. Bul. 112.
RHOAD, A. 0., and KLEBERG, R. J., JR. 1946. The Development of a Superior
Family in the Quarter Horse, Jour. Hered., 37(8) :227-238.
ROBERTS, E. 1927. Some Genetic Aspects of Disease Control, Amer. Soc. Anim.
Prod. Proc., pp. 47-52.
- - - and CARD, L. E. 1935. Inheritance of Resistance to Bacterial Infection in
Animals, Ill. Agr. Expt. Sta. Bul. 419.
TERRILL, C. E., and HAZEL, L. N. 1943. Heritability of Yearling Fleece and Body
Traits of Range Rambouillet Ewes, Jour. Amer. Sci., Vol. 2, No.4.
TYLER, W. J., and HYATT, G., JR. 1948. The Heritability of Official Type Ratings
and the Correlation between Type Ratings and Butterfat Production of Ayrshire
Cows, Jour. Dairy Sci., 31(1) :63-70.
VAN RIPER, W. 1932. Aesthetic Notions in Animal Breeding, Quart. Rev. Bioi.,
7:84-922. \
WHATLEY, J. A., JR. 1942. Influence of Heredity and Other Factors on 180 Day
Weight in Poland-China Swine, Jour. Agr. Res., Vol. 65.
WEIGHT, S. 1934. An Analysis of Variability in Number of Digits in an Inbred
Strain of Guinea Pigs, Genetics, 19 :506-536.
1937. The Distribution of Gene Frequencies in Populations, Natl. Acad.
Sci. Froc., 28:307-320.
1935. The "Analysis of Variance and the Correlation between Relatives
with Respect to Deviations from an Optimum, Jour. Genet. 30 :243-256.
CHAPTER XXI
SELECTION IN DAIRY CATTLE
We have now finished our delineation of the principles underlying the

improvement of animals. In this and the two following chapters we
must get down to practical cases and indicate how the principles can
actually be applied. \Ve will cover many phases of practical selection,
though we cannot, of course, cover all of them in all of their possible
permutations and combinations. Each choice of males or females, each
mating, each herd, each breed presents special problems. Only the
person on the spot at the moment can supply the best answer, make the
most nearly correct selection, and then only if he understands the basic
principles involved and has good judgment and a reasonable amount of
experience. We will start these practical chapters with a consideration
of selection in dairy cattle.
Dairy cattle are found on about 75 to 80 per cent of the farms in the
United States, and their products account for about 15 per cent of the
total farm income. Of the25million dairy cattle in the United States,
only about 5 per cent are purebred and registered, the remainder being
80 per cent grades and 15 per cent nondescripts. The average pro-
duction of all dairy cattle is estimated to be about 5,000 lb. of milk a year.1
Dairying is thus seen to be one of the most important phases of animal \
husbandry and _one great~y i~ ~eed of imp~ovement as far as ~h~ ma~hin-\\
ery of productlon, the mdlvldual cow, IS concerned. Dalrymg IS of j
special importance near the large centers of population. /
In one respect, selection of dairy cattle is somewhat easier than with
a
other classes of farm mammals, for a measure of cow's producing ability
in terms of milk and butterfat may be secured over her whole productive
life. We can thus have tangible evidence of her average yearly lactation
yield as well as that of her female offspring, though such actual records
are available on only 4 to 5 per cent of our dairy animals. In another
respect, dairy-cattle selection is more difficult than in other livestock
classes, since it is difficult, if not impossible, to estimate from the exterior
form what the internal functioning will yield in terms of milk and butter-
fat. In this class of livestock, too, the male cannot evidence the quality
for which the species is kept; i.e'., the bull yields no milk, so that his
genetic quality can only be measured through the performance of his
offspring, male and female. Selection is further complicated in dairy
580
SELECTION IN DAIRY CATTLE 581
cattle by the fact that breeders desire' their animals to evidence the
approved dairy type in addition to being high and profitable producers,
and the two things are not highly correlated.
Genetically, we are faced with the very serious problem that milk
production is the end result of a long chain of events caused by manifold
and complex physiological functionings, which probably means that
many genes and many kinds of gene interactions are involved. What we
are trying to do, therefore, is to ascertain the genetic causes of very
complicated physiological processes in a species which has many chromo-
somes and fmv offspring. With 30 pairs of chromosomes, 2 30 gametic
recombinations of paternal and maternal chromosomes are possible and
3 30 possible zygotes (to say nothing of dominance and epistatic effects,
crossing over, mutations, and chromosomal aberrations), and at best vve
can have only a few score or hundreds of daughters of a bull and three
or four or five daughters of a cow. We may get a few landmarks, but
the details of the genetic picture are quite likely to be blurred and the
whole seeming picture turn out to be a mirage.
Another obstacle militating against the effectiveness of selection in
dairy cattle is the fact that l!either type nor producti9!l are completely
heritable. For example, Lush,- N-Ort()n, and Arnold (1941), in a study
of the effects which selection of dams may have on sire indexes, report
heritability figures of 28 per cent for fat and 33 per cent for milk based
on Iowa D.H.LA. records and from Holstein H.LR. data figures of 25
per cent for fat when based on the first lactation and 30 per cent when
based on the second lactation. Lush and Straus (1942) found a value
of 17 per cent for heritability of butterfat production in dairy cattle.
They also cite Ward's figure of about 25 per cent as revealed by his
study of New Zealand cattle. And likewise, in dairy-cattle type, Tyler
and Hyatt (1948) report a figure of about 30 per cent for heritability of
type ratings between parent and offspring in dairy cattle.
The fact that type and production are not completely heritable does
not mean that we should not select from best-type and highest producing
cows. We must of course do this. What th~relatively low heritability
figures mean is that individual looks and·performance must be augmented
by as many other favorable indications as can be secured from perform-
ance of an animal's close relatives and the performance of its offspring.
In addition to all the above handicaps, the.environment can play mean
tricks on us and lead us astray. Milk production is a highly elastic
affair, easily influenced by men, money, materials; feeding, fussing,
fertilizing; health and general hocus-pocus, as well as heredity. Finally,
the cow is forced to live, produce, and reproduce under conditions very
far removed from what might be called natural ones.
In spite of the difficulties, enormous progress has been made in improv-

ing both the type and the productiveness of our dairy animals. We
have no pictures of the general type of cow that our savage, cave-dwelling,
or nomadic ancestors tended, but it seems certain that it would win
nothing but derision or a laugh if exhibited in a modern show ring. The
earliest cows produced enough milk to help nourish their calves for 1, 2, or
3 months, a few hundred pounds at the outside, whereas many thousands
of D.H.LA. cmvs average over 8,000 lb. of milk yearly, and the record
production by a single cow stands close to 42,000 lb. of milk in a year.
Progress has been made, and breeders have made it for the most part
with the presently considered clumsy tool of a belief in the old adage
that "like begets like," or phenotypic selection. Most breeders still
find this tool useful, but it can be sharpened considerably by considering
records of performance on the part of collateral relatives, as well as
ancestors and progeny.
A hunt for superior germ plasm in dairy cattle was instituted by the
Bureau of Dairy Industry in 1936. In all, a total of 1,097 herds of
dairy cattle were surveyed that had continuous records for a long enough
period to have proved at least two sires on the basis of comparing their
daughters' and their mates' records. Of these, 708 herds lo~~ted in
40 states and including seven breeds had complete enough data to be
included in the summary, of which 157 herds were deemed to have made
progress, the other 551 not. The 17 foundation cows, which ,vas the
average number, in the 157 herds averaged 447 lb. of fat; the 62 average
additions to these herds averaged 498 lb., giving a cumulative average of
487 lb. of butterfat. The remaining 551 herds had an average of 15
foundation cows with records averaging 433 lb. of fat; the 40 additions
averaged 431 lb. of fat with a cumulative average of 431 lb. A total of
4,309 sires were used in the 708 herds, and 2,242 of them could be proved
through five or more daughter-dam comparisons. About 32 per cent
of these bulls raised production of their daughters over their dams; 20
per cent held it; and 48 per cent lowered it. The average herd reported
on six sires, which left a total of 39 daughters whose production was
451 lb. of butterfat, while their dams had averaged 452 lb. A few herds
had made fairly steady progress, a few had steadily gone downhill in
production, most had had ups and downs. A few samples of this are
shown in Table 37.
When it is recalled that the above figures are for the better dairy herds,
one can well imagine what the total story on all dairy herds in the United
States would be. In general, it would appear that we have been breeding \
average cows to. average bulls, and we can, of course, expect to get "
average offspring. Better means of discriminating between our o\\"n
\
\
TABLE 37.-CASE HISTORIES OF BREEDING FROM BUREAU OF DAIRY INDUSTRY IN
TERMS OF YEARLY BUTTERFAT AVERAGES
,
Herd A Herd n Herd C Herd D
Foundation cows ......................

I 318 372 327 435
Daughters of 1st sire ................... 369 366 377 373
Daughters of 2d sire .............. ..... 408 363 421 336
Daughters of 3d sire ................... 495 374 346 291
female lines, and sharper tools for use in the selection of males are badly
needed.
The most important item in the dairy business is the individual cO\y.
If she is healthy, a good producer, long-lived, and able to transmit her
good quafitles to numerous' offspring, then there is a firm foundation for
the multitude of activities that must result before milk, butter, cheese. or
ice cream appear on the consumer's table. If, on the other hand, the
individual cow is not healthy, not a good producer, not long-lived, and J
worse still, probably capable of transmitting these poor qualities to her
offspring, then the whole of the dairy industry rests on a very insecure
foundation. It.. is clear, therefore, that dairymen cannot devote too
much time and thought to the matter of getting better CO\YS into their
dairy herds.
Selection Goals in Dairy Cattle ..!:_Breeders of dairy cattle want their
herds to exhibit many qualities of which the following is a partial list:
0.1'4 production, test, efficiency, persistency, longevity, disease resistance,
.' regularity of breeding, good disposition, easy milking, goodtype, trueness
t~ breed standards, etc. It should be obvious that, with ~o many things
clamoring for attention, considerable compromising is bound to ensue.
It would not be so bad if we had easily applied objective measurements
for all these wanted things, but for many of them, all we can have is a
subjective judgment, an opinion which may be right or may be wrong
and may be variable. Many things in our above list can be measured
objectively, but only over a period of years, yet we must select from
some of our cows while they are young. Our above list could perhaps b~ .'
grouped under two headings: (1) physical individuality or type, (2) phys- , /
iological individuality, which adds up to production and reproduction.
Dairy Type and Selection.-Dairy type has been more or less of a
moot subject for a long time. In fat stock type is production to a con-
siderable extent. In hogs, for example, a long deep body (loin and rib
chops and bacon), well-fleshed front and rear ends (shoulder, butt, and
ham), with smoothness, quality, and nonwastefulness indicates directly
to the eye that the five primal cuts make up a considerable portion of the
584 BREElJI1I;G AND IMPIWVEMEN2' OF FL1RM l1NIMALS
live animal. In dairy cattle the degree of correlation between type and 1
production is not nearly so high. Nevertheless, type does enter into the
selection picture for dairy cattle, and, when successful, it pays very good
dividends possibly as regards reproduction, probably as regards pro-
duction, certainly as regards sales merit. Type is an old means of
evaluating animals, having appeared in Roman animal hu~bandry
writings of nearly 2,000 years ago. In those far-off days the bovine was
used primarily as a draught animal, and the ideal animal ,,"as to be well
made; sound, deep-bodied; with long, thick neck; broad, high shoulders;.\
a wide, deep body; good rump; short, straight legs; good hoofs; ~nd at
smooth, soft hide. Included with these points, which obviously have
some bearing on power, ,,"ere stipulations that the animal have dark,-
horns, hairy ears, flat nose, black muzzle, tail reaching to the heelst i
and preferably be black in color-points which might be described as'
fancy rather than strictly utilitarian. With this for a starter, or everi
,,"ithout it, it is not surprising that dairy-cattle score cards continue to
carry aesthetic as well as utilitarian requirements.
The presently used score card for dairy cows of all breeds allots 30
points to general appearance, 40 points to dairy character and body
capacity, arid 30 points to mammary system. It seems impossible to
quarrel with the latter two divisions which have to do \\"ith size and
quality of body and udder. Many statistics are available ,yhich show
that the larger cows within a breed are, generally, more productive and
profitable than are the smaller ones. ; Quality is more difficult to measure,
but experience has shown that,. generally, cows of a lean, angular type,
of moderate-sized bone, of thin hide, and "with udders which, when
milked out, are soft and yielding, thus indicating no overdevelopment of
connective tissue, are higher milkers than those ,Yith the contrasting
attributes. But, whether one pays any attention to so-called "type"
in practicing selection, or not, he is probably going to select cows which
are lean in the neck, shoulder, and thigh; deep and wide in body and
flank; and carrying a capacious mammary system of high quality. So, we
repeat, that these 70 points on the present score card are fully justified
and are probably used by all progressive dairy-cattle breeders.
The only possible score-card argument then boils down to the matter
of the 30 points for general appearance. These 30 points demand that
the animal conform, within certain limits, to the generally accepted
breed standards and that its head and lean neck, its capacious body,
and its large, quality udder, show some decent relationship to each other,
and that the body be carried on fairly straight and well-placed legs
which (following Abraham Lincoln's stipulation) are at least long enough
to reach to the ground.
SELECTION IN DA IR Y CATTLE 585
It appears, then, that the critics of "type" make much ado about
nothing. True, how pretty a cow's face is (and judgments would differ
as they seem to do in our own species) , how straight her back, how level
her rump, how pretty her legs, probably have little to do with how
productive she is. But, both "dirt farmer" and "real breeder" know
FIG. 170.- Holstein-Friesian bull (top) Chip of Nettie and Aaggie; and (bottom) cow,
Rosehill Fayne Wayne--All-American aged bull and cow for 1948.
they must have size and quality of body and udder for high production,
and with the udder strongly attached-not sagging to within a few inches
of the ground and being kicked like a football at every step. The bodily
essentials for milk production we all agree on-some "practical" men
will take them no matter how poorly they are joined together, just as
long as they are still in one piece-the "breeder" wants the same things,
but wants them blended together into a symmetrical whole. Now it
should be obvious that the man who tries to get the necessities more
properly blended has a more difficult task, but he gets good dividends for
his efforts when successful, both in satisfaction and sales appeal as well
as to some extent in production and perhaps in longevity.
Some people apparently have the erroneous idea that "dairy type"
was conceived in one or a few minds sometime in the"dim past, and that
selection for this type gradually led to increased £!od1.!.£!ion. The exact
opposite is probably the truth-selection for production gradually
brought" dairy type" into being; i.e., cows which put excess feed into
milk rather than into fat on their backs are just naturally lean and
angular. It is certain, therefore, thatthe individuality (type) of a dairy
cow is an indication of ,vhetner the avimal will be a good producer or ng,t'3
Type is not an infallible guide to production, but it is, within limits, a
~ound guiqe. Sinc~ "like tends to beget like" (again not perfectly), \~e
can get some idea of how an animal will transmit from its o\vn type, and
,,-e should continue to use individuality in trying to measure the probable
transmitting abilities of our breeding females. Most breeders do select
heifers (and bulls) from their best-type and highest producing cows.
Type and production of daughters of these "best" cows do not always
equal or exceed the dams-the correlation between type or production
of dams and daughters is not perfect-but what we get from this proce-
dure is certainly well worth having.
/' There are two biological reasons why there is not a perfect correlation
bet\veen type and production of daughters and dams. The first ,reason
is that inheritance is a halving and sampling process. Each animal is
I likely to get ·some good and some poor heredit~erminers from each
parent. In some eggs (or sperm) the sample half may contain most of
the good hereditary units and a few of the poor ones, or most of the poor
ones and a few of the good ones. Many varying sample halves are
possible. So, our "best" cows do not always transmit fully ~heir own
excellencies, and can and do transmit differently to their various offspring
-the very nature of the hereditary process makes it almost certain that
this will be so. The other biological reason for the low correlation
between dams and daughters is the fact that each daughter must have
another parent (a bull), and he, too, transmits samples of his inheritance
-some better, some worse. In addition to these two biological reasons
for lack of complete type and production correlation between dams and
daughters, there are also the environmental differences in feeding and
management, which may lead to rather extreme differences between
dams and their daughters.
There is a decided relation, as numerous statistics show, between size
and produci'ng capacity in dairy cattle. Other things being equ'&l, the
\
SELECTION i N DAlBY CATTLE 587
big cow produces more and returns more over feed cost than does the
little cow in any breed or in grade animals. Again, this is not to say
that all big cows are good producers and all small cows poor producers.
Our experience contradicts such a statement, but it is true that we will
more often find high production in big dairy cows than we will in small
.FIG. 17l.-Ayrshire bull (top) Netherhall Swanky Dan who won 84 grand championships;
aud cow (bottom) Alfalfa Farm Ann 2ud, a many times grand champion cow.
ones. Therefore, good size for the breed or grade is a point to be kept
i.n mind in the selection of replacements.
Included in size is the matter of length, depth, and width of body.
The ,animals should be deep from the top of the withers to the floor of
the chest, should be comparatively wide through the chest region, should
be deeper through the rear than through the foreflank. The ribs should
;)88 BREEDING AND IMPROVEMENT OF FARM ANIAfALS
leave the backbone at as much of a right angle as possible and be wide

between. Such conformation gives ample room for the vital organs,
heart, and lungs, and generally indicates the proper bodily mechanism
for the consumption and digestion of large quantities of feed.
Another important consideration as to type is that of dairy quality.
By this is meant the form of the external characteristics, "\vhi'Chevidence
the probable ability and willingness of the animal to consume and digest
large amounts of feed beyond her maintenance requirements and the
ability also to convert this feed into milk rather than into flesh or fat.
These characteristics are evident to the practiced eye, much as they
may have been derided by those who like to scoff at type. Among them
are a clean-cut face; prominent, bright eye; wide, strong muzzle; long
lean neck; thinness and fineness over the withers; spread between the
ribs; length, width, and leanness through the rump; leanness through
the thighs; fineness of hair and of bone; and thinness and mellowness of
hide. These characteristics generally accompany the high milk producer
and can be accepted as marks indicative of good dairy production, not
as we have stated above, in an infallible sense, but by and large they
will hold true.
The third item, in addition to size and dairy quality, is that of udder
q~lity and capacity. The udder is the working portion of the dairy
cow. It should be large. It should be attached well up between the
hind legs and well forward on the underllne:" It should not be broken
away and pendulous, and it should have a level sole. The udder should
be of good quality; i.e., when milked out, it should be soft and yielding,
indicating the presence of enough, but no excess, connective tissue, ,yhich
does not, of course, aid milk secretion. The udder of a high-producing\.
cow is generally well supplied with veins that are evident to the eye. ~
The udder should show as little cleft between the right and left halves
as possible, and no cleft between the front and rear quarters as viewed
from the side. There should be four teats of uniform size placed toward
the four corners of the udder, though not, ideally, exactly at the corners.
The blood, carrying food materials to the udder, flows from front to
rear along the upper portion of the cow and then descends down through
the udder and flows back to the heart, through veins known as the milk
veins. These veins pass forward from the udder on the underline and
should disappear through the body wall somewhere near the front legs.
These veins should be large and generally tortuous. If large and tor-
tuous, this can generally be taken as indicating a good flow of blood
through the udder. The general type of the dairy cow should indicate
strength together with quality. The animals should be of good size but
show no coarseness. The practiced eye of the dairyman catches the
indication of femininity and quality in a dairy cow. There should be
evidence of strength throughout, as indicated by a relatively straight
topline, width and depth of chest, straightness and strength of legs
when viewed either from the side or the rear. The dairy cow is a hard-
working animal, and, if she is to perform her work over a long period
of time, she must be of a rugged, strong constitution. At the same time,
she must show femininity, quality, and animation.
Good-type animals are more likely to have good-type offspring than
are poor-type animals. For this reason, and even though the correlation
is considerably less than 1.0, it will pay to select from our better type
individuals. Figure 172 shows the type distribution of 5,616 daughters
of classified cows in one of our dairy breeds.
Daughters
Ex. V.G. G+ G F P
Ex. 22 85 95 86 6
V.G. 48 851 516 448 76 2
'" G+ 29 281 710 673 127 6

fl
AG 25 209 488 817 167 18
F 4 20 92 154 44 5
P 2 2 5 3
FIG. 172.-Distribution of classified daughters from classified dams.
Excellent cows do not beget only excellent daughters; poor cows do not
beget only poor daughters. However, we are more likely to get good-
type offspring from the three upper classification types than from the
lower three. The correlation is not high (about 0.2) but it is positive.
From a study of 3,738 paternal sisters and of 1,601 cows out of classified
dams, Tyler and Hyatt arrived at the conclusion that the heritability of
type was about 30 per cent; i.e., that offspring inherit about one-third
the superiority or inferiority of the parents' type. If we had the average-
type score of both the male and female parents, the correlation would
probably be higher.
Most of our dairy-breed associations now provide for type classifications
for ~r breeders. Some of the breeds print the types of the animal
cl~d. Such information is valuable, but it leaves a lot to be desired.
When the record shows a certain cow to be G+, we do not know whether
she was off in general appearance, body capacity, dairy quality, or
mammary system, or how much in each. When all a bull's daughters
590 BREEDIlIlG AND IMPROVEMENT OF FARM ANIMALS
average G+ in udders, we do not know whether they were scored down

for small udders, unbalanced udders, poorly attached udders, poor teat
placement, or something else.
The danger with type classification is twofold: (1) We are inclined to
assume we are getting more information than is actually the case. If we
want to know what kind of daughters a certain buH had, no amount of
data on any printed page will substitute equally for a trip to inspect the
daughters at first hand, and to round out the picture, we should study
their dams as well. (2) The other danger comes from using the data
superficially. Some breeders take the stand that they will not use any
bull whose dam has not classified as excellent. We have nothing against
the dam of a bull classifying "Excellent" j but if, as often happens, the
inf.erence is that the bull's daughters will be of superior type (because
his dam was "Excellent"), we must hasten to object. The only proper
way to use classification data as a selection tool is by comparing the
average of all of a bull's daughters with the average of the dams of those
daughters. If a bull was bred to cows which averaged 80 per cent in
type and his daughters averaged 85 per cent, you know that the bull was
a good transmitter of typej but even so you should strive to know a
lot more about the details than this general, over-all type of score average
gives you.
Still another dra\yback to type is that animals do look differently on
different days and in various stages of lactation. We find that we get
a truer judgment of a cow's genotype for production from an average
of several records than from anyone by itself. The same thing would
probably hold for type.
A recent study by Hyatt, Tyler, and Conklin (1949) with 102 Ayrshires
classified several times (6, 12, 18, 24+ months) showed that 5 per cent
classified the same as cows as they had as heifers, 51 per cent varied 1
grade, 38 per cent varied 2 grades, and 6 per cent varied 3 grades. Part
of this variation was due to type changes in animals, especially probably
in feet and legs and uddersj part was due to the fact that there were nine
official classifiers who worked on the project with no doubt some differ-
ences among judges and in the same judge on different days.!
The correlation between the average of the several ratings of each heifer before
first calving and the first and second ratings after calving were 0.37 and 0.40
respectively. With improvements in and standardization of classification
methods, particularly for heifers, the classification program may become valuable
1 HYATT, G., JR., TYLER, 'IV. J., and CONKLIN, C. T., The Relationship betweell
Type Ratings of Ayrshire Females as Young Heifers and as Cows, Jour. Dairy Sci.,
32(4) :380, 1949.
\ \,
SELEC1'ION IN DAIRY CATTLE 591
in helping breeders to cull the poorest type individuals from their herds at an
early age.
Since dairy type has sales value (more for some breeders than for
othe:rs)~'-ls partIally correlated with production and perhaps with longev-
ity, and is up to one-third heritable, it should receive attention from
dairy-cattle breeders in their selection plans and on a family as well as
individual basis. An animal which herself scores 95 per cent, but whose
parents, grandparents, uncles, aunts, cousins, half brothers and sisters,
and nieces and nephews average 80 per cent is not likely to be as good a
transmitter of type as one which herself scores 85 per cent and whose
close relatives also average 85 per cent. The breeder will be recompensed
for selecting from good-type animals, especially if the latter are members
of good-type families.
20,000 0
19,00'0 1 ,.......
0
i l7 o~
~
., 18,00D J/ I
I
I 0 7 .... ~ I
I
'i- 11,000
~ 16,000 ILr i i )'..
::E
c 15,000
L'_ II I ~
C5
~ 14,000
v I 0
1'\
1:',000 L I
v I
12,00°1.9 2 9 ;) 9 4 9 5.9 6·9 1.9
8.9 9.9 10.9 {1.9 1'1..9 13.9 /49
: /5.9
Age
FIG. 173.-0bservational and fitted curves showing the relation of 365-day milk yield
to age for Holstein-Friesian cattle. The observational curve is represented by small
circles. The smooth curve shows the fitted logarithmic curve for milk yield. (From
Gowen, Studies in Milk Secretion, Me. AUr. Expt. Sta. Report, 1920.)
Should a breeder pay attention to type in his selection of bulls? The

answer is yes-perhaps unfortunately so. If he keeps poor-type bulls
and uses their pictures in illustrated advertising, his farm will be avoided
as a source of breeding stock.
There is little correlation between the type of a bull and the type of his
get-there is no known correlation between the type of a bull and the
production of his get. If a bull himself is of good type and his direct
ancestors and collateral relatives are also and he is mated to "typy"
cows from "typy" families, he probably will sire good-type offspring,
and they do sell for more than those of poor type.
Selecting for excessively good type in dairy bulls has probably retarded
increased production to a considerable degree. It should be enough to
have our bulls average or better in type. Too much attention to type
makes us compromise on production. We do not favor selection for
production alone and paying no attention to type, nor do we favor
592 BREEDING AND IMPROVEAfEN1' OF FARM .1NIMALS
selecting for type alone and paying no attention to production. A

reasonable balance must be sought and it will not be easily found.
Production and Selection.-Since dairy cattle are maintained primarily
for the purpose of converting roughages, grain, and grain by-products
into l!)._~ and butterfat, it is not surprising that production bulks large
in most select{on schemes for dairy cattle. Like good type, high records
have advertising and sales value; they are 16 to 30 or 40 per cent heritable;
and since many studies have shown that the return over feed cost moves
steadily upward with increasing production, the matter of basing selection
mainly on production would seem to be justified. The manner in which
animal breeding still gropes in the dark because of a lack of factual data
on individual accomplishments may be visualized through the fact that
in the class where it is easiest to get objective records, namely, dairy
cattle, only about 4 per cent of cows enrolled are in D.H.LA. which
together with cows on AR test make the tested total probably less than
5 per cent. For more rapid progress in animal breeding, we need more
factual data on which to base our judgment in selection. It seems to
come awfully hard, and what we get is likely to serve advertising needs
more than it does genetic ones. Before the individual milk records of
cows can be most useful, they must be placed on some comparable basis
by means of conversion factors.
Conversion Factors.-It is a fact that most cows wjll cootimle-to--
produce more milk and butterfat yearly from the time they first freshen
as two-year-olds up to the a~f..~~ ~~_~ig_h_t_r_ears, and that ther:eafte~
the alll.D.ill!iJ>..ibutterfat and milk will gradually decline in amount. For
this reason age-conversion factors for amount of mil~haYf been devised,
generally from-the average productIOn of maii-y-:tll-o~sands of cows at the
various ages. Butterfat test generally drops slightly with advancing
age (increasing production), but the change is so slight that we generally
do not attempt to apply conversion factors to butterfat per cent.
The age-conversion factors for amount of milk are probably in reality
grmvth or weight-conversion factors. The decline in the size of the
factors from two to six to seven years of age corresponds to an increase
in size toward full maturity, the rise in the size of the factors from six
to seven years of age on is probably due to some general slowing down
with advancing age plus the likelihood of udder injuries. What we want
is a measure of the cow's productive ability at her prime-necessarily
compensated for in youth or old age.
Various sets of age- and times-milked conversion factors are given in
the accompanying tables. These tables show what cows do "on an
average" as they increase in age or in times milked. They are suitable,
therefore,\for converting groups of daughters' and dams' records to a
\
\.,
\
TABLE 38.-GENERAL CONVERSION FACTORS
To convert a 365-day record to a 305-day basis multiply by. . . . . . . . . . . . . . .. 0 .85
To convert a 305-day record to a 365-day basis multiply by. . . . ... . . . . . .. I . 17
To convert a 4-times-a-day milking to 3-times-a-day milking multiply by. . . 0.88
To convert a 4-times-a-day milking to 2-times-a-day milking multiply by. . .. 0.74
To convert a 3-times-a-day milking to 4-times-a-day milking mUltiply by. . .. I . 13
To convert a 3-times-a-day milking to 2-times-a-day milking mUltiply by . . .. 0 . 83
To convert a 2-times-a-day milking to 3-times-a-day milking multiply by. . . 1 .20
To convert a 2-times-a-day milking to 4-times-a-day milking multiply by. . .. 1.35
To convert 2 milkings per day for 305 days to 3 milkings per day for 365 days add 40 %
To convert 3 milkings per day for 365 days to 2 milkings per day for 305 days
subtract. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. ............................... 30 %
TABLE 39.-AGE-CONVERSION FACTORS, D.H.I.A. DATA, U.S.D.A.
Mixed average
Ayrshire, Brown Swiss,
of Ayrshire,
Age Guernsey, Milking Holstein
Guernsey, Jersey,
Jersey Shorthorn
and Holstein
1-6 1.343 1.71S 1.515 1.429

2-0 1.262 1.538 1.377 1.319
2-6 1.195 1.400 1.275 1.235
3-{) 1.141 1.286 1.203 1.172
3-6 1.099 1.196 1.131 1.115
44> 1.063 1.136 1.077 1.070
4-6 1.037 1.088 1.035 1.036
5--{) 1.020 1.052 1.017 1.018
5-6 1.008 1.028 1.006 1.007
6-0 1.000 1.012 1.000 1.000
6-6 1.000 1.006 1.000 1.000
7-0 1.000 1.000 1.006 1.003
7-6 1.006 1.000 1.012 1.009
8-0 1.012 1.000 1.018 1.015
8-6 1.018 1.000 1.036 1.027
9-0 1.024 1.006 1.054 1.039
9-6 1.035 1.012 1.072 1.053
10-0 1.047 1.030 1.090 1.068
10-6 1.064 1.048 1.114 1.089
11-0 1.082 1.072 1.138 1.110
11-6 1.100 1.096 1.162 1.131
12-0 1.112 1.114 1.192 1.152
12-6 1.124 1.132 1.222 1.173
13-0 1.136 1.144 1.252 1.194
13-6 1.148 1.156 1.282 1.215
144> 1.160 1.168 1.306 1.233
14-6 1.172 1.174 1.330 1.251
15-0 1.184 1.180 1.348 1.266
15-6 1.193 ' 1.186 1.366 1.279
16-0 1.199 1.192 1.378 1.288
I
'fABLE 4O.-FACTORS FOR REDUCING PRODUCTION RECORDS TO A TWICE-A-DAY

MILKING BASIS
Number of 3 times 4 times

days milked daily daily
5--15 0.9903 0.9831

6-25 0.9839 0.9721
26-35 0.9776 0.9614
36-45 0.9713 0.9509
46-55 0.9652 0.9406
56-65 0.9591 0.9306

66-75 0.9531 0.9208
76-85 0.9472 0.9111
86-95 0.9414 0.9017
96-105 0.9356 0.8925
106-115 0.9299 0.8834

116-125 0.9242 0.8746
126-135 0.9187 0.8659
136-145 0.9132 0.8573
146-155 0.9077 0.8490
156-165 0.9024 0.8408

166-175 0.8971 0.8328
176-185 0.8918 0.8249
186-195 0.8866 0.8171
196-205 0.8815 0.8096
206-215 0.8764 0.8021

216-225 0.8714 0.7948
226-235 0.8665 0.7876
236-245 0.8616 0.7806
246-255 0.8567 0.7736
256-265 0.8520 0.7668

266-275 0.8472 0.7601
276-285 0.8425 0.7536
286-295 0.8379 0.7471
296-305 0.8333 0.7407
standard basis as is necessary in indexing sires-they will not necessarily

work in individual cases.
The 'writer has proposed the set of factors shown in Table 41 for
converting Guernsey AR records to a standard 3 X, 365, mature basis.
Some AR records are for numbers of milkings other than 610, 730,
915, or 1,095. Conversion factors to take care of this are shown in Table
\
\
\
.I '.
TABLE 41.-AGE AND CLASS CONVERSION FACTORS FOR GUERNSEY AR RECORDS
365 day 3X 305 day 3X 365 day 2X 305 day 2X

Age
1,095 lVI 915 WI 7301\1 610 WI
Jr. 2 1.26 1.47* 1.51t 1.76t

Sr. 2 1.21 1.42 1.45 1.69
Jr. 3 1.16 1.36 1.39 1.62
Sr. 3 1.11 1.30 1.33 1.55
Jr. 4 1.08 1.26 1.30 1. 51
Sr. 4 l.04 l.22 l.25 1.46
Jr. 5 1.02 1.19 1.22 1.43
Sr. 5 1.01 1.18 1.21 1.41
Jr. 6 1.00 1.17 1.20 1.40
Sr. 6 1.00 1.17 1.20 1.40
Jr. 7 1.00 1.17 1.20 1..10
Sr. 7 1.00 1.17 1.20 1.40
Jr. 8 1.01 1.18 1.21 1.41
Sr. 8 l.02 l.19 1.22 1.43
Jr. 9 1.03 l.20 1.23 1.44
Sr.9 1.04 1.22 1.25 1.46
Jr. 10 1.05 1.23 1.26 1.47
Sr. 10 1.06 1.24 1.27 1.48
Jr. 11 1.07 1.25 1.28 1.50
Sr. 11 1.08 1.26 1.30 1.51
Jr. 12 1.09 1.27 1.31 1.53
Sr. 12 1.10 1.29 1.32 1.54
Jr. 13 1.11 1.30 1.33 1.55
Sr. 13 1.13 1.32 1.36 1.58
Jr. 14 1.14 1.33 1.37 1.60
Sr. 14 1.15 1.34 1.38 1.61
Jr. 15 1.16 1.36 l.39 1.62
* Column 2 times 1.17.

t Column 2 times 1.20.
~ Column 2 times 1.40. .•
42, and it would be unnecessary to have these conversion factors if all

records were reported as either of 305 or 365 days duration.
Finally in Tables 45 and 46 are to be found factors which will put
records of any sort onto a 305, 2 X, mature basis.
What we are interested in is a cow's ability to convert feed into food
efficiently and over a long lifetime. Since the fat per cent between
breeds varies and also the fat per cent among cows in the same breed,
Gaines and Davidson have suggested a method for converting milk of
any test to a given standard per cent of Fat Corrected Milk. The
standard most generally used is that of 4 per cent F.C.M. (Fat Corrected
Milk). The factors for this are 0.4 times tot!!l Hl:ilk 1-15 times total
butterfat.
596 BREEDING AND IMPROVEMENT OF FARM ANDVALS
'fABLJ1l42.-FACTORS FOR INCREASING PARTIAL 3X RECORDS TO ENTJRE 3X,305 AND

3X,365
305-day records 365-day records
No. of milkings factor No. of milkings factor

----
610-615 1.200 730-735 1.200
616-625 1.188 736-745 1.190
626-635 1.181 746-755 1.184
636-645 1.173 756-765 1.177
646-655 1.165 766-775 1.171
656-665 1.158 776-685 1.165
666-675 1.151 786-795 1.159
676-685 1.144 796-805 1.153
686-695 1.137 806-815 1.147
696-705 1.130 816-825 1.141
706-715 1.123 826-835 1.135
716-725 1.116 836-845 1.129
726-735 1.109 846-855 1.123
736-745 1.102 856-865 1.117
746-755 1.096 866-875 1.112
756-765 1.089 876-885 1.106
766-775 1.083 886-895 1.100
776-785 1.076 896-905 1.095
786-795 1.070 906-915 1.089
796-805 1.064 916-925 1.084
806-815 1.058 926-935 1.079
816-825 1.052 936-945 1.073
826-835 1.046 946-955 1.068
836-845 1.040 956-965 1.063
846-855 1.034 966-975 1.058
856-865 1.028 976-985 1.053
866-875 1.022 986-995 1.048
976-885 1.017 996-1,005 1.043
886-895 1.011 1,006-1,015 1.038
896-905 1.005 1,016-1,025 1.033
906-915 1.000 1,026-1,035 1.028
1,036-1,045 1.023
1,046-1,055 1.018
1,056-1,065 1.014
1,066-1,075 1.009
1,076-1,085 1.005
1,086-1,095 I
I
1.000
SELECTION IN DAIRY CATTLE 5\)7
TABLE 43.-AYRSHIRE AGE-CONVERSION FACTORS
1-3, 1-4, 1-5 ....... 1.412

1-6, 1-7, 1-8 ....... l.377
1-9, 1-10, 1-11 ..... l.338
2- 0 .............. 1.294 7- 0 ... 1.000 11- 0 ... 1.070 15- 0 .... 1.167
2- 1 .. ............ l.279 7- 1. .. 1.001 Il- l. .. 1.072 15- 1. ... 1.170
2- 2 .............. 1.264 7- 2 ... 1.002 11- 2 ... 1.074 15- 2 .... 1.173
2- 3 ... ' .......... 1.250 7- 3 ... 1.003 11- 3 ... 1.076 15- 3 .... 1.176
2- 4 .............. 1.236 7- 4 ... 1.004 11- 4 ... 1.078 15- 4 .... 1.179
2- 5 .............. 1.222 7- 5 ... 1.005 11- 5 ... 1.080 15- 5 .... 1.182
2- 6 .............. 1.211 7- 6 ... 1.006 11- 6 ... l.082 15- 6 .... 1.185
2- 7 .............. 1.199 7- 7 ... 1.007 11- 7 ... 1.084 15- 7 .... 1.188
2- 8 .............. l.188 7- 8 ... 1.008 11- 8 ... l.086 15- 8 .... 1.191
2- 9 .............. 1.176 7- 9 ... 1.009 11- 9 ... l.088 15- 9 .... 1.194
2-10 .............. 1.167 7-10 ... 1.010 11-10 ... 1.090 15-10 .... l.197
2-11 .............. 1.159 7-11 ... 1.011 11-11 ... 1.092 15-11. ... 1.200
3- 0 .............. l.149 8- 0 ... 1.012 12- 0 ... 1.094 16- 0 .... 1.203
3- 1 .............. 1.142 8- 1 ... 1.013 12- 1 ... 1.096
3- 2 .............. 1.135 8- 2 ... 1.014 12- 2 ... 1.098
3- 3 .............. 1.127 8- 3 ... 1.015 12- 3 ... 1.100
3- 4 .............. 1.119 8- 4 ... 1.016 12- 4 ... 1.102
3- 5 .............. 1.111 8- 5 ... 1.017 12- 5 ... 1.104
3- 6 .............. 1.103 8- 6 ... 1.018 12- 6 ... 1.106
3- 7 .............. 1.095 8- 7 ... 1.019 12- 7 ... 1.108
3- 8 .............. 1.089 8- 8 ... 1.020 12- 8 ... 1.100
3- 9 .............. 1.082 8- 9 ... 1.021 12- 9 ... 1.112
3-10 .............. 1.076 8-10 ... 1.022 12-10 ... 1.114
3-11 .............. 1.071 8-11 ... 1.023 12-11 ... 1.116
4- 0 .............. 1.064 9- 0 ... 1.024 13- 0 ... 1.118
4- 1. ............. 1.059 9- 1 ... 1.025 13- 1 ... 1.120
4- 2 .............. 1.054 9- 2 ... l.026 13- 2 ... 1.122
4- 3 .............. 1.049 9- 3 ... 1.028 13- 3 ... 1.124
4- 4 .............. 1.045 9- 4 ... 1.030 13- 4 ... 1.126
4- 5 ....... ' ...... 1.041 9- 5 ... 1.032 13- 5 ... 1.128
4- 6 .............. 1.036 9- 6 ... 1.034 13- 6 ... 1.130
4- 7 .............. 1.033 9- 7 ... 1.036 13- 7 ... 1.132
4- 8 ............. ' 1.031 9- 8 ... 1.038 13- 8 ... 1.134
4- I), . . . . . . . . . . . . . 1.028 9- 9 ... 1.040 13- 9 ... 1.136
4-10 .............. 1.026 9-10 ... 1.042 13-10 ... 1.138
4-11 .............. 1.024 9-11 ... 1.044 13-11. .. 1.140
5- 0 .............. 1.020 10- 0 ... 1.046 14- 0 ... 1.142
5- 1 .............. 1.018 10- 1. .. 1.048 14- 1 ... 1.144
5- 2 .............. 1.016 10- 2 ... 1.050 14- 2 ... 1.146
5- 3 .............. 1.013 10- 3 ... 1.052 14- 3 ... 1.148
5- 4 .............. 1.011 10- 4 ... 1.054 14- 5 ... 1.150
5- 5 .............. 1.009 10- 5 ... 1.056 14- 5 ... 1.152
5- 6 .............. 1.007 10- 6 ... 1.058 14- 6 ... 1.154
5- 7 .............. 1.006 10- 7 ... 1.060 14- 7 ... 1.156
5- 8 .............. 1.005 10- 8 ... 1.062 14- 8 ... 1.158
5- 9 .............. 1.004 10- 9 ... 1.064 14- 9 ... 1.160
5-10 ............. '. 1.003 10-10 ... 1.066 14-10 ... 1.162
5-11 .............. 1.002 10-11 ... 1.068 14-11 ... 1.164
6- 0 .............. 1.000
TABLE 44.-FACTORS FOR COMPUTING INCOMPLETE RECORDS TO A 305-DAY BASIS

By 5-day Periods. Ayrshire Data
Days Factor Days Factor Days Factor Days Factor

---
38.76 80 2.81 155 1.61 230 1.19
1-7
29.18 85 2.66 160 1.56 235 1.17
10
16.58 90 2.53 165 1.53 240 1.16
15
11.99 95 2.41 170 1.49 245 1.14
20
8.00 100 2.31 175 1.46 250 1.13
25
6.98 105 2.22 180 1.43 255 1.12
30
5.98 110 2.12 185 1.40 260 1.10
35
5.31 115 2.04 190 1.37 265 1.09
40
4.74 120 1.98 195 1.35 270 1.07
45
4.29 125 1. 91 200 1.32 275 1.06
50
3.93 130 1.85 205 1.29 280 1.04
55
3.62 135 1.80 210 1.27 285 1.04
60
3.37 140 1.75 215 1.25 290 1.03
65
3.16 145 1.70 220 1.23 295 1.02
70
2.96 150 1.65 225 1.21 300 1.01
75 I I
It would be impossible to determine superiority between a cow which
made 20,000 lb. of 3 per cent milk and 600 lb. of butterfat and one which
made 14,000 lb. of 5 per cent milk and 700 lb. of butterfat. Applying
the above factors, however, gives us (20,000 X 0.4) 8,000 +
(600 X 15)
9,000 = 17,000 lb. of 4 per cent F.C.M. for the first animal and (14,000
X 0.4) 5,600 + (700 X 15) 10,500 = 16,100 lb. of 4 per cent F.C.M. for
the second. The first cow transformed more feed energy into milk
energy than did the second without regard to the efficiency of the trans-:
formation. The figure 17,000 lb. of 4 per cent F.C.M. for the first cow
means that in making 20,000 lb. of 3 per cent milk she transformed as
much feed energy into milk energy as she would have done if she had
made 17,000 lb. of 4 per cent milk.
The breeder in any breed will have cows which vary quite a bit in
butterfat test. A method is available for converting milk of any test to
any standard. The procedure is based on the figures for converting
to 4 per cent F.C.M., namely, 0.4 of the milk plus 15 times the total
butterfat. One pound of 5.3 per cent milk is equal to 1 X 0.4 = 0.4
plus 0.053 X 15 = 0.795 or 1.195 lb. F.C.M. and the figures for convert-
ing milk of any test to a standard 5.3 per cent F.C.M. are obtained by
dividing (0.4 M. + 15 F.) by 1.195, 'which gives us milk times 0.3347
and total fat times 12.55.
These factors for the various breeds are found in Table 47. With
the breed factors in Table 47, a breeder can put the production of
all his C0lWS into the standard F.C.M. for his breed and thus smooth out
\
\
TABLE 45.-HoLSTEIN CONVERSION FACTORS
Age 2X,305 2X,365 3X,305 3X,365 4X,305 4X,365
1-6 1.515 1.287 1.257 1.06 1.121 0.953

2-{) 1.377 1.170 1.143 0.964 1.019 0.866
2-6 1.275 1.084 1.058 0.892 0.944 0.802
3 1.203 1.022 0.998 0.842 0.89 0.757
3-6 1.131 0.961 0.939 0.792 0.837 0.711
4 1.077 0.915 0.894 0.754 0.797 0.677
4-6 1.035 1.880 0.859 0.724 0.766 0.651
5 1.017 0.864 0.844 0.712 0.753 0.640
5-6 1.006 0.855 0.835 0.704 0.744 0.633
6 1.000 0.85 0.83 0.7 0.74 0.629
6-6 1.000 0.85 0.83 0.7 0.74 0.629
7 1.006 0.855 0.835 0.704 0.744 0.633
7-6 1.012 0.860 0.840 0.708 0.749 0.637
8 1.018 0.865 0.845 0.713 0.753 0.640
8-6 1.036 0.880 0.860 0.725 0.767 0.652
9 1.054 0.896 0.875 0.738 0.780 0.663
9-6 1.072 0.911 0.890 0.750 0.793 0.674
10 1.090 0.926 0.905 0.763 0.807 0.686
10-6 1.114 0.947 0.925 0.780 0.824 0.701
11 1.138 0.967 0.945 0.797 0.842 0.716
11-6 1.162 1.988 0.964 0.813 0.860 0.731
12 1.192 1.013 0.989 0.834 0.882 0.750
12-6 1.222 1.039 1.014 0.855 0.904 0.769
13 1.252 1.064 1.039 0.876 0.926 0.787
13-6 1.282 1.090 1.064; 0.897 0.949 0.806
14 1.306 1.110 1.084 0.914 0.966 0.821
14-6 1.330 1.130 1.104 0.931 0.984 0.836
15 1.348 1.146 1.119 0.944 0.998 0.848
DAYS MILKED'
305-308.. .... . .. 1.00 337-340 .. 0.92
309-312...... 0.99 341-344 .. 0.91
313-316...... 0.98 345-348 .. 0.90
317-320...... 0.97 349-352. 0.89
321-324...... 0.96 353-356 ... 0.88
325-328...... 0.95 357-360 .. 0.87
329-332 ........... 0.94 361-364. 0.86
333-336 ........... 0.93 365 0.85
* Applicable also to Table 46, hence omitted there.
differences in butterfat test. The most usual figures which breeds print
about their cows are those for total milk and total fat. These figures are
the least valuable to the breeder. What he needs to know are his
animals' fat test (because it does not generally pay to stray too far from
breed average in this respect) and total inheritance for production. He
{)OO BREEDING AND IMPROVEMENT OF FARM ANIMALS
TABLE 46.-GUERNSEY AND JERSEY CONVERSION FACTORS
Age 2X,305 2X,365 3X,305 3X,365

---~
1-6 1.343 1.142 1 ~ 115 o 940

2 1.262 1.073 1.047 0.883
2-6 1.195 1.016 0.992 0.836
3 1.141 0~970 0.947 0~799
3-6 1.099 0.934 0~912 0.769
4 1.063 0.904 0.882 0.744
4-6 1.037 0.881 0.861 0.726
5 1.020 0.867 0.847 0.714
5-6 1.008 0.857 0.837 0.706
6 1.000 0~850 0.830 0.700
6-6 1.000 0.850 0.830 0.700
7 1.000 0~850 0.830 0.700
7-6 1.006 0.855 0.835 0.704
8 1 ~012 0.860 0.840 0.708
8-6 1.018 0.865 0.845 0.713
9 1.024 0.870 0.850 0.717
9-6 1.035 0.880 0.859 0.725
10 1.047 0.890 0.869 0.733
10-6 1.064 0.904 0.883 0.745
11 1.082 0.920 0.898 0.757
11-6 1.100 0.935 0.913 0.770
12 1 ~ 112 0.945 0.923 0.778
12-6 1.124 0.955 0.933 0.787
13 1.136 0.966 0.943 0.795
13-6 1 .148 0.976 0.953 0.804
14 1 ~ 160 0.986 0.963 0~812
14-6 1 ~172 0.996 0.973 0.820
15 1.184 1.006 0.983 0.829
TABLE 47.-CONVERSION FACTORS TO VARIOUS F.C.M. STANDARDS
Conversion factors
Av. test
Milk Total fat
Ayrshire .... ~ . ~ ~ . ...

~ 4.06 0.396 14.84
Brown Swiss. 3.96 0.402 15.10
Guernsey~ .. ..... . .. 4.90 0.353 13.20
Holstein ~ ~ ~ . . . . . . . . .... 3~55 0.429 16.09
Jersey ..... ~ . .~ ..... 5.30 0~3347 12.55
"
"
SELECTION IN DAIRY CAT'l'LE GOl
should, therefore, know the fat test of each cow, and if he will convert
all production to his own breed standard F.C.M. test by means of the
factors in Table 47, he will have each cow's total production figure.
For more rapid progress in improving dairy cattle, we need more
production facts, and these facts must be standardized to some common
basis by means of age and times-milked factors, and the F.C.M. factors
will add a further refinement of the data.
Methods for Indexing Dairy Bulls for Production.-Since a dairy bull
yields no milk, his transmitting inheritance must be assayed by means
of his type, his pedigree or the production facts of his offspring. The
first two methods are not very accurate. Over the past 30 years, many
suggestions have been made for working out the third method. In
general, the idea is to try to get a genetic picture of the bull by studying
his daughters' records or by comparing them with those of their dams.
Pearl, Gowen, and Miner 1 (1919) suggested the quartile method of
~riving at a bull's transmitting ability. This method consists of plotting
the curve of variability of AR records, dividing it into quartiles (or
octiles), then classifying the offspring of each sire in terms of percentage
on the basis of the relative standing of each dam with her daughter. The
highest quarter of the AR records was called A, the second B, the third G,
and the fourth D. If a bull had 4 daughters, and 50 per cent, or 2, of
them were in the second quarter, whereas the respective dams were in
the first quarter; 1 daughter, or 25 per cent, was in the second quarter,
,,'hereas her dam was in the third quarter; and 1 daughter, or 25 per cent,
was in the fourth quarter, whereas her dam was in the first quarter;
the bull was then represented by the following formula, the dams' letters
being given first:
50AB + 25GB + 25AD
The example just given seems relatively simple, but on occasion this
formula for a bull can become very complicated; e.g., the formula for
Spermfield Owl:
12AA + 4AB + 4AG + 23BA + 12BB + 4BG + 8GA + 8eB
+ 4GD + 12DA + 4DB + 8DD
In any event, the method was complicated and too indefinite for general
use. In any given case the question for which a breeder seeks an answer
is, How many pounds of milk and what fat percentage will this bull
transmit to his daughters? No such definite answer is possible by the
quartile method.
1 PEARL, R., GOWEN, J. W., and MINER, J. R., Maine Agr. Col. Ext. Bul. 281,
(Studies in Milk Secretion VII), 1919.
Y~ (1925) advanced the theory that a sire's potential transmitting

ability might be expressed in terms of 4 per centJ}lilk by the formula
X = 2A - B. Here X = the sire's potential transmitting ability, A =
the daughters' average records converted to a 4 per cent fat basis, and
B = the dam's record converted to a 4 per cent fat basis. This is
tantamount to saying that the average production of the daughters falls
just halfway between the average production of the dam's and sire's
potential transmitting ability. We have here the adoption of the theory,
borne out quite generally in nature, that the crossing of the two extremes
will produce progeny about halfway between the parents if the character
is determined by multiple factors, as no doubt milk production is. It is
comparatively simple, of course, to find the bull's index, provided the
dams' and the daughters' average productions are known. This method
is unfortunate in that it hides in one figure details that a breeder should
know, viz., the potential amount of milk and the percentage of fat which
the bull transmits. The method of converting records to a 4 per cent
basis as practiced by Yapp was based on the above-mentioned Gaines
and Davidson 2 formula of 0.4 of the milk plus 15 times the fat equals the
F.C.M. to a standard 4 per cent basis.
Table 48 3 gives the factors for standardizing milk of any test to a 4
per cent F.C.M. basis.
TABLE 48.-FACTORS FOR CONVERTING WEIGHTS OF MILK OF OTHER FAT CONTENT TO
THEIR ENERGy-EQUIVALENT WEIGHT OF 4.0 PER CENT MILK
f* 0.0 0.1 0.2 0.3 . 0.4 0.5 0.6 0.7 0.8 0.9
- - - - --- - - - - - - - - --- ---
2 0.700 0.715 0.730 0.745 0.760 0.775 0.790 0.805 0.820 0.835
3 0.850 0.865 0.880 0.895 0.910 0.925 0.940 0.955 0.970 0.985
4 1.000 1.015 1.030 1.045 1.060 1.075 1.090 1.105 1.120 1.135
5 1.150 1.165 1.180 1.195 1.210 1.225 1.240 1.255 1.270 1.285
6 1.300 1.315 1.330 1.345 1.360 1.375 1.390 1.405 1.420 1.435
7 1.450 1.465 1.480 1.495 1. 5101 1.525 1.540 1.555 1.570 1.5~
*f ~ fat percentage.
Turner 4 (1925) proposed the following formula as a means of ascertain-

ing a bull's index: Sire's potential transmitting ability equals the daugh-
1 YAPP, W. W., Transmitting Ability of Dairy Sires, Amer. Soc. Anim. Prod. Proc.
December, 1924.
2 GAINES, W. L., and DAVIDSON, F. A., Relation between Percentage Fat Content
and Yield of Milk, Ill. Agr. Expt. Sta. Bul. 245, 1923.
3 PERKINS, A. E., A Simplified Procedure for Calculating Weights of Milk, etc.,
Jour. Dairy Sci., 20:129-132,1937.
4 TURNER, C. W., A Comparison of Guernsey Sires, Mo. Agr. Expt. Sta. Res. Bul. 79,
1925.
\
BELEC'l'ION IN DAIRY CA'l"l'LE 603
ter's fat production minus 0.15 of the dam's fat production divided by
0.85. This formula grew out of Turner's study in which he found that
for each increase of 100 lb. of fat on the part of the dams, the daughters
increased 15 lb. This formula of Turner's implies the genetic theory of
partial dominance on the part of the bull, since the daughters are seen to
progress 85 per cent of the way up or down to the bull's level from the
dam's average level of production.
Graves l (1926) suggested comparing the dams' and daughters' records
and i~tting the difference (plus or minus) stand for the bull's index. In
rating bulls under this system, Graves also considered the average milk
production of the dams and daughters, the average fat percentage of
dams and daughters, and the percentage of daughters making an increase
in amount of milk as well as the percentage making an increase in fat test.
Goodale 2 (1927) proposed the Mt. Hope bull index. This assumed
that, in matings between animals of unequal levels, ~ilk production is
on the average about seven-tenths of the distance above the level of the
lower parent, while butterfat per cent is about four-tenths of the distance
above the lower level. To get the Mt. Hope Index, compute the averag~
mature equivalent of the milk production of all daughters of the bull,
also the average mature equivalent of the milk productiOI; of the da~s
of these daughters and take the difference between these averages. •
If the daughters' average exceeds the dams' average, add three-
sevenths_ (or 0.4286) of the difference to the daughters' average to get
the bull's milk-index figure.
If the daughters' average is less than the dams' average, subtract
seven-third~ (or 2.333) of the difference from the daughters' average to
'get the bull's milk-index figure.
The index for percentage of butterfat is obtained by similar operations,
but with different fractions,
If the daughters' butterfat average percentage exceeds the dams'
butterfat average percentage, add 1J!!ee-halvet?, (or 1.5) of the difference
to the daughters' average to get the bull's butterfat index.
If the daughters' average is less than the dams' average, subtract t-wo-~
.thirds. (or 0.6667) of the difference from the daughters' average to get the
bull's index.
The above was known as the Precise Mt. Hope Index. Later it was
suggested to ignore the theoretical partial dominance of high milk yield
and low fat test and consider that the daughters fell halfway between
1 GRAVES, R. R., Transmitting Ability of 23 Holstein-Friesian Sires, U.S. Dept.

Agr. Bul. 1372, 1926.
2 GOODALE, H. D., Selecting a Herd Sire, Mt. Hope Farm Publication, 1927.
604 BREEDING AND IMPROVEMENT OF FARi'.! ANIMALS
dams and sire index in both milk and test. This is a variation of the
idea expressed by Yapp and others earlier.
Gifford 1 (1930) from a study of a group of Guernsey bulls concluded
that the daughters' records could be used as the index of the bull. In
other words, the dam's contribution is disregarded entirely, which is all
right provided the dams in each instance are an average of the breed.
Wright 2 has proposed a method for evaluating bulls that takes into
account the number of daughter-dam pairs available. He has clarified
the situation by distinguishing between three types of matings. The
first type, a mating of a bull with some individual cow, Wright says,
can be answered by using the method of multiple regression, provided all
the necessary correlation coefficients are available. Even this, however,
would seem to be rather a risky procedure in the present state of probable
heterozygosity of most of our dairy animals. The second type of matings
of a bull to cows in the same herd or of similar levels of production,
Wright says, can be foretold by the records of the previous daughters
of the bull.
For the answer to the question as to what record would be expected of
daughters of a bull, if the latter were mated with a random sample of
cows of the breed, Wright has proposed the following formula:
2 n
S =n + 2 A + n + 2 (20 - D)
where n = the number of daughter-dam comparisons; A = the breed

average in production; 0 = the daughters' average production; D = the
dams' average production; and S = the bull's index. This system is
based on the genetic theory of multiple-factor inheritance and gives added
weight to the breed average, where the number of daughter-dam com-
parisons is small and increasing weight to the daughter-dam comparisons
as the number of these increases.
At the present time, the dairy-bull index that is being most used is the
one which places the daughters exactly halfway between the production
level of their dams and the index of their sire, both for amount of milk
and percentage of fat. This is now generally known as the Equal-parent
Index. First, we must convert the milk records of cows to~almil
;as;;;'ated to some standard basis by means of conversion factors given
above; do the same thing to the daughters' records; and, when we have
gotten all the records onto a comparable basis, make a comparison
1 GIFFORD, W., The Mode of Inheritance of Yearly Butterfat Production, Mo. Agr.
Expt. Sta. Res. Bul. 144, 1930.
2 WRIGHT, S., On Evaluation of Dairy Sires, Amer. Soc. of Anim. Prod. Proc., 1931,
pp.71-;-78\
SELECTION IN DAIRY CATTLE G05
between the average production of the dams and the average production
of the daughters. The simplest way to ~o this is by the formula-twice
the daughters' averages for milk and test minus those of the dams. This
lets the daughters fall just halfway between the actual average pro-
duction of the dams and the estimated index of the bull, on the assumption
that both parents are jointly and equally responsible for the character-
istics of their offspring.
Dams Daughters
I
Age Conver- Con- Age Conver- Con-
Actual Actual
of Sion verted Test of sion verted Test
record record
cow factor record cow factor record
---
2 7,841 X 1.262 = 9,895 4.91 3~~ 7,422 X 1.099 = 8,157 5.21

2 8,263 X 1.262 = 10,428 5.06 4 8,347 X 1.063 = 8,873 5.10
3~~ 9,340 X 1.099 = 10,265 4.83 2 7,250 X 1. 262 = 9,149 5.00
6 11,521 X 1.000 = 11,521 5.21 3Y2 9,351 X 1.099 = 10,277 4.98
4 10,842 X 1.063 = 11,525 4.96 2?~ 10,461 X 1.195 = 12,500 5.03
6 9,560 X 1.000 = 9,560 5.01 3 7,340 X 1.141 = 8,375 5.12
6/63,194 6129.98 --_
6/57,331 6/30.44
9,555
-5.09
-
10,532 4.99
All records-305-day lactations

Average of dams ................................. . 10,532 - 4.99
Average of daughters .............................
Difference .......................................
. -9,555
- - - 5.09
. - 977 +
0.10
-_
Bull's index ..................................... . 8,578 - 5.19
FIG. 174.-Bull-index work sheet. For a worth-while index the first 6 to 10 daughters of
a bull must be tested and their average production compared with that of their dams. We
could select 6 daughters of almost any bull and thereby give him a fairly high index, which
might have a large advertising value, but its genetic value would be almost zero. Average
production over a period of years gives a much truer estimate than does the production
of anyone year.
A bull index is a very useful tool in the selection and breeding of dairy
~maTs/ though it should by no means be the only one. Figure~i7 4
shows the index of a given bull with the dams' records on the left and the
daughters' records on the right, both sets converted to standard mature
age, six years, 305 days lactation, twice-a-day milking. Since records
show that the fat test of milk produced by a cow is relatively constant,
i.e., changes little, although it does fall slightly with age, it is not custom-
ary or necessary to apply any conversion factors to the butterfat test.
The bull comes out with an index of 8,578 lb. of milk and a 5.19 per cent
test on 6 daughter-dam pairs. It would have been better, of course, for
the bull to have had 8 or 10 daughters rather than 6, and they must be
606 BREEDING AND IMPROVEMElIT OF FARM ANIMALS
the first 6 or 10 daughters; i.e., they cannot be a selected 6 or 10 and

yield anywhere near accurate results.
Regression Index.-One of the authors l recently examined the bull-
index problem and proposed a new index called Regression Index. The
regression of groups of daughters' records on those ofg:toups of dams·was
found to be 0.5.
The matter of regression may be easily illustrated Jty a graph, as in
Fig. 175.
In Fig. 175, we have assumed a breed with an average production of
_10,000 lb. If we graph the groups of cows from 6,000 to 14,000 lb.
(solid line), then the average of their daughter groups will fall about on
14,000 the broken line starting at _8l 000
lb. and rising to 12,000 lb. From
6,000-lb. dams, we "rrornrally expect"
{2,OOO (0 get daughters at the 8,000-lb.
level; i.e., they ,viII regress one-half
. Breed of the way back up to the breed
llo,ODO
average. Likewise, from 14,000-lb.
I'J dams, we "normally expect" their
daughters to regress one-half of the
way back down to the breed average.
Regression, however, is not a mys- .
FIG. l75.-Graph showing regression of terious natural law. It is due to two
groups of daughters on groups of dams. things: (1) the fact that phen0!'YPe
does not necessarjly gulY_2!li!'.:rw _genotype (especially in a quality like
i;
milk prodUctroii.-"\v}i:i~h so prone to envi~o~mental influences), and (2)
that animals producing toward the extremes are unlikely to be mated to
animals as extreme as th~.mselves.
If the breed average is 1.0&00 lb. of milk, we normally expect a group
of daughters averaging 8 1.Q0Q.lb. from a group of cows averaging 6,000 lb.
·Whether ·we would get this or not from any particular bull would remain
to be seen. From a dozen bulls we might get groups of daughters all the
way from 10,000 or 12,000 lb. down to 6,000 lb., but the aver~g_e~Jl_Qf_
._!>hem would be a~ou~ There is no mysterious force, regression or
any otlier, which will tend to move a 6,000-lb. herd up to 8,000 lb. If
the owner continues to use bulls with a transmitting level of 6,000 lb.,
his herd will stay at the 6,000-lb. level, but if he uses average bulls, it
will climb up toward 8,000 lb. Likewise, a 14,000-lh. herd will not
automatically drop to 12,000 lb. This herd is likely to fall back toward
the breed average because its owner ~e avl':lltge bulls .. If he·
-------
1
.
V. A., A New Metlwd 1m Indexing Dairy Bulls, Jour. Dairy Sci., 2.7(11):
RlCE,
921-936, 19,4. ~ / n
, I
uses bulls ,vith an inheritance of 14,000 lb. or more, his herd ,yill remain
at 14,000 lb.
Regression (first used by Francis Galton) is, in one sense, an unfortunate
term because to most minds it usually connotes going backward. In
the biological sense, it means, according to Webster, "the correlation
bet,veen parent and offspring when used as a meas~r~~.ri~."
Actually, of course, regression is the correlation times the ratio of the two
standard deviations, but since in much biological data the two standard
deviations are approximately equal, the correlation and the regression
often have approximately the same value in studies o.f he.redl.·t y . ~res
sion is commonly referred to as "the drag or boost of the breed." hen
animals g_et below the breed average, their offsPri~-"teiidto" chmb back
upt;;;ard the breed a;;erage; when they get above the breed average'1
they "tend to" fall back toward the breed average. In short, regression I
moves upward from below the breed average and downward from above
the breed averag9-
The regression index can be found by getting the difference between
his daughters' actual and "normally eXI?ected" production and adding
this difference to the breed average. This proposal differs from the
Equal:parent system in that the latter method deals with the actual
records of dams and daughters without specific reference to the breed
average, although, as will be shown, this feature could be included.
The Regression Index could be calculated from tables centered on the
respective breed average§_lillth for amount of milk and for butterfat test,
with daughters' expectation rising or falling from the breed average
one-half as fast as their dams.
If tables of normal expectation were not available, the normal expecta-
tion for any group of daughters could be ascertained by adding the
respective breed averages in milk and test to the dams' average in milk
and test and dividing by 2.
(The easiest ,yay of finding a bull's Regression Index would be to find
his Equal-parent Index in the usual fashion (twice the daughters' aver-
ages minus the dams), then add the respective breed averages for milk
and test to this Equal-parent Index and divide by two. In other words,
the Regression method simply regresses the Equal-parent indexes half-
way back to the respective breed averages.
Formulae for finding the Equal-parent and Regression indexes are as
follows:
Equal-parent = W + 2(d - e)
Regression = W + (d - e)
where W = breed average, d = daughters' actual production (or test)
and e = daughters' expected produetion (or test). Leaving the figure 2
out of the Regression Index formula automatically regresses Equal-

parent indexes.
To use the Regression Index, one must know the breed or population
average. This is something ,eve think every breeder should know in
order to more properly evaluate the standing of his own herd.
Various figures can be found which supposedly represent the various
Dairy Cattle Breed Averages. We were not able to get categorical
answers from all the breed secretaries as to the "average production of
their respective breeds." From our study and correspondence, however,
we could get the average butterfat test of the breeds. It seems that
there is relatively little difference in the average yearly butterfat pro-
duction of cows in the various breeds under similar conditions. This
figure appears to be about 410 lb. of butterfat yearly for mature cows on
twice-a-day milking under good Herd Improvement Test conditions.
From the known butterfat tests and the assumed 410 lb. of butterfat ,>ye
have constructed the figures shown in Table 49, which will have to do for
the " Breed Averages" until the various breeds can supply us with the
actual figures.
TABLE 49.-ApPROXIMATE DAIRY CATTLE BREED AVERAGES
Mature basis, 2 X, 305 days
_ _ _ _ _ _ _ _ _ I__ ~_I_ilk__ 1 Tes~1 Butterfat

Ayrshire.... .. .... 10,100 4.06 410
Brown Swiss..... . 10,350 3.96 410
Guernsey...... 8,370 4.90 410
Holstein..... 11,550 3.55 410
Jersey..... 7,740 5.30 410
The figures in Table 49, it should be noted, assume that cows are milked
for a full 305 days, twice a day, under good conditions. 1Ve realize that
not all purebred herds reach the figures indicated in Table 49. Also,
the basic figure of 410 lb. may be up to .j per cent too much 01' too little
for certain breeds. Until we can get more accurate figures from the
breeds, however, the above approximate must suffice. Actually, the
Ayrshire breed is now using the figure of 9,100 lb., which is the actual
average for this breed, but this average includes many cows ·which do
not milk the full 305 days. Those milking the full lO-month period
approximate the figure shown in Table 49.
With ,the values shown in Table 49, one can compute Regression
indexel3 f'01' bulls in any breed. ~he chief advantage of the Regression
\
\
\
~
Index is that, knowing his breed average, the breeder can quickly .arriy~
at a bull's apparent transmitting level. For example, if a Breed Average
was actually 10,000 lb. of milk with a 4 per cent test and a bull of that
breed is stated to have a Regression Index of 10,600 - 4.20%, we know
immediately that his daughters produced 600 lb. more than normally
expected (10,600 minus 10,000) and tested 0.20 better than normally
expected (4.20 minus 4.00) regardless of the level of the dams to which he
was mated. An Equal-parent Index of 10,600 milk and 4.20 test, on
the other hand, could have been secured from dams 8,000, daughters
9,300-dams 14,000, daughters 12,300, and from millions of other com-
binations. The Equal-parent Index is not very revealing because it
lacks any specific point of reference which the Regression Index has
automatically in the Breed Average:
In addition, since the Regression Index moves very high or very low
Equal-parent indexes halfway back to the Breed Average, its use in
pedigrees and other advertising material would help to keep purchasers'
hopes within reasonable bounds of possible accomplishment.
Bull1JLJlgX_!L~__IJ._Sele~t:ion Tool.-Some people prefer to use the actual
daughters' average as the -measure of a bull's transmitting ability without
standardizing the data even as far as age or number of milkings. In
this form, we think the data practically worthless, at least so far as
comparisons among bulls or prediction of future daughters are concerned.
When standardized for age, number of milkings, and environment, if
necessary and possible, the records of a bUll'S daughters are of great
value in delineating his hereditary level of transmission. If the mates
were a random average sample of the breed or population and the bull
was later to be mated to a random average sample of the breed or popu-
lation, such records are perhaps all that are necessary.
Some people standardize mates' and daughters' records and express
the results as simple pluses or minuses for daughters in amount of milk,
butterfat percentage, and total butterfat. This may be misleading
because it fails to consider the normal average regression of 0.5 of groups
of daughters on groups of dams. A bull bred to 6,000-lb. cows and
having daughters at 8,000 lb. would show a 2,000+ figure, ,...-hile one
bred'to cows averaging 14,000 lb. and having 12,000-lb. daughters would
show a 2,000 - figure, leading one to assign greater merit to the first
bull. However, if the breed or population average was 10,000 lb., the
normal regression of daughters on dams would call for exactly the figures
secured, 8,000 and 12,000. In short, these two bulls are equal in trans-
mission-both of them average rather than there being a spread of 4,000
lb. between them, as the 2,000 lb. plus and minus seem to indicate.
For more sound procedures in the selection of bulls, we need all the
610 BREEDIlv'G AND Il}!PROVEME.'Il'l' OF FARM ANIMALS
breeding facts we can get and we need to use the facts as logically and
completely as possible. To us this means using the standardized mates'
and daughters' averages in the form of a bull index.
; A bull's index is a numerical summation of standardized data from his
I
I mate's and his daughters' production figures aimed to express the bull's
inherited level of transmission for milk, butterfat test, and total butterfat.
'I'
, There are several factors which may limit the accuracy of an index.
Environment.-Since an index is based on the average production of a
bull's mates and daughters, it is only as trustworthy as the records
themselves. It is always possible that, either through design or chance,
the mates or the daughters may have been more or less favorably circum-
stanced for high production than were their opposites. If this has
happened, the index loses value. In assessing an index, it is necessary to
establish the "normalcy" of conditions under 'which mates' and daugh-
ters' records were made and to make as reasonable an allowance as the
facts seem to justify. Such allowances are arbitrary and may further
distort rather than correct for differences. If the error is brought about
by forcing a bull's daughters, the distortion will not be increased any
more in an index than it would be in using daughters' records alone.
If the distortion is brought about by hwering mates' production, the
distortion will appear in the bull's index but would not in the daughters'
average. One is, therefore, perhaps justified in discounting to some
extent an index on a bull which was bred to very low record cO\vs.
Since milk, test, and total fat are all elastic to some degree (some
perhaps more than others), it is apparent that neither daughters' records
alone nor an index can be infallible. Both of them are relatives, not
absolutes. Since an index considers all the data, while the daughters'
average considers only half or a little more of it, ,ye prefer the index
over daughters' records alone for the hereditary evaluation of a bull or
as a prediction tool.
Genetic Interactions.-Another point which may influence the accuracy
of an index grows out of the following facts: (1) the number of genes
affect!ng milk, test, and total fat is probably large and (2) the genes
probably interact with each other in a variety of ways other than in
simply additive or multiplicative fashion to produce a given result
(dominance or lack of it, epistatic effects, etc.) The practical breeder
lumps all these and perhaps others into what he calls "nicki~g." This
phenomenon is perhaps difficult of rigid proof, but many practical
examples are most easily explained by recourse to "nicking," and the
nature of the hereditary process would lead one a priori to expect it. If
the early use of a certain bull in a certain herd does result in a favorable
"nick," tl~en the index derived from those early favorable matings may
\
\
\, \
SELECTIO.fI; IN DAIRY CATTLE 611
be misleading so far as the results of the use of this bull on later cows of a
different genetic background are concerned (and this would also be true
of the use of his daughters' records alone).
Finagling.-Still a third factor may limit the accuracy of an index,
namely, selection. Studies have shown that 5 daughter-dam pairs are a
sufficient number on which to formulate an index, provided they are
unselected daughters, although we personally still prefer to have at least
7 or 8 daughter-dam pairs (as does probably everyone else). If only the
selected 5 best daughters (out of 10 or 20) are used in calculating the
index, then the index is of very questionable value and adding more
selected daughters would not help any. Selected daughters are invalid
for indexing a bull; they are also invalid in themselves for estimating the
genetic worth of a bull.
Unpublished data of Hyatt and Tyler of West Virginia on 473 Ayrshire
sires with 17 to 20 or more tested daughters from tested dams showed the
following regardirig average butterfat production.
Regression of
Correlation next 10 on 1st
7, 8, 9, or 10
1st 7 daughters and next 10 daughters ............. . 0.56 0.51

1st 8 daughters and next 10 daughters ... . 0.58 0.55
1st 9 daughters and next 10 daughters .. . 0.65 0.62
1st 10 daughters and next 10 daughters .. 0.65 0.62
These data also showed the following regarding Equal-parent Indexes.
Regression of index on
Correlation next 10 on 1st 7, 8, 9
or 10
Index on 1st 7 and index on next 10.. . I 0.41 0.41 •

Index on 1st 8 and index on next 10 ..... : : : 0.44 0.44
Index on 1st 9 and index on next 10 ....... '1 0.51 0.51
Index on 1st 10 and index on next 10 ....... ' 0.54 0.56
As Hyatt and Tyler point out, in this data the Regression Index on
the 1st 7, 8, 9, or 10 daughter-dam pairs becomes the Equal-parent
Index for the next 10. Apparently, if the first few daughters are above
or below average, future groups of daughters will tend to regress toward
the breed average and the same for indexes.
612 BREEDING AND IMPROVEMENT OF FARM ANI MA LS
Then what answer can we give to the question, " What about indexes
jn the selection of bulls"?
To us the answer seems to be about as follows:
1. Since we want to know how good (or bad) a bull is as quickly and as
completely as possible or on as few unselected daughters as possible,
we can get the answer a little more quickly with indexes, since it
requires ab.out 50 per cent more total daughters to provide as valid
an answer.
FIG. 176.-Any bull js "safe" jn tbis sort of pen. (From Wis. AUT. Expt. Sfa. Cir. 238.)
2. We must learn to take regression into account, to expect future

groups of daughters beyond the first 7 to 10 to regress about halfway
to the breed or population average. The Regression Index will
help to reduce our hopes to somewhere near probable realization.
3. In artificial-breeding publicity, we should try to make our clients
realize that:
a. If the average of the indexes of the bulls being used in artificial
breeding in any breed can be kept above the average of the cows
o{ that breed in the area (or any given herd), reasonable progress
in. increasing production can be expected.
\
\
b. If a bull's first few daughters have done 450 lb. or his index is
500 lb., the client must not necessarily expect his own daughters
by that sire to reach these levels. It depends on a lot of things-
especially the genetic worth and perhaps specific genetic make-up
of his cows and how he feeds and manages them. Better regress
his hopes to some realizable level.
c. If a bull's Regression Index on his first 7 daughters is 450 lb. of
butterfat, better use that as the Equal-parent Index for his
next 10 or 100 daughters.
d. If a bull's index stands at 450 lb. and the cows he ,,·ill be bred to
average 400 lb., then the resulting 1,000 daughters ,,,ill average
about 425 lb. But that average means that there ,,,ill be about
500 of them below 425 lb. Every artificial breeding patronizer
will expect his neighbors to be housing and caring for these 500
below-average daughters. Many will probably want to "raise
Ned" if they get one of them, so be prepared.
e. Because of the heterozygosity of our animals and the halving
and sampling nature of inheritance, no one can predict with any
accuracy ,,,hat a certain daughter by a fairly indexed bull and
out of a cow of known production will produce. An index on
daughters' records is useful only in judging what future groups
may do. The groups will inevitably show considerable vari-
ation, and there will probably be "squawks" from those on the
lower side of average.
Age of Bull to Buy.-It seems fairly obvious that the best buy in a bull
would be a proved sire, one that had already demonstrated his ability to
transmit good type, production, and all the other desirable things which
we want in a herd of dairy cattle. If the bull has done a good job of
transmitting in one herd, he stands a good chance of doing likewise in
another herd, but this is not an absolute certainty. Some bulls do 'well
in one herd, not very well in some other herd, and the reverse of this
occasionally happens too. If management is adequate in both herds
concerned, the differing transmitting abilities of bulls is probably genetic.
As the breeders say, the bull "nicked" well here, he didn't there. In
other words, the genes of the bull fitted well with the genes of the cows of
one herd but not with those of the other.
There are several difficulties with good adequately proved bulls: (1)
There are not many of them; (2) their price is apt to be high; (3) they' are
often getting on in years and losing some of their breeding vigor and
sureness; (4) they are often dead when we learn of them.
In other words, although good proved bulls are the best bet, few of us
are going to be apt to own them. Most of us, in other words, are going
to be using young bulls for some time to come.
If we can't get a proved sire six or eight years old, should we buy a
younger bull, say three or four years old? Some breeder may have such
a bull and not want to use him on his own daughters but be anxious to
keep him alive. If the bull has a good pedigree, if his daughters look
good in comparison with their dams, if the herd is clean, so that we are
not apt to be introducing disease into our herd by the use of this bull,
and if he can be purchased at a reasonable price or, better still, leased at a
nominal price, or if perhaps we find ourselves in a similar situation to this
breeder we and the breeder might then just exchange bulls for 1 or 2 years,
the bull problem would be solved for a while.
And so we might go on, all the ,yay down to calves. A six- or eight-
year-old bull might be proved, a four-year-old bull might have a group of
yearling daughters, a two-year-old bull a bunch of calves being born, a
yearling bull will have no offspring, but he has had a chance to grow out a
bit so we can get an idea about what he is going to look like. With the
calf, there is not much that we can determine perhaps except that he is
alive and with good luck ought to remain so for some time.
There are dangers and possible disappointments in the purchase of any
male. The problem is worthy of all the study and effort we can put into
it.
The principal dangers with the purchase of older bulls are both physio-
logical and genetic. In other words, the older bull used in some other
herd may bring disease with him when he comes to us, and it is very
easy to draw rather broad conclusions from rather fe,v data. Perhaps
a bull has half a dozen daughters that in their youth look a lot better
than their dams which have seen a lot of hard service, but how will 30
of his daughters look 5 or 8 years from now? It is very easy to fool our-
selves with a small" number of young offspring.
With a bull calf, we generally largely avoid the matter of his intro-
ducing disease into our herd. Genetically, we can know nothing about
him except what we can guess from his pedigree and his collateral rela-
tives. The older bull gives us a chance to see a few of his offspring-
generally too few and too young to be a very sure indication. The young
bull, although having no offspring, is generally on the farm where he was
bred. We, therefore, usually have a chance to study his sire and dam,
often many other of their offspring and other direct or collateral relatives
of the bull calf in question. Also the calf can usually be bought more
cheaply.
There. is no definite categorical answer to the question-is it better to
buy a t,~- to four-year-old bull or a bull calf-provided a good, definitely
SELECTION IS DAIRY CATTLE 615
proved bull is not available. There are many angles from which the
problem must be viewed, and the solution will vary with different
individuals and circumstances.
One thing is certain, we need to prove more bulls. Records show that,
on the average, out of every three bulls, one will give daughters that excel
their dams in production, one bull's daughters will produce less than their
dams, and one bull's daughters ·will produce at about the same level.
When a young bull is purchased, he preferably should be used on 12 or
15 cows when he is just at or past a year old and then be leased, loaned,
or just temporarily retired until 5 or 6 of his daughters have made records.
In this way the good transmitting bulls can be discovered and, when dis-
covered, used as extensively and as long as possible. The present system
is generally the exact opposite of the above. A bull calf is purchased, he
gets one to two crops of calves, and then he goes to the butcher to avoid
inbreeding. Thus lots of good young bulls have been real "boloney"
before their worth was realized, and by the same token, lots of poor bulls
get by on the" phony boloney" of 6 or 8 good daughters out of a total
crop of 60 or 80 daughters.
If a man is to be a successful breeder of dairy cattle, there are several
"musts" in his program. We think the most important one of all is that
he tryout young bulls before putting them into heavy service. This,
of course, is just another way of saying that to be successful, a breeder
"must" use a series of good, proved sires.
Many breeders are solving their bull problem by joining artificial-
breeding cooperatives, and the trend in this direction seems likely to
continue, provided the sire selection committees of the artificial breeding
cooperatives are able to do a thorough and painstaking job in the selec-
tion of proved sires and also in the matter of proving their own young
sIres.
Practical Shortcomings of the Progeny Test.-By means of the
progeny test we can, it would seem, get a fairly definite idea as to the
transmitting ability of a bull from a production standpoint. Ho\vever,
some practical considerations would indicate that even this method of
breeding with bulls which have been tested by the progeny method and
found desirable has serious shortcomings. In the first place, the metho~L
is e~pe_n.~~cause it implies the using of a one- or two-ye~r-~l(fbull
'On 12 Qr1.5. .<?.?~s and then setting the bull aside for a perio~_~rs._llll.d_
waiting t~milk yleIct corrlpaxes-"'ith that of their
dams. . The gestation period in""Ca~The heifers are not
br~d until they are eighteen months old, then follows their gestation
period of 9 months, and finally they must go through a 10 months'
lactation period. This is at the very least a total of 48 months, or 4 years,___.
616 BREEDING A ND IMPRO V E M E N T OF FA R M ANI M ALS
which must elapse after a bull reaches sexual maturity until his genetic
~ Quite OOVlOUS y suc a proce d never
be put into practice except by an infinitesimally small number of the
most affluent and farsighted breeders. In addition to the time and
financial aspects involved in such a system of breeding only from proved
sires, there is the additional fact that only a relatively small perce;tage
tIT- be proved good in any event.
FIG. 177.-A proved Guernsey bull and some of his daughters. Ten dams averaged 333
Ib of fat; 10 daughters averaged 5331b of fat. (Courtesy of W. D. Hoard & Sons Publishing
Co.)
Several state-wide inquiries in the United States have revealed that the
average age of dairy bulls in service is about 2H to 3Yz years. Many
dairymen will not keep an old bull because (1) of the danger involved,
(2) of a wish to avoid inbreeding, and (3) of the danger of sterility.
Lush and Lacy have analyzed this problem in a very thorough-going
way as the following quotation l will indicate.
Table 50 shows an estimate of the age distribution which might be expected
among dairy bulls if the value of proven sires w~re so thoroughly appreciated
1 Lus~ J. L., and LACY, M. D., The Ages of Breeding Cattle and the Possibility of
Having ~foved Sires, Iowa Agr. Expt. Sta. Bul, 290, 1932. .
SELECT ION IN DAIRY CA1"l'LE 617
that every bull thought worth using in the first place was kept (unless natural
caUses of death or sterility interfered) until he could be proven and if all those
proven to be distinctly better than the average of theit breed were kept after
the proving as long as they remained fertile and healthy. The figures in the
table represent the maximum which could be attained in the direction of using
as many proved sires as possible for dairy breeding. There are several rough
approximations in the table, as for instance the estimate that during the first
five years the losses from sterility and death from natural causes would only be a
little more than 2.5 per cent per year and that after the bulls had passed 8 years,
the losses from the same causes would run some 13 per cent per year and higher.
If these estimates are roughly correct, for every 1,000 bulls thought worth
trying as sires in the first place, approximately 900 would go past 5 years of age
and begin to be proven by the records of their daughters. Extensive data already
collected! upon proven sires indicate that about one-third actually lower the
production of the herds where they are used, another third do not change the
production much one way or the other, and one-third produce distinct increases
in the production of the herds where they are used. Therefore, it is likely that
of the 900 bulls which begin to be proven between 5 and 6 years of age, it would
rather quickly become apparent that about 300 of them were below the average
merit of their breed and these would be discarded promptly. Of course the
evidence would not all be available at once, and there would be many bulls about
which the breeder would still be in doubt when they were 6 or 7 years of age.
In Table 50, it is indicated that about half of the medium bulls would have their
mediocrity definitely enough proven so that they would be discarded between
the ages of 6 and 7 and the remaining half would be discarded the following year.
This would leave about 30(} bulls which would be proven to be good and would
still be alive at 8 years of age so that they would be available for further use.
An annual loss of 30 or 40 each year through death and sterility from causes not
under the breeder's control (only an estimate but probably not unduly high)
furnishes the basis for the remaining figures in Table 50. When the figures in
Table 50 are added, it is seen that there would be at anyone time only 1,345
bulls 8 years of age or older which would be clearly proven to be superior sires
among a total of 7,145 in use. This is not quite 20 per cent of the entire number
of bulls in service. One thousand fifty more (those between 6 and 8 years of age)
would be in the process of being proven and the very poorest sires would all have
been discarded from this group. This would make a total under such nearly
ideal conditions of about one-third of the sires in actual use which are either
proven to be good or are partially proven and are indicated to be at least of
average merit.
It becomes apparent, therefore, that, although the only method so far

proposed for ascertaining the transmitting abilities of our dairy cattle,
bulls especially, is· scientifically sound, it has very practical limitations as
! McDOWELL, J. C., and 'VINTERMEYER, W. Eo, Proved Dairy Sires, U.S. Dept.
Agr. Cir. 3, September, 1927.

TABLE 50.-AGE DISTRIBUTION OF DAIRY SIRES ATTAINABLE IF ALL POSSIBLE

EFFORTS WERE MADE TO PROVE SIRES AND TO USE AS LONG AS POSSIBLE
THOSE PROVED GOOD
No. of sires
Steps in the proving Age of
No. of sires
that go out
alive at this Reason for decrease
process sire of service
age
during yel1r
5-year period before the 1 1,000 Deaths and sterility from

daughters begin to 2 975 25 disease and accident
prove the sires 3 950 25
4 925 25
5 900 25
3-year period when proof 6 600 300 Disease and accident plus
is coming in and being 7 450 150 the culling of % which
culled accordIngly I prove to be distinctly
inferior. Further cull-
ing of >2 those proved
mediocre
8 300 150 Culling of the remaining
mediocre ones
Period after proof is 9 260 40 Deaths and sterility from

complete and only 10 220 40 disease and accident
the distinctly superior 11 180 40
sires are kept 12 145 35
13 110 35
14 75 35
15 40 35
16 15 25
far as expediting the creation of a more beautiful and more efficient

population of dairy cows.
Selecting a Young Bul1.-If one caimot or does not desire to buy an
older bull, does not choose to use bulls standing at artificial-breeding
units (and if everyone used these bulls and demanded that his cows be
bred to proved sires, there would be no one left to tryout young sires to
find the good ones and dairy-cattle breeding would be in for a tumble),
then he must gamble on a bull calf which he" must select on tYRe_ltnd
Redigree.
There has never been proved to be a best type for dairy bulls. There
is an accepted show-ring type, and the purebred breeder must perhaps
pay some attention to it. How much harm this has accounted for over
the past 100 years, it is impossible to say. There is but a small corre-
lation between type and production in cows. There is a very small
correlation between type of dam and type of daughter. There is prob-

ably no correlation between type of dam and production of daughter and
no correlation between type of sire and production of daughter. Then
why pay any attention to type in a bull? We do not know the answer.
We have often thought, however, that the dairy industry might move
ahead faster if type in bulls was given no standing and no bulls were
shown at fairs and expositions. We do not know what influence great
show bulls have had on the various breeds either in type or production.
We seriously doubt whether the good they may have done outweighs the
bad in either category, and ~we believe further that production might
move ahead faster if breeders could give only such consideration to type
in bulls as they choose of their own free \dll to give. We asume they
would give some, but perhaps not as much as they now feel they must.
Having to pay so much attention to type in bulls means that other more
important things can get less attention, and selection is, therefore, not so
effective as it otherwise would be. (In selecting a \yife (assuming man
does the selecting), we would have a fairly wide choice if we demanded
only a certain minimum of physical charm. If in addition we say, "She
must also bQ a good cook," our choice is narrowed. If ,,'e also stipulate
that" she" must have a million dollars, our choice practically vanishes,
assuming that there is no positive correlation among these things but,
as is more likely, a negative one)
In production we have a little more solid ground on which to stand and
can say something positive \yithout too many if's, and's, and but's. The
most important feature about a young bull is that he be the~j)ia good
prQY~iL sire, the latter animal being one whose Regiession Index for
milk and test are above the breed average (the farther above the better),
whose sons are bulls with good indexes, and whose offspring are of accept-
able type. We say this because \ye have found a good correlation bebyeen
size of index of sire and size of index of sons in several breeds; e.g., 569 sons
listed in Volume 15 of the H olstein-Friesl:an Red Book show a total pro-
duction 4 per cent fat-corrected milk correlation \yith their sires of
+0.31 ± 0.03 and a regression of +0.38. Like father, like son to a
considerable extent.
If the sire (call him A) of our young bull is a proved sire, this means
that he has daughters with records out of dams with records. We next
would group these records to see whether bull A "nicked)) any better
when mated to the daughters of certain bulls. If study showed that he
had "nicked" better when mated to daughters of bull F, then bull F,
if a bull proved good through his sons and daughters, is elected as mater-
nal grandsire of our young bull. Study of bull F might reveal that he
had done best when mated to daughters of bull M, which places bull M,
620 BREEDING AND I1\1IPROVEMENT OF FARM ANIMALS
jf a good proved sire, as the maternal great-grandsire of our young bull.

This gives us the three sires on the bottom line of the pedigree of our
proposed young bull. Now we must find a good daughter of bull M, a
cow with several good records and several good offspring to be the maternal
FIG. 178.-Guernsey bull (top) Cowham Farm Tress King, first-prize bull, three years and
over, and senior and grand champion, winner of the Langwater Trophy. Sire: Riegeldale
King's Philosopher. Dam: Rex's Tress of Cowham Farm. Breeder and owner: C. F.
Cowham, Jackson, Mich. Cow (bottom) Adohr Eldor Pearlette, first-prize female, five
years and over, and senior and grand champion. "Pearlette" won the Douglaston Manor
Farm Trophy for best AR cow, the Meadow Lodge Farm Trophy for best uddered cow,
and her breeder was the winner of the Guernsey I sland Perpetual Challenge Trophy.
Sire: Eldor of Adohr Farm. Dam: Escalon Pearlette. Breeder: Adohr Milk Farms,
Tarzana, Calif. Owner: Adobr Farms-Merritt H. and Rhoda R. Adamson, Carmarillo,
Calif. (Courtesy of American Guernaey Cattle Club.)
granddam (call her Y) of our young bull, and finally we must find a good
daughter of bull F and out of cow Y, a cow we will call X who has several
good records and several good daughters, and she will be the dam of our
young bull. \ This procedure is not theoretical; i.e., it is practical. One
\
might ask, "Where would one find that kind of situation?" The answer
is in any good herd which has used three good sires in succession. They
first used bull M and got a lot of his daughters. These were bred to
bull F to get a lot of his daughters. They are now using the good
proved sire A. In such a herd, if of fair size, there would be considerable
choice of cow family-of what we designated as cows Y and X. N at all
bulls selected in this way will turn out to be good ones, but four out of
five of them should. Since inheritance is a halving and sampling process,
there is no way to take all the guess out of breeding.
We have applied our thinking in the above example to production.
We could also apply it to type, so that if we picked a young bull of
acceptable type whose pedigree, analyzed for type as favorably as we
have just demanded it analyzed for production, we would, no doubt,
have a young bull which would sire good type.
Up to the present, a good proved sire has meant one whose daughters
have done well considering the level of the dams to which he was mated.
Proved bulls have been valued very highly especially by artificial-
breeding associations. This has created the temptation to try to make
bulls look as good as possible. Perhaps there has been some cheating-
feeding and managing dams on a much lower plane than daughters.
Such manipulations are easily possible, and there may nmy, or in the
future, be mediocre or poor bulls masquerading under rather high
indexes.
A further and more definite check on the probable genotypes of bulls
consists of studying how their sons perform. Daughters in one herd can
be manipulated to give apparently good results. Sons scattered over
several herds could not be juggled so easily. A bull can be proved
through his daughters when he is five or six years old. To prove him
through his sons will take about 2 years longer.
We think one good way to spot good bull prospects is to watch for
those bulls in any breed which begin to show up with good sons. If the
first few sons of a bull to become proved really look good, then we could
write to the breed association and get a list of all his registered sons and
perhaps find a few out of good cow families which might merit a trip to
study their offspring. In this way, we might find a good four- or five-
year-old son of a good proved sire (out of a good cow in a good cow
family) who, while not yet proved, will be in another year or two. If
the first few sons of a bull prove out well, it seems probable that more
will later. Studies do indicate that later sons of proved sires will not be
so good or so poor as earlier ones-will tend to regress toward the breed
average, but this is a possible source of bulls which has not been utilized
as much as it might be with advantage. Cow family must be scrutinized
622 BREEDING AN]) IMPROVEMEN'l' OF FARM ANIMALS
very carefully when using it, however, because early sons are likely to
be out of good cow families, but when the older bull begins to get a
reputation through his sons, it will be a temptation to the breeder to sell
his sons out of any old cows.
Another possible source of good herd sires is to induce the owner of a
well-proved sire to breed said sire back to a few of his own best daughters.
If we can get a calf bred that way, he will have 75 per cent the same genes
as his proved sire.
In the past, most breeders have used a good bull or two, then a pOOl"
bull or two, etc., ,vhich, together with the fact that within-herd heifer
selection has not been very soundly based, means that genetic improve-
FIG. 179.--8ir Bess Ormsby May 2nd. a great Holstein sire--result of mating of daughter
back to her sire. Sir Bess Ormsby May. (Courtesy of Mr8. W. S . Kellogg.)
ment in dairy cattle has been very halting. If the last three bulls which
a breeder has used have been good ones, then that breeder now has a good
herd. We often say that the place to buy your next bull is at some farm
where the last three bulls used have been good ones, and then ,ye get off
onto another topic before anyone has time to ask us the location of such
farms and, speaking seriously, they are rather few and far between.
Bull Transmitting Chart.- The art of animal photography has made
rapid progress during the last 20 or 30 years. To bring out photo-
graphically the best points of an animal and to minimize the weaker ones
is a very difficult procedure and one requiring an infinite amount of
patience and practice. When the job has been well done, however, it has
decided limitations from a breeding or selection standpoint, because the
photograph portrays only the phenotypic aspects of the animal. In
additioI\, the breeder also needs some method of getting a picture of th.€
animal that will portray his or her genotypic aspects.
\
\
SELECTION IN DAIRY CATTLF: 623
The following graphs are an attempt to supply this need, especially for
production.
Figure 180 shows a graph for the Holstein bull, King of the Ormsbys,
in regard to amount of milk. The 84 cows to which this bull ,yas mated
were grouped according to production to the nearest 1,000 lb. The
solid line of the graph shows that there was one cow in the 12,000-lh.
class, one in the 13,000-lb. class, two in the 14,000-lb. class, three in the
15,000-lb. class, etc. The broken line of the graph shmys that out of these
84 cows, this bull sired three daughters III the 15,000-lb. class, three
11r-,.-,.-..,--r--,--,-,---,--,
,
161--+-+-+--+-rl----+-+-j---1
151--+--+-+++-\,1--+--+-+---1
141--1:--+--tT--t-H--+++--l
'Dams~ Vi \ Daughter.
13 1--+-+--1I-"'-II---;I.>'-+''-+-t--l
13 I
1'2 121--+--+-;\--;',\--,I-+-+-+--1
II !i
II .'!l Ill--j--+-J H' \--I'i-l-1----1'----4
Dams \ ~.Dau!!n ~rs " 'I I
V>
10 I
I
I
I
I j I~ I~I-H'-\-+-+--+--1
~ 9 \
I
.... I /' \
~8
I \
~ 81--+--+-1'+-++-1+ !-++-+-+--I
't;1
I .
I ,
I
.g 1 ,: ~
1-\;+-'+--+--1
l:; 6 1,1 '\
I \
~ 6 I--l-t-I-i ':-T--f-L-tl--¥-t+-+--{
I,t',
15 JI\I
II
J+--\\
I 'I \ I 1
51--l-~,rt~-+I-t~lf-I-+--1
\: J\1
z
4
\
41--+-+rif'~~-+-~~~'~+--1
L I
I
\1 \
, , IAi
3
2
/ l I ~ 1\/ \\\
I 1\ V"! 1 V /1\4 /
'to 1'2 14 16 18 '20 22 '24 o·(0 28 ~O 32 '2.6 2.8 :!I.O 3.2 14 ;'.6 18 4.0 4.2 4.4
Pounds of Milk,in thousands fat Percentage
FIG. 180.-Daughter-dam chart for the FIG.181.-Daughter-dam chart for
Holstein bull, King of the Ormsbys. the Holstein bull, King of the
Ormsbys.
daughters in the 17,000-lb. class, five in the 18,000-lb. class, etc. The
dams' mean was 20,547 lb. and the daughters' mean 22,190 lb. The
standard deviation of the dams was 3,858 lb. and that of the daughters
3,465 lb., and the coefficient of variation of the dams was 18.8 per cent
and the daughters 15.6 per cent.
A desirable bull is, of course, one who moves the mean to the right;
i.e., increases average production and also lowers the coefficient of vari-
ation. If a bull is able to do both these things, it indicates that he is
more or less homozygous for the genes making for high production.
This sort of graph, in other words, presents a picture of the genotypic
make-up of this bull, which shows at a glance just ,,,hat sort of breeder
he is.
624 BREEDING AND IMPROVEMENT. OF FARM ANIMALS
Figure 181 shows in the same manner the genotypic make-up of this
bull from the standpoint of fat percentage. In this case the dams' mean
was 3.34 per cent, their standard
14
13
deviation was 0.288 per cent, and
~, Dams their coefficient of variation 8.62 per
12
1/\
II
10
"
1\ I VD;Jughfers
I ,
cent.
per
The daughters' mean was 3.54
cent, their standard deviation was
'"] 9 I
\ I , 0.256 per cent, and their coefficient of
~8 I I \ N ,, variation 7.23 per cent.
'151 I I 'I~
, , Figure 182 shows a graph for this
t II \1 \ '
\ \ ,I,,
~6 I
I same bull on the basis of 4 per cent
,j! 5
4 I- II
I
" \' \
F.C.M. The coefficient of correlation
bet\YE~en dams' and daughters' pro-
1/ I Ji
3
j_
I
,\ I I
, duction on a 4 per cent F.C.M. basis
2 \
\ \
was also calculated and found to be a
110 l'Z 14 16 Itl 20 22 24 26 2B 30 32 positive correlation of 0.57 ± 0.04,
Pounds of 4% F.C.M.,in thousands indicating that the level of produc-
FIG. 182.-Daughter-dam chart for the tion of these dams did influence con-
Holstein bull, King of the Orrusbys.
siderably the level of production of
their daughters, ,,"h~h adds weight to the argument for using both dams'
and daughters' records in computing indexes for bulls.
Herd Knowledge and Heifer Selection.-A breeder must start from
where he now is, with whatever heredity is now present in his herd. He
\yill hope to gradually improve the inheritance in his herd-primarily
through the use of better sires. Obviously his first task is to learn more
completely what he has.
Since he is going to pay some attention to type, that would be a good
place to start. He could arrange to have his herd classified, and by
assigning values of 95 for Excellent, 87.5 for Very Good, 82.5 for Good
Plus, 77.5 for Good, 72.5 for Fair, and 67.5 for Poor ratings by the
official classifier, he could work up the average score for his o\vn herd in
general appearance, dairy character, body capacity, mammary system,
and over-all rating. The breeder must go into type matters much more
thoroughly than this, however. He must know 'whether his animals are
good- or plain-headed, fine- or heavy-shouldered, strong- or weak-
topped, wide- and level- or narrow- and peaked-rumped, straight- and
strong- or crooked- and weak-legged, and especially about any general,
weaknesses in size, quality, contour, and attachments of udder and teat;
placing. Only with thorough, detailed, and clear-cut knowledge of his
own present status can a breeder be fully armed to start the search for
his next'. sire. Breeders can work up the data on their own herds, stu-
dents Qn'their college or school herd, and this type of study must be both
\
\
~I
....OJ
;::l
~~I
on '"
:;s~
0
lQ
0
Q) M
'd ....;::l
OJ
Q)
.... ~ OO~I
~-
(:i:l
0
p:; 6~
~
'"
Q)
-...
00
Q)
~blJZ0'\
..c:
;::l
on
A
i:::l----.I
1\
o
~ ~;:IIIIIIIIIIIII ]~I
~ ~~! ~
111111111111
] g~11111111111111j
~;:1111111111111211
11 !:IIIIIIIIIIIII ~a
~:lllllllllllill ~I
~ !~lllllllllllill ~
~ 1 j~~
~ ~
on
11111111I111
15....
til '0 ~ :SblJ..... 1:0 I=Q
,
Q) ~
S 8! $ ._
1b $ 0 al
Q)
"'"
Z ZPSA~ ~~ A
Ol
rn
;:l
"d
....
0
"'"
Ol
.D-
'"
Ol
p:; Ol
bll
A
...., ...."
0
~ ....rn
Ol
::r:: rJl
Ol "
Ol
;J
rJl
Ol
;J
rJl
Ol
Eo-< Eo-<
.~
.~
....,
=§
0
....
Eo-<
....Ol
1
;J
"
~ -< ::a" rJl
...., "'0
A
~
~
E-< d 0
...'"~
~
"
C,,)
'\-d
....
..Q "»
"a"0
.S
"00
;:l
....
\
.....0
"0
/
til
Ol "0 "....
Ol
.... 8 .::: 8 0
'fil
"d "
"d
rJl
"d "d " ~
I'i P I'i
~....
".... ".... "
....
Cl Cl
I
Ol
Ol
Cl Cl C'5
00
.... .-<
. !:II
"d c5
Ol
Po< I ~
(j)
....
;:l
~
::a
Q
to
0
e
"d
....
0
bll
~ '"
p:;
Ol
~ .S rJl
~p ~ »5
H
<Ii
;- "8 8 oOlp ....Ol .~ .'
rn
< ~ ....
0 ~ Po ~
...:
.... Ol Ol 'fil ~ 8
Ol
8 '"
....,
A Po
>.
rJl .... Ol Ol
J31 A" 0 Eo-< ~ " ~::r::p:;

SELECTION IN DAIRY CAT7'LE G27
thorough and critical. Presumably the next sire is going to begin to

correct the present herd faults; but before he can be intelligently chosen,
both as to individuality and family, the herd faults must be crystal
clear in the breeder's mind.
Improving our dairy cattle dependE?_ primarily upon our being able to
recognize the cows which are genetically in the upper half of the normal
distribution curve and to get them bred to genetic upper half males.
Since the heritability of type and production runs somewhere up to
30 per cent, it is obvious that whil~ selection from good-type, high PI:9=-
ducers would move us ahead, our rate of progress,v-oul(fbe slo~v-hecause
we woutd be -bulIding our hopes on a great deal of favorable-looking
variation which was not caused by additive genetic factors but by other
types of genetic interactions and by the environment. Genetic vari-
ations other than additive may be partially captured, but environmentally
caused variations, not at all. It is plain for all to see, therefore, that the
dairy-cattle breeder must get more factual data on the animals in his
herd, must organize the material into its most readily usable and most
valuable form, and plan his matings with a view to strengthening weak
spots and intensifying the good qualities he seeks if he is to hope to make
any real progress in his all-too-short lifetime.
Figure 183 is a sample record card for use in a purebred herd. A sheet
(this or a similar one) should be made out for each cow and kept up to
date at all times. The reason for this statement is twofold: (1) the
limited storage capacity of any brain, (2) the absolute necessity for
getting production and reproduction facts out where we can see them and
use them.
In our opinion there is only one best way to determine what cows to
save heifers from, namely, their own type and performance and their
family. If heritability of desirable things in dairy cows were high, we
could select from our better individuals and let it go at that. Herit-
abilities for the things we want are generally low, which means that
it will pay to consider family as well as the individual cow in deciding
which cows to save heifers from.
We described in the previous chapter how a herd could be set up into
its respective cow families. It takes a little time to do this; but when
once done, cow-family charts can be kept up to date with very little
effort-adding the new heifers when born and reaveraging the production
performance for living cows whenever they complete a new 305-day
period, or once a year. When records have been kept for a few years,
it will be possible to set the whole herd up graphically, as is illustrated
in Fig. 184; and when this can be done, the breeder can eliminate a con-
sideration of the retention of certain calves as replacements even before
they are born, because they are in poor families which he is planning to
eliminate gradually from his herd. Then he can devote his 'whole
attention to choosing the better calves in the better families to serve as
replacements.
Likewise, the task of choosing a bull calf from another breeder's herd
will be greatly simplified when the breeder can display a chart of his
whole herd as indicated in Fig. 184. We may hear of a breeder with a
good proved bull and decide we y,ould like a son of this bull. We go to
Sire I Sire 2 Sire3 Sire 4-
Dams' ewe. 12,371 3.47 12.872 3.50 13,541 3.52 14,2603.52
Daus' ave. 13,4253.52 13,532 3.52 14,163 3.63 15,8803.68
BulJ's index 14,4793.57 14,192 3.54 14,785 3.74 11,3003.84
FamiiyA
Cow 101
Record 4l 13,980 3.61 2l /6,74/ 3.BO
160 181
6 Lac.
13,372 3l 16,420 3.67 II 17,320 3.70
3.58 4l 15,'261 3.7/
2l 16,471 3.81
) - - - - - - - { 158 ) - - - - - - - - { 175
FOimilyB
Cow 115 - - - - . ( 4L 13,470 3.41 2L 12,0603.S0
Record
4 Lac.
12,270
3.21
2L 11,046 3.22
FIG, 184.-Two cow families of very different breeding worth.
this farm and study the female lines as charted and try to pick our bull
calf from a good female line, one in which each generation of females has
proved to be better than the dams. We find from the chart two or three
cows meeting this requirement, and then we go out to the barn to make
our choice between these calves on the basis of their own and their
close relatives' individualities. Such a scheme of selection gives a double
check. We not only get a son of a good proved bull, but we secure him
from a line of females that has shown steady progress generation by
generation.
Two cow families are shown in Fig. 184. Family A is a good one,
family B not so good. The chart is to be read in the following manner.
FoundatipIi cow 101 had 2 daughters by sire 1; one (137) averaged
SELECTION IN DAIRY CATTLE (\29
in 6 lactations 13,240 lb. of milk testing 3.62; the other (127) averaged
15,261 lb. of milk testing 3.71 in four lactations. In family B, founda-
tion cow 116 had a daughter by sire 1 that averaged in four lactations
11,431 lb. of milk testing 3.33. The daughter 137 of foundation cow 101
by sire 1 in turn left a daughter (159) by sire 2 that averaged in four
lactations 13,980 lb. of milk testing 3.61; whereas the daughter (133)
of foundation cow 116 by sire 1 left 2 daughters by sire 2, one of which
averaged in four lactations 12,470 lb. of milk testing 3.37; and the
other (147) averaged in five lactations 14,370 lb. of milk testing 3.13,
etc., through the rest of the chart. All the descendants of foundation
cow 101 (there were 20 in all) averaged to produce 16,020 lb. of milk
testing 3.75, and all the descendants of foundation cow 116 (13 in all)
averaged to produce 12,340 lb. of milk testing 3.34 per cent.
It is obvious that from a production standpoint selection in family B
should be stopped and selection of replacements confined to family A,
with due attention given to type, longevity, ease of milking, freedom from
disease, regularity of breeding, etc. If a breeder keeps records, he can
analyze his herd in the above fashion, and only then is he in a position
to select his female replacements intelligently. If he does this and uses
care in the selection of his sires, there is no reason why he cannot make
progress in building up a higher producing herd. Records provide the
oIlly intelligent basis for the art of selection.
- One pos~ible way to lump together the effects of inherent producing
capacity with regularity of breeding, freedom from disease, etc., is to
compute the average daily production for each cow from the day they
freshen as heifers until they leave the herd. This simply means dividing
a cow's lifetime production (standardized to a 4 per cent F.C.l\1. or par-
ticular breed basis) by the number of days she remained in the herd fol-
lowing her first calving. If she had high inherent milk-producing capac-
ity and persistency and dropped a calf on about the same day each year
for a number of years, her average daily milk production (milking days
and dry days) will be high. The dairy farmer prospers or not depending
on the average daily production of all the cows in his herd, so this figure
for each cow and for each cow family can be very useful in guiding
selection.
Some breeders find it difficult to think in terms of cow families arranged
as in Fig. 184 where we start at the left and read to the right in order to
go from ancestors to offspring, whereas in a pedigree we move from right
to left as age decreases. The material in Fig. 184 can be shown as a
pedigree if one wishes by simply listing all the off-spring of the females
on the bottom line of the pedigree.
Because of the limitations of space, we have shown only the lines
G30 BREEDING AND IMPROVEMENT OF FARM "LVIMALS
with living descendants and nothing but the records in the cow-family
chart, Fig. 184. The breeder should also list on his family charts all
pertinent data concerning the animals in each family, e.g., the typf'.
mature weight, number of services per conception, abortions, mastiti1:i,
difficult calvings, average time between calves, and then work up the
averages for each family.
All sorts of things, both good and bad, run in cow families; i.e., they
are more likely to be found in this family than they are in that one.
Here is a family ,,"hich averages 16,020 lb. of milk ·with a 3.75 test over an
average of 5.6 lactations per cow, has a type score of 83.2, requires 1.83
services per conception, is seldom bothered with mastitis or abortion.
FIG. 185.-Countess Chloe. foundation cow to which more than 90 per cent of the HolsteinI'!
at the University of Massachusetts now trace on the bottom line of their pedigrees.
Another family averages 12,340 lb. of milk with a 3.34 test over an
average of 3.1 lactations per cow, has a type score of 80.1, requires 2.21
services per conception, and is bothered a lot with mastitis and with
"casting the wi.ther;:;" a frequent accompaniment to calving. The
breeder's first job is to analyze his herd, know his best cow families and
broaden the base of his selection (make it more foolproof) by paying
attention to family, which is a better guide to genotype, especially in
young animals, than is their own individuality or phenotype.
It is genetically unsound to select dairy heifers solely on the phenotype
of their dams. It is not strange for a good phenotype to appear c;rccasion-
ally in a poor family. It is very questionable whether her offspring
should be saved. Speaking generally, it would probably be better to
save f,rom a less good phenotype in a good family than from a better
phen,otype in a poor family. Ideally, of course, we should save from
\
SELECTION IN DAIRl- CA'l'TLh: 631
good phenotypes in good families, the close relatives or family serving to

fill in many of the blanks which the animal's own phenotype can shed
little light on. If she is a young cow without any or ,,-ith only one record,
many pertinent questions cannot be answered directly-\\-ill she be a
regular breeder? resistant to mastitis? long-lived? a consistent producer?
etc. Time alone can supply the final and complete answers to these
questions. But if the cow-family chart shows that all her close relatives
have performed well in these various regards, the chances are certainly
greater that the present young cow will folIo,,' this family pattern.
The most necessary lesson for American breeders to learn is that of
thinking in terms of fami ly groups. We have analyzed many herds of
:FIG. 186.-Holstein bull, Clovercourt Ormsby Royal Blend, a ten-year-old bull which has
sired over 17,000 offspring in the New York Artificial Breeding Cooperative. (Cou'rtesy
0/ S . J. Brownell, Cornell University.)
dairy cattle and arranged the animals into their respective cow families.
Without exception we have found breeders saving as many or more
heifers from their poorer cow families as they were from their better
ones. Good and bad hereditary traits do run in families, and there is
ample justification for the old Scot's admonition to his son, "Sandy,
never marry the only good girl in a family." What animal breeding
most needs is to put a broader base (than individuality) under selection.
The best way to do this is to arrange one's herd into its female lines and
consider family along with individuality.
Actually we have not used individuality or phenotype very intelli-
gently. We are likely to take a cow's highest record as the measure of
her genotype. This is unsound because so many environmental happen-
stances can make a record large or small. Lush, from a study of repeat-
ability of cow's records, arrived at a figure of 0.4 for the usual run of
herds (higher possibly where environment has been well standardized

and corrected for). Since each performance of any trait yields additional
information although on a decreasing scale, Lush suggests the following
formula (based on repeatability of 0.4) for arriving at a cow's ability,
2n2n .
+3 tImes h er own average pus1 2n 3+ 3 .tImes the herd average,
n standing for the number of records. Therefore we can broaden the
base of selection based on individuality by properly weighting many
performances into an average.
We can widen it still further by con,;idering close relatives along with
the individual's phenotype. The clirrent herd average can perhaps best
serve as a point of reference. If a breeder's herd average is now 450 lb.
of butterfat, he can rate each CO\V by summating the following:
difference between C01\-'8 own average and herd average + or -

0.5 difference between her ,;ire',; Regression Index and herd average +
or -
0.5 difference between her dam's average and herd average + or -
0.5 difference between her daughter's average and herd average + or
0.5 difference between her full sister's average and herd average + or -
0.25 difference between her maternal grandsire's index and herd average
+ or-
0.25 difference between her maternal granddam's average and herd aver-
age + or -
0.25 difference between her half sister's average and herd average + or -
0.12~~ difference between her aunt's and uncle's performance and herd
average + or
This slim + or - is added to the herd average in order to get a more

accurate estimate of each cmy's genetic or breeding merit than her own
record alone can provide. In this way all the females in the herd can
be listed in the relative order of their merit. It must be understood
that this figure is relative, not absolute. If we know a dam's performance
through several records, we do not learn as much additional from her
parents as the above figures indicate. It is just a rough \Yay of gathering
together the data from close relatives to give a sommyhat more accurate
measure of an animal's probable genotype than her o\yn phenotype itself
can yield. Type and other qualities considered important by the breeder
can be treated in similar fashion.
We cannot see the genes in the chromosomes of our dairy cattle-we
probaQly would learn nothing if we could, because it seems unlikely that
they bear any labels as to what they are or do. By keeping records,
SELECTION IN DAIRY CATTLE fi33
the breeder can begin to get some notions as to the probable kinds of
genes his animals possess. This, in faet, is the only reason for keeping
record".
Other Considerations in Selection of Dairy Cattle.-There are many
other considerations to be borne in mind in "electing dairy cattle, only a
few of which ,,·ill be mentioned here.
From a practical standpoint a dairyman desires to have a herd of cows
that are not too high-strung and nervous, are easy milkers either at hand
or machine milking, and have teats \\'hich are neither too large nor too
small and are well placed on the udder. He also desires cows that calve
easily and without complications and those not particularli';~lbj~-ctto
going off feed or prone to develop udder troubles, such as leaky teats,
mastitis, etc. He hopes to develop a generally vigorous strain of animals
and a strain that maintains high production for a long time during each
lactation period and remains productive over a long period of years. The
matter of longevity is especially important, for it takes about the first 2
years of a cow's productive life to repay the cost of growing her up to
two years of age, or her cash cost if purchased. In addition, the range of
selection is seriously curtailed if cows stay in the herd but a short time;
in other words, the faster the rate of turnover, the less effective can
selection be. To evaluate properly all the considerations that have been
discussed in this chapter makes it imperative for the breeder to keep
records. Many forms have been developed for this purpose, a sample of
which is shown in Fig. 183. Most of the characteristics for which or
against which we select in dairy cattle are the result of the combined
action of heredity and environment. Only with fairly complete recorded
data is the breeder enabled to discover the strains and families within his
own herd that have a preponderance of the genes which he desires to
increase and a minimum of those which he would like to rid himself of.
Finally, the matter of efficiency of feed conversion should receive some
attention in making selections. Several investigations have demon-
strated that variations in ability to convert feed in excess of maintenance
requirements into milk exist among dairy cows. A study by W. T.
Smith and one of the authors on the matter of efficiency of feed conversion
among 42 cows in the University of Massachusetts Experiment Station
herd in 136 lactations over a period of 13 years showed efficiencies rang-
ing from 18.12 per cent to 39.6 per cent, with an average efficiency of
29.25 per cent. One cow put 39.6 lb. of total digestible nutrients into
the milk pail for each 100 lb. of total digestible nutrients fed, another
only 18.12 lb. In general, the higher producing co\\"s were more efficient.
Any cow's rate of efficiency was fairly constant at all ages, being some-
what lower, of course, while the animals were still growing. One hull's
634 BREEDING AND IMPROVElvIEN l' OF FA RJf A N IM ALS
daughters were 5.21 per cent less efficient than their dams, another bull's
daughters were 4.17 per cent more efficient than their dams. That
efficiency of feed conversion has a hereditary basis seems quite likely.
To measure things of this nature requires more feed weighing and record
FIG. 187.-Brown Swiss bull (top) Bradenhurst Royal ChaLlenger 76657, Gr'and Champion
bull at the 1949 National Brown Swiss Show. He was owned and shown by Charles A.
Choate, Old Elm Farm, Winona, Minn.; and cow (bottom) Royal's Rapture of Lee's Hill
115541, Grand Champion cow at thE! 1949 National Brown Swiss Show. Just 2 weeks prior
to the show she completed a national champion production record for the breed of 29,095.7
lb. of 4.22 per cent milk, 1228.84 lb. of butterfat ill 365 da.ys, 3 times, as a five-year-old.
She is owned by Lee's Hill Farm, Morristown, N.J. (COU1·te8Y of Brow'll Swis8 Cattle Breeders'
A8sociation. )
keeping than the average breeder is equipped to provide, but most

breeders can point out cows in their barns that require smaller amounts
of high-priced concentrates than do others for the same amount of pro-
duction. \ The high correlation between amount of production and effi-
\
SELEC'l'ION IN DAIRY CATTLE 635
ciency of feed conversion makes it likely that selection aimed at higher
production will also to a considerable extent take care of the matter of
efficiency of feed conversion.
Only when a breeder knows his own herd average, both as regards
anatomy and physiological functioning, is he in a position to tackle the
all-important question of selecting his next sire.
Pedigree Estimates of Dairy Cattle.-The fact that records are quite
often available on the females' and daughters' averages or indexes on
the bulls in dairy-cattle pedigrees creates a considerable temptation both
to investigators and to breeders to try to devise means of apportioning
values to the different ancestors and other relatives in order to arrive at
a p~cise genetic value of the pedigreed animal. Sometimes the estimates
come very close to the way the ammal breeds, sometimes the miss is
extensive. Sometimes one way of apportioning value seems best, at
other times some other way. If the females' records were never greatly
influenced by the environment and the indexes of the males were secured
without very great effects from "nicking," from the environment, or from
daughter selection, a useful scheme of evaluation might be achieved.
Breeders should study carefully the pedigrees of any animals which
they contemplate buying. They should try to ascertain (1) the con-
ditio~s under_'IhiclLthe ..records were made and (2) how much w;-arriif
out of u~favorable materi:'Jhas been indulged in. TIleY should pay
particuiiu atte~~p ancestors and collateral relatives, but
relatively littleto-'remOt'e ones. They should i~de as many close
collateral relatives as possible in arriving at their estimate. They should
realize that if they can get all the pertinent facts about an animal, they
will learn relatively little more by going back further in the direct lines
of the pedigree. They should know that only the most favorable data is
likely to live in public print. They should be wary of the auctioneer's
prattle about animal X being a half brother to a grandson of the Grand
Chlj,mpion at the 'iVhatsis Expose. They must acknowledge the fact that
knowing a pedigree ever so well still leaves one in ignorance of just what
genes a sire and dam actually transmitted to an offspring.
As said else,yhere, we like to study the bottom line of i1--:pedigree y~ry
thoroughly and demand -that those three slres- and cows be well-proved
animals. If one must arrive at a pedigree estImate By aSSIgmng-Values
to ancestors, we believe in doing it on the basis of degree of relationship
and without compounding values in dired ancestral lines.
Summary.-It is hoped that the discussions of this chapter have made
it evident that the task of selecting dairy animals should be based quite
largely on carefully recorded data. Records of production properly
standardized can yield an index of a bull's transmitting abilities for
636 BREEDiNG AND IMPROVEMEN'l' OF FARM ANIMALS
amount of milk and butterfat test. Records on the cows in a breeder's

herd can be so arranged as readily to reveal the better female lines in
the herd. Records of many other qualities besides production, e.g.,
health, disposition, type, regularity of breeding, etc., must be available
FIG. 188.-Milking Shorthorn bull (top) Revelex Daisy's Warrior, Imp. Grand Champion
bull at Eastern States Exposition and Michigan State Fair, 1949. Owned by ~Iystery
Farm, Hope, R.I.; and cow (bottom) Maidstone Dairy Queen, Grand Champion female at
Eastern States Exposition, 1949. Owned by Last Chance Ranch , Lake Placid, N.Y.
(Courtesy of American Milkino Shorthorn Society.)
if a breeder is to be eventually successful in establishing desirable qualities

in and eliminating undesirable ones from his herd. When sufficient
records litre available in a pedigree, it would seem to be possible to predict
with cOrllliderable accuracy what a projected or newborn individual can
be expected to transmit.
\
Selection in dairy cattle is admittedly complex because many genes

are involved, the environment plays such a large role, and we must
necessarily select for and against many things simultaneously.
In breeding his own replacements, a man should try to buy bulls whose
pedigrees can be filled in with the sort of records that will give some
idea as to what the bull in question may transmit to his daughters. If
one cannot get a proved bull, he should at least try to get a son of a
good proved bull out of a daughter of a good proved bull, and the inclu-
sion of this daughter in a good family should receive at least as much
consideration as her own record.
In the purchase of bulls one should avoid those with no tangible
evidence of freedom from disease, with sufficient age to be proved but
lacking proof, with "holes" in their pedigrees, with a serious lack of the
marks of dairy quality, with some good but many poor close relatives,
with evidence of weak constitution and general lack of size, with pedigrees
too incomplete in terms of proved dairy production.
In the purchase of cows, one should patronize only breeders and dealers
of unquestioned integrity and never buy cows with no records of pro-
duction, with no tangible evidence of freedom from disease, with meaty,
pendulous, lumpy, or broken-away udders, with a lack of body capacity
and udder capacity, with short necks, chunky, meaty bodies, with narrow
pelvis, short, narrow, droopy rumps, with evidence of weak constitution
and general lack of size, with a lot of poor close relatives.
In the purchase of heifers, one should avoid those with no tangible
evidence of freedom from disease, with no prospect of a sound, large,
quality udder, with marked beefy tendency, with a marked deficiency of
size for age, with a lack of proved dairy quality in their close relatives,
with sire of unknown or poor dairy-transmitting qualities. _
The breeder must keep production records on all the females in his-
herd year after year for the twofold purpose of spotting the "boarder" ~
and of providing himself an opportunity to select wisely in his best cm'(.l
families. -
Our first job as animal breeders is to:
Analyze. Find out what we have. We do this by means of complete
record keeping.
and, second, to
Synthesize. Fit, together known hereditary materials in our animals
into more efficient and more beautiful forms, and we do this by systems
of breeding and selection.
Successful dairying depends, in the final analysis, on the inherent
milk-producing qualities of the individual CO\YS in the dairy herd. For
efficient, economical production they must be healthy and endowed by

nature with the bodily mechanism and temperament that will permit
them to consume large quantities of feed and to convert this feed into
large quantities of high-quality milk over a long period of years. In
order to ensure continuing success for the industry, breeders must con-
trive 'mys to create increasingly more efficient animals, and this is only
possible through the intelligent application of facts established through
honest and unprejudiced record keeping.
References
Books
CALDWELL, W. H. 1930. "The Story of the Glenwood Girls," Henry W. Leeds,

Westville, N.J.
1925. "Langwater Guernseys," The American Guernsey Cattle Club,
Peterboro, N.H.
DECKER, F. N. 1923. "Kriemhild Herd," F. N. Decker, Syracuse, N.Y.
ESPE, D. L. 1941. "Secretion of Milk," Collegiate Press, Inc., of Iowa, State
Gow, R. M. 1936. "The Jersey," The American Jersey Cattle Club, New York.
GOWEN, J. W. 1925. "Manual of Dairy Cattle Breeding," The Williams & Wilkins
Company, Baltimore.
GRAVES, R. R., and FORIlMAN, ]\1. H. 1936. "Superior Germplasm in Dairy Herds,"
U.S. Dept. Agr. Yearbook, pp. 997-1139.
HILL, C. L. 1917. "The Guernsey Breed," F. L. Kimball Co., W'aterloo, Iowa.
LINSLEY, J. S. 1885. "Jersey Cattle in America," Burr Printing House, New York.
MEIlIDALE FARMS. 1929. "Dairylike Majesty Imp. 198188," Meridale Farms,
Meredith, N.Y.
PIRTLE, T. R. 1926. "History of the Dairy Industry," Monjonnier Bros. Co.,
Chicago.
PRENTICE, E. P. 1935. "Breeding Profitable Dairy Cattle," Houghton Mifflin
Company, Boston.
PRESCOTT, 1\1, S., and PIlESCOTT, \V. A. 1923. "Holstein-Friesian Foundations,"
Holstein-Friesian World, Inc., Lacona, N.Y.
- - - et al. 1930. "Holstein-Friesian History," Holstein-Friesian "Vorld, Inc.,
Lacona, N.Y.
SANDERS, A. H. 1926. "The Cattle of the World," The National Geographi('
SIBLEY, J. R. 1929. "The Owl-interest Family of Jerseys," John R. Sibley, Spencer,
Mass.
SMITH, A. D. B., and ROBINSON, O. J. 1933. "The Genetics of Cattle," Biblio-
graphia Genetica X, The Hague.
TURNER, C. W. 1939. "The Comparative Anatomy of the Mammary Glands (with
Special Reference to the 17dder of Cattle)," University Cooperative Store, Uni-
versity of Missouri, Columbia, Mo.
WASHBON, W. E. 1948. "Dairy Sire Directory and Line-Breeding Guide," 97 Ken1
Blvd., Salamanca, N.Y.
~> 1950. "Dairy Sire Directory," 97 Kent Blvd., Salamanca, N.Y.
SELECTION IN DAIRY CAT7'LE 639
Bulletins and Paper8

ALDRICH, A. W., and DANA, J. W. 1917. The Relation of the Milk Vein System
to Production, Vt. Agr. Expt. Sta. Bul. 202.
BAKER, T. A. 1926. Transmission of Butterfat Percentage by Holstein-Friesian
Sires, Del. Agr. Expt. Sta. Bul. 145.
BAKER, T. A., and TOMHAVE, A. E. 1941. The Improvement of a Holstein-Friesian
Herd through the Use of Sires with Superior Pedigrees, Del. Agr. Expt. Sta. Bul.
231.
BARTLETT, J. W. 1944. Can We Breed up Holstein Test? Holstein-Friesian World,
41(20) :13.
- - - , PFAU, K. 0., and TUCKER, H. H. 1934. The Inheritance of High Butterfat
Percentage in Holstein-Friesian Cattle, N.J. Agr. Expt. Sta. Bul. 572.
- - - , REECE, R. P., and COWLING, J. D. 1940. The Inheritance of Color in the
Milk of Guernsey Cattle, Amer. Soc. Anim. Prod. Proc., pp. 74--75.
BOGART, R., and IBSEN, H. L. 1937. The Relation of Hair and Skin Pigmentation
to Color Inheritance in Cattle, etc., Jour. Genet., 35(1) :31-59.
BROCKEI,BANK, K E., and WINTERS, L. ~L 1931. A Study of the Methods of Breed-
ing the Best Shorthorns, Jour. Hered., 22 :245-249.
BRODY, 8., and RAGSDALE, A. C. 1935. Evaluating the Efficiency of Dairy Cattle,
Mo. Agr. Expt. Sta. Bull. 351.
BURRINGTON, 'V. D., and WHITE, G. C. 1925. Inheritance of the Percent of Fat
in a Holstein Herd, J our. Dairy Sci., 8(3) :215-230.
CANNON, C. Y., and HANSEN, E. N. 1939. Expectation of Life in Dairy Cows,
Jour. Dairy Sci.,22(12):1025-1032.
CHAPMAN, A. B., and DICKERSON, G. E. 1936. The Relation of Age at First Calving
to Butterfat Production in the First Five Lactations, Amer. Soc. Anim. Prod.
Proc., pp. 52-55.
COPELAND, L. 1938. The Use of Records in Evaluating the Inheritance of Cows and
'in the Proving of Bulls, Jour. Dairy Sci., 21(10) :651-661.
1937. Persistency of Production in Jersey Cows, etc., Jour. Dairy Sci.,
20(3) :151-158.
1931. The Contribution of the Dams in Inheritance of Milk and Butterfat,
Jour. Dairy Sci., 14(5) :379-394.
CSUKAS, Z. 1939. The Genetics of the Lactation Curve, Animal Breeding in the
Light of Genetics, 7th Internatl. Congo Genet., Section D, pp. 48-51.
DAWSON, J. R., and KOPLAND, D. V. 1948. A Breeding Experiment with Holstein
Cattle, etc., U.S. Dept. Agr. Tech. Bul. 965. •
DICKERSON, G. E. 1941. Estimates of Producing Ability in Dairy Cattle, Jour. Agr.
Res., 61 (8) :561-586.
- - - - and CHAP~IAN, A. B. 1940. Butterfat Production, Reproduction, Growth
and Longevity in Relation to Age at First Calving, Amer. Soc. Anim. Prod. Proc.,
pp. 76-81.
DICKEY, H. C., and LABARTHE, R. 1945. Predicting the Transmitting Ability of
Young Dairy Sires for Milk Production, Butterfat Test, and Butterfat Produc-
tion, Jour. Dairy Sci., 28 :893-900.
DINKHAUSER, F. 1939. Yield and Inherited Yield in Northwestern German Breeds
of Black Spotted Lowland Cattle, with Special Reference to the Influence of the
East Friesian Lines, II. Schleswig-Holstein, Zuchtungskunde, 14:10-1039.
Dairy Sci. Abs., Vol. 1, No.2, August, 1939.
640 BREEDING AND IMPROVEMENT OF FARM ANll"IALS
1939. Yield and Inherited Yield in Northwest German Breeds of Black

Spotted Lowland Cattle, with Special Reference to the Influence of the East
Friesian Lines, III. Ems and Osnabruck, Zuchtungskunde, 14 :175-184. Dairy
Sci. Abs., Vol. 2, Ko. 1, May, 1940.
1938. Inheritance of Buttcrfat Content, Deut. Land. Tierzucht, No. 52,
pp. 1013-1015, Dairy Sci. Abs., Vol. 1, May, 1939.
EDWARDS, J. 1932. The Progeny Test as a Measure of Evaluating the Dairy Sire,
Jour. Agr. Sci., 22 :811.
- - - and HUNTER-SMITH, J. 1932. The Importance of the Progeny Test in
Dairy Cattle, Jour. R.A.S.E., 93.
FOHRMAN, M. H. 1926. Official Records as Material for Studying Inheritance of
Milk and Butterfat Production, Jour. Dairy Sci., 9(3) :286-293, May.
- - - and GRAVES, R. R. 1939. Experiments in Breeding Holstein-Friesian
Cattle, etc., U.S. Dept. Agr. Tech. Bul. 677.
FOLLEY, S. J., and YOUNG, F. G. 1938. The Effect of Anterior Pituitary Extracts on
Established Lactation in the Cow, Roy. Soc. London Proc. Ser. B, 126(842): 45-76.
FOWLER, A. B. 1932. The Ayrshire Breed of Cattle, A Genetic Study, Jour. Dair!!
Res., Vol. 4, No.1, December.
GAINES, W. L. 1940. Live Weight and ::\1ilk Energy Yield in Holstein Cows, Jour.
Dairy Sci., 23(3);259-265.
1935. Correction Factors and Germ Plasm in Dairy Cattle Breeding,
Amer. Soc. Anim. Prod. Proc., pp. 50-53.
1931. Size of Cow and Efficiency of Milk Production, Jour. Dairy Sci.,
14:14-25.
1927. Measures of Persistency of Lactation, Jour. Agr. Res., 34(4) :373-383.
- - - , DAVIS, H. P., and MORGAN, R. F. 1947. Within-cow Regressi9n of Milk-
energy Yield on Age and Liveweight, .I our. Dairy Sci., 30 :223-278.
GARNER, F. H. 1932. A Study of Some Points of Conformation and Milk Yield in
Friesian Cows, Jour. Dairy Res., 4:1-10.
GIFFORD, W., and ELTING, E. C. 1930. The Mode of Inheritance of Yearly Butter-
fat Production, Mo. Agr. Expt. Sta. Res. Bull. 144.
- - - and - - - . 1928. The Effect of the Ages of Sire and Dam on the Average
Butterfat Production of Offspring in Dairy Cattle, Jour. Dairy Sci., 11(1):1-9,
January.
GILMORE, L. O. 1950. Inherited Non-Lethal Anatomical Characters in Cattle:
A Review. Jour. Dairy Sci., 33(3):147-165.
GOODALE, H. D. 1928. Selecting a Herd Sire, Mt. Hope Farm, Williamstown, Mass.
GOWEN i J. "'. 1934. The Influence of Inheritance and Environment on the Milk
Production and Butterfat Percentage of Jersey Cattle, Jour. Agr. Res., 49(5):
433-465.
1933. On the Genetic Constitution of Jersey Cattle as Influenced by
Inheritance and Environment, Genetics, 18 :415-440. .
1933. Conformation of Parents as Related to Milk Secretion of Daughters,
Jersey R.M., Jour. Agr. Sci., 23(4) :514-518.
1933. Conformation of the Cow as Related to Milk Secretion, etc., Jour.
Agr. Sci., 23(4):485-513.
1931. Body Pattern as Related to Mammary Gland Secretion, Natl. Acad.
Sci. Proc., 17 :518-523.
- - - . \1927. A Resume of Cattle Inheritance, Maine Agr. Expt. Sta. BioI. Lab.
Bul. 1'05.
\
\
\,
! \
SELECTION IN DilIRY CA_1'TLE 641
HJ27. Productivity of Guernsey Cows of American or Island Origin,

Maine Agr. Expt. Sta. Bul. 341.
1926. Genetics of Breeding Better Dairy Stock, Jour. Dairy Sci.,
9:153-170.
1926. Judging of Dairy Cattle and Some of Its Problems, Jour. Hered.,
17(1):13-26.
1925. The Size of the Cow in Relation to the Size of Her Milk Pnduction,
Jour. Agr. Res., 30 :865-869.
1923. Conformation and Milk Yield in the Light of the Personal Equation
of the Cattle Judge, Maine Agr. Expt. Sta. Bul. 314.
1921. Report on Progress on Animal Husbandry Investigations, Maine
Agr. Expt. Sta. Bul. 274.
1918. Report of Progress on Animal Husbandry Investigations, .Maine Agr.
Expt. Sta. Bul. 299.
GRAVES, R. R. 1925. Improving Dairy Cattle by the Continuous Use of the Proved
Sire, Jour. Dairy Sci., 8(4):391-405, September.
HEIZER, E. K, et al., 1938. Nicking in Dairy Cattle, Amer. Soc. Anim. Prod. Proc.,
pp.67-72.
HOOPER, S. S. 1919. Inheritance of Jersey Colors, Jour. Dairy Sci., 2(4):290-293,
July.
HUNT, R. E., and NEWMAN, W. S., JR. Selecting Holstein-Friesian Sires for High
Yearly Production, Va. Agr. Expt. Sta. Bul.
HYATT, G., JR., TYLER, W. L., and CONKI.IN, C. T. 1949. The Relationship
between the Type Rating of Ayrshire Females as Young Heifers and as Cows,
Jour. Dairy Sci., 32(4):375-380.
JOHNSON, L. A., BARTLETT, J. W., and COPELAND, L. 1940. A Study of Nicking in
Jersey Cattle, Jour. Dairy Sci., 22 :709-718.
KRUGER, L. 1939. The Determination of Performance Value, Genetic Value, Genes
and Genetic Quanta, Animal Breeding in the Light of Genetics, 97th Internatl.
Genet. Cong., Section D, pp. 45-47, Dairy Sci. Abs., Vol. 2, No.1, May, 1940.
LESCH, H. 1938. Genetical Analysis of Fertility and Milk Production, J ena I naug.
Dis., Dairy Sci. Abs., Vol. 1, No.1, May, 1939.
LFSH, J. L. 1944. The Optimum Emphasis on Dam's Record \Yhen Proving Dairy
Sires, Jour. Dairy Sci., 27(11) :937-951.
1933. The Bull Index Problem in the Light of Modern Genetics, Jour.
Dairy Sci., 16(6):501-523, November.
- - - , HOLBERT, J. C., and WILLHAM, O. S. 1936. Genetic History of the Holstein-
Friesian Cattle in the United States, Jour. Hered., 27(2):61-72.
- - - and LACY, M. D. 1932. The Ages of Breeding Cattle and the Possibilities
of Using Proven Sires, Iowa Expt. Sta. Bul. 290.
- - - and SHRODE, R. R. 1950. Changes in :Milk Production with Age and Milk-
ing Frequency, Jour. Dairy Sci., 33(5) :338-357, May.
- - - , NORTON, H. W., III, and ARNOLD, F. 1941. Effects Which Selection of
Dams May Have on Sire Indexes, Jour. Dairy Sci., 24(8) :695-721.
- - - and STRAUS, F. S. 1942. The Heritability of Butterfat Production in Dairy
Cattle, Jour. Dairy Sci., 26(11) :975-982.
MADSEN, K. 1932. Inheritance of Milking Capacity, Nature }v[ag., 129:165,
Jan. 30.
MANRESSA, M., and DRAPO, D. 1938. Studies on the Breeding Habits of Cattle,
Col. Agr. Philippine Univ., Vol. 26, March.
642 BREEDING AND L'VIPROVEMEN '1' Of!' FARM ASIMALS
MCCANDLISH, A. C. 1922. Influence of Age at the Time of Freshening on Produc-

tion of Dairy Cows, Iowa Agr. Expt. Res. Bul. 73.
" - - , GILLETTE, L. S., and KILDEE, H. H. 1919. Influence of Environment and
Breeding in Increasing Dairy Production, II. Iowa Agr. Expt. Sta. Bul. 188.
McDOWELL, J. C. 1928. Comparison of Purebred and Grade Dairy Cows, U.S.
Dept. Agr. Cir. 26, 1928.
MCPHEE, H. C., and WRIGHT, S. 1925. Mendelian Analysis of the Pure Breeds of
Livestock, III. The Shorthorns, Jour. Hered. 16:205-215. 1926. IV. The
British Dairy Shorthorns, Jour. Hered., 17:397-401.
MISNER, E. G. 1938. Relation of Size of Cows to Production, etc., N.Y. (Cornell)
OLSON, T. :\1., and BIGGAR, G. C. 19,22. Influence of Purebred Dairy Sires, S. Dak.
PARSONS, C. H., and RICE, V. A. 1930. Strains and Systems of Breeding in the
1,000-pound Fat Producers of the Holstein-Friesian Breed, Holstein-Friesian
World, ]\Iar. 22, Mar. 29, and Apr. 5.
PUTNAM, D. N., and BOWLING, G. A. 1940. Selecting Sires to Control Butterfat
Test, Ayr. Dig., 26(3) :3-5.
REED, O. E. 1938. Breeding Dairy Cattle for ·Efficient Production, Jersey Rul.
and Dairy World, 57:775, 820-822.,
RICE, V. A. 1944. A New Method for Indexing Dairy Bulls, Jour. Dairy Sci., 27
(11):921-936.
1941. Many Are Called-Few Should Be Chosen (Selective Registration)
Holstein-Friesian World, Feb. 1.
1939. The "400" Guernsey Bulls, Guernsey Breeder 8' Jour., No. I, Nov. 15,
and Dec. 1.
1937. Selection the Key to Better Breeding, Jersey Bul. and Dairy World,
June 9.
1935. A Genetic Analysis of the May Rose Family of Guernseys, Guernsey
Breeders' Jour., Dec. 15.
1933. Concerning Bull Indexes, Guernsey Breeders' Jour., Apr. I, Apr. 15,
and J\1ay l.
1932. The Great Ghost Conversion Factors, Ouernsey Breeders' Jour.,
July 1.
}<')32. Should a Bull's Dam Have a High Record, Guernsey Breeders'
Jour., Jan. 15.
SAVAGE, E. S., and CROWE, E. F. 1936. Forecasting Daughters Production from
Parental Pedigrees, Guernsey Breeders' J onr., 49 :330-332.
SEATH, D. M. 1940. The Intensity and Kind of Selection Actually Practiced in
Dairy Herds, Jour. Dairy Sci., 23(910):931-951.
- - - and LusH, J. L. 1940. Nicking in Dairy Cattle, Jour. Dairy Sci., 23 :103-
l13.
SMITH, A. D. B. 1928. Inbreeding in Jersey Cattle, Brit. Assoc. Adv. Sci. Rpt.,
69 :649-655.
1937. A Statistical Inquiry into the Inheritance of Milk Yield in Three
Herds of Dairy Shorthorn Cattle, Jour. Dairy Res., Vol. 8, No.3.
- - - and- ROBINSON, O. J. 1931. The Inheritance of Milk Yield, Internatl.
Dairy Cong., Copenhagen, Eng. ed. 127.
S.MITH, J, A. B., and DASTUR, N. N. 1940. Studies on the Secretion of Milk Fat,
Biochem. Jour., 34(7) :1093-l107.
STAFF, C. 1935. Interpretation of Milk Production Records by Use of Standard

Production Graphs, Larrowe Milling Co., Detroit, Mich.
SWETT, \Vo "To, GRAVES, R. Ro, and l\IILLER, F. \V. 19280 Comparison of Conforma-
tion, Anatomy, and Skeletal Structure of a Highly Specialized Dairy Cow and a
Highly Specialized Beef Cow, Jour. Agro Res., 37:685-717.
THOMPSON, A. A. 1931. Color in Guernsey Milk, Guernsey Breeders' Jour., Nov. 15,
Dec. 1, and Dec. 15.
TURNER, C. W. 1927. The Mode of Inheritance of Yearly Butterfat Production
(Jersey), Mo. Agr. Expt. Sta. Res. Bul. 112.
- - - . 1925. A Comparison of Guernsey Sires, Mo. Agr. Expt. Sta. Res. Bul. 79.
Many buUetins on endocrine function in lactation published by Mo. Agr.
Expt. Sta.
TYLER, W. J., and HYATT, G., JR. 1948. The Heritability of Official Type Ratings
and the Correlation between Type Ratings and Butterfat Production of Ayrshire
Cows, Jour. Dairy Sci., 31(1):63-70.
WEIRETHER, F. J., et al., and KLEIN, L. A., et al. 1941. A Preliminary Report on the
Effect of Colloidal Silver Oxide on Bovine Mastitis, Amer. Jour. Vet. Res.,
2(3) :141-151.
WILLARD, H. S., and QUAYLE, W. L. 1936. Producing Better Dairy Cattle, Wyo.
WINKJER, J. G. 1939. Cooperative Dairy Bull Association, U.S. Dept. Agr. Farmers'
Bul. 1830.
WOODW ARD, T. E. 1945. Some Studies of Lactation Records, Jour. Dairy Sci.,
28 :209-218.
YAPP, W. W. 1938. Relation on Regression and Selection, Amer. Soc. Anim. Prod.
Proc., pp. 272-277, Dairy Sci. Abs., Vol. 1, No.3, November, 1939.
YODER, D. :VL, and LUSH, J. L. 1937. A Genetic History of the Brown Swiss Cattle
in the "Gnited States, Jour. Hered., 28(4):154-160.
CHAPTER XXII
SELECTION IN MEAT ANIMALS

by
DR. K J. WARWICK
Basically the task of the breeder of meat animals is to produce animals

which will more efficiently convert vegetable products (some edible by
man and some not) and inedible waste animal products into nutritious
human food products of the quality which will appeal to the human
palate. In a world with constantly increasing human population and
a consequent narrowing of the margin of safety between world food needs
and potential world production, it is essential that this conversion be
made as efficiently as possible. From the standpoint of the individual
commercial-livestock producer, profit depends upon byo things: (1) the
cost of production and (2) quality of product as reflected in selling price.
Thus, his interests in efficiency are identical with those of the population
at large so far as the kind of animals needed is concerned.
In many ways the task of selection in meat animals is simpler than i;;
the catle in dairy cattle. The principal favorable factors are: (1) Many
characters of economic importance in meat animals are expressed by
both sexes; (2) many such characters are expressed before sexual maturity,
thus making selection on the basis of individuality possible at much
younger ages; and (3) for at least certain characters, e.g., carcass quality,
there is a greater correspondence between type of the live animal and
true value than there is between type and production in dairy cattle.
Arrayed against these advantages are certain disadvantages which
limit progress. Of these the fact that we have no single end point which
can be used to measure value is perhaps the most important. As pointed
out in an earlier chapter, selection for more than one thing reduces the
amount of selection which can be directed toward anyone thing. Since
meat animals must be selected simultaneously for economy of production
and for quality of product and since both of these things depend upon a
number of factors some of which may be biologically antagonistic, the
difficulty of the meat-animal breeder becomes quickly apparent. Added
to these things is the fact that quality of product can best be evaluated
only in the carcass and that an animal studied in the carcass cannot be
used (or breeding!
\
644
\
SELEC'1'IO N IN Ml!JA'1' ANIMALS
It cannot be emphasized too strongly that while as an art the breeding

of meat animals is old, it is very new as a science. Thus future research
work may well change some of our current concepts.
The question of whether we should strive to breed animals which
are good in terms of all-around value and then breed them "pure" or
"straight" for commercial production or whether it is more feasible to
develop certain strains which are superior in one character and then
cross them ,dth strains superior in ot.her characters for commercial
production is one ·which must be considered at present as unanswered.
Both procedures have been used in the past, although the philosophy
which has largely dominated the thinking of American livestock people
has been that of producing desirable purebreds and then using them or
lS9.- Grand Champion barrow, 1948 National Barrow Show, Austin, Minn.
·FIG .
Owned and shown by Portage }"arms, Woodville, Ohio.
their high grades commercially. Even where crossbreeding has been

practiced extensively (as "o:ith hogs), crosses have often been made to
take advantage of heterosis between breeds which have each been
selected along similar lines. The system of using strains which are
selected for all-around merit (even if used in crossing) is the most preva-
lent in the United States. Most of the material in this chapter assumes
a breeding system aimed at the production of animals of all-around
merit. Some mention will be made of the use of divergent strains for
crossing.
The best example of crossing markedly different types in the United
States is the system of crossbreeding sheep followed in many range areas
in the West. Animals of fine-wool breeding (principally of the Ram-
bouillet breed) possess the vigor, hardihood, and flocking instinct neces-
sary for range-sheep production but lacking in most other breeds. The
fine-wool types do, however, lack size and mutton conformation. In an
over the face shouid be indicative of a strong wool-growing tendency and

therefore indicative of fleece production. These two assumptions served
as the basis upon which several breeds of fine-wool sheep were selected
for an extremely wrinkled condition and upon which both certain fine-
wool and medium-wool breeds were selected for an extensive face covering
-extending in extreme cases clear to the muzzle.
FIG. 191.-Dressed carcasses of large-, intermediate-, and small-type hogs, slaughtered at

approximately 225 lb. in weight. (From O. G. Hankins, A Study of Carcass Characteristics in
Relation to Type of Hog, Amer. Soc. Anim. Prod. PTOC., 1940.)
Systematic studies over a period of years especially by the U.S. Depart-

ment of Agriculture (Spencer and Hardy, 1928) and the Texas Agricultural
Experiment Station (Jones et al., 1944) have shown that the wrinkly
types do produce more pounds of grease wool than the smooth types.
The differences practically disappear in scouring, however, so that the
weight of clean wool produced by the two types is almost equal. In
addition, the staple length was longer in the smooth types, and fiber
diameter was more uniform. Observation indicates that the smooth
types a e easier to shear and less susceptible to fly-strike. Thus, the
evidence seems to be clear that the wrinkly types do not have the wool-
\,
SELECTION i N MEAT ANIMALS 651
producing superiority claimed for them and that they are undesirable in
other respects.
Extensive studies have likewise shown that extreme face covering is
undesirable. Probably the deficient production is dependent upon the
FIG. 192.-Ab01'e, a champion B type Rambouillet ram showing beavy neck folds and a
moderate degree of wrinkling on the body . B elow, a range Rambouillet ram almost free
from wrinkles. Note the difference in degree of face covering in these two rams. (Courtesy
oj Bureau of Animal InduatTy, U.S . Department of AOMcultw·e.)
excessive growth of wool covering the eyes and causing "wool-blindness"

which interferes with the ability of the animal to feed properly. Figures
in Table 51 taken from a recent report of Terrill (1949) indicate that
under range conditions Rambouillet ewes with covered faces are seriously
deficient in lamb production. The last column iI1 t4e table, indicating
\'IIet \ YC ,e
& An r 1 t l
~~~ lJo.l'('~_A~,
652 BREEDING .AND I M PROVEM ENT OF Ff!RM AN I MAL S
FIG. 193.-Variation in degree of face covering in sheep. The animals shown in E and Fare
definitely "wool blind." (Courtesy of Western Sheep Breeding Laboratory, Du~ois, Idaho.}
a difference of 11.1 lb. in lamb produced per ewe bred, is especially

striking.
TAIlLE 51.-RELATION OF FACE COVERING TO LAMB PRODUCTION
% lambs Pounds
No. % ewes % lambs % live Av. % lambs
weaned of Jamb
Face of lambing born of lambs of weaning weaned
of live weaned
covering ewe of ewes ewes lambs weight, per ewe
lambs per ewe
years bred lambing born lb. bred
bred
born
---
Open .... 286 95 .5 126.7 91.9 89.0 76.3 99.0 i5.5
Partially
covered 845 95.4 124.2 92.3 88.1 74.8 96.3 72 .1
Covered. 1,557 91.9 119.7 90.0 88.6 73.4 87.7 64 .4
I I , ,
SELECTION IN MEAT ANIMALS 653
Not only were the covered-face ewes inferior in lamb production, but
\yere not importantly superior in wool characters. A few statements
from Terrill's paper! give a concise summary of the work:
Ewes with open faces produced 11.3 percent more lambs and 11.1 more pounds
of lamb per ewe bred than those with covered faces. Ewes with partially covered
faces weaned 8.6 percent more lambs and 7.7 more pounds of lamb per ewe bred
than those with covered faces. Differences in fare covering within these groups
were associated with corresponding differences in lamb production. These ad van-
tages for ewes with open faces occurred in spite of three periodic clippings around
the eyes of all ewes subject to wool blindness.
About 46 percent of the advantage of open-faced ewes was due to a greater
number of lambs born per ewe lambing; 26 percent was due to higher weaning
weights; 19 percent was attributed to a higher proportion of the ewes becoming
pregnant; and 9 percent was due to greater viability to weaning of offspring.
Open-faced ewes excelled covered-faced ewes in lamb production at each
year of age. The greatest advantage for open-faced ewes in pounds of lamb
per ewe bred was found at 3 years of age followed in order by 2, 4, 6, and 5 years.
The yearling grease and clean fleece weights and staple lengths of 2,499 Ram-
bouillet ewes and the lifetime grease fleece weights of 798 Rambouillet ewes were
slightly greater for covered-faced ewes than for those with open faces. The
differences were not significant except for staple length and were not large enough
to be economically important.
The great economic importance and high heritability of face covering indicate
that it should receive as much or more attention in selection than any other trait
in sheep if wool blindness is a problem.
The moral to be drawn from such studies is obvious, namely, that
present-day breeders should critically examine existing breed standards
and, if necessary, research should be carried out to determine whether
they are compatible with the production of the most efficient and profit-
able animals. Obviously, such critical tests should be applied to fads
which are certain to be started in the future before and not after they
have become so widespread as to seriously handicap the usefulness of a
breed.
Evidence that responsible leaders among breeders are aware of the
necessity of this approach comes from the fact that the American Here-
ford Breeders' Association is currently (1950) sponsoring research in
cooperation with four state experiment stations on the question of type
in that breed. The need for such work arose from the facts that (1)
over a long period of years, show rings have tended to favor an increasingly
thick, compact, early maturing type of animal which seemingly has been
accompanied by some loss of size and ruggedness, and (2) during recent
,years an ultracompact type known as" comprest" has arisen in the breed
1 Jour. Anim. Sci., August, 1949, pp. 360-361.
and has found favor in some livestock exhibitions. Objective evidence

on the relative value of these various types for commercial beef produc-
tion is lacking, and the officers of the association are to be commended
for their research approach to the question.
Research on the problems at the Colorado Station has to date shown the
following trends: 1
1. There were no consistent or great differences in pounds of feed required to
make a pound of gain in steer calves of conventional or small types. Small-
type steers gained less per day and consumed less feed per day. Rate of gain
and efficiency were not associated between these groups.
2. Small-type steers when killed at the same degree of fatness as conventional-
type steers weighed about 150 pounds less than the larger type.
3. The percentages of carcasses in the different wholesale cuts were almost
identical for the two types of steers. The carcasses of the small steers
weighed 402 pounds, those of conventional type weighed 500 pounds at the
same slaughter grade.
4. Cows of large, intermediate, and small types have eaten amounts of hay
during the winter almost directly proportional to their body weights.
Heifers, full fed, consumed hay relative to body weight, but required con-
siderably more hay per unit of body weight than did cows.
The work as being carried out at the other three stations includes
steers of small, medium, and large types with representatives of each type
being fed out under three different regimes, namely (1) fattening by
full feeding as calves, (2) grazing as yearlings followed by a dry-lot fatten-
ing period, and (3) fattening on grass as two-year-olds. Quite possibly
the same type of steer will not prove to be best under all three types of
management.
The position an individual breeder should take if some fad he knows
to be uneconomic should sweep his breed is a question for thought.
Should he follow the fad in order to produce animals for which there is
an immediate market or should he make the financial sacrifice of staying
with an unpopular but basically better type? That this is not a hypo-
thetical situation is illustrated by the story of Peter Mouw of Orange
City,. Iowa.
During the period around 1900, and particularly during the few years
immediately following the turn of the century, the Poland-China breed
of hogs was swept by the so-called "hot-blood" craze in which selection
was for an extremely small, short-legged, refined, early maturing type
of hog. Mr. Mouw c,nd a few other breeders believed this to be a
basically unsound type of hog and continued to produce a more rugged
1 Beef ,Bteeding Research in Colorado, Progress Report, Colorado Agr. Exp. Sta.,
July 6, 1950 (mimeo).
\
\
\
"big type," which was often the subject of ridicule at the shows. The
tide finally turned, however, and overnight these hogs became the most
popular in the breed. Adherence to an ideal paid off with a small
fortune to Mr. Mouw in this case.
Many breed trade-marks are neither positively nor negatively corre-
lated with productivity but are nevertheless a problem for breeders.
The amount and distribution of white in breeds with white color markings
apparently is very low in heritability. Genetic knowledge at present
indicates that rigid adherence to a set color pattern is difficult, if not
impossible, to fix. Therefore, if the breed requires certain markings,
3 portion of young animals will be ineligible for registry regardless of
their excellence in other regards. Discarding these animals automati-
cally lowers the amount of selection which can be practiced for more
important characters and cannot help being a drag on the progress of
the breed. A statement several years ago by the late E. M. Harsch,
secretary of the Hampshire Swine Registry Association, in which he said
that "a few white hairs on the hind legs or a black spot the size of a
dime in the belt is truly a long ways from a pig's heart" seems to express a
common-sense attitude toward minor deviations from a set color pattern.
To summarize our discussion of type, it can be said that the breeder
should endeavor to select for the fundamentals of good meat type----=-Items
known to be positively correlated with true worth and regarding which
our ideals are not likely to change greatly within the forseeable future.
On the other hand it is imperative that he not espouse a new fad until
it has been shown to be clearly superior to existing types. He should
avoid placing undue emphasis on "frills" or "fancy points," even if
'they are in themselves harmless.
Use of Production Records as a Basis for Selection in Meat Animals.-
Since meat animals are kept for profit, it would seem axiomatic that
selection should be directed principally toward those things which have
a direct influence on productivity (in the broad sense) and thus on
profitability. If type were perfectly correlated with productivity, it
couldlogically serve as the sole basis for selection. Siqce, as we have
seen, this correlation is at best of a low order, a priori reasoning indicates
that selection could be better if based on actual production records
themselves. Selection for type will then be automatic in so far as it is
correlated with productivity.
The ease of measuring speed, milk production, and egg production, in
racehorses, dairy cows, and laying hens, respectively, led to the use of
production or performance records in selecting breeding animals in these
classes of stock long before such records were used for meat animals.
The problem of using production records in meat animals is more
656 BREEDING AND IMPROVEMENT OF FARM AN!J1/ALi)
difficult than in the classes of stock just mentioned because no one

measurement accurately evaluates the worth of an animal. Extensive
studies have been made of the factors which should be considered in
breeding meat animals, however, and today we have a fairly clear idea
of the things which should be considered, although the exact importance
which should be given to each factor is not known. It probably varies
under different conditions. For example, in producing for a discriminat-
ing market, quality of product may be the most important factor influ-
encing profitability. On the other hand, in producing for a market
which does not pay for quality, economy of production is the most
important.
Specific record-of-performance procedures for each class of livestock
have been developed which should serve as the basis for selection, regard-
less of whether selection is based on individuality, pedigree, or progeny
test. The use and limitations of such procedures will be discussed in
later sections on each class of animal.
Certain general procedures must be followed if production records an~
to be kept and used for selecting breeding animals. Accurate pedigree
records must be kept, and each individual animal must be positively
identified by means of ear notches, ear tags, brands, or tattoos. Accurate
records of birth dates must be kept, and weight records must be taken
at certain specified ages. In so far as possible all young animals should
be raised under similar conditions of care and management, since records
of relative growth rate would be worthless from a genetic standpoint
if animals had been fed differently. Animals should be kept under
practical conditions, since records made under forced conditions (such
as with nurse cows) may have little relationship to true value under
normal conditions.
Thus, it can be seen that the keeping of production records represents
an expenditure of time and labor-an expenditure which is highly worth
while, however.
The exact record forms to be used will depend upon the person using
them. Some forms which have been found satisfactory are given later in
this chapter. They will serve as a source of ideas for the person who
desires to design his own records. -
Although production testing may be worth while as an aid in culling
the female side of a herd or flock, it loses most of its effectiveness if sires
without production records are continually introduced into the herd.
Thus, most of the discussion in the remainder of this chapter assumes
either (1) a closed herd with home-produced sires being used or (2) the
purchase of sires from production-tested herds. In the latter case a
purchaser\should study the environment of the herd where sires are
\
produced in relation to that of his own herd to be sure the production
records do not depend upon skillful management alone.
Relation of Environment to Selection in Meat Animals.-H has long
been recognized that animals of improved meat types tend to resemble
unimproved types when raised under suboptimal nutritional conditions
-hence the origin of the old saying that "half the breeding goes in the
PLAN OF EXPERIMENT
LIVE WEIGHT GROWTH CURVES TO BE SECURED BY QUANTITATIVE CONTROL
OF THE PLANE OF NUTRITION.
(DIFFERENCES IN SHAPE OF GROWTH CURVES BETWEEN ALL TREATMENTS TO BE
ACCENTUATED IF POSSIBLE)
200----------
200 -------------------------------- lb. 200-----------:3000oys
lb. 2001b.
I Kill
180
'">
:.J
20 24 28 32 36 40
Age-Weeks
FIG. 194.-Plan of experiment to determine the influence of plane of nutrition at different
periods with respect to age upon growth rate and carcass composition in bacon hogs.
(Courtesy of Dr. C. P. Mc1IIeekan and Dr. John Hammond, Cambridge University, Cambridge,
England.)
mouth." Moulton et al. (1921 and 1922) showed that in cattle the major
tissues, bone, lean, and fat, were differentially affected by a low plane of
nutrition, with bone growth being affected the least and fat deposition
the most. Several other workers have confirmed this basic tendency
in other species of farm animals.
McMeekan (1940-1941) made a detailed study of the growth and
carcass composition of pigs of an inbred Yorkshire strain when fed from
birth to a slaughter weight of 200 lb. on various feeding regimes. Pigs
were divided at birth into two groups, with one fed on a high plane and
the other on a low plane for the first 16 weeks of life. At that time half
the pigs on each plane of nutrition were shifted to the other so that four
658 BREEDIN G A N D I MPRO V FJMEN'1' OF FA R M ANI M ALS
nutritional regimes resulted, namely, groups fed on (1) a high plane from
birth to slaughter (high-high), (2) a high plane from birth to 16 weeks
and a low plane from that time to slaughter (high-low), (3) a low plane
from birth to 16 weeks and a high plane thereafter (low-high), and (4) a
low plane from birth to slaughter (low-low).
It was found in this work that if the pig is grown so slowly as to be
stunted early in life, the early growing parts of the carcass, such a
head, legs, and bone and to a lesser extent muscle, are permanently
reduced in size. When these animals are later put on a high :plane of
nutrition, the later developing parts, particularly fat) grow rapidly and
result in a carca, s too fat for bacon. Conversely, pigs fed from birth
on a high plane have a rapid development of the early growing parts and
FIG. 195.-Influence of plane of nutrition upon carcass composition of bacon pigs as illus-
trated by cross sections at the last rib. Pig 83, killed at 168 days of age, maintained on high
plane of nutrition throughout life. Pig 85, killed at 196 days of age, was changed to a low
plane of llutrit ion a t 16 weeks. Pig 89, also killed a t 196 days, grown On a low plane of
nutrition the fir st 16 weeks. Pig 99 was maintained on a low plane of nutrition t hroughout
and required 315 day s to reach the standard slaughter weight of 200 lb. (Cow·tesy of Dr .
C. P. lvIcMeekan and Dr. John Hammond, Cambridge University, Cambridge, England.)
if later put on a low plane of nutrition produce carcasses "ith a high
content of lean. In a similar experiment with sheep Verges (1939)
showed that carcass composition may be materially altered by controlling
the shape of the growth curve.
These experiments and others which could be quoted point definitely
to the fact that environment exerts a control over development. The
animal breeder is vitally interested in the questions of (1) how to properly
evaluate known differences in environment so far as their effects on
potential breeding stock are concerned, and (2) what type of environ-
ment should be used in breeding foundation stocks for ultimate commer-
cial use.
The old saying, that "fat is a good color," shows that the first of these
problems has long been recognized. From the standpoint of the sales
value of an animal it often happens that putting on a degree of finish
far and above what the animal needs for proper development will improve
the appearance (apparent type) enough to be a profitable investment for
the sellek This is apparently the principal reason for the high and often
excessive degrees of finish so often seen on meat animals in our shows and
sales rings.
The question the practical breeder, interested in true improvement
more than immediate sales value or show-ring glory, has to answer is
how he can logically make allowance for differences in environments to
which various animals have been subjected during growth. There
appears to be no absolute rule which can be laid down, but certainly
the growth rate and apparent type of an animal should be discounted
somewhat if it is knmvn to have been forced unduly. Conversely, an
animal reared under poorer-than-average conditions should be given some
credit over and above what its record and appearance would entitle it to.
In some cases, such as the difference in growth between twin and single
lambs, it is possible to determine the average effect of a certain environ-
mental factor and make suitable adjustment for it.
The question of what environment a breeder should provide for his
animals is a broad question for ,yhich no one answer can be given. It has
sometimes been assumed that animals should be bred in the best environ-
ment possible and that the commercial producer who will use the animals
produced should strive to bring the environmental conditions under which j
he raises his stock up to the same level.
This viewpoint is unrealistic because there are vast areas, such as
range country, subtropical areas, etc., ,,,here environmental conditions
can in all probability never be made ideal but which can if properly used
support large numbers of meat animals and add tremendously to our
meat and wool supplies. The same argument applies to the utilization
of much feed that would otherwise be waste in areas where more intensive
agriculture prevails.
The prevailing opinion among animal-breeding specialists in the United
States seems to be that animals should be bred and selections made
under the same environmental conditions where their descendants will
be used commercially. This general plan would seem to have much
to commend it since adaptation to the environment at hand is the first
essential for successful production.
The folly of departing far from this plan can be ilbstrated by the
general low productivity of our common breeds of beef cattle when
placed under tropical or subtropical conditions where their thermo-
regulatory mechanisms are inadequate for keeping body temperatures
down. The development of these breeds in temperate climates has not
fitted them for use in hotter areas.
At the present time there is a considerable amount of effort being put
into the development of strains of beef cattle suitable for use in the
southern regions of the United States and containing various percentages
•
660 BREEDING AND nJI>!WT'EMEN'l' OF Fl1RJf l1NDfALS
of the blood of Zebu or Africander cattle which carry more genes for heat
and insect resistance than are present in our beef breeds. The best
known of these is the Santa Gertrudis (Rhoade, 1949) which is already an
established breed. It would seem axiomatic that development of these
types should continue in that region, even though regions with better
feed conditions might be found.
Hammond (1940, 1947) has developed a point of view w4ich differs
somewhat from that outlined above. He feels that, if the environment
is so poor as to put a virtual ceiling on development, there will be little
chance for inherent differences in gaining ability, type, or production to
be expressed and that under such conditions there is little or no chance
for a selection program to be effective. He therefore recommends that
if we are attempting to make improvement in a given character We
should place the animals in an environment 'which will develop that
character to the utmost and then make selections from among those
which show it to the greatest degree, even if the descendants are later
to be used under poorer conditions.
No experimental evidence is available to indicate which of these rather
divergent viewpoints is correct. The most common procedure to date
in the United States seems to have been more in line with Hammond's
viewpoint. Our purebred herds and flocks have for the most part been
maint'1ined in areas where feed conditions are good and other environ-
mental conditions not extreme. Animals from these herds have then
provided fO\lndation stock for commercial production in all areas of the
country. This situation, together ,vith the nation-wide promotional
activities of many of our breed associations, has sometimes tended to
result in animals of the same general types and breeds being raised in
virtually all parts of the country regardless of their adaptation.
Since meat animals are often maintained in this country under range
conditions or other types of environment which may not be adequate
in many cases to produce animals with enough finish to meet market
demands, it is common to use such areas as breeding grounds, and to
move the potential market animals into an area where better feeding
conditions prevail for a fattening period before marketing. Such animals
must thus have the abilities to (1) survive and grow under suboptimal
conditions and (2) fatten readily and produce desirable carcasses when
given the opportunity.
Therefore, the logical system to use in producing sires for use in range
herds and flocks would seem to be to maintain the purebred breeding
herd under typical range conditions and raise the potential sires to
weaning there. A preliminary culling could be made at weaning, elim-
inating those which are undersized or otherwise undesirable. The
remaining animals could be put on feed and further culling done on the
basis of rate and efficiency of gain and apparent body type after a feeding
period.
Selection in Swine.-From a breeding standpoint the following items
are of primary importance in determining the efficiency and profitability
of swine:
1. Litter size. / "./
2. Weight per pig and per litter at weaning time.
3. Daily gains between weaning an~ing. j
4. Efficiency of feed utilizatiOil':' v
5. Body type or carcass desirability. 1./
Although greatly influenced by herdsmanship, the number of pigs
farrowed and raised and their weight at weaning time together form
the best measures we have of prolificacy and mothering ability of sows.
They are particularly important since they have a tremendous influence
on profitability. On the average, a sow will have consumed a total of
1,51)0 to 2,_QOO_lb. of feed during the period between breeding and the date
her litter is weaned. If this all has to be charged against a litter of only
four or five pigs, the chance of eventual profit is small. In addition to
number of pigs farrowed and raised, the weaning weight of the pigs is
important since it has been definitely shown that the pigs which are the
heaviest at weaning time reach market weights the soonest.
Daily gain from weaning to market weight is important because the
fast-growing pig is on hand fewer days, thus requiring less labor and
fewer days of shelter and exposure to risk of disease. Another important
consideration is that rate of gain and efficiency of feed utilization are
related, the correlation having been shown in several studies to be about
+0.70. With this relationship and the difficulties of keeping feed records
under practical conditions, most breeders will be content to select for
efficiency of feed utilization indirectly through selecting fast-gaining pigs.
In selecting primarily for rate of gain, there may be danger of selecting
for excessive fatness since, as discussed later, these characters may be
correlated to some degree.
As is to be expected, the estimates of heritability for various characters
in swine (Table 52) show considerable variation from study to study but
do show general trends. The heritability of litter size at birth is evidently
low, only two of the nine estimates being over 22 per cent. Most
investigators have found heritability of this important character to be
less than 20 per cent. Three estimates of heritability of litter size at
about weaning age range from 17 to 31.7 per cent, indicating that this
character is likewise low in heritability. Similarly, weaning weight of
litter is apparently low in heritability.
GG2 BREEDING AND IMPROVEMENT OF FARM ANIMALS
TABLE 52.-EsTIMATES OF HERITABILITY FOR VARIOUS CHARACTERS IN SWINE
Herita- Method used to

Charactpr bility, determine heritability Reference
%
Litter size at birth 17 Maternal half-sib litters Lush and Molln (1942)
1O}
17
18
Estimated from published Lush and :\Iolln' (1942)
reports of various workers
34
44
13 Maternal half-sib litters Hetzer et al. (1940)
14.5 Average of 3 methods Stewart (1945)
21. 7 Intrasire regression Cummings et al. (1947)
24 Average of 2 methods Blunn and Baker (1949)
Litter size at wean- 17 Maternal half-sib litters Lush and MoUn (1942)
iug 22 Average of 2 methods Blunn and Baker (1949)
31.7 Intrasire regression Cummings et al. (1947)
Litter size at 70 20 Maternal half-sib litters Hetzer et al. (1940)
days
Weaning weight of 18 Maternal half-sib litters Lush and Molln (1942)
litter 7.4 Intrasire regression Cummings et al. (1947)
37 Average of 2 methods Blunn and Baker (1949)
Survival 40.3 Intrasire regression Cummings et at. (1947)
Pig weight at 56 0 Intrasire regression Comstock et at. (1942)
days of age 15 Paternal half-sib Baker et al. (1943)
0 Paternal half-sib Nordskog et al. (1944)
13.6 Paternal half-sib Krider et al. (1946)
Pig weight at 180 14 Intrasire regression Comstock et al. (1942)
days of age 20 Paternal half-sib Whatley (1942)
62 Intrasire regression Whatley (1942)
30 Intrasire offspring-dam Whatley (1942)
correlation
40 Full sibs, not litter mates Whatley (1942)
30 Regression of variance to Whatley (1942)
genetic relationship
I 23 Paternal half-sib and in-
trasire regression
Whatley and Nelson (1942)
19 Line differences due to Krider et ai. (H)46)
selection
23.9 Paternal half-sib Krider et al. (1946)
Market score at 33 Average of ,2 methods Whatley and Nelson (1942)
slaughter
---
Conformation score 20 Intrasire regression Stonaker and Lush (1 ()42)
Type of score:
Within strains 38 Paternal half-sib Hetzer el al. (1944)
Between strains 92 Paternal half-sib Hetzer el al. (1944)
SELECTION IN MEA.T ANilvIALS 663
All estimates of the heritability of individual pig weights at 56 days of

age are low, indicating that weTght at that age is largely a function of
the nursing ability of the sow rather than of the pig's mvn genes. After
·weaning, however, the pig's own genes evidently exert their influence,
and heritability i~~re3:s~Jo possibly about 30 per cent at ISO days of age.
Heritabilities of various type and conform~.!.ion scores averaK~around
30 per cent in -studies where only animals of basically similar types were
studied, but in one study where strains of different types were considered,
an estimate of 92 per cent for the heritability of type score was found.
The latter figure probably explains why breeders have been able to
alter hog types rather rapidly upon occasion in the past, while the figure
of 30 per cent within strains indicates why further improvement often
seems to come so slowly in a herd already fairly good.
In order to put the above figures in more practical termf', let us consider
a hypothetical example. If the inheritance of these characters were
assumed to be all due to additive gene action, if we assume that litter
size weaned and litter wean'1i g·weight are each 20 per cent heritable,
and ISO-day weight and conformation each 30 per c~nt heritable, and
that we have herd averages for the various items as indicated, then the
progress to be expected from one generation of selection would be as
shown herewith.
A f
\
Selected
A I I Expected
Vt·· ,Herita- perform-
h:~~ individ-
uals
Ise Iec ,IOn
d t
a van age
,.
blhtv, % ance next
. generation
.
Av. No. of pIgS weaned, ..... ,I 6,0
I~--~----I---
6.6 +0,6 20 6.12
Av.litterweaning weight" .. ,.!IS0,0 19S.0 +lS,O 20 IS3,60
Av. individual ISO-day Fcight.11SO,0 195,0 +15.0 30 lS4.50
Av.typescore* ....... , ...... 1 5 7 +2.0 30 5.6
* Graded on a scale from 1 to 9 with 1 being inferior and 9 excellent.
The selection advantages indicated for each character are approxi-

mately of the size it has been found possible to make in herds of the
Regional Swine Breeding Laboratory when selections are made largely
upon the basis of production records.
The gains expected for anyone character are relatively small and,
even if they were all attained genetically, environmental differences from
year to year are great enough to make it difficult to know whether progress
is being made until records have been kept for several years.
There are several reasons for believing that the figures given are them-
lSplves too optimilStic when all factors are being considered at the same
time. The bases for this statement are: (1) Recent unpublished studies
at several stations! participating in the Regional Swine Breeding Labora-
tory ha,ve shown that the expected advantages from selection have not
been realized. These studies are complicated by the fact that most of
the lines studied are being inbred and this may be depressing performance
more than realized. Then, too, the studies cover a period of only a felV
years. (2) There is a continuous automatic selection for litter size
which has apparently been going on for centuries without raising average
litter size above its present modest average.
This last point is something which has apparently not been considered
in previous animal-breeding texts and which may need a word of explana-
tion. To understand this, it is necessary to differentiate between
"average litter size raised" per sow farrowing and" average size of litter
in which raised" for the pigs raised. To make this clear, let us consider
an actual example of a group of 12 litters born to gilts in an inbred line
on a Purdue University farm in 1948.
Size of litter No. of Total No. of pigs raised Litter size X

(~o. of pigs raised) litters in litters of this size No. of pigs raised
--
8 1 8 64
7 1 7 49
6 5 30 180
5 2 10 50
4 1 4 16
2 2 4 8
Totals ..... ,. ....... 12 63 367
"Av. litter size raised" per sow farrowing = 6H2 = 5.25

"Av. size of litter in which raised" per pig raised = 36%3 = 5.82
If a person selected breeding stock from this group of litters entirely
at random so far as litter size was concerned, he would be making a
positive selection of 5.82 - 5.25 = 0.57 of a pig per litter over the average
of that generation. This is, of course, due to the fact that the larger
litters have more pigs available. The previously mentioned Regional
Swine Breeding Laboratory study indicates that a high percentage of
the selection practiced for litter size is of this automatic type.
It is possible that in breeders' herds animals from small litters are
better grown and are therefore selected for breeding purposes with a
resultant lack of positive selection for litter size. The fact that there
1 Iowa, Nebraska, Missouri, Illinois, Indiana, Wisconsin, and Oklahoma stations.
The data are now being summarized for pUblication by Dr. G. E. Dickerson of
:\fissouri. '
SELECTION IN MEAT ANIMALS GG5
is by no means an absolute negative correlation between litter size and
rate of development, together with the fact that many breeders have
consciously selected for large litters, makes this seem unlikely.
If, as seems likely, litter size is failing to respond to selection as rapidly
as it theoretically should, the reason is probably either (1) prolificacy
depends to a certain extent at least upon "overdominance," with the
best performers being heterozygous for a larger-than-average number of
gene pairs, or (2) there is a negative correlation between prolificacy and
certain other factors so that when we select for some of these, such as
rate of gain, we indirectly select for smaller litters and override any
actual positive selection we have made for litter size.
No positive evidence for" overdominance" has as yet been advanced
·with swine, although the possibility that it may exist cannot be over-
looked. Dickerson and Grimes (1947) and Dickerson (1947) have
presented convincing evidence that in the material with which they
worked rapid gain was correlated genetically with (1) rapid fat deposition
(r = around 0.60) and (2) low feed requirements (r = 0.78). The data
strongly suggested that poor suckling ability may be caused by the same
genes responsible for rapid fat deposition and low feed requirements.
Blunn and Baker (1947) working with a different herd of hogs found
no significant genetic correlation (actually slightly negative, r = -0.04)
between rate of gain and carcass fatness as estimated from back-fat
thickness. No data on feed requirements were kept in their experiment.
Winters et al. (1949) in an experiment where rate of gain was controlled
by limiting feed intake during certain periods of the feeding period in
some lots found that the slower gaining lots had lower feed requirements
and produced leaner carcasses. They rightly point out that genetically
slow rate of growth may be quite a different thing from environmentally
controlled slow growth. They also point out on the basis of other data
from their swine-breeding project "that a high rate of gain, economy of
feed utilization, and a quality carcass can be combined."
Quite aside from the experimental data on the subject, a priori reason-
ing (always a dangerous procedurel) might be taken as indicative of a
negative relationship between maternal ability and rapid fattening.
Bacon hogs are commonly reputed to raise larger litters than most lard
breeds. Likewise in cattle, no one has yet succeeded in combining the
best beef type with maximum milking ability.
If there is a strong negative association between rate and economy of
gain and maternal ability, the increased use of crossbreeding or crossing
of inbred lines for commercial production is suggested. Dickerson (1947)
makes the following suggestion l from his data:
1 Iowa Agr. Expt. Sta. Res. Bul. 354, October, 1947.
666 BREEDING AND IMl!ROVE1IIENT OF FARM ANIMALS
It suggests that maximum performance can be secured only through judicious

crossing of different strains of swine. For example, sows of a cross that has
exceptionally good milking ability and prolificacy mated to boars from a strain
that excels in rate and economy of post-weaning gains would give maximum
litter performance.
Regardless of whether this association exists, it is far from absolute and

few lines or strains of hogs are being developed by selection based on
one character only. The economics of producing breeding stock make
it probabJe that a line extremeJy deficient in maternal ability but excep-
tionally good in growth rate or vice versa could not be profitably used.
Thus, we will have to continue selecting animals for reasonable excellence
in all qualities. It is true, however, that existing swine breeds differ
significantly in certain characteristics, with some being strong in one
character and some in another. These differences may depend upon
chance fluctuations in gene frequency or upon the type of selection
which has been practiced in the various breeds. As inbred lines are
developed, they likewise differ from each other. Judicious crossing will
take advantage of these differences and may prove to be the method
which will give maximum performance.
Even if reasonably high merit in most characteristics is accepted as
the aim in swine breeding, a question yet to be answered is that of what
system of selection will result in maximum total progress over a period
of several years or generations. Should selection be solely for one trait
until it is improved, then for a second until it is improved, etc., until
finally all characters have been improved to the desired levels? Or
should selection be made simultaneously for aU desired traits by using
some index of net merit arrived at by adding into one figure the credits
or penalties given each animal for its superiority or inferiority in each
trait? Finally, should a method be used in which all traits are considered,
but animals are culled if they do not reach a certain level for each trait
regardless of excellence in other traits? Terming these the "tandem,"
"total score," and "independent culling level" methods, respectively,
Hazel and Lush (1942) arrived at the conclusion that the tandem method
of selecting first for one thing and then for another could be expected to
be considerably inferior to either of the other two methods. The total-
score method of selecting for several traits simultaneously should be
superior to the method of independent culling levels. The latter may
be useful in cases where some culling has to be done before all traits
can be evaluated.
In spite of the admittedly slow rate of progress usually made in select-
ing swine, there is ample evidence that our present hogs show marked
improvement in growth, fattening ability, and other characters over

their wild ancestors, thus showing that selection can be effective in
bringing about improvement.
The best data showing definite improvement in swine over a period of
years come from the records of animals fed out in the testing stations in
Denmark as part of the swine progeny-testing system in that country.
The Danish system of swine breeding, as summarized by Lush (1936),
is based upon a group of state-recognized breeding centers, where a high
percentage of the boars used for commercial production in the nation
are produced. The breeding centers are visited twice annually by a
committee representing farmers' organizations and cooperative bacon
factories and scored for numerous items including type and prolificacy
of breeding stock, management and general appearance of farm, and
health of animals. Each year the breeding centers are obligated to
send enough test litters of four pigs each to average two pigs per sow
in the herd to testing centers. The test litters are fed out under standard
conditions and slaughtered at about 200 lb. live weight. Records are
kept of rate of gain, efficiency of feed utilization, and several carcass
measurements and scores.
Annual reports are issued which include the above information for
each center. These reports are designed to serve as a guide for the
purchase of breeding stock and the establishment of breeding programs.
The first of the testing stations was established in 1907, and the work is
still progressing under a plan which has changed but little since its
inception, except for an interruption in the First World War.
The following figures, calculated in some cases from graphs presented
by Lush (1936) and in other cases from graphs and figures presented by
Clausen (1949), indicate some important changes which have taken place
in the performance of Danish Landrace pigs fed out at the testing stations:
___ 11909-1910 1924-1925 1947-1948
Av. daily gain, lb.. . . . ..... 1.18 1.32 1.46

Av. feed required per unit gain.. . ,1, 377 357 319
Av. age at 20 kg. live weight, days .... . 69 74
Av. age at 90 kg. live weight, days ......... " .. 1 186 180
Class 1 bacon sides. . . . . . . . . . . . . . . . .. .. . . . . .1 40% 90%
Detailed data (Clausen, 1949) indicate that progress in lowering feed

requirements per unit of gain has been slight during the past dozen years
or so. The figures show that increases in rate of gain while on test have
been accompanied by increases in age at 20 kg. live weight. This suggests
that selection for rapid gains on test may result in indirect selection
for lowered milking ability in sows and thus for lower gains in pigs during
the nursing period. Although rate of gain on test has increased by 11
per cent since 1924-1925, the decrease in age at 90 kg. live weight has
been only about 3 per cent. In spite of this possible antagonism, it
must be remembered, hawever, that the decrease of 6 days in age at 90
kg. live weight represents a real gain.
FIG. 196.-Graph showing trends in carcass length, back-fat thickness, and belly thickness
in carcasses of Danish Landrace hogs fed out and slaughtered in testing stations in the Danish
swine progeny-testing program. (Redrawn by T. W. Perry oj Purdue University from graph
presented by Clausen, 1949.)
The accompanying figure, adapted from Clausen (1949), shows that

progress has been made in increasing body length, increasing belly thick-
ness, and decreasing back-fat thickness. Progress during the past few
years has not been so rapid as earlier, however. Doubtless this fact is
an illustration of the principal that progress becomes more difficult in
populations already greatly improved by selection.
Regardless of whether progress is becoming slower or whether some
traits may be somewhat antagonistic to each other, real progress has
been made in improving the Danish Landrace hog while this program
has been in operation. The Yorkshire breed, maintained in smaller
numbers than the Landrace, has changed in a roughly parallel fashion.
Since the reports serve as only a guide to breeders, and since management
SELECTION I N M EAT A NIMALS 669
of the test litters has undoubtedly improved over the years (although
known changes in rations or management are slight), it is impossible to
determine how much of the improved performance can be attributed to
genetic improvement resulting from the program.
Several other European countries and Canada have progeny-t.esting
programs which are carried on under plans more or less similar to the
one in Denmark.
In the United States production testing on an organized basis has been
slower in ' developing, but several programs initiated during the past 20
years indicate that American breeders are coming to a realization of the
FIG. 197.- Ring Gold Lady Dora PR149, 753956, the first sow of any breed to produce eight
qualifying litters in a P .R. program. Shown here with her sixth qualifying litter, the 10
pigs of which sold at auction for a total of $4,365. (Courtesy of Hampshi re S wine R eoistry
Association, P eoria, Ill.)
need for such procedures. Active effort.s are being made to devise
programs which will be practical with the huge hog population and under
the management systems used in swine production here.
In i938 the United Duroc Record Association initiated a program of
"Production Registry" which is designed to give recognition to sows
having the ability to farrow and raise large litters and nurse them properly
to weaning. Recognition is also given to boars. Most of the other
breed associations developed somewhat similar programs during the
following years under rules which varied in some particulars. In 1946
all the associations united in adopting a single set of rules which (as
changed somewhat in 1948) are as follows:
I. Farrowing reports.
Farrowing reports on nominated litters must be witnessed by one disinterested
party of unquestionable standing in the community to verify size of litter and
earmarking of litter. This farrowing report must be sent to th e breed record
association office within 5 days after the farrowing date.
(jiG BREEDING AND IMPROVEMENT OF FARM ANIMALS
II. 56-day weighing reports.

All nominated litters are to be weighed between 51 and 61 days of age and
reported to the breed record association within 3 days after pigs are weighed.
Ali litter weights will then be calculated to a uniform 56-day basis.
This 56-day weighing must be witnessed by one official, other than the owner
of the litter, such as a county agent, 4-H Club agent, vocational agriculture
instructor or supervisor, coliege livestock extension specialist, cow tester, director,
or representative of the party designated and endorsed by, previous to the weigh-
ing, one of the aforenamed officials. It shall also be the duty of the official
witness to check the accuracy of the scales used for the "Weighing.
FIG. 198.-A P.R. Duroc sow, New Era's Type 185962. At the ti me of her death in 1946
she had raised 116 pigs ia 12 litters for a lifetime average of 973 per litter. As well as being
outstanding from a production standpoint, she was a noted show sow, being named reserve
All-American aged sow in 1943. (Courtesy oJ United Duroc Record Association, Peoria, Ill.)
III. Qualifying litters.

A Production Registry qualifying litter shall be as follows:
A. For sows over 15 months of age at farrowing:
To raise, without the aid of a nurse sow, 8 or more pigs to an official 56-day
litter weight of at least 320 lb.
B. For gilts (15 months of age or younger at farrowing time):
To raise, without the aid of a nurse sow, 8 or more pigs to a 56-day litter weight
of at least 275 lb.
IV. Qualifying sows.
Sows will be officially admitted in the respective ,breed association Production
Registry after having raised two qualifying litters. However, recognition
for the first qualifying litter of any sow call be indicated on pedigree certificates
where her name appears by affixing a star to her registry number. Production
Registry sows will be clas ified according to the number of qualifying litters
raised, such as 2-star, 3-star, etc.
\' . Qualifying boars.
A boar may qualify for official Production Registry as follows:
A.. When he has sirecl15 or more official PR qualifying litters. Recognition for
this hono\ will be designation as a PR Boar.
\
SELECTION IN MEAT ANIMALS G71
B. When he has sired five 2-Star PR daughters, or ten I-Star PR daughters or an

equivalent combination of both. Recognition for this honor will be designa-
tion as a I-Star PR Boar. (In computing an equivalent combination each
I-star daughter will count one point and each daughter with two or more
stars will count two points, the required total for any combination being
ten points.)
C. When a boar has qualified once on the record of his PR daughters he may
qualify for another star for each additional ten points on the basis of PR
daughters, points to be calculated as in (B) above.
Entry blanks can be secured from the respective breed associations.
Starting from small beginnings, the program has made steady progress,
although as yet including only a small fraction of the sows of any breed.
Following is a tabulation of the number of litters nominated and qualify-
ing in 1948 in two of the major breeds:
Ko. of litters No. of litters %

Breed
nominated qualifying qualifying
Duroc ...... .......... 2,476 747 30.2

Chester White ...... ....... 128 43 33.6
I
This program represents a constructive start on the difficult problem
of getting production records on purebred swine. Some may be inclined
to criticize it for not getting enough data, but in order to secure partici-
pation, it must be kept simple at least until experience dictates the
necessity of more extensive records. The value of weaning weights as
indicators of genetically influenced growing ability is not clear. It is
an established fact that the biggest pigs at weaning grow faster there-
after and reach market weights sooner'thanHght:er\~eaning pigs. It is
possible that the same genes affect the growth rates of pigs during the
pre- and postweaning periods. However, available evidence indicates
that weaning weight is very low in heritability-apparently being largely
determined by the maternal ability of the dam. Daily gain at any stage
of a pig's life is dependent upon size already attained, the heavier pig
making a bigger gain. Comstock et al. (1942) found that weaning weight
was correlated with rate of gain from weaning to 200 lb., but not with
rate of gain from 50 to 200 lb. Thus, the heavier pig at weaning appar-
ently reaches market weights sooner not because of inherently greater
growing ability, but because it (1) is in a more advanced phase of the
growth curve at weaning and (2) has already attained a greater per-
centage of the final weight.
From the standpoint of the practical hog producer, however, ~.ge.at(J)
rnarket weight is the important con_sideration. Since heavy weaning
pigs reach market weights in fewer days, selection for them may be
effective even if it is more directly selection for sow maternal ability than
for gaining ability as such.
It is well at this point to introduce a word of caution in order that
swine breeders may not be tempted to make the same mistake that our
dairy-cattle associations made with their AR testing. The mistake that
practically spoiled all the good that might have come from this system
was the fact that a man could test as many daughters of a bull as he
wanted to. Naturally, the best daughters were tested, the poor ones
not. This gave a very distorted picture of the breeding worth of hun-
dreds of bulls. Any bull can be made to look good if the breeders test
only a few of his best daughters. Likewise, any boar can be mad.e to
look good if they put only a few of his very best daughters into PR
testing. When this is done, they completely defeat the very purpose of
the system, which is to determine the genetic breeding worth of the
animals. Thus, all a sow's litters should be considered in evaluating her,
and the records of all a sire's daughters that farrow should be used in
evaluating him.
At least one breed association is developing a program of herd testing
in which weights are taken on all litters in the herd, not only at 56 days
of age but at six months. This is a most constructive step and overcomes
the danger of selective testing inherent in the PR program. Some
breeders are now following the commendable practice of weighing all
pigs and publishing the records made, be they good or bad, as a part of
the pedigree in sale catalogues. An example of an actual pedigree
printed in the catalogue of a recent purebred Hampshire sale is given in
Fig. 199.
Another word of caution should be introduced against undue forcing
of litters because of the advertising values sometimes attached to extreme
weight records. Records made under practical conditions will be of
much p;reater value in assessing the genetic worth of the animals. The
prima y purpose of "Production Registry" is to reduce the number of
small litters weighing less than 275 lb. and to increase the number in
the commendable range of 350 to 450 lb.
Another significant development during the past few years has been
the growth of sow-testing programs in various states. Probably the
most highly organized of these is the Wisconsin Swine Selection Coop-
erative. This program is designed to provide information which will
aid the breeder in making intelligent selections from his herd for breeding
purposes.
The breeder ear-notches each pig individually at birth and records
litter size. The cooperative furnishes forms for keeping records and
employs field m~n to aid the breeders taking weights of pigs at approxi-
SELEcrION IN MEAT ANIMALS 673
mately 154 days of age. The weight records are sent to the University
of Wisconsin where they are adjusted to 154 days of age. An index is
then calculated for each pig and the records returned to the breeder in
time for him to use the index in selecting breeding stock.
LOT No. 24 LITTER No. 62
CONNER VERBENA
GILT, Tag 2501, 56 Day Weight 30.6 Ibs., Dinner Plates 14
(Pedigree of Litter 62)
Farrowed Jan. 27, 1948 Average birth wgt. 3.1Ibs.
Pigs Farrowed 13 Average 56 day wgt. 30.49Ibs.
Pigs Raised 9 Bred by Conner Prairie
CONNERS COMMANDER 348641
4 litters avo 8.75 pigs born
4 litters avo 5.5 pigs raised
Average birth wgt. 2.39 Ibs.
Average 56 day wgt. 28.851bs.
CONNER TOP DECK 361355 ...
14 litters avo 9.43 pigs born CONNER ALBRIGHT 879126
14 litters avo 7.1 pigs raised 5 litters avo 10.6 pigs born
Average birth wgt. 3.19 Ibs. 5 litters avo 6.8 pigs raised
Average 56 day wgt. 32.69Ibs. Average birth wgt. 3.03 Ibs.
Average 56 day wgt. 33.88 Ibs.
SOUTHWIND ROCK & RYE 303037

27 litters avo 9.1 pigs born
27 litters avo 6.26 pigs raised
Average birth wgt. 3.09 Ibs.
Average 56 day wgt. 31.67Ibs.
CONNER SILVER STAR 896490
3 litters avo 10.66 pigs born MORNING STAR 785214
3 litters avo 9.33 pigs raised 7 litters avo 11.57 pigs born
Average birth wgt. 3.04Ibs. 7 litters avo 10.29 pigs raised
Average 56 day wgt. 30.9 Ibs. Average birth wgt. 31bs.
Average 56 day wgt. 26.15Ibs.
FIG. 199.-Pedigree of a gilt sold in a recent Hampshire auction in which production records
of all animals were given. (Courtesy of Conner Prairie Farm, Noblesville, Ind.)
As an example of the index, let us assume we have a pig-one of a

litter of eight, all of which were raised to five months-and that this
pig weighs 175 lb. at five months:
Points
1 point allowed for each pig farrowed (up to 12) . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2 points allowed for each pig raised to 5 months ........................... , 16
Total litter weight at 5 months divided by 100 (let us assume the litter weighed
1,300 lb.) ......................................................... ' 13
Ylo point for each pound over 75 lb. for the individual pig at 5 months. (Our
175-lb. pig at 5 months would be scored 100 X Ylo)..... .. .... .... .. ..... 70
Pig index. . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 107
674 BREEDING AND IMPROVElvIENT OF FARM ANIMALS
Herd No.._ __
.t=
C.
ro<Il E
bO
ro<Il c.0 .s0
0 0
<Il .t=
z 5l- '6 0..
"E '6 .B
'"
:I: '0 c::
2
0
<Il ,.,
li;
'"
u
0'" 0::'"'" ::J
III d:
~~...J
i':'
}t!o::
.., .t=
c::
'" zS c:.
>- 'E" ~ :8
0::
Cl
Z '" LLI
'"
E
Z
« .., ...., ~ -g .g~ E .2
c::
.., .., ..,a ,., ..,
Q) oj 0
Cl 3: ~ 3: oj oj u
>-Cf) g~ "0 oj
:ll "5
t:~"E -c
Q)
.f! .!!! <D a ~ 0

Cf) 0 <D
.S: E on on
c:: ~ c:: '"
.t=
.0 .0
0::--' <.>
~ 3: "
[:!
'"
2 .<'! '"
~~~ 0 0 3: l'! <Il
0 Z Z ~ a ~ a III 0..
::J
'"
0 Q: -c
:z_3:
:;)~c5l
'"E
Q)
-;;; 0::
W,,-
::lO.g
o "S
0::1-"5
:;)zc::
Q_w-
:::;:
I-
0::
«
Q_
W
Cl
~
Q)
c::
.iE., c::
0
:;::;
ii
-=..,c: '"
E2
~ 0
~-E"i
-sro'"
Q)
~ 'IJ, 0 0
III on
N
.c u
'"
<Il
N
'" '0 '0 '" tl
-:;; ~.~=~roC1l
iti
~ ;,:. ;:. ~ :?o ~ ~!S.§ ~ .& Q)
>O'~~.i(; 0
""
'6 ~
.0 E .0 0
u
If)
'"
0.. (ij a5 '"
0 a5
a
w
(.)
:::>
a
0
a:
0- f')
IJ)
a:
w
§
....0
0
a:
0
(.)
W N
a:
ci
c:
~
0 ,.,'" ~ ,.,'" ,.,'" ~
'" '" '0<0'" 0
'0
"E
u
0
.s
~
~.~~
~c:
.E.2
tno. "0.,
~
~
.<::
-ro '"'"- - -'"
'" "0 '0
"0
N
<0
I{) If)
~
""
c:
.",
~
r::
ci
Z ·i ~§ g t
:c a; a;'" '"
a; ~ .,'"
~
.g o "'u ~
.Q .Q .Q
~>- z:::J z:::J z:::J

E E E 0;
~
~ ~~
u ~ ai
ci -0
:::J
'0
::; Ui z >- «> .,in
0
li:
oz
Programs of the type described above are very helpful in expediting

the necessary book work in a production-testing program. Then, too,
they have an official standing which increases their value somewhat over
and above the value of private records.
The keeping of personal records is to be urged, however, if suitable
organized programs are not available. These records need not be
elaborate, but should include pedigree records on each litter, records of
the number of pigs farrowed and weaned, and weights of the individual
pigs at weaning (usually 56 days of age) and at some later time, preferably
TABLE 53.-FACTORS FOR CONVERTING PIG WEIGHTS TO STANDARD AGES OF 56 AND
154 DAYS
Factors for converting individual Factors for converting individual

pig weights to 56 days* pig weights to 154 days t
Age Factor Age Factor
47 1.2812 145 1.108

48 1.2424 146 1.095
49 1.2059 147 1.082
50 1.1714 148 1.070
51 1.1389 1411 1058
52 1.1081 150 1.046
53 1.0789 151 1.034
54 1.0513 152 1.023
·55 1.0250 153 1.011
56 1.0000 154 1000
57 0.9762 155 0.989
58 0.9535 156 0.978
59 0.9318 157 0.968
60 0.9111 158 0.957
61 0.8913 159 0.947
62 0.8723 160 0.937
63 0.8542 161 0.927
64 0.8369 162 0.918
65 0.8200 163 0.908
* Ca.lculated from the formula
where Y = corrected weight

Z = actual weight
X = age in days
t Calculated from the formula
142.5 ]
W = Z [ O.OO32143X2 + O.58X - 23
where ll'" = corrected weight
Z = actual ",eight
X = age in days
about five months of age. Factors have been worked out by Whatley
and Quaife (1937) and Lush and Kincaid (1943) for adjusting individual
pig weights to 56 or 154 days of age, respectively. These make it
possible to weigh several litters at once and convert all records to a
standard age for use in comparing the records of various individuals.
Assuming production records are being kept, the breeder is faced with
the problem of how to evaluate them in selecting breeding stock for
retention in the herd. How much emphasis should be put on weight
ruRDUE UNIVERSITY
Animal Husbandry Department
Litter Record, Farrowing to Weaning
Sow Number: Litter Number' _ _ _ _ _ _ __
S,w's Sire Sire of Litter.._ _ _ _ _ _ __
&lw's Dam Breeding D&tc _ _ _ _ _ _ __
Date of birth, sow Farrowing Date _ _ _ _ _ _ __
Litter (1st, 2n:l., etc.) Date Vaccinated _ _ _ _ _ __

fOW'S weight: at b r e e d i n g _ _ _ Date weaned _ _ _ _ _ _ __
at 109- 112 day
at weaning
Sow's general appearance Li t tel" s general appearance.

at farrowing: Good, MediUlTl, Poor at farrowing: Good, M"dil.Ull, Poor
No. good teQ.ts: R_ _ L _ __
ot weaning: Good, 11edium, Poor at weaning: Good, Medium, POOl'
No. functioning teats: R _ L _ _
Birth Weaning (56 days Datal

Pie: Sex Weight .Act. wt. at alCul. 5G No. Remarks, defects,
ro. (lbs.) days days wt. Nipples abnonnalities, disposal
_ _ l _ , L - - L_ _ _ _~~~_ _J __ _ _ _ _ _ _ __ _ _
FIG. 202.-An example of a record sheet used to keep records on a Jitter of pigs to weaning
age. (Courtesy of Dr. J. A. Hoefer, Purdue Uni~er8ity, Lafayette, Ind.)
for age, how much on litter size, how much on litter weaning weight,
how much on body type, etc.? Noone answer can be given to this
question. Hazel (1943) in an analysis of the factors involved in con-
structing selection indexes, pointed out that the emphasis which should
be placed upon the different characters depends upon (1) the relative
economic importance of the various traits, (2) the variation in each trait,
(3) the heritability of each trait, and (4) the phenotypic and genetic
correlations between each pair of traits. Obviously, these constants may
vary from herd to herd and even in the same herd over a period of time.
Using data from the Iowa Station swine herd of inbred Poland Chinas,
G78 BREEDING AND IMPROVEMENT OF FARM ANIMALS
Hazel employed a multiple correlation method to construct selection

indexes. Using information on the pig's own I80-day weight (W), the
pig's own conformation score at market weight (8), and information
on litter size and weaning weight expressed as "productivity of the
dam" (P), he arrived at the following index:
I = 0.136W - 0.2328 + 0.164P
An apparently slightly more efficient index including terms for the average
,veight and score of the pigs in the litter was constructed. Since the
coefficients of these terms vary with the number of pigs per litter, this
index does not lend itself well to presentation as an illustration.
An interesting feature of Hazel's index is the negative attention
assigned to score. This is due to the fact that weight and score are phe-
notypically correlated with the pigs biggest for their age also tending to
receive the highest scores. Thus, selection for heavy pigs automatically\
selects for better type, and the type score can be discounted. ..-....
It is not knovv1J. whether the values given by Hazel would apply to
other herds and breeds of swine. From his analysis and from other
information, it appears probable, however, that as long as present
marketing practices prevail in the United States considerably more
attention should be paid to the items measuring sow productivity and
weight for age than to conformation.
Under usual conditions, the performance of an animal itself is the
best sin!ll~-indication of its genotype. Thus, for characters which can
be measured in the individual, selection based on individuality should
be practiced at least to the extent that no individuals of below-average
merit should be kept, regardless of the performance of relatives. PedigreC\
and family selection are a useful supplement to individual selection, and I
in selecting young animals for such things as prolificacy and nursing ,
ability, this type of evaluation is the only type possible. The general
plan of retaining the best individuals from the best families appears to be
sound in hogs.
~(1947) in a searching analysis of the relative value of individual
and family selection pointed out that if either method is absolutely
a
ineffective the other will be also. He found that combination of individual
and family selection is at least equal to either type of selection alone and
is distinctly superior when (1) genetic relationship within family is
moderate and the phenotypic correlation within family is much smaller
but well above zero, and (2) when the phenotypic correlation is distinctly
larger than the genetic relationship within family. The latter case i::;
interesting s~nce for this to occur family averages must depend more upon
environmental conditions peculiar to each family than upon genetic
SELECTION IN MEAT ANIMALS 6'i9
similarity. Under these conditions family averages playa negative part

in combination selection. Family selection is most necessary for those
traits in which members of the family resemble each other least, i.e., for
traits which are low in heritability.
FIG. 203.-Modern sow types. Chester White sow (above) upreme Fancy 2nd, grand
champion 1947 TIlinois State Fair. Owned and shown by Harry Brett, Decatur, Ill.; and
Duroc sow (below) Tiptop Princess 3rd, grand champion 1948 Nebraska State Fair.
(Courtesy of Chester White Swine Record Association and Duroc Record Association.)
Due to the approximate doubling of the generation interval, progeny

testing (considered here in the strict sense as being the plan followed in
which young sires are progeny-tested on outside females before being
used in the herd) appears to be of limited usefulness with swine (see
Dickerson and Hazel, 1944b). Progeny performance should, however,
be a major consideration in determining which boars to keep for use in
a herd a second or third breeding season. Sows to be retained should be

selected on the basis of the performance of a first litter. The number of
pigs farrowed and weaned and their growth to weaning are largely
maternal characters. Thus, selection for them is really individual selec-
tion superimposed upon the family selection by which the sows were
first selected as gilts. After weaning, the performance of a pig depends
to an increasing extent upon its own genes. Thus, selection of sows upon
the basis of postweaning performance of a first litter is partially progeny-
test selection.
Dickerson and Hazel (1944a) have arrived at the decision that maxi':)
mum progress can be made in a closed herd or a breed if approximately',
one-third to one-half of the sows are kept for a second litter and a very few (
of those which performed best in their first two litters are retained for J\
a third and fourth.
A Regional Swine Breeding Laboratory was established with head-
quarters at Ames, Iowa, in 1937, under the authority of the Bankhead-
Jones Act of 1935. According to the director, Dr. W. A. Craft,l
The principal objective of the laboratory is to discover, develop, and test
procedures of breeding that will result in improvement of swine. The followi~g
four phases of study are involved: Systems of breeding, measures or standards of
appraising hogs and their carcasses, heritability of various characters, and
physiology of reproduction. Emphasis is being placed on productiveness of
sows, growth rate of pigs, economy of gains, vitality, and desirability of carcasses.
At the present time the state experiment stations in the Middle West-
ern region have active programs in this project, and several other states
are carrying on work of a similar nature and cooperating informally
with the laboratory. The general procedures used 2 are as follows:
A promising supplement to conventional breeding practices that is being
studied at the laboratory is the development of inbred lines combined with selec-
tion for productiveness of individual animals. This permits selection from
different lines of breeding, use of inbred boars for top crossing onnoninbred stock,
and hybridizing inbred lines within and between breeds. Inbreeding to some
degree is necessary to form distinctly different lines of stock, especially within
breeds; it expedites the forming of lines that differ widely in their heredity.
Production recording is being practiced as it helps in locating not only the superior
lines but also the individuals and litters within lines that have the best heredity.
Systems of breeding including various degl'l:~es of inbreeding of 3 to 5
per cent per generation (four sires in a herd closed to outside blood), of
6 to 7 per cent per generation (two sires in a herd closed to outside blood),
1 U.S. Dept. Agr. Misc. Pub. 523, July, 1943.
2 Ibid. ,\
SELEC'l'ION IN MEAT ANIMALS 681
of 11 to 12 per cent per generation (one sire in a herd closed to outside

blood), and of somewhat greater intensity in a few lines carried for a
few generations by full sib matings are being studied. Inbreeding is
being carried on both within established breeds (33 of the present inbred
lines) and within new strains produced by crossing two existing breeds
(4 of the present inbred lines).
Results to date indicate that, on the average, some depression in
productivity occurs with increases in inbreeding, in spite of continuous
selection for productivity. This is more serious in some lines than in
others. Dickerson et al. (1947) using data from four Regional Swine
Breeding Laboratory herds showed average decreases per 10 per cent
increase in inbreeding of 0.2, 0.4, and 0.5 pigs per litter at birth, 21 days,
and 56 days of age, respectively. Average 56-day weight per pig was
not affected, but there was a 3H-Ib. per pig decrease in 154-day weight
for each 10 per cent increase in inbreeding.
The past few years most stations involved in this \york have been
testing inbred lines in crosses with each other and in top crosses with
noninbred stock. A few of these results were quoted in a previous
chapter. A recent statement by Dire~tor W. A. Craft l summarizes these
results:
Crossing of three or more lines restored the losses in all traits; in some cases
crossing of three or more lines produced increases in most of the traits studied as
compared to the performance of the parent lines before inbreeding or as compared
with noninbred stock used in some trials as controls. Some, but not all, inbred
boars have sired litters which exceed, in weight and numbers of pigs, litters by
noninbred boars at 5 or 6 months of age; some, but again not all, boars from
crosses of two inbred lines have sired litters which beat litters by other noninbred
boars at 5 or 6 months of age; crossline sows have proved to be good pig raisers;
crosses of inbred lines of different breeds have shown more hybrid vigor than line
crosses within a breed; genetic divergence of stock of different breeds seems to
provide gene patterns differing enough to produce more hybrid vigor than lines
related to each other such as is the case when they are of the same breed. Study
of data completed recently for 70 litters of line crosses within the Poland China
breed compared with 72 litters of crosses of lines between breeds shows a differ-
ence of 0.66 pigs at birth, 4.2 pounds per pig at weaning and 12 pounds per pig
at 5 months in favor of the lines of different breeds. Hybrid vigor appears to
be as real in pigs as in corn, but there are economic problems involved in obtaining
it which are more difficult to overcome than is the case in corn.
The lines produced by mild, medium, and intense inbreeding appear to
differ little in average merit as inbred lines or in their usefulness in
crosses. Since rapid inbreeding results in a considerable saving in time,
1 Quoted from a speech given in April, 1948, at the University of Illinois Swine Day.
it now appears that the most feasible method of producing inbred lines
is to start many small inbred lines and inbreed intensively at first. The
least promising lines can then be discarded and the remaining ones tested
in crosses.
Although several of the inbred lines produced thus far in this study
appear to have real value and may in themselves exert a beneficial
influence on American swine production, the work of the laboratory
must be considered primarily a study of methods. As promising methods
are developed, breeders are adopting them and can be expected to
develop better lines than the relatively few developed thus far by the
experiment stations. If the methods used continue to show promise,
this program can be expected to exert a considerable influence over
future swine-breeding practices.
Selection in Sheep.-In colonial America, sheep were grown almost
entirely for the purpose of producing wool. This led shortly to the
creation, from Spanish ':\lerino importations, of the American Merino
breed, which became very numerous along the Eastern seaboard, with
Vermont at that time the leading state in number of sheep, owing in
large measure to the influence of William Jarvis, who had a farm at
Weathersfield, Vt., and at the time was serving as American consul in
Lisbon, Portugal. The wool clip in the early colonies was only 2 or 3 lb.
per animal per year, but through selection and improved feeding and
management it has been continuously raised to an average level of 7 or
8 lb. for the present sheep population (all types) of the United States.
From the Vermont .M.erino, there were later developed the various
Delaine Merinos for the purpose of producing a fine wool of sufficient
length for combing. Most of the numerous Merino breed associations
that had sprung up were merged in 1906 into the American and Delaine
Merino Record Association, with three other smaller associations still
existing. Merino type varies from a small thin animal of poor carcass
quality but heavily wrinkled and producing a fine, dense, heavy fleece of
short wool (Type A), to a larger, fairly heavy, and low-set animal pfo-
ducing a good carcass but generally lacking wrinkles and with a heavier
but less fine and dense fleece (Type C or Delaine), and a third inter-
mediate form (Type B).
From Spanish :Merino foundations, the French developed a fine-wool
strain of sheep, the Rambouillet, of much greater size than the American
Merino. Many of these animals were brought to America about the
middle of the nineteenth century, and for a time they were a very popular
breed. Interest then declined, to be revived about 1890. The Ram-
l~ouillet is a large animal producing a heavy, dense fleece. As with the
American Merino, there are variations in amount of skin wrink,les and
SELEC'l'lON IN MEAT ANIMAL/:> 683
at some shows the animals are divided into Band C types. They were
brought to America to serve primarily as wool producers. In recent
years, greater attention has been paid to fleshing qualities, more by
some breeders than others, but the carcasses are not of so high a grade
generally as those from the Down breeds. This breed has been called
the most typical dual-pmpose breed in America, yielding something in
i.
FIG. 204.-Different types in lamb carcasses. (CottTtesy of Bureau of Animal I ndustry,
U.S. Department of Aoric~lltttre,.)
density and fineness of fleece to the American Merino and to the Down
breeds in quality of carcass. ! Under these conditions, we would expect
to find wide degrees of variation in the breed.
Robert Bakewell, the noted English livestock breeder of the eighteenth
century, greatly improved one of the long-wool breeds, viz., the Leicester.
He devoted his major attention to fleshing or fattening qualities and
.quite neglected consideration of wool. The result was a fairly large
breed of sheep with rather long and coarse wool and a carcass of a coarse
grain, which when finished was overlaid with several inches of fat. To a
684 BREEDING AND IMPROVEMENT OF FARM ANIJULS
greater or less extent, these qualities are found also in the other long-wool
breeds, Cotswolds, Lincolns, and Romneys. The long-wool breeds were
created, therefore, primarily to serve as meat animals supplying the large,
fat legs and roasts then in demand. In recent years, more attention has
been paid to wool, and attempts have continually been made to improve
the quality of carcass in these breeds. The fine-wool breeders generally)
pay particular attention to fleece, less to carcass qualities. The long- It
wool breeders, on the other hand, pay more attention to carcass, less to )
fleece. '
The various "Down breeds" of sheep, Southdown, Shropshires,
Oxfords, Hampshires, etc., have been developed to fill the void between
the fine-wool and long-wool breeds. They are, in a sense, a happy
medium between two extremes with the very great added advantage of
producing very superior carcasses from the standpoint of quality. The
various Merinos and the Down breeds have been bred and kept relatively
pure in the East and Midwest. In the range country farther west, the
foundation females were largely grade Rambouillet ewes. On these ewes,
there have been used all the types of American and Delaine Merinos and
Rambouillet rams. Later, rams from the long-wool breeds were used,
particularly Cotswolds and Lincolns, in the attempt to improve size and
fleshing, and rams from the Down breeds were used to try to improve
carcass quality still further. Finally the use of long-wool rams on
fine-wool e\ves in the attempt to get good sheep from the standpoint of
both wool and mutton led to the formation of the Corriedale breed in New
Zealand (imported to the United States in 1914) and the Columbia,
Panama, Targhee, and Romeldale breeds in the United States.
As can be seen from the above discussion, there are a great many types
of sheep in existence. Sheep are raised in the United States under a
greater range of environmental conditions and management systems than
is true of any other class of farm animal. The first job of the sheep\
breeder is, therefore, to select the type and breed of sheep best adapted 7
to the conditions under which operations will be carried on and where\
breeding stock will be sold. It may well be that more breeds are needed)
to fill needs in certain areas.
The second task of the breeder is to improve the productivity of the
type of sheep selected. Although in former years many sheep were kept
almost exclusively for wool production with aged wethers often being
run for this purpose, the sheep industry is today almost universally on a
dual-purpose basis, i.e., both wool and meat animals (usually lambs) Ile
major products, with the relative importance of the two varying c .
siderably in various areas. For example, in the semiarid regions of the
Southwest, w~ere ranges are so sparse as to make the production of
\
\
1'\
milk-fat lambs almost impossible, lambs usually sell as feeders at lighter

weights and lower prices than those at which milk-fat lambs sell in areas
where their production is possible. High-quality fine wool can be pro-
duced under these conditions. Thus, wool is relatively a much more
important product than in better range areas or under farming conditions
such as prevail in the Midwest.
Regardless of the relative emphasis given to wool and lamb production,
the general objectives of sheep breeders are very similar, namely, to pro-
duce as many pounds of lamb as possible per 100 lb. of ewe and as much
high-quality wool as possible. They must pay attention to prolificacy, .
ease of lambing, milking ability, and natural fleshing. As with dual-
purpose cattle, some of these qualities are more or less antithetical. If he
gets too great propensities for wool production and milking ability,
natural fleshing is apt to suffer, and vice versa. In addition he must
pay attention to general vigor and longevity, since susceptibility to
diseases, parasites, or too rapid turnover of the flock all reduce imme-
diate profit and lessen the effectiveness of selection.
As if all these things were not enough to worry about in a breeding
program, the sheep industry is plagued with a greater diversity of breed
trade-marks than is the case with other classes of farm animals. As
pointed out earlier, some of these, such as excessive face covering, are
actually detrimental to productivity. The modern breeder might be
well advised to place emphasis on production records rather than winning
show-ring type.
The largest and most highly organized project devoted to sheep-
breeding research in the United States is the Western Sheep Breeding
• Laboratory established at Dubois, Idaho, in 1937. It has operated since
that time under the guidance of Director J. E. Nordby. It is utilizing
the facilities developed by the Bureau of Animal Industry Sheep Experi-
ment Station, which was organized at Dubois in 1917, and in addition the
experiment stations of the 11 Western states and Texas are cooperating.!
The main objective of the program is the improvement of sheep through breed-
ing. The improvement is measured by the application of utility standards
as they concern commercial production under ranching conditions. Production
is measured on the basis of wool yield, and in pounds of lamb of the desirable type
and finish. Obviously, the staple length and the quality and uniformity of fibers,
as well as the shrinkage, are important factors that are studied with the yield of
wool. All useful type characteristics that influence lamb production, ranch
adaptability, and general serviceability are closely integrated in the improvement
program. Under ranching conditions at relatively high altitudes, where winter
1 From mimeographed statement on the Western Sheep Breeding Laboratory and
the U.S. Sheep Experiment Station by Julius E. Nordby.

•
686 BREEDING AND IMPROVEMENT OF' PARM ANIMALS
storms are frequently severe, the objectives in sheep improvement must also
include a consideration of those characteristics in a fleece that protect the sheep
against weather extremes. The fleece must be relatively long and "lofty" as
well as dense and must hold together along the topline and cover well the
underline.
Attention in selection is being given to both meat and wool production

but with meat qualities 1 emphasized somewhat more than those relating
to wool.
Besides we are interested in the mature form of mutton sheep on the basis of
its value for producing the right kind of market and feeder lambs, and replace-
ment ewes. While obviously the mature size, form, and fleshing quality of a
sheep are important, the ranchman is perhaps fully as concerned with weight,
form, and fleshing qualities at the time of weaning the lamb, because it is at this
time that weight and form, but principally fleshing will determine if the lamb is a
market lamb or a feeder lamb. The milking quality in ewes is observed just
after lambing, and this is further checked in their ability to produce vigorous and
fleshy lambs. In the selection program, therefore, much emphasis is placed
upon the finish in lambs at the time of weaning. Due consideration is given to
twins in this connection. Stud ram prospects are scored carefully for market
qualities at weaning, because a ram lamb that has the acceptable growth, rugged-
ness, form and finish at weaning time is probably more apt to be a potential sire
of lambs of the same kind, than is a more slowly maturing ram that does not
develop these qualities before he is more mature. Vigor is an essential quality
in the selection of ranch sheep. Some ewes, and rams also, "wear" uniformly
well under heavy service until they are seven or even eight years of age. Others
begin to decline early in life and are on the" cull" list at five or before.
Selection is being practiced against too heavy face covering that might
\ lead to "wool blindness" and also against heavy skin folds.
<
Work is being done with the Rambouillet, Columbia, and Targhee

breeds, with a few other breeds being used in certain studies and in test
crosses. Inbreeding is being done with about 30 Rambouillet lines and
8 to 10 lines of each of the other two breeds being in process of formation.
N oninbred control stocks in which the same type of selection is practiced
as in the inbreds are also being carried to see whether the greatest progress
can be made with or without inbreeding.
Many publications on factors associated with productivity in sheep,
methods of evaluating breeding animals, heritability of various traits,
and other things about which information is needed in properly carrying
out a research program in sheep breeding have already come from the
laboratory. Many of these will be referred to later in this chapter.
1 Ibid.
•

•
II
~
,
~ ~
1~
"
~
{
o
g,
II
.,ll ) I1111 LLllll ill
~ ~~ [[1-1111 IIIII "
fi~~I~~ ~~- \ \ I I I I \ I I i I I I \ ~
U H+H-I I I I I I I I I f' !
~! l~ )))))1-\1-1111-11 ~
il ITITrrm-11nI j n
&1miTITTlI-II-)) ~,H
-frlfITfll-iII ITIT ml
11111 i II II -[11 ~l!~
-H-f--:--;--:-t--t-t-t-t~1 -:- - - >.
i; ml mml rl-I m~
~ mn--III-III~,Iii ~~!~~
•
SELECTION IN MEA'l' ANIMALS 689
Definite progress has been made in improving sheep, especially for

some of the more highly heritable characters. Because of the relatively
slow rate of reproduction in sheep, inbreeding advances more slowly than
is the case with hogs.
The keeping of production records on sheep and their intelligent
evaluation in making selections necessarily entail a considerable expendi-
ture of time and effort. Due to the variety of types of sheep it is difficult
to suggest procedures for taking and evaluating records which will be
universally applicable. However, the follmving items have been included
in one form or another in most record-of-performance precedures sug-
t
gested to date:
1. Prolificacy of ewes (regularity of lambing and twin production).
In s-;;me areas where grazing conditions are poor, twin lambs may be a
disadvantage, but generally twin-producing ewes are more profitable.
2. Birth weights of lambs. This item is of no value in itself but is
related to vigor at birth and to rate of gain after birth.
3. Weaning weights of lambs. Taken individually but also used to
measure total lamb production of a ewe. This latter may be best
expressed as pounds of lamb weaned per ewe (in areas where costs are
largely on a per head basis) or as pounds of lamb weaned per 100 lb. of
ewe (best in areas where costs are more related to actual amounts of
feed eaten than to total numbers).
4. Type and finish of lambs at weaning. These important factors reflect
market value to a large extent. Attempts to evaluate these items
objectively by means of measurements have as yet met with little success.
Although unsatisfactory at best, the most successful means of evaluating
these characters has been through the use of scores based on subjective
appraisals of merit.
1
5. Fleece weight and quality.
6. Regular annual or semiannual weights of all breeding stock.
The evaluation of performance records is greatly complicated by
environmental factors. Some of these, such as the effects of sex, age of
dam, type of birth (single or multiple), and age at weaning, can be
measured and definitely taken into account.
If these things are not systematically evaluated, selection may be
biased in favor of lambs having the most favorable conditions for growth.
A study of the lambs born during a 14-year period (1921-1934) in the
Hampshire, Shropshire, and Southdown flocks at the National Agricul-
tural Research Center at Beltsville, Md., showed that their environ-
mental factors had apparently influenced selection to an undue degree
when selection of lambs began at about t\lree months of age. In summa-
fi90 BREEDING AND IMPROVEMENT OF FARM ANIMALS
rizing this studyl tho authors recommend some practices which should
minimize this tendency.
The results showed that type of birth, time of birth, and birth weight have
influenced the selection of breeding animals. Single lambs have been preferred
to twins, early lambs to late ones, and lambs that were heavy at birth to light
lambs. With the exception of a portion of the effect of birth weight, the effects
of these factors on selection are considered to be environmental in nature, and
therefore, reduce the chances of selecting the genetically superior animals for
breeding purposes. This weakness in the present method of selection can be
at least partly overcome in one of the following ways:
1. When selections are made at weaning or a similar early age, the animals
may be divided, first by sex and then into single and twin groups within sexes.
These last two groups should then be divided according to ti!I1e of birth. If the
range in date of birth is 6 weeks) the lambs may be divided into two groups
having a range of 3 weeks each. If lambing occurs over a longer period, more
groups may be necessary. With the animals so divided, those in each group will
be on a more comparable basis than if considered as a whole, and selections can
be made within each group. This division does not take into consideration
differences in birth weight, and if data on this point are available they may be
kept at hand for use while making selections. Selections would, of course, be
based on the characters in which the breeder is primarily interested, such as
mutton form, fleece, and perhaps other factors such as pedigrees of the animals.
This method eliminates comparison of single lambs with twin lambs and of
early lambs with late ones. Assuming that the proportion of individuals possess-
ing superior germ plasm is equal in the groups of singles and twins and that
there is a similar proportion in the early and late group in each type of birth,
the chances of actually selecting these superior animals would be greater if this
suggested system was followed.
2. Selection may be made at a standard age, such as 6 months, rather than on a
certain date. Weights at this age could be adjusted for the effects of birth fac-
tors, and any advantage or disadvantage kept in mind when evaluating each
animal. The weights of all animals could be adjusted to a standard basis, such
as single males of a standard birth date and weight, for use in making such allow-
ances for the effects of birth factors and sex. All animals reaching the standard
age within a given ''leek might be observed on one day during that week. By this
method each individual is compared with an ideal at a standard age, with allow-
ance for any unfavorable environment due to birth factors. The effect on selec-
tion would be similar to that of the first suggested method.
3. Selections might be postponed until an age when the effects of the birth
factors on development discussed herein have disappeared. Little effect of type
and time of birth and birth weight on development, as measured by weight,
remains at 12 months. Retention of all animals to this age is not practical for
the sheep breeder who must sell his discarded lambs when they reach market
1 PRILl,lPg, R. W., and DAWSON, W. M., Some Factors Affecting Survival, Growth,
and Select:ion of Lambs, U.S. Dept. Agr. Cir. 538, 1\.)40.

\
SELECTION IN MEAT ANl11;JALS (j91
weight, but it may be a worthwhile procedure in experimental breeding, unlesfl

data taken at younger ages can be shown to be sufficiently representative of the
animals' potential development.
Another application of the results of the study presented herein is in the com-
paring of groups of animals, such as the offspring of two or more rams in progeny
testing. If the number of offspring from each of two rams is sufficiently large
that practically equal numbers of singles and twins, lambs of heavy and light
birth weights, and males and females are available, and if the time of birth of
the lambs is comparable, then a comparison of the uncorrected weights of the two
groups at a standard age would be valid. However, most young rams are tested
on a small number of ewes. If the resulting small number of offspring of a ram
are to be compared with those of another ram at an age when the effects of birth
factors still persist, the figures given in this circular, or similar figures, can be
used in making necessary adjustments. Adjustments for sex can be made in
like manner. All animals might be adjusted to a basis of male singles with date
of birth and birth weight constant.
Recent extflnsive studies at the Western Sheep Breeding Laboratory

(Hazel and Terrill, 1945a, 1946b, and 1946d; Terrill, Sidwell, and Hazel,
1947, 1948a, and 1948b) have shown that under range conditions the
effects of sex, type of birth and rearing, and age of dam have important
TABLE 54.-DIFFEREXCES IN PERFORMANCE OF RANGE SHEEP DUE TO SOME ENVIRON-

• ~IE:-.rTAL CONDITIOl\S
Weaning characters Yearling ewe characters
Clean
Body Staple Body Staple
fteeee
weight, length, weight,
weight, I length,
lb. crn. lb. em.
lb.
-~~---- _----_
Rambouillpts
Advantage of:
Rams over ewes ..... ...... . 8.3* -0.48*
Mature dam over 2-year-old
dam ..................... 6.1 * 0.19 2.6 0.17 0.12
Singles over twins ...........
Columbias, Targhees, and
9.2* 0.05 I 6.0* 0.23* 0.30*
Corriedales
Advantage of: I
Rams over ewes ............ 10.8* -0.54*
1
Mature dam over 2-year-old I

dam ..................... / 8.7* 0 52* 4.16* 1.04t 0.46
Singles over twins ........... 1 11.7*
I 0.02 7.12* l.03t 0.12
I \
* Signifies statistic is large enough to account for 2 % or more of the total variation in that trait.
t Grease fleer.e weight.
692 BREEDING AND IMPROVEMEN1.' OF FARM ANIMALS
effects on body weight at both weaning and yearling ages, on fleece

characters, and on type and condition as evaluated by subjective scores.
Age of lamb at weaning also had a significant effect upon weaning weight.
The environmental factors were relatively unimportant in affecting;
scores for face covering or neck folds.
Table 54, prepared from data in the papers just mentioned, shows the
magnitude of some of the differences observed. In general, the differ-
ences tend to become smaller with advancing age. The reader should
recognize that these figures will not necessarily be applicable to other
flocks under other environmental conditions.
In a large operation, the necessary data to arrive at figures such as the
above can be collected and the records all corrected to a single standard,
preferably by adding the appropriate number of pounds to the weights
of lambs in the lower groups to adjust them to the level of the heavier
group.
In actual practice similar results may be obtained by sorting lambs
into groups according to sex, age of dam, type of birth, and age of lamb
and selecting the same proportion from the various groups.
Heritability estimates of a number of traits in sheep as given in Table
55 indicate that fleece characters, face covering, and neck folds are, in
general, more heritable than body weights, and that subjective scores of
body type and condition are considerably less heritable.
In a breeding program the relative emphasis to be placed upon the
various characters depends upon (1) the economic importance of the'\
character, (2) its heritability, and (3) the level of performance alre!J,dy )
attained in the flock. The latter two factors are interrelated since>
average performance near either end of a scale of desirability will be
indicative of an approach to genetic fixation with a resultant low herit-
ability. Hazel and Terrill (1946c) point out a situation which illustrates
this fact. Jones et al. (1944) found a heritability of 51 per cent for neck
folds in a moderately wrinkled Rambouillet flock, Terrill and Hazel
(1946c) obtained an estimate of 38 per cent in a less wrinkled Rambouillet
flock, while in Columbia and Targhees with only a slight degree of
wrinkling the apparent heritability was only 8 per cent.
In a breeding program intense selection should first be practiced for \
traits which are economically important and highly heritable. Progress
should be rapid in this phase of the breeding program; and after desirable)
levels of performance have been attained for these characters, attention
can be turned to other less heritable traits y,here progress can be expected
to be slower, but nevertheless important once it is attained.
r
As with other classes of livestock, observation indicates that selection
has been effeptive over the years in modifying type and productivity in 1
\
*00 *00 *00 *00'1J

*
'"0 '"0 '1j '"0'"0
o 00 00
..c:: ..c::..c::
+' +'+'
..c::..c::
+'+'
Cl) Q) Cl) Cl)Cl)
S SS SS
C'l C'l C'l C'lC'l
...............
o 00
.....0.....
0
~ ~~ ~~
694 BREEDING AND I M PROVE MEN7' OF FARM ANI MAL S
sheep. Numerous examples, including the work of Neale (1946) in

selecting for staple length in Rambouillets and the work of Warwick
et al. (1949) in selecting for resistance to stomach worms, can be cited to
show the possibilities.
Terrill (unpublished data) in a study of data from the Western Sheep
Breeding Laboratory has compared the amount of progress actually
made for several traits in Rambouillet sheep during the period 1940-1948
'with the genetic progress expected. The latter is calculated by com-
paring the averages of the animals which later became parents born in
FIG. 207.- A model hropshire ram, bred by George McKerrow and Sons Compan.v ,
Pewaukee. Wis.
each year (suitably weighted for number of offspring produced) with

the average of all sheep born in that year. These selection differentials
were divided by the average age of parents and multiplied by the herit-
ability of the trait being studied in order to give an estimate of expected
annual genetic progress.
In evaluating his results, it must be kept in mind that trends in environ-
mental conditions could have been responsible for some of the observed
trends in productivity. Year~ing fleece weights, weaning body weights,
and weaning body scores for condition all trended slightly downward in
average merit, in spite of positive selection for improvement. Improve-
ment was shown in staple length, type scores, face-covering scores, and
neck-fold s,cores. An index based on the above items, and thought to
\
I.
•
reflect all-around merit, is used as the basis for selecting breeding animals
in this flock. The average index has increased during the period of this
study almost as much as would be expected, and Terrill concludes that
the average merit of the sheep in the flock has improved.
Individual performance and family averages in production are the
basi~it£ms--upon which selection in sheep must be based. This follows
frcimtneTacts that Crr
although some culling of breeding animals may be
practiced on the basis of the first lamb crop or two, they must initially
be selected for use in a flock upon the basis of their own and relatives'
performance, and (2) even if a progeny-testing program is followed with
prospective stud rams before they are used in the main flock, only a
small percentage of the most promising rams can be progeny-tested,
these must be selected on the basis of individual and family performance.
An organized program in which prospective stud rams are progeny-
tested in an auxiliary flock prior to being used in the main flock can be
useful under certain conditions. Dickerson and Hazel (1944) concluded
that a program of progeny-testing yearling rams and using the best ones
in the main flock as two-year-olds or three-year-oIds, should increase
progress by about 20 per cent and 5 per cent for weanling and yearling
traits, respectively, with characters having heritabilities as low as 10 per
cent. For characters with heritabilities of 30 per cent, progeny testing
would be expected to increase progress slightly for traits which could
be measured at weaning age, but would be expected to decrease progress
for those traits which could not be measured until yearling age. This
latter situation occurs because of the additional year which it takes to
evaluate yearling characters of the progeny.
Progeny testing can be expected to be still more effective if rams can be
progeny-tested as lambs instead of yearlings and the best performing
ones used in the main flock as yearlings.
Selection in Beef Cattle.-Selection in beef cattle has long been based
primarily upon show-ring winnings or type evaluations based upon show-
TABLE 56.-PERFORMANCE RECORD OF FOUR CHOICE BEE~' SHORTHORN STEERS*
Feed to produce 400 lb. of gain

Steer Days on Daily gain,
number feed pounds
Grain, pounds I Hay, pounds
--_
I 239 1.71 1,930 I- 864
2 225 1.77 2,251 926
3 306 1.35 2,796 1,174
4 287 1.38 2,809 1,768
• KNAPP, B. JR., Ext. Anim. HU8bandman, June, 1940.

.9g~ BREEDING A ND IMPROVEMENT OF FAR M ANIMALS
ring standards. The fact that the beef market tends to be a discriminat-
ing one makes it necessary to pay considerable ~atten~ beef
cattle, but much more informat~ed regarding therclation
between type and carcass quality.
FIG. 208.-Appearance may not be a good indicator of a bull's breeding value. Calves sired
by the bull at the top averaged 42 lb. heavier at the end of a feeding period and produced ~
higher grading carcasses than calves sired by his more attractive half lirother shown in the
lower photo. (Courtesy of B1LreaU of Animal Industry, U.S. Department of Agrictdture.)
Although all breeding work with farm animals is slow and expensive,
such work with beef cattle is especially so because of slow reproductive
rates. Much information has been accumulated during the past 15 to 20
years, however, which when widely applied promises to aid in the develop-
ment of more profitable types of cattle.
Winters and McMahon (1933) in a series of experiments in which steers
were fed i~dividually, demonstrated rather wide differences between
\
•
individuals in gaining ability and in efficiency of feed utilization. Further
extensive work by the U.S. Department of Agriculture has extended these
observations (see Table 56 for some extreme cases), and it has been shown
that sire progenies differ considerably in these same characters.
The few heritability studies thus far reported for beef cattle (Table 57)
have almost all been from the same herd and seem to be unreasonably
TABLE 57.-HERITABILITY OF VARIOUS CHARACTERS IN BEEF CATTLE
I
Herit-
;\Iethod used to
Character ability, Reference
estimate heritability
%
Birth weight. ..... . ..... 53 Paternal half-sih corrp- Knapp and Clark
lation (1950) *
11 Paternal half-sib corre- Dawson et al.
lation (corrected birth (1947)
weights)
, Va
e mng weI g ht '" . 28 Pate mal half-sibe orr ~- K napp a nd Clark
lation (1950)
Live.weight at 15 months ... 86 Paternal half-sib corre- Knapp and Clark

lation (1950)
92 Sire offspring regres- Knapp and Clark
sion (1950)
Gain during feeding period .. 65 i Paternal half-sib eorre-i Knapp and Clark
lation (1950)
Rate of gain in feed lot. .. I: 77 Sire offspring regres- Knapp and Clark
Slon (1950)
Score at weaning ..... ., .. 28 Paternal half-sib corre- Knapp and Clark

lation (1950)
0 Sire offspring regres- Knapp and Nordskog
i sion (1946b)
Slaughter grade ..... ... 45 Paternal half-sib corre- Knapp and Clark
lation (1950)
Carcass grade. ...... 33 Paternal half-sib corre- Knapp and Clark

lation (1950)
Area of eye muscle ......... 68 Paternal half-sib corre- Knapp and Clark
lation (1950)
* KNAPP, BRADFORD, JR .. and CLARK, R. T., Revised Estimates of Heritability of Economic Charac-
ters in Beef Cattle (unpublished manuscript). This paper represents an extension of studies previously
reported by Knapp and Nordskog (l946a and 1946b) but includes about fOUI times as much material
(1)8 BREEDING AND I]'rlPROVEMENT OF FA.RM ANIMALS
high for those characters reflecting feed-lot efficiency and carcass quality.
However, even if further studies should indicate that the figures upon
·which these estimates are based were not typical of beef cattle in general
and that the estimates need to be revised downward, it does appear
likely that heritabilities are high enough for selection to be effective in
bringing about improvement in these characters.
Several record-of-performance procedures have been proposed for beef
cattle (Winters and McMahon, 1933; Sheets, 1932; Winters, 1940; and
Bureau of Animal Industry, 1941), and the factors usually considered
important will be listed and discussed briefly.
1. Birth Weight.-Although no one would advocate breeding beef
cattle for extremely heavy birth weights, it has been shown that, within
the normal weight range, heavier calves at birth tend to grow more
rapidly later (see Dawson et al. 1947). Thus, birth weights are a desirable
but not indispensable item in performance records.
2. Weaning W eight.-Weaning weight is important to all beef cattle-
men because, in general, gains made prior to weaning are cheaper than
those made later. The producers of calves for slaughter at weaning or of
feeder calves are especially interested in heavy weaning weights because
their income depends upon weight available for sale at that time. It has
been assumed that growth rate from birth to weaning is largely a function
of milk production of the dam. Gifford (1949) found gross correlations of
+0.60, +0.71, +0.52, and +0.35 between daily milk production of 57
Hereford cows and daily gain in weight of their calves during the first,
second, third, and fourth months, respectively. After the fourth month
there was no significant relationship between the two items. From this
data it would appear that, while milk production exerts an influence on
calf growth, it is by no means the only factor involved.
It has long been assumed by animal husbandmen that heavier weaning
weights are associated with rapid gains during a subsequent fattening
period. Knapp et al. (1941), however, found virtually no correlation
bet\veen these items. They also found that weaning weight was nega-
tively correlated with efficiency of gain during a subsequent fattening
period. Since the heavy calf at weaning gains as rapidly after weaning
as a lighter one, he will reach a given market weight in fewer days because
of his headstart. In all probability the lower efficiency of the heavier
weaning calves is due to the fact that they have reached a more expensive
part of the growth curve, and not to an inherently lower efficiency. Thus,
these observations do not constitute an argument for selecting light
weaning calves.
3. Gain and Efficiency of Feed Utilization during a Feeding Period.-Beef
cattle are ordinarily fed a heavy grain feed for a period of time prior to
\
\
\ \,
being marketed. The animal which gains rapidly during this period has
to be fed fewer days, with a resulting saving in labor and a more rapid
turnover of capital. The rapid-gaining animal also tends to be the most
efficient. This relation is by no means absolute, however.
Since no other characters seem to be highly enough related to feed-lot
performance to be of predictive value, feeding tests are needed to measure
this important character. How such feed tests should be conducted is
not a settled question.
Earliest proposals were for a feeding period of a fixed number of days
after weaning. This is by far the most convenient method, but has the
disadvantage that the animals may be in very different portions of the
growth curve during the period of similar chronological age. In general,
the heavy weaning calf will be at a disadvantage in efficiency evaluations
because he will be heavier throughout the feeding trial and therefore in
a more expensive part of the growth period. This method is, therefore,
not well suited for measurement of efficiency (see Knapp and Baker,
1944), but since there is little or no relationship between weaning weight
and subsequent rate of gain, it does provide a convenient and acceptable
method of evaluating rate of gain for use in selection.
It was later proposed (Black and Knapp, 1938) that animals on test
be fed through a given weight range, the suggested period being from
500 to 900 lb. live weight. They showed a very high relationship between
rate and economy of gain using this method.
Guilbert and Gregory (1944) pointed out that animals differing in
rate of maturity may differ greatly in composition at a given weight.
Since the energy required to produce fat is greater than that required to
produce lean meat and bone, it is apparent that the early maturing,
rapidly fattening calf would be at a disadvantage in evaluating efficiency.
They recommend feeding to a definite degree of finish in an effort to
control the composition of the gains.
Although there is no basic relationship between size and efficiency,
many of the overhead costs of livestock production tend to be on a "per
head" basis. Thus, as Brody (1939) has pointed out, it may be more
economical to produce the same amount of product from fewer animals
of large size than from a larger number of smaller animals.
It would appear, therefore, that the method of selecting on the basis of
rate and economy of gain during a constant weight period would be
satisfactory from a practical standpoint if it were combined with selection
for satisfactory carcass finish at the desired weight.
All investigators are agreed that the feeding period should be one of
full- or self-feeding. Knapp and Baker (1943) found that genetic
differences were not expressed between sire progenies if a limited feed
FIG. 209.-Angus bull (top) Prince Sunbeam 249th, 1948 Grand Champion bull at Inter-
national Livestock Show, shown by Ellerslie, Charlottesville, Va.; and cow (bottom) Eileen-
mere's Effie W., Grand Champion female, 1948 International Livestock Show, shown by
J. Garrett Tolan Farms, Pleasant Plains, Ill. (Courteay of American Aberdeen Angus
Breeders' A8sociation.)
intake put a ceiling on rate of gain. Knapp and Clark (1947) found that
genetic differences in rate of gain were not fully expressed until late in a
lopg feeding period. The apparent heritabilities of gain were 10, 54, and
84 per cent in the first, second, and third 84-day periods of a 252-day feed-
ing period, respectively. Thus, the chance of getting very much genetic
information from a feeding period of less than 168 days appears to be
slight.
\
•
Sl.'JLl~C1'j()N IN MEAT ANliVlALS 701
4. Type and Carcass Evaluation.-Type in the live animal can be best

expressed in terms of a grade or score given to each animal at the end of
the feeding period. For the steer progenies slaughtered, carcass grades
should be taken as the best possible evaluation of type.
The above discussion points out the records which are valuable in beef
production. Although the most efficient method of using performance
records in selection cannot be stated with certainty, it does appear that
each breeder can work out a system which, I1lthough it may not be
perfect, will be distinctly superior to no records.
Each calf should be evaluated on the basis of his own birth weight,
weaning weight, rate of gain, efficiency of gain, and body type.
A dam should be evaluated on the basis of her calves' performance,'
with special emphasis on weaning weight since it refiects maternal ability.
More than one record is desirable for evaluating a cow, since the corre-
lation between the weaning weights of calves from the same cow is only
about +0.50 and between type grades only +0.33 (Koger and Knox,
1947). Koger and Knox found, however, that a heifer whose first calf
was decidedly poor (barring accidents such as crippling) was never a
heavy producer later in life. Thus, some culling of first-calf heifers can
be safely done with little or no chance of discarding a valuable cow.
Efficiency records on individual animals can only be obtained if indi-
vidual feeding is practiced. This expensive practice is probably not
practical for the average breeder. He can, however, group feed and get
individual weight records, and if the groups are fed by sires can get sire-
progeny efficiency records.
Theoretically, sires can best be evaluated upon the basis of the per-
formance of a number of offspring, preferably at least eight to ten. Since
rate and economy of gain are apparently relatively high in heritability,
it may be that progeny testing is not needed for sire evaluation. The
U.S. Department of Agriculture has recently been experimenting ·with
the method of evaluating prospective herd bulls upon the basis of their
pe.rformance under a heavy grain-feeding program for a period of 6 >~
months. Those animals performing best can then be used for breeding
in the regular herd at about fifteen months of age without waiting the
additional 2 years necessary for progeny-test information.
Results to date indicate a high correlation between the performance of
a sire and that of his get. This is to be expected if heritability is high.
This method needs extensive trial in other herds.
Full feeding of prospective breeding heifers is ordinarily not practiced,
but many prospective show animals are full-fed without apparent harm.
If heifers are to be properly evaluated for gaining ability, it is desirable
702 BREEDING AND IMPROVEMENT OF FARM ANIM.i"lLS
FIG. 210.-Hereford bull (top) MW Larry Mixer 1st, champion Hereford bull of the 1949
American Royal Livestock Show, Kansas City, Mo. Bred and shown by the Milky Way
Hereford Ranch, Phoenix, Ariz. ; and Hereford cow (bottom) JJ Gertt"Udis, reserve cham pion
Hereford female of the 1949 American Royal Livestock Show, Kansas City, Mo. Bred
and shown by Jack Haley, Escondido, Calif. (Courteay of American Here/ord Association.)
to full-feed them for a period, although it is recognized that this may not
be possible in some cases.
A cooperative beef-cattle breeding program has been recently started
by the U.S. Department of Agriculture and cooperating state experiment
stati'Ons, with Dr. R. T. Clark of Denver, Colo., as coordinator.
Inbreeding and performance testing are being carried on by a number of
experiment stations. Due to the slow rate of reproduction in cattle,
this program can not be expected to produce results so rapidly as the
Regional Sheep and Swine Breeding Laboratories. The progress of this
program will be watched with interest.
FIG. 211.- Lassie of W.T. 92607, Grand Champion Red Poll female at the 1949 Indiana
State Fair. The Red Poll is a dual-purpose breed of cattle. (COU1·tesy of Purdue University,
Lafayette, Ind.)
Selection in Dual-purpose Cattle.-Because dual-purpose cattle are

kept both for milk and beef production, it is obvious that breeders should
use both the index commonly in use in dairy herds as well as the one now
in process of formation for beef animals. The problem here is similar to
that found in all classes of livestock, viz., the securing of sufficient data so
that selection may be practiced in keeping with the known laws of
inheritance, in other words that it rest on a sound genetic basis.
Dual-purpose cattle selection is not very different from that prevailing
in all other classes of livestock, for most breeders try to select for more
than one thing at a time. Here we are selecting for two major things
at once, milk and beef, and in addition the two things are to some degree
antagonistic. We know, therefore, at the start that we will probably
never combine both things in their extreme degrees in anyone animal.
The beef breeds have many genes for beef production and a few for milk
production owing to the breeding and selection that has been practiced
in the past. The opposite situation, of course, prevails in the dairy

breeds. Dual-purpose cattle breeders want about a 50-50 combination
of beef and milk genes. Discovering the animals that come closest to
having such a genetic make-up by testing and record keeping, followed
by intelligent mating and the use of some degree of inbreeding, s1.ould
eventually produce superior strains of dual-purpose animals.
Summary.-In this chapter we have reviewed the evidence that
hereditary differences exist in virtually all characters importantly related
to the efficient production of high-quality meat-animal products. It
has been pointed out that there is a definite relation between visual
appraisals of type and carcass quality of enough importance that indi-'
vidual appearance will continue to be used as one criterion of quality ..
Visual type appraisals are, however, poor indicators of productive effi-
,ciency and must therefore be supplemented with production records if
maximum progress is to be made. The breeding of meat animals is just
l.iJeginning to be treated as a science, and we can look forward to an
increasing rate of development in the future.
References
Books
HAGEDOORN, A. L. 1946. "Animal Breeding," 2d cd., Crosby, Lockwood & Sons,
London.
HAMMOND, JORN. 1940. "Farm Animals, Their Breeding, Growth, and Inherit-
ance," Edward Arnold & Co., London.
KELLY, R. B. 1946. "Principles and Methods of Animal Breeding," John Wiley &
Sons, Inc., New York.
LusR, JAY L. 1945. "Animal Breeding Plans," 3d ed., Collegiate Press, Inc., of
Iowa State College, Ames, Iowa.
NICROLS, J. E. 1945. "Livestock Improvement in Relation to Heredity and
Environment," 2d ed., Oliver & Boyd, Ltd., Edinburgh and London.
WINTERS, L. M. 1948. "Animal Breeding," 4th ed, John 'Viley & Sons, Inc.,
New York.
BAKER, M. L., HAZEL, L. N., and REINMILLER, C. F. 1943. The Relative Impor-
tance of Heredity and Environment in the Growth of Pigs at Different Agee,
BLACK, W. H., and KNAPP, B., JR. 1938. A Comparison of Several ::\lethods of
Measuring Performance in Beef Cattle, Amer. Soc. Anim. Prod. Proc., pp. 103-
107.
HLUNN, C. T., and BAKER, M. L. 1949. Heritability Estimates of Sow Productivity
and Litter Performance, Jour. A nim. Sci., 8 :89-97.
- - - and - - - . 1947. The Relation between Av..erage Daily Gain and Some
Carcass Measurements, Jour. Anim. Sci., 6 :424-431.
BRODY, S. 1939. Factors Influencing the Apparent Energetifl Efficiency of Produc-
ti,\e Processes in Farm Animals, Jour. Nutr., 17 :235-251.
•
SELECTION IN MEAT ANIMALS
705
Bureau of Animal Industry, lI.s. Dept. Agr. 1941. Record of Performance Proce-
dure for Beef Cattle, A.H.D. Mimeo. 37.
CARROLL, W. E. et al. 1929. Swine Type Studies. 1. Type in Swine as Related
to Rate and Economy of Gain, Illinois Agr. Expt. Sta. Bul. 321.
CHAPMAN, A. B., and LUSH, J. L. 1932. Twinning, Sex Ratios, and Genetic Vari-
:lbility in Birth Weight in Sheep, Jour. H ered., 33 :473-478.
CLAUSEN, HJALMER. 1949. Beretning om sammenlignende Fors¢g med Suin fra
statsanerkendte Avlscentre, (Report about the comparisons between swine from
the state recognized breeding centers, 1947-1948), Ejvind Christensens Forlag,
Copenhagen.
COMSTOCK, R. E. et al. 1942. Measures of Growth Rate for Use in Swine Selec-
tion, Jour. Agr. Res., 65 :379-389.
CRAFT, W. A. 1943. Swine Breeding Research at the Regional Swine Breeding
Laboratory, U.S. Dept. Agr. Misc. Pub. 523.
CUMMINGS, J. N., WINTERS, L. M., and STEWART, H. A. 1947. The Heritability of
Some Factors Affecting Productivity of Brood Sows, Jour. Anim. Sci., 6 :297-304.
DAWSON, W. M., PHILLIPS, R. W., and BLACK, W. H. 1947. Birth Weight as a
Criterion of Selection in Beef Cattle, Jour. Anim. Sci., 6:247-257.
DICKERSON, G. E. 1947. Composition of Hog Carcasses as Influenced by Heritable
Differences in Rate and Economy of Gain, Iowa Agr. Expt. Sta. Res. Bul. 354.
- - - and GRIMES, J. C. 1947. Effectiveness of Selection for Efficiency of Gain
in Duroc Swine, Jour. Anim. Sci., 6 :265-287.
- - - and HAZEL, L. N. 1944a. Selection for Growth Rate of Pigs and Produc-
tivity of Sows, Jour. Anim. Sci., 3 :201-212.
- - - and - - - . 1944b. Effectiveness of Selection on Progeny Performance as a
Supplement to Earlier Culling in Livestock, Jour. Agr. Res., 69 :459-476.
- - - et al. 1947. Performance of Inbred Lines and Line Crosses in Swine, Jour.
Anim. Sci., 6 :477. (Abs.)
GIFFORD, WARREN. 1949. Importance of High Milk Production in Beef Cows
Found Overestimated, Jour. Anim. Sci., 8 :605-606. (Abs.)
GUILBERT, H. R., and GREGORY, P. W. 1944. Feed Utilization Tests with Cattle,
Jour. Anim. Sci., 3 :143-153.
HAMMOND, J. 1947. Animal Breeding in Relation to Nutrition and Environmental
Conditions, Cambridge Phil. Soc. Biol. Rev., 22 :195-213.
HAZEL, L. N. 1943. The Genetic Basis for Constructing Selection Indexes, Genetics,
28 :476-490.
----- and LUSH, J. L. 1942. The Efficiency of Three Methods of Selection, Jour.
Hered., 33 :3\)3-399.
- - - and TERRILL, C. E. 1946a. Heritability of Type and Condition in Range
Ramboui!let Lambs as Evaluated by Scoring, Jour. Anim. Sci., 5:55-61.
- - - and - - - . 1946b. Effects of Some Environmental Factors on Weanling
Traits of Range Columbia, Corridale, and Targhee Lambs, Jour. Anim. Sd.,
5:318-325.
- - - and - - - . 1946c. Heritability of Weanling Traits in Range Columbia,
Corriedale, and Targhee Lambs, Jour. Anim. Sci., 5:371-377.
- - - - and - - - . 1946d. Effects of Some Environmental Factors on Fleece and
Body Characteristics of Range Rambouillet Yearling Ewes, Jour. Anim. Sci.,
5:382-388.
- - - and - - - . 1945a. Effects of Some Environmental Factors on Weanling
Traits of Range Rambouillet Lambs, Jour. Anim. Sci., 4 :331-341.
- - and - - - . 1945b. Heritability of Weaning Weight and Staple Length in

Range Rambouillet Lambs, Jour. Anim. Sci., 4 :347-358.
HETZER, H. 0., DICKERSON, G. E., and ZELLER, J. H. 1944. Heritability of Type
in Poland China Swine as Evaluated by Scoring, Jour. Anim. Sci., 3 :390-399.
- - - , LAMBERT, W. V., and ZELLER, J. H. 1940. Influence of Inbreeding and
Other Factors on Litter Size in Chester White Swine, U.S. Dept. Agr. Cir. 570.
JONES, J. M., et ai. 1946. Inheritance of Skin Folds of Sheep, Jour. Anim. Sci.,
5:154-169.
~--, et al. 1944. Influence of Age, Type, and Fertility in Rambouillet Ewes on
Fineness of Fiber, Fleece Weight, Staple Length, and Body Weight, Tex. Agr.
KOGER, M., and KNOX, J. H. 1947. The Repeatability of the Yearly Production of
Range Cows, Jour. Anim. Soc., 6:461-466.
KNAPP, B., JR., and BAKER, A. L. 1944. Correlation between Rate and Efficiency
of Gain in Steers, Jour. Anim. Sci., 3:219-224.
-~- and - - - . 1943. Limited vs. Full-feeding in Record of Performance Tests
for Beef Cattle, Jour. Anim. Sci., 2 :321-327.
- - - , - - - , et al. 1941. Record of Performance in Hereford Cattle, Mont. Agr.
- - - and CLARK, R. T. 1947. Genetic and Environmental Correlations between
Growth Rates of Beef Cattle at Different Ages, Jour. Anim. Sci., 6:174-181.
- - - and NORDSKOG, A. W. 1946a. Heritability of Growth and Efficiency in
Beef Cattle, Jour. Anim. Sci., 5:62-70.
- - - and - - - . 1946b. Heritability of Live Animal Scores, Grades and Certain
Carcass Characteristics in Beef Cattle, Jour. Anim. Sci., 5:194--199.
KRIDER, J. L. et al. 1946. Effectiveness of Selecting for Rapid and for Slow
Growth Rate in Hampshire Swine, Jour. Anim. Sci., 5:3-15.
LUSH, J. L. 1947. Family Merit and Individual Merit as Bases for Selection, Amer.
Nat., 81 :241-261, 362-379.
1936. Genetic Aspects of the Danish System of Progeny Testing Swine,
Iowa Agr. Expt. Sta. Res. Bul. 204.
- - - and KINCAID, C. M. 1943. Adjusting Weights of Pigs to an Age of 154 Days,
Regional Swine Breeding Laboratory Res. Item 25.
- - - and MOLLN, A. E., JR. 1942. Litter Size and Weight as Permanent Charac-
teristics of S:)ws, U.S. Dept. Agr. Tech. Bul. 836.
McMEEKAN, C. P. 1940-41. Growth and Development in the Pig with Special
Reference to Carcass Quality Characters, Parts I - V, Jour. Ag;. Sci., 30 :276-343,
387-436,511-568; 31:1-49.
MOULTON, C. R., TROWBRIDGE, P. F., and HAIGH, L. D. 1921. Studies in Animal
Nutrition. 1. Changes in Form and Weight on Different Planes of Nutrition,
Mo. Agr. Expt. Sta. Res. Bul. 43.
- - - , - - - , and - - - . 1922. Studies in Animal Nutrition. II. Changes in
Proportions of Carcass and Offal on Different Planes of Nutrition, Mo. Agr.
Expt. Sta. Res. Bul. 54.
NEALE, P. E. 1946. Corrective Sheep Breeding, N. Mex. Agr. Expt. Sta. Bul. 334.
NELSON, R. H., and VENKATACHALAM, G. 1949. Estimates of the Heritability of
Birth Weight and Weaning Weight of Lambs, Jour. Anim. Sci., 8:607-608.
(Abs.)
NORDSKOG, A. W., COMSTOCK, R. E., and WINTERS, L. M. 1944. Hereditary and
Environmental Factors Affecting Growth Rate in Swine, Jour. Anim. Sci., 3 :257-
273. \
\ ! \
,
SELECTION IN MEA T ANIMALS 707
PHILLIPS, R. W., and DAWSON, W. M. 1940. Some Factors Affecting Survival,
Growth, and Selection of Lambs, U.S. Dept. Agr. Cir. 538.
- - - , SCHOTT, R. G., and SPENCER, D. A. 1945. The Genetics, Physiology, and
Economic Importance of the Multinipple Trait in Sheep, U.S. Dept. Agr. Tech.
Bul.909.
RHO~DE, A. O. 1949. The Santa Gertrudis Breed. The Genesis and the Genetics
of a New Breed of Beef Cattle, Jour. Hered., 40:115-126.
SEMPLE, A. T., and DvORACHEK, H. E. 1930. Beef Production from Purebred,
Grade, and Native Calves, U.S. Dept. Agr. Tech. Bul. 203.
SHEETS, E. W. 1933. Evaluating Beef Cattle Performance for a Register of Merit,
Amer. Soc. Anim. Prod. Proc., 25:41-47.
SPENCER, D. A., and HARDY, J. 1. 1928. Factors That Influence Wool Production
with Range Rambouillet Sheep, U.S. Dept. Agr. Tech. Bul. 85.
STEWART, H. A. 1945. The Inheritance of Prolificacy in Swine, Jour. Anim. Sci.,
4:359-366.
STONAKER, H. H., and LUSH, J. L. 1942. Heritability of Conformation in Poland-
China Swine as Evaluated by Scoring, Jour. Anim. Sci., 1 :99-105.
TERRILL, C. E. 1949. The Relation of Face Covering to Lamb and Wool Production
in Range Rambouillet Ewes, Jour. Anim. Sci., 8:353-361.
- - - and HAZEL, L. N. 1946. Heritability of Face Covering and Neck Folds in
Range Rambouillet Lambs as Evaluated by Scoring, Jour. Anim. Sci., 5 :170-179.
- - - and - - - . 1943. Heritability of Yearling Fleece and Body Traits in
Range Rambouillet Ewes, Jour. Anim. Sci., 2 :358-359. (Abs.)
- - - , - - - , and SIDWELL, G. M. 1948a. Effects of Some Environmental Factors
on Yearling Traits of Columbia and Targhee Rams, Jour. Anim. Sci., 7 :181-190.
- - - , - - - , and - - - . 1948b. Effects of Some Environmental Factors on
Traits of Yearling and Mature Rambouillet Rams, Jour. Anim. Sci., 7 :311-319.
- - - , - - - , and - - - . 1947. Effects of Some Environmental Factors on
Yearling Traits of Columbia and Targhee Ewes, Jour. Anim. Sci., 6:115-122.
VERGES, J. B. 1939. Effect of Nutrition on the Carcass Quality of Suffolk Cross
Lambs, Suffolk Sheep Yearbook, Ipswich.
WARWICK, B. L. et al. 1949. Selection of Sheep and Goats for Resistance to
Stomach Worms, Haemonchus contortus, Jour. Anim. Sci., 8:609-610. (Abs.)
WHATLEY, J. A. 1942. Influence of Heredity and Other Factors on 180-day Weight
in Poland-China Swine, Jour. Agr. Res., 65 :249-264.
- - - and NELSON, R. H. 1942. Heritability of Difference in 180 Day Weight
and Market Score in Swine, Jour. Anim. Sci., 1 :70. (Abs.)
- - - and QUAIFE, E. L. 1937. A Method for Correcting Pig Weights to a 56-day
Basis, Regional Swine Breeding Laboratory Res. Item 2.
WINTERS, L. M. 1940. Records of Performance for Meat Animals, Empire Jour.
Expt. Agr., 8 :259-268.
- - - and McMAHON, H. 1933. Efficiency Variations in Steers, A Proposed
Record of Performance, Minn. Agr. Expt. Sta. Tech. Bul. 94.
- - - , SIERK, C. F., and CUMMINGS, J. N. 1949. The Effect of Plane of Nutrition
on the Economy of Production and Carcass Quality of Swine, Jour. Anim. Sci.,
8:132-140.
ZELLER, J. H., and HETZER, H. O. 194,1. Influence of Type of Hog on Production
Efficiency, U.S. Dept. Agr. Cir. 698.
CHAPTER XXIII
SELECTION IN HORSES
The creation by selection of the various types of horses from one or a

few ,vild progenitors has been one of man's most interesting and most
helpful accomplishments. The development of more efficient power to
accomplish physical labor is simply the story of an advancing civilization.
As a savage, man's only source of power was his own muscle, and in order
to survive he had to use it so continuously that little time or energy was.
left for cerebration. At some time the idea dawned of using animal
power to supplement or replace his own limited supply, and many animals
have been so used, among others the dog, ox, yak, camel, llama, goat,
elephant, ass, and horse. Later, water, steam, gas, electricity, and gaso-
line were used, and now the tremendous power of the atom is about to be
harnessed for constructive progress.
The following figures l suggest the very rapid rate of increase in mechan-
ical power in the "Cnited States.
Total horsepower I Horsepower

Year
in prime movers per capita
1860 15,793,000 0.50

1900 64,193,000 0.90
1940 1,230,816,000* 9.35
* For tractors and automobiles drawbar equivalent was used rather than rated horsepower.
If we calculate that 1 horsepower is the equivalent of 10 man power,

it is evident that each person living in the United States in 1940 had as
his servants the equivalent of about 93 slaves. (Like the writer, you
probably wish at times that some of these 93 were a little nearer at hand.)
Savages existed with no source of power save their own muscle, and
without the adoption of outside sources of power we would still be eking
out a precarious existence as savages.
In 1787, the year the Constitution was framed, the surplus food
produced by 19 farmers went to feed 1 city person. In recent years 19
1 Figures supplied through kindness of Z. R. Pettit, chief statistician for Agricul-
ture, Bureau of the Census, Department of Commerce, Washington, D.C.

\ 708
SELECTION IN HORSES 709
people on farms have produced enough food for 56 nonfarm people plus
10 living abroad. 1
This is probably due, for the most part, to the fact that the modern
farmer has more efficient sources of power. In ancient times a few
peofJle could live well and enjoy leisure because they had enslaved othm
human beings who were compelled to work for their keep. Today oUI
slaves in America are machines.
Among the animals that have been domesticated and whose energy
has been used to supplant human muscle, the horse is the most interesting
and the most important. Next to the dog, among the larger animals, the
horse apparently has the greatest capacity for responding to man's
friendship. From the earliest times he was both helpmate and com-
panion. The maximum efficiency of both man and horse during work,
according to the researches of Brody and Trowbridge,2 stands at 24 per
cent, with steam engines at 15 per cent, gasoline engines at 18 per cent
"at the belt" and about 13 per cent "at the drawbar," and Diesel
engines at 35 per cent. Many other factors, of course, are involved in
the choice of the" best" source of farm power.
In addition to supplying power, the horse had filled a great variety
of human needs. History, unfortunately, is too largely the story of
war and conquest, and, although "struggle" is the keynote and first
necessity of a nation's or an individual's progress, it is to be hoped that
less devastating forms of struggle than war can speedily be devised.
Associated with the famous warriors of the past, there generally is found
their favorite horses, e.g., Alexander the Great and Bucephalus, Napoleon
and Marengo, Washington and Nelson, Grant and Jack, Lee and Traveler.
These men's names live in history; their physical likenesses, generally
astride their favorite horse, on canvas or in bronze. The place of
cavalry in war is too well known to need repetition here, and the same
is true of the place of the horse in moving supplies, guns, and other
equipment. During the First World War the United States shipped
abroad for war purposes nearly 1 million horses and over ~ million
mules. And, in spite of motorization, the horse played an important
part in the Second World War, for Germany is estimated to have used
over 200,000 horses and mules in the Polish campaign and 750,000 in the
conquest of the Low Countries and France.
In addition to his services in work or war the horse has filled, and will
1 MCCRORY, S. H., HENDRICKSON, R. F., and COMMITTEE, Technological Trends
in Relation to Agriculture, reprint from Report of Subcommittee on Technology to
the National Resources Committee, 1937, p. 99.
2 BRODY, S., and TROWBRIDGE, E. A., Efficiency of Horses, Men, and Motors, "Mo.

71tf"'- BREEDING AND IMPROVEMENT OF FARM ANIMALS
probably continue to fill, a great place in pleasure and sport. In colonial

America, he provided the most rapid means of land transportation. The
American Standardbred and Saddlebred finally grew out of this in the
Southern colonies, and even today Virginia, Tennessee, and Kentucky
retain first place in the production of this light-horse type. The Middle
and Northern colonies developed a somewhat heayier general-utility
horse, which provided the basis for the later draft-horse development.
Riding, racing, and polo still claim the attention of millions of Ameri-
cans, and there appears to be little likelihood of abatement of this
interest.
As a source of power in cities and on farms the horse is having serious
competition from the gasoline engine. Both the motor and the horse
have advantages, some of which are listed below:
Motor Horse
Advantages Advantages
More work in given time Lower initial investment and depreciation
No permanent unsoundnesses or disease Power of reproducing
Can work day and night Live on home-produced feeds
No care nor fuel while not working Flexibility-use 1 or 30
Less man power needed Help maintain soil fertility
Use for belt power and custom work Less actual cash outlay
The advantages of one are to a certain if not the full degree disad-
vantages of the other. Whether motors will entirely displace horses as
a source of farm power is for the future to answer. The American farmer,
like workers in all other fields, is becoming more and more efficient,
i.e., he produces more in less time and with less human physical effort.
This tends to create a puzzling economic problem, for it means we cannot
profitably consume all that is produced if all work as many hours a day
or a week as they did under less efficient methods of production. As
individuals, we are not interested in efficiency for efficiency's sake but
only in translating efficiency into better living for ourselves and our
families. If our efficiency can be made to mean a greater net income or
the same income from less hours of labor, the average man is definitely
interested. Hours of labor have shown a steady decline for the past
40 years; it is probable that the bottom in decreased labor time has not
yet been reached.
The farmer is on the lookout for more efficient sources of pow~r. If
engineers can provide machines that can do the general run of farm
work more efficiently and more cheaply than can the horse, it seems
certain that farmers will find ways to buy these machines. Table 58
shows something of the trend in forms of power on the farm.
\
TABLE 58.-POWER ON FARMS
1920 1930 1940 1945
Horses and mules .......... 25,742,000 19,124,000 13,931,000 8,274,000

Tra~tors .......... " ...... 229,332 920,021 1,567,430 2,458,628
Farms with electricity ..... ........... 841,310 2,032,316 2,787,624
Trucks ................... 150,000* 900,385 1,047,084 1,490,300
Automobiles .............. 2,000,000* 4,134,675 4,144,136 4,148,275
* Estimate.
In any event the farm horse is going to have to try to meet this compe-
tition. It seems unlikely that the farm horse will be totally replaced
by machines, and it is evident that horse breeders must use every possible
means to improve their methods of selection so that only sound, durable,
and efficient types of horses will be produced.
Many farmers who keep horses and raise colts do not have sufficient
numbers of animals to justify the keeping of a stallion, which means that
they must patronize various and sundry owners of stallions. In the
early part of the twentieth c~ntury, there were many unsound and poor-
type stallions and jacks standing for public service at low service fees in
the Midwest, and many also that were improperly registered. In order
to protect both the owners of good purebred sires and the owners of
mares, some states inaugurated stallion-enrollment or licensing services,
which required the payment of a fee for enrollment or licensing, provided
the stallion was adjudged sound upon inspection by a veterinarian, and
the inclusion of a copy of the license in all advertising. Other states
were, of course, forced to follow suit to prevent the dumping within
their borders of inferior stallions from states that had adopted such
legislation. This is the only class of animals in which legislation has
been invoked as an aid in bringing about improvement. Twenty-two
states from New York and New Jersey on the Atlantic to Washington
and Oregon on the Pacific and including New Mexico and Oklahoma in
the South now have stallion registration boards, and a national asso-
ciation of these state boards was formed in 1910 and is still functioning
with Prof. W. L. Blizzard of Oklahoma as its president and Prof. R. B.
Cooley of Indiana as secretary.
Although the state laws for licensing stallions and jacks vary somewhat,
in general it may be said that to be eligible for public service a stallion
must be free from sidebone, ringbone, bone spavin, curb (when accom-
panied by curby formation of the hock), glanders-farcy, maladie-du~
co'it, urethral gieet, and mange. He must be registered in a nationally
recognized registry association or be advertised as a grade. In some
"tates only sound purebred stallions are eligible for a license. In other
states an unsound stallion or jack may be licensed, but the nature of the
unsoundness must be shown on the face of the license.
The main purposes of stallion-enrollment laws were and are to prevent
misrepresentation in advertising and to prohibit the use of sires thought
likely to spread infectious diseases or a conformation that would not
stand up under hard work and, conversely, to encourage the development
of a sound breeding program for horses. That these stallion-enrollment
laws have accomplished their main purposes cannot be questioned, and,
although it is impossible to estimate their financial advantages to breed-
ers, farmers, and the respective states, their contribution to a better
average type of horse is admittedly very great.
General Bases for Selection in Horses.-In general there are five l
types of horses, l)iz., draft, heavy harness, light harness, saddlers, and ')
ponies. They are maintained in the final analysis for but one thing,
their ability to move for its own sake, as in the various types of racing,
driving, hunting, riding, polo-playing, etc., or their ability to move loads,
as in various forms of draft work. The criterion of selection in horses
has always been their ability to do certain things plus, in some instances,
)Juili: p~auty of appearance. In the heavy and light harness horses, the
pony, and the Saddle Horse used in the show ring, perhaps equal weight
has been given to appearance and performance. In the hunter, utility
saddler, draft, and various types of racehorses the main influence shaping
selection has been performance, with somewhat less attention paid to
appearance. The distinct types of horses have evolved quite naturally,
as breeders selected those which excelled in the various types of perform-
ances, and so it has come about that in horses perhaps more than in any
other class of livestock "~!L_~_£roduction." Draft power calls for a
compact, heavy, low-set, heavily musclelt,-iind clean-boned animal "with
a relatively docile disposition. Speed calls for a lithe, lean, relatively
long-muscled and light-boned animal with plenty of animation and a
highly developed nervous temperment. The individuality of the animal,
its phenotype, because this so closely correlateswithits performance, is
and should be the most important consider~tion in selection.
Pedi~..L if relatively complete in terms of records of performance
(work, speed, show_ winnings, etc.), can be of very great usefulness in
providing a safer basis for-selection; and progeny performance, as in all
classes of livestock, provides the one unaSs~est of an animal's
breeding merit. Performance~.cords are available for various types
of racing horses, which provides something comparable to milk-produc-
tion records in cattle. To a limited extent, show winnings are available
for heavy, and light harness horses and Saddle horses. The dynamometer
is available for the testing of draft power in heavy horses, but thus far
it has not been used in any practical way to help in the selection of draft
horses.
The principles of inheritance in horses are similar to those in all other
clas~es of livestock. The horse is the product of its genes times its
environment. In this class of livestock, the enVironment in the form of
training plays a much more important part than is true in any other
class of livestock. Proper feeding is essential if any animal of any class
is to grow out to the normaCsetby-its particular genes. Cattle, sheep,
and swine require no particular training, whereas training in the horse
is fully as important as good genes. An offspring of Dan Patch and Lou
Dillon would hardly be expected to do 1 mile in 2 minutes if it had grown
up on pasture and spent the years from three to five pulling a grocery
or laundry wagon.
1\1ore also in horses than in other classes of livestock do we need balance
between the several parts of the body and between the physical and the
mental make-up. We can overdo compactness in a draft horse until we
reach the point where we have an animal that cannot move effectively.
We want length of leg in a racehorse, but we can reach the point where
further increase in length of leg detracts from rather than adds to speed.
Likewise, we can have too phlegmatic a draft horse or too high-strung
and nervous a Thoroughbred.
The halving and sampling nature of inheritance prevails in the horse
as well as complex interactions between groups of genes providing favor-
able and unfavorable "nicks." We still have far too many horses ill-
adapted, both from a genetic and from a training standpoint, for the
purpose they are supposed to fill. With the number of horses declining,
it is of great importance that ...ve use all means available, and perhaps
design new ones, to measure both the performance and the transmitting
qualities of those remaining, so that the less efficient may be the first to
go and the better genes discovered and put together in more desirable
and homozygous combinations.
Buying Horses.-Buying a horse, even for the initiated, is something
of ~gamble. This is due to the fact that, in the sh0rt space of minutes
0; hour-s with the knowledge that many faults can be temporarily covered
up through clever manipulation, we must evaluate a complex physical
and mental make-up for use over long hours of work through days
stretching into years.
As -with all other classes of livestock, the first consideration is that of
general health. Horses that have been shipped or have passed through
sales stables may have had the opportunity of picking up various germs
sueh as thosp, causing; pleuropneumonia (shipping fever) or influenza,
Such ailments very seldom prove fatal, but, if carried home to a group
of horses, considerable inconvenience may be caused.
Since the horse is maintained for his ability to transform energy into
useful work or to provide pleasure and companionability, one of the
most important things to be appraised is his disposition. Whether or
not there are genes for obnoxious dispositions in animals or man is not
known, but something does occasionally give rise to bad dispositions,
and whether genetically or, probably more often, environmentally caused,
we do not want them. Undesirable dispositions may manifest themselves
in the stall as halter pulling, kicking, biting, crowding, stall walking or
trotting, weaving, cribbing, wind sucking, manger chewing, tail rubbing,
etc., and outside the stall as an unwillingness to be harnessed, balking,
running away, stumbling, etc. Work becomes doubly hard when it
must be performed with the aid of recalcitrant beasts, and in a horse
maintained for purposes of pleasure the first prerequisite is good manners.
Even if not dangerous or particularly obnoxious, many cf these faults
and vices consume considerable energy for which the owner gets no return.
In addition to securing healthy horses that are free from objectionable
manne~dyices.-Ql!_e inl!St,~[~urse, select horses whose types fit them
for the specific functions that they are intended to fulfill. This, of course,
has many gradations from ponies to draft horses. Limitations of space
and the general availability of excellent discussions on types of horses
for specific requirements in book and bulletin form make both impossible
and unnecessary similar discussions at this point. Since the value of a
horse depends so largely on his ability to move from place to place, his
feet and legs should receive first and major attention or, said more
concisely, "no foot, no horse." Fully as important, however, are the
matters of SOlinG ,vind and good eyes. Any unsoundnesses of feet or
legs, inability to breath easily and normally, or any impairment of vision
should be avoided in purchasing horses. Since he is not a gift horse,
better look in his mouth too in order to check on the soundness of his
teeth as well as his age.
One final consideration remains, after having checked on a horse's
freedom from vices, his disposition, his soundness of foot and limb, wind,
mouth, and eyesight, his action, his-general (ype c" blilidas to-the work
or service it is intended to have him perform, viz., his adaptability for
specific work!
A horse may possess proper conformation, be sound, and have good action yet
still not be well adapted for a specific work; consequently it is very essential
that he be thoroughly examined at the work for which he is wanted. If the
1 REESE, H. H., How to Select a Sound Horse, U.S. Dept. Agr. Farmers' Bul. 779,
pp.21-22, \
\
\
horse is to be used for heavy hauling for draft purposes, steady pulling under all
conditions is an indispensable quality. For harness use the horse should drive
promptly and freely with an easy, rapid gait and an alert expression, taking just
sufficient hold of the bit to be in hand without causing the driver to pull on the
lines. The saddle horse should have an easy, prompt mouth, style, graceful
carrtage and should stand quietly to be mounted and dismounted.
Some horses are difficult to harness and object to taking the bit in their mouths;
others jump when an attempt is made to place a saddle or harness on their backs;
while still others offer a great deal of resistance to having the crupper placed
under their tails, which, if due entirely to general muscular strength and tension,
may be an indication of endurance. While being hitched up or mounted the
horse should stand quietly and should start promptly but quietly on command.
For any purpose the following vices should cause the animal to be rejected:
balking, backing, rearing, kicking, striking with the forefeet, or running away.
Less important vices are: throwing the head up or down, shying, scaring, breaking
loose when tied, resting one foot upon the other, grasping the bit between the
teeth, rolling with the harness on, or switching the tail over the lines. Occa-
sionally the last named vice causes the horse to kick, in which case it becomes
dangerous.
Enlargements or scars (due to deformity, unusual mishap, or uncommon
disease) not conforming to any of those discussed should cause a horse to be
rejected unless the nature of the cause and the detriment to the value and useful-
ness of the animal is self-evident.
Experience gained by examining large numbers of horses will aid in quickening
the eye and judgment, thereby making it possible to perceive readily any unusual
condition, but it should be remembered that a hurried examination is liable to
prove a disappointment: consequently plenty of time should be taken in making
the examination, because time is much cheaper than money tied up in an unsatis-
factory horse. In some countries nine days are allowed by law to the purchaser
in which to learn of the serious forms of unsoundness or vice in a horse, so that in
this country it would seem fair to allow at least a day for a trial when practicable.
If possible, get a history of the animal, and while you are about it get a history
of the person having it for sale. So many defects may be covered up by such
unfair methods as drugging that it is a good plan to make purchases only from
persons with good reputations.
Horses offered at auction sales should preferably he thoroughly examined
previous to their being brought into the ring; at least they should be tried out in
compliance with the rules of the sale before time for settlement.
Finally, it is well not to form the habit of seeing only the defects, for horses,
like people, are seldom perfect; consequently in judging them weigh the good
qualities against the bad. A horse should be valued by the amount of service
it will perform rather than by its minor shortcomings.
In case of the purchase of mares i£t;ended to serve in whole or in part as
brood mares, attention should"oegiven to their pedigrees, their collateral
relatives, and their offspring in addition to their own indIVIduality.
'i'16 BREEDING AND iMPROVEMENT OF FARM ANIMALS
Breeder Must Know His Own Band First.-In spite of the fact that
horses by and large have been selected on the basis of a phenotype lending
itself to the efficient performance of specific tasks for a considerable
period of time, they are as yet far from homozygous for desirable genes.
The best appearing mare may not beget the best-type foal when br~d to
stallion A although she might if bred to stallion B. It is a worth-while
and sometimes very surprising exercise to lead out our animals by families,
standing the foundation female at the head of each separate female line
and arranging her offspring behind her so that all the offspring of each
successive sire used will form a line at right angles to the female lines.
Foundation Mares - Pentoila's Favorite Daffodil's Belle
Stallions
~
Colin
I
Lily
Revelation
A
Laura Lady Bene
/~ - ~
Laletto Lena * Imdy Lou * La Belle*
~t_~:h:'Lt
Konchapet
Konhysop
Konhopecar II Lucille * Lorn,a* Hylea* Louise * Kathleen * Lana*
• Indicates still in Brood Mare Band

FIG. 212.-Family chart of University of Massachusetts Percherons. Only femaleR in
living lines shown.
This is easily done with horses and provides the opportunity for studying
the whole band of mares with their offspring, of sizing up and evalu'ating
the contributions of both sires and dams and arriving at an estimate of
the good and bad features of the group as a whole, and of visualizing the
trend which the group as a whole is apparently taking.
The story of the Percheron breeding at the Massachusetts State College
is shown diagrammatically in Fig. 212. Two closely related mares,
Pentoila's Favorite 164611 and Dope's Primrose 163421, born in the
spring of 1920, were purchased from Ohio State University in 1923. The
dam of the latter was the maternal granddam of the former, while the
maternal great-grandsire of both was the same animal Diamant 30018,
and both mares were sired by Libretto 121447. They therefore were
40 per cent r~lated. Both mares were in foal to Colin 166024, a grandson
\ \,
SELEC'l'lON IN HORSES 717
of Carnot 66666, when purchased and dropped filly foals, Lilly and Rose.
These two mares (see Fig. 214) together with their filly foals appeared
to be an excellent foundation for a band of Percherons together with a
good homebred mare Daffodil's Belle.
Ht>wever, the Dope's Primrose line quickly faded out. She was fifth
Futmity filly at the International in 1921, fifth three-year-old mare at
Ohio State in 1923, second aged mare at Eastern States Exposition in
1926. Besides her daughter Rose by Colin born in 1924, she left two
FIG. 213.-Percberon stallion Libretto.
colts by Revelation 181000, called Bay State Reliance 204527, which was
fu·st as a colt in 1930 and as a two-year-old stallion in 1932 at Eastern
States and sold as a stallion, and Jerry, a male born in 1927 and used on
the farm as a gelding. She did not conceive to breedings in 1924, 1925,
1928, and 1930 and bore a dead male foal in 1928. Her daughter Bay
State Rose 186214 bore a male colt in 1928 that was altered. She failed
to breed in 1928 and thereafter was used as a work horse until 1938
when she was sold. Dope's Primrose was sold in 1932. This mare,
although a good individual and purchased as a foundation mare, made
no contribution to the herd, and her line has entirely disappeared.
The other marc, Pentoila's Favorite, has been a fairly successful brood
mare through having left one really good daughter Bay State Lilly
186213 by Colin in 1924. Pentoila's Favorite left 2 other daughters,
Favorell and Novelle, by outside studs, which were not kept because of
deficiencies in type, as well as 2 by Revelation 181000, 7.'iz., Rosalind 'in
1928 and Rosabelle in 1929. Rosabelle dropped 1 filly foal that was
sold, and she herself was sold in 1934. Rosalind dropped 3 dead foals,
also 2 that ,Yent bad behind and 1 filly that was sold. In addition to
FIG. 214.-Foundation mares, Dope's Primrose and Pentoila's Favorite.
these 5 females, Pentiola's Favorite also dropped 5 stud foals, 1 of which

died as a foal, 1 was altered, and 3 were sold as stallions. Pentoila's
Favorite was first-prize filly foal at the Ohio and Indiana State Fairs in
1920, third filly at the Percheron International Futurity in 1921, junior
and reserve champion at Indiana in 1921, first-prize aged mare at Eastern
States in 1924.
Bay State Lilly 186213, a daughter of Pentoila's Favorite born in 1924,
has been one of the most outstanding brood mares at the University of
Massachusetts. She was first-prize mare in 1931 at Eastern States,
first and grand champion in 1932, second-prize mare in 1933 and 1934,
und fhst-prize mare and foal from 1928 to 1933, inclusive. Four of her
sons have. been sold as stallions (Review, Renown, Leander, and B. S.
Laletto)~ 2 were working geldings on the farm (Rob Roy and Let's Go).
\
\
1
SELECTION I N HORSES 719

One daughter, Lillian, which had been first as a filly and yearling at
Eastern States and second as a two-year-old, was sold.
Two other daughters of Lilly, Laura and Lady, by Revelation 181000
have been used as brood mares at the University of Massachusetts.
La.u1·a was third-prize filly foal at Eastern States Exposition in 1931,
first-prize yearling in 1932, and stood 'econd in 1933, first in 1934,
~econd in 1935, fourth in 1936, second in 1937, and first mare and foal
FIG, 215.-Percheron mare B ay State Lill y.
in 1940. One son, Leader, by Lafayette 204281 and one daughter, Lala,
by Laletto 208524 have been sold as breeding animals, and one daughter,
Laurel, by Dragon Jr. 113939 and 2 by Laletto 2085424, Leta and Lau-
rette, were brood mares on the farm. Their type through their rear
quarter and the set of their hind legs, however, resulted in the discard of
this entire line.
The other retained daughter of Lilly was called Lady. She was first
filly foal and junior champion at Eastern States Exposition in 1932,
first in her class in 1933, second in 1934 and 1935. Her first colt in 1935
hy Dragon Jr. was altered, her second and third colts by Laletto in
1936 and 1937, Lena and Lady Lou, are brood mares on t he farm, the
latter having been fu'st-prize two-year-old mare and junior champion at
FLG. 216.-ReveJation aqd his two best daughters, Laura and Lady,
Eastern States Exposition in 1940. Lady was sold in 1938 because of

a bad front foot.
Lena and Lady Lou (full sisters and daughters of Lady) are still on
the farm, and at present all our horses trace to them with -the exception
of two.
Lena, had KonLena and KonLeader in successive years, both sired by
\
- \
lConchapet and also had Konilla by the same stallion. When mated to
Konhysop, she had Lensop, a young mare still on the farm; and when
mated to Koncarno, she had another filly, Carno Carrie, that we had to
destroy after she broke her leg. For the last two years Lena has had
finY'foals by Konhopecar II. The first one, Lorna, ~was Junior Champion
mare at Eastern States Exposition in 1949 and won her class of eight
yearling fillies. The second filly, Lucille, shows a great deal of promise,
and we expect much from this pair of full sisters in our breeding program.
In this same family line, Konilla has had a stallion foal that was sold
as a gelding and has had a filly, Hopeful, by Konhopecar that has been
discarded; she was one of the poorest in size and type of five good fillies
I
by Konhopecar II. Lensop had one filly by Konhopecar II that we
call Hylea. We, therefore, have Lorna, Lucille, and Hylea as potential
female breeding replacements that trace to Lena.
Lady Lou had KonLouis, a gelding, that we disposed of and KonLady,
both sired by Konchapet. KonLady is a brood mare on the farm and
foaled ,vith twins prematurely in 1948. Lady Lou had two fillies by
Konhysop that are both brood mares on the farm; Lousop and Kon-
hysop's Lady. Lady Lou had a stud foal by KonCarno that we altered,
and now she has a promising yearling filly named Louise, sired by Kon-
hopecar II. Lousop had a stud foal by Konhopecar II that was sold
as a breeder, and Konhysop's Lady has a filly foal, Kathleen, sired
by the same stallion, that probably will be retained as a brood
mare.
The third foundation mare, the homebred Daffodil's Belle, has been
an average good brood mare. Her daughter Betsey by Revelation
18lO00 was second-prize foal at Eastern States in 1931, but she and all
her offspring have been sold. The other daughter, Belle, also by Revela-
tion 181000, was not shown. Belle had a daughter, LaBelle, by Laletto
and four other foals. LaBelle is the only one present as a brood mare
on the farm. LaBelle, when bred to Konhopecar II, had a filly, Lana,
that in 1949, as a yearling, was second in a class of eight fillies at Eastern
States Exposition.
So, after a start with the foundation mares in 1923, and using nine
different stallions, we now have two good full sisters-Lena and Lady
Lou-as brood mares on the farm. Both are good-type mares and have
transmitted satisfactorily; so satisfactorily that at present all of our
Percherons, except LaBelle and Lana, trace to Lena and Lady Lou.
LaBelle is a rather plain-bodied mare but has high-quality underpinning.
When she was mated to Konhopecar II, she had Lana, and we are much
pleased with her individuality.
Our younger mares, Konilla, KonLady, Lensop, Lousop, and Kon-
722 BREEDING AND IMPROVEMENT OF FARM ANIMAL~
hysop's Lady are all relatively untried as breeders, and we have much to
learn about their transmitting ability.
We have bred and raised a lot of horses since 1923. Most of these
animals have been used on the farm and have done their work in an
acceptable manner. The stallions have been carefully selected in regard
to their individual type and have cost from $1,000 to $1,500. Revelation
181000 was used for 7 years (1925 to 1932) and sired two very good
offspring, Laura and Lady, a lot of good ones, and some poor ones. He
was an imported horse that stood second to Hesitation, the grand cham-
I·'w. 217.-Lafayette. the grand champion Percheron stallion. Eastern States Exposition,
1933. Owned by Massachusetts State College.
pion at the Internat.ional in 1923. He was grand champion at the

Eastern States Exposition in 1925, 1920, 1927, and 1930.
Lafayette 204281, a son of Sir William 168112, he by Laet, was used
from 1932 until his death from coronary thrombosis in 1935. He was
grand champion at the Eastern States Exposition in 1933 and 1934. He
was an excellent individual but left no impression in our herd, leaving
only one daughter which was culled.
Laletto 298524 was purchased in 1935 and used until the purchase in
1945 of Konhopecar II. Like Lafayette 204281, Laletto is a grandson
of Laet 133886, and his dam is by Libretto, the sire of the two originally
purchased mares. There is, therefore, just a slight trace of line breeding
to LibrettI;) 121447 in the pedigrees of the daughters of Laletto at present
in the h'iJrd. Laletto was third-prize aged stallion at Eastern States in
1936 and 1937 and senior champion in 1941.
Konchapet, Konhysop, and Koncarno, all sons of Koncarcalyps, bred
by Madrey Farms, were used successively in ow' herd. We purchased
Konhopecar II (by Koncarcalyps and out of Caruona IV's Hope) in
1945 as a yearling. He is a full brother to the Lynnwood Farm's horse,
Ktlncarhope, that was grand champion at the National Perchel'on Show
in 1947. We sho-"ved Konhopecar II to the grand championship in 1949
at Eastern States Exposition, where he defeated another good son of
Koncarcalyps, Koncaruo, and also a good son of Topper, owned by
FIG. 218.-Percheron stallion Laletto.
Floyd Hill of North Woodstock, N.Y. We are expecting a great deal

from ·the get of Konhopecar II. His yearlings and foals at present show
every indication of general improvement over our brood mares.
, As is indicated in the above discussion, some of the female lines are
better than others. A glance at Fig. 212 shows that practically all the
animals represented trace to old Pentoila's Favorite. The job of breeding
and improving horses is not an easy one. The story has been told here
for the purpose of illustrating the fact that rapid progress in the breeding
of a species of animals that reproduce at such a slow rate as does the
horse is hardly to be expected.
Selecting the Fillies to Save.-Only after a breeder has made a close
study of the mares in his own band with suitable records and tabulations
of their strong and weak points is he in a position to start on his search
for a stallion to mate with them. It is obvious, of course, that the male
should be selected both on the basis of holding the desirable qualities
which the breeder already has and also, and more especially, of bolstering
the weak points shown by the mares. The mares at the University of
Massachusetts have from the beginning been a little too light in IDOlle
and a little too shallow to suit us. In addition there is a tendency for
the hind legs to be a little close together and not to set under the horse
as much as they should. In our search for our next stallion, these points
will receive major attention.
In selecting mares or fillies either from outside or in one's own herd,
acceptable type for the class of horse concerned should, of course, be the
first criterion regardless of the fact that mares do not always breed
exactly as they look. In other words, one has a greater expectancy of
getting drafty offspring from draft mares than from Thoroughbreds, and
VIce versa.
In addition to acceptable type, one should try to select from good
female lines in which as many as possible of the animals have been
Ol"lhe aeslrea type. An obvious weak point in the University of Massa-
chusetts horses, as discussed above, is the fact that Pentoila's Favorite
had 4 mediocre daughters and 1 good one, Lilly. Lilly in turn had 2
mediocre daughters and 2 good ones, Laura and Lady, and the latter, a
somewhat poorer individual, is evidently transmitting in a more accept-
able fashion than her somewhat superior typed sister, Laura. In this
whole female line, there are evidently both good and bad genes, the
former not being so numerous :::s one might wish. It is very important,
therefore, that both the direct ancestors (those occurring in the pedi-
gree) and the collateral relatives (half brothers and sisters, cousins,
aunts, uncles, etc.) receive some consideration. Because, generally, only
the most acceptable animals are saved and used for breeding purposes,
the pedigree of the direct ancestors of an animal is bound to look pretty
good and will tell us, in general, the best that we can expect. The
collateral relatives, on the other hand, which do not generally appear in
the pedigree or, if included, contain only the desirable references, will
tell us some of the bad things we may expect by selecting their family.
The pedigree of the University of .Massachusetts filly Leta can be made
to look very good, as is shown in Fig. 219. _
..The pedigree (Fig. 219), with some of the better show winnings listed,
looks 7ery good. How about some of the collateral relatives? Pentoila's
Favorite had 1 good daughter Lilly and 4 not very good ones; 1 good son,
2 or 3 average ones, and 1 poor one. Lilly had 2 good daughters and
2 mediocre ones, 6 sons ranging from good to poor. Revelation had
quite a ~ahge of offspring, making his total breeding value about average.
\
Laura's offspring on the whole (4 daughters and 1 son) have not been
nearly so good as their dam in draft quality. Laletto has sired a range
of offspring from very good to very poor. Leta, the filly pedigreed in
Fig. 219, is a fairly good individual, but she is not so good as Fig. 219
mal~es her appear, and, although she has some good genes, it would seem
from a study of her collateral relatives that she probably also has a lot
of poor ones and would probably not be a particularly successful breeding
Laletto
2d prize Ohio, Indiana
1935, 3d prize East-
ern States Exposi-
tion 1936, 1937
Revelation
Leta
2d International 1932
1st prize foal
Grand champion East-
1937
ern States Exposi-
tion 1925, 1926, 1927,
Laura 1930
1st Eastern States 1932
2d Eastern States 1933
1st Eastern States 1934 Colin
1st Eastern States 1940
Lilly
1st prize Eastern States Pentoila's Favorite
1931 1st filly Ohio and
1st and grand cham- Indiana 1920
pion Eastern States Junior and reserve
1932 champion Indiana
1st mare and foal 1928-- 1921
1933, inclusive 1st mare. Eastern
States Exposition
1924
FIG. 219.-Good-looking pedigree of Leta.
or transmitting mare. Complete pedigrees in terms of record of per-

formance of both direct and collateral relatives can be an important
aid to selection based first of all on individuality.
The real test of a breeding animal is, of course, its ability to transmit
its own--cleSirable qualities. This is the progeny test, which should be
utilized wherever"possible to do so. If we can find good individuals,
whose complete pedigrees will bear the light of careful scrutiny and who
have demonstrated their ability to transmit their own good qualities, we
are on perfectly sound ground in making our selections, although we have
yet to learn how this triply desirable animal will" nick" with one of our
own. Obviously, it is not always possible to get all this information,
in which case a breeder should get all the information available and then
"cross his fingers" until he sees how things turn out.
In the selection of brood mares, it is essential that attention be given
to their probable and actual ability to bear young over a long period z'
of years. Their probable ability in this regard can best be judged born (
their complete pedigrees (both direct and collateral relatives). In other
words, in the selection of mares one should select only in families that
have a high natural rate of fertility. Like all other attributes having a
genetic basis, this varies among different families in all the classes of
livestock. Demonstrated abilities by the direct and collateral relatives
is the best guarantee that a breeder can get for any new animal going
into his band of females.
Actual ability to reproduce can be ascertained from the performance
of older mares as well as from current examinations of their genital
tracts by competent veterinarians, and the latter type of examination
can and should be used on young mares. .
Finally, in the selection of mares or fillies that hard-to-describe quality
of femininity should receive considerable emphasis. The general term
for it is probably quality, as evidenced in fine texture of bone (different,
of course, with different types of horses) and fineness of hide and hair,
together with indications of alertness and intelligence and a certain
nobility of carriage and demeanor. The mare should be well balanced
throughout, with a roomy middle and ample width through the hips and
croup, and must have a sound, well-developed udder.
Selecting the Stallion.-The s~fest plan in securing a stallion is
i
obviously that of getting one that has already demonstrated his ability
as a sire in the type and performance of his get, in other words, a proved
sire. Unfortunately not many good, proved stallions are available at
any price. Most breeders, therefore, will have to select and use young,
unproved studs.
These should be healthY, of the desired type and disposition from a
naturally fertile strain. Preferably they should be by a proved stallion,
the majority of whose offspring were desirable and out of a daughter of
a good proved stallion, said daughter having preferably demonstrated
her ability to transmit desirable qualities to her offspring. Admittedly,
such animals are scarce, complete records of ancestors not generally
available, etc. Nevertheless we must make a start sometime in getting
such information, and it is unlikely to become available until prospective
buyers start asking for it.
Just as indications of femininity are desired in the female, so do we
desire indications of masculinity in the stallion. Here we expect a little
more compactness and scale, a little heavier bone, a well-developed
forequarter and chest. We want spirit and masculinity in our stallions
but of a sort that proper training has rendered obedient.
Sire Indexes in Horses.-As yet few objective measurements of
performance that might serve as a basis for an index have been developed
f&r'the horse. H. H. Laughlin has evolved a mathematical yardstick
for measuring racing capacity in the Thoroughbred through a study
of the breeding and racing records of some 10,000 of these animals. He
has measured quality of performance for a single race on a definite
mathematical scale by developing correction factors for sex, age, weight
carried, distance run, and speed, on the basis of actual performance of
this breed under varying conditions and yielding standards that are the
"smoothed best" which the breed has accomplished up to the present
iime. By means of such standards the true racing capacity of any horse
can be measured, and, if he runs a sufficient number of races under
standard conditions, the one figure for racing capacity (derived from
the several "quality-of-performance" figures for separate races) can be
compared with a similar figure for any other horse, and the question as
to which was the better racing horse can be definitely answered, although
the two horses may have never raced against each other.
When the racing capacities of a male and female together with their
(lose, direct, and collateral relatives can be ascertained, then it is possible
to formulate a futurity index or hereditary-promise level for their off-
spring. Because racing capacity involves the whole organism, it is
reasoned that multiple genes are concerned, and therefore equal ,,,eight
is given to both the sire and dam in arriving at the futurity index of their
offspring. In this system the breeding factor of the sire is obtained by
taking one-third of the average racing-capacity figure of his sire and darn
plus one-third of the racing capacity of the sire himself plus one-third of
the average racing capacity of the foals that he has already sired. The
breeding factor of the dam is arrived at by a similar procedure, and the
average of the sire's and the dam's figures is called the futurity index of
their potential or actual offspring. This" prediction-index, 'when it
contains only a few ancestors, each doubtfully stressed, gives a low
prediction-value; but when it comprises a highly representative group
.. of close antecedent blood-kin, each properly stressed, then the prediction-
value of the specific formula is high."l
It would, of course, be possible to work out similar prediction values
for any type of racing horse, where the one criterion for selection is that
of speed, or for show horses, where the criteria are type, action, and
I LAUGHLIN, H. H., Raeing Capacity in the Thoroughbred Horse, Sci. Monthly,
28:210,310,1\)34.
,0»
OJ
manners, provided accurate and complete records of performance are

~vailable.
Phillips, Brier, and Lambert have made some preliminary studies
of tests designed to measure the speed and efficiency of light horses.
The tests were designed to measure speed, both at the walk and tre>t'in
carriage and at the trot under saddle, length of stride, and respiration
and heart rate before and after exercise. 1
From the results it is evident that tests designed to measure speed, length of
stride, and respiration and heart rate are subject to rather wide variations when
applied to the same horse at different times. It is apparent~ therefore, that, if
such tests are used for the purpose of detecting differences between horses, the
tests must be repeated several times under carefully controlled conditions. How-
ever, the degree of accuracy required will depend upon the magnitude of the
differences that it is desired to detect. In most of the analyses presented here,
the variability of the material used (i.e., of the 14 horses) relative to the various
sources of error was high enough that the observed differences between horses
were significant. In a more uniform group of horses it would probably be neces-
sary to improve the accuracy of the test in order to differentiate with any degree
of certainty between animals with respect to any given character.
It should be emphasized that the results presented in this bulletin deal only
with the degree to which results of each of the tests agree when obtained on the
same horses at different times. Before these tests, or any other tests of perform-
ance, should be applied generally, careful studies are needed to determine the
relationship of the tests to actual performance under practical working conditions.
Although the machine has replaced the horse as a source of draft

power to a very considerable extent, there is unmistakable evidence that
interest in light horses for pleasure and sport is on the increase. No
better safety valve for the effects of our high-speed, machine civiliza-
tion can be imagined. It is to be hoped that much of the former breed-
ing of draft horses in our farms can be gradually replaced with the breed-
ing of light horses. This is a new field which has opened up in the past
few years and one that seems to hold much in the way of possibility,
especially for farmers located near the large centers of population.
In the draft horse, we also need objective tests for measu.ing working
abilities. These animals exist for the one purpose of doing useful work.
As stated by Phillips, Krantz, and Lambert, 2
1 PHILLIPS, R. W., BRIER, G. W., and LAMBERT, W. Y., A Study of Some Problems
Involved in Measuring Performance in the Horse, U.S. Dept. Agr. Bur. Anim. Ind.
Mimeographed Rpt.
2 PHILLIP~, R. 'V., KRANTZ, E. B., and LAMBERT, 'V. Y., The Accuracy of :VIeasure-
:nents and Shores of Draft Horses, A mer. Soc. of Ardm. Prod. Proc., 1938, p. 82.
\
SELECTION IN HORSES 72U
Objective measures of the characteristics which contribute to maximum ability
to do work are badly needed. Such measures might include maximum pulling
ability for a short period of time, staying power with a heavy load over long
periods, walking speed with a standard load over a given distance, and ability to
lliIaintain weight doing a given amount of work with a given allotment of feed,
alo~g with careful records of the occurrence of soreness and unsoundnesses.
When such methods are developed and a sufficient number of records is obtained
under standard, experimentally controlled conditions of training and testing, and
the animals used have been scored and measured at comparable ages, it will be
possible to arrive at some estimate of the value of any given measure or score.
The dynamometer is available and has been used for many years to
get the maximum puliing ability of horses for a short period. Records
of these contests together with measurements of the animals competing
have shown definite correlations between height and weight and ability
to pull, but no other measurements have shown definite positive corre-
lations. "Staying power with heavy loads over long periods" and
"walking speed \vith a standard load over a given distance" might also
be secured by means of the dynamometer.
So far as the writer knows, the dynamometer has not as yet been used
in any constructive manner as an aid in the selective breeding of draft
horses. The dynamometer contests 1 are a very interesting sporting
event and generally draw a much larger crowd than does the orthodox
judging of horses on the basis of type. In fact, it has happened more
than once that the type judging has had to be temporarily suspended
until the breeders and showmen of the purebreds have returned from
watching the pulling contest. In other words people in general are more
interested in seeing what a horse can do than they are in admiring him for
what he may look like.
Type in Horses.-Horses have been select~g_9n the basis of the type
best suited to perform given tasKs--ror--liuudreds of years. Thi~-has led
gfiUlualli-to the c;~ation of distinct and separate types for racing, riding,
driving, and pulling. Selecting the speediest horses gradually produced
the racehorse type, selection for power gradually produced the draft-
horse type. The speed and saddle types are now definitely established,
and there seems to be little prospect that they will be changed in any
fundamental way. The task of the future in these types, then, is further
refinement to add slightly to their speed, symmetry, balance, action,
and behavior; and by means of keener selective tools based on records of
performance, measurements, and progeny tests to render them more
homozygous for the respective qualities desired.
1 See PHILLIPS, R. W., MADSEN, M. A., and SMITH, H. H., Dynamometer Tests of
Draft Horses, Utah Agr. Expt. Sta. Cir. 114, 1940.

730 BREEDING AND IMPROVEMENT OF FARM ANL11ALS
The draft horse was created by selection for the particular purpose of
moving heavy loads at a fair speed. For this task the criterion was the
greatest weight possible in the smallest compass. The heavy draft horse
was a remarkably efficient animal for the job he was supposed to do.
The railroads, trucks, and tractors have made their appearance dul'in-g
the past 100 years, however, to challenge the horse in this field, and
largely because of their greater speed they have supplanted him so that
the market for the ton-upward horse has shrunk almost to the vanishing
point. Draft-horse type, in other words, is changing to a somewhat
smaller, more active model, with greater quality. In particulars, the
type will remain what it always has been, a short, compact, well-set-up,
and straight-moving type, but, in place of weighing from a ton upward,
he will weigh from a ton downward. He will be a medium-sized draft
horse, since the city market for the heavy sort has disappeared, the
farmer never did like the extremely big ones, and in so many instances
now power machinery has to do the very heavy work. The new type
will stand 15:3 to 16:2 hands, the stallion weighing 1,900 to 2,100 lb.,
the mares 1,700 to 1,900 lb. in good condition.
There are two ways of creating such a horse, one by mating large
stallions to small mares (or small stallions to large mares) and the 'other
by mating medium-sized stallions to medium-sized mares. If the former
method is used, and we put in the two extremes, we will generally get
the mean in the immediate offspring, but the mixed inheritance of these
animals can be expected to come out in all sorts of combinations in their
offspring. In other words, we can expect to have to do a lot of weeding
out if such a plan for creating the medium-type horse is used. Because
the average run of farm mares tends to be on the small side, there would
still seem to be a place for the good-sized stallion of good quality.
The other plan of mating medium size to medium size should give less
extreme variations and lead more quickly toward the desired goal.
With the system either of mating unlikes to produce a medium or of
mating intermediates in types, it goes without saying that selection must
be practiced and the offtype animals discarded. Less of this will be
necessary with the matings of animals that resemble each other closely
in general type. One should not deceive himself, however, by thinking
that two animals which are similar in appearance are also similar in their
genetic make-up. Many different combinations of genes may give the
same general phenotypic appearance. We must remember that, in
dealing with commercially important characters of our animals, we are
probably dealing with hundreds or thousands of genes. This is very
different from crossing a rough, white guinea pig (RR bb) and a smooth,
black one (rr\BB) and getting rough, black offspring (Rr Bb). And even
\
\
\, \
in this simple case, dominance interferes with our progress, because we

cannot distinguish offhand between the following rough, black guinea
pigs (RR BB), (RR Bb), (Rr Bb), (Rr BB). We have no conception of
the dominance, partial dominance, epistatic, and other probable interrela-
fions involved in the thousands of genes in our larger animals. There
are many combinations of coins that will make a dollar, e.g., 10 pennies,
4 nickels, 2 dimes, and 2 quarters, or 5 pennies, 3 nickels, 3 dimes, and
1 half dollar, etc. Likewise many combinations of genes may give the
same phenotype.
If animals are to breed true, they must be made homozygous in the
bulk of their genes. Mating animals which are alike phenotypically
will have little effect in making them homozygous in their genes, and
likewise mating unlikes will have little effect in making them more
heterozygous in their genes. In short, whatever system is adopted for
attempting to create a true-breeding strain of intermediate-sized draft
horse must involve both selection and some form of inbreeding in order
to render the animals relatively homozygous or in the breeder's language
to "fix the type."
This whole business of changing the type (weight) of our draft horses
means many a headache for the judge in the show ring. It is going to
take real courage to turn down a good big horse in favor of a good medium-
sized one. The big one "will excel in weight and draft power, the smaller
one to offset this should be expected to excel in action and movability.
Care must be exercised too that the desire for somewhat less weight than
formerly does not express itself in either a chunk, easy keeping to be sure,
but generally lacking in movability, or in an upstanding, light-boned,
and light-middled animal ,Yhich is apt to be a hard keeper, lacking in
power and stamina, although probably excelling in freedom of action or
movability. It is in just such situations as this one involving a change
in type that the art of the breeder has its best opportunity of expression.
If the art is backed up by or based upon a sound knowledge of the genetic
principles involved, it will eventually evidence its inherent soundness.
Art alone, however, is not enough (or it works too slowly with too many
mistakes); nor is a knowledge of principles in itself enough. For con-
structive breeding at a tempo to suit the modern age both art and science
are essential.
Breeding EfficieIl~y.in Horses.-In order that the breeder may have a
clear pictu-re" of th~ breeding efficiency of his stallions and band of mares,
careful records should be kept of the occurrence of heat periods in the (
mares, number of services required to settle each mare, percentage of '
mares settled by each stallion, birth weight of foals, duration of gestation, _j
etc.
A stuilyl of the breeding performance of 43 mares at the United States

Morga " Horse Farm, Middlebury, Vt., from 1928 to 1938 shows that
132 mares conceived, foaled, and raised their foals.
10 mares conceived, foaled, but foals died before weaning.
6 mares conceived but aborted. • oj)
73 mares failed to conceive.
221 total mares bred.
Summarizing, we see that 221 breedings produced 132 foals which
lived-a breeding efficiency on the part of these mares of 59.7 per cent
in terms of livable foals.
Morgan Horse Breeding by the U.S. Department of Agriculture.-The
United States l\Iorgan Horse Farm was established at Middlebury, Vt.,
in 1907, as a result of a gift from Col. Joseph Battell. The farm at present
is composed of approximately 1,000 acres. The primary object of the
farm is to conduct research that will yield results of value to all horse
breeders and users. Since ::\Iorgans were already on the farm, it was
logical that the "'ark be done with them. The use of Morgans is inci-
dental to the main objective of conducting research that ~will yield infor-
mation of value to horsemen. Pasture improvement, research, obser-
vations on reproduction, devices for training and testing are studied and
data collected and analyzed.
Performance testing was initiated in 1940 with the horses in an attempt
to learn more about the suitability of the individual, line, or sire group
for pleasure purposes, riding, or driving. Objective measures-qualita-
tive and quantitative-have been made and some little data collected
as results of the tests. At Middlebury the horses are scored for con-
formation, general appearance, feet and legs, and action at one year,
two years, and three years of age. All horses are measured for such
things as height at the withers; depth of chest, heart girth; circumference
of the fore cannon, rear cannon, etc. Weight records are maintained
from birth up to three years of age on all horses.
The actual performance tests are designed to give a definite measure
of speed for a mile at the normal ,,'alk and trot and the length of stride
of these gaits, both under the saddle and in the harness. Each horse
carries 20 per cent of its body weight under the saddle and pulls 60 per
cent of its body weight in the harness. These tests are made on a ~~-mi.
track.
Endurance is measured under the saddle on a cross-country ride of
about 11 mi. Approximately 50 per cent of this distance is covered
1 LAMBERT, W. V., SPEELMAN, S. R., and PHILLIPS, R. \Y., The Reproductive
History of the Stud at the United States Morgan Horse Farm from 1928-1938, Amer.
Soc. Anim. PrOd. Proc., 1939, pp. 358-365.
••
at the trot, 10 per cent at the canter, and 40 per cent at the walk. In
addition to the time record, each horse is scored for ease of handling,
response to commands, degree of fatigue, and ease of gaits by the rider.
Recuperative power is measured in harness by a test on the track at
th~ trot for 5 mi. This is done by studying data on respiration and
heart rates as well as time elapsed in completing the distance.
Results to date on the above types of performance studies are an
approach to the many problems involved. It is obvious that there is
variation from year to year and variation due to riders and trainers.
Every attempt is being made to standardize these environmental
influences as much as possible. Temperament or the "desire to go"
undoubtedly has considerable effect on the performance of the individual
horses. It has been found that temperature and humidity also affect
results greatly, particularly the endurance tests. Very many con-
clusions cannot be made until some of these variable factors are controlled
even more than at present.
These performance tests and correlative studies involving them and
other measurements are a complex problem. The results over the next
few years should be of value to all breeders of horses for saddle purposes.
Summary.-We have seen in this chapter that in spite of drastic
reductions in the numbers of horses in the United States during the
past 30 years, due to the inroads of various forms of mechanized power,
the horse is still very much with us as a source of power on the farm
and is filling an ever larger place in pleasure and sport. In the past,
selection of horses has been based almost_ \Vholly 0rt individuality.
Because inheritance in horses follows the one basic pattern found in
all animate forms of life, we have now come to realize that sound selection
must consider records of performance of direct and collateral relatives
and of the animal itself in addition to the consideration of individuality.
We are badly in need of practicable measures of performance in the
horse. Only when these have been devised and put to work in the
securing of essential records, and the records used intelligently, together
with continued full considerations of individuality, will our selection in
this class of livestock be on a sound footing. When the above methods
have revealed the inherently better strains of horses, we will face the
further task of intelligent line- and closebreeding, so that we may create
fairly true-breeding strains.
References
Books
BROWN, W. R. 1929. "The Horse of the Desert," Derrydale Press, New York.
CARTER, W. H. 1923. "The Horses of the World," The National Geographio
CREW, F. A. E., and SMITH, A. D. B. 1930. "The Genetics of the Horse," Biblio-
graphia Genetica, VI, The Hague.
DuHAYS, C. 1886. "The Percheron Horse in America," Orange Judd Publishing
Co., Inc., New York.
HAMMOND, J. 1940. "Farm Animrus, Their Breeding, Growth and Inheritance,"
Longmans, Green & Co., Inc., New York. • ~
HARIUS, W. J. 1934. "The History of Bourbon King 1788," The Judson Company,
Cleveland, Ohio.
LINSLEY, D. C. 1860. "Morgan Horses," C. M. Saxon, Barber and Company, New
York.
NORBY, J. E., and LATTIG, H. E. 1937. "Horse: Selecting, Fitting and Showing,"
Interstate Printers and Publishers Co., Danville, Ill.
SANDERS, J. H. 1893. "Horse Breeding," J. H. Sanders Publishing Company,
Chicago.
SKINNER, J. S. 1856. "The Horse" (Youatt), Blanchard and Lea, Philadelphia.
SUSAUNE. 1932. "Famous Saddle Horses," The Farmers' Home Journal Co.,
Louisville, Ky.
U.S. Department Agriculture. 1942. "Diseases of the Horse," U.S. Government
Printing Office, \Yashington, D.C.
WALL, J. F. 1936. "Practical Light Horse Breeding," The Monumental Printing
Company, Baltimore.
WOODRUFF, H. 1871. "The Trotting Horse of America," J. B. Ford and Co., New
York.
ANDERSON, W. S., and HOOPER, J. J. 1917. American Jack Stock and Mule Pro-
duction, Ky. Agr. Expt. Sta. Bul. 212.
BRANDT, A. E. 1927. Relation between Form and Power in the Horse, Arner. Soc.
Agr. Engin. Trans. 21(PM):3 to 4.
BRODY, S., and TROWBRIDGE, E. A. 1937. Efficiency of Horses, i\ien, and Motors,
Mo. Agr. Expt. Sta. But. 338.
CALDER, A. 1927. The Role of Inbreeding in the Development of the Clydesdale
Breed of Horses, Roy. Soc. Edinb. Proc., 47 :118-140.
DAWSON, W. M. 1934. The Pulling Ability of Horses as Shown by Dynamometer
Tests in Illinois, ArneI'. Soc. Anim. Prod. Proc., pp. 117-121.
FULLER, JAMES G. 1931. Horses for the Farm, Wis. Agr. Col. Ext. Cir. 244.
HARPER, M. W. 1921. Raising Colts, N.Y. (Cornell) Agr. Expt. Sta. Bul. 406.
HARVEY, A. L .. 1936. Using Horses on the Farm, Minn. Agr. Expt. Sta. Spec. Bul.
145.
HUDSON, R. S. 1939. Guides for Horse Buyers, Mich. State Col. Ext. Bul. 197.
Horse Association of America. Horses, "Mules-Power, Profit (1934).
Horse and Mule Power (1937).
1930. Making History with Horses, Mich. State Col. Spec. Bul.
Learn to Judge Your Horses and Mules, Leaflet 196 and many others,
Wayne Dinsmore, Secretary, Union Stock Yards, Chicago.
JOHNSTON, P. E., and WILLS, J. E. 1933. A Study of the Cost of Horse and Tractor
Power on Illinois Farms, Ill. Agr. Expt. Sta. Bul. 395.
LAMBERT, W. V., SPEELMAN, S. R., and PHILLIPS, R. W. 1939. The Reproductive
History of the Stud at the United States Morgan Horse Farm From 1928 to 1938,
reprint from 32d Amer. Soc. Anim. Prod. Proc.
'.
SELEC'1'lUN IN HORSES 735
LAUGHLIN, H. H. 1934. Racing Capacity in the Thoroughbred Horse, Sci. Monthly,
38 :210 and 310.
NEWELL, PAUL F., and GOODELL, C. J. 1938. '~'ork Stock Feeding, Management
and Production, Miss. Agr. Col. Ext. Bul. 95.
PJilT,TET, Z. R. 1933. The Farm Horse, U.S. Dept. Com., Bur. Census.
PHILLIPS, R. W., BRIER, G. W., and LAMBERT, W. V. A Study of Some Problems
Involved in Measuring Performance in the Horse, U.S. Dept. Agr., Bur. Anim.
Indus., Anim. Hush. Div.
- - - , KRANTZ, E. B., and LAMBERT, W. V. 1938. The Accuracy of Measure-
ments and Scores of Draft Horses, Amer. Soc. Anim. Prod. Proc., pp. 77-83.
- - - , MADSEN, M. A., and SMITH, H. H. 1940. Dynamometer Tests of Draft
Horses, Utah Agr. Expt. Sta. Cir. 114.
- - - , SPEELMAN, S. R., [l,nd \VILLIAMS, J. O. 1942. Horse Breeding Research at
the United States Morgan Horse Farm, Vt. Horse and Bridle Trail Bulletin. Jan.
RIGGS, ELMER S. 1932. The Geological History and Evolution of the Horse, Field
Mus. Nat. His., Chicago, Leaflet 13.
ROMMEL, G.::\1. 1908. The Preservation of Our Native Types of Horses, U.S.
Dept. Agr. Bur. Anim. Indus. Cir. 137.
SCHWARTZ, B., IMES, M., and \VRIGHT, ,Yo 1936. Parasites and Parasitic Diseases
of Horses, U.S. Dept. Agr. Cir. 148.
SOLANET, E. 1930. The Cirollo Horse of South America, Jour. Hered., Vol. 21,
No. 11, Nov.
'WILLIAMS, J. 0., and JACKRON, W. 1936. Improving Horses and Mules, U.S.
Dept. Agr. Yearbook, pp. 929-946.
CHAPTER XXIV
RETROSPECT AND PROSPECT
We have now completed our discussion of the principles underlying

the creation of better livestock. We tried to start at the beginning, to
get some idea of the probable origin of life on the planet Earth something
over 1 billion years ago. We concluded that something comparable to
the gene must have appeared fairly early in this process. These entities,
capable of certain reactions among themselves and with their immediate
environment, the cytoplasm of the cells, and endowed with the power of
exactly reproducing themselves,' seem to have been proved without
reasonable doubt to be the mechanism of inheritance in all living forms
today. .
From the point of view of logic, it seems probable that the genes have
always performed this function. In the earliest stages, there perhaps
was formed wave after wave of the simplest sort of living material that
flourished for a short time and then quickly faded out of existence as
living things without leaving offspring. At some stage, however, we
must suppose that bits of this living material in some way became
endowed with the property of reproducing themselves, so that as the
old died, the new was born. Once started and with the occasional pro-
duction of something a little different from the parent form we have the
basis for an evolutionary process. At present, so far as we know, a
new and somewhat different living thing can arise only from other living
things by means of the genes. They are today the fundamental mecha-
nism of life, as evidenced through reproduction. It seems likely that
they have always fulfilled this role.
We hurriedly traced the process of evolution until it finally gave rise
to the larger farm animals and to man himself. We sketched the story
of man's progenitors until modem man, Homo sapiens, appeared on the
scene some 25,000 to 50,000 years ago. Then we saw man take one of
his greatest forward strides some 8,000 to 12,000 years ago when he began
to domesticate plants and animals, which better ensured his food supply.
We then sketched the progress that man has made through historic
time in breeding, selecting, and perfecting his. animals for specific pur-
poses, and finally we find ourselves in the twentieth century with well-
nigh perf~6t individual animals in all the classes of livestock, but with
" 736
RETROSPECT Al'·.jD PROSPECT 737
the average type and performance of our livestock well below an efficient
and profitable level.
Our next task was l;hat of learning some of the details of the workings
of the intricate process of reproduction. In the lower forms of life,
tlris. process is relatively simple consisting merely of fission or division
of the parent into two offspring. Next came budding, one or a few
unspecialized cells having the power of creating a new individual. In
time, specialized cells were developed for the reproductive function.
Eventually, they came to be specialized in different ways, the union of
two differentially specialized cells being necessary for the creation of a
new life. Originally, these different cells were produced by the same
organism, but they finally came to be produced by different members of
the species, which gave rise to the sexes. At first, these male and female
sex cells were passed to the exterior, where fertilization and growth took
place. As organisms grew in complexity, so did the generative apparatus,
so that in the higher forms of life we have, besides the glands producing
the sex cells, a complicated mechanism of accessory glands and organs
under the control of distantly located endocrine glands, with fertilization
and embryonic development taking place internally, the latter proceeding
for many months and resulting eventually in the birth of a completely
organized, though somewhat helpless and dependent, individual.
The researches during the past half century, and particularly during
the last decade, have done much to explain the reproductive processes
and the factors which limit maximum reproduction.
One of the most amazing technological advances in animal production
has been the development, of techniques for, and the rapid acceptance
by livestock producers of artificial insemination. The use of this method
on a few thousand dairy cattle in the United States in 1938 had been
extended to nearly 3 million in 1949, and it appears that, in some intensive
dairy areas, artificial insemination may soon be the predominant method
of breeding.
During the 10-year interval that the artificial insemination program
has been developing in the United States, the methods of processing and
utilizing semen have been greatly improved. Earlier estimates that the
semen of one bull could be used to service 1,000 cows annually have been
revised upward to 10,000 cows. Semen was at first diluted 1:4 and
transported within a radius of 20 to 30 mi. The regular dilution rate is
now 1: 100, and experimental work indicates that it may be extended to
1 :400 or even higher. Semen is now being transported between states
by bus, rail, and air on a regularly scheduled basis.
Much progress has been made in storing sperm, but the maintenance
of maximum fertility in stored semen leaves much to be desired. As
738 BREEDI NG AND I MPROVJ<J1I1J;JN'l' OJ!' Jl'A_RM A_NIMALS
yet few samples of bull semen give satisfactory fertility when stored
for more than 3 or 4 days and other farm animals even less. The fact
that sperm motility, but not fertility, can be maintained for many months
in quick-frozen semen; that sperm retain their fertilizing capacity for
up to 5 months in the female reproductive tract of the bat; and th",t
FIG. 220. -The "targets" at which Holstein breeders are shooting. Such models shoul d
prove invaluable in fixing type.
the queen bee can produce fertile eggs for 3 to 4 or even up to 7 years,
indicates to the physiologist that there are broad horizons ahead.
The endocrine regulation of the testis is fairly well understood, but
it is as yet impossible to completely regulate fertility in the male by
artificial means. Too many sires are deficient in reproductive capacity.
Whether their limitations are genetic, physiological, pathological or
environmental remains to be learned. One of the important questions
before attempting to restore fertility in males is whether or not the
\
\
••
HE'PROSPECT AND PROSPECT
cause is heritable. We cannot stress too much that breeding animals

must first of all be of high fertility.
In the female the widespread use of the principles of breeding manage-
ment now at hand would increase the breeding efficiency of most herds
a~d. flocks. Unfortunately, 100 per cent efficiency is seldom achieved
at present. One of the most puzzling problems is the failure of « appar-
cntly normal " animals to conceive.
FIG. 221.-May Rose II, foundress of t he May Ito e family of Guernseys, noted for type
as well as production. (Courtesy of American Guernsey Cattle Club.)
As new information and techniques become available, the diagnosis

of reproductive disturbances becomes more exact. Difficulties which
were previously lmrecognizable are now becoming known, and it is hoped
that methods of correction will follow.
The recovery and transfer of fertilized ova from one rabbit to another
has been repeatedly accomplished for nearly three-quarters of a century.
Much interest in the possibilities of utilizing this technique in cattle is
being shown. It is well known that the ovaries of cattle, sheep, and
other species can be caused to produce ova in greatly increased numbers.
It is likewise known that many of these ova do not develop normally.
As _described in previous chapters great numbers of fertilized ova may
be produced by appropriate endocrine treatment, and in a few instances
fertilized eggs have been successfully transferred from one sheep to
another. These techniques are far from the practical stage in 1949,
but much optimism, especially in the popular press, is being shown.
A.Ttificial parthenogenesis has already been achieved in one mammal,
the rabbit. Whether or not a scientific procedure for accomplishing this
in the larger animals will ever be achieved remains to be seen. WHat
a boon it would be, for instance, in cattle if a particularly good cow could
be made to reproduce parthenogenetically, provided, of course, that she
herself was relatively homozygous for the desirable genes, because we
·FIG. 222. -The Guernsey bull Foremost Prediction bred by Emmadine Farm, Hopewell
Junction, N.Y., owned by McDonald Farms, Cortland, N.Y. (Photo(Jraph Courtesy of
Strohmeyer and Carpenter.)
presumably ,,,ould still have the sampling feature of inheritance as

reduction took place in the formation of the mature egg.
The rapid development of endocrinology has done much to explain the
basic principles of growth, reproduction, lactation, egg production, and
many other economic charactel·s. The l'elationships of the genes and
hormones are only beginning to be understood. In· any complex char-
acter such as milk production, it seems logical that nearly every organ
system contributes either directly or indirectly to the process. In some
cases, for example, the thyroid gland may limit maximum milk pro-
duction. In this case the administration of the thyroid hormone may
have a stimulatory effect. In the same way growth might be limited by
the anterior pituitary, thyroid, or gonads. If it were known that the
thyroxine seyretion of a yOlmg animal was suboptimum, the administra-
\
:;. .,
;
•
RETROSPECT AND PROSPEC1' 741
tion of this hormone might have a stimulatory effect on growth. With
an increase in knowledge it may eventually be possible to correct such
deficiencies or limiting factors. The cause of such difficulties should
not be overlooked. If the deficiency has an environmental cause much
:nay be gained by its correction. If the deficiency is primarily ~enetic,
there is considerable question as to whether such lines should be propa-
gated. It is evident that in man numerous characters that would be
lethal or semilethal under natural conditions are constantly being repro-
duced. The livestock breeder can practice much more rigid selection
than man seems inclined to accept for himself. Deviation from such
standards should be given careful thought.
Tremendous strides have been made during the last 50 years in unrav-
eling the heretofore puzzling and seemingly mysterious ways of hereditary
transmission. Although the exact nature of genes is still unknown,
experimental inquiry has yielded a wealth of information as to the varied
nature of their reactions in determining the presence of specific characters.
For reasons of expediency and cost, most of these discoveries have
occurred in small, rapidly breeding species. Enough work has been
done with the larger animals, however, to reveal the very great proba-
bility that the genes behave in these animals in approximately the same
manner as has been so abundantly proved to be the case in the smaller
species. It is true that we do not have for any farm mammal even a
beginning of a chromosome map, but this may be no great loss or depriva-
tion, for, as far as the writer knows, the chromosome map of Zea mays
has not served any utilitarian purpose in the recent remarkable progress
in corn breeding.
Work on the chemistry of the gene and the biochemical effects of gene
action should provide the animal breeder of the future with powerful
working instruments. The fact that in some of the lower forms vitamin
and amino-acid synthesis are so closely controlled by specific genes may
eventually contribute to the identification of such genes and their
functions in higher organisms.
The clear-cut demonstration of the influence of X rays and other
radiations on mutations in lower forms and the effects of chemicals, such
as mustard gas and colchicine, are shedding new light on the causes
of variations. It is not impossible that when the gene, the virus, the
enzyme, the hormone, and the vitamin-to mention a few-are under-
stood, and the information integrated, that evolution may be speeded
up or molded more to man's purpose. In the past we have had to be
content with what nature provided in the way of mutations. Perhaps
the time is coming when we can induce beneficial mutations by artificial _
means and thus speed up the process of animal improvement.
742 BltEEDING AND IMPROVEMENT OF FARM ANIMALS
An understanding of the basic principles involved in hereditary trans-

mission gives the modern breeder a compass, with the north, south, east,
and \"est quarters clearly marked, even though it is impossible yet
completely to "box the compass."
Animal breeding is still an art, but it is coming to rest more and mGre
securely on scientific foundations. We know that any animal arises
from the union of a specific egg and a specific sperm; that the qualities
FIG. 223.-Belgian stallion Jay Farceur, grand champion International Livestock Exposi-
tion 1938,1939, and 1940. Bred by J. W. Billman, Grand Junction, Iowa, owned by B. C.
Borneman, Danville, Ill.
of the new individual depend entirely on the genetic make-up of the

egg and sperm, regardless of the phase of the moon, direction of the
wind, etc., when fertilization takes place; that proper feeding and care
of the pregnant female and the offspring after birth are necessary to
bring its inherent capabilities to full fruition; that the sorts of germ
cells which any animal can produce are determined by the genetic
materials supplied it by its parents; that any animal passes on a sampJe
half of its o~vn inheritance, one member of each pair of chromosomes
to each of ~ts offspring; that age of itself has no influence on these chance
\
••
RETROSPECT AND PROSPECT 743
combinations of hereditary units, i.e., other things being equal, an animal
will transmit similarly at one, two, ten, or twenty years of age.
And on the negative side we know that blood does not function in
inheritance; that the kind of blood an animal has depends on the genetic
~ontent of the egg and sperm which gave rise to it, not vice versa; that
acquired characters, telegony, maternal impressions, saturation, and
other beliefs of a superstitious age can be consigned to the limbo of the
never-was. In inheritance, genes and chromosomes are everything, all
else nothing.
It is obvious, therefore, that we will have better livestock only when
we have succeeded in so mating our animals that they automatically
produce better germ cells, better in terms of containing the genes and
gene combinations making for more desirable qualities. Each planned
mating which a purebred breeder makes actually covers two generations
in the sense that it is designed not only to produce a good immediate off-
spring but one which will in turn be able to beget better offspring.
There are two major tasks awaiting the livestock industry as a whole.
The first is the matter of analysis, or the finding out what we have
genetically in our animals as they exist at present. What our livestock
will be like 10 or 100 years from now depends on two things: (1) what
genes are present in animals now living and (2) our ability to discover
the good genes or induce better ones through artificial mutations and to
increase their number at the expense of poorer genes.
We! have made remarkable progress in animal breeding in the United
States since the days of the early settlers. Tough, wiry, streamlined
cattle have been largely replaced by blocky, low-set, high-quality bull-
ocks; our sheep have three or four times as much wool spread over a
much more desirable carcass; our hogs are more prolific, more efficient,
and speedier converters of grass and corn; our hurses are heavier and
better fitted in every way to relieve the physical burdens of food pro-
duction; and om dairy cattle produce a greater quantity of a higher
quality product than they did in Pilgrim days, in Civil War days, or even
in the First and Second World War days. Our methods may have been
somewhat of a patchwork of guess, intuition, hocus-pocus, and ballyhoo,
but what we had we used in typical, energetic American fashion. N'ever-
theless, we owe a tremendous debt of gratitude to former generations
of breeders for the legacy of genes which through their efforts are now
ours. In the march of progress, our breed associations have had a
1 Most of the material in the next few pages appeared originally as an article by the
author entitled Our Breed Associations and was published in the March, 1939, issue
of Successful Farming. It is reprinted here by permission of the Meredith Publishing
Company, Des Moines, Iowa.
744 BREEDING AND I MPRO VEMEN l ' OF' FARM A iVI MA £ S
prominent part. That they have tended to be a bit conservative IS

probably a good thing.
A breed association is a group of men banded together for the pedigree-
ing, protection, and promotion of theh' particular breed. Most of these
associations were organized after the year 1875, making them less thll.
75 years old at the present time. That they have accomplished great
good, no one acquainted with the facts would for one moment doubt.
Like all things human, however, they have had and still have their fail-
ings. After 75 years of effort, and because they are recording societies,
they should, it would seem, have records to which a breeder could turn
for fairly accurate guidance in his job of breeding better livestock. They
should now, in other words, be a real help to breeders, because they were
organized for that purpose alone and long enough ago for results to have
been achieved.
The greatest "lack" in breed associations today is the lack of records
of performance. A pedigree that consists merely of names of ancestors
without production performance of the ancestors and collaterals is of
little value.1 If, however, the pedigree could be made complete in terms
\
\
\
•t
of records or characteristics, it would be an invaluable guide in the

selective mating of animals. For example, if you have brown eyes and
your parents, grandparents, great-grandparents, brothers, sisters, cousins,
qunts, and uncles all had brown eyes, then we would be reasonably safe
in ~ssuming that you would transmit only brown eyes to your children.
In livestock, substitute for brown eyes such characteristics as hams,
wool characters, loins, rounds, size of litters, efficiency of feed conversion
or fattening, etc. Before we can hope for rapid progress in livestock
breeding, a way must be found to get records of such performance
characteristics on our breeding animals.
The U.S. Department of Agriculture, in cooperation with state colleges
of agriculture, inaugurated surveys in 1936 seeking to discover superior
breeding animals backed by adequate records. In beef and dual-purpose
cattle, the 48 state colleges and experiment stations were able to furnish
practically no data that might be useful in locating superior breeding
stock. They were asked also to submit the names of breeders who
might have such superior stock, and for the whole of the United States
11 stations furnished the names of 64 men-who were not surveyed
because of the small likelihood of getting any data of real value. It
was the same story in swine, although several stations were carrying on
research projects dealing with the inheritance of characters of outstand-
ing economic importance. "Because of the lack of results from the
survey of experiment-station herds, contacts with owners of private herds
were postponed."
I t was the same general story all over again with sheep and horses:
no records anywhere to prove breeding worth. In dairy cattle, more
tangible results we're secured from the state colleges and experiment
stations, and the survey extended to dairy-cattle breeders. In all, 1,097
herds were listed, and 708 herds in 40 states were included in the final
summary. A total of 4,309 sires in seven breeds were found whose
daughters' production could be compared with that of their dams. These
bulls were bred to about 30,000 cows whose production averaged 452 lb.
of butterfat-the resulting daughters averaged 451 lb. of butterfat.
These, remember, are the selected best herds in the country. One farm,
after using more than 12 bulls over a 35-year period, ended with approxi-
mately the same production it started with.
The first stage of our animal husbandry in America was the introduction
and mating of cattle, sheep, swine, and horses by the early settlers.
There were no breeds in those early days, just animals. Most of our
breeds arose in England, Europe, and America largely during the nine-
teenth century and largely, be it noted, through inbreeding to intensify
desirable characteristics and labels (black for Angus, black and white for
740 IJREEDING AND IMPROVEMEN7' OF FARM ANIMALS
Holsteins, red for Durocs, etc.). Along with this came the formation of
breed associations to record pedigrees, to protect the purity of and
promote the use of various breeds. Standards of perfection were set up,
and the breeds gradually approached the standard"by the simple process
of weeding out those animals which failed to approximate the breed
standards. It was early realized that some animals which were them-
selves good representatives of the breed lacked the ability to transmit
their own desirable qualities. By and large, however, the purebreds
were better individuals and better transmitters than grades or scrubs.
This led to the second stage in American animal husbandry, the use
of purebred sires for grading up native or grade herds. The demand
FIG. 225.-Grand champion steer, a Hereford , at the 1949 International Livestock Exposi-
tion, shown by Pecos County 4H Club, Ft. Stockton, Tex. (PhotoGraph. Courte8Y of J. F.
Abernathy Livestock Photo Company.)
for purebreds and the lack of rigid criteria for measuring breeding worth
led to the use of many pOOl" purebreds as breeders-to the detriment
of animal husbandry in general and the breed in particular. The worst
feature of a poor sire from either the breed's or the individual breeder's
standpoint is the fact that, if he is used, the harm does not stop with
him but breeds on for many generations. The contribution of purebred
sires on the whole to our American aninial husbandry, though, is beyond
calculation. Although some bad sires have been used, the total effect
in better loins, rounds, hams, legs, racks, wool, milk, and power is both
enormous and well recognized. The breed associations have been great
promoters, and they have performed their function in this second stage
of animal development with marked success.
The thi~ stage, that of getting more complete and accurate production
records sp that animals may be proved for breeding value, lies in the
\
,
t
future, and time alone can tell how well the breed associations will
respond to this responsibility. Up to the present they have done very
little. They have worked hard to preserve the purity of their respective
breeds through recording genealogies and also to promote their breeds.
~_Vue, the dairy breeds have had their AR work for nearly 50 years.
Although it represented a considerable advance over the old, recordless
days, its value should be apparent after this lapse of time, so that we
could now write to a dairy-breed secretary and find out just where we
could buy a bull that, when mated to a herd of cows of some known
production level, would leave a group of daughters of some certain higher
production. This, however, is not yet possible. This does not imply
that the secretary would be acting as sales agent for some particular
breeder or group of breeders, but rather that the breed had secured facts
regarding the transmitting abilities of animals in their respective breeds
and had published the facts for breeders and buyers to use as they saw
fit. And before we criticize the breed secretary, we should call to mind
the fact that he is simply a person hired to carry out the desires of the
breeders as expressed through the breed's executive committee or board
of directors.
The value of AR work, from a breed-improvement standpoint, has
been negligible or worse. Thousands of excellent records have been
made and many animals have been sold at long prices because of them,
but we know now that an individual's performance and its progeny may
be very dissimilar. AR testing is like going into some city, finding the
five prettiest girls in the town, and then inducing eligible young bachelors
to believe that all the girls in said city are beautiful, and implying proba-
bly that they are also good cooks. No, selective testing is not worth
the money it costs because it gives a distorted picture of the facts. The
poorest bull that ever lived would probably have sired a few good daugh-
ters out of really good cows.
The herd test is a lot more satisfactory (records on all animals in a herd
year after year), and such testing is now available in all the dairy breeds.
I ts use should be pushed much harder than is generally being done, and
at the same time selective testing should be discouraged. Then, after
the records are secured, they should be made readily available to breeders
in a form that they can use in selecting breeding stock. The best way
is on the basis of a comparison of the daughters' records with those of
their dams. Som~ of the breeds are now doing this.
A fine system of recording D.H.I.A. records and getting valuable data
quickly into the hands of the breeder has been set up by the Bureau of
Dairy Industry, in cooperation with state colleges. If a bull is being
used in a D.H.LA. herd, we can write to our state extem;ion dairy
specialist and get the breeding facts about the bull as soon as he has
five daughters with records from dams with records, as these facts are
published several times a year.
From a breed-improvement standpoint the dairy-breed associations
could well afford to discourage, and shortly give up entirely, selective
testing (as some already have), and try to induce their breeders to test
all their cows every year. All the bulls could then be proved while they
are still young and vigorous.
The breed associations are now becoming cognizant of their responsi-
bility to provide their members with unselected factual data regarding
the animals in their respective breeds. Plans are being developed by
many breeds to provide breeders with unselected factual data on which
constructive breeding and selection programs can be founded. The plans
in the various breeds vary and are being refined as new needs are indi-
cated. Since the plans are in a state of flux (and perhaps always should
be), we do not print them herewith but anyone interested can secure
them by writing to the breed secretaries as listed in the Appendix. 1
The greatest good can accrue from the breeds securing and publishing
records of performance of sires, since the series of sires a breeder uses
makes or breaks his herd. If the breeds will do this, fewer mistakes will
need to be made by breeders in selecting sires. If, in addition, breeders
will study and chart their own female lines, selecting always from the
best and letting the poorer lines die out, great impetus will be given to
the creation of finer livestock.
Our greatest need is records to prove breeding worth-records as to
speed and economy of gains, quality of product, reproductive efficiency,
amount and character of wool, longevity, resistance to disease, pulling
and staying power, soundness, etc. The job is for each class of stock to
set up its own standards and to find practical, economical methods of
measuring its animal population against the standards. The purebred
breeders, through their associations, should take the lead. It is a
challenge that it would seem they cannot ignore.
The other major task awaiting the livestock industry, after that of
getting records of performance, is that of fitting together these known
ingredients in our animals into strains of animals which will surpass in
performance anything that we have known in the past. This is the true
art of breeding. In essence, it is merely applying the known principles
of reproductive physiology and genetics through systems of breeding and
selection in order to attain quickly and surely some preconceived ideal.
Selection might be thought of as the anvil and systems of breeding as
1 The American Dairy Cattle Club (based on production rather than on pureness of
pedigree) sec~ta'ry is Leland W. Lamb, 213 East Seneca St., Ithaca, N.Y.
\
\
•,
the hammer. Animals are tested in the fire of record keeping from which
the best emerges in a somewhat purified state. Taking this tested
material, the breeder with his hammer of breeding system shapes it
according to his will agairist the anvil of selection.
) 1'he regional animal-breeding laboratories are carrying on extensive
research in inbreeding and outbreeding. Lines are being made relatively
homozygous through inbreeding which retain desirable qualities that
can be perpetuated in true breeding strains. Crosses between lines are
sometimes yielding highly desirable results, and many new strains ha\'e
been and will continue to be made by outbreeding followed by varying
It
FIG. 226.-Grand champion wether, a Southdowu, at the 1949 International Livestock
Exposition, shown by the University of Kentucky. (Photo(JralJh Courtesy of J. F. Abernathy
Livestock Photo Company.)
degrees of inbreeding. Now that the problems of systems of breeding

have been attacked on a logical and comprehensive basis, we can con-
fidently look forward to more rapid animal improvement.
The first problem which must be solved we can properly call analysis.
Its method is that of recording essential data about the performance of
our animals, and its success depends largely on our ability to separate
the environmental differences from those caused by inheritance, for only
the latter are capable of being incorporated into the material passed
along from parent to offspring. This second problem we can therefore
call that of synthesz's, the building up of strains of animals that are rela-
tively homozygous for the qualities which make any given class of live-
I:ltock desirable and profitable.
Up to the present, we have tried to practice this sort of synthesis on
the information that we could get by studying individuality and pedigrees
consisting merely of names, with perhaps a modicum of highly selected

records to try to make pedigrees look as favorable as possible, extreme
care being exercised to keep out any unfavorable data. It is, of course,
human to try to put our best foot forward, and a breeder who told the
whole truth about his animals would no doubt, temporarily at least; Be
at a great disadvantage compared \vith his fellow breeders who were
telling only enticing half-truths. Our selection has been based, in other
words, on phenotype and made-to-sell pedigrees.
The means of putting a firmer foundation under the whole of our
livestock-breeding industry are now at hand. How should we proceed?
1. Study individuality with even greater care and diligence than we
have in the past.
2. Get and print relevant breeding and performance data on our
breeding animals. In other words, apply the progeny-performance test
so that individuality can be more scientifically appraised and pedigrees
written which will aid rather than hinder a breeder in arriving at a fair
estimate of a young animal's probable performance and transmitting
abilities.
3. By systems of breeding render the more desirable strains homozy-
gous by inbreeding or create ne\\' strains by outbreeding.
The past 30 or 40 years have furnished the basic scientific principles
in terms of reproductive physiology, hereditary transmission, and the
control of genital diseases for a sound art of animal breeding. Coopera-
tive efforts among breeders are providing the practical working organ-
izations for putting these scientific discoveries into effect.
Many agencies are now cooperating in livestock-improvement en-
deavor. The Bureau of Animal Industry "Better Sires-Better Stock"
campaign has been instrumental in ridding several whole counties of
grade and scrub sires. In many states the extension service working
through the county agents is carrying on detailed programs of grading,
selection, and culling in all types of livestock, and in some states the
college or university is loaning sires to cooperating groups of producers.
These projects involve the securing and intelligent appraisal of actual
transmitting performances, and many of them involve county and dis-
trict shows with special attention given to "get-of-sire" classes. In
some states county livestock-improvement associations have been formed,
and in addition to livestock improvement through the testing of sires
and the greater use of proved sires, grading and management demonstra-
tions, and competitive showing, such things as "livestock protective
associations" and "cooperative grazing associations" are developing.
Such cooperative efforts of producers are in many instances being aided
by banks, stockyard companies, railroads, civic clubs, etc.
)
1
RETROSPECT I1ND PROSPECT 751
Regionallaboratories for livestock improvement have been established

by the Bureau of Animal Industry and the Bureau of Dairy Industry,
in addition to the National Agricultural Research Center at Beltsville,
Md., to cope with the problem of analyzing our present breeding animals
alld,to devise methods through systems of breeding and selection for
rendering them pure in their inheritance for commercially desirable
qualities. Last, but by no means least, our breed associations are
devising projects aimed at discovering the best in each respective breed
and are beginning to foster plans to weed out the poor purebreds.
Constant improvement is being made in the techniques for measuring
genetic make-up. Many of these techniques can be used in practical
FIG. 227.-Grand champion barrow, a Berkshire, at the 1!)49 International Livestock

Exposition, shown by Ohio State University. (Photoqmph Courtesy of J. F. A bernathy
Livestock Photo Company.)
livestock breeding. In the beef-cattle breeding project at the U.S.

Range Livestock Experiment Station and the Montana Agricultural
Experiment Station, it has recently been shown that there is a high
correlation between the performance of the progeny of proved sires and
that of his sibs; between the rate of gain of young bulls and the rate of
gain of their progeny, and a high correlation between rate of gain and
efficiency of gain. This opens up the po sibility of greatly speeding up
beef-cattle improvement. In short, rates of gain and weight at fifteen
months seem to be highly heritable so that performance of young bulls
themselves can be used as a valid selection tool rather than having to
wait for performance of their offspring. This short cut over older
methods plus the purification of more efficient lines within breeds by
inbreeding promise much more rapid improvement in our beef-cattle
breeding operations. It is possible that the method can also be adapted
to other classes of meat animals.
752 BREEDING AND IMPROVEMENT OF FARM ANllJ.fALS
Some such speeding up is very badly needed in dairy cattle. Possibly

solutions will be found in biological analyses of hormone levels on which
presumably milk production finally depends.
The two major problems that still confront the breeding and improve-
ment of livestock can be solved only by proving sires and by crertilig
viable and desirable inbred strains to be used either as homozygous
" pure lines'" or in crosses among themselves, "controlled heterozygosity."
The graph shown as Fig. 113 indicates the distribution of 703 indexed
Holstein-Friesian bulls. A lot of these bulls are poor, most of them are
average, and a few are exceptionally good. In other words, in Holsteins
we are generally breeding average bulls to average cows, from which, of
course, the only logical expectation is that ,ve will get average offspring.
There is no reason to suspect, if we had the breeding data on all the males
in use in all the other classes and breeds of livestock, that we would get
any different sort of picture.
Since our animal populations in their commercially important features
probably simulate a normal curve genetically, the task of the livestock
industry is tc devise procedures that will reveal the facts about our
animals so that those with the better gene complexes may be used in
breeding operations and the poorer ones discarded. Since the amount
of culling that can be practiced with females is decidedly limited (i.e., we
could not overnight dispose of 50 per cent of our dairy cows, desirable as
that might be from a genetic standpoint, provided we knmv ·which ones
to discard), the problem must be solved largely in terms of males. The
procedure both for the industry as a whole and for the individual breeder
is obvious. It consists of testing out and proving by means of records
of performance of their offspring illctny times the number of males now
being proved. Only in this way can the best males from a genetic
standpoint be discovered. When they have thus been discovered, both
they and their proved good sons should be used extensively, which proba-
bly means the wider adoption of the technique of artificial insemination.
The other major problem is that of carrying on intensive inbreeding
experiments on a large enough scale so that they may have a fair chance
of being successful. This the individual breeder is generally not equipped
to do. The regional animal-breeding laboratories sponsored by the U.S.
Department of Agriculture and the individual states are now at work
on this problem. If enough strains can be started in closebreeding, a
few of them can probably be carried through until they emerge after a few
generations as relatively pure strains. Experimental crossing of such
strains with the purpose of bolstering the weak points of each can perhaps
lead eventually to relatively pure breeding lines in animals, at least in
1- degree compa.rabl~ with those already created in plants. There are
many difficulties involved in such a program, but they would not at this
time seem to be insurmountable. If such "pure lines" can be created
and sires made available to breeders to be used in artificial-breeding
associations, or in bull, boar, ram, or stud associations, and if the
irid~'vidual breeder win keep such records as will enable him to select
intelligently in his best female lines, animal breeding will begin to
approach a scientific level. As pointed out by Dr. Hugh C. McPhee,
what we seek in our breeding animals is merit. Genetically, this can
be achieved either through homozygosity or perhaps, as in the case of
corn breeding, through controlled heterozygosity.
Within the past 20 years a concerted, scientific attack on the problems
involved in improving our livestock through breeding has been launched
by the Bureau of Animal Industry, and the Bureau of Dairy Industry of
the U.S. Department of Agriculture. We are indebted to Dr. Hugh C.
McPhee, assistant chief of the Bureau of Animal Industry, for the
following discussion of its work:
REGIO~AL SWDiE BREEDING LABORATORY
The Ilegional Swine Breeding Laboratory is one of several laboratories estab-

lished in cooperation with state agricultural experiment stations under authority
of the Bankhead-Jones Act of 1935. Each of these regional laboratories was
established by the Secretary of Agriculture of the United States at the request of
the directors of the state agricultural experiment stations in the region repre-
sented. The region covered by the swine breeding laboratory is that of 13 Corn
Belt States. A plan setting forth the objectives and procedures was approved
by the directors of these experiment stations and officials of the U.S. Department
of Agriculture and recommended to the Secretary of Agriculture. The Secretary
authorized the establishment of the laboratory in 1937 with headquarters at
Ames, Iowa, and assigned it to the Bureau of Animal Industry for administration.
The general plan provides for an annual meeting of the technical representatives
to formulate and recommend the program of research to be pursued.
Objeclives.-The primary objectives of the laboratory are (1) to discover,
develop and test procedures of breeding and selection which may aid hog pro-
ducers in improving hogs in respect to performance and carcass, (2) to investigate
precisely the use of inbred lines for improving the breeding value of pure breeds
and for use in pork production, (3) to enlarge knowledge concerning the genetic
effects of inbreeding and inheritance of charaetel's in swine, and (4) to evaluate
and demonstrate application of such knowledge in swine breeding.
Procedures.-The general procedures for the research undertaken in the various
projects up to the present time relate to devising and testing methods of selection,
developing and using inbred lines in various types of crosses, determining the
consequences of inbreeding swine, evaluating the carcass desirability of various
lines and crosses, determining the heritability of performance characters, and
some of the physiological aspects of the breeding performance of SOWl'( and boars.
754 BREEDING AND IA1PROVEMENT OF' FARM ANIMALS
Three systems of breeding have been followed:

1. Inbreeding for the purpose of establishing inbred lines to permit converging
in single and double crosses within lines and topcrossing of inbred boars on non-
inbred sows. Three intensities were used: four-sire, two-sire, and one-sire closed
herds. l '
2. Selection based entirely on rate of growth in a herd closed to outside blood

in which the inbreeding is kept at a negligibly low level.
3. Selection in a herd started from crossing 2 established breeds and subse-
quently closed to outside bloods, selections being based on performance and
carcass merit.
REGIONAL BEEF CATTLE BREEDING PROGRAM
The Bureau of Animal Industry is conducting an extensive beef cattle breeding

program in cooperation with the state agricultural experiment stations in the
western, north-central, and southern regions of the United States. This program
differs in several respects from the regional breeding laboratories established
under authority of the Bankhead-Jones Act. The financial support comes from
the Research and Marketing Act of 1946. There is a regional experiment station
advisor and a technical committee for each region to advise with the national
roordinator on matters pertaining to initiation of projects, procedures and alloca-
tion of funds. The national coordinator is also in charge of other beef cattle
research for the Bureau of Animal Industry and at present has headquarters at
Denver, Colorado.
Objective.-To improve beef cattle production with respect to quality and
efficiency through use of sib tests in predicting breeding value of lines and com-
binations of lines of beef cattle and through propagation and distribution of
superior lines and combinations.
Procedures.-Utilizing available basic research information on how to develop
better lines and how to measure genetic variability this new program will test,
synthesize, and put to use in the field new lines of superior beef cattle. State
experiment station and privately-owned unrelated herds of Hereford, Angus,
Bhorthorn, and other pure breeds of cattle and lines derived from combinations of
lines of these breeds will be sib tested. Superior lines and combinations of lines
will be propagated and distributed through cooperating state experiment stations
to beef cattle producers for further testing under field conditions.
WESTERN SHEEP BREEDING LABORA'l'ORY
This regional laboratory was established in 1937 by the Secretary of Agriculture

acting upon a request of the experiment station directors in the 12 far-western
states, including Texas. The headquarters were placed at the U.S. Sheep
Experiment Station, Dubois, Idaho, at the request of the station directors for the
reason that the Bureau of Animal Industry already had available at that point a .
large physi()al plant consisting of buildings and sufficient land for spring, summer,
fall and winter range for a large number of sheep. X 0 experiment station in the
region was similarly equipped to maintain the number of animals required hy
the proposed research program.
The laboratory is assigned to the Bureau of Animal Industry for administration
and is in charge of a director, who is also in charge of the program of the U.S.
S~e!jp Experiment Station carried on at the same location. These two programs
are separate and distinct in regard to budget and animals, but are fully coordi-
nated by the director and his staff. In the beginning the program of the Western
Sheep Breeding Laboratory was limited to the Rambouillet breed. The program
of the U.S. Sheep Experiment Station includes breeds other than the Rambouillet,
principally Columbia and Targhee. It also embraces other lines of research
such as range sheep management, sheep feeding, and range vegetation.
An annual meeting of technical representatives of the cooperating state experi-
ment stations is scheduled, at which time the research progress is reviewed and
plans are adjusted for the coming year. Between such meetings the director
visits the cooperating experiment stations and reviews current projects which
are being conducted as part of the over-all program.
Objective.-The main objective is to improve Rambouillet sheep for lamb and
wool production under range conditions.
Procedures.-To attain this objective, basic breeding methods are employed,
heritability analyses are made of the various utility factors and selection is
based on production as it is measured under range conditions. Emphasis is
placed primarily on the quantity and quality of lambs produced; the length,
quality, and quantity of clean scoured wool and upon the adaptability and
longevity of the sheep.
Development of systems of breeding for locating strains of Rambouillet sheep
which may possess combinations of genes that will improve strains with which
they may be crossed includes:
1. The development of inbred strains or lines by mating animals as closely
related as possible with emphasis on selection for all characters of economic
importance.
2. The development of inbred lines with special reference to very important
characters that are of economic significance in range sheep such as mutton form,
length of staple and open faces.
3. The development of non-inbred control groups.
For the purpose of developing methods by which undesirable characteristics
can be eliminated from Rambouillet sheep, the inheritance of abnormalities in
the growth of wool, hairiness in the fleece and skin folds as wrinkles are being:
studied. Supporting projects in physiology of reproduction are also being carried
on with special reference to sexual maturity of ram lambs, quality of semen in
relation to fertility and factors affecting fertility of ewes.
The Bureau of Dairy Industry has made some notable contributionE'
to better breeding through continued use of proved sires, having raised
butterfat production in their Holstein and Jersey herds by about 200 lb.
per year. This bureau has also contributed much to our knowledge of
inbreeding and is now embarked on a program of crossbreeding.
756 BREEDING AND Il11PROVElvIEN1' OF FARM ANIMALS
Not so many years ago animal breeding was thbught by many to

involve a considerable amount of magic, which could be invoked by
only a select few on the basis of a God-given instinct. Genetics has
dispelled this myth and made it evident that anyone who is willing to
take the pains to analyze his animals in terms of their producing' and
transmitting abillties can also use, with varying degrees of success to be
sure, the only tools which ever have been or will be available for animal
improvement-breeding systems and selection.
A proton and electrons, then elements emerge-

They organize and synthesize to form a living surge.
Then animals and man appear in orderly progression:
Teamed, they move with quickened pace-a jitting, joint possession.
;
APPENDIX
Livestock Record Associations
The following list of livestock record associations, as of Sept. 1, 1950, is presented
through the courtesy of 1\~r. H. J. Brant, secretary of the National Society of Livestock
Record Associations with headquarters at 282 South \Vabash Street, Wabash, Ind.
This society was organized in 1911 and represents over 350,000 breeders of purebred
livestock. The purposes of the society are stated as follows:
"The objects of this Society shall be to advance the interests of member registry
associations, (1) by devising and perfecting methods for preserving pedigrees of pure-
bred animals; (2) by endeavoring to secure the enactment of equitable laws relating to
the purebred livestock industry and to livestock record associations; (3) by securing
the adoption of just freight and express rates on exhibition and breeding stock; and
(4) by originating or participating in any and all other activities which will in the
judgment of the Directors of the Society advance the interests of breeders of purebred
livestock through their respective registry associations."
Beef and Dual-purpo3e Cattle
American Aberdeen-Angus Breeders Association, Union Stock Yards, Chicago H,
Ill., Frank Richards, Secretary.
American Brahman Breeders' Association, 2711 South Main Street, Houston 2, Tex.,
Mrs. G. R. Sunday, Secretary.
American Devon Cattle Club, Inc., Meredith, N.H., W. J. Neal, Secretary.
American Galloway Breeders' Association, Henry, Ill., Rank C. Forbes, Secretary.
American Hereford Association, 300 West 11th Street, Kansas City, Mo., Jack Turner,
Secretary.
American Kerry and Dexter Club, Independence, Iowa, Roy Cook, Secretary.
American Polled Hereford Breeders' Association, 1110 Grand Avenue, Kansas City 6,
Mo., Don Chittenden, Secretary.
American Polled Shorthorn Society, Union Stock Yards, Chicago Il, Ill., C. D. Swaff('r,
Secretary.
American Red Danish Cattle Association, Fairview, T\lich., Clifford Shantz, Secretary.
American Scotch Highland Breeders' Association, Henry, Ill., Rank C. Forlws,
Secretary.
American Shorthorn Breeders' Association, 1:'nion Stock Yards, Chicago, Ill., Clinton
K. Tomson, Secretary.
Red Polled Cattle Club of America, 3275 Holclredge Street, Lincoln, 3, "'ebr., F. A.
Sloan, Secretary.
The American Milking Shorthorn Society, 4122 South union Avenue, Chicago 9, IlL,
'V. J. Hardy, Secretary.
Dairy Cnttle
American Guernsey Cattle Club, Peterborough, N.H., Karl B. Musser, Secretary.
American J(,rsey Cattle Club, Sixth and LOllg, Columbus 15, Ohio, Floyd Johnston,
Secretary.
757
758 BREEDING AND Il1IPROVEMENl' OF FARM ANIMALS
Ayrshire Breeders' Association, Brandon, Vt., C. T. Conklin, Secretary.

Brown-Swiss Cattle Breeders' Association, Beloit, Wis., Fred S. Idtse, Secretary.
Dutch Belted Cattle Association of America, Anamosa, Iowa, Thomas Stimpson,
Secretary.
Holstein-Friesian Association of America, Brattleboro, Vt., H. ,,yo Norton, Jr.,
Secretary.
Draft Horses
American Clydesdale Association, Union Stock Yards, Chicago 9, Ill., Miss Margaret
Coridan, Secretary.
American Shire Horse Association, 319 East 4th Street, Des Moines, Iowa, E. F. Fox,
Secretary.
American Suffolk Horse Association, Clinton, N.J., L. B. ·Wescott, Secretary.
Belgian Draft Horse Corporation of America, 'Vabash, Ind., H. J. Brant, Secretary.
Pereheron Horse Association of America, 809 Exchange Avenue, Union Stock Yards,
Chicago 9, Ill., Mrs. Anne Brown, Secretary.
Light Horses
American Albino Horse Club, Inc., Butte, Neb., Ruth Thompson, Secretary.
American Hackney Horse Society, Room 1737, 42 Broadway, ='lew York, N.Y., Mrs.
J. Macy Willets, Secretary.
American Quarter Horse Association, 1405-b \Vest 10th Avenue, Amarillo, Tex.,
Raymond D. Hollingworth, Secretary.
American Saddle Horse Breeders' Association, 929 South Fourth Street, Louisville,
Ky., C. J. Cronan, Jr., Secretary.
Appaloosa Horse Club, :l\Ioscow, Idaho, George B. Hatley, Secretary.
Arabian Horse Club of America, 111 West Monroe Street, Chicago 3, Ill., Frank Watt,
Secretary.
Cleveland Bay Society, White Post, Va., A. ]\Iackay Smith, Secretary.
IVlorocco Spotted Horse Association of America, Greenfield, Iowa, LeRoy Fritz,
Secretary.
National Quarter Horse Breeders' Association, Houston 5, Tex., J. M. Huffington,
Secretary.
Palomino Horse Association, Reseda, Calif., Jim Fagan, Secretary.
Palomino Horse Breeders of America, Mineral Wells, Tex., Dr. H. Arthur Zappe,
Secretary.
Pinto Horse Society, Concord, Calif., George ]\1. Glendenning, Secretary.
Tennessee \Valking Horse Breeders' Association of America, Lewisburg, Tenn., Miss
Syd Houston, Secretary.
The Jockey Club (Thoroughbred), 250 Park Avenue, New York 17, N.Y., l.1arshall
Cassidy, Secretary.
The Morgan Horse Club, 90 Broad Street, New York 4, N.Y., F. B. Hills, Secretary.
United States Trotting Association (Standardbred), Goshen, N.Y., Neil R. Gahaa,
Secretary.
Ponies
American Shetland Pony Club, P.O. Box. 2557, South Bend, Ind., Wayne C. Kirk,
Secretary.
Welsh Pony Society of America, 417 West Engineering Bldg., University of Michigan,
Ann,Mfor, Mich., Frank H. Smith, Seeretary.
, .""._
\
APPENDIX 759
Jacks and Jennets

Standard Jack and Jennet Registry of America, P.O. Box 375, Garden City, Kan.,
Mrs. N. E. Hineman, Secretary.
Sheep
American Cheviot Sheep Society, Oneonta, K .Y., :\frs. Katherine S. Turrell, Secretary.
American Corriedale Association, 100 North Garth Street, Columbia, Mo., Rollo E.
Singleton, Secretary.
American Cotswold Registry Association, Union Stock Yards, Chicago 9, Ill., F. W.
Harding, Secretary.
American and Delaine-Merino Association, Marysville, Ohio, Walter 1\1. Staley, Jr.,
Secretary.
American Hampshire Sheep Association, 72 'Yoodland Avenue, Detroit 2, ::'I1ich., l\Irs.
Helen Tyler Belote, Secretary.
American Oxford Down Record Association, Clayton, Ind., J. :\1. :'IIcHaffie, Secretary.
American Rambouillet Sheep' Breeders' Association, San Angelo, Tex., Geneva Cald-
well, Secretary.
American Romney Breeders' Association, 200 Dairy Building, Corvallis, Ore., H. A.
Lindgren, Secretary.
American Shropshire Registry Association, Lafayette, Ind., Chas. F. Osborn, Secre-
tary.
American South down Breeders' Association, 212 South Allen Street, State College,
Pa., 'Yilliam L. Henning, Secretary.
American Suffolk Sheep Society, Moscow, Idaho, C. ,Yo Hickman, Secretary.
Black Top and National Delaine-1\Ierino Sheep Breeders' Association, Houston, Pa.,
1. Y. Hamilton, Secretary.
Columbia Sheep Breeders' Association of America, Box 2466, State College Station,
Fargo, N.D., 1\1. L. Buchanan, Secretary.
Continental Dorset Club, Hickory, Pa., J. R. Henderson, Secretary.
Karakul Fur Sheep Registry, Friendship, Wis., L. K. Brown, Secretary.
Montadale Sheep Breeders' Association, 61 Angelica Street, St. Louis, Mo., E. H.
Mattingly, Secretary.
National Lincoln Sheep Breeders' Association, College Manor, Apartment 10, East
Lansing, Mich., Harry Crandell, Jr., Secretary.
National Suffolk Sheep Association, Middleville, Mich., C. A. Williams, Secretary.
National Tunis Sheep Registry, Fulton, N.Y., Ralph E. Owen, Secretary.
Texas Delaine-Merino Record Association, Brady, Tex., George H. Johanson,
Secretary.
Goats
American Angora Goat Breeders' Association, Rocksprings, Tex., Mrs. Thomas L.
Taylor, Secretary.
American Goat Society, Inc., :'Ilena, Ark, R. D. '''eis, Secretary.
American Milk Goat Record Association, Sherborn, Mass., Miss Mary L. Farley,
Secretary.
Swine
American Berkshire Association, 410 South 5th Street, Springfield, Ill., Noel F. Titus,
Secretary. - -
American Essex Swine Association, 1335 East 2d Street, Muscatine, Iowa, Mrs. J. J.
Lighthall, Secretary.
American Spotted Poland-China Association, Moberly, Mo., Van G. Sutliff, Secretary.
American Yorkshire Club, Wallace Building, LaFayette, Ind., Bob Shannon,
Secretary.
Breeders Chester White Record Association, 330 Royal Lnion Building, Des l\1oinC!:s (,~
Iowa, L. 'V. Drennen, Secretary.
Chester 'Vhite Record Association, Rochester, Ind., Levi P. Moore, Secretary.
Hampshire Swine Registry, Commercial National Bank Building, Peoria, Ill., R. L.
Pemberton, Secretary.
Kentucky Red Berkshire Swine Association, Lancaster, Ky., Hogan Teater, Secretary.
National Hereford Hog Record Association, Chariton, Iowa, Harris Sellers, Jr.,
Secretary.
Kational Mule Foot Hog Association, DeGraff, Ohio, O. C. Kreglow, Secretary.
National Spotted Poland-China Record Association, 3153 Kenwood Avenue, Indian-
apolis 8, Ind., Fned L. Obenehain, Secretary.
O.I.C. Swine Association, Brookville, Ohio, ':\lrs. Jane I\oper, Secretary.
O.I.C. Swine Breeders' Association, Goshen, Ind., Harry C. Miner, Secretary.
The Poland-China Record Association, Galesburg, Ill., C. G. ·McCahan, Secretary.
Tamworth Swine Association, Hagerstown, Ind., R. H. Waltz, Secretary.
United Duroe Record Association, Peoria 3, Ill., B. R. Evans, Secretary.
NAME INDEX
A Berthold, 113
Bickersteth, 60
Adamson, 620 Bilek,243
Agassiz, 80 Bissonnette, 111, 146
Alexander of Macedon, 33 Blakeslee, 405, 409
Allen, 147,251,439 Blandau, 152, 155, 161, 167
Almquist, 118, 278 Blizzard, 711
Altenburg, 404 Bloom, 223
Ames, 16 Blunn, 665
Anaxagoras, 69 Bogart, 178-179
Anaximander, 69 Bohren, 147
Ancon, 381 Boissonnade, 2gn.
Anderson, 119, 124, 126, 181-183,276 Bos, 217, 511
Andrews, 110, 112, 117-118, 123, 130, Boule, 43
147, 176, 178,206,269 Brahe, 304
Aristotle, 70 Branham, 458
Arnold,581 Branton, 184
Asdell, 177, 179,205,211,252 Bratton, 276-2i7
Auerbach, 399 Brett, 679
Bridges, 214, 376, 383-384,404,406,408-
B 409, 435, 438-440, 442-443
Brier, 728
Babcock, E. B., 214n., 216, 218-219n., Briggs, 178, 210
401 Brody, 699, 709n.
Babcock, H. E., 11, 418, 421 Broom, 45
Bacon, 78 Brown, 442
Baer, von, 99, 293 Brownell, 631
Bagg, 430 Bruno, 71, 78
Baker, 174,469,515,665,699 Bufion, 71, 78-79
Bakewell, 16, 23-24, 293, 295, 509-510, Burrows, 267
513,517,564,646,683
Barnard,73 C
Barry, 127
Bartlett, 207-208, 515 Caesar, 5~1
Bateson, 291, 295, 365, 415 Camerarius, 2\)3, 304
Battell, 732 Card, 427, 573
Baugess, 120, 126 Carducci, 60
Beadle, 381 Carroll, 466-468, (;49
Beard, 148 Casida, 1.55-156, 200, 221, 277, 28G-287
Beck 271-272 Caspari, 381
Beeling, 405, 40g Castle, 309n., 311n., 351, 381, 414-415,
Berg, 125 429,511
Berliner, 110, 112, 182,273,276 Cheng, 277
761
7fi2 BREEDING AND IMPROVEMENT OF FARM ANIMALS
Choate, 634 E
Clark, 469, 697n., 700, 703
Clausen, 215n., 216, 218-219n., 401, 418, Earls, 243n.
421, 667-668 East, 295, 498, 517
Cole, 144, 151, 155, 190 Eaton, 342n.-343
Collings, 24, 513 Edwards, 209
Columella, 30 Ellerslie, 700
Comstock, 269, 671 Ely, 257
Conklin, 590 Empedocles, 69
Cook,293n. Engle, 155
Cooley, 711 Ephrussi, 381
Copernicus, 304 Epicurus, 69
Coppini, 199 Erb, 111,212
Correns, 294, 306, 311 Espe, 253
Craft, 680-681 Esplin, 210.
Creighton, 371, 385 Evans, 147, IS.5, 267
Crew, 446
Cruickshank, 16, 513
Cuenot, 295
Culbertson, 468n. Fairchild, 293, 305
Culley, 509 Finlay, 430
Cuvier, 50, 80 Flemming, 29,1
Frank,251
D Frankhauser, 440
Fraps, 382
Dalton, 69 Freeman, 394
Danforth, 439 Friedman, 209-210, 228
Darwin, Charles, 78-84, 290, 294, 305,
G
420,528
Darwin, Erasmus, 79 . Gaines, 595, 602
Datura, 405 Galen, 98
Davenport, 219, 295, 423 Galileo, 304
Davidson, 595, 602 Galton, 294, 300, 607
Davis, 155, 174, 177, 179,282 Gardner, 250
Dawson, Charles, 42 Garrett, 700
Dawson, W. M., 690n. Gates, 409
Dempsey, 151 Gentry, 16, 510
Descartes, 71 Gifford, 604, 698
De Vries, 80, 294, 301, 306, 311 Gillette, 463
Dickerson, 516, 664n.-66S, 679-681 Gilmore, 217
Dimmock, 209 Goethe, 80
Dinnussen, 147 Goldblatt, 125
Dobzhansky, 325n., 440 Goldschmidt, 437
Doisy, 147, 439 Gomez, 252
Dubois, 39, 45 Good,523
Dukes, 157 Goodale, 603
Dunkle, 353 Gowen, 569, 573n., 591, 601
Dunn, 107, 317, 325n., 334n., 372-3i3n., Graves, S13-S14, 572n., 603
380, 383, 406n.-407, 437n., 499n. Green, 269
Duthie, 16 \ Gregory, 381, 699
Dvoracheki647 Grew, 98
\
NAME INDEX 763
Griffith, 428 J
Grimes, 665
Griieter, 253 Jarvis, 682
Grummer, 200 Johannsen, 301, 532
Guilbert, 699 Johnson, 125, 215
"G11nn, 268 Jones, 498, 650, 692
Guyer, 430 Jordan, 155
H K
Haeckel, 99n., lOOn.
Haley, 702 Kant, 78
Ham, 126, 270, 273, 293 Karpechenko, 410
Hammond, 175, 433, 647-648, 657-658, Kellogg, 622
600 Kepler, 304
Hankins, 650 Khishin, 547
Hansen, 430 Kildee, 463
Hardy, 650 Kincaid, 677
Harsch,655 King, 218, 448,511,513
Hart, 210 Kirkpatrick, 110
Hartman, 130 Kleberg, Richard, 476
Harvey, 71, 98 Kleberg, Robert J., Jr., 476
Hays, 294, 511 Knapp, 469, 696-700
Hazel, 666, 677~680, 691~692 Knox, 701
Hemen, 125 Koenigswalde, von, 41
Hendrickson, 709n. Ki:illiker, 100, 127,294
Herman, 277 Ki:ilreuter, 294
Hertwig, 294, 430 Koger, 701
Hetzer, 536, 649 Krantz, 728
Hewer, 513 Kuhn, 381
Hill, 723
Hillman, 742 L
Hisaw, 231
Hodgson, 212 Lacy, 616
Hoefer, 674-675, 677 Lamarck, 71, 79, 83, 89, 420
Hofmeister, 294 Lamb,74Sn.
Hogue, 109 Lambert, 482, 573, 728, 732n.
Holt, 273 Landauer, 381
Hooke, 91, 98, 290 Laqueur, 251
Homeman, 742 Lardy, 275
Huggins, 125~126 Lasley, 119, 178, 272
Hughes, 190, 445, 513 Laughlin, 727
Hunter, 260 Lawrence, 381
Hurst, 95, 405n. Lawritson, 261
Hutchings, 206~207 Leeuwenhoek, 71, 98, 126, 2(;9, 273, 293
Hutchinson, 431 Leibnitz, 78
Hyatt, 581, 589~590, 611 Leisering, 137
Lewis, 130, 155, 261
I Lillie, 444
Ibsen, 328 Lincoln, 584
Irwin, 342 Linnaeus, 78
Iwanoff, 260 Little, 318, 430
764 BREEDING AND IMPROVEMENT OP PARIV[ ANIMALS
Loessl,260 Morris, 215

Long, 147 Moulton, 657
Loomis, 40n., 85 Mouw, 654-655
Lucretius, 69 Muller, J., 100
Lush, 434, 468n., 487n., 523-524, 535, Muller, 295, 398, 402-403, 443
581,616,631-632,666-667,677-678 Mumford, 177,431,510
Luzzi, da, 98 Murphree, 155-156, 286
Lydekker,61 Murphy, 573n.
M N
MacArthur, J. M., 347
MacArthur, J. ''1'., 347 Nalbandov, 153, 202
yfcCandlish, 463 N athusius, 401
McClintock, 371, 385 Nealc, 694
McClung, 295 Nelson, 252
McCrory, 709n. Newton, 71, 304
McDougall, 428 Nibler, 261
McDowell, 617 Nichols, 176, 646
McKenzie, 110, 112, 117, 120-121, 176- Nilsson-Ehle, 345
178, 182, 273 Nordby, 685
McLellan, 125 Nordskog, 697n.
McMahon, 696, 698 Norton, 581
Mc~{eekan, 657-658
McPhee, 753
o
Madsen, 729n. Oken, 99
Malpighi, 98 Opperworth, 45
Malthus,81 Osborn, 573
Mangelsdorf, 382 Osborne, 80
Margolin, 515 Owen, 2'd4, 342n.
Marsh, 16
Marshall, 424, 431 P
Mason, 53
Matthew, 80-81 Painter, 295, 405
Matthews, 249 Palmer, I\)9
:\{ayer, 119, 126, 272-273 Parkes, 433
Meites, 254-255 Pasteur, 71
M®~,~W~300,W2,W5,~,3W Patten, 105, 144
Mercier, 111 Patterson, 441-442
Miller, 120, 125-126,210,267 Pavlov, 428
Mills, 176 Pearl, 300, 601
Mimura, 130 Pearson, 294, 300, 523
Miner, 601 Pei,41
:\1:i1-llcr, 515 Perkins, 602
Moeller, 130 Perry, E. J., 261
Moench,273 Perry, T. W., 668
Mohler, 205 Peterson, 257
Mohr, 406 Pettit, 708n.
Moore, 117, 125-126 Pfiffner, 252
Morgan, 177 ,\-t07 , 295, 348, 356n., 365, Phillips, no, 117, 127,130, 175-176,213,
368, 37.;1, 376, 383, 407, 436, 439, 273, 275, 318, 690n., 728-729n.,
442-443\ 732n.
NAME INDEX 765
Pincus, 129, 152,285, 409 Sharp, 107, 376

Prince, 278 Shearer, 468n.
Punnett, 355 Shull, 295,419,517
.. .
Sidwell, 691
Q Simpson, 155
Sinnott, 107, 317, 325n., 334n., 372-
Quaife, 677 373n., 380, 383, 406n.-407, 437n.,
Quesenberry, 174, 477 499n.
Quick, 117 Smith, H. H., 182, 430, 633, 729n.
Quinn, 257 Snedecor, 347
Snyder, 342, 358n., 428n.
R Spallanzani, 71, 126, 250, 305
Speelman, 573, 732n.
llagsdale, 277 Spemann, 384
Hatner, 243 Spencer, 82, 650
Redfield, 423-424 Spuhler, 530
Redi,71 Stadler, 399
Reece, 252, 254, 515 Stanley, 74 .
Reese, 714n. Stern, 371-373, 376, 385, 443
Regan, 516 Stevens, 295
Reineke, 112, 255 Stewart, 177, 215-216
Remak,l00 Stockard, 429-430
Rhoad, 475'lL., 660 Stoughton, 73
Rice, 606 Strasburger, 294
Roberts, 427, 456-468, 573 Straus, 581
Robeson, 190 Stricker, 253
Robinson, 176 Strong, 250, 426
Robson, 399 Sturtevent, 376, 383, 407-408, 440, 442-
Rowan, III 443
Sutton, 212, 295
S Swammerdam, 98
Swett, 249, 572n.
St. Augustine, 70 Sykes, 146
St. Hilaire, 79
St. Thomas Aquinas, 70 T
Salisbury, 111, 124, 177, 269, 271-272,
276-277 Tanabe, 177, 221
Sanders, 503 Taylor, 16
Savage, 127, 2i3 Terrill, 178, 651, 653, 691-692, 694-
Schleiden, 91, 98, 200, 294, 305 695
Schnetzler, 109 Thales, 69
Schoetensach, 42 Thomas, 476
Schrader, 440 Thompson, 420
Schultz, 440 Thuret,294
Schwann, 71, 90, 98, 290, 294, 305 Toutain, 3On.
Scott, W. W., 125 Townshend, 24
Scott, Sir Walter, 51 Trentin, 251, 253
Scott-I"v1oncrieff, 381 Trimherger, 155, 174, 179,282
Seiler, 447 Trowbridge, 70!Jn.
Semple, 647 Tryon, 428
Shaffner, 112, 123, 129,269 Tschermak, VOIl, 294, 306, 311
Webster, 573n.
Tull,23-24
Turner, 112, 148, 209-210, 251-255, 476 Weisman, 271
Weismann, 80,217,294,421-422,511
Turner, C. D., 250, 253
Turner, C. W., 250, 602-603
Twain, 485
Tyler, 515, 581, 589-590, 611
Wells, 80--81
Wbatley, 677
Wheeler, 110
Wbeldale, 381
.. .
Tyrell, 118
Whitney, 130
V Williams, 136, 273
Wilson, 295
van Beneden, 294 Wimsatt, 129
Van Demark, 130 vYintermeyer, 617
van Dyke, 390 Winters, 478, 517, 665, 006, 698
Varro, 295, 564 W odsedalek, 218
Verges, 658 Wolff,99
Vernon, 448n. Woodward, 45, 513-514
Vesalius, 98 Wriedt, 509
Vilmorin, 294 Wright, 130, 381, 400, 414, 496, 511, 513,
Vinci, da, 70 535-536, 541, 604
Vir chow , 294
Von Guaita, 217 Y
W Yapp, 602, 604

Young, 119, 155
W aldeyer, 294
Wallace, 84
Walter, 422, 426
z
Ward,581
Warwick, 156, 163, 200, 286, 353, 644, Zelany, 383, 407
Zeller, 175,649
694
SUBJECT INDEX
A Alleles, 306, 313, 322
multiple, 339-342, 357, 401
Abiogenesis, 69 Allotetraploids, 410
Abnormalities, 94--96, 169, 173 American Aberdeen Angus Breeders'
genital, 201, '203, 221-223, 245, 281, Association, 700
342-343 American Dairy Cattle Club, 748n.
Abortion, 150, 164, 211, 246, American and Delaine Merino Record
and brucellosis, 205-208, 246 Association, 682
Acquired characteristics, 419-420 American Guernsey Cattle Club, 620, 739
inheritance of, 79, 82 American Hereford Breeders' Associa-
arguments against, 428-429 tion, 653, 702
practical phase of, 431-432 American Jersey Cattle Club, 199
requirements for proof of, 421-428 American Milking Shorthorn Society, 636
summary on, 429-431 American Pacer, 21
transmission of, supposed method, 84, American Saddle Horse, 20-21, 66
420 American Society of Animal Production,
(See also Characters; Secondary sexual Committee on Investigations, 481
characteristics) American Trotter, 21
Acrosome, 127 American Trotting Register, 424
Adohr Eldor Pearlette, Guernsey cow, 620 Amino acids, 298
Adrenalin, 353 Amnion, 168
Adrenals, 113, 222, 255-257 Amoeba, 73, 88, 90
Afterbirth (see Chorion) Amphibians, 76, 101
Age and testes function, 109 Amphimixis, 419, 424
Age-conversion factors, 593, 595, 597, Anaphase, 93-94, 367
676-677 Anatomy, 80
Age distribution of dairy sires, 618 comparative, 85
Agricultural Economics, Bureau of (see Ancon sheep, 400
U.S. Department of Agriculture) Andalusian fowl, inheritance of blue in,
Agriculture, 47 329,331
Department of (see U.S. Department Androgens, 110, 113-115, 148,222
of Agriculture) Androsterone, 113
English, 22-28, 293 Anestrus, 143
Greek, 32-34, 59, 293 Angus cattle, 23, 318-319, 413, 432, 465,
medieval, 28-30, 293 469,472
Oriental, 34--36 Angus-Hereford cross, 319
prehistoric, 47 diagram of, 318
Roman, 30-32, 59-60, 293 Animal husbandry, 30-36
Albinism, 415 Bureau of (see U.S. Department of
Alfalfa Farm Ann 2nd, Grand Champion Agriculture)
cow, 587 Greek, 32-34
Algae, gO growth of, in America, 745-747
Allantois, 164-165, log Oriental, 34-36
767
768 IJREEDING AND IMPROVEMENT OF FARM ANIMALS
Animal husbandry, Roman, 30-32 B

Animal Industry, Bureau of (see U.S.
Department of Agriculture) Babcock test, 294
Animals, and civilization, 49-50 Babirussa, 64
definition of, 68 Backcross, 364-366, 442
domestication of, 47-66 Bacteria, 73 •
farm (see Farm animals) Bacteriaphages (bacterium eaters), 74
improvement of, 103 Balance and selection, 575-576
meat (see Meat animals) Baldness, inheritance of, 353
origins and progression of, 68-96 Bang's disease (see Brucellosis)
purebred (see Purebreds) Bankhead-Jones Act of 1935, 680, 754
related, breeding of, 485-526 Banteng, 57, 59
unrelated, breeding of, 452-483 Bar gene, 383
(See also Livestock) Barrenness, 220
Anterior-lobe hormones (see Pituitary of Dutchess Shorthorns, diagram of,
gland) 216 ,
Anthropoids, 77 Bartholin, glands of, 122
Ape man of Java, 39 Bates' Dutchess family of Shorthorns,
Arab horses, 56 214-215
Archaeopteryx, 76 Bay State Lilly, Percheron mare, 719
Archaeozoic Age, 40 Beef cattle, 3, 455, 468, 551, 745, 751
Argali,61 breeding of, 174-175
Artificial breeding associations, 181, 221, crossbreeding in, 472, 47()-477
261, 283-284, 621 fertility of, 178
Artificial insemination, 5, 150, 174, 178, regional breeding program for, 754
180-182, 184, ] 86-188, 209, 221, selection of, 647, 653-654, 696-703
260-287,474-475,552,737-738,752 triple crossing in, 469
advantages of, 261-264, 287 Beef production, 16
of cow, 279-281 Beefmaster cattle, 476
and breeding efficiency, 282 Bees, 32, 419
diagrams of, 280-281 Belgian horses, 551
problems of, 284-285 Berkshire swine, 513, 751
definition of, 260, 287 Beth of Amherst 3rd, prize cow, 391
disadvantages of, 264 Bharal, 61-62
equipment for, 279 Bidder's organ, 446
history of, 260-261 Bighorns, 61
technique of, 279-283 Biology, 77-78, 82
Ascheim-Zodek test, 228 Biometry, 299-300
Ascorbic acid (vitamin e), 213 Birds, 76, 115
Asexual reproduction, 100 reproduction of, 110
Asia Minor, agriculture in, 34-36 Bison (buffalo), 57-59, 459
Ass, 30, 53-54, 218, 457, 459 Black Death, 26
Atavism, 432 Blackhorse, 23
Aurochs, 59 Blastocoele, 168
Autosomes, 92, 348, 434, 438, 441, 447, Blastocyst, 149, 164
451 Blastula, 164, 168
Autotetraploids, 410 Blending inheritance, 330
Ayrshire, Association, 3 Blood tests, 85
Ayrshire cattle, 23, 353, 391, 394, 448, Blood types in man, 341-342
540, 570,590~593,597-598,6oo,608, Boars, 110, 118-119, 121-122, 180, 182-
611 \. 183, 268, 278, 669-671
\
! \
SUBJECT INDEX 769
Boars, wild, 65 Brown Swiss cattle, 487, 540, 570, 593,
(See also Swine) 600,608
Bos, genus, groups included in, 57-60 Brown Swiss Cattle Breeders' Associa-
Boskop man, 4-5 tion,634-
Bovidae, 57, 60, 480 Brucellosis (Bang's disease), 190, 205-
·Bo·wman's capsule, 101 208, 223, 246
Bradenhurst Royal Challenger, Grand Buffalo (see Bison)
Champion bull, 634 Bull indexes, 601-613
Braham cattle, 459, 472, 476-477, 547 as selection tool, 609-613
Breed associations, 6, 17-20, 743-744, work sheet for, 605
746-748,757-760 'Bull pen, safe, 612
Breeding, animal, 1-36 Bull transmitting chart, 622-624
analysis in;749 Bulls, 180--181, 207-208, 211-212, 262-
art of, 742, 748 263,269,273,278,284,345,347,391
basic idea in, 562 424-425,464-465,538-540,580,591-
definition of, 16 637, 695, 701
in England, 22-25, 86 best age at which to buy, 613-615
genetics of, 12-14,301--302 dairy, age distribution of, 613
as hobby, 1 methods of indexing, 601-613
as livelihood, 1 exercising paddocks for, 183
opportunities in, 14-17 fertility of, 177-178
and ovulation, 154-155 young, selection of, 618--622
pre-Mendelian, 292-293 (See also Cattle)
progress in, 743-744 Burdizzo forceps, 116
since 1900, 294--296 Byzantine Empire, agriculture in, 29
and records (see Records)
scientific basis of, 742-743, 753
C
status and problems of, 11-14
Calcium, 125--126
synthesis in., 74-9-750
Calf, in utero, abnormal positions of, 239
systems of, in related animals, 485-
normal position of, 237
526
vaccination of, 206
and sterility, 217-219
young, care of, 240
in unrelated animals, 452-483
California, University of, 515--516
and variation, 411-412
California Experiment Station, 210
plant, 46--47 California Favorite, Grand Champion
pre-Mendelian, 293-294 steer, 465
Breeding efficiency, in horses, 731-732 Canidae, 51
Breeding management, of female, 187- Capra, 60
192 Capsella, 422
of male, 179-187 Carbohydrates, 184, 210
Breeds, American, 66 Carnegie Institution of Washington, 1{)6
beginnings of, in England, 22-25 Station for Experimental Evolution of.
competition in, 19 517 -
differences in, 549-551 Carnivora, 51, 77
rise of, 20--22 Castration, 110, 114--118,209, 445-4J<i
and selection, 453, 549--552 bloodless, 116
trade-marks of, 551-552 of female, 147-148, 157, 178
(See also Purebreds) (See also Spaying)
Bridge of inheri1;ance, 418 Catenations, 405
Bronze Age, 47 Cattaloes, 459
~
770 BfiEEDING AND IMPROVEMEN'l' OF FARM ANIMALS
Cattle, 29, 32, 35, 48-49, 423, 457, 459 Characters, acquired, requirements for
beef (see Beef cattle) proof of inheritance of, 421
breeding of, 187-189,221 schematic representation of, 420
breeds of, 21, 23 development of, 418-419
chromosomes in, 92 hereditary transmission of, 299-~01..
dairy (see Dairy cattle) 325, 329, 479, 741-743
domestication of, 57-60 inherent, 423
dual-purpose (see Dual-purpose cattle) and selection, 555-556
in England, 21, 23-25 Charbray cattle, 476-477
families of, 57 Chester White swine, 21, 66, 467
on farms in United States, 2 Chester White Swine Record Association,
inbreeding of, 513-516 679
lethals in, 343-344 Chiasma diagram, 368
origin of, 59-60 Chick embryo celis, 95
purebred (see Purebreds) Chickens, 32, 113,212,267,278,282,427,
seasonal fertility of, 111 430,447
(See also Bulls; Calf; Cows; Steers;
(See also Fowl)
specific names of cattle)
Chihuahua dogs, 52-53
Cattle associations, 7, 757-758
Chip of ~ettie and Aaggie, Holstein-
Cattle problems, 187-189
Friesian bull, 585
Cell diagram, 90
Cell division, 93-96, 128, 166 Chorion, 164--165, 169
Cell theory, 90, 94, 98-99, 290, 294 Chromatids, 108, 369-370, 373, 377, 404
Cells, classification of, 91 Chromatin, 92-94, 107-108, 139
definition of, 90 Chromosome groups, Trimerotropis, 297
development of, 90-91 Chromosome maps, facing 366, 374-375
differentiation of, 169-170 Chromosomes, 12-13, 21, 73, 75, 82, 86,
ova as, 94 91-93, 107-108, 128, 137-140, 153,
sex, 737 16]-162,294-296,302,312-315,332,
(See also Germ cells) 394-395, 417
Cenozoic Age, 39 definition of, 308
Centromere, 367 diploid,92
Centrosome, 91, 94, 162 distribution of, 129
Cervidae, 57 aberrations in, 410-411
Cervix, 102, 130, 1511-L60, 170-171,202 random, 140
Characters, acquired, 419-420 division of, 92-96, 140, 218, 308
and anatomical modifications, 421- and duplication, 407-408, 410
422 and genes, 91-93
dcfinitions of, 419-420, 423 maternal and paternal, 141
and environmental modifications, numbers of, in farm mammals, 93
422-423 in grain, 92
and functional modifications, 423- in man, 92-93, 285, 299
425 reduction of, 408
inheritance of, 420, 425 and segregation, 311-312, 332
arguments against, 428-429 sex, 92, 138, 214,348,437-438,441,451
practical phase of, 431 and sex determination, 434
summary on, 429-431 Classification, 84
and pat~ological modifications, 425- Clitoris, 103, 123, 161
427. Closebreeding, 486, 518, 521-f.i22
and psychological modifications, Clovercourt Ormsby Royal Blend, Hol-
427-428 stein bull, 631
SUBJECT INDEX 771
Coat color, in cattle, 328, 541 Cows, scrub, 419, 462
in mice, 325-328 (See also Cattle)
in rabbits, 340 Craven Longhorns, 24
Coincidence, 377-379 Creation, separate, theory of, 77-78, 80
Color blindness, 350-351 Creepers, 342
• d.iagram of, 351 Crisscrossing, 472-473
Color inheritance (see Coat color) Crista urethra, 120
Colorado Agricultural Experiment Sta- Cro-Magnon man, 45-46
tion, 654 Crossbred, definition of, 482
Colostrum, 242-243 Crossbreeding, 217-218, 295, 413, 456-
Columbia sheep, 475, 684, 686, 691-692, 457, 464--472, 483, 645
755 diagram of, 349
Columbia Univ~rsity, 295, 365 effects of, within breed, 480-482
Comb form, factors for, 317 of hooded rats, 416
Comet, Shorthorn bull, 503-505 for market, 472-475
pedigree of, 504, 506 and new breeds, 475-479
Conner Verbena, Hampshire gilt, pedi- of sheep, 645--646
gree of, 673 uses of, 479-,j.80
Conversion factors, 592-601 Crossing over, 295, 306, 366-371, 386--
D.H.LA., 593 387, 404
Copulation, 105, 114, 123, 130, 180, 192, cytologically, diagram of, 372
220 diagrams of, 370-374, 377
Cornell University, 177-178, 184, 271- genetic, cytological proof of, 371-373
272, 275-276, 631 percentage of, 373-374
Cornua (see Horns of uterus) problems on, 387-389
Corona radiata, 164 schematic, 367
Corpus.albicans, 163 unequal, diagram of, 407-408
Corpus luteum, 134, 141-143, 147-149, variation from, 397-398
163-164,170,220-221 Crossover percentage, 368-369
absorption of, 163 and linked genes, 370
development of, 163 Cryptorchidism, 118, 202
formation and retrogression of, 144 Cumulus oophorus, 137
functions of, 163-164 Cytology, 82, 86, 92, 290, 294, 434
Correlation studies in classes of livestock, Cytoplasm, 88, 90-92, 137, 297, 444
570-572
Corrie dale sheep, 17, 475, 684, 691 D
Cosmic or astronomic time chart, 40
Cotswold sheep, 475, 684 Dairy-breed associations, 589
Cotyledons, 165-166, 169 Dairy cattle, 25, 249, 256-258, 454-455,
Countess Chloe, Holstein cow, 630 464, 537, 550-551, 555-556, 568-
Cousin relationship, 488-489 570,752
Cow fantilies, 628-630 artificial insemination of, 279-282
charts for, 627-628 auction records of, 5-6
Cow ham Farm Tress King, Guernsey conversion factors for, 593-601
bull, 620 crossbreeding of, 471
Cowper's glands, 114, 122 distribution of type of, 540, 589
Cows, 149, 177, 187, 220, 232, 234--238, on farms in United States, 2-3
241,246,255-258,263-264,279-281, feed conversion in, 633
292,347,352,391,397-398,423-425, fertility of, 177-178
448-449 judging of, 584, 590
records of, in proving sires, 538 number of, in United States, 580
772 flREEDV·;G AND IMPROVEMEj'v''P OF FARM ANIMALS
1)airy cattle, pedigree estimates of, 635 Donkeys, 35, 49

progeny tests for, 564 Dope's Primrose, Percheron mare, 716-
purebred (see Purebred8) 718
record production of, 14-15 Dorset sheep, 353
records of, 582-583, 590, 592-596, 598, Down sheep, 472, 684
745 Drosophila melanogaster, 86, 89, 2e5-"'"
selection in, 580-638 296, 320, 331, 348, 355-356, 365,
Dairy Cattle Breed Averages, 608 368-369,371-374,377-378,381-383,
Dairy Herd Improvement Associations 385-386,398-400,402-406,408-409,
(D.H.I.A.), 537, 581-582, 592-593, 412, 419, 434, 439-442, 444, 447, 545
747 diagram of germ cells in, 435
Dairy Industry, Bureau of (see U.S. eight chromosomes of, 140
Department of Agriculture) inbreeding of, 511
Dairy products, consumption of, 6, 9-10 linkage map for, 376
Danish Landrace swine, 477 nondisjunction in, 435
Darwinism, 81-84, 88 notch deficiency in, 407-408
Death during parturition, 231, 240 relation of sex chromosomes in, 438
Defects, heritable, transmission of, 430 sex-linked factors in, 349-350
Deficiency, 406 Drosophila simulans, 406
Denmark, swine breeding in, 667 Dual-purpose cattle, 16
Dental formulae, 54 selection in, 703-704, 745
Determiners, 356 Ducks, 32
Deviations, 294, 418 Ductless glands, 352
Devon cattle, 23 Duplication, 407-408, 486
Dexter cattle, 343 and inversion, diagram of, 408
Dicotylidae, 63 Durham Shorthorns, 24
Diestrus, 143 Duroc-Jersey swine, 21, 66, 110,323,394,
Diethylstilhestrol, 147 466-468, 481
Dihybrid, 303, 365-366 Duroc Record Association, 549, 679
Dihybrid crosses, 314-316, 332 Dwarfs, 94
diagram of, 315 Dynamometer, 729
Dihybrid ratio, 364 Dystocia, 246
Diploid, 307
Diploid number, 408 E
Dipnoi,76
Disease, 11 Ectoderm, 99
of dairy cattle, 6 Egg (see Ovum)
genital, 223 Egg production, effect of light on, 111
inheritance of, 425-427 Egypt, agriculture in, 34
resistance to, 573 Eileenmere's Effie \V" Grand Champion
Dishley sheep, 401 cow, 200
Dogs, 47, 49, 2f30 Ejaculatory ducts, 120
breeds of, 53 Elastration, 116
domestication of, 50--53 Electron microscope, 73
origin of, 52-53 Elephant, 65
species of, 51 Emaseulatome, 117
Domestication of animals, 47-66 Embryo, 164, 166, 226, 299
Dominance, 313-315, 321-322, 324-325, development of, 166-170
333-334, 3~5-346, 554 Embryology, SD
aspects of, 329-332 comparative, 84-85
Dominant, 30(\ early discoveries in, 99-100
\
SUBJECT INDEX 773
Enca~elllent, 98 Evolution, mechani8111 of, 82, 86, 88-8\\

Endocrine glands, 352 modern theory of, 79
functions of, 220 organic, 75-81
ovaries, 134, 146-150 Ewes, 190-Un, 210, 232, 237-238, 246,
placenta, 160, 227 282,286
. , testes, 105, 109 Exercise, for male animals, 183-186
Endocrine secretions and sex determina- for pregnant females, 235, 246
tion, 443-446, 451 Experiments of :\Iendel, in hyhridization,
Endocrinology, 147 309-311, 348
Endoderm, 99 with peas, 402
England, beginning of breeds in, 22-25, summary of, 310
36
farming in, :;l3, 25--28 F
Environment, 78-79, 82-83, 86, 356, 409,
412,422,424,432,554,610,657-661, F, (first hybrid generation), 307
691 F2 (second hybrid generation), 307
heredity vs., in selection, 546-547 Fallopian tubes, 127, 130--131, 157-158,
and reproduction, 222-223 166, 170
role of, in evolution, 89-90 Family and selection, 566-568, 627-631,
and somatic variation, 418-419 678--679, 695
and training, 291~292 female, chart for, 568
Enzymes, 122, 381 Farm animals, 1, 10, 90--91
Eoanthroplls dawsoni (Piltdown man), 42 characteristics of, 537-541
Eocene period, 39, 85 improvement of, means of, 541
Eohippus, 85-86 marking of, 245
Epididymis, 104, 118-119 newborn, care and feeding of, 242-244
Epistasis, 315, 318n., 334, 393, 524, 554 and principles of inheritance, 379-380
definition of, 322 reproductive cycle in, 143
dominant, 323 sperm of, 180
incompletely duplicate, 323 taxinomic arrangement of, faring 66
other types of, 323-324 value of, per head, 537
recessive, 323 (See also Livestock)
summary on, 324-32.5 Farm income from livestock, 1
Epithelium, 121-122 Farrowing pens, 232-233
Equal-Parent Index, 538-530, 604, 607 Fat Corrected Milk (F.e.M.), 595, 598,
Equidae,53 600--602, 624, 629
Eqlllls caballus, 54, 57, 85-86 Fat deficiency, 184
Przhevalski, 54 Fecundity, 199-200
Estradiol, 147 Fertility, 118, 174-176, 181,215,726
Estrogens, 142, 147-150, 170, 222, 226, definition of, 198
228,230-231,251-254 and exercise, 183, 185
Estrone, 147 lowered, 198--214,219-223
Estrous cycle, 142, 150--151, 187,220--221 and inbreeding, tahle of, 217
Estrus (heat), 142-145, 150--151, 170, and nutrition, 112-113, 175, 184-185,
179-180,187-192,195,227 209-214, 223
and gestation, table of, 235 in selection, 572-573
Evolution, 290, 410, 420 Fertilization, 1, 91, 103, 128-121), 134,
definition of, 77 139-140,157,161-163,166,737-739
dynamic, 423-424 Fetal monsters, 245
evidence for, 84-85 Fetus, 159, 164-16.~, 221l, 230, 236-238,
of horse, 8.5~87 249,286
..
774 BftEEDING AND IMPROVEMENT OF FARM ANIMALS
Fetus, abnormal positions of, 238-240 Genes, definition of, 306

presentation of, types of, 238 and development, 380-384
Feudalism, 26, 28 ... ;lominant, 414, 432, 456, 533
Fillies, selection of, 723-726 _. duplication of, 407-408
Filtrable viruses (filter passers), 73-74, 92 _interaction of, 610-611
Financial King's Interest, Jersey cow, 4AS multiple, 344--348 ., ...
Fish, 101 '--1"ethal action of, 342-344
Fission, 100, 443, 737 Winear order of, 373-374
Fissipedia, 51 "and mutations, 453, 531-532
Flushing, 210 general statements about, 401-403
Follicle (see Graafian follicle) variations from, 398-399
Follicle-stimulating hormone (FSH), 141- number of, in man and farm animals,
142, 153, 221 531
Food, 418 physicochemical nature of, 298, 355-
of animal origin, 6, 9-11 357
(See also Nutrition) recessive, 412, 432, 456
Foremost Prediction, Guernsey bull, 740 recombinations of, table of, 396
Fowl, 49 and sex determination, 441
comb form in, 330 sex-linked, 349
diagram of, 317 partially, 351
crossing of, 329-331, 351-353 size of, 357, 380
diagram of, 352 summary on, 357, 386
(See also Birds; specific kinds of fowl) (See also Heredity; Inheritance)
Freaks, definition of, 411 Genetic equilibrium, 412
Freemartin, 188, 203, 444 Genetics, 12-14, 17, 21-22,82, 86, 290-
French Charollais cattle, 476 302
FSH (see Follicle-stimulating hormone) content of, 291
Futurity index, 727 definition of, 290-291, 332
modern theory of, summary of, 385-
G 38\)
problems of, 296-301
Galactin, 254 problems on, 334--338
Galloway cattle, 23, 465, 472 Genic balance, 444--445
Gametes, 307-308, 311,313,394,396,449 theory of, 437-440
female, 139 Genital ridge, 100
functional, 107 Genitalia, anatomical defects of, 202-203
(See also Ovum) embryological development of, 100-103
Gastrula, 164 female, 134-142, 148, 157-161, 163-
Gaur, 57 166, 186
Geese, 32, 110 diagrams of, 135-137
Gemmules,420 sperm survival in, 129-130
I).,.,Gene frequencies, changing of, 399, 541- male, 109-123
" 546,550 malformation of, 118
Gene manifestations, 355-357 mechanical injury of, 203-204
Gene pairs, 320-322, 332, 339, 364, 370- Genotype, 307, 316-317, 319-320, 324-
371,397, 542-543 325, 396, 483, 534, 537, 554-555, 557
table of, in crosS, 320 Germ cells, 1,82-83,89,91-92, 100, 102-
Genes, 12-13,75,86,88,91-93,186,291- 103, 312-314, 364, 424, 428
292,295-296,299,301,313-317,417, division of, 93-96, 106, 367-369
740 . origin of, 96
changes in blo.cks of, 404-417 table of types of, 394-395
,,
SUBJECT INDEX 775
Germ plasm, 18, 20, 88, 91-92, 96, 301, Herd tests, 18, 747
418-419, 425 Hereditary transmission of character-
schematic representation of continuity istics (see Characters)
of, 421, 429 Heredity, 1, 3, 12~14, 84, 96, 103, 215--
Germinal epithelium, 102, 105, 135-137 217, 223, 292~294
• -<Germinal variation, theory of, 80 in beef cattle, 697
Gestation, 230 definition of, 307
and estrus, table of, 235 vs. environment, in selection, 546~547
ovarian hormones during, 226-227 in horses, 713
table of, 234 principles of, 304~333, 339~357, 364~
Giants, 94 387
Giraffidae, 57 in sheep, 692~694
Goats, 32, 35} 60~63, 457 in swine, 661~662
milk yield of, 15 Hereford cattle, 23, 318~319, 413, 415,
Gonadotropic hormones (see Hormones) 465, 469~470, 472, 476, 746
Gonads, 102, 114, 116 Hermaphrodites, 444
Ganium, 75 Hermaphroditism, 100, 102
Graafian follicle, 99, 135~143, 151, 153, Hernia, 202
170, 220~221 scrotal, 118, 202
diagram of, 144 Heteroploids, 307, 409
Grading, 459~465 Heteroploidy, 408~409
Grain surpluses, 10 Heterosis (hybrid vigor), 218, 456-457
Greece, 304 definition of, 482
agriculture and animal husbandry in, Heterotypic division, 108, 138
32~34 Heterozygosity, 13, 17,295,301,496-499,
Grimaldi race, 44~45 731
Guernsey cattle, 21, 391, 448, 471, 540, controlled, 752~753
550, 570, 593~595, 600, 604, 608, 616 Heterozygote, 307, 329
Guinea fowl, 32 Hinny, 218, 457, 459
Guinea-pigs, 430, 511~512 Hippopotamidae, 64
Gynandromorphs, 442~443 Hippopotamuses, 64
Hoard, W. D. & Sons, Publishing Co., 616
H Hogs, 30, 32, 48~49, 454, 555~556
carcasses of, 658, 668
Hambletonian 10, 424 nutrition of, 657~658
Hamprace swine, 477 in United States, 2, 15
Hampshire Down sheep, 475 (See also Swine)
Hampshire swine, 21, 477 Holstein cattle, 21, 393, 415, 448, 471
Hampshire Swine Registry Association, 487,54O-541,544~545,550,570, 581,
655, 669 593, 599~600, 608, 738, 755
Haploid, 307 Holstein-Friesian cattle, 487, 515, 539,
Haploid number, 408 591, 752
Haploidy, 409 Holstein-Friesian Red Book, 619
Harvard University, 429 Holstein indexes, graph of, 393
Health in selection, 573~574 Hominidae, 41
Heat (see Estrus) Homo heidelbergensis (Heidelberg man),
Heidelberg man, 42~43 42~43
Heifers, selection of, 637 Homo neanderthalensis (Neanderthal
through herd knowledge, 624--633 man), 43~44
Herd books, 17, 299~300 Homo sapiens, 38, 42~46, 66
Herd Improvement Test, 608 (See also Man)
776 B[tEEDING AND IMPROVEMENT OF FARM ANIMALS
Homotypic division, 108 Horses, selection in, 708-733

Homozygosity, 13, 17,301,496-499,731, summary on, 733
752 sire indexes in, 727-729
Homozygote, 307, 329 as source of power, 57, 708-711, 733
Hormone levels, 14 "steppe," 55
Hormones, 140-141, 255-257, 352, 444- Thoroughbred, 17-18 ...
446, 740 type in, 729-731
distribution of, 114 types of, 53-54, 712
female (ovarian), 142, 146-150, 170, wild,55
252 (See also Fillies; Mares, Stallion;
effect of, on genital organs, 148-149 specific names of horses)
during gestation, 226-227 Hunting stage, 46
and ovulation, 151-154, 161 Hybrid corn, 517
and secondary sex characteristics, Hybrid vigor, 456-457, 681
149 definition of, 482
structural formula of, 146 Hybridization, 457
follicle-stimulating (FSH) (see Follicl",- (See also Experiments of Mendel)
stimulating hormones) Hybrids, 140, 218-219, 293, 301, 305,
gonadotropic, 108-109, 112, 140, 142, 308-310, 330
146, 153, 219-222 commercial, 458-459
in terstitial-cell-stim ula ting (lOS H) , definition of, 306
109 intergeneric, 410
lactogenic, 253-255
luteinizing (LH) (see Luteinizing hor- I
mone)
male (androgen), 105, 109-110, 113- Ibex, 62
ll5 ICSH (see Interstitial-ceIl-stimulating
pituitary, 252 hormone)
sex, 147,209,220,222,353 Illinois, University of, College of Agri-
thyroid (thyroxin), ll2 culture, 466
Horns of uterus, 158-159, 170 Illinois Experiment Station, 182, 517,649
Horse associations, 8, 758 Implantation, 163, 170
Horse problems, 191-192 Inbred line, definition of, 482
Horses, 25, 30, 32, 35, 48-49, 423-424, Inbreeding, 217-218, 295, 411-412, 456,
457,551, 568 476, 486-488, 681, 752
adaptability of, 714--715 through brother-sister mating, 501-504
Arabian, 56 coefficient of, 495-499, 515
breeding efficiency in, 731-732 table for, 501
buying of, 713-715 conclusions on, 522-525
domestication of, 53-57 examples of, 505-506
draft, 728-731 experimental data on, 511-517
evolution of, 85-87 formula for, 495-496
on farms in United States, 2-3, 15 good results from, 510
fertility of, 176 harm from, 510
"forest," 55 through parent-offspring mating, 500,
Greek, 33, 55 521
lethals in, 343 pedigree of five generations of, 524
and mechanical power, 708-710 role of, 506, 508-509
origin of, 54-56 summary on, 525
pedigrees of, 71,2, 724-725 usc of, 517-518
for pleasure purposes, 3, 57 lncasement, 98-\lH
\
SUBJECT INDEX 777
Incross, definition of, 482 Iowa State College, 434
I~crossbred, definition of, 482 Iron Age, 47
Independent assortment, 306, 311-313
Indexes, 538 J
bull, 601-613
• _ equal-parent, 538-539, 604, 607 Jackal, 51
futurity, 727 Jacks, 711, 759
pig, 673, 678 Java, ape man of, 39
regression, 606-613 Jay Farceur, Belgian stallion 742
sheep, 694-695 Jennets, 45\), 75\) ,
sire, in horses, 727-729 Jersey cattle, 17,448, 462, 471,540, 550,
Indiana Experiment Station, 126, 182 570, 5\)3, 600, 608, 755
Individuality ,and selection, 552-557, JJ Gertrudis, champion Hereford cow
678,695 702 '
Indo-European race, 35
Infertility, 112, 188, 228, 285 K
Infundibulum, lol, 170
Inheritance, 79, 82-83, 103, 294-296, Kansas Agricultural Experiment Station
299-300, 302, 305, 322, 339 3i8n. '
of acquired characters, 419 / Kidneys, 101
(See also Characters, acquired) King of the Ormsoys, Holstein bull,
blending, 330 charts for, 623-624
bridge of, 418 King Ranch, 475-476
gene mechanism of, 86, 88-89, 96, 355-
357, 412, 417
gene theory of, evidence sup'porting,
384-385 Labia majora, 103
in hooded rats, 414 Lactation, 170, 210, 220, 250, 253-258,
laws of, 306 353
:\1endclian theory of, 311-313, 329, 354 Lactogenic hormone, 253-255
multiple factors in, 344-348, 357 Lafayette, Grand Champion Percheron
physical basis of, 307-308 stallion, 722
principles of, and mental traits, 427- I..aletto, Percheron scallion, 723
428 Lamarckism, 431
probability in, 354-355 vs. Weismannism, 420-421
problems on, 357-363 Lambs, carcasses of, 683
sex-influenced, 352-354 production of, in relatioll to face Cover-
sex-linked (see Sex-linked factors) ing, 652-653
summary on, 357 study of, 689-691
Insemination, artificial (see Artificial Lancashire cattle, 24
insemination) Landrace swine, 468, 477-478, 667-668
Interference, 306, 377-379 Lasater Ranch, 476
International Livestock Exposition, 403, Lassie of W. T., Grand Champion cow
465 703 '
Intersexes, 437-438, 440 Leicester sheep, 24, 475
In tersti tial-cell-stim ula ting hormone Leta, Percheron filly, pedigree of, 72.5
(ICSH), 109 Lethal action of genes, 342-344
Inversion, 400 Lethals, 342-343, 447, 573
Invertebrates, 38, 68 balanced, 403
Iowa Agricultural Experiment Station, induced, diagram of, 402
463, 467-468, 655n., 677 list of, 343-344
.,
•
778 BllEEDING AND IMPROVEMENT OF FARM ANIMALS
Leydig cells, 104, 113 Mammogen theory, 253

LH (see Luteinizing hormone) Man, 77-78, 104, 353
Libretto, Percheron stallion, 716-717 blood types in, 341
Light, 418 color inheritance in, 346
effect of, on reproduction, 222-223 early, and animal domestication, 38-66
on sexual cycle, 146 early forerunners of, 39-46 ... 00>
on spermatogenesis, 111-112 lethals in, 344
Lincoln sheep, 475, 684 modern (homo sapiens), 38, 42-46, 66
Lincolnshire Ox, 23 sex-linked factors in, 350-351
Linebreeding, 476, 480, 486, 518-523 sex ratios in, 434
Linin network, 92 • time chart of, 40
Linkage, 295, 306, 365-366, 374-377, 386, Manor system of farming, 26
397 Mansfield, Morgan stallioJ;l, 744
Linkage groups, limitation of, 30(), 379 Mares, 178, 209, 227-228, 232,234-238,
relations of, 379 242, 246, 260, 281-282, 726
Livestock, characteristics of, 537-541 (See also Horses)
income from, 1 Markhor,62
number and value of, in United States, Marsupials, 76
2-4,9 Massachusetts, University of, 630, 633,
practical improvement of, 453-483 716, 724
... (See also Farm animals) Massachusetts Experiment Station, 211
Livestock record associations, 757-760 Massachusetts State College, 448, 716
Livestock shows, 19, 571-577 Maternal impressions, 431-432
Livestock trends in United States, 3-4 Maternity barns or pens, 231-232
Longhorn cattle, 23 Mating, :Mendelian formula for system
Loyal Alumnus 4th, Grand Champion of,498
steer, 466 Mauchamp sheep, 401
Lumen, 106, 108, 160 May Rose II, Guernsey cow, 739
Luteinizing hormone (LH), 109, 141-142, Meat, consumption of, 9-10
153, 163, 221 production of, 66
Lymantria dispar, 437 Meat animals, 3, 23, 66, 583, 751
and environment, 657-661
M judging of, 646-647
and production, 568, 570-571
Maidstone Dairy Queen, Grand Cham- selection in, 644--704
pion cow, 636 through production records, 655-657
Maine Agricultural Experiment Station, Mediastinum testis, 118
566,569 Meiosis, 367, 394
Major MacFarlane, champion gelding, of germ cells, diagram of, 107, 141
463 Meiotic divisions, 106
:Mammals, 76-77, 103, 110, 131, 249 Mendelian formula for system of mating,
classes of, facing 64 498
farm, sex ratios in, 434 Mendelian principles of inheritance, 290,
(See also Farm animals) 298, 300, 306, 329, 354, 365, 386
germ-cell differentiation in, 449 Mendelian terms, 306-307
higher, sex establishment in, 446 Mendelism, 307
placental, 76 essence of, 308-309
Mammary glands, 226, 241 Menstruation, 143
development of, 249-253 Merino sheep, 27, 353, 400, 472, 475,682-
endocrine regulation of, 251-253, 258 684
stimulation of; 252, 257-258 Mesoderm, 99, 101
\
! \
SUBJECT INDEX 779
Mesohippus, 85 Morgan horses, 20, 66
Mesonephric duct, 101 breeding of, 732-733
Mesonephros, 101 Morula, 168
Mesosalpinx, 157 Mouflon sheep, 62-63, 648
Mesozoic Age, 40 Mt. Hope bull index, 603
~etal ages, 47 Mullerian duct, 102
Metaphase, 93-94, 367 Mules, 431, 457-459, 517
Metestrus, 142 on farms in United States, 2-3
Mice, 426, 428, 430, 433 Multifactor crosses, 319-320
breeding for coat colors in, 319, 325- lVlultiple births, 156, 162, 215
329 Mutabiljty. iSh 79
inbred, decreasing fertility in, 217 ~pposition to, 80--81
Microscopes, 73, 269 Mutation, 80, 88, 222, 295, 343, 373, 424,
Middlehorn cattle, 23 741
Milk, ejection of, 256-258 definition of, 411 -
of goats, 15 direct, 401
yield of, 4, 6 gene, general statements about, 401-
average, 6 403
record, 14 variation from, 398-401
(See also Lactation) reverse, 401
Milk fever, 229, 241, 256 theory of, 301
Milk production, 59-60, 455, 471, 538- Mutilatjons,-421
540, 550, 555-556, 564-565, 569-570, MUtual Ormsby Jewel Alice (Holstein
580-624, 629 cow), pedigree of, 492-493
Milkdale Aristocrat Rag Apple, Hol- MW Larry Mixer 1st, champion Here-
stein bull, 261 ford bull, 702
Mind, development of, 77
Mineral deficiencies, 213-214, 229, 241 N
Minnesota Experiment Station, 215, 472-
473,482 " National Agricultural Research Center,
Minnesota No.1, boar, 478 Beltsville, Md., 689, 751
Minnesota No.2, boar, 478 National Barrow Show, 645
Miocene Age, 85 National Dairy Show, 448, 510
Missouri, University of, 272, 276 National Research Council, 213
Missouri Experiment Station, 177, 182 Natural selection, 83-84 .
Mitosis, 367 Neanderthal man, 43-44
germinal, HO Nebraska, University of, 177
Modifications, 419-420 Neolithic Age, 44, 47-49, 53, 62
anatomical, 421-422 Nephrotome, 101
environmental,422-423 N ether hall Swanky Dan, Ayrshire bull,
functional, 423-425 587
pathological, 425-427 New England Homestead, 4-5, 8
psychological, 427-428 New Era's Type, Duroc sow, 670
Moisture, 418 New Jersey Agricultural Experiment
'Monestrus animals, 145 Station, 515
Monkey, Santa Gertrudis bull, 476 New Stone (Neolithic) Age, 43-44, 48-49,
Monohybrid, 306 53, 62
Monohybrid crosses, 313-314, 332 N ew York Artificial Breeders' Coopera-
Montana Agricultural Experiment Sta- tive, 261, 631
tion, 469, 477, 751 N ew York State, study of seasonal fer-
Montana No.1, Hamprace hog, 477 tility in, 111
New York State Fair, 555 Ovum, ova, 1, 83, 91, 96, 99-100, 102,
Nicking, 412-413, 457,473, 610, 619, 713 134-142, 158, 220
Nondisjunction, 214, 435--437 cyclic development of, 138-142
Notch deficiency, inheritance of, 407 developmental organization of, 298-
Nucleus, 73, 91-94, 108, 137, 139, 170 299
Nutrition, of bacon hogs, 657-658 fertilization of, 161-163, 170 .' -
of beef cattle, 698-700 fertilized, transplantation of, 285-287
and fertility, 112-113, 175, 184-185, vitality of, 155
209-214 Oxen, 25, 35, 59-60
of mother, 241 Oxford Down sheep, 475
of newborn, 242-244
of sheep, 658 P
(Sec also Food)
Nymphomania, 189, 221 P (parental generation), 307
Paedogenesis, 100
o Paleolithic (Old Stone) Age, 43-49, 55
Paleontology, 80, 85
Paleozoic Age, 40
Ohio Experiment Station, 190
Ohio State University, 716 Pangenesis, 78, 82
Oklahoma Experiment Station, 155, 210, Parallel induction, 418
261, 481 Parathyroid, 229, 256
Parthenogenesis, 100, 162, 384, 409
Old Stone (Paleolithic) Age, 43-49, 55
artificial, 740
Oligocene period, 39, 85
Parturition, 134-137, 164, 193,204, 220,
Open-field system of farming, 22, 26
226, 230--242
Orchitis, 206, 219
act of, 236-238
Organic evolution, 75-81
assistance at, 238, 240
defi,nition of, 77
care of mother at, 241-242
Organisms, evolution of (sec Organic
care of young at, 240--241
evolution)
complications of, 245-247
generation of, 68-72 duration of, 237
single-celled, 72-77 physiological regulation of, 230--231
Oriental agriculture, 34-36 place of, 231-233
Orthogenesis, 84, 89-90 and sanitation, 233
Os uteri, 135 preliminaries to, 235-236
Outbreeding, 456--483, 525-526 significance of, 231
Outcross, definition of, 482 time of, 233-235
Outcrossing, 480, 483 and types of presentation, 238
Ovarian stroma, 136,-137 Pasang, 62
Ovaries, 96, 102-103, 113, 134-140, 151- Peccaries, 64-65
157 Pecora, 48, 57
functions of, dual, 134-135, 170 Pedigrees, 17,48,215,491,746
hormonal, 146-148, 170, 353 bracket-type, 558, 561
hormonal regulation of, 140-142 of bulls, 492-493
schematic diagram of, 144 of cattle, blood and chromosome distri-
Ovis, 60--61 bution in, 494
Ovogenesis, 135, 136, 152 dairy, estimates of, 635
Ovulation, 151-155 early or old-style, 558--559, 561
and breeding ,time, 154-155, 179, 187- female side of, 560--562
191,220 ' of five generations of inbreeding, 524
super-, 155--1,56 of horses, 712, 724-725
I
\
SUBJECT INDEX
781
Peuigrees, identity of, 489 Poland-China swine, 21, 66, 215
, 466,
to illustrate linebreeding, 520 468, 478, 482, 649, 654
of inbred pigs, 514 Polycotyledonary animals, 165
of outbred animal, 498 Polyestrous animals, 142
pictorial, Galton's law, 495 Polymerization, 74
.....schematic, 488-491, 559 Polyploids, 295, 307, 409
and selection, 553, 557-563 commercial interest in, 410
of swine, 673, 678 Polyploidy, 409-411
Peking man, 41 Ponies, 423
Pembroke cattle, 23 Pony associations, 758
Penis, 102-103, 114, 122-123, 202, 266- Population of United States, farm mam-
268 mal, 3, 11
Pentoila's Favohte, Percheron mare, 716, human, 3, 6, 10
718, 723-725 Position-effect theory, 408
Percentage of blood, 494-495, 501 Posterior-lobe hormones (see Pituitary
Percheron horses, 545, 551 gland)
family chart of, 716-723 Poultry, mass selection in, 566
Performance records (see Records) (See also Fowl: specific kinds of
Perissodactyls, 54 poultry)
Pfliigers egg cords, 102, 138 Preformation, 98
Phacochaerus, 64 Pregnancy, 142, 157, 173-174, 177, 195,
Phallus, 103 226-230, 245
Phenotype, 307, 316-317, 320-321, 324- diagnosis of, 227-229
325, 332, 483, 534, 536-537, 552-554,. endocrine changes during, 229
557 premature, 178
pseudo-, 229-230
Physicochemical entities in heredity,
termination of, 230
296-298
(See also Abortion)
Physiology, 98
Prepotency, 413-414
Pigeons, 32
Primates, 39, 143
Pigs, 26, 465-468, 477
Prince Sunbeam 249th, Grand Cham .
inbred, pedigree of, 514 bull, 700 PlOn
in United States, 2
Private ownership of land, 26, 28 33
(See also Swine)
Probability, 354-355 '
Piltdown man, 42 Production, milk (see Milk product' )
Pinnepedia, 51 " lOn
m selectlOn, 508-572, 655-657
Pithecanthropus I!rectus, 39-41, 43 Production averages and selection 574-
Pittsfield, Mass., first livestock show, 571 575 '
Pituitary gland, 142, 229, 253 Production registry of swine, Purpose f
hormones of, anterior-lobe, 150, 153, M2 0,
163, 219, 230-231, 252, 254 rules for, 669-671
posterior-lobe, 230, 257 Proestrus, 143
Placenta; 164-16G Progeny-performance records (see Rec-
endocrine functions of, 227 ords)
Plant culture, 304 Progeny-performance test, 553, 563 695
early, 46-47 practical shortcomings of, 615-618
pre-Mendelian, 293-294 and selection, 564-5G6, 679-680
Plants, disease-resistant, 573 Progesterone, r'42-143, 147-150,220 226
somatic variation in, 418-419 231,252-254 ' J,
Pleistocene Age, 39, 41-42, 62, 85 functions of, 163-164, 170, 230
Pliocene Age, 62, 85 Progestin, 148
Prolactin, 254 Rats, sex ratios in, 448

Prolificacy, 199-200, 215-216 white, 428
Pronephros, 101 Record card, for dairy herd, 625-{)26
Prophase, 93-94, 367 for pigs, 677
Prostate, 102, 121-122 for sheep, 687-688
I'rotein, 74-75, 12\'> lDr sows, £,74-075 .--
Proterozoic Age, 40 Records, 744-747
Protoplasm, 90 of beef cattle, 696, 698-701
Pseudopregnancy, 229-230 breeding, 233, 244-245
Psychology, animal, 180, 186, 258 of dairy cattle, 582-583, 592-596, 598,
Psycho zoic Age, 40 624-633, 636-637
Puberty, 114, 135, 138, 187 of female families, 558
age of, 177 performance, 691, 696~ 596
I'urdue Experiment StatiDn, 100-111, production, 569
118, 154, 157, 177 of meat animals, 655-657, 669-671,
Purdue University, 664, 703 687-689
Pure lines, 532, 752-753 Red Dane cattle, 451
Purebred breed associations, 18-20, 479 Red Glamorgan cattle, 23
formation of, 6, 17, 21 Red Polled cattle, 703
Purebred-crossbred comparison, 467 Reduction, 435
Purebreds, 17-20,413-414,453,459-465, Regional Swine Breeding Laboratory,
479,74\'> 481, 663-604, 680-681, 753-754
cattle, 17-18 Registered animals, 5-6, 17
milk production of, 4 number of, table, 7-8
value of, 4-6 Regression, 606-607
importation of, 6 Regression Index, 606-613, 619
records of, 14-16 Relationship of animals, 96, 488-495
coefficient of, 499-505, 535n.
Q collateral, 488-493
direct, 488, 4\)1, 4\)4-4\)5
Quaggas, 431, 457
Relaxin, 230-231
Quarter Horse, 21
Quartile method for indexing bulls, 601 Reproduction, 3, 68, 86, 91, 98, '737
chronological order of, 170
R and environment, 112, 222-223
female's part in, 134-171
Rabbits, 226, 228, 231, 254, 276, 430, forms of, 1OD-101
739-740 male's part in, 98-131
crossbreeding of, 339-341, 346 and nutrition, 209-214, 223
Race, 307 and physiological disturbances, 219-
Raceborses, 56 222
Radiation, atomic, and reproduction, 222 Reproductive efficiency, 173-196
Rambouillet sheep, 15,353,472,475,645, definition of, 200
651, 653, 682-683, 686, 692, 694, 755 influence of age on, 176-183
Rams, 112, 123, 180-182, 191, 219, 267- standards of, 173-176
2G8, 273, 278 Reproductive failure, causes of, 200-201
Range Livestock Experiment Station, Reproductive phenomena, table of, 194
Miles City, Mont., 175, 469, 477, Reptiles, 76, 88, 101
573, 751 Rete testes, 104, 106
Rats, )18, 148, 226, 228, 252, 430 Revelation, Percheron stallion, 720
hooded, 414-415 Revelex Daisy's Warrior, Grand Cham-
inbreeding of, 511 pion bull, 636
\
SUBJECT INDEX 783
Reversion, 432 Selection, and hooded rats, 416

Rhinoceroses, 54 in horses, 708--733
Rhodesian man, 45 individuality and, 552-557
Ring Gold Lady Dora, Hampshire sow, and market, 547-549
669 in meat animals, 644-704
~dents, 77 beef cattle, 696-703
Roman Empire, agriculture in, 30-32 dual-purpose cattle, 703-704
Rosehill Fayne Wayne, Holstein-Frie- sheep, 682-696
sian cow, 585 summary on, 704
Royal's Rapture of Lee's Hill, Grand swine, 661-682
Champion cow, 634 and modification, 419
Ryeland sheep, 27 natural, 528
., pedigree and, 553, 557-564
S production, type, and show ring in,
553, 568--572
Saddlebred horses, 710 progeny test and, 553, 564-566
St. Bernard dogs, 52-53 purpose of, 576
St. Stephens College, 426 summary on, 576-577
Salmonella pullorum, 427, 573 Self-fertilization, 498
Saltations, 83 graph of, 499
San Carlos Indian Reservation, 178 Selfing, 411-412, 498
Sanitation, 244, 266 Semen, 110, 120, 122-123, 179, 182, 184,
during parturition, 233, 241 186
Santa Gertrudis cattle, 475-476, 660 chemical composition of, 125
Scandinavia, cattle breeding in, 29 dilution of, 275-277, 279, 284, 737
Science, definition of, 68, 78 evaluation of, 268-274
Scouring, 241 examination of, 267
Scrotal hernia, 118, 202 methods of collection of, 264-268
Scrotum, 102-103, 114, 116-118, 203 pH of, 125-126, 274, 450
Seasonal periodicity, 145-146 physiocochemical properties of, 123-
Secondary sexual characteristics, 105, 126
109, 113-115 processing and shipment of, 274-279,
and ovarian hormones, 148 737-738
Segmental interchange, 404 purity of, 274
Segregation, 306, 311-312, 333 recovery of, from vagina, 267
diagram of, 312 and spermatozoa, quantitative data
Selection, 1, 3, 21, 53, 79, 293, 528--577, on, 124
748-749 tests of, 124-126
artificial, 529 analysis of, 126
averages and, 574-575 volume of, 268--269
balance and, 575-576 Sf·minal fluid, 100
basis of, 529-537 Seminal vesicles, 120-121
and ]jteeding, 453-456 Seminiferous tubule, 104-106, 108--110,
breeds and, 549-552 118, 120
in dairy cattle, 580-638 semischematic figure of, 105
effect of, on sex ratio, 448-449 Semiplacenta, 165
family and, 566-568 Serosa, 159-160
fertility and health in, 572-574 Sertoli, cells of, 102, 105, 108
and gene frequencies, 455 Service, 173
as guide to variation, 414-416 amount for males, 180~ 183
heredity vs. environment in, 546-547 excessive, 181-182
784 l!i'REEDING AND IMPROVEMENT OF FARM ANIMALS
Service, first, 174, 180-183 SinanthropU8 pekinensis (Peking man),

injury from, 204 41-42
Sex chromosomes (see Chromosomes) Sir Bess Ormsby May 2nd, Holstein butl,
Sex control, 449-451 522
Sex determination, 102, 433-451 Smithfield Market, 24
chromosomes and, 434-437 Sni-a-Bar ranch, 454
conclusions on, 449 Soil fertility, 1, 10
endocrine function in, 443-446, 451 Solo man, 41, 45
genes and, 441-443 Soma, 92, 419, 424, 432
Sex-linked factors, 348-351 Somatic mitosis (see Cell division)
in Drosophila, 349-350 Somatoplasm, 15, 91, 96
in man, 350-351 Southdown sheep, 749
WZ type of, in poultry, 351-352 Sows, 232-233, 237-238,., 240-241, 282,
XY type of, 348 391
Sex ratios, 433-434 (See also Swine)
conditions influencing, 447-448 Spaying, 157, 445
effect of selection on, 448-449 Species, mutability of, 81
Spermatic cord, 104, 113
and environment, 447
in King's inbred rats, 448 Spermatids, 105
Spermatocytes, 106, 108
Sex reversal, 445-447
Spermatogenesis, 109-110, 120
Sex-differentiators, 441 ,,.,
diagram of, 108
Sex-mosaics, 442-443
effect of light and temperature on, 111-
Sex-producers, 441
113, 117
Sexual cycle, 142-145
seasonal, 110-111
Sexual maturity, 108, 110
Spermatogonia, 105-106
Sheep, 15, 21, 29-30, 32, 35, 48-49, 50,
Spermatozoa (sperm), 1, 83, 91, 99, 102,
228, 353, 421-422, 457, 472, 475,
104-113, 118-119, 121-126, 151,450
533,551
abnormal,273-274
breeding of, 175-175, 178-179, 190 anatomy of, 125-127
in England, 22-28 dual function of, 127-128
groups of, 51 metabolism of, 272
lethals in, 344
motility of, 91, 108, 119, 126, 129-131,
origin and domestication of, 62-63 182, 259-272, 738
purebred, 17 duration of, 271
selection of, 648-553, 582-696 physiology of, 127-129
in United States, 2 rate of travel of, 130
(See also Ewes; Lambs; Rams; specific and semen, quantitive data on, 124
names of sheep) survival of, 130
Sheep associations, 759 in female genital tract, 129-130
Sheep Experiment Station, 755 transport of, through uterine tubes,
Shorthorn cattle, 15-17,23,313,329-330, 130-131,170
465,469-470,472,457,523,541,547, types of, 128
593,595 Stallion registration boards, 711
Bates' Dutchess, 214, 216 Stallions, 118, 180, 183, 191-192, 203-
Celtic, 59 204, 267-258, 278, 711-712
herd book for, 17 selection of, 726-727
SIlow ring and selection, 553, 571-572, Standardbred Horse, 56, 710
618 Standardbred Register, 18
S~ropshire sheep, 594 Steers, 391
Silken Lady's Ruby of F., Jersey cow, 199 Shorthorn, performance record of, '696
SUBJECT INDEX 785
Sterility, 120-121, 140, 173-174, 186, 188, Tapirs, 54

200, 202-203, 209, 222-223, 247 Targhee sheep, 684, 686, 691-692, 755
Tarpan horse, 55
• definition of, 198
genetic causes of, 214-217 Taxonomy, 84
Teat~ 249
inherited, 215-217
, ~nd systems of breeding, 217-219 Telegony, 431
Teleology, 69
temporary, 117
Telophase, 93-94, 367
Sterilization, 152
Temperature, 418
Stilbestrol, 252-253 effect of, on reproduction, 222-223
Stillbirth, 240 on sexual cycle, 146
Stimuli, external, 418 on sperm and semen, 271, 274, 277-
Stroma, 102
279
ovarian, 103, )36-137
on spermatogenesis, 112-113
Successful Farming, 743n.
Tennessee Walking Horse, 21
Suffolk horses, 544
Teratology, 80
Suffolk sheep, 648 Teratoma (fetal monsters), 245
Suidae, 63-64 Testes, 102, 106, 117,203
Suina, 64 and age, 109-110
Superovulation, 155-156, 287
dysfunction of, 109
Supersexes, 438
Supreme Fancy 2nd, Grand Champion of fowl, 109
functions of, 105, 131
sow, 679 hormonal, 113-114, 119, 131, 146, 353
Sus, genus, domesticated and wild species
hormonal regulation of, 108
of,64-65
(See also Castration)
Sussex cattle, 23
Swine, 63-66, 232, 433, 472-473, 477-478, Testicles, 96, 102, 117
Testosterone, 105, 113-114, 118
481, 551
Tetrad,108
breeding age of, 177
Tetraploids, 410
breeds of, 66
Texas, University of, 441
crossbred, analysis of, 469 Texas Agricultural Experiment Station,
fertility of, 175, 216
inbreeding of, 513, 516-517 650
Texas Agricultural and Mechanical Col-
lethals in, 344
origin and domestication of, 65-66 lege, 458
Texas Longhorns, 476
progeny tests for, 564
selection of, 649-650, 654-655, 657- Theca, 137
Thoroughbreds, 20, 727
658, 661-682
pedigrees for, 17-18
sex ratio of, 448
(See also Boars; Hogs; Pigs; Sows; Thyroid, 229, 255
specific names of Swine) Thyroxine, 112
Tiptop Princess 3rd, Grand Champion
Swine associations, 759-760
Swine problems, 189-190 sow, 679
Sybil's !;i,amboge, Jersey bull, 505-508, Top Set, Duroc boar, 549
Topcross, definition of, 482
510 Topcrossbred, definition of, 482
pedigree of, 507
Synapsis, 106, 108, 138-140, 366-367, Topincross, definition of, 482
Translocation, 404
377,405
reciprocal, 404-406
T TrichononiaHis, bovine venereal, 207-208,
246
Tahr, (\2
Trihybrid, definition of, 307
Tamworth Rwine, 478
Trihybrid crosses, 316, 318 U.S. Range Livestock Experiment Sta-

'i'rimerotropsis, eight chromosome groups tion, Miles City, Mont., 175, 469,
of,297 477, 573, 751
Triple crossing, 468, 473-474 Urials,61
Triplets, 199, 204, 246 Urinary organs of bull, 104
Triploids, 410 Urogenital ridge, 101
Trophoblast, 164-165, 168 Utah Station, 210
Trotting horses, 20 Uterine milk, 166
Tubuli recti, 104, 106 Uterine mucosa, 159
Tunica albuginea, 102-104, 136 Uterine tubes, 102, 130
Ture, 62 Uterus, 102, 131, 147, 149-150,156, 158r,
Turkeys, 110, 267 159,162,167,170-171,226,231
Twins, 188-190, 203, 246, 444, 531 horns of, 158-159
Two-gene ratios, modified, 32~323, 325 masculinus, 120
Type, in dairy cattle, 581-592
in horses, 72\)-731 V
in meat animals, 646-655
Vaccination of calves, 206
U Vagina, 102, 130, 148, 160-161,266-267
artificial, diagram of, 265
Udders, 249-250, 256-258, 585, 588, 590 Vaginal epithelium, 160
involution of, 250-251, 255 Vaginitis, 208-209
Ultraviolet-light photography, 73 Variance, 534, 554
Unguiculates, 51 and environment, 546
Ungulates, 48, 63, 77 hereditary, 535-537
UnisexuaJism, 100, 3\)7 Variation, 3, 53, 80, 96, 528
Unit characters, 306 autogenetic, 392
United Duroc Record Association, 669- breeding systems and, 411-412
causes and types of, 391-3\)2
670
from chromosomal aberrations, 403-
United States, cattle in, 2-3
404
dairy cattle in, 203 j
! from crossing over, 3\)7-398
hogs in, 2, 15
definition of, 307
pigs in, 2
environmental, 82-83, 8\), 301
sheep in, 2, 15
exogenetic, 392
U.S. Department of Agriculture, 104,
functional, 415
165, 168, 176, 205, 237, 23\), 464n.,
from gene mutations, 398-401
468, 475, 650, 6\)7, 701, 703, 745, 752
genetic, 417-419, 447
Bureau of Agricultpral. Economics, 11 kinds of, 533-534
B-Ilreau of Animal Husbamiry, 16,-1~5,<-'j germinal, 82, 89, 301, 3\)2, 429
195; 206, 651 . ..
guided by selection, 414-416
'! Bureau of•.Anirp.al. Industry, (\83 l 6\)5, hereditary, 84
698, 751.' 75~ plus and minus, 415
Farm, 467, 477 j principles of, 390-432
"'Sheep Experiment Station, 685, ·755 .;' random, 84
Bureau of Dairy Industry, 166, 186, from recombinations, 394
261,471, 537-538, 565, 572n., 582, somatic, 3\)2, 418
747,751,753, 755 environment and, 418-419
United Statel'l Livestock Sanitary Asso- as tool, 392-394
ciation, 206 transgressi ve, 34()
United Stat~\ Morgan H()r~e Farm, 732 types and frequency of, -!J6-417
SUBJECT INDEX ; 787
Vas deferens, 119-120 Wheat-germ oil, 213
Vasa efferentia, 119 Wisconsin, University of, 155-156, 207,
Vasectomy, 120 221,286
Vertebrates, 38, 68, 75-77 Wisconsin Agricultural Experiment Sta-
Vesicula seminalis, 104 tion, 182, 612
~tigial structures, 85 Wisconsin Swine Selection Cooperative,
Veterinarians, 101, 116, 1~18f;"192;:: 672
195,247 Wolffian duct, 101-102
Vibrio fetus, 208 >. .' Wool, 23, 25, 48, 63, 649-653, 682-{l86,
Villi, 165 689
Viruses, filtrable, 73, 92 short, 27
Vitamin deficiencies, 211-214, 242, 246 (See also Sheep)
Vitamins, A, 11_, 211-213, 242-243, 246·
B,212 x
B-complex, 298
C,212-213 X rays, 2H5, 398, 741
D, 211, 243 and reproduction, 222-223
E,213
and reproduction, 211-21.,:':·'· y
. -<Volvox, 75 . ""'
'/ Vulva, 103, 161 .. ' . Yak,57-58
Yeast, 73
W
Yorkshire swine, 394, 467-468, 478
Wadjak man, 45
Wart hogs, 64-65 Z
Washington, State College of, 687-688
Wavemaster Stilts, Grand Champion Zehras, 53-54, 431, 457, 459
boar, 549 Zebroid, 459
\Yeismannism, 421, 431 Zebu, 426, 459
West Highlands cattle, 23 Zona peUucida, IG4, 166
VI'estern Sheep Breeding Laboratory, Zoogeography, 85
Dubois, Idaho, 15, 652, 685, 691, Zygote, 90-91, 96, 103, 138, 162, 164,166
694, 754-755 169 '

Breeding and Improvement

Uploaded by

Copyright:

Available Formats

Breeding and Improvement

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Breeding and Improvement

Uploaded by

Copyright:

Available Formats

~MsGRAW-HILLPUBLICATIONS IN THE

BREEDING AND IMPROVEMENT

ADRIA:\"CE AND BRISON' Propagation of Horticultural Plants

VICTOR ARTHUR RICE ,.

EVERETT JAMES WARWICK

.\.PAU C~l~"" U •••• ~ l

NEW YORK TORONTO LONDON

McGRAW-HILL BOOK' COMPANY, INC.

Olpyright, Hl26, 1934, 1942,1951, by thEl McGraw-Hill Book Company, Inc.

Vl PREFACE TO l'HE FOURTH EDl'l'ION

II, Mechanisms of Reproduction; continuing delineation and refinement

VICTOR ARTHUR RICE

in lower organisms are difficult or impossible of exemplification in the

PREFACE.TO THE THIRD EDITION . . YI

SECTION I. ANIMALS AND MAN UP TO THE PRESENT

SECTION II. MECHA~ISMS OF REPRODUCTION

SECTION III. MECHA~ISMS OF HEREDITY

SECTION IV. THE ART O}" BREEDING

XXI. Selection in Dairy Cattle . 580

ApPENDIX. Livestock Record Associations . 757

NAME INDEX . . 761

SUBJECT INDEX. 767

THE NATIONAL GEOGRAPHIC MAGAZINE

Successful animal breeding depends primarily on the proper choice of

aesthetic value. The horseman wants speed and stamina; power,

Horses on farms Mules on farms Dairy cattle

Year Number Number Number I

1880 10,903 $ 53.74 1,878 $ 61. 74 11,754 $ 23.31

Cattle other than

1880 31,593 $ 15.75 44,867 $ 2.18 44,327 $ 4.40

amount of milk of a high quality over a lengthy period. The sheepman

Year ::\0. of all farm

1880 120,990,000 $ 1,576,636,520 50,156,000

* Horses, mules, cattle, sheep, swine.

in human population. It shows that the breeding of livestock is one of

TABLE 4.-shLK AND FAT PRODUCTION OF PURE BREEDS*

\ No. of Av. No. Av. pounds Av. pounds

Holstein (AR) ................ 60,375 315 16,440 570.0

* New England Homestead, Springfield, !\lass., Sept. 24, 1949.

Value of Purebred Animals.-Table 1 shows the low value of the aver-

On the average the monetary value of purebreds always exceeds that

Aberdeen-Angns-male, Prince Eric of Sunbeam, $40,000; female, Erianna B.

Clydesdale-male, Baron of Bucklyvie, $47,500.

Breed Membership 1948 All-time

Hereford .................... 1 16,000 352,888 5,825,000

Shropshire ................. . 12,160 16,523 1,117,000

TABLEl5.-NUMBER OF REGISTRATIONS IN THE VARIOUS PURE BREEDS. *-(Cont'inued)

Breed Membership I 1948 All-time

Per~heron .............. '1 6,500 251 247,893

American Rabbit and

TABLE 6.-MEAT CONSUMPTION PER CAPITA, UNITED STATES

Year Beef Pork * Lamb Veal All meats

1900 67.0 65.3 6.5 5.2 144.0

highest possible level. In such a program there will surely be found a

as far as color is concerned. An RR (red) or a rr (white) animal is pure

to probable performance, we can intelligently weed out the unfit, breed

55~ seconds (Greyhound), at the pace, 1 minute 55 seconds (Billy

An idea of the opportunity for improvement within a particular breed

tant breed associations formed in the decade 1870-1880; 13 in the next

or not to do so. Any breeder is free to attempt to change the presently