Early  Journal  Content  on  JSTOR,  Free  to  Anyone  in  the  World  
This  article  is  one  of  nearly  500,000  scholarly  works  digitized  and  made  freely  available  to  everyone  in  
the  world  by  JSTOR.    

Known  as  the  Early  Journal  Content,  this  set  of  works  include  research  articles,  news,  letters,  and  other  
writings  published  in  more  than  200  of  the  oldest  leading  academic  journals.  The  works  date  from  the  
mid-­‐seventeenth  to  the  early  twentieth  centuries.    

 We  encourage  people  to  read  and  share  the  Early  Journal  Content  openly  and  to  tell  others  that  this  
resource  exists.    People  may  post  this  content  online  or  redistribute  in  any  way  for  non-­‐commercial  
purposes.  

Read  more  about  Early  Journal  Content  at  http://about.jstor.org/participate-­‐jstor/individuals/early-­‐

journal-­‐content.    

JSTOR  is  a  digital  library  of  academic  journals,  books,  and  primary  source  objects.  JSTOR  helps  people  
discover,  use,  and  build  upon  a  wide  range  of  content  through  a  powerful  research  and  teaching  
platform,  and  preserves  this  content  for  future  generations.  JSTOR  is  part  of  ITHAKA,  a  not-­‐for-­‐profit  
organization  that  also  includes  Ithaka  S+R  and  Portico.  For  more  information  about  JSTOR,  please  
contact  support@jstor.org.  
VOLUME XXVIII NUMBER 4

BOTANICALGAZETTE

OCTOBER 1899

THE COMPOUND OOSPHERE OF ALBUGO BLITI.

CONTRIBUTION FROM THE HULL BOTANICAL LABORA-
TORY XVI.
F. L. STEVENS.

[Concluded from p. 176.]

RIPENING OF THE OOSPORE.

THE process by which the oospore attains its mature condition

is easily divided into three stages, which are marked by changes
that take place in the oospore wall, as follows: (I) a period
during which the exospore is partially constructed, and which
terminates with the beginning of (2) when the primary endospore
is laid down and the exospore nearly completed, and (3) a period
during which the secondary endospore is formed and the oospore
becomes fully ripe and ready for its winter rest.
The nuclei in the oosphere withdraw from the immediate
vicinity of the primitive wall soon after fusion of the sex nuclei,
or even before this act is completed, and the strands of cytoplasm
toward the center thicken, leaving those of the periplasm but
little more than slender supporting threads. This condition is
difficult to describe but is well represented in fig. 89. The
vacuoles have previously enlarged and the cytoplasm is more
granular than at any earlier stage. The primitive wall becomes
about o 75 p thick.
225
2 26 BOTANICAL GAZETTE [OCTOBER

At this time a peculiar substance appears in the interstices of

the ooplasm and fills the large vacuoles. It is not seen in younger
stages, but is constant at this period and takes a rusty color when
Flemming's triple stain is used. When the substance first
appears the tint is barely perceptible, but later it assumes a brick-
red color. Slightly stained sections show a film in the vacuoles,
often so curled up that an edge view is obtained, thus showing
that the vacuoles actually contain a stainable substance. This
substance is present in stages such as those represented in figs.
9 1, 9 2, 93, 9 4-
As the primitive wall reaches maturity it becomes covered
by a layer of semi-transparent substance which is penetrated by
straight pores. The condition is illustrated infig. 92, where a
portion of this layer lies in such a position that both a surface
and an edge view appear at one time. As the pores are close
together and straight the structure reminds one of the hymenial
surface of the Polyporei. Since the main substance of the layer
is nearly invisible, the pores appear dark and look like papille
or cilia growing out of the primitive wall when seen in edge-view
(fig. 92). A condition was noted by Wager in A. candidus, and
referred to as the " columnar condition," which from the descrip-
tions and drawings seems similar to this. There appear some-
what later in this semi-transparent layer peculiar saucer-shaped
masses of a dark substance of unknown nature (fig. 9I). They
are the first indication of the ridges so characteristic of the
mature epispore. There is but little periplasm at this period,
although the nuclei of the periplasm are apparently still undimin-
ished in number. However, they stain very lightly and show
only a membrane and nucleolus. They frequently mass together
in bunches, but there is no evidence that they disorganize to
furnish material for the exospore as is described for A. cauldidus.
To complete the endospore the saucer-shaped regions, previ-
ously mentioned (fig. 9I), extend laterally and run together, thus
forming the ridges. This gives the epispore the structure shown
in fig. 93. Nuclei persist much longer outside of the exospore,
but they seem to be functionless, and their history was not followed
I899] THE COMPOUND OOSPHERE OF ALBUGO BLITZ 227

further. The mature endospore consists of two layers that may

be called the primary and secondary endospores according to
their development. The primary endospore appears as an even
layer lining the primitive wall, at first agreeing precisely in stain
reaction with the gelatinous material of the vacuoles, which is
still present (fig. 93). The primary endospore, however, soon
refuses this stain and gives the reaction of cellulose with chlor-
iodide of zinc. A variation in the normal sequence was found in
one oospore in that the epispore was completed before any trace
of endospore was to be seen.
The conspicuous feature of the third period is the laying down
of the secondary endospore inside of the first. It is of cellulose,
and is about equal to the primary endospore in thickness. The
two layers of the fully developed endospore are clearly evident
when stained, also when they cleave apart in a perfectly regular
manner as they frequently do (fig. 97). There seems to be a
pause in the laying down of cellulose when the primary endospore
is completed, as is shown by the fact that perceptible changes
occur in the development of the epispore after the first and
before the second endospore is formed. Fig. 91 shows the
earliest trace of the secondary endospore that was seen. When
the second layer of cellulose is completed the gelatinous material
has completely disappeared from the vacuoles. The fact that
this substance appears just before the laying down of the primary
endospore, the agreement in their staining reactions while the
endospore is young, and the disappearance of all of the gelatinous
material simultaneously with the completion of the secondary
endospore, are significant facts. They indicate that this substance
is intimately connected with the formation of the cellulose walls
of the spore. It appears to be transferred directly from the
vacuoles to the exterior of the protoplasm, there to change to
true cellulose.
Finally, various food substances accumulate in the center of
the oospore, taking the form of one very large irregularly shaped
globule (fig. 97). Around this mass lies a zone of cytoplasm
containing the fusion-nuclei in resting condition. Their number
228 BO 744IC iz GA7Z'T1
7/.77 j () CTOBER

is about one hundred, so it is not probable that there has been

any multiplication since fertilization. Mitotic figures have never
been seen in the oospore after that act, and it is probable that
the oospore persists in this condition until the following spring.

OILS AND OTHER FOODS.

A thorough discussion of this subject would involve a much

more elaborate microchemical study than was undertaken. It
seems desirable, however, to record a few observations since
these extra-protoplasmic substances are very conspicuous at cer-
tain periods of development and their presence often introduces
serious difficulties of technique. No writer has given the sub-
ject the attention that it deserves, and detailed study would
doubtless yield valuable results.
There are three substances which, when present, always
appear as nearly spherical globules with the general appearance
of oil, and thev have undoubtedly been described as such by
many previous observers. Further study may show that they
are not true oils, and this seems quite probable, as they do not
answer to all of the microchemical tests; but in the lack of a
precise knowledge of their chemical nature they may be con-
sidered here under that name. All of them are found in the
serial sections cut from paraffin, that is, in material which has
passed in bulk through the weak alcohols and rested in 75 per
cent. alcohol for weeks or months; which has passed to the
paraffin through chloroform and been cleared of paraffin by
immersion for a few moments in xylol; and finally which has
undergone the baths accessory to the stain employed.
The first oil is very abundant, existing in far greater quantity
than either of the others, and during more stages in the devel-
opment of the fungus. It is found in the young oogonium, the
oosphere, and the periplasm. Diagrams showing its distribu-
tion in the important periods of development are given in the
plates. Fig. 16 presents it for an oogonium of about the age
shown in fig. 15. Fig. 63 illustrates its distribution at zonation,
and it should be noticed that the drops are small where the
1899] THE COMPOUND OOSPHERE OF ALBUGO BITI 2 29

meshes of the ooplasm are fine, but large in the periplasm

where the meshes are more coarse. Fig. 72 shows the condition
of an oogonium at a stage similar to fig. 70, a period some-
what later than z)nation. The distribution is much the same
because the protoplasm has changed but little. Fig. 8I, a stage
after fertilization, indicates that as the meshes of the ooplasm
become coarser the globules become larger; apparently some of
the globules fuse to form the larger drops. A later stage
similar to fig. 93 shows that the oil of the periplasm is nearly
exhausted during the building up of the epispore.
This most common oil is found in unstained sections as
blackish or brown drops when the material has been killed by
Flemming's fixing fluid. It is usually absent from material fixed
by chrom-acetic acid, or if present is in very small quantity.
If the material is allowed to remain several (six) days in the
chrom-acetic fixing agent the oil appears in considerable quan-
tity, very much as it does in Flemming's material, but of lighter
straw color instead of brown. When material fails to show the
oil, it seems to be really absent, because several days' immersion
in I per cent. osmic acid solution fails to show its presence.
Therefore, it seems that the osmic acid of the Flemming's fixing
fluid, or the long immersion in chrom-acetic acid, so acts on
the oil as to render it less easily soluble in the fluids that it
meets between the killing and the time when the sections are
stained. This oil when present takes a characteristic and beau-
tiful red color from the safranin in the Flemming's triple
stain combination. If it be very dark from the action of the
osmic acid, it is frequently necessary to employ hydrogen per-
oxid to remove the blackness, in order that the best results
may be obtained from the stains. It is partially on account
of this oil that material killed by Flemming's fluid is not so
favorable for cytological study as that fixed by the chrom-
acetic acid.
The second oil-like substance is present only in very small
qluaItit, possibly eight or ten small globules in an oogoniutr.
It is evident in both Flemnling and chrom-acetic material, is
230 BOTANICAL GAZETTE [OCTOBER

stained black by Heidenhain's haxmatoxylin, but may be distin

guished from the first oil by the fact that it does not take the saf-
ranin when applied in the usual way. It is figured in the plates
as black dots (figs. 51, 56, 59, 6z, 61, 68, 69) , in the antheridium
in fig. 85, in the periplasm in fig. 8o. No difference in stain
reaction was observed between this oil-like substance and the
central globule of the ccenocentrum, and they may be of similar
nature.
The third oil is only found in the maturing oospore, first
appearing while the secondary endospore wall is developing as
globules of a clear honey-yellow color in the meshes of the proto-
plasm (fig. 95'). It soon accumulates at the wall, and fig. 9,/ shows
large drops in close contact with the forming secondary endo-
spore, which has broken away from the primary endospore, prob-
ably owing to the impact of the knife in cutting. The drops
finally become larger and more numerous, as is shown in fig. 96.
From a study of subsequent stages it seems probable that these
oil drops later break away from the wall irregularly, and unite to
form the several large drops that are often found in older
oospores. This results finally in the condition shown in fig. 97,
where the entire central region is occupied by a curious irregular
globular structure which stains much as the oils have stained in
previous stages, but which is certainly not of fluid consistency.
It probably represents reserve food material. If this structure
represents a genuine oil drop in the living spore, we may have
to do here with the shrunken and collapsed vesicle or membrane
that encased it in life, and whose contents have been lost during
the processes of technique. Its appearance would accord well
with this view. From its size and the time of development in
the oospore it is surely the structure described as oil by previous
investigators, and is characteristic of the winter oospore.

GENERAL CONSIDERATIONS.

It seems well before closing to take a general view of the

facts that have been presented, and of their relation to previous
knowledge; also to point out more clearly than was possible
I899] 7TWE COMPOUND OOSPHERE OF ALBUGO BLITI 23 1

in earlier pages the broader significance of the phenomena

described.
A mitotic division of the nuclei of fungi seems to have been
first noticed by Sadeback ('83) in Exoascus, and has since been
observed in various forms and carefully studied and described in
a few papers.
The occurrence of a mitotic division in the oogonium of
Albugo was discovered by Wager ('96) in A. candidus, and con-
firmed by I3erlese ('98) in A. Portulace. Wager notes that the
nuclei enlarge and divide, leaving one daughter nucleus imbedded
in the central body while the other nuclei pass to the periplasm.
The one division increases the nuclei from about ii5 to about
double that number. Berlese says the nuclei divide several
times in A. Portulaca, increasing the number from 30 or 40 to
about 200. The account given in the present paper describes two
approximately simultaneous divisions affecting all of the nuclei
in both antheridium and oogonium, and these mitoses result in
the formation of the sexual elements, numerous male and female
nuclei. The second division is strikingly different from the first
in the appearance of the nuclear elements, particularly the
chromatin. This condition suggested the possibility of a reduc-
tion of chromatin, but careful study revealed no convincing
evidence. The mitoses are characterized by the intranuclear
formation of the spindle, the intranuclear centrosomes, the per-
manence of the nucleolus, and the entire absence of extra-
nuclear radiation.
Wager described the disappearance of the nucleolus in early
prophase, followed by the formation of chromosomes and then
by the spindle development. He inclines toward the view that
the spindle is derived from the lining. The membrane persists
up to metaphase, and Wager did not follow the division further.
Berlese, from observations similar to those of Wager, namely,
the disappearance of the nucleolus in prophase, argues that the
spindle fibers are derived from it. The behavior of the nucleo-
lus in A. Bliti is different from that described by the writers
mentioned, in that the structure remains apparent inside of the
232 BOT'AYA1GCAI GAZETTE [OCToBEIR

nucleus through all stages of mitosis. However, since Wager

studied this question only incidentally, and as Berlese gives no
figures and his account is very brief, a detailed coml)arison
of the species is impossible. l3erlese reports from twelve to
sixteen chromosomes, and seems to have been able to count
them during the fusion of the sexual nuclei. Neither Berlese
nor Wager give details of the mitosis later than prophase.
Wager reports that as near as he call estimate there are from
twelve to sixteen chromosomes shown in the mitotic figures. In
A. Bliti six are found with certainty in some anaphase nuclei,
and twelve appear in some metaphase nuclei with equal certainty.
It may be that when twelve are counted the chromnosomcs have
already divided, and that they really belong to two rather than
to one nucleus. However, this is not certain, and there is some
evidence that makes it appear that there is a reduction in the
number of chromosomes during the first mitosis, but this cannot
be considered as proved.
Centrosomes were not observed in either A. candidits or A.
Porhtu/ce, or in any other of the Phycomycetes, so far as the
writer is aware; but they have been described in earlier papers
for Ascomycetes (Gjurasin '93, Harper '95), and for Basidomy-
cetes (Wager '92, Juel '98).
WVithsuch fundamental differences as have been indicated, it
is useless to attempt to establish a type of mitosis for Albugo,
or to attempt to determine the relationship of the group through
cytology. The type here described for A. Bliti, while present-
ing a few deviations from the mode of mitosis in vogue among
higher l)lants, is in no way a departure from the forms well
known among the lower typ)es of plants and animals.
It is impossible to generalize on the facts of morphology pre-
sented in this paper, because siml)le processes of fertilization
have been described in all forms where homology wcould be
sought. Until the behavior of the nuclei of other species of
AlbugTo is known, it is impossible to say which is the anomalous
form of the genus, A. candiulus or A. Billt. From a l)reliminary
study of A. ca(didus it can be affirm-led that A. B/iti is in most
I8991 THE COMPOUND OOSPHERE OF ALBUGO BUTf 233

ways far more favorable for study; its nuclei are larger, its peri-
plasm more abundant, its developmental stares more strongly
marked, and its antheridial tube larger. Albugo candidus, how-
ever, has a remarkable ccenocentrum, which will be much easier
to study than that of A. Bliti, owing to its much greater size and
more pronounced stain reaction. A problem of great impor-
tance lies in a comparative study of the ccenocentra of the
genus.
The characteristic massing of the cytoplasm to form a rudi-
mentary oosphere in A. Bliti, differing thus from the vacuolate
oosphere of A. caudidus, is not a wider divergence than might be
expected in different species; nor is the variation in the cceno-
centrum more than what might be regarded as a specific differ-
ence. If such variations are found to be more marked in other
species the way may be clear to trace the relationship between
plants with one oosphere and those with several oospheres in
each oogonium; between forms which differentiate their peri-
plasm after the manner of Vaucheria, and others that follow the
habit of the Saprolegniaceic in a parietal rather than a central
massing. It must be left to future research to make clear the
relationship that must exist between the multiple fertilization
illustrated by Albugo Blili and simple acts of fusion between
sexual elements.
It may be that cytological investigation will show remarkable
variations in many respects in this genus, and establish a chain
of derivative forms. The Saprolegniacece are said to range from
parthenogenesis to complete sexual fertilization. Should Albugo
prove to be similarly rich in habits the present knowledge of
relationships will be much increased.

METHODS, MATERIALS, AND STAINING REACTIONS.

The material upon which this investigation was based con-

sisted of leaves, stems, and flower clusters of Arnar-atlizs retro-
l7exus L. and A. /iybridus L. bearing the fungus. It was collected
at Syracuse, N. Y., Columbus, Ohio, and Chicago, Ill. In all
cases the species seemed to be unquestionably Afluzigo Bliti Biv.
234 BOTANICAL GAZETTE [OCTOBER

(C. Amaranti Schw. C.amarailtacearum Zal.). The form described

by Zalewski ('83) as a different species was not met. Oospores
are very abundant on both leaves and stems, producing on the
former characteristic blister-like patches that assume a blackish
hue if the oospores ripen in sufficient quantity. In the stems
their presence may be predicted from peculiar swellings, usually
accompanied by a reddish coloration, the entire plant often
being thus affected. In partially diseased plants the oospores
are likely to be found in the inflorescence, which reacts much as
does the stem, becoming swollen and red. The parts most favor-
able for study are the stems, but leaves and flowers often sec-
tion more easily. In killing the material undesirable parts were
cut away, and the portions apparently favorable were cut into
small bits; leaves were scored, and stems and peduncles were
cut in pieces about 2 to 4cm long, deep incisions being made
every 2mm to give ready access to the killing agent. The killing
with suitable solutions was apparently perfect, and was as good
midway between incisions as where the solution immediately
reached the tissue.
The killing agent giving the best results was chrom-acetic
acid of the following formula: chromic acid o.8 per cent., acetic
acid o.5 per cent. in water. The material was usually left in this
solution from twelve to eighteen hours, then washed in five or
six changes of water, allowing about two hours between changes.
It was then successively transferred to 12, 25, 50, 75 per cent.
alcohol, remaining about two hours in each grade, and was left
in the last grade until it was practicable to imbed in paraffin. A
variation in the above method, by which the material was left
six days in chrom-acetic acid gave interesting results. It ren-
dered the oil in the protoplasm much less soluble, but caused
the loss of many of the details of the mitoses.
Flemming's chrom-osmic-acetic acid was employed in the
same manner as the chrom-acetic acid, but was not so useful,
since the sections were very much darkened by the osmic acid,
thus necessitating elaborate methods of bleaching before a
desirable stain could be obtained.
18991 THE COMPOUND OOSPNEIE OF ALBUGO BLITZ 235

Other killing agents used were corrosive sublimate in satu-

rated aqueous solution; hot corrosive-acetic-sublimate in alco-
hol; Carnoy's fluid; absolute alcohol; Hermann's fluid; and
Merkel's fluid. Most of these gave far inferior results to that
obtained by the chrom-acetic acid, and none surpassed it in
effect.
In order to imbed in paraffin the material was transferred
through 85, 95, and ioo per cent. alcohol to a mixture of absolute
alcohol and chloroform, first of one third then of two thirds
strength of chloroform. The specimens were left about two
hours in each fluid, and were finally placed in pure chloroform.
After the material had been left in chloroform for an hour a
quantity of paraffin was added, and two hours later the material
was warmed on the bath. After two hours of gentle heat it was
removed to a warmer position, and later the most of the chloro-
form was poured off, melted paraffin substituted, and the whole
kept in the bath at a temperature of about 55?. It was found
advantageous during the whole process of imbedding to use
very shallow dishes, as by this means most or all of the chloro-
form is driven off gradually by evaporation. Material was left
in the bath in soft paraffin three or four hours, the paraffin being
twice changed in the meantime to insure the removal of all
chloroform. It was then cast in a cake, the final paraffin hav-
ing a melting point of about 62'. Sections 3 to 5 ,) in thickness
were cut with a Jung sliding microtome and fastened to the slide
with Mayer's albumen fixative.
Flemming's triple stain, used with chrom-acetic acid material,
gave the best results. This stain demands the greatest atten-
tion in its use or failure is inevitable, as is well known by all
who employ it. In general, the best results were attained by a
bath of 30-60 min. in safranin, followed by a rinsing more or
less prolonged in acid alcohol. The time here is entirely a mat-
ter of judgment, varying with the result desired; 30-90 sec. was
a most usual time. After running down through the alcohols
the slide was placed in saturated solution of gentian-violet for
from 5-45 min. It was then rinsed in water and placed in orange
236 BOTANJCAL GAZETTE [OCTO3ER

G from 5-25 sec.; a longer time may do no harm but probably

I 0- I 5 sec. is always sufficient. WVipe away excess of liquid and
flood the slide twice with absolute alcohol, allowing the second
lot of alcohol to remain on the slide until sufficient of the gentian-
violet has been removed. The time required will depend on the
material, the length of time it was in the gentian-violet, and the
result desired. Drain rapidly with filter paper and flood with
clove oil for one minute; drain, follow by cedar oil and cover
in xylol balsam. If properly stained the host cell wall should
be a light violet, the chromatin of the spirem and the chromo-
somes blue, nucleolus and centrosomes red, and cytoplasm
slightly yellowish.
Hxmatoxylin stain was used, applied after the method of
Heidenhain (iron-alum 2 hrs., hxmatoxylin I2-I8 hrs., followed
by the slow extraction of iron alum till the proper degree is
reached). This treatment gave some beautiful results in contrast
with the Flemming stain, and was particularly valuable in demon-
strating the achromatic portion of the nuclear figure. Hartog's
nigrosin-carmine stain, as used by Wager and Berlese, was
tried repeatedly on corrosive sublimate material, but the results
were far inferior to those afforded by Flemming's triple or
Heidenhain's hIrematoxylin stains. However, it is possible to
demonstrate, even with this combination, the presence of many
nuclei in the oosphere and in the antheridial tube, and to rec-
ognize the principal features of the mitotic figures. Other
stains employed were Delafeld's hrmatoxylin, Biondi-Ehrlich,
and cyanin-erythrosin, but they were distinctly inferior in their
results.
The following stain reactions were presented in the best prep-
arations, and were attained by the Flemming's triple stain
unless otherwise stated: chromatic blue or violet, black with
Heidenhain's hxmatoxylin ; nucleoli red, black with Heidenhain's
hxmatoxylin; centrosomes as nucleoli; spindle fibers dark
blue; cytoplasm yellowish; granules, mentioned on page 163,
are only seen in preparations stained with Heidenhaiin's hnma-
toxylin, and then black.
1899] THE COMPOUND OOSPHERE OF AIBUGO BLIT1 237

SUM MIARY.

i. The oogonium when cut off from the parent hypha con-
tains about 300 nuclei, which enlarge and divide mitotically
while the oosphere is being differentiated.
2. The oosphere is differentiated through a massing of the
cytoplasm of the oogonium. By this process the nuclei, usually in
stages of mitosis, together with the vacuoles, are expelled from
the central region, and there results a dense and coarsely vacuo-
late periplasm. This condition occurs when the antheridial
tube is very short.
3. There is a stage called zonation in which the nuclei, usually
in metaphase, are lined up around the ooplasm, some of the
spindles lying across the definite boundary that separates ooplasm
from periplasm. In the telophase of this mitosis about fifty
daughter nuclei are found in the ooplasm.
4. The antheridium contains at first about thirty-five nuclei
which divide twice mitotically, and simultaneously with the
division in the oogonium and oosphere.
5. Previous to the entrance of the antheridial tube a papilla
is found projecting from the oogonium into the antheridium.
6. The antheridial tube penetrates slowly, reaching the
ooplasm at the time of zonation, later entering the oosphere and
appearing as a conspicuously multinucleate structure. When it
opens there are discharged about one hundred male nuclei which
fuse with the female nuclei in pairs.
7. The sexual nuclei differ in form; the sperm being elon-
gated and the egg spherical.
8. A peculiar central body, the ccenocentrum, develops as
the oosphere matures and disappears before fertilization. Its
function is unknown. There is some evidence of its being a
dynamic center of the compound oosphere.
9. The mitoses are alike in the oogoniumn and antheridium.
The spindle is intra-nuclear and there are no extra-nuclear
radiations. The centrosomes are very prominent at metaphase,
and are intra-nuclear. They could not be distinguished in the
238 BOTANICAL GAZETTE [OCTOBER

resting nucleus. The nuclear membrane persists until after

metaphase and the nucleolus is present throughout the division.
iO. The primitive wall of the oospore first appears when the
antheridial tube opens. Later the epispore is laid down upon it
by the periplasm.
ii. Two endospores are formed by the ooplasm after the
development in the vacuoles of a peculiar substance which dis-
appears as the endospores reach maturity.
I 2. After the complete encasement of the oospore it becomes
rapidly filled with food-stuffs. A large central oil-like drop is
present during the winter condition,
I 3. The fusion nuclei pass the winter in the resting condition
without further perceptible change.
THE UNIVERSITY OF CHICAGO.

BIBLIOGRAPHY.

BEHRENS, go. Einige Beobachtungen uber die Entwickelung des Oogons

und der Oosphare von Vaucheria. Ber. d. deut. bot. Gesells. 8: 314.
BERLESE, 98. Ueber die Befruchtung und Entwickelung der Oosphare bei
den Peronosporen. Jahrb. f. wiss. Bot. 31 I59.

DANGEARD, go. Recherches histologiques sur les champignons. Le Bot.

i8go: I25.

GJuRASIN, 93. Ueber die Kernteilung in den Schlauchen von Peziza vesicu-
Zosa Bull. Ber. d. deut. bot. Gesells. ii : 113.
HARTOG, 95. On the cytology of the vegetative and reproductive organs of
the Saprolegniacee. Trans. Roy. Irish Acad. 30: 645.
HUMPHREY, 92. The Saprolegniacex of the United States with notes on
other species. Trans. Am. Phil. Soc. I7: 63-I48.
HARPER, 95. Beitrag zur Kenntniss der Kernteilung und Sporenbildung im
Ascus. Ber. d. deut. bot. Gesells. I3: 67.
ISTVANFFI, 95. Ueber die Rolle der Zellkerns bei der Entwickelung der
Pilze. Ber. d. deut. bot. Gesells. 13: 456.
JUEL, 98. Die Kernteilung in den Basidien und die Phylogenie der Basidio-
myceten. Jahr. f. wiss. bot. 32: 36I.

LEGER, 95. Structure et developement de la zygospore du Sforidinzia gran-

dis. Rev. g6n. Bot. 7: 48I.
I899] THE COMPOUND OOSPHEERE OF ALBUGO BLITI 239

OLTMANNS, 95. Ueber die Entwickelung der sexualorgane bei Vaucheria.

Flora 80: 388.
SADEBECK, 83. Untersuchungen ilber die Pilzgattung Exoascus. Jahrb.
der wiss. Anstalten zu Hamburg -: IOI.
TROW, 95. The karyology of Saprolegniacee. Ann. Bot. 9: 609.
95. Trans. Roy. Irish Acad. 30: 649.
WAGER, 90. Observations on the structure of the nuclei in Peronosyhora
5arasitica, and on their behavior during the formation of the oospore.
Ann. Bot. 4: I27.
- 93. On the nuclei in Hymenomycetes. Ann. Bot. 6: I46.
96. On the structure and reproduction of Cystohus candfidus Lev.
Ann. Bot IO; 295.
ZALEWSKI, 83 Zur Kenntniss der Gattung Cystopus. Bot. Cent. I5: 2I5.
ZIMMERMANN, 96. Physiologie des pflanzlichen Zellkernes 129, I46.

EXPLANATION OF PLATES XI-XV.

All figures are from material killed in chrom-acetic acid and stained with
Flemming's triple stain, unless otherwise indicated. The figures were sketched
with an Abbe camera, using the following combinations of lenses: Zeiss 2mm
ap. I.30, with compensating ocular i8 and I2; also Bausch & Lomb 112,
with ocular no. 4. These combinations give respectively the following
magnifications when projected to the table level: 3300, 2200, and I500
diameters. P/ate XI was not reduced in reproduction. All other plates are
three fourths of the original scale.

PLA TE XI.

All drawings represent a magnification of 3300 diameters, and were

studied with Zeiss lenses.
FIG. I. Resting nucleus in mycelium, lining faint, nucleolus prominent.
FIG. 2. Nucleus in flowing cytoplasm near entrance of oogonium.
FIG. 3. Nucleus in antheridium, spirem stage, membrane faint.
FIG. 4. Prophase: small drops accumulated on the lining network, the
thread itself thinner, nucleus becoming spindle-shaped, no centrosome visible.
FIG. 5. A stage somewhat later, lining threads have almost entirely disap-
peared, chromatin grains scattered through nucleus, nucleolus to the right,
and remains of a lining strand to the left.
FIG. 6. Nucleus more spindle-shaped, globules arranged in the equato-
rial plate, at the poles round bodies about equal to the chromatin dots in
size, no spindle fibers, the definite line bounding the whole of the nuclear
m embrane.
240 BOTAANICAL GAZETTE [OCTOBER

FIG. 7. Similar tofig. 5. The longitudinal lines probably chromatin which

has not reached the equator, nucleolus at the left. This nucleus was crowded
by a mass of others, hence its short form.
FIG 8. Similar to fig. 7. Fibers visible toward the poles. This spindle
was formed in one of the strands supporting the forming oogonium, and was
consequently much elongated by tension. It is from the oogonium shown
in fig 62.
FIG. 9. Chromosomes at the equator, spindle fibers very apparent at the
poles, but not visible at the equator, nucleolus to left, membrane intact and
enclosing the spindle.
FIG. iO. From oogonium shown in fig. 6o. Cross section of a spin-
dle, twelve chromosomes apparent (stained by hematoxylin from Flemming's
material).
FIG. II. Spindle mature, chromosomes closely grouped at the equator,
centrosomes prominent, spindle brilliant and clear, nuclear membrane pres-
ent but poorly stained (fig 13).
FIG. 12. Chromosomes splitting, membrane visible with nucleolus
enclosed.
FIG. I3. Nucleolus large, nuclear membrane very definite, daughter
chromosomes ready to separate. Only those in the highest focus are shown,
several others being found at a deeper focus. The stain was particularly to
show the membrane, and was not suitable for centrosomes.
FIG. 14. Anaphase: chromosomes separating, nucleus lying about mid-
way, centrosomes still visible, whole nucleus staining dark and membrane
indistinguishable from spindle fibers.
FIG. I5. Chromosomes nearing the poles, centrosomes not distinguished
from them, nucleolus midway, slight traces of spindle fibers stretching across
the middle space, cytoplasm in the ends of the nucleus stains darker than
that of the central area.
FIG. i6. Whole spindle elongated, chromosomes massing together at the
poles.
FIG. 17. Similar to fig, 16, but in a crowded position. Compare with
fig. 7.
FIGS. I8 and I9 are from the periplasm of an oogonium of about the age
shown infig. 67, slightly younger than shown infig. 68.
FIG. i8. Spindle fibers collapsing in the middle leading to the sepa-
ration of the daughter nuclei, the fibers constituting the origin of the mem-
brane of the daughter nucleus.
FIG. i9. A young daughter nucleus, nucleolus, membrane, and chromo-
somes.
18991 THE COMPOUN-D OOSPEP'RE OF ALBUGO BUZT 24I

FIG. 20. Same as fig. 19 in resting stage, nucleolus prominent, linin

faint.
FIGS. 2I and 22 are from the same oosphere, arid in the condition shown
in figs. 68-70.
FIG. 21. Same as fig. 20 but passing into the spirem stage.
FIG. 22. Nucleus elongating preparatory to division. Compare fig. 4.
FIG. 23. Breaking of skein into chromosomes, centrosomes apparent.
FIG. 24. Spindle forming inside of nuclear membrane, nucleolus lying
outside of the spindle.
FIG. 25. From an oospore of the condition shown in fig. 70. Spindle
lying completely inside of the nuclear membrane, chromosomes grouping at
the equator, and centrosomes well defined.
FIG. 26. Metaphase: chromosomes splitting, membrane, centrosomes, and
fibers clear. Fibers apparently of about the same number as the chromosomes.
FIG. 27. Daughter chromosomes ready to leave the equator.
FIG. 28. Anaphase: chromosomes well separated, centrosomes visible,
spindle fibers crossing the middle space, chromosomes six in number. From
same oospore as figs 2.1 and 27. Compare with fig. z6.
FIG. 29. Later anaphase: chromosomes near the poles, area in which
they rest darker stained than central portion.
FIG. 30. Similar tofig. iS. Spindle fibers collapsed and daughter nuclei
read) to separate. From same oospore as fig. 29.
FIG. 3I. A male nucleus from the entrance of antheridial tube. This
nucleus is the same as the one marked X infig. 73.
FIG. 32. A nucleus (sperm) in the tip of the same tube that contained that
shown in fig-. 3I, also marked X. WVall of the antheridial tube, dense
cytoplasm surrounding nuclei, sperm with nucleolus and mass of chromatin
at anterior end.
FIG. 33. Tube open, elongated and pointed sperms escaping, showing a
verve faint lining network, one female nucleus shown. This is part of the
drawing, shown infig. 85.
FIG. 3A. A sperm approaching a female nucleus, lining more prominent
than in fig. 33.
FIG. 35. A sperm in contact with a female nucleus, becoming more nearly
round, and its lining still more prominent.
FSIGS. 36 to 40. Various stages of fusion, from oospores of the general
appearance represented infi . 90. During fusion the whole nucleus becomes
more darkly stained with gentiln-violet.

1 1G. 4 1. Cewnocentruim with globule at its center, female nucleus near by,
ordinary vacuoles of the oosphere near the marg-in.
2 42 BOTANICAL GAZET.2E [OCTOBER

PLA TE XII.

Magnification in all figures i5oo diameters.

FIG. 42. A portion of mycelium, showing nuclei with prominent nucleoli.
FIG. 43. Nuclei and cytoplasm flowing into a developing oogonium, nuclei
and vacuoles elongated and angular, nuclei too darkly stained to show
structure.
FIG. 44. The septum below the oogonium, a portion of mycelium and
posterior end of oogonium, oogonial nuclei assuming the spirem stage,
mycelial nuclei still in resting condition with prominent nucleoli but
enlarged.
FIG. 45. Oogonium and antheridium, nuclei of both in the spirem stage.
FIG. 46. Shows location of oil drops in an oogonium of the condition pre-
sented in fig. 5. They were all drawn at one focus; a slight change of
focus would have brought vastly more into view.
FIG. 47. Showing adhesion of the oogonial Hautschicht to its wall in the
neighborhood of the antheridium, nuclei overstained.
FIG. 48. Adhesion as in fig. 47. Oogonial wall partly and irregularly
corroded away on the side toward the oogonium.
FIGS. 49-55. Stages in the perforation of the wall preparatory to the
entrance of the antheridial tube.
FIG. 49. Optical section of papilla, wall partly corroded away and bulging
toward the antheridium. The dense protoplasm represented in black was
stained a deep red by the safranin.
FIG. 50. Sectional view of a condition slightly older than fig. 49. The
shaded portion of the separating wall took the stain differently from the rest,
and was apparently in the last stages of dissolution.
FIG. 5 I. A papilla of different shape in a stage similar to the last.
FIG. 52. Optical section, stain as infig. 49.
walls very thin, slightly shrunken
FIG. 53. A papilla becoming bubble-like,
as though due to imperfect killing, watery vacuoles apparent, their contents
staining homogeneously with the gentian violet.
FIG. 54. Bubble-like papilla expanding irregularly in all directions, con-
tents highly vacuolate, wall extremely thin.
FIG. 55. Wall almost perforated, but showing no marked bulging in either
direction, o, oogonial, a, antheridial side.
FIGS. 56, 57. Young antheridial tubes with a cellulose wall and surrounded
by a dense sheath of the protoplasm of the oogonium.
FIG. 58. Protoplasm beginning to collect in masses, nuclei approaching
metaphase, some shown in polar view (preparation from material fixed in
Flemming's agent and stained by hxrmatoxylin).
18991 THE COMPOUND OOSPHERE OF ALBUGO BLITI 243

FIG. 59. An intercalary oogonium, slightly older condition than fig. 8,

cytoplasm distinctly in masses, nuclei in early prophase, antheridial tube just
entering bearing no nuclei, a haustorium at extreme right.

PLA TE XLII.

Magnification in figs. 6o, 6z, 62, 63, 64, 65, 68, 69, 70, 72, I 500 diameters;
in figs. 66, 67, 71, 3300 diameters; in fig. 73, 2200 diameters.
FIG. 6o. Nuclei in metaphase, protoplasm massed in a few centers,
spindles very clear and brilliant (stain hematoxylin from material fixed in
Flemming's agent).
FIG. 6i. Nuclei in late prophase, the protoplasmic masses coalesced to
form one, vacuoles mark the juncture last made, a few nuclei not yet floated out.
FIG. 62. Nuclei approaching metaphase, spindles much elongated, all not
yet extruded from the central region, mitosis in the antheridium, antheridial
tube shown at its typical position for this stage.
FIG. 63. Diagram of disposition of oil drops in an oogonium of the con-
ditioni shown in fig. 65 (zonation).
FIG. 64. Nuclei nearly at metaphase, zonation almost complete, oil-like
drop to the right in ooplasm, antheridial tube present showing no nuclei.
FIG. 65. Zonation: nuclei near metaphase, ooplasm sharply differentiated,
many spindles lying directly across the boundary between ooplasm and peri-
plasm (stained by haematoxylin from Flemming material). See also fig. 64.
FIG. 66. Shows spindle in metaphase lying across the film between
ooplasm and periplasm.
FIG. 67. Nucleus in anaphase directly across the boundary film of the
ooplasm, also half of a late anaphase cut diagonally.
FIGS. 68, 69. Consecutive sections of an oogonium just after the division
of the nuclei, showing the position and shape of the antheridial tube. Fig.
69 shows the ccenocentrum which had just passed its maximum development
(see fig. 71).
tube and differentiating oosphere, oosphere nuclei
FIG. 70. Antheridial
in mitosis. Fig. 74 was taken from an adjacent section of the same oogonium
and shows the ccenocentrum.
FIG. 7I. Completely developed ccenocentrum, the central globule sur-
rounded by three regions of differentiated protoplasm. From a stage of zona-
tion where the nuclei were in anaphase (as infig. 67) and the daughter nuclei
were about to pass into the ooplasm.
FIG. 72. Diagram of distribution of oil at the stage shown in fig. 70.
FIG. 73. Antheridium and tube of the age shown in fig. 70, showing
multinucleate contents; 7, film; i, one of the female nuclei; A5, periplasm;
x, x, nuclei, represented also in Plate XI, figs. 31, 32.
244 BOTANICAL GAZETTE [OCTOBER

PLA TE XI V.

Magnification in figs. 74, 8o, 82, 87, 88, 1500 diameters; fig. 76, 2200
diameters; figs. 75, 77, 78, 8&, 85, 86, 53oo diameters.
FIG. 74. Ccenocentrum and differentiated oosphere with dividing nuclei.
The broken empty portion is where the protoplasm shrunk away from the
antheridial tube. Section adjacent to that shown in fig. 70.
FIG. 75. Ccenocentrum, in an oogonium, at anaphase period of zonation,
consisting of three small globules evidently fusing, surrounded by a region of
denser protoplasm, granules resembling small oil drops scattered through the
ooplasm.
FIG. 76. Similar to fig. 73 but somewhat older, antheridial tube showing
many nuclei, slightly torn near tip, antheridium becoming vacuolate; A, peri-
plasm; b, film; i, female nucleus.
FIG. 76 a. Portion of the tip of the same tube, found in an adjacent
section.
FIG. 77. View of the end of an unopened antheridial tube, wall not
apparent, sperms very numerous and crowded, each showing a dark nucleolus,
ooplasm slightly shrunken away from the tube.
FIG. 78. Section near the base of the same antheridial tube, showing
very thick wall in contrast with the extremely thin film covering the apex.
FIG. 79. Diagram to show position of preceding sections. Fig. 77 was
from a section just above the line aa,- fig. 78 from one just below the line bb.
One intermediate section was not drawn.
FIGS. 80, 8oa. Adjacent and approximately longitudinal sections of the
opening antheridial tube of the same oogonium, showing the primitive wall
and the increased vacuolation of the ooplasrn which frequently shrinks away
from the primitive wall.
FIG. 8i. Diagram of the distribution of oil in an oogonium of the age
shown in fi~g-.8o.
FIG. 82. Antheridial tube discharging sperms, antheridium shown above,
primitive wall very young, ooplasm becoming vacuolate, antheridial tube cut
slightly oblique so that its base is in an adjacent section.
FIG. 83. Diagram to show relation of sections presented in figs. 84, S,
86: fig. 84 was cut from above the line aa, and tangential to the tip of the
opening tube; fig. 85 from between the lines; fig. 86 from just below the line
bb.
FIG. 84 (see fig. 83j). Sermn nuclei leaving the antheridial tube and
approaching the female nuclei.
FIG. 85 (see fig. 83). MNlassof sperms escaping front tube. The tube may
be traced to the left as a nCass of darkly stained structureless protoplasm.
BOTANICAL GAZETTE, XXVIIH PLATEX[

2. 8

i70 i i Id. '0' : 0 I3 . I5 = e22. '2 B

:1.~~~~~~~~~~

:: ffi t 29
~~~~~~~28 0 ,
:R..0024. 2 , ~~ 27B e
a4
4;*5
At
. '1\
'26 . a

X-.s

j~
3t be I:~~~~~3
32
-
'3
jaw-d3 b

's AZ.
\
-e,,;
k'?
>,b,~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~41
0.its>ash
28Mt~~-.3 40
18991 THE COMPOUND OOSPHERIK OF ALBUGO BLITZ 245

The sperms in a mass appear dark but individually are hyaline except at
anterior end which bears the nucleolus. See also fig. 33, Plate XI.
FIG. 86 (see fig 83). The base of the antheridial tube filled with dark
staining cytoplasm and few nuclei.
FIG. 87. Just after discharge from antheridial tube, periplasm as infig.
82, several small masses of nuclei apparently both male and female sur-
rounded by denser cytoplasm (these masses seem to arise by the breaking
apart of the contents of the antheridial tube and soon disappear), primitive
wall well developed, antheridium above, ooplasm coarsely and irregularly
vacuolate.
FIG. 88. Nuclei fusing in pairs, primitive wall distinctly thickened. See
Plate XI, ftgs. 36-4o.
PLA TE XV.

All figures magnified I5oo diameters.

FIG. 89. Protoplasm collecting in dense network toward the center of
the oospore, leaving light peripheral strands where the endospore is soon to
appear.
FIG. 90. Shows the remains of the antheridial tube in the periplasm, with
no trace of its former presence in the ooplasm. The oospore is of the age
shown in fig. 88.
FIG. 91. Primitive wall mature, exospore forming, now consisting of a
porous semi-transparent mass with imbedded disks which are to form the
ridges, vacuoles in oospore filled with gelatinous substance, nuclei of proto-
plasm overstained.
FIG. 92. Somewhat younger than the last figure, showing structure of
exospore more clearly, presents edge view and also a fragment bent back
and giving a surface view, gelatinous substance in vacuoles.
FIG. 93. Exospore nearly formed, primary endospore complete, vacuoles
still containing gelatinous substance.
FIG. 94. Rudimentary secondary endospore, gelatinous substance in
vacuoles.

FIG. 95. Oil forming in oospore.

FIG. 96. Spore walls complete, including double endospore, vacuoles
with no gelatinous substance, oil accumulating in large drops on the endospore
walls.
FIG. 97. Winter conditions of exospore, the ridges higher than in previous
stages, large central oil-like mass, nuclei in sporoplasm. Section not directly
through middle so that the endospore appears thicker than it really is.
FIG. 98. Exterior of ripe oospore.