.Euclid in Greek Book I With Introduction and Notes Text

EUCLID IN GREEK
CAMBRIDGE UNIVERSITY PRESS

C. F. CLAY, Mamagek
LONDON : FETTER LANE, EC

NEW YORK ; G. P. PUTNAM ’S SONS
BOMBAY
CALCDTTA MACMILLAN AND CO., LTD.
MATO? AS
TORONTO :M. DENT AND SONS, LTD.
TOKYO : MARUZEN-E&BDSQIEI-EAISBA
ALL RIGHTS RESERVED

EUCLID IN GREEK
BOOK I
WITH INTRODUCTION
AND NOTES
BY
SIR THOMAS L. HEATH

K.C.B., K.C.V.O., F.R.S.
SC.D. CAMB.
HON. D.SC. OXFORD
CAMBRIDGE
AT THE UNIVERSITY PRESS
1920
;
PREFACE
N these days when Greek is supposed to be on
I its trial and EucKd happily defunct, it may well
seem a wildly reactionary proceeding to suggest to
teachers a combination of the two, a piling (so it
might be thought) of one inutility on another. But,

first, we must bear in mind that it is only compul-
sory Greek that is threatened : when that is gone, the
study of Greek wiU be no whit less necessary to a
complete education. Generation after generation of

men and women will still have to go to school to the
Greeb for the things in which they are our masters
and for this purpose they must continue to learn
Greek. Again, Euclid can never at any time be more
than apparently in abeyance; he is immortal. Ele-
mentary geometry will also continue to form part
of a complete education; and elementary geometry
isEuclid, however much the editors of text-books
may try to obscure the fact.
But I am not here concerned to argue the case
of Euclid against other text-books of geometry.
The aim of this book is to maintain an opinion which
I have long held that, if the study of Greek and
Euclid be combined by reading at least part of
Euclid in the original, the two elements will help

each other enormously. In the first place, boys
VI PREFACE
learning Greek in the higher Forms in schools will
generally have some knowledge of elementary geo-
metry. Even if this is read in some text-book other
than Eudid, all the technical terms and phrases will
be the same, and in any proposition of Euclid that
may be taken up the course of the proof will be
easily divined, almost by simple inspection. Hence,
in translating the Greek, the student will really be
translating something quite fanuliar. Now every
one knows that, when beginning the study of a
foreign language, it is quite a good plan to read
familiar chapters of the Bible in a translation into
the particular tongue. The advantage to the student

of Greek of reading in Greek the familiar proposi-
tions of Euclid’s first book will hardly be less.
Secondly, from the point of view of learning geo-

metry, much advantage will be gained by having,
as it were, to spell out the Greek. The beginner in
geometry needs to learn a good many things by
heart, especially technical terms, many of which,
bdmg reflections of the Greek, will be the better
understood if the Greek forms are known. How can
any person who has only had such words as theorem^
prohlem^ isosceles, paraUelepifed explained to him in
English apart from their derivation get any such
dear idea of their significance as the person who
knows them as Oewpyfia, hrocTKchcs, TropoX-
XiyXen-MTcSov? Again, persons with no particular
aptitude for mathematics find a diflBculty in memor-
ising the course of the proofs and avoiding confufion
PREFACE vii
between propositions which are nearly allied, e.g, as
converses. I cannot but think that the formality

and deliberation of Euclid’s exposition, involving a
certain amount of repetition of stereotyped phrases
or steps, combined with the necessity of translating
the Greek word for word wherever they occur,
would be a very effective way of impressing the
technique and the whole content on the mind of
the learner.
A last consideration is this. I remember well (it
was in the days long ago when most people were

under the impression that Books VII —X of Euclid
were lost in Greek) the thrill of pleasure I felt when
I first took up the Greek text of Euclid, then not
too easy to get hold of, for there was no Teubner
text as there is now, and one had to look in College
libraries for copies, which in any case were not too
convenient to handle. I cannot but think that at
all events the studious boys in the higher Forms of
schoolswho have already got a fair grasp of Greek
and know the amount of geometry corresponding to
the first book of Euclid would be really interested
to see the actual language in which the old Alex-

andrian taught the youth and pupils of maturer age
in his own day, and so to put themselves in the place
of their fellow-students of twenty-two centuries
ago. It is in the confident belief that there wiU be
many senior boys at school and students at the Uni-
versities, to say nothing of students of riper years,
who will in this way come to read Euclid with more
viii PREFACE
zest and more profit that I offer them this Kttle
book
It is desirable to add a word with regard to the
notes. I am convinced that there is no subject
which, if properly presented, is better calculated
than the fundamentals of geometry to make the

schoolboy (or the grown man) think. This is the
object of the notes; and, if it is attained, it is
worth while, whatever the schoolboy’s after career
may be.
T. L. H.
5. xL 19
CONTENTS
INTRODUCTION
EUCLID 3
WORKS OTHER THAN THE ELE-

MENTS s
EARLIER WRITERS OF ELEMENTS lo
CONTENTS OF EUCLID’S ELE-

MENTS II
THE ELEMENTS IN GREECE . IS
CHANGES IN THE TEXT . . 20
LATER HISTORY OF THE ELE-

MENTS 24
EUCLID IN EDUCATION , . 36
GREEK TEXT OF BOOK I . 41
NOTES Ill
INDEX OF GREEK TERMS . 231
INDEX OF PROPER NAMES . 237

INTRODUCTION
INTRODUCTION
EUCLID
ANYONE writing on Euclid would wish to be
jr\. in a position to give some particulars of the
Kfe and personality of the man who wrote a great
classic the reign of which, at least among the civi-
lised nations of the West, can hardly have been

paralleled by that of any book except the Bible.
Unfortunately however we have only the most
meagre information about him. It was apparently
the fate of men of science in Greece as elsewhere to
make little noise in the world as such. A still greater
man, Archimedes, only appears in history because
he figured as a defender of Syracuse during the
siege of that place by MarceUus in the Second Punic
War; his mechanical appliances (engines of war)
were a constant terror to the Romans, and it was
these which made him famous, while he himself
thought meanly of mechanics and all practical arts
and devoted himself, heart and soul, to the theoreti-
cal investigation of abstruse problems in pure
mathematics.
Euclid is claimed by Produs as a Platonist, but all
that can be assumed is that he received his mathe-

matical training at Athens from the pupils of Plato;
for most of the geometers who could have taught
him belonged to that school, and it was in Athens
that the older writers of Elements had lived and
taught.
I 2
4 INTRODUCTION
It is clear that he was in date intermediate be-
tween the first pupils of Plato (and, in particular,
Eudoxus, who lived approximately from 408 to 355
B.c.) on the one hand and Archimedes on the other
(for Archimedes quotes him textuaUy). He is
assigned to the reign of the first Ptolemy (306-283

B.C.), and he flourished probably about 300 b.c. or
a little earlier. He taught and founded a school at

Alexandria, for we are told that Apollonius of Perga,
the author of the great treatise on Conics^ spent a
long time with the pupils of Euclid at Alexandria.
There are some stories of him which one would
like to believe true. Stobaeus teUs us that some
one who had begun to read geometry with Euclid,
when he had learnt the first proposition, inquired,
‘What shah I get by learning these things?^ where-
on Euclid called his slave and said, ‘Give him
threepence, since he must needs make gain out of
what he learns.’
In another story he appears as the instructor of
Ptolemy in geometry. Ptolemy evidently, like some
moderns, found the subject long-drawn-out, for he
asked whether there was not any shorter way to
geometry than that of the Elements,, to which
Euclid replied that there was no royal road to
geometry.
The Arabs, who assimilated Greek geometry with
avidity, have more circumstantial accounts of Euclid,
to the effect, e.g., that he was the son of Naucrates
and grandson of Zenarchus, a Greek born at Tyre
EUCLID 5
and domiciled at Damascus. But the Arabian

stories seem to be partly due to misunderstandings
and partly invented in order to gratify a desire
which the Arabs always showed of connecting
famous Greeks in some way or other with the East
(the same predilection made them describe Py-
thagoras as a pupil of the wise Salomo, Hipparchus
as the exponent of Chaldaean philosophy or as the
Chaldaean, Archimedes as an Egyptian, and so on).
They even invented an explanation of the name of
Euclid, which they variously pronounced as Uclides
or Icludes, making it a compound of Ucli^ a key, and
Dis, a measure, or, as some say, geometry, so that
Uclides is equivalent to the key of geometry!
WORKS OTHER THAN THE

ELEMENTS
The Elements^ though the most famous, was only
one of many treatises by Euclid. He seems indeed
to have covered the whole field of mathematics as
then known.
In enumerating the other known works of Euclid,
we will begin with those which, like the Elements,
were concerned with elementary geometry.
The Pseudaria, or Fallacies, which seems to be
irretrievably lost, was a sort of foil to the Elements.
In the efforts of geometers to discover new theorems
or to solve new problems, many cases had arisen
where the authors, although basing themselves on

6 INTRODUCTION
the true principles of the science, had through
errors in reasoning or method deduced conclusions
which were false. Euclid’s book seems to have been
a sort of guide to the causes of error, and to have
contained illustrations of the false deductions, with
practical hints to enable the beginner to detect the
flaw in the argument, to refute it, and, by dint of

practice in distinguishing the true from the false, to
avoid such errors in his own work.
The Data (ScSo/jtâ) is fortunately preserved in
Greek, and the full text with the commentary by

Marinus has been newly edited by Menge (Heiberg
and Menge’s Euclidis opera omnia^ vol, vi, Teubner,
1896). A translation of the Data was also in-
cluded in the later editions of Simson’s Euclid
(though his text left much to be desired). The
form of the Data is that of propositions proving
that, if certain things in a figure are ‘ given ’ (in
magnitude, in position, or in species^ i.e. in shape)
something else is ^ given,’ that is to say, can be
actually determined. This collection of Data^ con-
venient for reference, enabled the solution of small
subsidiary problems occurring in a larger investi-
gation to be taken for granted instead of being
worked out on each occasion. We may, by way of
example, quote one enunciation, namely that of
the proposition which gives the equivalent of the
solution of a general quadratic equation. ^ If two
straight lines contain a parallelogram given in
magnitude in a given angle, and if the sum of the
OTHER WORKS 7
straight lines be given, then shall each of them

be given’; that is, if xy=h^^ a given area, and
xA- y— a given length, then x and y are both
given, i.e. can be determined.
Ilepl SiaLpeo-eiDv î^XCov. a tract On Divisions
(of figures), is lost in Greek but has been discovered
in the Arabic- Woepcke found it in a ms. at Paris
and published it with a translation in 1851. It is
expressly attributed to Euclid and corresponds to

the description of it by Proclus. There is no doubt
that the tract edited by Woepcke is not only Euclid’s
own work but the whole of it. It has been restored
and published with full commentary and intro-
ductory chapters by R. C. Archibald (Cambridge,
1915). The divisions of figures are divisions into
other figures either like or unlike; thus a triangle is
divided into triangles or into a triangle and a quad-

rilateral; a figure bounded by an arc of a circle and
two straight lines drawn from a point to the ex-
tremities of the arc is divided into two equal parts;
from a circle a given fraction of it is cut ofi between
two parallel chords; and so on.
To higher geometry belong the following.
Three books of Porisms. These are lost, and all we
know of them is contained in the Collection of
Pappus. There have been several attempts at
restoration, but none that can be regarded as dis-
posing of the subject. This much is clear, that the

Porisms belonged to higher geometry and contained
propositions forming part of the modern theory of
8 INTRODUCTION
transversals and of projective geometry. It con-
tained the basis of the theory of anharmonic ratios.
The Surface-Loci (tottoi irpo^ irTLâveia), This
too is lost, and nothing is known of it except from
Pappus, who mentions the book as part of the
‘
Treasury of Analysis * (tottos avaXvoficvos) and
gives two lemmas on it. It seems to have dealt with
such loci as are cones, cylinders and spheres. One
of Pappus’s lemmas states and proves completely the
focus-directrix property of the three conic sections,
which may therefore be taken to have been assumed
in Euclid’s work as known.
The Conics. This treatise too is lost, having
evidently been superseded by the great work of
Apollonius about a century later. Pappus says of it,
‘The four books of Euclid’s Conics were completed

by Apollonius, who added four more and gave us
eight books of Conics.’ It is probable that Euclid’s
work was already lost in Pappus’s time, for he goes
on to speak of ‘ Aristaeus who wrote the still extant
five books of Solid Loci coimected with the Conics.’
‘Solid Loci’ {(rT€p€ol tottol) in Greek terminology
were actually conic sections. Probably Aristaeus’s
work was on conics regarded and Euclid’s
as lod,
treatise, though general in scope like that of Apol-
lonius, was confined to those properties which were
necessary for the analysis of the Solid Loci of
Aristaeus.
The Fhaenomena. This is an astronomical work
and is still extant. A much interpolated version
OTHER WORKS 9
appeared in Gregory’s edition o£ Euclid, but it has

now been edited by Menge from an earlier and better
source {Euclidis opera omnia^ voL viii, Teubner,
1916). The book consists of 16 (or 18) propositions
of sphaeric geometry and is based partly on Au-
tolycus’s work On the moving sphere, and partly
on an earlier text-book of Sphaerica of exclusively
mathematical content.
The Optics. This book too survives and is in-
cluded in Heiberg and Menge’s edition (vol. vii,
Teubner, 1895). The Catoptrica (theory of mirrors)

included by Heiberg in the same volume is not by
Euclid, and Heiberg suspects that in its present
form it may be due to Theon of Alexandria (fourth
century a.d.), the editor of the Elements.
Euclid is also said to have written on the Elements
of Music, Two treatises are attributed to him in
our MSS. of the Musici. One is the lectio Canonis
(jcaraTo/ji^ Kavovos), the theory of the musical
intervals. Its genuineness is however doubtful.
Jan, the editor of the Musici, thought it genuine,
on two main grounds, (i) that the form and style
agree well with what we find in the Elements and
(2) that in an ancient commentary on Ptolemy’s
Harmonica Euclid is thrice mentioned as the author
of the Sectio, and almost the whole of the treatise
except the preface is quoted in extenso. On the other
hand Tannery disputed the authenticity, and the
latest editor, Menge, who finds in the tract a number
of expressions and at aU events one proof which are

lO INTRODUCTION
not in Euclid’s manner, concludes that in its present
form it is not Euclid’s but is more likely to have been
excerpted by some not too well equipped compiler
from the fuller Elements of Music mentioned by
Proclus and Marinus. The other treatise, an Intro-
duction to Harmonics (eloraywyT] apyuoviKrj), is not by
Euclid but by Cleonides, a pupil of Aristoxenus.
Both treatises are included in Heiberg and Menge’s
edition, vol. viii (Tcubner, 1916).
EARLIER WRITERS OF ELEMENTS

The first writer of Elements was, we are told,
Hippocrates of Chios (fl. in the second half of the

fifth century b.c.), a great geometer, who was
famous for two other achievements, (l) the reduc-
tion of the problem of doubling the cube to that of
finding two mean proportionals in continued pro-
portion between two given straight lines, (2) the
quadrature of certain lunes which can be squared by
means of the geometry of the straight line and circle.
From the account of these quadratures we can judge

to some extent how far the Elements had been de-
veloped up to Hippocrates’s time; he is quite familiar
vnth. the main propositions of Euchd’s Book in, and
he is himself said to have proved that the areas of
circles are to one another as the squares on their
diameters (the theorem of Eucl. xii, 2). Next Leon,
who came between Plato (429-348 b.c.) and Eudoxus
(408-355 B.C.), put together a more careful collec-
tion, the propositions in it being at once more
EARLIER WRITERS II
numerous and more serviceable. The geometrical

text-book of the Academy was written by Theudius
of Magnesia, who, with Amyclas of Heradea,
Menaechmus the pupil of Eudoxus, Dinostratus
the brother of Menaechmus, and Athenaeus of
Cyzicus, consorted together in the Academy and
carried on their investigations in common. Her-
motimus of Colophon is also said to have ‘ discovered
many of the elements,^ but no text-book after that
of Theudius is mentioned. Theudius therefore must
be taken to be the immediate precursor of Euclid.
Euclid himself no doubt made full use of Theudius
as well as of aU other available material. Euclid is
naturally silent as to his relation to his predecessor;

but we find in Aristotle (who was fond of geo-
metrical illustrations) indications of proofs of certain
which he must have taken from
propositions, proofs
some recognised text-book (probably that of Theu-
dius) and which differ from Euclid’s proofs. Some
light is thus thrown on the changes which Euclid
made in the methods of his predecessors.
CONTENTS OF EUCLID'S ELEMENTS

A SHORT indication of the contents of the Elements
wiU not be out of place. Book i is in three sections.
The first section deals mainly with triangles, their
construction and their properties in the sense of the
relation of their parts, the sides and angles, to one
another, and the comparison of different triangles
12 INTRODUCTION
in respect of their parts, and of their area where
they are congruent; it also treats of ‘adjacent’ and
‘vertically opposite’ angles made by two straight
lines; and it contains a few simple problems of con-

struction, the drawing of perpendiculars, and the
bisection of a straight line and an angle respectively.
The second section beginning with i, 27 establishes
the theory of parallels, leading to the proposition

(i, 32) that the sum of the angles of any triangle is
equal to two right angles. The third section begins

with I, 34 which introduce the parallelogram
33,
for the first time, and this section of Book r with
Book II carries the subject up to the transformation

of rectilineal areas of any shape into parallelograms
of any shape, rectangles, and ultimately squares.
Book III consists of the elementary geometry of the
circle, and Book iv solves the problems of inscribing
in circles and circumscribing about circles (i) triangles
of any shape, (2) the regular polygons which can be
inscribed or circumscribed by means of the elemen-
tary geometry of the straight line and circle, namely
the square, the pentagon, the hexagon, the decagon
and the regular polygon of fifteen sides. The content
of practically the whole of Books i, ii, iv was Py-
thagorean; and although we know little of the
Pythagorean contributions to the geometry of the
circle. Book in also may well be to a great extent
Pythagorean, as it is certainly pre-Euclidean.
Book V expounds the theory of proportion in its
most general form, applicable to magnitudes of all
CONTENTS OF THE ELEMENTS 13
kinds, both, commensurable and incommensurable.

The whole theory is due to Eudoxus.
Book VI applies the new theory of proportion ex-
pounded in Book v to plane geometry. Much of this
Book is also Pythagorean and, in particular, the pro-
positions dealing with the most general cases of the
ap^plication of areas (Props. 27-29, which are equiva-
lent to the solution of the most general forms of the
quadratic equation in algebra so far as the roots are
real), and the construction of a rectilineal figure
equal to one given rectilineal figure and similar to

another. The treatment of these questions is more
perfect than the Pythagoreans could give, in so far
as Euclid is able to use the perfected theory of pro-
portion, whereas the Pythagorean theory of pro-
portion was only applicable to commensurable mag-
nitudes and consequently their geometrical investi-
gations which involved the use of proportions were
not, so to speak, watertight.
Books vii-ix deal with arithmetic in the sense of
the elementary theory of numbers, i.e. rational
numbers; the finding of the greatest common
measure and the least common multiple, the theory
of proportion applicable to rational numbers, pro-
positions about numbers of different kinds, prime
numbers, plane and solid numbers, squares, cubes
and similar plane and solid numbers, series of num-
bers in continued proportion (i.e. in geometrical
progression), the summation of a geometrical pro-
gression, the formation of 'perfect’ numbers.
INTRODUCTION
Book X contains a highly developed theory of
irrationals, and is, in form, the most finished of all
the Books. Euclid here owed much to the original

investigations of Theodoras of Cyrene and Theae-
tetus of Athens.
Books XI, XII and xiii are on solid geometry.
Book XI corresponding roughly in the character of
its contents to Book i. Book xii to Books ii and vi,
and Book xiii to Book iv, in plane geometry. Book

XIII shows how to construct all the five regular
solids and to inscribe them respectively in given
spheres, and ends with a proof that there can be no
other regular solids. Book xiii is, in content, largely

due to Theaetetus.
The Books purporting to be Books xiv and xv of
the Elements are not by Euclid. Book xiv is by
Hypsides and is an interesting supplement to
Book XIII. The main problems com- solved are the
parisons (i) of the sides (i.e. and
edges) of the cube
the icosahedron respectively inscribed in one and
the same sphere, (2) of the surfaces and solid con-
tents respectively of the dodecahedron and icosa-
hedron inscribed in one and the same sphere. The
results are those which Apollonius had established
in a tract on the subject. Book xv is of much less
interest, though it is also supplementary to the

geometry of the five regular solids. One part of it at
least was written in the sixth century a.d. by a pupil
of Isidoras of Miletus, architect of the great church
of St Sophia at Constantinople (about 532 a.d.).
—
THE ELEMENTS IN GREECE 15
THE ELEMENTS IN GREECE

There is no doubt that Euclid’s Elements at once
superseded all the earlier text-books, and took its
place as the accepted authority on the subject.

Archimedes, for instance, refers to Euclid by chapter
and verse, as it were, and so do all later Greek
geometers. It is true that — ^to say nothing of the
attacks upon the Elements by those who, like the
Epicureans, objected to the whole of mathematics
the methods of proof etc. were not to pass un-
questioned for very long. Only a century later no
less a person than the ‘Great Geometer,’ Apol-
lonius of Perga, undertook the role of reformer. The
suggestions of Apollonius seem to have been con-
tained in a certain Genial treatise KodoXov
TrpayfmT€La)y which apparently related to elemen-
tary geometry. In particular, Apollonius gave a new
general definition of an angle, tried (vainly of course)
to prove certain of the axioms, and gave alternative
constructions for the problems of i, 10, ii and 23.
These last seem to show that Apollonius wished to
give a more practical turn to the beginnings of the
subject ; he seems in fact to have gone on the tack
of combining theoretical with practical geometry
which has found so much favour of late. So true is
it that ‘
there is nothing new under the sun or,
as Aristotle forcibly puts it, ‘

it is not once nor twice
but times without number that the same views
make their appearance in the world.’ But the
i6 INTRODUCTION
autkorily of EucEd remained unshaken. Proclus
could say truly that in all the essentials for a text-
book of Elements Euclid’s system ^ will be found
superior to the rest indeed we hear of no rival
to it in Greek geometry.
It was natural that so great a classic should become
the subject of commentaries, more or less systematic,
by later Greek mathematicians. We possess the
commentary of Proclus on Book i; but the rest are
lost except for the considerable number of extracts
preserved in Proclus or in the Arabic.

I. The commentary of Heron of Alexandria
(date stiU uncertain, the limits being 50B.C. and
the third century a.d.) was evidently rather elabor-
ate. It is cited by Proclus, but more extensively by
the Arabian commentator an-Nairizi (died about
922). The last reference to Heron in an-Nairizi is
in the note to viii, 27, which shows that Heron’s

commentary must have reached at least as far as
that point. It contained a few general notes, but

seems to have consisted, in the main, of what formed
the ordinary commentator’s stock-in-trade, dis-
tinctions of cases, alternative proofs, with a few
additions to or extensions of Euclid’s propositions
(notably in connexion with i,
47 and ui, 20 re-
spectively). It would appear that Heron was the
first to adopt the quasi-algebraical method of
proving the propositions of Book ii (from the second
proposition onwards) without a figure (except one
line), as consequences of Prop. i.
2. Pappus (end of third century a.d.) wrote a
commentary which was doubtless fairly complete.

Produs on Book i makes a few references to Pappus
by name, and probably Pappus is drawn upon in
other places where no name is mentioned. It is
certain that Pappus wrote a commentary on Book x
in two parts; fragments of this survive in an Arabic
translation in a Paris ms. discovered by Woepcke.
Lastly, Eutocius quotes a note from the commentary
of Pappus on the Elements^ the proper place of
which is probably Book xii.
3. It is probable, though not certain, that

Porphyry (about 232-304 a.d.), the Neo-Platonist,
wrote a systematic commentary on part at least
of the Elements. Proclus on Book i quotes some
alternative proofs as given by Porphyry; they are
of no particular importance.
4. Simplicius (fl, about 500 a,d.), the commen-

tator on Aristotle, wrote a commentary on the
beginning of the Elements^ i.e. the definitions,
postulates and axioms, which is preserved in an-
Nairizi’scommentary already mentioned. It is
among other things, an account of
interesting for,
an attempt by a certain ‘ Aganis ’ to prove the
Parallel-Postulate.
5. I have left till the last the commentary on

Book I by Proclus (410-485 a.d.). The importance
of this can hardly be over-estimated, since it is one
of the two main sources of information on the
history of Greek mathematics, the other being the
i8 INTRODUCTION
great Collection of Pappus. Proclus succeeded
Syrianus as head of the Neo-Platonic school. He
was an acute dialectician and pre-eminent among
his contemporaries in the range of his learning.
Although he was much more a pliilosopher than a
mathematician, he was thoroughly versed both in
mathematics and in astronomy.
The commentary seems to have been founded on
lectureswhich Proclus gave to beginners in mathe-
matics. But there are signs that it was revised and
re-edited for a wider public, since he speaks in one
place of ‘ those who shall come upon ’
his work,
and passages are included (e.g. about the cylindrical
helix, the conchoid and the cissoid) wliich could
hardly have been understood by beginners.
The importance of the commentary from a
historical point of view lies in the fact that Proclus
had access to a number now
of sources lost. The
following may be mentioned; Eudemus’s History, of
geometry (the loss of which is one of the gravest that
fate has inflicted on us), Geminus On the theory
of ike mathematical sciences^ commentaries on

Euclid by Heron, Pappus and probably Porphyry,
a work on elementary geometry by Apollonius of
Perga, Ptolemy On the Parallel-Postulate^ books on
The Angle by Eudemus and Syrianus respectively,
Apollonius’s tract On the cochlias (cylindrical helix),
a whole book of Posidonius directed against the
Epicurean Zeno, Carpus’s Astronomy.
The most famous passage in the whole of the
2
commentary is the so-calledEudemian ’ summary*

*
of the history of geometry up to the immediate

predecessors of Euclid, ending with the words,
‘Those who have compiled histories bring the
development of this science up to this point. Not
much later than these is Euclid, who put together
the Elements, collecting many of the theorems of
Eudoxus, perfecting many others bv Theaetetus,
and bringing to irrefragable demonstration the
things which had only been somewhat loosely proved
by his predecessors.’ It is certain that the summary
up to this point must have been made up from

scattered notices found in the great work of Eu-
demus; it is hardly possible that it can be textually
quoted from Eudemus, and it seems more probable
that it was taken from a compendium of Eudemus’s
history made by some later writer.
It is not certain whether Produs continued his
commentary beyond Book i. He intended to do so,
but, at the time when the commentary on Book i
was written, he had not begun to write on the other

Books and was uncertain whether he would be able
to do so; for he says at the end, ‘For my part, if
I should be able to discuss the other books in the
same manner, I should give thanks to the gods; but,
if other cares should draw me aside, I beg those who

are attracted by the subject to complete the ex-
position of the other books as well, following the
same method, and addressing themselves to the
deeper and better defined questions involved.’ As
2 —
20 INTRODUCTION
however the scholia which we possess show that
thdr authors had the commentary on Book i in

the same form as we have it, while they contain no
trace of notes which Proclus promised in certain
passages of the commentary on Book i, it is more
than probable that his commentary extended no
further.
CHANGES IN THE TEXT

It would have been surprising if the text of a classic in
such constant use and so much canvassed by a succes-
sion of commentators had not, in process of time,
suffered a certain amount of alteration. A dividing
line in the history of the text is marked by the new
edition brought out by Theon of Alexandria in the

fourth century a.d. It is well known that the title-
page of Simson’s edition of Euclid claims that in it
‘the errors by which Theon or others have long ago

vitiated these books are corrected and some of
Euclid’s demonstrations are restored and readers
of Simson’s notes will remember the familiar phrases
in which, when anything in the text does not seem
to him satisfactory, he says that the demonstration
has been spoiled, or things have been interpolated
or omitted, ‘
by Theon or some other unskilful
editor.’ Now most of the mss. of the Greek text
prove by their titles that they proceed from the
recension of the Elements by Theon; they purport
to be either ‘ from the edition of Theon ’ or ‘ from
the lectures of Theon’ (exTro <rwovo‘L(av rod ©cwvos).
CHANGES IN THE TEXT 21
Theon himself says, in a passage of his Commentary

on Ptolemy’s Syntaxis, ‘ But that sectors in equal
circles are to one another as the angles on which
they stand has been proved by me in my edition of
the Elements at the end of the sixth book.’ Hence
it is clear, not only that Theon edited the Elements,
but that the second part of vi, 33 containing the
proof of this proposition and found in nearly all the
MSS. was added by him to the original.
The passage from Theon just quoted is the key to

the whole question of Theon’s changes in the text
of the Elements; for when Peyrard found in the
Vatican ms. 190 a text which contained neither the
words from the titles of the other mss. quoted above
nor the interpolated second part of vi, 33, he was
justified in concluding, as he did, that in the Vatican
MS. (P) we have an edition more ancient than
Theon’s. It is clear too from a marginal note that
the copyist of P or rather its archetype had before
him the two recensions and systematically gave the

preference to the earlier one. We are thus more
fortunate than Simson in that we can form our
judgment of Theon’s recension by the actual docu-
mentary evidence furnished by a comparison of the
Vatican ms. P -with the mss. which we may, after
Heiberg, conveniently call the Theonine mss.

We have naturally to look elsewhere for evidence
of interpolations in the text before Theon’s time.
Besides internal evidence, Heiberg uses for this pur-
pose (l) some valuable fragments of ancient papyri,
z% INTRODUCTION
and (2) tlie fresh evidence furnished by an-Nairizi’s
commentary with its rich quotations from Heron,
which in several cases enable us to detect differences
between our text and the text as Heron had it, and
to identify some interpolations which actually found
their way into the text from Heron’s commentary
itself (notably the alternative proof of in, 10 and
the substantive proposition in, 12).
The interpolations made before Theon’s time
may be thus classified:
(1) Alternative proofs and additional cases, the
former often betraying themselves by introductoiy
words such as ^
We can prove this more readily ’ or
^more shortly thus.’
(2) Lemmas, of which there are a considerable

number, especially in Books x-xiii.
(3) Porisms (or corollaries). Sometimes parts

only of porisms are interpolated, and sometimes the
interpolation consists of a sort of proof (as if the
interpolator thought the porism not obvious
enough).
(4) Scholia, sometimes betrayed by the use of

the words KoXet and €#cdX€o-€, ‘he calls’ and ‘he
called’ (sc. Euclid).
(5) Some whole propositions, c.g. i, 41, iii, 12

(due to Heron), and at least one proposition (117)
of Book X containing the traditional proof (referred
to by Aristotle and doubtless Pythagorean) of the
fact that the diagonal of a square is incommensur-
able with the side.
CHANGES IN THE TEXT 23
(6) A number of definitions (e.g. the definition

of a ^ segment of a circle ’
in Book i and that of
‘reciprocal figures,’ vi, Def. 2) and axioms (most of
those after Common Notions 1-3).
It would appear that the text was most spoiled by-
interpolations in the third century a.d., since Sextus
Empiricus (about 180 a.d.) had a correct test, while
lamblichus (early in fourth century) had an inter-
polated one.
With regard to the changes introduced by Theon
the following notes may be added. Although Theon
took some trouble, in cases where his sources differed,
to follow older mss., his object was less to discover
and restore from mss. the actual words written by
Eudid than to remove difficulties that might be
felt by learners in studying the book. We have
(i) alterations made by Theon where he found, or
thought he found, blots in the original, (2) emenda-

tions intended to improve the form or diction of
Euclid; but he seems to have given the most atten-
tion to (3) additions designed to supplement or
explain Euclid. Under the head of additions come
the second part of vi, 33 relating to sectors^ a second
case to VI, 27, some porisms and some alternative
proofs. Under the head of explanations come {a)
repetitions of the argument where Eudid had said

‘
For the same reason,’ (jb) intermediate steps supplied
where Euclid’s argument seemed too rapid and not
easy enough to follow, and (c) additions of a word
or two for the sake of dearness and consistency. On
24 INTRODUCTION
tlie other hand he made (d) some omissions which
were designed, like the additions, to reduce the
language to a standard form. In particular, Theon
seems to be responsible for the frequent omission
of the words owcp ISct SeZât (or -Trot^crat), Q.E.D.
(or F.), at the end of propositions. A curious point
is that, when Euclid added a porism (corollary) at
the end of a proposition, it was his habit to omit the

words oTTcp eSct Sctfat (iroii}crat) at the end of the
proposition proper and to insert them after the
porism, as if he regarded the latter as being actually

a part of the proposition. In the Theonine mss., on
the other hand, the words are sometimes omitted
and sometimes inserted at the end of the proposition
instead of at the end of the porism.
There is no doubt that Theon’s edition was
approved by the persons for whom it was written;
and it was almost exclusively used by later Greeks,
with the result, as above mentioned, that the more
ancient text is preserved to us in only one complete
MS.
LATER HISTORY OF THE

ELEMENTS
We have next to consider the fate of the Elements
in countries outside Greece.
Italy was not a favourable soil. Cicero is the first
Latin author to mention Euclid; but he also tells us

that, while geometry was held in high honour among
the Greeks, the Romans limited its scope by having
LATER HISTORY 25
regard only to its utility for measurements and

calculations. The Roman agrimensores were satisfied
with the minimum of theoretical geometry; cf.
Balbus de mensurisy where some of the definitions
of Book I are given. Again, the extracts from the
Elements in a fragment attributed to Censorinus (fl.
238 A.D.) are confined to the definitions, postulates

and axioms. By degrees, however, the Elements
passed into the curriculum of a liberal education.
Martianus Capella {circa 470 a.d.) describes the
effect of the mention of the enunciation of the
problem how to construct an equilateral triangle

on a given straight line in a company of philosophers,
who forthwith broke out into encomiums on Euclid.
But the Elemeftis were probably at that time read
in Greek.
Magnus Aurelius Cassiodorius (born about 475
A.D.) says that Euclid was translated into the Latin
language by Boetius. A letter of Theodoric to
Boetius also contains the words ^
for in your trans-
lations... Nicomachus the arithmetician and Euclid

the geometer are heard in the Ausonian tongue.’
The translation can hardly be represented by the
so-called Geometry of Boetius which has come down
to us, for this by no means constitutes a translation
of Euclid. Even the version of it in two Books
which was edited by Friedlein is not genuine, but
appears to have been put together in the eleventh
century from various sources. It begins with the
definitions of Book i; then come the postulates (five)

26 INTRODUCTION
and tBe axioms (three), followed by the enunciations
only (without proofs) of the propositions of Book i,
of ten propositions of Book ii and of some pro-

positions from Books iii and iv; these are followed
by a literal translation of the proofs of Euclid i, 1-3,

which shows that the Pseudo-Boetius had before
him a Latin translation from which he extracted
these three proofs.
The Elements next passed to Arabia. The Caliph
al-Mansur (754-775), as the result of a mission to
the Byzantine Emperor, obtained from him a copy
of Euclid among other Greek books, and again the
Caliph al-Ma’mun (813-833) acquired mss. of Euclid
among others from the Byzantines. The first trans-
lation of the Elements into Arabic was made by
al-Hajjaj b. Yusuf b. Matar, who seems to have
translated them twice, first for Harun ar-Rashid
(786-809) and then for al-Ma’mun. Six Books of

the second version survive in a Leiden ms. (Cod.
Ldidensis 399, l) which is being published, with a
Latin translation, by Besthorn and Heiberg. The
object of al-Hajjaj seems to have been less to give a
faithful reflection of the original than to write a
useful and convenient mathematical text-book.
This may be the reason why a fresh translation was
undertaken by Ishaq b. Hunain (the son of the more
famous translator Hunain b. Ishaq al-Tbadi). Ishaq
died in 910, but already in his lifetime the trans-
lation was revised by Thabit b. Qurra (d. 901). The
translation by Ishaq was made direct from the
LATER HISTORY 27
Greeks and Thabit consulted Greek mss. in making

the revision. The original version by Ishaq does
not seem to have survived, but the Ishaq-Thabit
version is extant in two mss. at Oxford cont aining
Books i-xiii and one at Copenhagen containing
Books v-x. Thabit’s version, which is said to be a
model of a translation of a mathematical text, was
translated into Latin by Gherard of Cremona (1114
-1187), MSS. were discovered in 1904 by Axel
Anthon Bjornbo which seem to contain this trans-
lation. It was made from a ms. containing a revised
and critical edition of Thabit’s version; the trans-

lator also compared the Thabit version with other
Arabic versions, and apparently vvith some trans-
lation from the Greek. The third form of the
Arabian Euclid actually accessible is the edition of
NasTraddin at-Tusi (1201-1274), which exists in two
forms, a longer and a shorter. At-Tusi’s work was
not a translation of Euclid’s text, but a re-written
Euclid based on the older ALrabic translations.
In 747 the Arabs who crossed from Africa to Spain
established themselves there under a separate cali-
phate. In due course Arabian learning also found a

home in the West and, under the most enlightened
of the caliphs, schools and libraries were established
at such places as Seville, Cordova and Granada. We

hear of a palace at Cordova containing a library of
600,000 volumes, though the figure may be an
exaggeration. It was in Spain that the author of
the first extant translation of the Elements into
28 INTRODUCTION
Latin, Atiielliard of Bath, got his copy of the Arabic
version. Athelhard is said to have studied at Tours
and Laon and to have travelled in Greece, Asia
Minor and Egypt as well as Spain; he went to
Cordova disguised as a Mohammedan student. His
translation of the Elements was made about 1120 a.d.
(Athelhard was however not the first to translate
the Elements into Latin; there are traces of an
earlier translation apparently made from the Greek,
which must have been in existence before the
eleventh century. That some sort of translation, or
at least fragments of one, were available before
Athelhard’s time even in England is indicated by
the old English verses:
The clerk Euclide on this wyse hit fonde
Thys craft of gemetry yn Egypte londe.
Yn Egypte he tawghte hyt ful wyde,
In dyvers londe on every syde.
Mony eiys afterwarde y understonde
Yer that the craft com ynto thys londe.
Thys craft com into England, as y yow say,
Yn tyme of good Kyng Adelstone’s day,
which would put the introduction of Euclid into
England as far back as 924-940 a.d.)
Closely related to the translation by Athelhard
is that of Johannes Campanus (Campano), who is
mentioned by Roger Bacon (1214-1294) as a
prominent mathematician of his time and who was
chaplain to Pope Urban IV (1261-1281). It would
appear that Campanus used Athelhard’s translation
(the definitions, postulates and axioms, and the 364
LATER HISTORY 29
enunciations are word for word identical in tiie two

translations), but developed the proofs by means of
another redaction of the Arabian Euclid. Cam-
panus’s translation is the more complete, and his
arrangement is different from Athelhard’s in that
in Athelliard the proofs regularly precede the
enunciations, whereas Campanus follows the usual
order. It has been suggested that Athelhard’s trans-
lation may have been gradually altered by successive
copyists and commentators who had Arabic originals
before them until it took the form in which Cam-
panus gave it to the world.
The first-fruits of the discovery of the Greek
masterpieces in the Arabic translations were soon
seen in the works of a very brilliant mathematician,
Leonardo of Pisa, called Fibonacci (about 1175-
1 250). Leonardo assimilated all that the Arabs could

tell him about arithmetic and algebra (this is clear
from his Liber Maci)^ and similarly in his Practica
geometfiae (1220) he gives in substance all that he

could learn from the Elements of Euclid (whom he
often mentions by name), and from Archimedes’s
works about the measurement of plane figures
bounded by straight lines, solid figures bounded by

planes, the circle and the sphere, besides the
elements of trigonometry as given by Ptolemy.
What is remarkable about Leonardo is the complete
grasp which he shows over the whole of the subject-
matter. There is nothing about him of the dif&-
dence of the novice who adheres painfully to the
30 INTRODUCTION
text and form of the work drawn upon, hugging the
shore as it were. On the contrary he launches out
boldly, and alters the form of exposition at his
pleasure and with the utmost certainty, yet abating
nothing from absolute rigour of proof and often
adding clever developments of his own. It is extra-
ordinary that, after such a send-off, geometry should

have languished completely for two centuries and
more. Nothing in the least comparable to Leo-
nardo’s work appeared in the fourteenth and fif-
teenth centuries.
It was not tiU after the invention of printing that
the period of stagnation came to an end. In 1482
appeared the first printed edition of the Elements,
This fine book was printed at Venice by Erhard
Ratdolt, and contained Campanus’s translation.
The book has a margin of 2j- inches, and in this
margin are placed the figures of the propositions.
Ratdolt claims that, while up to that time no one

had succeeded in printing diagrams, he had after
much labour discovered a method by which figures
could be produced as easily as letters. How eagerly
the opportunity of spreading geometrical knowledge
was seized upon is shown by the number of editions
which followed in the next few years. Even the
year 1482 saw two forms of the book, though they
only differ in the first sheet. Others came out in
i486 and 1491 respectively.
In 1505 Bartolomeo Zamberti (Zambertus)
brought out at Venice the first translation from the
LA TER HISTORY 31
Greek text of the whole of the Elements. Zamberti

quite unfairly attacked Campanus as ‘ that most
barbarous translator ’
because he had used such
terms as ‘
helmuain ’
and ‘ helmuariphe Zamberti
does not seem to have had the penetration to see
that Campanus was translating, not from the Greek,
but from the Arabic.
The editio prince^ of the Greek text appeared in
1533 at Basel. The editor was Simon Grynaeus the
elder (d. 1541), and the preface is addressed to a
notable Englishman, Cuthbert Tonstall (1474-
1559), the author of a book on arithmetic, De arte
sufûtandi libri quatuor. Unfortunately the two

Mss. used by Grynaeus are among the worst, so that
his text could not claim to be authoritative. Yet it
remained for a long time the source and foundation

of later editions of the Greek text.
1570 saw the first English translation of the whole
of the Elements. The translator was Sir Henry
Billingsley, who had a varied career. Admitted
Lady Margaret scholar of St John’s College, Cam-
bridge, in 1551, he is said also to have studied at
Oxford, though he did not take a degree at either
University. He was afterwards apprenticed to a
London haberdasher and rapidly became a wealthy

merchant. Sheriff of London in 1584, he was
elected Lord Mayor on 31 December, 1596, on
the death, during his year of office, of Sir Thomas
Skinner- From 1589 he was one of the Queen’s four
‘customers,’ or farmers of customs, of the port of
32 INTRODUCTION
London. In 1 591 he founded three scholarships for
poor students at St John’s College, Cambridge.
Billingslejr’s translation is a magnificent book of 928
foKo pages (excluding the long preface of John Dee).
Billingsley was evidently a competent Greek scholar.
This appears from marginal notes in the copy of the
Greek text used by him (nowin Princeton College,
U.S.A.), which contain conjectural emendations of
the text as well as comments on the translations by
Athelhard and Campanus from the Arabic. The
commentary in Billingsley’s edition includes all the
most important notes that had ever been written
from the time of the Greek commentators onwards.
The elaborateness of the book may be illustrated by
the fact that the diagrams of Book xi are nearly all
duplicated, one figure being that of Euclid, and the

other an arrangement of little pieces of paper (tri-
angular, rectangular, etc.) pasted at the edges on to
the pages of the book so as to admit of being turned
up and made to show the real form of the solid
figure depicted.
The most important Latin translation is that of
Commandinus of Urbino (1509-1575), which was
the basis of most of the later translations up to the
beginning of the nineteenth century, including that
of Simson and therefore of those editions also,
numerous in England, which give Euclid ^
chiefly
after the text of Simson.’ The date of publication
was 1572. Commandinus seems to have used, in
addition to the Basel editio princess, some Greek ms..
LATER HISTORY 33
so far not identified. He not only followed the

original Greek more closely than his predecessors,
but added to his translation some ancient scholia as
well as good notes of his own.
The Latin version by Clavius (Christoph Klau),
which appeared in 1574, ^ translation but a
re-written version, with proofs compressed or added

to whenever the editor thought they could be made
clearer. The book contains a very extensive and
valuable collection of notes. This version also went
through several editions.
In 1654 appeared Andre Tacquet’s Elementa

geometriae flanae et solidae containing the eight
geometrical Books arranged for general use in
schools. This edition had a great success; many
times re-issued up to the end of the eighteenth
century, it held its own as the standard text-book in
use on the continent until superseded by Legendre.
Barrow’s Euclidis Elementorum lihri xv breviter
demonstrati (1655) compressed the whole of the
Elements into a very small compass by abbreviating
the proofs and using a large quantity of symbols
(which, Barrow says, were generally Oughtred’s).
In the preface to the edition of 1659
he would not have written it but for the fact that
Tacquet gave only eight Books of Eudid. There

were several editions of the book up to 1732 (those
of 1660 and 1732 and one or two others are in
English). It remained a standard text-book till the
beginning of the eighteenth century.
34 INTRODUCTION
1703 is the date of the great Oxford edition of
the Greek text hj David Gregory, which, until
the appearance of Heiberg and Menge’s new text,
was still the only edition of the complete works of
Euclid. In the Latin translation accompanying the
Greek text Gregory mainly followed Commandinus;
the text itself was based on the editio frincefs except
in certain special passages where, with the help of
John Hudson, Bodley’s librarian, some of the Greek
MSS.bequeathed by Savile to the University were
consulted.
In 1703 William Whiston, Lucasian Professor at
Cambridge, brought out an edition of Tacquet’s
Latin version of the eight geometrical Books. A
second edition followed in 1710 and a third (in
English and abridged) in 1714. The book went
through several more editions and was the standard
English text-book on elementary geometry until
superseded by Simson’s.
In 1708 appeared an English translation of Books
i-vi, XI and xii from Commandinus’s version by
Dr John Kdll, Savilian Professor of Geometry at
Oxford. The book was revised and corrected in
1723 by Samuel Cunn. From the number of
editions (1749 saw the eighth and 1782 the twelfth)
we may conclude that it had considerable vogue.
Simson’s first edition, in Latin and in English,
appeared in 1756 at Glasgow; it contained the first
six Books with the eleventh and twelfth. The title-

page claims that ® in this edition the errors by which
LATER HISTORY 35
Tkeon or others have long ago vitiated these Books

are corrected and some of Euclid’s Demonstrations
are restored.’ It is true that many of the faults were
not due to Theon or even to others, and that, in
preferring certain alternative proofs and making
certain alterations in the text, Simson did not
always show sound judgment. But on the other
hand there are many cases in which he found real
blots and showed how to remove them. This re-
mark applies especially to Book v (on proportion)

where Simson’s acuteness both in criticism and
emendation is seen at its best. The notes attached
at the end of the work describe the changes made
by Simson as compared with the Greek text and the
reasons for the changes, and they contain, in
addition, many valuable explanations and obser-
vations. Simson’s version fully deserved to become
a standard text-book; and some thirty editions
attest its actual success.
The translation by James Williamson of the whole

i, Oxford 1781, vol. ii, London
thirteen Books (vol.
1788) was,up to 1908, the last which reproduced

Euclid word for word, though from the imperfect
text then in use.
Peyrard’s Greek text published in three volumes
between 1814 and 1818 represents the first approach
to a better text in so far as it adopted or recorded
the readings of the Vatican ms. 190 (P). It was
accompanied by translations into Latin and French.
A most valuable edition of Books i-vi in Greek and
36 INTRODUCTION
Latin is that of J. G* Camerer (and C. F. Hauber) in
two volumes published at Berlin 1 824-25 • The
Greek text is mainly based on Peyrard, although the
Basel and Oxford editions were also used. The
collection of notes is probably the most complete
that exists, and is weH-nigh inexhaustible.
E. F. August’s Greek text (1826-29) containing
Books i-xiii further improved on Peyrard’s; but all
earlier editions of Euclid in Greek are now super-

seded by the definitive text edited by Heiberg and
Menge in the Teubner series. The Elements (edited
by Heiberg) with prolegomena, critical notes,
scholia, etc., appeared as vols. i-v in the years from
1883 to 1888.
A new translation, from Heiberg’s text, of the
whole of the thirteen Books of the Elements^ with
introduction and commentary, is now available
(T. L. Heath, Cambridge, 1908, three volumes).
EUCLID IN EDUCATION
In adding a few notes on the part played by Eudid
in education we need only begin with the century
following that in which the first Latin translations
by Athelhard and Gherard of Cremona were made.
Some idea of the general neglect of geometry in the
thirteenth century may be gathered from remarks of
Roger Bacon (1214— 1294). He tells us that in his
time the University of Paris paid little attention
to mathematics, and he has nothing good to say
EUCLID IN EDUCATION 37
of those who taught there. At Oxford he says that

there were few students who (unless driven to it by
the rod) cared to do more than three or four pro-
positions of Euclid, and it was in consequence of this
that the fifth proposition was called Êlefuga^ fuga
miserorum,’ or, as we might say, ‘escape from
troubling ’
or ‘ end of troubling ’ on the
(the play
words IA.COS and elementum can hardly be expressed
in English). In the revised regulations of the Uni-
versity of Paris in 1366 it was laid down that candi-
dates for the Licentiate must have heard lectures on
some mathematical books (unspecified) ; and similar
regulations are said to have been made a year or two
later at Oxford and Cambridge. The preface to
Demonstrations on the first six Books of the Elements
by Oronce Fine (1536) says that at that time the
University of Paris required from all who aspired to
the laurels of philosophy the most binding form of
oath that they had attended lectures on the first
six Books of Euclid; there was no examination-test.

Similarly at the University of Prag in 1367 the
regulations prescribed lectures on six Books; at
Vienna in 1389 candidates must have attended
lectures, on one Book for the bachelor’s degree, and
on five Books for the Licentiate; at Cologne in 1398

the corresponding requirement for the Licentiate
was limited to three Books; and so on. At Oxford
between 1449 and 1463 the only mathematical sub-
jects read were Ptolemy’s astronomy and the first
two Books of Euclid; and the conditions were no

3* INTRODUCTION
doubt similar at Cambridge. The Edwardian
statutes of the University of Cambridge laid it
down that freshmen should begin with mathe-

matics as being the best foundation for a liberal
education; and Euclid is mentioned as the text-book
of geometry. In 1 550 John Dee, Fellow of St John’s
College, Cambridge, was lecturing in Paris, in
English, on Euclidean geometry to large audiences
and —contrary to the prevailing practice—gratuit-
ously. Sir Henry Savile (1549-1622) who founded
the Savilian Professorships of Geometry and
Astronomy in 1619, gave lectures on Euclid, also
free, in 1620. There were thirteen of these lectures,
and they were published in 1621. They did not

extend beyond i, 8, but they are valuable because
they grapple with the difficulties connected with
the preliminary notions, the definitions, postulates,
etc., and the tacit assumptions. Briggs, the first
Savilian Professor of Geometry, took up the subject
where Savile left off. But it was during the seventy
years or so from 1660 to 1730 when Wallis and
Halley were Professors at Oxford, and Barrow and
Newton at Cambridge, that the study of Greek
geometry was at its height in England. After
Newton’s time attention would of course be mainly
directed to the new analytical methods in mathe-
matics and natural philosophy; and the Greeks
(except Euclid) fell into the background. But it
israther surprising that as late as 1799 ‘a knowledge

of the first two Books of Euclid, algebra to simple
EUCLID IN EDUCATION
and quadratic equations, and the early chapters of
was still considered
Paley’s Evidences of Christianity
sufficient to secure a position in the senior opdmes’
(W. W. Rouse Ball, History of Mathematics at
Cambridge^ p. 198). It is conjectured that Euclid
became a school text-book about the middle of the
eighteenth century when boys came to stay longer
at school and therefore required to have some pre-
liminary knowledge of mathematics before going up
to the University (Gow, History of Greek Mathe-
maticSy p. 208).
GREEK TEXT OF BOOK I
STOIXEIflN a'.
"Opoi.
a\ ^^HMEIOK icTTLV, oiv fJLipo<s ovOei'.
VpapufM] Sc fjJrjKO^ d^XaT€5.

rpa-LCjU,^? Sc 'Trepara OT^/xcta.
S^. lEuvOeta ypapup/iQ icrru/, cf tcrou rot?
iavrrjs G-rjpueLOLS KCtrat.
€', ’E7rt<j?>dv€ta Sc ccrrtF, o pjrjKOî koI ^rXdro? p.6vov
5" . ^^j^TLipaveLas Sc Tripara ypafifiat.

’EttwtcSos cTTt^dvctd iarriv, 'qrv^ c^ tcrov rats
€< ()’
5
iavTjjs evOetaL^ KiiraL.
'q\ ’EttwtcSos Sc ycovta early cy iirLTriSta Svo
ypap.p.wv aiTTopievoav aXXrjXuiV koI pJq hr evOetas
K€Lp.€y<ov Trpos aWyjXas roiy ypap^fjiwv kXlotk.
O'. ''Orav Sc at 'Trcptôvo'at r^y yayviav ypapipxu

eu^ctat eSaty, evOvypafipLO^ /coA-ctrat ^ ywyoi.
i'. Ôray Sc evOeia ctt eu^ctay ora^ctaa rds c^c^s
ytoytas taas aXXi/jXaLS 'rot^, op0^ CKarcpa rSy
lacoy ywytSy cart, Kat 17 eKpeo’rqKVLa evOâ
KaOeros KaXetrat, c^* lyy c^ccmy/ccy.
ta'. ’Ap./5Xcta yoyyCa. corty /tet^wy 6p0rj<s.
ijS'. ’O^cta Sc -q iXdo-orcJV 6p0q<:.
ty. Ôpos k(rriv^ o rtvos cart iripas.

44 STOIXEION a'.
%-)(^lJLai. eOTt TO VTTO TtVOS ^ TLV(OV opctiv Trc/otcô-
fJiCVOV.
Le'. KvfcXos icTTL ay^fJLCL eiTLTreSov virb fjitag ypapfir}^
TrepiexppLevov [^' KaXeiraL irepcKbipeLCLj, irpog
a<l> Ivos a'qp.€LOv rtav ivrbs rov oîJ/xaTOS k€l-

pL€V(jav iracraL at TTpocnriTTrovacu evOeToj. [ttôs t^v
Tov kvkXov 7r€pL<j>ip€Lav] Lcrat oXXiqXaL^ elcrCv.
lS^, Kerr/aov Se tov kvkXov to cnqpiCLOv /coXetTat.

L^. Aia/x,€Tpo 9 8c TOV kvkXov ioTLV evOeid tls 8ta
rov Korrpov rfypiuiq kcu êpaTOVfieyifj i<f> iKorepa
rd p.€p7j VTTO T^s rov kvkXov r-cpt^cpcwz,?, lyrt?
Kol Bixo- ripveL rov kvkXov.

Vff. ‘H/AtICV/cXtOV Sc COTt rO ir€pi€)(Opi€VOV O^pua VTTO
rc rJJs StafLcrpov Kat r^s aTroXapâvopivTqs vir
OLvrrjs TTcpt^cpc/as. Kevrpov Sc rov ‘qpuLKVKXLOv
ro avTOj o koI rov kvkXov iorriv.
L^. ^y^para evOvypappA iari rd inrb evSeiSiv Trepte-

yopeva, rpCirXevpa pev rd vird rpiiov, rerpd--
wXevpa Sc Ta vô r€(r€rdp(jt)v, iroXvTrXevpa Sc ra

xnrd 'TrXeLovfvv y r€a-crdp<ov evOcubv ‘Treptcyopeva.
K. T(vv Sc rpLirXevpoiv (ryTqpdrwv hroTrXevpov pdv

TpLyoivov koTi TO Ta9 Tpcis to*as typv TrXevpdî
i<ro(rKcX€<s Sc to tois Svo povas tcras ^yov r-Acvpas,
crKoXifjvdv Sc TO Tas rpeis dvLcrov^ irXcvpas.
K0,'. "Êri Sc ra>v rpLTrXevpoav oryypdr<ov opOoywviov
phf rpCywvov iorrt rd eyov 6p6rp^ ytovLav, dp^Xv-
ydviov Sc TO ôv dp^Xetav yivviav, S^vydvLOv
Sc TO TCL? rpcts 6^£ias ôv yoivias.
K^. TSv Sc rerpoTrXevpfov ayyjpdrtvv rerpdynavov pev
ioriVy o laroTrXevpdv ri €<ttl koI opOoydjvLOv,

STOIXEION a. 45
iT€p6jJLr}K€s 8c, 0 6p6oymiov ovk lororXevpov

Se, poju.j8os Se, 0 laoTTkeupov ovk opOoymvtov
Se, po/ijSoaSes 8c to rag aTTCvavriov irXevpdj; re
Kal ymlws icras aXk'qXms ^
irA-eupov eo’TLV ovrc opôydaviov* tol 8c Trapa
raura Tcrpar-Acrpa rpoTreîa KoXeLcrdoi.
Ky. HapdXkrjXoL elcnv ci^cwUj airtvcs ^ tw avrw

CTTiTTcSo) oS(rat Kttt iKôXXopLevcu eU aireLpov
c<^’ c/carcpa ra puipyj hrl pciyScrcpa crvp.Tuirrovo'LV
aXXrjXai^,
AlT7]fLara.
a\ aTTO wavrog (rrjp,eLov im iray (nipJiov
evOeiav ypap.pir]v dyayctv,
jS". Kal 7re7r€paG-p.h/7]V ciêiav /card to crvv€)^s hr
evdctas iKjSaXcLv,
y\ Kal -Travrl K^pw fcal 8ta(myp.aT6 kvkXov ypd-
êcrdoL,
8^ Kal Tacras rdg opflds ymias to-as dXXTyXais
ehcu.
€. Kal cdv ctg Sdo evdelxis evOcia ipLiruirrovora rdg

5 \
cyrog Kat
\S\\»\/
ra avra
cttl p^^pyj
Zl''
ywvtas Ovo opoiov
iXdo'crovas TTOiy, iKjSaXXofihas rdg Svo eutfcidg
inr wTreipov avpTTLTrreLVy €<f>

d piipyj eimv al t(ov
8t;o 6p6!Xv cXdo’O'Oveg.

46 STOIXEION a.
K-OLval evvoiaL.
a. To. rw avT(a Lcra kcu oXXi^Xol^ iarlv tcra.
Kat iav tcrots tcra TTpooredy, ra oka iarrlv tcra.
y. Kat iav avo icruav i(ra dâtpe^'j, rd KaraXct-

7r6fJL€vd icTTLV lcra.
Kat rd i^âppLOôvra hr aXXrjXa tea dXXrjXoL^

ioTLV.
rf. Kat TO oA.oj' Tou piepov^ pLelôv [etrrti^].
/
a.
*Eirt T^9 SoO€t(jrr}<s cvOelas TrcTrcpacrptenys Tpt-

yiDvov IcrdirXevpov ovo'TiQO-acrOaL.
ESTO Aet
^ 8o6&cra euêtd nrcirepacrpievrf ^ AB.
eTTL ttJ? AB evOela^ Tpty(avov lcr6~
5 irXevpov (rvaTrj<ra(r6aL.
STOIXEION a'. 47
Kot/wo fJLCv rS A SLaom^fJuiTL Se rw AB kvkXos

y€.ypd<l> 6 (a 6 BP A, koI ttoXiv KarTp(o pikv r<o B 8ta-
<rrrjp.aTi Se t(S BA kvkXo^ yeypa.<^ 9(d 6 ABE, kolL
diro Tov r cr7}p,€LOv, KaO* o ripivovcnv ttAAiÂ-ovs ot
kvkXol, iirl rd A, B a-qp.&xL hr€êu')(6(acrav evOelaL lo

ai BA, BB.
Kat errel to A (Tr]fL€LOv KCVTpov ioTt rov BAB
kvkXov, icnq iarlv ^ AB ry AB* TraXiv, hrei to B
crrjpLeLOv Ko/rpov iarl rov BAE kvkXov, loti icrrlv 17
BB ry BA. iSeixO?/ Sk kcu ^ BA ry AB lorp 15

€Karepa dpa rtav BA, BB ry AB iariv tar), rd Sc
TW avTw L(ra koI aXXiqXoi^ iartv to’a* koI ^ BA dpa.
ry BB eoTLV tarj* at rpcL^ dpa at BA, AB, BB wrat
dXXyjXaLS ctcrtV.
iaoTrXevpov dpa icrrl rd ABB rpLytavov* kcu ow- 20

iarrarai c-ttI t^s doOcCaryjs evOelas Tr&repacrp.anfjs
AB.
pETTt 7t}s SoOeiarrjs dpa eu^ctas TrcTrcpao’jU.enys
rpLytvvov laOTrXevpov arwecrrarcu^ OTrcp eScc Trotijo’ai.
/S'.
npo? Tw So^^t OTyp-CMp ry SoOeLcrrj evOeia tcnfjv
cu^ctav OecrOaL.
STO to jih/ SoOey orjpieiov rd A, y Sc So^ctora

E evOeia ^ BB* Set S7 irpos tw A arjp.€L<D tq
SoOeiary evOeCa ry BB tonyv cv^ctay OecrOcu, 5
4S STOIXBiaN a.
*E7r€^€u;(0CD yap oltto tov A OT^/xetoi; iirl to B

(rrjp.€tov evOeia ^ AB, /cat (TwearTaTO) ctt’ aurijs
rptyoDVOv lo’OTrXevpov to AAB, Kat CKjScjSAiôr^coo-av
€7r’ cu^cta? Ttttg AA, AB evdeloL at AE, BZ, #cat
lo K^rpo) pi^ t(3 B StatmypLaTt 0€ t(3 BP kvkXo^s
yeypd<fiO(o b PH®, Kot Trdktv K^Tpw r<S A Kat

SiaonypaTt tw AH kvkXos y€ypd(j>0(j) b HKA.
’B-aret ow TO B cn^ptctbv KO^Tpov Icttl rov PH®
<kvkXov>, tCTT/ €OTtv BP rg BH. ttoXiv, hrel to A
15 (n7/t€iov K&rpov eort tov HKA kvkXov, icrv} icrrlv y

AA ry AH, Sv AA ry AB iortv. XoLTry dpa
rj icriy
y AA Xoiiry ry BH eortv ibeLxOy be koI y BP ten/].
ry BH toy eKarepa dpa twv AA, BP ry BH icmv

icry. Ttt Se tw avr^ icra Kat aXXyXoLS icrrlv t<ra*
20 Kat AA apa t^ BP lortv toy.

Hpog dpa r<S boOevri cnypctw t<S A T^ SoOetojj
evOaa ry BP ten; eu^€id KctTat y AA* OTrep eSci
TTOt^crot.
ST0IXEI12N a. 49
Avo hoOeicrS^v evOeimv avLcriDV wiro Tjj^ fieiôvos
rrj iXd(raovL l(T7]v evOeiav dêXetv.
E2TOSAN
w r,
at So^cttrat Svo evOecaL dvicroL at
/x.€t^ft)v €OT<i> ^ AB* Set 817 d-TTo

AB,
ttJ?
/itctfovos ttJs AB ttJ cXacrarovi ry P lo-yv euOeiav 5

d^cXcti/.
Kcicr^w TTpbs Tw A crrjfJi^LiD ry P evOeia ten) y

AA* Kal K&Tpca p.€v t <5 A BLaanrjpAiTL 8c tw AA
KVKXog ycypd^^ft) o AEZ.
Kat c-zret TO A <rrjpL€Lov Kcinrpov coTt rov AEZ 10
kvkXov, urq cortv y AE ry AA* dXXd Kat 17 P rjj
AA cartv Icrr]. iKaripa dpa twv AE, P ry AA

COT6V ten/* wtTTC Kat AE ry P iamv lcttj.
Avo dpa SoêttrSv cv^ctaiv dviaraiv ratv AB, P

uTTo Tr}9 peiôvos rrj<s AB ry iXd(r<rovL ry P tony 15
dcfyypyrat y AE* o-ircp ISct irotôrai.
H. E. 4
50 3TOIXEION a.
S'.
’Eav Svo rpcycava ms Svo TrXevpas [rats] Sval

TrXeupais teas e)(j^ eKarepav iKaripa koI t^v
yoavuLV ry yoivta taryv rr/v vtto twv l(to)v
€u0ei(t)v “TTcptcôfL^v, Kal ryv ^denv ry jSderei
5 Lcryv cêt, koI to rptyinvov rw rptytovo) Icrov
loratj /cat at XotTrat ymvlaL rats Xotirats

ycovtats teat eeovrat l/carepa iKaripa, ixfi
as at teat TrXojpat wrorîvovcnv.
St;o rpCywivaAEZ ras Svo wXcvpd^s

ra ABP,
Ttts AB, Ar AE, AZ
rats Svet TcXeupals rats
teas eôvra IxaTepav iKaripa rrjv pJiv AB ry AE

Tî/ Se Ar ry AZ /cat ywvtav virb BAP ytoyla
ry {mb EAZ icrTjv. Xiyoi ort /cat /Octets ^ BP âect

TQ EZ toTj €crT«/, Kttt TO ABP rpiymvov rtS AEZ
15 rptytovitt teov lerat, Kat at XotTrat ym/Cai rats Xot-
TTats ywvtats teat ceovrat iKaripa iKaripa^ vefi as
at teat ^Xevpal viroreLVOvo-LV, y pkv vtto ABP tQ
vTrb AEZ, y Sc vnrb APB ry virb AZE.
’Eâpptofo/jtôv yap tov ABP rpiydyav iirl to
20 AEZ Tptywvov Kat Ttêptôv too /t^ A aypucCov hrl
2
STOIXEION a\ 51
TO A crr)iJLeLOv rrj^ Sf AB evOetag cTn t^v AE, kâp-

JJM(r€L Kol TO B Ol^fL^LOV IttI TO E Sttt TO LOlfJV €LVCU
T17V AB rg AE* l<^fLppxiiTa.<rq>i 8^ r^s AB hrl t^v

AB icfiappLOcret kcu ^ AP €v$€La hrl r^v AZ Bia to
IfTTpf etvat T^v wro BAP yoavCcLV Ty xnro EAZ* coore 25
Kol TO P crripL€LOV to Z crqpLUQV i<l}apfWcr€L Bia
TO r^v AP ry AZ. dXXa fiyv Kal
toyjv ‘ttoXlv €H/ai
TO B E i(f>7)pp.6K€L‘ mCTTC jStto-t? y BP €irl

IttI TO
Pd(nv Tr}v EZ k<f>appi6<r€L, el yap tow /i^ B «ri to
E e<âppj6aavTO<i rov Sc P evl to Z y BP j8a<rts cirt 30
ryv EZ ovK i(l>appL6aeLy Bvo evOeiai yinpCov Trepi-
€ôv<rtv oirep iarrlv aSvvaTOV. eijiappLOcret apa y BP
jSacrts €7rt Tyv EZ kol toy avry ecrrcu' iSare Kal
oXov TO ABP rplymov hrl oXov to AEZ rptymrov

e<l>appL6aei Kal urov avrS ecrroL, Kal at XoLTral ycavLat 35
CT-t TCLS Xovira9 ytovias eKjyapfiocrovcn Kal icrai avrais
eorovTcUy y pL€v wo ABP ry wo AEZ y 8c wo

APB Ty wo AZE.
*Eav apa 8vo Tpiytova ras 8vo irXevpas [rat?] 8vo

TrXevpais to’as exS ^KaTcpav c/carcpa icat Tyv ytovtav 40
ry ycDvta tcryv e^y ryv wo T<av tcrwv ev&eiiov êpue-
Xppihyv, Kal ryv jSao-tv ry jSao'ct tayv c^ct, icot to

rpLyasvov t<S rpiytavta laov eoTOL, #cat at XotTrat
yoivtat Tats Xotvais ycovtats urat co-ovrat eKarepa '
cicarcpa, v<j> as at taat TrXevpal wroreLVovcriv orrep 45

cSct Setât.
4 —
5^ 2 T 0 IXBI 0 N a.
/
€ .
Twv to-ocTKcXaiv rptycovcov at irpos ry pdcrei

yusviajL tcrcu dAAiJXats elariv, koI TrpocreKjSXrj-
OeicrSiv rSiV lctcdv evêtiov at wro rrp/ pdciv

ycavtat tcrat dXA^Xats Icrovrat.
5 T7 STO TptytDvov to-ocTKcXes to ABF Lcrrjv ej(Ov rqv

JL-/ AB TrXevpdv tq AT irXajpa, /cat -TTpoacK^Sc/SX*^-
crOnxrav hr evOîas rat^ AB, AB €v 0^m at BA,
PE* Xiym oTt wo ABB yiAvla. ry vtto ABB

tcnj €<rrtv, y Se wro BBA ry vtto BBE.
10 €iXri<\>Oia yap cm Trjs BA TV)(pv crypclov to Z, Kai
d<l>rjpyor6io otto petiôvo? t^s AE ry iXacraovL ry

AZ Icrq AH, Kttt
7) hrcêuxOiocrav at ZB, HB evOeiat,
hrel ovv lanj iarlv yp^ AZ ry AH y Sc AB

ry AB, Svo By at ZA, AB Svoi Tats HA, AB taat
15 euTiv l/carcpa eKarepa* kcu ywviav KOivyv TrepiexovcrL
Tyv wo ZAH* jSao-ts dpa y ZB ^daei ry HB toy
STOIXEION a.
/
53
eoTtv, Kal TO AZr Tpiymov rw AHB rpiyoUva) Icrov
eoTOL, Kol at Xonrol ymiai rats XotTraw ycovtW teat

ecrovTai l/carcpa emrepa, a? at teat irkevpal
v7roT€tvovariv, rj piv vto AFZ wro ABH, ^ 0€ 20

wro AZr ^TTO AHB. feat IttcI oXiy rj AZ 0X17
AH lertv tony, (Sv 17 AB rp AF eenv teiy, Xotir^

apa rj BZ XoiTry rrj TH ianv lotj. iSet^dr} Se Kal
77
Zr TiJ HB L<nf}* Svo 8^ at BZ, ZF ovet rats FH,
HB teat €tety l/carepa iKaripa* Kal ymia rj viro 25
BZF ymta ry viro FHB teiy, kol jSaets avrSiV
KOLvrj fj BF* Kat to BZF apa rpiymov r<3 FHB

TptyaSvo) teov lerat, /cat at XotTrat ymm rat? Xot-
Tcak ymioJL<y teat leovrat haripa harlp^ as
at teat irXevpcu wrorctvouetv* ton; apa certv ^ phf 30
wo ZBF rg wo HFB g 8c wo BFZ rg wo FBH.

cTTCt oilv oXg g vtto ABH ymia oXrj rg wo AFZ
ycovta k^d')( 6 teg, w g wo FBH rg wo BFZ teg,

yj
Xotirg apa g wo ABF XotTrg rg wo AFB certv
teg* /cat ctet xpos rg jSdereL tot) ABF Tptytoi/ou,

35
eSet^flg Sc Kttt g wo ZBF Tg wo HFB teg^ /cat
ctetv wo Tgv jSaetv.

T5 v apa teoe/ccXwv Tptycovo)!' at Tpo5 Tg jSctect
ywvtat teat dXXgXats ctetr, /cat Trpoee/cjSXg^cieSv
t£v tew cufictw at wd Tgi^ jSdeti' ywtdt teat 40
dXXgXats ceovTat’ OTrep eSet Setôt.

54 STOIXEION a.
*Eav rpiytovov al Svo ycovwu tcrat aAAijXcus
<w<rtv, Kol al VTTO rots to-agycovtas wroTCLVOTJorat
TrXevpal tarai aXkrfXai^ Icrovrai.
STO ABF 1017^ €;(ov t^v vtto ABP

rpiymvov to
^
E ywviav ry APB ymiq.* Xkyia on koI TrXevpa
vtto
y AB vXevpa tq AP icrriv unj,

d yap avLcro^ iariv y AB rfj AP, y iripa avTtbv
ju.€t^(ov ioTLV. €<rr<o p.^(fyiv y AB, koI a<l>yprj(r6(ii
OTTO p.€LÔVOS TTJS AB iXoTTOVl Ty AP LCrTf

y
10 AB, Kol €7r€^€VX^<«> V
’EttcI ow Xcry icrriv y AB ry AP KOLvy Se rj BP,
Svo 8y al AB, BP 8vo rats AP, PB tcrat ctcrtv

cfcarcpa cxarcpa, /cat yavia y vtto ABP ywvta ry
VTTO APB ioTiv toy pdcis dpa y AP jSaorct ry AB
15 toy iarcv, /cat to ABP Tptywvov rtS APB rpiydvm
icrov Icrrat, to eXaorcrov rw jocôvt- oTrcp droTrov
ovK dpa dvicros iariv y AB ry AP* tay dpa.

"Eav dpa rpiydivov al Svo yasvCai tcrat aXXyXcus
cSortv, /cat at wro tols tcras ytovtas viroreLVOvarai
20 TrXevpal tcrat dXXyXais ecrovrar oirep tSci Seiâi.

STOIXEION a. 55
r.
’EttI t7]s avTrj^ evOetas 8vo rais aurats euêwus

aXXm Suo cu^ciat tacu. iKaripa c/carepa ov
(TVdTaO’^crovTai Trpos aXX<o koI aXX(a <rrjpL€L<a
iTTL ra avra pîpr] ra avra irepaTa €)(0V(rat
TOLS cf ^PX^S evOeiats- 5
l yap Svvarov, iirl rrjs avr^^ evOeuL^ AB

E Svo rats avTOLS evOeCais rats AF, FB aXXaL Svo
cuêtat at AA, AB to-ai eKorepa l/carepa orwccrra-
Twcrav Trpos aXX(o Kai aXk(o (rrjpieL(d rw tc F Kat A
€7rt ra aâ p.ip7j to. avra irepara exovaac, euare lo
tonyv €tvat ttJv pev FA ry AA to avro Trepan expvcrav
avry TO A, Tr;r Se FB ry AB to aô Trepan eôvo'ar
auT^ TO B, Koi hreêuxOoi y FA.
TEttcI ovv tcnff icrrlv rj AF ry AA, lay iarl koX
ytovla rj vtto AFA ry vtto AAF* pid^mf apa rj vtto 15

AAF ttJs vtto AFB* ôAAw apa y vtto FAB fiel^wv
ioTL Tys VTTO AFB. ttoXiv €7r€i lay icrrlv y FB ry
AB, lay iarl Kal yiovla y vtto FAB yovia ry mro
AFB. iSeIxOy 8 c avr^S kcu TroXXio pLelîov oirep
iorlv dSvvarov. 20
OvK apa hrl rys ainys evOeias Bvo raXs avratg
56 STOIXEION a.
evOeiaus aXXat Sijo evOeiaL «rat CKarepa iKaripa
OTJora^^crovTat Trpos oAXo) koI cny/Aetw J-ttI to,
awa fiipy} m avra iripara eôvcrai rats ^PXV^

25 evOeiaiS’ oTrep eSet Sctlcu.
r
'
7 -
“Eav 8uo rpLyi3yvcL tcls 8vo TrXeupas [rats] 8i;o
TrXevpaTs LacL<s Ifcarepav Ifcarepa, Se
/cat T^y ^dcrtv tq jSacret tcnyv, /cat r^v 'ywvtW

yoyvLa urrjv c^€t wo rtov tcrcov evOeiwv
5 7repte;(Op,^v.
T0 ABE, AEZ ras 8uo TrXcupas

E2
8vo TplymoL tcl
Ttts AB, Ar rats 8vo TrXevpats rats AE,AZ

tcras e^ôvra eKarepav iKaript^ r^v p^ AB AE
ry
T^v 8c AE ry AZ* cx€Tft) 8c Kttt jSacrtv ttjv BE
10 jSaorct EZ tcnyy Xô) OTt /cat ytovta vtto BAE

17
ycovta ry wo EAZ ccrrtv tcny.
’E<âp/x,oôp.ôi/ yap toij ABE rpiydvov hrl to
AEZ rptycovov xat Ti^cpôv tou p.ev B oTypctov cttI

TO E OT]pi€LOV TTJS Sc BE €V 0 C«I$ CTTt TiyV EZ €«âp-
15 pOO*Ct Kat TO E Cr7]p,€LOV CTTl TO Z Sta TO tOT^V €LVai

STOIXEIfiN a. 57
T^v BP Ty EZ' fôpjiUKraffijs 8^ t^s BP in ryv
EZ i<l>apfjmowi koI al BA, PA m ras EA, AZ.
(I yap ^diTK ph' y BF in jSaffiv t^v EZ i<l>ap-
pa(Ta, ai 8e BA, AP vkajpal in ras EA, AZ m

iijioppjoffmm aXXa TapaXkôvmv ws ot EH, HZ, 20
<7v<rra^y<rmai in ryi avrrjs ai^ftas 8w tws awals

€vS((ais aX\m Suo evSiiai hax imipa kanpa irpos
aKhp m SXka (njji£Ui in ra owa jtipnj ra am

tipara eôvm. ov (rn'toToyrat Se* m apa iâp-
/io(o/i^s T^s BP ^dcrm in t^v EZ jSacrtv ok 25
iijxtpimcmn m tu BA, AP zXoipdi in ras EA,
AZ. i^jwcrmiv m ymla y wro BAP

dpa' wore
kl ymlav t^v wro EAZ i^pdau m iotj airy
COTOl.
"Eav dpa 8vo rptywra tos 8uo irXevpd^ [toIs] 8vo 30
TrXeupaii icras (](y kaupav karepa (cat rijv ^wof
ry jjaca iotjv koi ywvtait ry ywvla mpi
£^€i ryv VTO mv iw euâwv ircpieôp^v mep
(8ei 8(iâi.
58 2 T 0 IXSI 0 N a.
T'^v SoOeiorav ymvLav €vûypa/A,aov St^®- Tc/xctv.
STO ^ So^clora yQ)^^ €v$vypafj.fio^s ^ vtto BAP.

E Set S'^ aVT^V Bl)(a T€
IJL € LV .
'EtlX'^<l>$(i) €7ri T^s AB TV)(ov aryjfJieLoy to A, Kat
5 a<j>ijfyyj(r 6 <ii arro t^s AP AA icrq 17 AE, Acat e-ire-
êu^ô) 17 AE, Ktti crwcoraTO) cttc AE Tpeymvov

to-OTrXeupov TO AEZ, koI cTreêvx^w ^ AZ- Aeyo)
OTt 7} VTTO BAP ywVLCL St^tt rirpLipTaL virb AZ

evêtas.
10 *E7ret yap lott} iariv rj AA tq AE, KOtv^ Se 17
AZ, Svo 0^ at AA, AZ Suert rat? EA, AZ tcrat ettrtv
€/carepa CKaripa, kol âtrt? AZ pdcrei ry EZ Icrrj

ioTLV' ymta dpa ^ wrb AAZ ywvta vtto EAZ
toTy eariv.
15 “H dpa SoOeLora ycovtix €v 6vypap.pLOs 17 vtto BAP

Stâ rirfiTfraL vtto rrjs AZ cvêtW OTrep ISct Trotijerat.
.
STOIXEinN a. 59
t
I,
T^v SoOelcrav evOeiav TreirepcLa'fieinrjv SL)(a TepeLV.
STO ^ Soêtcra evOeia Tre7r€pa(rfJL€VT} rj AB* Set

E Srj r^v AB eWelav ‘iren’epao'p.a/rjv Stâ T€p.€LV.
SiivccrraTW hr avr^s rpCymfOv iaOTrkevpov to
ABr, ml T€T/i7j(r0w y vwo APB ymta Stâ ttJ TA 5
evOcCa' Xôt) on rj AB evOela 8c)(a rerp/irjraL Kara
TO A (nrjpuov.
’EttcI yap loT] eo-rtv ^ Ar ry TB, KOLV^ Sc 17
PA, Svo S^ at AP, PA Svo rats BP, PA tcrat ctertv

eKarepa imrepa’ koll yfavia vtto APA ytovia ry 10
wo BPA lo-Tj egrrtv* jSaorts apa y AA jSao'ci tq BA
iViy ccTTtV.
'H apa BoOeiaa evOeia Treircpao-pj^ y AB 8t;(a
rhprjTOL Kara to A- oirep eSet Tronjcrai.
la
Ty Bodetoy evdeia diro too Trpbs avrp Soôtos

(rqpLUOv TTpbs opOas ycavCas cvOeiav ypappLrjv
dyayetv.
STO jutev So^ctcra €v 0 €ia y AB to Sc So^w

E 17
(TypêLOV hr avrrjs to P* Set 8y dirb tov P 5

6o 2T0IXEI0N a.
anjfiuov ry AB evQeiq. Trp6<s opOas yaivias evOetav

ypapip.^ dyayciv-
EiXi^Bia iirl Ti}? Ar tv)(6v on/ju-etov to A, /cat
KCio’^ci) FA tcny^ PE, /cat cruvccrraTO) cttI t^s AE

lo rptycovov tcroT-Xeupoi/ to ZAE, /cat ^ ZF*
X^co oTt ry SoScLo-rf evOeta ry AB dxo tov irpos
avry 8oâ/T09 <rt}pL€Lov rov F 7rpo5 opOas ywtas

€vd€ta ypap.pLrj yKrai y ZF.
*ETr€t ydp ttny ccrrlv rj AF ry FE, Koivy 8 c y
15 FZ, Svo By at AF, FZ Bvcrl rats EF, FZ tcroi claw
c/caTcpa l/cttTcpa* Kat jSdcrts -17 AZ jSdo-et ry ZE tOT]
coTtv- ycovtd dpa y vwo AFZ ywvCa ry vtto EFZ ten;

icTTLv Kat elcriv orav Be ev&eta ctt* cu^ctdi'
ora^ctcra Tas c^c^? ywvtas taas aXkyXat'S ‘irot^,
2^0 op^^ c/cttTCpa Twv y&vmv cortv

tcrctfv opîj dpa eortr
c/cttTcpa tSv wo AFZ, ZFE.
dpa BodetGT} evOeta ry AB aTro tov Trpos aoTiJ*
BoO&fros (ryptetov rov T Trpos op^ds yoivtas evOeia
ypapupiy yKrca y FZ- orn-ep eBei Trot-^crat.

STOIXEION a. 6i
c/y.
’Em T^v SoOuorav eiJ^ctav aTreipov airo tov

SoOofTos cnfjfieLOv^ o ftij iariv ejr avr^s,
KaOerov eWeioy ypafjup.^ ayaycLV.
TO AB
E5
T7 Soêura evOeta amtpos 17 to §€
So 6 h^ oijjuueLOV, 0 fjirj iarcv iir avT^s^ to. P* Set 5
«rt T^v Soêwrav evOciav aTrctpoy T^y AB d-n-o
TOO BoOevTOs cnj/xetov too T, o firj eorty hr aisr^s,

KaOerov evOeiav ypafi^pJ^v dyaycty.
lEjiXi^<l)0<o yap em ra erepa p-^prj rrj^ AB evOeiag
rvxov OT^ftetov to A, /cat Kevrfxp pJev t<S T Siaxm^puLTL 10

Se T(p PA kvkXo^ yeypdc^^© d EZH, /cat Terp,yj(rO<o
^ EH edêtd St)(a KttTa to 0 , /cat eirefei^âjcray at
PH, P0, PE evOciar A.ey« ort em T^y Soêtcrav

ed^ctdy diretpoy t^v AB dxd too SoOhros (nqpelov
TOV P, o /ly io-TCV hr adnjs, KaêTOS y/crai rj P@. 15

*E7r€t yap urrj eorty ^ H0 Ty ®E, KOty^ Se rj
@P, 800 8^ at H0 0P 800

, Tats E@, @P taat etaty
e/carepa eKaTcpa* Kat )

3dcrts PH jSdaet t^ PE eorty
toiy* ytoydi dpa ^ wd r0H ywta drd E@P etrrty
62 STOIXEinN a.
20 Icrq. Koi cwriv cc^c^S. orav 8e evOîa hr euêiav

(rradclcroL ras €^€^§5 yiavim tcras aXh^Xais TOiy,
opdrj eKaripa rlXv lartav yoiVLwv eortv, koI k<^€-
crrqKVLa evOeia Kaderos /caXetrat e<^* i(j>€OT7jK€y.
’EttI ôOeicrav apa evOctav aTreipov r^v AB

25 airo rov So^hros anqpeiov rov T, o p-rj koTiv hr
avrij^, /caêros yKrat ^ F©* oxep eSet Trotijcrai.
’Eav €v 6 &jol hr evBeiav araOcLcra ytovtas Troiy,
ÔL 8vo op^tts Bvalv opâts icra? TroLt^crei,
(oy
Y@EIA yap T15 *7 AB hr euOeiav rr/v FA ara-

E Betau ycDvtas TroieLTio rag viro FBA, ABA- Acyca
5 oTi at VTTO FBA, ABA ywyiat *5T0t Svo opOai etcrtv ^

Suortv opOoLLS tcrat.
El ow icrrj iarlv wo FBA wo ABA,

860 opOoL etcTLV. el B crr^peLov
Se ov, y)(B(o airo rov
TQFA [cv^cta] TTpos 6p6a^ BE* at apa wo FBE,

10 EBA 8vo SpOat euTLV' koI hrel wo FBE Swt raig rj
wo FBA, ABE larj icrriv, kolv^ irpoaKeLcrOat ^ wo

EBA* at apa wo FBE, EBA rpto-t rats wo FBA,
.
2 TOIXEION
r
a 63
ABE, EBA taac euriv, TraXty, iirel ^ vtto ABA dvcrl

rats VTTOABE, EBA lctt] kariv^ kolv^ 'irpo(rK€L(r$(o 17
wro ABP* ai apa viro ABA, ABF TpLcrl rats 15
ABE, EBA, ABF to-at eio-LV. eB€L)(drj<rav Se xal at
VTTO FBE, EBA rptcrt rats awaLS to’at* ra Sc
avT<o tea Kal oAXiyXots ierlv tea' Kal ai vtto FBE,
EBA apa rats vtto ABA, ABF teat eleiv aXka al

VTTO FBE, EBA Svo opdat cieLV koI al viro ABA, 20
ABF apa Svciv QpOaX^ teat ilelv.
’Eay apa evOcta iir evOeiav araêtcra ycDVtas 'srot^,
^TOL Svo opOa^ ^ SvertT/ opdai^ icras irovqeeL' oirep

eSet Setât.
iS'.
’Eav 7rpo5 TLvi cv^cta kcu tS irpbs CLvry erjpeiio
Svo cv^ctat pi] iirl ra avra p€pyj Ketpevat,
TCLS €<j[>e^s yiovtas Svetv opOaX^ teas iroLwetv,

hr cv^ctas caovTat aXkrjXoLS al cv^ctai.
POS yap AB
n eYjp€Lip t<3
TLVi €u^€6a
B
avra pepi] Kctpcvai ras
ry
Svo cv^ctat al
/cat t<S
BF,
mpos aiyrg
BA p^ hr\ ra
yoivias ras viro ABF,
5
ABA Svo opOais teas irovelraêay' X^co on hr

(vOeias ierl ry FB y BA.
.
64 2TOIXEION a.
10 Et yap prq iari tq BP hr cvêia'S y IB A, earo)
ry PB eTT cvOeta^ y BE.

*E^€t ovv euOeLa -q AB hr evOelav ttjv PBE
iîamjKey^ at apa inro ABP, ABE ywvtat Svo 6p0ai9
tcrai elartv eia’l Se /cat at wro ABP, ABA Svo opOai^
15 tcroti* at apa vtto PBA, ABE rats vtto PBA, ABA
tcrat €tortv. KOLvq a(jyQp'qo‘0(si ^ wo PBA* XoLTrrj
apa wo ABB XoiTry rf wo ABA icrriv tarr],
cA.ao'O'wv /tetôvt* wcp lortv aSwarov. ovk apa

w* €V0€tas eoTtv *§ BE TQ PB. opotojs S&ôpev
20 oTt ovSe oAAiy rts -n-X^v ttJ? BA* w* eoêta? apa
loTty ^ PB rg BA.
*Eav apa xpos rtvt ev^cta /cat t<3 ^rpos aw^ cn^peuv
Svo euêiat pir] hrl ra ahra piprq /cctpcrat ras
ymla^ Bvaiv opâts ttras 'Trottoo'tv, w euêtas ccrovrat
25 aXXi}Xat9 at evOetac^ OTrep eSei Sctât.
/
Ae
’Eav 5vo evOcuu ripLVitxriv oAXi^Xas, Tas /cam
KOpvifyrjv yoivtas t<ras dXXi^Xats ttolovctlv.
YO -yap dudeiat at AB, PA TcpLvh-fDcrav dXX-jJXag

* Kara to E (nypetov* Xô) drt tony cortv 17 p.€V
STOIXEIOK a. 65
VTTO AEr yiovta tq ^6 AEB, y Se vtto EEB ry 5

VTTO AEA.
TETTct yap evOcla y AE ctt evOeiav T A c^coTTy/cc
ya)vta5 iroiova-a m? wo TEA, AEA, at apa vtto
TEA, AEA ydivtai hvow opBcws taai euriv. TraKa/,
hrei euêta y AE ctt* cu^ctar ryv AB iîaryKc 10
ycDVtas -TTOtovcra ras wo AEA, AEB, at apa vtto
AEA, AEB ytovtat Svarlv opOcu? tcraji elcrCv.
Byaav 8c wo TEA, AEA Svcrlv opdai^ torot*

/cat at
at apa wo AEA raZs wo AEA, AEB taat

PEA,
elcTLv. KOLvy a<j>ypyo-9(o y wo AEA* koL-rry apa y 15
wo PEA Xouiry ry wo BEA ttnj iariy. o/xoto)? By
SeiÔyo'^raL ort /cat at wo PEB, AEA tcrai cttrtV.
TEav apa Bvo evOeloL r€pi.v<a<nv aXXyXas, ras Kara
KOpfwfjiyv ytovtas ttras aXXyXouis Troiovanv omrep cSci
8ct|at.
[Ilop^cr/xa.
*E/c 8-iy Tovrov Kjiavcpov ort, cay Svo cu^ctat rc-

p.V(ji(rLV aXXyXas^ ras “/rpos ry Top/y ywvCas Terpacrtv
opBcus to*as irotijcroi/o-ty.]
IIayT05 rpuytavov /xtas TSy ‘TrXevpwy vpocr&c-
pXyOeLcrys y €/ctos yaxvla cKarcpa? ro/y eyros

Kat aTrevavTLOv ywvifav /xcô/y Itnly,
ESTO
Tpiyavov to ABP, Kat irpoareKpe^XyaOta
avTOv pLia TrXevpa y BP CTrt to A* Xcyo) ort y ^
H. E. 5
66 STOIXEION a.
€KTO? yovta ^ VTTO AFA /let^cov eorti' cKaripas twi/
^Tos iccti aTrevavriov rwi' wo FBA, BAP ywvtoii'.
T^Tfj/i^G-Ou) Ar Si3(a Kara ro E, /cat cTriêox^cwra

rj
•37BE iKê^XricrOii} hr €v9aas hri to Z, koI K€LcrO(j>

10 ttJ be y EZ, /cat hreêvx^^ V
taifj
^ Ar IttI H.
*E r€t ow LOT) ioTLV ^ /t^ AE ry EP,
7 Se BE
Ty EZ, Svo 8-^ at AE, EB Svo-t rats TE, EZ tcrat

ciortv CKarepa cKarepa* Kat ywta wo AEB ycovta 17
15 ZEr ter?; ioTLV Kara Kopvt^^v yap- jSaert?
apcL 17 AB pd(T€L Ty Zr LOT] i(rTLv, Kal TO ABE

Tplyiovov Tw ZEr Tpiy<av<p io’TLV urov, /cat at XouTral
ywj/tat rats XotTrats ywvuus tcrat cto-tv l/carepa l/ca-
ripa, v<j> a? at «rat TrXevpal worctvoi/crtv teny apa

20 eoTtv 17 BAE Ty vtto ETZ. Se €OTtv 17
wo EFA T^s wo ErZ- peiôiv apa ^ wo APA

T75 wo BAE. 'Opotcos 8“^ t5 ? bp rerpypiirrjs Stâ
heixBrjcrerca kclL wo BPH, TOvreorTtv ^ vtto APA,
-7
pct^cov /cat T79 wo ABP.
25 EEavros apa rpiymov pms twv irXevpSiV irpocrc/c-

:STOIXBION a. 67
^)<.r}0€i(Trj<: ^ e/cro? ymvta CKaTcpas twv koX
airevavTiov yinVLUiv icrriv oirep eSet heiâi.
c^.
IlavTOS Tpiyiovov al Svo yojvuu Svo opOtov
iXacraouh cto’t Trdvrr} peraXapIBavopevai.
ESTH ABF
rptywvov to ABF* Xiyto otl tov
Tpiydvov at Svo ycovtat Mo opOfov iXarrovi^
eifTi Trdvrr] peraXapjBavop^cu. 5
’EKj8 €j8A.?}<râ> yap ^ BF cm to A.
Kat circt rpiydvov tov ABF cktost eorrt ywvta ^
VTTO AT A, pdôyv cart T7S c^to? /cat d-n-cvayrtW ttJs
ABF. /coti^ 7rpo(TKdcr6(a y vrro AFB* at dpa

VTrb AFA, AFB tQ>v vtto ABF, BFA ptetôvcs cIo'lv, 10
dXX* at wrd AFA, AFB Suo dpâts tcrot cwrtv at
dpa VTTO ABF, BFA Svo opOiov iXdcrcroves eurtv.
bpjom^ Sy Miôpev ore koll al inrb BAF, AFB Mo
6p 6 <bv €kdo-(rov€*s ehri kcll CTt at viro FAB, ABF.
navTOS dpa rpiydvov al Mo ywvtat 8vo opOatv 15
ekdo-crovh ctert TrdvTrj pLeraXappavopevar drrep cSci
Set^t.
5—2
63 STOIXBIflN a.
L7}\
IlavTos rpiymov rj [idl<i)V irkcvpa Trjv fidôva
ymiav VTOreiveu
2TO yap Tpiymov to ABF paCpva. eôv rrjv
E AT vXeupav AB* Xeyio on Kal ymia rj
5 VTTO ABF loTt TqS VTTO BFA.

’EttcI yap p€t{(ov Icrriv y AF Trjg AB, /ccto’flco ttJ
AB LOTj 7j AA, Koi iireêvx^co rj BA.

Kai hrd rpiydivov tov BFA Ikto^ iom ymva rj
VTTO AAB, pdîov icrrl rrjs oro? koI aTrevavriov tt}s
lo VTO AFB* L(nj Se 17 vtto AAB ry vro ABA, iurd
Kcu TrXeupa y AB ry AA ionv lay pcLîov apa kcli
y VTTO ABA rij? vwb AFB- ttoXAw apa y vtto ABF

pdfyiv coTfc T^s mro AFB.
narrow apa rpiymov y pud^m TXcvpa ryv pdôva
15 ymiav vrordvcf oirep 28 a

2T0IXEIQN a. 69
iff.
navTO? rpiymov vtto ttjv fiîôva ymiav ^

/ict^CDV TrXevpa wrorcivci.
STO rpcymov to ABF jxctôva iôv r^v vtto

E ABP yconav ttjs vto BFA* Xeyct) on Kal TrXojpa
Tj Ar TrAeupas AB peiQm €crr«'. 5
El yap prj, Jfoi Icrq iorrlv ij kV rp AB ^ lAdcr-

orcDi/- Lcrrj piv ow ovk lorty rj kT AB* lOT] yap
av Tjv KCLL ytuvta
^ ^0 ABF tq vtto AFB* ovk eom
Se* OVK apa iorlv Tf AF rp AB. ovSe pijv
Icnj
lkd<r(r(jiv AF rrjs AB* ekdo-crm yap av rjv

i(Tnv fj 10
/cat ymvia vtto ABF

rj viro AFB- ovk eon Se*
OVK apa ^kdaorm iorrlv y AF ryg AB. Se
on ovSk Lcnj iarcv. pet^m dpa iarlv AF t^s AB. fj
IlayTOS dpa rpiydvov virb rifv peiôva ywvLov 7j
pd^m TrXevpa VTorecvcr direp eSct Setât.

70 STOIXEION a.
/
fC .
IlavTos TpLy(Dvov at Bvo ‘TrXevpal

ftetôvis €to-t Travry p.eraXappav6pL€vai,
B (C)
STH yap rptyoivov to ABF* Xiyo) otl tov ABF

E rptyoivov at Svo ‘jrXcvpal p€tôvh
5 ctcrt Trdvrrj BA, AF
fLeTaXapâvop-evat, at ph^
BF, at 8c BF t7]<s AF, at 8c BF, FA -nj? AB.

AB,
Afq')(Oid yap ^ BA cttI to A crqpeLOV, /cat /ccia-^w
TQ FA ten; ^ AA, /cat hreêvxOio ^ AF.

’Ettci ow lot; coTtv ^ AA rrj AF, ton; cerrt /cat
lo ywvta 7j VITO AAF tq vtto AFA* /xet^wv dpa y vtto
BFA T^s VTTO AAF* /cat hr^l rptytovov cort to AFB

petôva ôv t^v ô BFA yoiviav t^s -utto BAF,
ô Sc T^v p€Lôva y(dVLav ^ p€L^(ov TrXcvpa inroreLvei,
AB dpa rrjs BF lort pet^tav, lorj 8c AA t^
'q
15 AF* p€Lôv€^ dpa at BA, AF Trj^ BF. bpotiog Sy

Secôpey otl kol at par AB, BF t^s FA paôv&i
datv, at 8c BF, FA t^s AB.
IlavTOS dpa Tpiydvov al Svo TrXevpal tt; 5 XotTnJs
peiôves elai Trdvrr} peraXappavopeyoL* orrep cSci
20 Setfat.
STOIXEiaN a. 7?
jca.
*Eav Tpiyiavov IttI fjLLa.>: rtav TrXevpwv airo rOiv
Treparaiv Svo evOelai evros orvorraO(oo'iy, at
ovoTaOeia'aL twv XoiTriov rov rpLyaavov Svo

TrXevpmv iXdTTOV€9 ph/ ecrovraif pîôva Sc
y(jtiVLav TrepUôvcriv.
PirONOY yap ToS ABF /uas twv irXevpfav

T TjJs Br OTTO ro>v -TTcpaTov
iirl
T(DV B, P Bvo cu^ctat

€VT05 crvv€aTdr(0(rav at BA, AF* Xeyo) ort at BA,
AF Twv XotTrwv Tov rpiywyov 8vo TrXevpIav rmv BA,
AF iXdcrcrov€9 ctcrtv, petôva Sc ymiav xcpt- lo
€)(Ov<n T^v vTTo BAF r^s wo BAF.

AtT^xô) yap ^ BA c^rt to E. /cat CTrct Travros
rptycrtvou at Suo irXevpal rrj^ Xonr^g petôvh elcriv,
Tov ABE dpa Tpiydivov at Svo TrXeupal al AB, AE

T^s BE pdôvis cto-tv koivtj TTpocr/ccto-^Q) ^ EF* at 15
dpa BA, AF T0V BE, EF eurtv. ‘irdXiv,
CTTCt Tou FEA rptywyov at Svo irXevpal at FE, EA
ttJs fa ptctôycs eixnv, Koan] Trpo(jK€L(rd<o ^ AB* at
FE, EB dpa Ttav FA, AB fictiôvc? ctati/. oAXa

Twv BE, EF p£Lôv€<s iSeixOrfcrav al BA, AF* TroXXio 20
dpa al BA, AF rtov BA, AF pîôvh ehriv.
IlaXiv, c-TTCt TravTos rpiydvov ^ c/ctos ywvta rgs
72 STOIXEION a.
eFT09 /cal CLTref/avriov iLeiîov coTtV, tov FAE apa

rpiyfavov ^ ^ wro BAF
c/ctos ycovta cart t^s
25 wo FEA. S6a ravra TOtwv koll tov ABE TpLywvov
^ c/CTOS
ywta rj vtto FEB wo BAF.
pLet^mv iorl tjJ?
aXA,a 1-5? wo FEB pLei^mv ^ wo BAF*

'3roAA<3 apa ^ wo BAF pîô^v ecrrl t^s wo BAF.
*Eav apa TpLyuivov hrl juLiJag riav irXeupuiV obro tSv

30 -jTcpctTW Soo eiOeLCLL €wos ovoTaOioarLV, al crvaraOcL-
crai T(av Xonrtav tov Tpiymvov Suo TrXevpiov cXarrovcs
elaiv, pL€Lôva Se yoiviav Treptôvatv* OTrcp eSct
Sctât.
fC^\
"E/c Tpmv cvOeioiv, at €Lcnv tacu Tptxri tolls
BoOclo-ols [ev^cAts], Tpiyivvov avarr^croLcrOaL'
Set 8^ ras 8vo rJ}? Xotir^s ptelôvas uvojl
Travryj pteraXapiPavopLiyas [8ta to koI iravTOS

5 Tpuymvov rag Svo 'irXevpas ttjs Xolit^s ptei-
ôvas €ivai Travry pueTaXapLâvoptiyas],
STOSAN at Soêtcrot rpets euêtat at A, B, F,

E CUV at Svo T^s XoLTnjs /tetiôves eoreuo'av irdvry
STOIXEIQN a. 73
jw,€TaAxL/jtj8avo/x€i/at5 at ph A, B -nj? T, at 8c A, T
Bj icat ?Tt at B, r A- Set he rm Imv lo
rats A, B, P Tplymov (jv(rn\(Ta(TQai.
’EKfC€t(r6o) Tt 5 ciêta
^ AE TreT€paL(rphif] ph /cara
TO A aTcipos 8c xaroL to E, /cat /cetVfio) rg ph A
ten/ AZ, Tj) 8c B t(n/ ZH, ry olV Icrrj jj H®-
/cat /COT/Ko ph T(! Z, Staon/joaTi 8c tw ZA kv/cXo? 15
yeypaj>d(i) 0 AKA* TroAtv /ccrrpo) ph t(S H, hwLtrrfj-

pan 8c t (3 H® ku/cXos yeypa^&m 0 BLA®, /cat cre-
levxOmav ai KZ, KH* Xeyo ort l/c TpiSr ci^ctiv
tSv tew Tats A, B, P rptywov (TweerraTat to KZH.

’EttcI yap to Z enj/tetov Kcvrpov cen toS AKA 20
kvkXov, tarj Imv ^ ZArg ZK* rj ZA rg k

imv ton/, /cat
^ KZ apa rg A cortv ttn/. TraXtv,
circt TO H <rrjp€L0v Kevrpov cort to? AK® kvkXov,

L<rg imv yj H® rg HK* dXXa H® tj P imv
ten/* Kat ij KM. apa rji i cemv mj. cort oc Kat 25
ZH rg B wn/- at rpets apa eu^ctat at KZ, ZH, HK

rpLoi Tats A, B, P teat ctetv.
TE/c Tptw apa cv^ctw rSv KZ, ZH, HK, at ctetv
teat Tptet rats 8 o 0 cteats cr^ctais rats A, B, P, rpt-
ywov oruveeraTat to KZH* oTrep eSet Toigeat. 30

74 :STOIXEION a.
/cy.
npos Trj ^vOeia koI tw Trpos avTrj (rqp^Ly
ry SoOeLory yiavla evOvypdfiiJuo Icrqv ycovtW

^dirypappjov OT.JO'T^cracrât.
STli y phf SoOetaa evdeta y AB, ro Sk wpb<s

E avry crrjpeiov ro A, y Sk Bo6ei(ra ymvta evOv--
ypapipos y vtto APE* Set Sy wpos ry SoOeioy euOeia

TQ AB Kol t(3 TTpbs avry cn^/xctw t(o A rp Bod^Lony
y<i)VLa €v6vypdp,p(a vy vtto APE Icryv ymiav evOv-
ypappxiv cruoTycraarOcu,.
10 ^LkyÔ<ii i<j> eKaripay rwv PA, PE TV)(6vTa ay-
jji€La TO, A, E, /cat kjrel^€d)(0(si
y AE* /cat l/c rptcuv
€v6€MVy at ctcrtv tcrat rpial rats PA, AE, PE, rpt-
AZH, ware layv etvat Tyv pjkv
^(ovov avveardro} to
PA r^ AZ, Tyv Sc PE ry AH, /cat Irt ryv AE ry ZH.
15 ’E'/ret ovv 8vo at AT, PE 8vo rats ZA, AH tcrat
ctcrtv l/carepa iKaripa, Kal pdats y AE jSacret ry

ZH ten;, ytovLa dpa y wo APE y<ovia ry wo ZAH
ioTiv toy.
npo5 dpa ry 8oO€Lay evOeta ry AB /cat r<S irpos
20 avry aypetto tw A ry 8o6€tay ytovta evOvypdppto

rywo APE lory ytovia evOvypappog crwiaraTat y
wo ZAH* OTrep cSei Trotyaai,
5
STOIXEION 75
K?!,
"Eav 8vo Tptytova ras 8vo TrXevpoL^; [rats] Svo

TrXevpaLS to-as ^tôripav eKaripa, rrjv Be
yoiVLav T^s ymfLas pueitfiva r^v viro rutv
i<j(Dv evOeLwv ‘irepLexppLGnfjv, koI ttjv pdcriv
T^s ^dcreuis pLclôva lêt. 5
STO Svo Tpcycova rd ABF, AEZ ras Svo 'srXevpds

E Ttts AB, Ar roTg Svo irXcupats rat? AE, AZ
ttras ôvra iKarepav eKaripa^ phf AB ry AE
rrjv §€ AP jy AZ, •§ Se Trpos rw A yuyvia rrj^ 7rpb<s
nS A ymlas p.€L^<av Icrro)* Xô) OTt /cat Poxtl^ y lo

Br pdaem ttjs EZ p-etifcov kcrriv.
TETret yap p.ctâ>vBAF yayyia tj}? wo EAZ

^ vtto
ycovtas, crweoraTO) Trpos AE evOeia, kcu t<3 wpos

ry
a-ury aypeiiv r<a A ry wo BAF ywvta Xoy y wo
E AH, /cat /cetcrô) oworipa twv AF, AZ lot? ^ AH, 1
/cat
’Eiret ovv toy iarlv y ph^ AB ry AE, y Sc AF

ry AH, Suo By at BA, AF Bvcrl Tal<s EA, AH wrot
eicTLV eKarepa c/carcpa* /cat ywvCa y vtto BAF ytavCa
ry VTTO EAH toTy* pda-L^ dpa y BF pdorei ry BH 20
ccrTtv Lcry. irdXtv, cxei lay iarlv y AZ ry AH, toy

76 2 T 0 IXEI 0 N a.
icrrl Kcl vtto AHZ yo)vta ry vtto AZH* ^€t^(ov
apa y wo AZH wro EHZ- iroXX^ apa piCL^mv
icrrlv y EZH Ty^ vnro EHZ. Km €7r€t rptywov

viro
25 icrri TO EZH /Actôva ryv mro EZH ymviav

TYj^ VTro EHZ, wo 8c ryv pLCiôva yiovLOV y p.eL^<m
irkevpa 'uttotcavci, jULCt^cav apa Kat irkeupa y EH rys

EZ. Loy Se y EH ry BP* fjL€L^(Ov apa Kat y BP
rys EZ.
30 "Eav apa Svo rptyan'a ra? 8vo 'irXeupas Svcrl
T-Xcupats «ra5 e)(y iKaripav cKarcp^ Sc yoivlav
Ty£ yiavias petôva t)(ji ryv vtto rwv taav evêuov

'irepi€)(ppAvyv^ Kal ryv pdcriv ttJ? jSaorcws peiôva
c^ci* OTTcp cSct Sctât.
/
«€ .
*Eay Svo rpLyma ras Svo irXcvpas Swl TrXcvpats

tcra? cp(2; hcaripav iKaripa, ryv Sc pdarivrys
pdaems fictôva c^îy, Kat yotviav rrj'S
yaivCas pelfyva c^ct wo twf eifOeiiDV
S TrepLexopevyv.
E
to-as
B
2T0
A
TttS
A
Svo rpLywva rd
AB, AP
ôvra knaripav CKaTcp^
(C)
ABP, AEZ ras Bvo ^Xcwpag

Tats Svo ^Xcupats Tats
Tiyv AB
AE,
ry AE,
AZ
2TOIXEION a. 77
TTjV Se Ar Ty AZ* âxTis Sc rj BP âorews EZ

fJLUÔiV €OTQ)* X^/(0 on KCLl yOivLcL ^ VTTO BAr yOiVLaS lO
r^S VTTO EAZ fJLELîOV luTLV.
Et yap firj^ yroi tcrrj cortv aur^ *5 eXao'O’wv tcrjy
yxv ovy ovK ecrny tj vtto BAP ry vtto EAZ- Lcnj

yap ay rjv kol pdcris y BP pdarei ry EZ* ovk %<m
oc. ov/c apa ury ^crri ytovLa y vtto JdAI ry t/tto 15
EAZ' ovSc pyv iXdar<r(ov ecrriv yBAP vtto vtto
EAZ* IXaorcrtov yap av yv koX pdcn^ y BP pdareios
rys EZ* OVK cart Sc* ovic dpa IXaorcrwv coriv

BAP yiovCa rys vô EAZ, iSeiÔy Sc ort ouSc wn;*
pjetÔDv dpa eorlv y wro BAP rys wo EAZ. 20
’Eav dpa Svo rpCyma ras Svo irXcvpa? Suo^l
irXcvpats €^(2^ l/carcpav cKarepa, Sc pdaiv

rys Pdcrem ftctôva c;^?/, /cat ycvviav ywnag
peCôva c^ct t^v wo rSv torw cvOeuav 7r£pL€)(opanpr
OTrep ISct Sctât. 25
fcS'\
’Eav Svo rptywm rag Svo ycovtas Svct ywvCoLS

uras eyri kKarepav kKaripa /cat plav TrXevpdv
/aa TrXcvpa tcnjr ^rot irpog rats tcratg
yoavLaus y ryv worcn'oucrai' wo plav rdiv

Icrwv ycDVtoiv, xat rag Xotiras irXcupag ratg 5
Xotâtg TrXcvpatg to'ag l^ct [c/carcpav l/carcpa]
/cat Xoviryv yusviav ry Xonry ywvCa.
STO Svo rpCyoiva rd ABP, AEZ rag Svo yovtag

E rag wo ABP, BPA Svort roZg wo AEZ, EZA
wrag ôurra c/carcpav eKaripa, ryv pJky vjto ABP ry
wo AEZ, T^v Sc wo BPA ry wo EZA* Sc
78 ST 0 IXBIi2 N o'.
Kol jxLav 'TrXevpoLV /xta irX.€vp(l toTyy, '/Tporepov t^v
TTpo^ rais tcrais yoiVLcus t^v BF rfj EZ- otl
Kcu Tas Xot7ra5 TrXevpa^ rat? XotTrats TrXcwpat? wras
15 €^€1 iKaripav iKaripq^ rrjv p.ev AB AE Sc

AF AZ, fcat XotTTTiv ytaviav ry Xonrg ym/la,
ryv VTTO BAF Ty vtto EAZ.
Et yap avLtros icmv y AB ry AE, pXa avriov
p£L^<iiV kcTTlV. €0-T<l) fl€LÔyV TJ AB, Kttt /CCtCTÔ) rfj
20 AE lay y BH, Kal iw€^€vxOoi) y HF.

’EttcI ovv lay icrrlv y ph' BH ry AE, y Sc BF
ry EZ, Bvo Sy at BH, BF Svort rats AE, EZ ttrat
cicrtv CKarcpa cfcarcpa* Kat yoyvia y inro HBF ycovta

ry VITO AEZ toy Iotlv ^do’cs dpa y HF ^daei ry
25 AZ lay io-TLV, kol to HBF rpiywvov t(S AEZ rpt-
y(ov(o toov ctrrtv, /cat at Xoaral ywvtat rat? XotTrats
yoyvMLS tcrat corovrot, as at ttrat TrXeupal vtto-
Tcivovcnv lay dpa y vttoHFB yiovla ry vtto AZE.

dAAa y vrro AZE ry viro BFA VTroKCiTai toy /cat y
30 VTTO BFH dpa ry vtto BFA toy ioTiv,
y iXacrofOv
ry pelôvi' OTrcp dBvvarov. ovk dpa avtcros iariv y
AB ry AE. tcry dpa. cort Sc /cat
y BF ry EZ toy
Suo By at AB, BF Sucrt rats AE, EZ tcrat etatv
hcaripa e/carcpa* Kat ycov^ y mrb ABF ymla ry
:STOixEiaN a. 79
VTTO AEZ ianv lctt]- jârris apa 07 AF jSao-et Trj AZ 35
lar} ioTLV, koX Xoltt^ yoivia tj vtto BAP rfj XoLiry

jiDvla Trj VTTO EAZ ten/ io-riv.
’AXXa 8^ rraXiv loTowrai/ at vtto ras teas ywias

rrXcvpal inroreiVOvaraL t(rai^ «S 5
/
AB rfl AE* Xeyco
ttoXlv otl /cat at Xoirral rrXevpal rats XotTrats rrXev- 4^
/oats teat leoi/rat, 07 /t^ AP r^ AZ, 17 8e BP ry EZ
/cat ert 17 Xoltt^ yeovta 07 BAP Xolttq ytavia
ry VTTO EAZ tei/ ioTiv.
Et yap avteos eertv 07 BP tq EZ, /ata avrwv

p€tô)v ioriv, ecTTO) /leiîoVy ei Swarov, 17 BP, /cot 45
K^LcrOoy ry EZ ton/ t/ B©, /cat cxeêr;^^ci) 07 A©, /cat
€7r€t t(T7j icrrlv

y p^h/ B© t§ EZ, 7 Sc AB ry AE,
Svo 8-^ at AB, B© 8vet rats AE, EZ teat ctetv
iKaT€pa l/carcpa* /cat ycnvias teas Trcptôi/eti'* pdervi
dpa y A® /Saeet ry AZ ton/ lertV, /cat to AB© 5^

TpLymvov rS AEZ rptywci) leov eertv, /cat at XotTrat
ywvtat rats XotTrats ywvtais teat eeovrat, as at

teat TrXcvpot viroreivovaiv lay dpa eertv o/ wro
B©A ycoj/ta v-TTO EZA. aXXa vtto EZA ry 77
VTTO BPA ccrrtv ton/* rpLyiavov Srj rov A®P y c/cros 55

ywvta wro B®A teo/ leri Ty evros /cat drrG/avTLOv
-i/
ry inrd BPA* direp dSvvarov. ovk dpa avteos icmv
y BP Ty EZ* Lcry dpa. lert 8c /cat

^ AB Ty AE
ton/. Suo 8^ at AB, BP 8vo rats AE, EZ teat elcrlv
exarepa iKaTcpa* koX ycovtas teas Trcpiôuet* paxrm

dpa y AP pdau tq AZ ten/ lero', xat to ABP
rpCycavov rw AEZ rpiycovo) teoy xat Xouiry yoovta y
VTTO BAP XoLTry ytavCa ry xnro EAZ tei/.
TEay dpa Svo rpiyiova ras Soo ycoi/tas Svet yo/vtots
8o STOIXEION a.
^5 wra? €)(7j eKaripav iKarepa koI pLiav TrXevpav /x-ta
Trkevpa lu-tjv ^tol Tr)v Trpbs rats tcroL^ yôviaLS, ^ rip'
VTTOTCivovG’av ^6 piav T<oy urm/ ycoviSv, koX tcls
XotTras Trk^vpwi rats koiTrai^ TrkevpaLS wras l^ct fcat
TT^v koirnjv ycdVLav rg Xonry yaiviar o^rep ISci Sctât.
at.
*Eav €ts Svo evOcMS €V$€La ipTTLTTTOvcra ras iv-
aXka^ ywvtas wra5 aXXrjXais irapaX-
krjkoL ecrovrai dXX^Xats at cvOeLiu.
IS yap Svo eudeia^ ras AB, TA evOcia Ipirt-

E TTTOucra -7 EZ ras kvaXhh^ ywvtas rots inro AEZ,
EZA tora? dXX-^Xats 'Trotetro)* X^co ort TrapdXkrjXo^
icrnv nq AB Tg TA.
Et yap purj, l/cjSoXXo/tcvai at AB, TA avpTrecrovv-
rai TfTOL ejTL ra B, A piprj rj hrl ra A, T. eicjSejSXTy-
10 crOmcrav Kal ov^mirriTOKrav hrl ra B, A pifyq Kara

TO H. rpiymov 8*7 roi) HEZ ij Iktos ytovila vtto
AEZ Lcrrj ioTL rg Irros Kat airei/avriov rg inro EZH*

07r€p loTtv dSui/ai-ov ovk apa at AB, TA eicjSaX-
XopueyoL cnJ/x-wecrowTot €7rt ra B, A />tep7. ofioitos
15 8^ hcLxOria-eraL ort ov8c €7rt ra A, T* at 8€ hrl
prj^hepa ra piprq avpiruTTrovaraL irapdXXvjXoC eiaLV

‘TrapdXXrjXos apa icrriv y AB ry FA.
STOIXEION a. 8i
’Eav apa ci9 Bvo evdeLO.^ cuêta iaTriTrovcra to?

hâXkd^ yft)]/ia5 t<ras dXX^Xat? 'TrapaXXrjXoi
lorovrat at eu^ctat* oirep ISet Setât. 20
IC7j\
’Eav €ts 8t;o evSeias euêta ifjL'irLTrrovara
c/cros yinvCav ry ivTog koI aTrevayrtov kcu
i-TTL Ttt avra t(rqv TTOcy y rag iyrog koI

hn ra avra pîpy Svaiv opdaig tcrag^ irapaX-
IS yap Svo evOeiag rag AB, TA evOela epiri-

E TTTOVda y EZ rrjv iicrog yiovLav Trp/ vtto EHB
ry hrrog kcu aTrcvavTiov yidvia ry vvb H®A tcryv
TTOietTO) y rag ivrog koI eirl ra avra pipy rag vtto
BH@, H@A Bvartv opOoag tcras* Xiym otl irapak- 10
XyXog Icrriv y AB ry TA.

yap Lay iariv y inro EHB ry mro H®A,
"E^rct
aXXa y inro EHB ry hrb AH® ioTiv lay, Kal y

vurb AH® apa ry vtto H®A iariv lay Kai ela-LV
ivaXXd^' 'irapdXX.yXog apa iariv y AB ry TA. 15

HoXty, irrel at virb BH®, H®A Svo bpOaig laai
elatv, €tcrt Sc /cat at virb AH®, BH® Svalv opOaig
Sz STOIXEION a.
icrai, al apa inro AH®, BH® rats BH®, H® A

tcrai eicro'* kolv^ a<jiripnfj(r$(t) 17 vô BH®* XoLirr] apa
20 y VTTO AH® XoLTr§ ry mro H® A cortv icnp koL elaiv
haXkd^* TrapaXXyXo^ apa kcrnv y AB TA.

*Eav apa ets ^vo evOeia^s ^vOeux l/xTrArrovcra t^v
Iktos yijxvua/ ry cvros Kal dTrcvavrtoy koI cttI to.
avra p^^py tcnjv Troiy y rds hrro^ Kal iirl ra avra

25 pipy Bvcrlv opOais uras, TrapaXXyXoL ecrovroL al
evOeiat* &Trep eSct oeiâi.
k6 \
*H eh rds 'nrapaXXyXov^ evOeia^ evOeia ipiri-
Trrovcra rds tc cyaXXd^ ywytas uras dXXyXoL^

TroLeX Kal ryv inros ry eyrds ml aTrevavrCov
i<rqv ml rd? iyTog koI cttI rd avra pipy

5 Svcrly opdoLS tcras.
(f)
IS yap TrapaXXyXovs evOeCas rds AB, FA evOeXa

E e/AirtuTCTO) 17 EZ* Xiy<a OTt rds iyaXXd$ yiDvCas
rd? wrd AH®, H®A wras iroieX ml r^y cktos ytaviav

xnrb EHB ry cyrds #cal oLTrewavrCov ry wro H®A
10 Loyv Kal rds eyrbs kou hrl rd avra pipy rds inrb
BH®, H®A Bvotlv opScus wra5.
2 TOIXEION a. 83
El yap ai/KTos iarriv 17 AH@ tq vtto H@A,

pita avT(DV fM€L^(i}v ioTLV, eoTO) ^ AH®*
KOLv^ TTpocrKCLadio 17 wro BH®* at apa vtto AH®,
BH® tS>v vtto BH®, H®A p.&ôvi^ eiatv. dXA,a ai 15
virb AH®, BH® ^vcrlv opOaZ^ taai elaCv, [/cat] at
apa VTTO BH®, H®A Svo opOS^v iXacraroveg elariv.

at Se OLTr iXacrcrovoiv 17 Svo opOSiV e/cySoAAojU.ci'at €ts
aTTctpov avprnruirrovcTLV' at apa AB, TA eKpaXXo-

p.€vax €ts aireipov avpjir€<rovvTaL' ov arvpLTrLTrrovari Sc zo
Sta TO 'jropoXA.'iJXo'u? avroJi mroKeiaOaL' ovk apa
avioros coTtv ^ VTTO AH® rg vtto H®A* lctt} apa.
dXXa ^ wrb AH® ry vtto EHB iariv arq' xat 17
VTTO EHB apa rq vtto H®A iariv urq. KOtvq
q inro BH®* at apa vrrb EHB, BH® 25

'irpooTKeCcrOto
rats V7rb BH®, H®A tcrat cto-A. oAAa at wrb

EHB, BH® Svo opOoLLS laca ckrtv* /cat at viro BH®,
H® A apa Stjo op 0ats teat ehriv.
*H apa €t9 ras TrapoAA^Xous evOeLas evO^Ta e/t-
TcuirravG’a ra^ re kvaWb^ yiovias to*ag oXXqXaK 30

'TTOtet /cat T^ c/cTOS errog Kat d-arevarTtov ttriTv /cat
ras eVTOS Kat cm rd avrd p.€fyr} St/cty opOods tcras*
OTTcp eSet Setât.
6-a
8+ STOIXEIilN a.
X'.
Al avry cvOîa TrapdkXTjXoL koI oAXr]XaL<5

clcrl TrapdXXTjXoL,
A -- ”/ B
E e> Z
/"> “TO
r A
(C)
/ CP)
TA EZ
E STQ
EKaripa T<av AB, ry 7rapdXXr)Xos‘
X^(o oTi Kal y AB ry FA eotl ‘jrapdXXruXos.
5 ’Efo^tTirero) yap ch avras evOeia 17 HK.
Kal etteI eIs TrapaXXrjXov^ evOelu^ ras AB, EZ
EvOeia Ep'ir€7rT(s)KEy 'q HK, Ixrq dpa oj wo AHK ry
VTTO H0Z. vaXiv, etteI eIs TrapoAXiyXovs EvOeias ras
EZ,FA EvdEia lpfir€Trr(aKEy HK, 77 lctt} eotlv wo
77
10 H®Z ry VTTO HKA. €ÊC)( 0 Se Kal rj 77wo AHK

TQ wo H©Z urrj. kcu wo AHK 77 dpa vff wo
HKA EGTLV Lorq' Kol eIctlv evqXXjo^. TrapaXX7}Xo>s
dpa Ecrrly 77 AB FA.
[Al dpa TQ avTQ evOeux wapdXXrjXoi kcu oXX^Xai^s
15 Etxrl 7rapaAA77Xoc*] wcp eBel Safat.
STOIXEION a. 85
W.
Aca Tov SoOh/Tog (rqfieuyv ry 8oO€L(rg €vO€La
TrapaX\7jXov evdeiav ypafip^v dyayetv.
5T0 to SoOev o^fi€LOV TO A, Sc ôOeiaa

E
fJL€v 07
evOeia 17 BP* Set Std tow A cnqpLUOv Ty BP

evOeCa âpakkr^Xov evOeiav ypap.p.^v dyayco', 5
EtX77<^0<l) eTTL T^<S BP TVXOV ar7fpL€LOV TO A, KOU
€7r€^€vx^<*> V owccTTaTO) TTpos AA evOeta
Kol rw T-pos avrQ 0T7pe«u tS A ry viro AAP ytavCa
lurj 77 VTTO AAE- Kttt kKp^pXyidOdi hr evOetas tq
EA eu^cta 77 AZ. 10
Kat CTTct cts Svo cd^cids tols BP, EZ evO^La ip.-
TTLTTTOVO-a tJ AA TO-S cvaXXd^ ytavLOS to.? wo EAA,

AAP to’o.s ttAA-^Xais Treiroui^Keir, 7rapaA.A77Xos dpa
Wtr 77 EAZ TQ BP.
Aid TOV SoÔTOS dpa OTjpeCov tov A ry So^cio^ 15
extOeia ry BP TrapaXXrjXos evOela. ypappy yKTai 77
EAZ* oTTcp cSci TToiycrai.

86 STOIXEiaN a.
\0 .
IlavTos TpLytavov fuas rtov TrXevpSiv rrpocreK-

pXrjOeCoTj^s ^ c/cTOS ywta Svcrl rats ôs
#cat aTreuavTLOV Icrq itrrLV, koX al evros rov
Tpiyihvov rpets ywvLOL Svcrlv opOals tcrcu ctcrtr.
5 T7 STO rpiytmfov ro ABP, koI 7rpo<r€K^€pXri(r9(i>

JL^ airrov fua TrXevpa ^ BP hrl to A* Xcyo) ot6 17
€icTOS yiavla wo APA ten/ eort Svo"! rat? oros

K<u aircvavTiov rats wo PAB, ABP, /cat at etros
Tov rpiy^vov rpeis y(ov(ai ai wo ABP, BPA, PAB
10 Svo’tv opOcLis tcai elcrtv.
yap 8ta tov P cryjpLeiov tq AB evOcia

TrapdXXTjXos y PE.
Kal hrd irapaXXTfXos iariv 17 AB tq PE, jcat ets
avra^ ipjreirTdiKev 17 AP, at evaXXa^ ycovtat at vtto
15 BAP, APE Krat dXXyXats ctatv. ttoXlv, iirel irapaX-
XtjXos iernv rj AB ry PE, icat cts awas ipu7rhrT(t)K€i/
€v$€La y BA, tj Iktos ytovta wo EPA teny cart

17
rg euTOS Kal aTrevavriov ry wo ABP. iSelÔy Se

wo APE
Kttt 7] Ty wo BAP toT7* 0X17 apa y vtto
20 APA y(dVLa lory eoTt Soert rats ci/ros /cat a-jrcvavTtW

rats wo BAP, ABP.
STOIXEION a'. 87
KoLV^ TTpOCTKCLO-Oo} VITO APB* ttt apa VTTO AT A,
APB rpiG-l rats ABF, BPA, PAB tcrai eurtv.

aXX al VTTO APA, APB Bvalv opOals tcrat elcrlv kcu
at “yTTO APB, PBA, PAB apa Bvarlv opOaX^ vcrai 25
cicTLV.
IlavTos apa TpLy(avov fuas rtav TrA.eupSv Trpoor-

eK^XTjSeiarrj^ y c/ctos ycavia Bvcrl ratg ei/ros koI
anrei/avTLOV icrq i(rTLV, /cat at eiôs rov rpLymvov
rpcts ycovtiat Svarlv opdcus tcrat ctcrtv* mrep cSet Sctât. 30
V-
At ras t<ras rc /cat âpaXXijAoy? €?rt rot aura
/x,€pi7 hrLêvyvvovaraji cu^ctat /cat awat tcrat
T€ /cat TTopdXXrjXoi ctcrty.
& A
STOSAN tcrat T€ Kttt TrapaXXTjXoi at AB, PA,

E /cat hrL^cuyvvrmo'av aura? em ra avrX piprq 5
evOeuiL al AP, BA* X^cd OTt /cat at AP, BA tcrcu
T€ /cat TrapdXXriXoC ctcrtv.
T&7r€j€Tj;(^ci) tJ bp. Kat c-irct TrapaXX-qXo^ Icrriv r
AB ry PA, /cat et? awas ipLireirriOKeu ij BP, at

croXXa^ ymtoL al xnrb ABP, BP A tcrat dXX^Xcus 10
€tcrtv. Kttt CTTct ten; Icrrtv AB
FA /coti^ Sc y
BP, Stjo S^ at AB, BP Svo rats BP, PA tcrat cterty*
/cat ycovta 17 wo ABP yovta wo BPA tarj*

88 STOIXEION a.
apa
PaxTL^ AP pd(r€L tq BA
'q ia^rtv taq, koI to
15 ABP TpiyiDvov T(o BPA Tptydivtd tcrov ioTiv, koI at
XotTral ywvlat rat? Xoi^rats ytavLam lctoli ecrovrai €#ca-
T€pa iKarepa, v<j> d<s al tcraL TrXevpal vTroTeivovcrLV’
tarq dpa ry vtto APB yaivta ry vtto PBA. koI iirdi
€ts 8vo ^vOeiaq ras AP, BA evOeia lp.'tCuirrov<r<L

y
20 BP ras ôAA^ yoivia^ tcras aXX.yXais 'ireiroCyKcv,
TrapaXXyjXog dpa iariy yj AP ry BA. iBeiÔy Se

airy /cat tcrrj.
At dpa Tag teas r€ /cat yrapaXXyXov^ iirl to. avrd

pipy hri^€vyvvov(rai evBeTkt /cat awat to'at re /cat
2^ yrapaXXyXoL ftertv oTrep ISet Secâi.
XS'.
Twv yrapaXXyXoypdppwy '^^mpLwv at direvayTLOV
yrXevpcu re Kat ytoyiOA, torat dXXyXaL<s clcriVy
Kat oy Sidperpos avrd SC)(a ripyei.
ISTO 7rapaXXyX6ypappx>v x^^piov to APAB, Sta-

E perpos Se awov 17 BP* Aeyo) ort toO APAB
TrapaXXyXoypdppov at dyreyavrioy TrXevpcd re /cat
ywviaji LcroL dXXyXxas eioriv^ /cat 77 BP Staperpos

awo Stâ ripya.
’Exet yap TrapdXXTjXos iariv y AB t^ PA, /cat
2 T 0 IXEIQN 89
€ts avras ifLirenTWKCv ei'Saa 'q BP, at h/aXKa^ yco- 10
vtat at VTTQ ABr, BPA arat dAA'^Xats elcriv. ttoXiv
€7ret TrapaXkrjXo? ioriv 17 AP ry BA, K(u eU awas

kpjirkimuK&f -q BP, at ei/aXAaf ywvtai at "utto APB,
PBA L<raL aXkqXaL'S etcrty. Svo oq rplyiovd icrri ra
ABP, BP A rag Svo ycovca^ rag vto ABP, BP A ^5
Svcrt ratg vtto BP A, PBA tcrag €;(OVTa c/carepav
eKaTepa /cat /jtwiy TrXevpav pia TrXeupa i<rqv rqv Trpog
ratg tcratg ywvtatg /cot^r aiTwy BP* jcoi rag

XoLTras OLpa TrXevpag ratg Xot-iratg tcrag cfct cKarepav
eKwripa /cat r^v Aot^v ycdvtav vq XovTry ywvta* ten; 20

apa 97 pkv AB irAcvpa PA, 97 Se AP ttJ BA, /cat
€Tt t<r97 loTtv 97 VTTO BAP ytovâ wo PAB. /cat

€9r€t tin? IcTTLV
q vtto ABP ywvta tq vtto BP A,
97 Se VTO PBA Tq VTO APB, oXq apa wo ABA rj
0A.97 TT} VTO APA eerrtv ten/. iS€L)(0q Se /cat 97 wo 25

BAP wo PAB tcr97.
Twv apa TapaXXqXoypdfjLfjuov ^(opuav at aTrevav-
TtW TXevpal re Kat ymiox tcrat otXXrjXca^ elatv.
Aeyo) 81] OTt xal 19 8iap.eTpog awot 8txa 7ep,v€t.
wet yap ten/ cortv 17 AB PA, KOLvq Se q BP, 3°

8uo 8^ at AB, BP BP terat eieriv
Suert ratg PA,
iKarepa l/carepa* /cat yeovtix
^ wo ABP ymia r§
VTO BPA t<rq. Ktu pdcris apa q AP tq AB l(rq.
Kal TO ABP [apa] rpiyiovov tw BPA rpiyolva) ttrov
eerrtv. 35
'H apa BP Stap-eTpog Stâ T€pv€L to ABPA

TapaXXqXoypappLOv wep Set Setât.
90 STOIXEION a.
Xe.
Ta TrapaXkTjXoypafifJLOL tol eTrl tt}? avr^g pdorem

ovra Kal iy ratg avrcu^ TrapaXXqXoLS i(rcL
dXKrikous kariv.
TO ABPA, EBFZ
E2
'n-apaXkyjXoypapL/m ra ctti
Ttjs avrrj^ fidcrem BP Kal a/ rats avrals
vapaXXrjXoL^ toas AZ, BP* X^<o otl i<rov lari to
ABPA Tw EBPZ 'irapaXX7jX(yypdfiiJ,(p.
*E7r€l yap 'irapaXXrjXdypap.piov ia-ri to ABPA, urri
eoTLV ^ AA rg BP. 8ta Ta avTa 8*^ Kal EZ ry

10 BP eoTtv tony* wore Kal AA ry EZ icrriv tony* Kal
-q
KOLv^ -q AE* oXTy dpa AE oXyy ry AZ ccrriv t<ny.

coTt 8c Kal y AB ry AP teny* 8vo 8*7 at EA, AB
8uo Tats ZA, AP t(rat cto'lv cKaTcpa CKaTcpa* Kal
yo)vta -ly wro ZAP EAB cernv tony
ymvta ry wro iy
15 cKTOs rg cinros* jSao-ts EB ySacct rj ZP toy

apa ly
icrriv, /cat to EAB rpiyiovov rip AZP rpiydvio icrov
loTTctt* KOLvov d<j>ypyo'6 (i> to ABDE* Xolttov dpa to

ABHA TpaTTc^tov Xonrip tw EHPZ Tpairc^to) corlv
to-ov* KOLVOV 7rpocrK€LcrO<j) TO HBP rptymvov oXov

20 dpa TO ABPA TrapaXXyXoypapipLOv oX<p riS EBPZ
'irapaXXyXoypdp.pAp t<rov ierrev.
STOIXEION a. 91
Ta apa TrapaXkyjkoypaiiim ra im ryjs avT^s

pdcrem ovra koI iv tols avrais TrapaXX.ykois laa
aXXyXoL^ i(TTLV* oirep eSet Setjat.
Xr'.
Ta TrapaXXrjXoypapipa ra hrl icraxv pda-eaav
dvra Kcu kv rats avrats TrapaXXyXoi^ tea

oiXXyXoK icTTLV.
STO TrapaXXyXoypappia ra ABFA, EZH© iirl

E tetov pdeetav ovra t®v BF, ZH xat a/ reus S
avTOL^ yrapaXXijXoi^ rats on
A@, BH* Xey(o teov
ioTL TO ABFA yrapaXXyXoypapbpjOv rw EZH@.
^'7r€ê6x0(oeav yap al BE, F@. kcu cr-et terrj
eerlv y BF ry ZH, aAAa y ZH ry E@ Iotlv tcry,
Kol y BF dpa ry E© ianv very, curl Sc Kat yrapaX- 10
XyXoL. KCU hnêvytrvoveiv avras al EB, ©F* at Sc
ras ttras re xat yrapaXXyXov^ ern. to. avrd p^py

hnêvypvovecu tecu re kcu TrapdXXyXoL cun [/cat at
EB, ®F dpa txrai re cart Kat irapdiXXyXoi^. -jrapoA.-
XyXoypappjov dpa cart to EBF©. kol keriv teov 15
r<3 ABFA* pdeiv re yap avnS ryv avryv eêc rr^

BF, Kat iv rots avrois yrapaXXyXoLS cortv avrS tcus
BF, A©. Sta ra avrd 8y xal to EZH© rcS avrw

92 2 T 0 IXE 10 N
rS EBr© ioTLV tcov* wcrre koI to ABBA TrapoA-

20 XrjXoypaixfJLOv tw EZH® cortv itrov.
Ta apa TrapakXrjXoypafiiia ra CTrt wrcov pdcretav
dvra Kcu tolls avrcds TrapaXXiqXots tcra dXXT^XoLS

€(rrty* oirep deiât.
xf-
Ta Tptyo)va ra €?rl av-nj? pdcr£(as dvra koI
iv TOLLS avTOLS ‘TrapaXX’^XoLS la-a oXXtjXols
ioTLV,
STO rpiytova to, ABF, ABF €7rl t^s avrrjs

E pdL(T€<iis T^s BF Kal €v Tats avrats TrapaXXi^XoLS
TOLLS AA, BF* Xeyoi ort i(roy €crrt to ABF TpLytovov

Tw ABF Tpiywv©.
^K^ e^XiljorOKii ^ AA OKaTepa to. p.ep7
/
iwi to.
E, Z, /cat Sta p.hr tov B t^ FA TrapdXXyjXos

10 BE, Sta Se tov T rrj BA TOLpdXXTjXos V
TrapaXXTjXoypappxiv dpa icrTiv l/carcpov twv EBFA,
ABFZ* KCLL dariv to'a* ext t€ yap t^s avTTjs fidareds
etort tt}s BF /cat ev Tots avTous TrapaXXi^Xois tolls
BF, EZ* /cat eoTt tov ph/ EBFA TrapaXXrjXoypdp-

1
5 pov ‘^pj.LTv TO ABF Tptywov' ^ yap AB Stdperpos
avTO repvev tov Se ABFZ irapaiXXyjXoypdppxyv
5TOIXEION a'. 93
rjf/.L(rv TO ABF Tpiyayvov' ^ yap AF Buijjierpo^ avro

SL)(a re/tm. [ra Se rSiv lo’<ov -^[jiLfrq taa aXkrjkoL^
coTtv]. LOTOV apa ia-Ti to ABF rptyiavov t(S ABF
rpLytovw. 20
Ta apa rpiyaiva ra iirL t^? avrijs jSaorcft)? ovra
KoX hr Tai^ avrais TrapaXkrfkoLS tcra oAAiÂots ecrTtv*
oirep eScL Sctifat.
\r}\
Ta TpLywva ra iirl lorm Paor^m ovra koI iv

Ttttg aâts TrapoAAiJXot? tcra aXkyjXoL^s lortv.
ABF, AEZ IttI lg-wv pdorcfav

E3
Tfl Tpiyoyva ra
Tiov BF, EZ KOL iv rols avTats irapaXKrjX.ov:
Tai9 BZ, A A* X^w OTL tcrov €<rr\ to ABF Tptymvov S
T(S AEZ rpiytavip.
^KpepXyjcrOia yap ^ AA iKarepa ra pîpmj
TCI H, ®, KOt hid p.kv rov B ry FA irapoXXyXo^

y BH, Sta Se rov Z ry AE vapdXXyXo^
y Z®. 7rapaXX7jX6ypapipx>v dpa icrrlv iKarcpov twv io
HBFA, AEZ®- /cat tcrov to HBFA t<S AEZ®* hri
T€ yap icrcov jSdcrecov etcri tQv BF, EZ /cat iv rats
avrat? rrapaXXyXoL^ Tat5 BZ, H®* /cat €OTt toS phr
HBFA irapaXXTjXoypdfipiov ypacrv to ABF rpiycavov

94 STOIXEION a.
^5 17 yap AB StapiCTpos awo 8 t;(a Te/xvet* tov Se AEZ®

TrapaXkTjXorypdfJLfWv rjpLLcrv to ZEA rpCytovov' yap
AZ StdpLerpo^ avro S/;j(a ripivei [ra Se rwv tcrti>v
’qplaifi Lcra aXXrjXoL? cortV]. tcrov dpa io’rt to ABF

Tptywvov t«3 AEZ rptytovo).
30 Ta dpa Tpiymfa rd iirl t(r(i>v pdcr€(i)v ovra koI ev

rats avTois TrapaXkrjkoi^ taa dXXrjXoL^ Icrriv onrep
ISct
xd\
Ta L(ra rpCyoiva rd hrl r^g ainr^g ^daeiag dvra
Kcu hrl rd avrd p.€prj Kal ev rais avroLig
TrapaXki^XoLS icrrCv.
STO tcra rpiyiova rd ABP, ABF iirl Tijs avrrjg

E Pdtrewg dvra Kal iirl rd avrd p,epnf] rijg BF*
Xeyo) OTL Kol ev toxs avraig TrapaXXyXoLS icrriv.
*E7r€^€v)(0<o yap y AA* ort TrapdXXjjXog
ioTLv AA ry BF.
El yap pnj, ^;(9<£) 8ia tot) A crqfjidov ry BF evOeia
10 TrapdXXrjXog y AE, koi hreêvîo 17 EF. wrov dpa

iarrl TO ABF rptyayvov tw EBF TpiyoJvw* hrl re yap
T^s avTjjg pdoreds eariv avr<a tt}s BF koI ev raig
avTOLLg 'jrapaXXTjXoLg, dAAa to ABF t© ABF eoriv
i(rov icai to ABF dpa tw EBF torov Icrri to peiôv
STOIXEION a. 95
TttJ iXda-o-ovr OTrep icrTtv dSwarov* ovk dpa TrapdX-

Xr]X6g i<TTiv ^ AE t-q BP. opLotm otl
ovS’ dXX'q Tig t-X^v t^s AA* ^ AA dpa ttJ BP lort
‘TrapdXXrjXog.
Ta dpa lara rpiyaiva ra hrl r^s avr^g pdcreiog
dvTa Kal iirl rd avrd fi^prj koll hf raig avraig irapaX- 20

XTjXoig icTTiv OTrep eSei Sctâi.
r
fl.
Ta Lcra TpCyiava rd hrl la<av ôjcteidv dvra Kal
hri rd avrd fJiipy} kqI eu rtug avraig rrapaX-

XrjXoig ioTLV.
2TO ura rpiyiova rd ABP, PAE ejn urav

E pd(r€{ov rwv BP, PE Kal IttI ra avrd fîpTf, 5
X€y(o art Kal iy rats ain’ollg rrapaXXriXoLg iariv.
*E7re^€V)(0(o ydp ^ AA* X^<o OTt TrapdXXyfXog
ioTiv Tf AA ry BE.
Et ydp prj, ^)(0<o Sid rov A ry BE TrapdXXrfXog
rj AZ, Kal iir€^€vx0<o y ZE. ttrov dpa earl to ABP 10
rpiycavov rw ZPE rpcycSyw* hrC re ydp taivv pd(r€dv
ctcrt T(av BP, PE Kal ey rots avTodg TrapaXXyXoig

raig BE, AZ. dXXd rd ABP rpcycjvoy urov iarl rtS
APE [rpiyajvw]' Kal ro APE dpa [rptycovov] mtov
96 2T0TXEI0N a.
1 5 ecTTt Tw ZPE rptytinvio to fieiôv r<S iXdcr(rovv orrcp
ccrrty dZvvaTOV' ovk dpa 'srapdXk7)Xos ^ AZ ry BE.

opiOLOis 8^ BetÔjLeu otl ovB^ qXXt) 'ttX^v t^s AA*
^ AA dpa TQ BE ecrrt 7rapdXX7}Xo^.
Ta dpa Lcra rpiytova tol iirl urwv pdo-euiv dvra

20 Kal kirl rd avTC. pcprj Kal iv rats avrats TrapaXXi^XoL^
ioTLV direp eBet SeiâL.
fia.
*Eav TrapaXXrjXoypapLpLOv rptycovo) pdaiv re

T^v avT^v Kal ey rods avrats TrapaXXi^XoLS y,
BvTrXdarLOv i<m to ‘irapaXXyXoypapipLOV rov
rpiydivov.
S TTAPAilAHAOrPAMMON ydp to ABPA rpi-

X X ydv<o Tw EBP ^dcTLV TC €^CTQ> T^v avTyv ryv
BP /cat ev TOLLS avracs irapaXXyXoLS corw Tatg BP,
AE* Xiyui OTi BeirXdaLOv icrn to ABPA irapaX-
XrjXoypapLpxiv rov BEP rpiydvov.
lo ’E7rc^€v;(6a) ydp y AP. Icrov By ecrrt to ABP
rpCytovov t (3 EBP rpiydxvm' hri re yd.p rys avrys
j^dcrems iariv aur<S rys BP /cat ev Tats avraTs

rrapaXXyXoLS rods BP, AE. dXXd. rdABPA rrapaX-
XyXdypapLfjLov SiirXacrtov eort rov ABP rptyiavov y
5
:STOIXEiaN a. 97
yap AP BidpLerpo^ avro BL)(a ripiver oaorrc to ABPA ^5

TrapaXXrjXoypapsfjuov Kat rov EBP rpLytavov iart
BnrXacruov.
*Eay apa jrapaX.A.iyXoypappov rpiyoivw fidcrtv t€
^X2? aVT^v Kal ev ral^ a-urat? TrapoAX'jjXots 'Q,
St-jrXao-toy loTt to TrapaXkTjXoypafji/JLOV rod rpiydvov' zo

OTrep ISct SetôL.
Tw So^^Tt Tptytuyu) tcrov '7rapaXXy}X6ypapLfiov
aucmâraarOai iu Ty SoSetcry ycavLa evBv-

ypdpLpxa,
ESTO
to phf Tptywvoy to ABP, y Se Soêtcra
ytjivia eo^vypappos § A* Set tw ABP rpi- 5
ycdv(d t<rov 'jrapaXXyXoypap.piov <rvarTy(racr0aL cy t§
A ywta €v6vypdp.pxa.
Terp.y(rO(D y BP icaTcl to E, /cat eTrefevxô)
§ AE, Kttt crvv€<rrdro) irpos ry EP evOeca xai tw

Trpos airy ayp-eCto tw E A yoiyia ten; y xnro
t§ 10
PEZ, Kat Sta Tov A ry EP TrapdXXyXos yx0<o y
AH, Bid Sc tot) P ry EZ TrapdXXyXo^ y^B*^ y PH*
vapaXXyX6ypap,px>v dpa kerri to ZEPH. /cat circt
lot; coTty y BE ry EP, uroy ecnri Kat to ABE

rpLyiOvov t<S AEP Tptycovcp* hrl tc yap Lcro>v fidacdv 1
H. E. 7
98 STOIXEION a.
eicTL T^v BE, Er Kol h/ t(U£ avraL^ TrapaWi^Xois

rats Br, AH* SL7rXd<rioy dpa icrrl to ABF rpCyiavov
Tov AEr rpLydvov. tern Se koI to ZEFH 'rrapaX-
XTjXoypapLpov hnrXdaLov tov AEF rptydvov' ^denv
20 Te yap avT(a avr^v €;(€£. feat €V tols avraXs ianv
avT(p irapaXX'jXoLi' tcrov dpa icrrl to ZEFH irapaX-
XrjXoypappov rta ABP rpiydyw. Kal r^v vtto
FEZ ywvLav larqv tq ooBclcrri Ty A.

T<r dpa BoBci'n Tptyoivw nS ABF t(Tov irapak-
25 XyXoypapipov (rwierraTOJ. to ZEFH iy yayyia ry

VTTO FEZ, yns irrrlv tary ry A* OTrep cBcl TroiycrajL,
p/y\
Ilavros TrapaXXyXoypdpLpLOv twv irepi ryv Bid^

perpoy TrapaXXyXoypdppo>y rd irapaTrXypiD-
pxrra tora dXXyXois i(rriv.
^TO irapaXXyXoypappxiv to ABF A, Stap.€Tpos

E Sg aôv y AF, ircpi Be ryv AF TrapoAAiyXo-
ypappa pev goto) rd E@, ZH, Ta Be Xeyopeva
TrapaTrXypiapara rd BE, KA* Xeyo) on urov iarri
TO BK irapaTrXripfupa tw KA TrapairXypiapjan.
STOIXEION a. 99
’EttcI yap TrapaXXTjXoypapifJLOv IcrTt to ABPA,

Sta/x€Tpo5 Se avTOv AT, tarov icrrl to ABB rplywyov lo
t<3 APA rptycovo). TraXiv, hrei TrapaXXrjXoypappLOv
icTTL TO E®, StapCTpOS ttl&TOi; loTTtV AK, tCTOy
€<jTt TO AEK Tpiyctivov tw A®K Tpiywyw. Sia ra

avTO. Srj Kai to KZE KHP icrrcv tcov,
Tpiyayvov t<3
€7r£L ovv TO p€V AEK A0K rpiyiLvio
rpLyaivov t(3 15
ioTTiv taov, TO Se KZP t<S KHP, TO AEK Tpiyaivov
pLera too KHP tcrov icrrl rw A®K Tptywyw p,€Ta

TOO KZP* €<TTL 3e Kal oXov TO ABP Tpiya^vov oAo)
Tw AAP Lo-oy XoLTTov ttptt TO BK TrapaTvX'qpwfia
XoittQ tcS KA TrapaTrX'fjpmp.aTL kdTiv tcrov. 20
HavTOS apa TrapoXXiqXoypdpLpov x^piov twv vepl
T^v BtapiCTpov TrapaXXTjXoypdpLpLOiv to, irapaTtX'qpd-
ptaTa icra. dXXrjXovi icrtv* oirep ISct

zoo STOIXEION
/iS'.
Ilapa T^v oo$€L(TQ.v cu^€tav t<3 ôO&tl rpiyiavto

uTov TrapaXXyjXoypafifiov ‘irapaâXeiv ev tq
oo$€L<Ty yiavia evOvypdpLpâ.
STO rj p}v Sodelara (vOeux AB, to Se Bo6hr

E Tpiyiavov TO T, ^ Sc Soêwra ywvta €v6vypap.p,o*s
^ A* Set Stj T-apa ri^v 8o^€t(rav evOelav r^v AB t<5
BoO^l Tpiywvw Tto r t<rov TrapaXX.7jX6ypap.pjov nrapa-

jSaXetV ev wiy t^ A yayvia,
r TpiycSvw torov TrapaXX7jX6ypap.pjov

^wecTaTO) Tw
lo TO BEZH ^ ywvta t^ vrrd EBH, -q iarriv icrq ry A*
Kcu KeiaSoi (ootc Itt* evOeias eivai rrp/ BE tq AB,
Kol Q ZH CTTt TO ®, KOLL Sttt TOV A OTTOTCpa
tSv BH, EZ TrapoXX^Xo^s V €7r€êu;(â)
^ ®B. Kttt circi CIS TrapctXXT^Xovs Tas A®, EZ cv^cta

1 5 heireaev q ®Z, at apa vtto A®Z, ®ZE ywvCai Svaiv
opOcus elariv laat. at apa wro B®H, HZE Svo
6p6^v iXda-crovh cicrtv at Sc dxo cAacraovcov q Sdo
opOwv et9 oTreLpov iKôXXopevai (rupTriTrrQvcrLv at
STOIXEinN a. 101
©B, ZE apa ^K/SaXXofi^vaL avfnreo’ovvrcu.

arO(o(rav Kal crvfLTn'rrT&oxrav Kara to K, koI Slcl tov 20
K (ryfj.€Lov oTToripa tQ)v EA, Z0 TrapdXXyjXo^
^ KA, Kal iK/Se/BX'ô-Ooicrav at ©A, HB iirt ra A,
M (Trjp.€La. 'irapaXXrjXoypap.jjiov dpa icrrl to ©AKZ,

Siapêrpos Se avTOv ©K, irepl r^v ©K TrapaX-
X'qXoypaiip.a pilv rd AH, ME, ra Xeyofieva 25
wapaTrX'qpdypiara rd AB, BZ* icrov dpa icrrl rd AB
TO) BZ. aXXa to BZ t (3 B rpiycdvip icrrlv tcrov Kal

TO AB apa TO) r ecrriv tcrov. Kal €7ret tcrrf io'rlv y
VTTO HBE ywvta rrj vtto ABM, dXXa vrrd HBE
y
ry A ioTtv tcry, Kal y vtto ABM dpa rrj A yarvta 30
» V V
ecrrtv tcry.
Jlapd ryv ôOetcrav dpa dOuav ryv AB rcS SoOevrt

rptycDVip r<S T tcrov rrapaXXyXoypaptptov rrapa/Bi^Xy-
rat to AB iv ycovca ry vtto ABM, y ecrrtv tcry ry
A* OTrep eSet rrotycrat. 35
102 STOIXEION a.
lie.
Tw hoBoni evOvy/idfiiJAa tcrov 'TraâXkrjXoypafjL-
fwv <rv(TT7j(Ta(T6aL kv Soddarrj ywvta eu-

OvypdpLpM.
5TO TO phf evOvypapipuov to ABF A,

E So^cib-a ymta €v6vypap.pjo^ ^ E- Set
y
8^ tw
8c
ABFA evSvypdppxo to'ov irapaXkrjXoypapipov (tvottJ-

aacrOai iy rg So^curiy ycovta vg E.
^'jr€^€v;(^<i) rj AB, #cat aw€crTdT(i) to ABA rpt-

yoivw LOTOV irapaXXgXoypapifiov to Z^h/rg wro ®KZ
10 ymu^ rj eortv tony Tg E* K<d irapa.ôpXrj(TO(ii irapd
Tjgv H© evdetav tw ABF Tpiydvfo tcrov TrapaXXgXo-

ypcLfipm TO HM ci/ rg vtto H®M ymia, rj kcrriv
txrg rg E. kclL eirel
g E ywvta OKaripa rwv wro
®KZ, H©M ccTTtv Lcrrjj KOI g VTTO ®KZ dpa rg vtto
15 H©M coTtv LCTT). KOivg TTpofTKOLcrOfA

g xnro K®H-
at dpa VTTO ZK®, K®H toT? wo K®H, H©M tcrat
ctflrtv. aXX at vtto ZK©, K®H Svcriv opOcus tcrat

ctertv* /cat at wo K®H, H®M dpa Svo opOais icraL
:STOIXEION d. 103
etcrtV. TTpos Brj rivi evOeia ry H© koI tS Trpo^ avry

crqpi€L(a rta ® Stjo evOeiat ai K©, ©M p.y idi ra 20
avra pdprj K€Lp.€vaL ras €<j>€^s ywvLas 6vo opOals
Lcrag ttolovo-lv hr evOeta? apa icrrlv y K© Trj ©M.
Kol iirdi eh 7rapa\XyXov<s rhs KM, ZH evOela a-
hreaev y ©H, at ivaXXa^ yissvisLi at vtto M©H, ©HZ
torat aXXyXoL^ eurtv, Koan] TTpoaKeLcrSa} tj vtto ©HA* 25
at apa M©H, ©HA rats vtto ©HZ, ©HA
inro ttrat
elcrtv. aXd at vtto M®H, ©HA cuo opBcus teat
€t<rtv* feat at vtto ©HZ, ©HA apa 8vo 6p0a2<s laai
euTLV iir euêtas apa eariv y ZH ry HA. Kat iirel
y ZK ry ©H tcry re koI TrapdXXyXoq iartv, oAXa 30

fcat y ©H ry MA, /tat
y KZ apa ry MA lory re kcll
irapaXXyXo^s corti/* Kat iTriêvyvvovcTLV avras evêtat
at KM, ZA* Kat at KM, ZA apa torat re Kat irapaX-

XyXoL eicTLV' 'irapaXXyXoypap.p.ov apa eerrt to KZAM.
Kat hreX icrov corrt ro pjkvABA rpLym/ov t Z© <3 35
TrapaXXyXoypdp.p.co, to Be ABB t HM, oXor dpa
(3
TO ABBA €v6vypappx)v 6X<o to KZAM irapaXXyXo-

ypdfkpM eoTiv tcrov.
T« dpa BoOevTL evdvypdpbpxo rio ABBA tcrov
'TrapaXXyXoypappuov (TuveoTarai to KZAM ev ycovta 40

ry VTTO ZKM, y Icttlv lory ry Bodeioy ry E* direp
eBei TTOLycraL.
104 ::§TOIXEION a.
5T0 BoOeio-a €vOaa y AB* Set aTro t^s

E AB evOcLas rerpdywvov dvaypatj/ai,
5 AB evêta dim rov irpo^ avrp cny/x-etov
Tov A “TTpos 6p6d^ 7] Ar, Kcu Kî<r6(o TQ AB icrq y

AA* fcat Sta fiev rov A crripjîov ry AB irapdXkrfXos
T^)(6io rj AE, Sta Se rov B orrjpieLOV ry AA irapaX-
XyXos ^v^Q) y BE. UapaXXyXoypapLfjLOv dpa i<rrl ro
lo AAEB* LCTT} dpa iarlv y pkv AB ry AE, y Se AA
ry BE. dXXa y AB ry A A eartr lay' at reo'arapes

dpa at BA, AA, AE, EB tcrat dXXyXoL^ uariv' ioS-
irXevpov dpa iori ro AAEB irapaXXyXoypappjOv.
X^<o Sy on Kai opOoydvLOv. iirel yap ets irapaX-
15 XyXov^ rds AB, AE euêta h/iirârey y AA, at dpa
WTO BAA, AAE yayviai 8vo opOais tcrat eixTiv. 6p6y
5
2T0IXEI0N a'. 105
Sk rj VTTO BAA.* 6 p 6^ apa koI ^ vtto AAE. twv Sk

TrapaXXTjXoypdppojv ^(DpiW at OLTrei-avriOv —Xeupat
T€ Kal yiavidi t<rat aXXrqXau; etcrtv opOi] apa. /cat
€Kar€pa t(ov airevavriov rmr wro ABE, BE A yiDVLwv 20

opOoymnov apa ccrrt to AAEB. €Q€L)(97j Bk /cat tcro-
TrXevpov.
Tcrpaywov apa kcrrLv' koI icrrtv diro Trjs AB

evOcLas ai/aycypapLpL€vov OTrep ISct irot^crat.
/^?'*
’Ey rots opOoytDvtOLS rpiyiavois to dvo rrjs Trjv
opO^v ytoviav worcti/ovoTy? TrXevpds Tcrpd-

yo)vov L(rov ecrrt rots d^ro twv opOrjv
y<j)VLav 7r€pL€XOV(Tiav TrXeupwr Tcrpaycoyots.
STO rptycovov opOoy&viov to ABP opO^v eôy 5

E wo BAr ywvtav’ Xey(0 6ri to arro rys BP
r^Tpdyoyvov taov icrrl roig dirb rSyv BA, AP Terpa-
ywvots.
*Avayeypdtxj)6(j) yap dirb pkv ri}? BP Terpaytovov
rb BAEP, dxo Se tojv BA, AT rd HB, ®P, Kat Otd 10
Tov A oTToripa twv BA, PE TrapdAAiyXos V

/cat eircêuxOaio-av al AA, ZP. /cat e-Tret opîj eorrtv
CKttTepa T<ovwo BAP, BAH ywvtwv, Trpbs Brj rivi

€V^€ia rg BA /cat t<3 ^pos avr^ <rqpjeup tS A 8uo
€vO€iaL at AP, AH IttI Ttt awd /tcpi; Kupevox 1
io6 3TOIXEION a.
Tos ytrivias ^vow opBai^ tcra? 7rou)V<nv hr
euOeias apa icrrlv i] FA ry AH. Sia ra avra. 8^ Kat
4] BA tt} a® €<JTtv hr euê^S. Kal hrei tcrrj ecrrlv
7j WTO ABF yiovia ry vtto ZBA* opdr) yap l/carepa*
20 KOLvrj Trpo(TK€L(r6<t> y ABF* oXrj apa y viro ABA

oX^ ry ZBF iimv tcrrj. teal hra urrj lortv y
VTTO
ph^ AB ry BF, y 8e ZB ry BA, 8vo 8y al AB, BA

8^vo rat5 ZB, BF tcrat eialv iKaripa hcarepa' koI
ytavCa y vtto ABA ywvux ry vtto ZBF tary- paam
25 apa y AA jSatret ry ZF [eoriv] toy, /cat to ABA

rptyiovov rtS ZBF rpiywvw iorlv tcrov Kat [eort]
ToO phr ABA rpLy(Dvov SwrXaortov to BA TrapaX-

kyXaypappx)v âxriv tc yap ryv avryv eôvat ryv
BA /cat €V Tats aurots etat TrctpaXXyXoL^ rats BA,
30 AA* Tov 8k ZBF rptyw/ov StTrXao'tov to HB T€Tpa-
HOIXEmN a. 107
ymov' jSwriv re yap taXiv rrjv avr^ t^v
ZB K(u ^ Tais awais m irapaWi^koK rws ZB,
Hr. [to Se w «ra)v StirXacna m aXXijXws mar]
uTOv apa ml Kal to BA irapoXXijAoypa/ijnov to HB

rerpaym. o/iom (Tnletiym/ieam rm AB,BK 35
kixSôrerai Kal to PA JopaXkrjk^fpap/m icrov
0r Terpaymip' oXov apa to BAEP TiTpaymov oditi
TO is HB, 0r TCTpaywyois mv mar. Koi mi ro
/iw BAEr Terpdymv awo t^s BP rnypaê/, rot
§£ HB, ®r airo rSv BA, AT. to apa hto t^s BP 40
irXeupas TCTpaywov «rov €<rri tob ato Tuy BA,
Ar jrAcupfiy Tcrpaywyois.
’Ey TOis ofôymm TptywyoB to airo t^s
T^y op% ywiay woTeimarji irXsupas TerpaTOVoy

W fOTi TOB aw Twy T^y op^y [yoywy] Tepiexoo- 45
(rfiy xXojpSy Tcrpaywyoir mrep fSa 3fiât.

io8 STOIXEION a.
"Eav TpLy(avov to otto fuobs Toiy TrXevptov
T€Tpdy(ovov 1(TGV y TOts diro tq)V A-otTrSv tov
TpLydvov 6vo TrXeofmv Terpaywvois, Trcpi-
€)(opL€trrj ymla vriro rm Xouttwv tov Tpiydvov
5 8vo TrXevpwv opSy ioriv.
PirONOY yap TOV ABF to t^s BF

T vXevpds T€Tpdy<i)vov tcrov ccrro)
outto fiias
toTs diro tcov
BA, AF irXevpwv Tcrpayfavovs' Xeym dri opOrj icrTtv

^ VTTOBAF ymvia.
o yap dirb tov A on]pL€LOv rrj AF evOciq. Trpos
opOas ^ AA Kal Keiadw Tp BA Ixrq AA, koI hre- 17
'C€V)(Bo> ^ AF. CTTCt L(rq iarrlv

^ AA tq AB, tcrov
ioTL Kcu TO a/TTo TT]S AA reTpdiyoivov tw aTro Trj<s
AB T€Tpaywv(a. kolvov irpoorKacrOa} to diro t^s AF

5 Terpdywvov' to. dpa dwo t<ov AA, AF T€Tpdy<jt)va tara
ioTL ToX^ diro TGiV BA, AF reTpaywvoes. ^Xa rots
p€v diro Twv AA, AF tcrov €<7T6 to aTro t^s AF- opO^
yap kdTLv 7} VTTO AAF ycovta- rots Sc dirb riav BA,
AF tcrov idTL to diro Tjjs BF- wrOKUTai ydp‘ to dpa
:STOIXEIQN a. 109
an-b r^s AP Tcrpaycovov tcrov icrrl rS awb rrj^ BP 20

TCTpaywvw* woTC koX TrXevpa ^ AP BP ccrrtv
Lcrt}' Kal hrei Icnq iarrlv
^ AA tq AB, kolv^ 0€ ^
AP, Svo at AA, AP Bvo raZs BA, AP lorcu elcrtv'
Kttfc jSacrts 17 AP jSatrei tq BP toTj* ywvta apa ^ vtto

AAP ywvta ry vtto BAP [IcrTti/] tony. 6p 0y 0€ y 25
VTrb AAP- apa koI y wrb BAP.
opOvj
*Eav apa rpiyiavov to dn-o /x.tds tSv TrXevpSiy re-
rpdyisivov taov y rots dxd roiv XoiTrmv rov rpiydvov

Svo irXevpwv rcrpayclivot?, 9/ trepLexopanq ym/Ca virb
tSv XotrrSiV rov Tpiywvov Svo TrXevpwv 6p9y ioriv 3®
dircp eSet Sctfat.
NOTES
NOTES
THE DEFINITIONS
DEFINITION I.
W
ov=€K€Lvo
ITH the relative ov
a demonstrative
ovy ‘that of which,’
we have to understand
pronoun to which it refers;
And with /xepos
ovOa/y ‘no part,’ we must supply icn-Ly ‘there is.’
The whole sentence then means ‘A point is that of

which there is no part’ or ‘A point is that which has
no part.’ In other W'ords, a point is indivisible; and
other writers have expanded the dehnition by add-
ing that a point ‘has no magnitude,’
The Greek word for ‘point,’ ayjfJLetovy really
means a ‘mark,’ something, say a dot, to mark where
the point is. But the mark itself is not the point.
Any mark or dot that we can make, on a sheet of
paper, for example, has some size; you can always
divide it, or suppose it to be divided, into (say)
two parts, which are therefore both smaller than

the original mark or dot. This process of division
would never come to an end, and some other
method is necessary if from a dot which has size
and position we are to pass to a point which has
position but no size. We can only do so by abstract
tiafiy i.e. by mentally disregarding the magnitude of
the dot and regarding the point as indivisible and

having position, but possessed of no other attributes.
H.E. 8
114 - NOTES
This, in substance, had already been stated, before
Euclid’s time, by Aristotle. Aristotle in fact says
that a point is that which is indivisible in respect of
quantity and has position, that we can make no

distinction between a point and the place where it
is, and that, a point being indivisible, no accumula-

tion of points, however far it be carried, can give us
anything continuous, such as a line; it is only by
motion in space that a point can generate a line.
If you cut a line into two parts, or if two lines cut
one another, the cutting in either case is at a pointy

which is where the line is cut or the lines cut one
another; its position or place is determined by the
division or intersection (though of course we have
in like manner to think of the lines as mathematical
lines, i.e. as having no breadth whatever). Similarly
any line which has an end (or ends) ends in a point,
which is simply where it ends.

Before Euclid’s time the Greeks generally called
a point arriyfiT^, a. puncture (from ‘to prick’).
This indicates another way of marking a point, by

(let us say) pricking the paper with the point of
a very fine needle. But even this puncture, however
much smaller it isthan any dot we could make
with pen or pencil, gives an indentation of some
size, and is therefore scarcely nearer to an abstract
mathematical point than a dot is.
The objection to Euclid’s definition is that it

states rather what a point is not than what it is; we
reaUy have to get our conception of a point by other
DEFINITIONS i, z 115
means, namely by the process of abstraction above

indicated, before we are in a position to understand
Euclid’s definition. This is why other ancient
writers preferred other definitions. Plato, we are
told, objected to recognising points as a separate
class of things at all, and regarded them as a ‘geo-
metrical fiction.’ He preferred to conceive a point
as being merely the ‘beginning of a line’; alterna-
tively he spoke of ‘ indivisible lines.’ But, as Aristotle
says, even indivisible lines must have extremities:
hence an indivisible line (even supposing that there
is such a thing) must contain at least two points, and
cannot therefore be the same thing as a point. The
Pythagoreans connected the definitions of a point
and a monad, or unit, saying that a point is a ‘monad
having position.’ The monad or unit being regarded
as indivisible, this means something indivisible which
has position, and is therefore equivalent to Aris-
totle’s definition above quoted.
DEFINITION 2.
aorkares, neuter of the adjective dorkarijs
(from a- privative and TrXaTos, ‘breadth’), meaning

‘ breadthless ’ or ‘ without breadth.’ ‘ Aline is breadth-
less length.’
This definition is quoted by Aristotle^ and it may

safely be attributed to the Platonic school, if not to
Plato himself.
The idea of a line is less of an abstraction than
that of a point in so far as a line has one positive
8—2
ii6 NOTES
realisable diaracter, tizmAj length. As was remarked
in the school of Apollonius of Perga, ‘the great
geometer,’ we have a notion of a line when we ask

the length of a road or a wall; when we say it is a
mile, or so many feet, long, we ignore all question
of breadth or thickness. A line is, as Aristotle called
it, ‘magnitude extended one way’ (jLeyeOos ev

Staorarov), or, as we say (if the line is straight),
‘magnitude in one dimension (only).’
The idea of breadthlessness may be obtained by

considering a Hne as that which divides a surface or
that which is the intersection of two surfaces (just
as a point is a division of a line or the intersection of
two lines). Consider (i) the division between the
light and the dark when a shadow is thrown on the
ground; if it had had breadth, it would have been

either a part of the dark surface or a part of the
light, but it is neither and is only the division be-

tween the two. Similarly (2) with (say) the edge of
a cube; it is the breadthless intersection of two ad-
joining faces where they meet. But both the line
which divides the dark shadow from the illuminated

area and the edge of the cube have a length which
can actually be measured.
Another way of regarding the line was to conceive
it as ‘the flux of a point,’ i.e. what is described by
a point which moves in space. A material point such
as we can mark on paper has some size; hence by its
motion, if it were moved, it would describe a line

which has some breadth or thickness and is visible
DEFINITIONS 2, 3 117
to the eye as the material point itself is; but a mathe-

matical point which has no size would (if it could
be moved) describe a mathematical line, tvhich has
no breadth or thickness or visibility.
ypafLfj.7}^ lincy of course includes a curve as well as

a straight line.
DEFINITION 3.
-TTcpara, nom. pi. of ircpas, '

limit ’
or ‘
ex-
tremity.’ Understand icrTi: ‘ (the) limits (or ‘ ex-
tremities ’) of a line are points.’
This remark is an explanation rather than a
definition; it is a property of a point, as defined,
that it may be the end of a line. Plato, as we have

seen, is said to have preferred to regard a point as
the ‘beginning of a line’ rather than as an indepen-

dent class of thing, a point being, according to him,
merely a ‘geometrical fiction,’ not a real thing.
Aristotle alludes to a definition of a point as ‘a limit
of a line’ (rrepas ypapp.'^s), but objects to it on the
ground that it defines what is prior by means of

what is posterior, a point being in the order of
thought prior to, or more fundamental than, a line,
while a line is similarly prior to a surface and a sur-
face to a solid. Aristotle contrasts what is prior in
the order of thought with what is prior relatively to
us. Relatively to us, a solid is prior to a surface, a

surface to a line, and a line to a point. This is be-
cause a solid is nearer to sense than a surface is ^t is
the solid, as Aristotle says, which most of all ‘falls

n8 NOTES
under sense,’ i.e. is apprehended hy sense), and
similarly a surface is nearer to sense than a line, and
a line than a point. Consider a solid cube^ say a
piece of metal in that shape; it is a thing which we
can see and feel; and a mathematical cube is the
same thing with all but its geometrical attributes
omitted, but it still has three dimensions. Any one
of the faces is more remote from
a surface, but this is
sense, since we have to abstract all idea of depth or
thickness and to think of an area with no thickness.

A further step in abstraction is necessary in order
to apprehend a line in what we call an êdge’ of
the cube, one side of the plane face; for we have to
abstract the breadth of the face and leave only its
(breadthless) length. A third step in the process of

abstraction is required in order to apprehend a.
mathematical point in the corner of the cube which

is one end of the edge which we have taken.
Euclid, in giving this definition, shows that, while
he prefers, on principle, the scientific definition of
a point, he recognises another, the current, view of

it, which, as we have seen, was that of Plato.
DEFINITION 4.
This famous definition is full of difficulty; it is
difficult to translate the Greek, and it is not less
difficult, when we have translated the words, to

assign an intelligible meaning to them. A straight
line is any (line) which (^ts) lies (#cetrat) lo-ov
/C.T.A.’ The words cf ta-ov (where the adjective

to-os has almost the force of a substantive) are

generally used by Plato and Aristotle in the sense
of ‘on a footing of equality’; once Aristotle speaks
of ‘asking a question ii tarov/ which means asking a
question impartially, or without bias one way or the
other, i*e. without showing any expectation of one
answer being given rather than another. These uses
suggest that the words must here mean ‘ evenly
placed,’ ‘without bias,’ i.e. without inclining one
way or the other. Hence ‘
evenly ’ is not a bad
translation of tcrov. Next, the dative rots i<j>
iavry}^ cny/ictot? seems to be constructed with

t(Tov rather than with Kctrai, and we have the
following as the result: ‘
a straight line is any (line)
which lies evenly with the points on itself.’
Simson substituted for ‘

the points on itself ’
the
words extreme points,’ no doubt in order to
‘its
make the signification clearer; but Euclid’s words

are more general, since they take into account, not
only the ends of the line (which would not even
exist if the line happened to be unlimited or infinite
in length), but all the points on it. That is, Euclid
says that the straight line lies evenly with respect
to any two (or any number of) points on it; which

may be taken to mean that, whatever portion of a
straight line is taken, say between two points on it,
it shows with reference to these points no swerve or

unevenness towards one side or the other; both
sides are alike, in the sense that, if the line were
supposed to be turned over by revolution about the

120 NOTES
two points as poles, or turned the reverse wa^, as
between the tvsro points, with or without such
revolution, it would still occupy precisely the same
position (which would not be the case if it had any
bend towards either or any side in any part of its
length).
In modern text-books the idea is explained by
illustrations rather than by a formal definition. If
we consider a stretched string, e.g. a plumb-line, or
a ray of light entering a room through a small hole,

wê get a notion of a straight object, and by elimina-
ting all thickness from this straight object we arrive
at the abstract conception of a mathematical straight

line, of which the straight line that we draw is an
imperfect representation.
The only pre-Euclidean definition of a straight
line which has been handed down is that given by
Plato, who called it ‘that of which the middle
covers (literally ‘stands in front of,’ i.e. obscures
the view oi^ the ends,’ relatively, that is, to an eye
placed at either end and looking along the line.
This implies an appeal to sight, which would verify
(if the straight line could be seen end-on) that there
is no swerve in any part of the straight line in any
direction. Plato’s notion would therefore appear to
be in effect the same as that of Euclid according to
the above interpretation. It seems, in fact, probable
that Euclid got his notion from Plato’s definition,
but, as he saw that a purely geometrical definition
should be independent of any sense-perception
DEFINITIONS 4, 5 I2Z
(which belongs to physics rather than geometry), he

tried his best to express the same thing in general
terms without any appeal to sight.
Other ancient dehnitions are given by Heron of
Alexandria and Proclus, e.g. (i) ^ a line stretched to
the utmost’; (2) ‘a line such that ail its parts fit
on all (other parts) in all ways’; (3) ‘that line which,

when its extremities remain fixed, itself remains
fixed,’ i.e. the line which does not change its position
when it is turned about its extremities as poles (this
is equivalent to definitions given in modem times
by Ldbniz, Saccheii and Gauss).
The definition of a straight line as ‘the shortest
distance between two points ’ is also ancient; the fact

is stated, not as a definition, but as an ‘assumption’
(XafjLâvofjLevov), by Archimedes. This definition
again has been used, in modern times, by Legendre.
These alternative definitions, so-called, as well as
Archimedes’s assumption, are really not definitions
but rather statements of properties of a straight line
consequent on a straight line being, as Euclid says,
‘any line which lies evenly with the points on

itself-’
DEFINITION 5.
hrtâv€ia^ ‘surface,’ means literally that feature

of a body (the outside) which is apparent to the eye
(eiruliavrjs^ from hcL and the root of <l>aCv€cr$aL,
‘to appear ’)-

o— cKCLvo o, ‘that which.’
IZZ NOTES
We are told that the Pythagoreans called a surface
Xpota, which seems to have meant skin as well as
colour. There is this much of appropriateness in the
word that it is the colour of the surface which en-

ables it to be seen.
The definition of a surface corresponds to that of
a line. As a line has length but no breadth or depth,
a surface has length and breadth but no depth.
Aristotle calls it ‘magnitude extended, or continu-
ous, two ways^ {hrl ovo or Sixff); in the same way,
if a surface is a plane surface, we say that it is in two
dimensions. As a line is terminated or divided at a
point, and a surface in a line, so a solid is terminated
or divided by a surface.
We get an idea of a surface by considering, say,
the surface of a piece of water which is the boundary

between the mass of water below and the air above,
but has no thickness and is neither water nor air (if
it had thickness it would have to be one or other,

water or air); and similarly by looking at a shadow
on the ground, since a shadow has no depth (it does
not penetrate the earth).
As a point by its motion in space generates a line,
so a line by its motion in space (otherwise than along
its own length) generates a surface.
DEFINITION 6.
The remarks on Def. 3 apply mutatis mutandis

to this definition. Euclid does not d^ne a line as a
limit, extremity, or division, of a surface; this, as
DEFINITIONS 5-7 m
Aristotle says, would be unscientific; but be recog-
nises this aspect of a line by adding the remark in
Def. 6 as an explanation, not as a definition.
DEFINITION 7.
i^rh-eSog, ^plane’ (adjective, from hr[ and the

root -wcS; cf. 'n-oiis, fes^ a foot). The etymology of
the word shows that the idea is that of something
you can stand on, i.e. flat. hrvTrSov^ the neuter, by
means a ^plane.^
itself,
This definition follows exactly, mutatis mutandis^

the definition of a straight line, lirwrcSos hniôLveux^
‘plane surface,’ taking the place of ewêta ypajw.^77,
‘straight line,’ and T<w9...€iu^€tais, ‘the straight

lines,’ replacing Tots...cnyp,€tot$, ‘ the points.’
Hence we must translate thus: ‘A plane surface is
any (surface) which lies evenly with the straight
lines on itself.’ The explanation, too, of the
definition must correspond to that of Def. 4.
Proclus tells us that ‘a plane surface is a surface
the middle of which covers the ends.’ This is an
adaptation of Plato’s definition of a straight line.
What is meant is that, if we could see a straight line,

and if we were to place any two straight lines in the
plane in such a position that one line exactly covered
the other as we looked at them (this assumes that the
straight lines can be so placed), the plane would be
seen as a mere line coincident with the nearest of the
two lines; no part of the plane would bulge on either
side of the straight line seen by us. Euclid, then,
124 NOTES
seems to liave tried to express the same fact in
general terms, vpithout any appeal to sight.
Simson substituted in his Euclid a different
definition to the effect that ‘A plane surface is that
in which, any two points being taken, the straight
line joining them lies wholly in the surface,’ a
definition which is equivalent to others found in
Heron and Theon of Smyrna. Archimedes states,
as an assumption^ that, of all surfaces which have the

same extremities, the plane surface is the least (in
area) : this corresponds to his remark that, of aU lines
connecting two points, the straight line is the least.
DEFINITION 8.
The words of this definition should, for the
purpose of the translation, be taken in the following

order: CTTwreSo? yovta loriv 17 #cA,«rts Trpos dAAiJXa?
Bvo ypafLfjiwv Iv €7rt7reSa> aTTropuiyiav aXX'qXfov koI
hr KCifjLeytov. kXlo-is (from /cXIveiv) means
‘inchnation’ or ‘leaning.’ airrofji.h^fav (pres. part, of
aTrreorOaL, ‘to meet,’ not as a rule ‘to touchy which
is k<j>a.TrT€(rOaji) and Keifji&iav (pres. part, of KeurOat,
‘to lie’ or ‘to be placed’) both agree with ypafJLfjuav,
and aXk-qXwv is the objective genitive after airro-

pL&o>Vy ‘meeting one another? Trpb^ aXXi^Xas goes
with kXlo-l^ Svo ypapupMVy the ‘

inclination of two
lines to one another? We translate therefore: ‘A
plane angle is the inclination to one another of two
lines in a plane which meet one another and do not
lie in a straight line.’ The words twv ypafipuav be-
fore icA.t<rts are unnecessary and should be ignored;

they- were no doubt added in view of the separation of

kA-wtis from Svo ypafifLiav by as many as nine words.
The definition purports to cover an angle in-
cluded by a curve and a straight line, or by two
curves, as well as an angle formed by two straight
lines. This is evident from the fact that the recti-
lineal angle is separately described in the next
definition. This being so, the words Vhich do not
lie in a straight line’ are strange, since, if the lines
containing the angle are curves, it would be more
appropriate to say ‘which are not continuous with
one another.’ It looks as though Euclid really
intended to define a rectilineal angle in this de-

finition and then, on second thoughts, as a con-
cession to the then common recognition of curvi-
lineal angles, altered ‘straight lines’ into ‘lines’ and
separated the definition into two.
The evidence suggests that Euclid’s definition of
an angleas an ‘inclination’ was a new departure.
The word kXIo-ls does not occur in Aristotle, and we

gather from him that the idea commonly associated
with an angle in his time was rather deflection or
breaking of lines (fcAao-ts), i.e. a bend made by a
broken line. Apollonius defined an angle as a ‘ con-

tracting of a surface or a solid at one point under a
broken line or surface (respectively)’; the aspect of
an angle here emphasised is convergence to a point.
The converse aspect of an angle as the divergence of
lines from a point was also recognised; for some
writers, including Plutarch, said that ‘the first dis-
126 NOTES
tance under the point ’ (to 'irpwrov SidarTyffm wo to
cr'qiJLCLov) is the angle. To realise this, take two
points on the lines respectively at equal distances
from the point of intersection and very near to it,
and consider ‘the distance under the point’ as being
the distance between the two points so taken; we

have then a sort of measure of the rate of divergence
and therefore of the size of the angle.
DEFINITION 9.
The straight lines which form an angle are said
to ‘
contain ’
or ‘ enclose ’ (TrepLe)(€Lv) the angle : at
‘TrepLôvcrcu rqv ytovtav ypa/t/tat are therefore ‘the

straight lines containing the angle.’ eu^vypa/t/tos
(from €v 0v 9. ‘straight,’ and ypapijuLy, ‘Hne’) is recfi^
lineal^ i.e. formed by straight lines.
DEFINITION 10.
oTa^ciua, ‘set up,’ aorist passive participle of

lardvoL. an adverb meaning ‘successively,’
‘next in order.’ As Aristotle says, ‘that is successive

which is after the beginning and has nothing of the
same kind between it and that to which it succeeds.’
The word is technically used, as here, in the ex-
pression at- ymvLOLy to mean ‘the adjacent
angles,’ 6p0^ iKaripa riov i<r(ov ywvuov karCy ‘each
of the equal angles is right,’ i.e. is a right angle.
r/ ifj>€crrqKvia enfleta, ‘the straight line standing

(on the other),’ l^coriyicvta being the fern, of the
perf. participle active (used in the intransitive sense)
from c^tcrrdvat.
DEFINITIONS 8-ii 127
KO^cTOs, perpendiculary means literally *let fall’
(verbal adjective from KaOievcLi); tlie full expression
is 77 KaOeros evOeta ypa/A/xoy, ^ the perpendicular

straight line.’ The notion is that of a straight line
let fall on the surface of the earth, a plumb-line. We
learn from Proclus that in ancient times the per-
pendicular was described as dizvm gnomon-wise (Kara
yv<afjLovcL), because the gnomon (an upright stick, and
hence the perpendicular needle of a sun-dial) was
set up at right angles to the horizon.
i<l> hrl r^v ypafLfx^v i(j> the first kir\ being

constructed with KoSero^^ ^perpendicular’ (as we
say) ‘ to the straight line on which it stands.’
The fuU translation then is: ‘When a straight

line set up on a straight line makes the adjacent
angles equal to one another, each of the equal angles
is right, and the straight line standing on the other
is called a perpendicular to that on which it stands.’
DEFINITION II.
dpjSXvs is opposed to ‘sharp,’ and means

‘blunt,’ i.e. with the edge or point taken off or
blxtnted; hence, tvhen used of an angle, obtuse. We

may translate ‘an obtuse angle is an angle greater
than a right (angle),’ although the Greek idiom
prefers the definite article in such cases, ^the (angle
which is) greater than a right angle.’ 77 opOrj^
is of course reaUy the subject: ‘the angle (or that

angle) which is greater than a right angle is obtuse’
would be the literal translation.
128 NOTES
DEFINITION 12 ,
oivg, ^ sharp,’ or, when used of an angle,

'acute.’ Aristotle discusses the question whether
the right angle is firior to the acute or vies versa.
He concludes that the right angle is prior in notion
and in being defined^ the definition of a right angle
is independent, whereas an acute angle is defined

hj means of a right angle. Only as matter could the
acute angle be called prior, i.e. in the sense that a
right angle can be made up as the sum of a number
of acute angles (two or more), in which case the
acute angles are parts of the right angle, and in that
sense are the material of which it is made up.
DEFINITION 13.
opos, ‘boundary,’ The same word is used in

Greek for ‘definition’; ‘that which defines'^ would
in English, also, be true of a boundary, o = c/ceZvo
o, ‘that which.’
'Tripas, ‘limit,’ or ‘extremity,’ as usual. With
Aristotle opo5 and iripas are synonymous.
DEFINITION 14.
<r)(7jfjja.y ‘figure’; wcpLexpfjiôv^ ‘contained’ or
‘enclosed,’ as usual.
Before Euclid’s time ideas of ‘figure’ were more

vague. Thus Plato says that roundness or the round
(circular) is a figure, and that the straight is so too.
Aristotle too classes ang/sy straighty circular as species
of figure. Plato and Aristotle seem however to be
DEFINITIONS 12-15 129
here speaking of ‘figure’ in tlie sense of shape rather

than of ‘figure’ in our sense. Plato comes nearer to
Euclid’s idea of figure when he savs that ‘figure is
an extremity of a solid,’ ‘that in which the solid

terminates.’ Aristotle too elsewhere uses language
not unlike Euclid’s, distinguishing plane figures of
two kinds, those contained by straight and circular
lines respectively. The definition of ‘ figure ’
is
however probably Euclid’s own. It is clear from his
use of opos, ‘boundary,’ in the definition that he

would not regard a straight line as being a figure,
and probably not an angle either.

Posidonius defined ‘ figure ’ in the very short phrase
ff€pas omy/cA-ctov, ‘confining limit’ or ‘extrenaity.’
Since Euclid says that a figure is Hhat which is
contained by any boundary or boundaries,’ he in-

cludes in the idea not merely the boundary or con-
tour but what is inside it, the content, as well.
DEFINITION 15.
kvkXos, ‘
circle o^fMa hrLTreôv^ ‘ a plane
figure,’ i.e, a figure lying in a plane.
ypaiJLfjLyjs, ‘line,’ here used in the sense of a curve.

‘A circle is a plane figure contained by one line.’
The MSS. add, in further explanation of the one

‘line’ which encloses the circle, the words y koAcctcu
TTcptcêpeux, ‘which (line) is called circumference.^ But
it is probable that these words were interpolated by
some one who had in view the occurrence of the
word Tr€pi<j>€p€La without any explanation in Defs.
H. E. 9
^
130 NOTES
17 and 18 foDowing. The word however,
in the sense of contour without any special mathe-
matical signification, was well understood before
Euclid’s time, and he was therefore entitled to use
it in its ordinary meaning without defining it mathe-
matically. The meaning is of course that of some-
thing carried round (^rept and cf. the ad-
jective TTcptêpT^s, ^ round’ or ‘circular.’
TTpos -qv, ‘to which’ (sc. ypapupn^v), is constructed

wdth TrpocnrLTTTOvcraL (TrpooTrt'Trrw), ‘falling on which,’
i.e. ‘drawn to which.’
a<jy ivos (rqp.€LOV rOiV ei'ros rov cr)(rip,aTOS K€Lp.^(oVy
‘from one point cf (i.e. among) the (points) lying
within the figure.’ A literal translation of the

relative clause Trpos TjV.-.wraL aXAî/Xats ctcrtV would
be intolerable in English (‘on wrhich all the straight
lines which fall from one point among those lying
within the figure are equal to one another’), and we
have to say ^zohich is such that all the straight lines
falling upon it from one point among those lying
within the figure are equal to one another.’ The
wmds 7rpo5 T^v tot) kvkXov T-epwêpetav after cvOeiat
are obviously interpolated and should be ignored in
translating.
Passing from the definition of figure in general to

definitions of particular figures, Euclid naturally
gives first the definition of the figure which has only

one boundary, being bounded by one line, and then
proceeds to take in order the figures which have
more than one boundary (two, three, four, etc.).
DEFINITION 15
TJie definition of a circle contains nothing that

was new in substance. The same thing had been
said before in terms less formal but more terse.
Thus Plato says ^Round (i.e. circular) is, I take it,
that the extremes of which are every way equally
distant from the middle’; and Aristotle uses the
expression ^the plane equal (i.e. extending equally
all ways) from the middle’ for a circle, while he
speaks elsewhere of ‘the circular (literally ‘cir-
cumferential-lined,’ TrepLtpepoypajULiJLov) plane figure

bounded by one line.’
Observe that with Euclid and the Greek mathe-

maticians generally the circle is not merely what
we see drawn, the circumference, but the whole of
the plane figure included by it. Thus where we
might speak of a straight line meeting a circle the
Greek would be careful to say that the straight line
meets the circumference of the circle. There are
however a few exceptions even in Euclid; cf. iii, 10
‘A circle does not cut a circle in more points than
two.’
• Observe also that the definition says nothing on
the question whether any figure answering to the
definition can exist or not. This is not stated until
Postulate 3, where the possibility of drawing a
circle, with a given centre and ‘distance,’ is assumed.
It is true to say of definitions in general that they
state nothing as to the existence or non-existence
of the thing defined. The definition explains the
meaning of a term, and it is only necessary that it
9—2
132 NOTES
should be understood. The fact that the thing de-
fined exists has to be proved (or postulated) later,

and in geometry is generally proved by actual con-
struction. Thus that a right angle exists does not
follow from Def. lo but only from Prop, ii of
Book I, where it is actually drawn. Similarly with
a square, which is defined in Def. 22, but is not
shown to exist till it is actually drawn in i, 46.
DEFINITION 16.
Kevrpov, centre.
^ ’ ‘
And the point ’ (namely the
point within the circle which is such that all the
straight lines drawn from it to the circumference
are equal) ‘
is called (the) centre of the circle.’ The
word Kevrpov was regularly used even before
Euclid’s time; it means literally that which makes a
puncture (KevretVy ‘to prick’), e.g. the stationary
leg of a pair of compasses when used to draw a
circle; hence it comes to be used for the place (the
centre) where the puncture is made.
DEFINITION 17.
(^, diameter, literally ‘that which

8td/jt€Tpos
measures through ’ (Std and fierp€Lv), was the

regular word, in Euclid and in the classical Greek
writers, for the ‘ diameter’ of a square or a parallelo-
gram as well as for the diameter of a circle; Sta-

ydivto<s, ‘diagonal,’ i.e. passing through from angle
to angle (8ta and ywvta), was a later term.
€v$€Ld TLs, ‘any straight line’; ‘drawn’
(fern, of the perf. participle passive from dyeiv, a word
DEFINITIONS 15—17 133
regularly used for ‘

drawing ’ lines in geometrical
constructions) ; 'irepaTOVfjiiyri (Trepardto), ‘

limited ’ or
‘terminated’ (cf. Trepan, ‘limit’ or ‘extremity’).
i<f>^ cKarcpa ra p-ipr}, ‘in both directions,’ literally
‘towards both the parts,’ where ‘parts’ must be

used in the sense of ‘regions’ (cf. Thuc. ii, 96).
best translated ‘and any such (straight line),’
i.e. any diameter.
Uxo. TCju-vet, ‘bisects,’ literally ‘cuts in two’;
St;(a Tc/Jiv€tv, like 8tva Statpeiv, regularly implies
cutting into two equal parts.
We thus arrive at the translation ‘A diameter of
the circle is any straight line drawn through the
centre and terminated in both directions by the
circumference of the circle, and any such straight
line also bisects the circle.’
Proclus relates that the first to prove that a circle

is bisected by its diameter was Thales. We are not
told how he proved it. Presumably the proof would
be based on the symmetry of the circle with respect
to any diameter, however this was brought out, e.g.
by turning one of the portions into which the
diameter divides the circle over, applying it to the
other portion, and proving that the two portions

coincide exactly when so applied.
The last words ‘and any such (straight line) also
bisects the circle’ are omitted by Simson and the

editors who followed him as not belonging to the
definition but stating a property of the diameter as
defined. All the same they are necessary where they
’
134 NOTES
occur; for, without this explanation, Euclid would
not have been justified in describing as a rmf-circle
(in Def. i8) a portion of a circle bounded b^ a
diameter and the circumference cut ofF by it.
DEFINITION i8.
y/jLLKVKkLovy ‘ semicircle ’ (^/xi= semi and kvkXo^s).

The construction (ro) o^fJia to Tr€pL€)(6p.a/ov
is
K.T.X., ^the figure contained by the diameter and

—
-lyiro
oLTroXafijSavofjLa/yj^, ^cut off’ (pres, participle pass,

of airoXafjLfidveLvjy agreeing with ir^pi^^petas. vtt
avr^s, ‘by it,’ i.e. by the diameter.

Observe that ^cpLcfeipeui is used here, as often, for
a fart of the circumference, i.e. what we call an

‘arc’ of a circle.
The full translation is ‘A semicircle is the figure

contained by the diameter and the circumference
cut off by it.’
The last words, Keurpov Se rov t^pukvkXiov to avroy

o Kal Tov kvkXov icTTLVy ‘And the centre of the semi-
circle is the same as that of the circle’ — ^literally ‘the
centre of the semicircle is the same (point) which is
also (the centre) of the circle’ —come from Pro-

clus’s commentary and not from the mss. Proclus
remarks, rather absurdly, that the semicircle is the
only plane figure that has its centre on its peri-
meter (!). The fact is that, as Scarburgh said in The
English Euclide, a semicircle as such has no centre,
properly speaking; the ‘ centre ’
is not the centre of
the semicircle but only of the circle of which it forms
DEFINITIONS 17-19 135
part. Tke words are probably better left out alto-

gether.
DEFINITION 19.
In defining particular figures Euclid (as we said)

takes them in an order corresponding to the number
of their boundaries. He begins with the figure (i.e.
the circle) which has ojie boundary; he proceeds

then to the figure which has namely the semi-
circle (which bounded by its circumference and
is
its diameter); and after that he takes the figures

which are bounded by three or more lines, in order.
All the figures defined in Defs. 19-22 are bounded

by straight lines and are therefore included in the
class of rectilineal figures.
evOvypafjifxos (from evOv^ and ypayLp^rj)^ recti-
lineal^ meaning, as Euclid immediately explains,

‘contained by straight lines.’
rpeirXevpov, ‘three-sided’ or ‘trilateral’ {irkevpa

— ‘side’); TerpdrrXeupov^ ‘four-sided’ or ‘quadri-
lateral’; TToXihrXeupovy ‘many-sided’ or ‘multi-
lateral.’
With rpioiVy wo recrcrdpwvy wo ttXciovwv ^

Tcororapo)!' understand €v 0 €L(hv: TrXctdvwv ^ rco-o-apw,
‘
more than four.’
The latter part of this definition distinguishing
‘three-sided,’ ‘four-sided’ and ‘many-sided’ figures
is probably due to Euclid himself, since the words

rpLirXevpovy rerpaarXevpov and 'TroXwrXevpov do not
appear in Plato or in the genuine Aristotelian
136 NOTES
writings. Bj Hs use of TerpaTrXeupovy quadrilateral,
Euclid no doubt wisbed, once for all, to put an end
to anv ambiguity in tbe use of the word rerpayiavov,
literally a ‘four-angled (figure)/ which had some-

times been used in the general sense of anj four-
sided figure, and to get it formally restricted to the
square, for which it is the regular term.
DEFINITION 20.
The regular word for ‘triangle^ is rpiyiavov,
literally ‘three-angled (figure),’ and this definition

distinguishes triangies as being of three kinds
according to the relations between the lengths of
the sides, while the next definition classifies them
with reference to their angles.
la-OTrXevpov, ‘equal-sided’ or ‘equilateral.’
In translating to ras rpeis icras ôv TrXeupd^ we

take the words in the order to €)(ov tus rpets TrXevpws
Lera'S, ‘that (triangle) -which has its (literally ‘the’)
three sides equal.’ Similarly in the definition of the
second kind of triangle (the isosceles) we take the
words in the order to ôv ras Svo TrXeupas jxovas
L<ra9, ‘that (triangle) which has two of its sides
(literally ‘the tw'O sides’) alone equal’; and in the
definition of the third kind of triangle we translate
the words in the order to rpeeg TrXevpag
oLvterovg.
ieroo’KeX'qg, isosceles, meaning ‘with equal legs^

(cr/c^s). The word is used by both Plato and
Aristotle.
DEFINITIONS 1^21 137
o'KttX'jyi/ogj scalene, also used by Aristotle of a

triangle with, no two sides equal. Proclus seems to
connect the word with crKoêLv, ^to limp’; others
make it akin to or/coXtds, ^crooked/ âslant.’ Apol-

lonius uses the word scalene of an oblique circular
’
cone. Plato in one place applies the term ^ scalene
to an odd number in contrast to ‘isosceles’ used of

an even number {Euthyphro, 12 d).
DEFINITION 21.
opOoyiDVLO^ (from opOog and yiavLOL), ‘right-
angled’; ajxpXvydiVLOs, ‘obtuse-angled,’ and 6 ^-

‘
yoiVLos, acute-angled,’ are similarly formed.
The descriptions of these triangles respectively
require careful translation, ro eôv 6p 6^v ywviav,
‘that (sc. triangle) which has a right angleî to
côv appXdcLV y(j}VLav, ‘that which has an obtuse
angle^ I to ras rpeis 6 ^€tas ôv ywvta^ (which we

must take in the order to eôv ras rpels ytavias
d^€ias), ‘that which has its three angles acute.’

The difference in form between the first two ex-
pressions and the third is due to the fact that, while
an acute-angled triangle has all its angles acute, the

other triangles can only have one angle right or one
angle obtuse, as the case may be, and the other two
angles must in either case be acute. This follows
from the fact (proved in i, 32) that the sum of the
three angles of any triangle is equal to two right
angles.
; \
138 NOTES
DEFINITION 22.
Quadrilaterals are now classified with reference

to the attributes ^
equilateral ’ and ^
right-angled ’ or
the absence of them respectively. The alternatives
are clearly that a quadrilateral may be
(1) both equilateral and right-angled: this is a
square',
(2) right-angled but not equilateral: Euclid says

that this quadrilateral is an oblong (irepojjrqK^q)
(3) equilateral but not right-angled: a rhombus

or
(4.) neither equilateral nor right-angled.
Dealing with the last class, Euclid distinguishes,

from the rest, the figures which, though they have
not all their sides equal, have the opposite sides
equal respectively, and the opposite angles equal
respectively, in pairs. These figures he calls by the
name rhomboid. All the rest of the class he calls by
the general name trapezium (rparri^Lov, diminutive
from rpaireâ, a table, i.e. ‘a little table’).
It is clear that ‘right-angled’ in this definition
means ‘ having all its angles right,’ because, although

for a square, which has all its sides equal, it is
sufficient to say that one of its angles is right, this
is not the case with the oblong which is not equi-

lateral. As a matter of fact, the quadrilateral figure
(a square) which has all its sides equal and one angle
right has all its angles right; but the quadrilateral
which has its sides unequal and one angle right has
DEFINITION 22 139
not necessarily any more of its angles right, whereas

what Euclid calls an oblong is really what we call a
rectangle, a parallelogram with all four angles right.

Tcrpaytavov^ literally ‘four-angle,’ was already a
square with the Pythagoreans, and it is so most
commonly in Aristotle; but in a few passages
Aristotle evidently uses the word with the wider
signification of ‘quadrilateral.’ By introducing the
new word rerpaTrXevpov for the quadrilateral in
general Euclid wished no doubt to confirm the
conventional restriction of rcrpaywyoy to a square,
iT€p 6pLTjK€<s means literally a figure with sides of
different lengths, but it was conventionally appro-
priated even by the Pythagoreans to the oblong, in
other words a rectangle,

po/x.ôs, rhombus, is apparently derived from
pipuêiv, ‘to turn round and round’; it meant among
other things a ‘spinning-top,’
pop-ôctSes, rhomboid, literally ‘rhombus-shaped.’
auTrei/avTLov, ‘opposite’ (cf. h/avrlo^), an adverb

used as an adjective qualifying nrXevpds and ycovta?
respectively : ‘
that which has its opposite sides and
angles respectively equal to one another.’ The
rhomboid is really a parallelogram. Observe here
too that it is not yet proved that any such quadri-
lateral exists; that it exists is first proved in i, 34.
Euclid does not anywhere define a parallelogram.
The definition in any case could not have preceded
the definition of parallel straight lines (Def. 23);
but, when we come to parallelograms in Book i.
14-0 NOTES
Euclid introduces them as ^parallelogrammic areas
or, more exactly, ^parallelogram areas’ (TrapaXXTjXo-
ypafjifjLa x<»>pta), evidently regarding the expression

as self-explanatory (— areas contained by parallel
lines) ;
after which, in the same proposition (i, 34), he
shortens the expression into ‘parallelogram,’ a word
which is, in reality, still an adjective with )(0}pLov
understood, although we make a substantive of it.
Though Euclid, in the above definition, classes as
trapezia all quadrilaterals which are not squares,

oblongs, rhombi or rhomboids, the term was gener-
ally restricted by later writers to quadrilaterals having
two sides (only) parallel. It is used in this sense by
Archimedes and Heron, and apparently by Euclid

himself in his book On divisions (offigures).
Euclid makes no use in the Elements of the terms

oblongs rhombus and rhomboid. The explanation of
his inclusion of definitions of these figures is prob-
ably that they were taken from earlier text-books.
DEFINITION 23.
TrapdXXyjXoSy ‘parallel,’ means literaUy ‘along-

side one another’ (irapd and aXXrjXa^).
eKfioXXopeycUyêing produced.’ iKfidXXetv (‘throw
out ’) is the regular expression in Greek geometry for
producing a straight line.
€t5 dxeipov. aTTCipos (a- privative) is connected

with rripa^y ‘Emit’ or ‘end,’ and means ‘without
limit’ or ‘ infinite.’ dwetpov is adverbial, meaning
‘without limit’ or, as mathematicians commonly
say, ‘indefinitely.’ Simson translated ‘ever so far,’

wMcli of course is not far enough.
i<l> eKarepa ra ‘in both directions,’ as in
Def. 17.
€TrL pLTjSirepa: understand ra p^^pTjy ‘in neither
direction.’
‘Parallel straight lines are any (straight lines)

which, being in the same plane and being produced
without limit in both directions, do not meet one
another in either direction.’
x\nother viev/ of parallel straight lines held in
antiquity was that they are such that, however far
they are produced in both directions, they always
remain the same distance apart. The distance be-
tween them is of course the shortest distance, which
is measured by a straight line perpendicular to both
the straight lines. Proclus quotes a definition given
by Posidonius to this eflFect: ‘Parallel straight lines
are those which, (being) in one plane, neither con-

verge nor diverge, but have all the perpendiculars
equal which are drawn from the points of the one
(straight line) to the other.’ This definition is how-
ever (as Saccheri pointed out) unsatisfactory be-
cause, before such a definition can be used^ it has
to be proved that the geometrical locus of points
equidistant from a given straight line is also a
straight line.
It is dear that with Aristotle the general notion
of parallels was that of straight lines which da not
meet^ as in Euclid.
THE POSTULATES
W
mand
E come now to five assumptions known as
The Greek term is ah-rjim, a de-
Postulates.
or fostulate^ formed from alreiv, ‘to ask’ or

‘demand.’ Euclid does not here use the substantive
but a part of the verb, namely 3 rd pers.
perfect imperative passive. This use of the perfect
imperative passive is very characteristic of Greek
geometry, and it is very expressive. Thus yr^crOo)
means literally ‘Let it have been demanded,’ not
merely ‘Let it be demanded.’ Similarly, in con-
structions, we have y€ypdiij>0(Oy ‘let it have been

described’ or ‘drawn,’ and (from ayeiv) ‘let
it have been drawn’; that is to say, the one word

means more than ‘let it be described (or drawn)’;
it is equivalent to ‘suppose it described (or drawn).’
This elegant and concise mode of expression is not
possible in English which has not the richness in
inflexions that the Greek has; and we have in prac-
tice no alternative but to translate by the present
imperative, ‘let it be demanded,’ and the like. In
the present case, in order to keep tho technical term
postulate^ it is convenient to say ‘Let the following
be postulated.’
The nature of a Postulate is well brought out by
Aristotle, who discusses at considerable length the
terms definition^ bypothesisy postulate and axiom in

relation to one another. In demonstrative sciences
THE POSTULATES H3
such as geometry it is necessary to begin from certain
assumptions which are not capable of demonstra-
tion: otherwise there would be an infinite series
of demonstrations. One class of indemonstrable as-

sumptions is that of ‘axioms’ (in Euclid ‘Common
Notions’), for which Aristotle also uses the alter-
native terms ‘
common (things) ’ (ra Kotvd) or
‘
common opinions ’ (kolvoI Soôl). Axioms are
general principles, common to all sciences, and self-
evident though incapable of proof, e.g. that ‘one

of two contradictories must be true’ or that ‘ when
equals are subtracted from equals, the remainders
are equal.’ Any person capable of reasoning at all
must accept such assumptions as these without

proof, and every one must grasp and hold them
firmly if he is to learn anything at all. A postulate is
also a necessary assumption, but it differs from an
axiom in that it is not self-evident, or a necessary
truth 'j
the learner need not be convinced of it by
the reason that is within him; it is a thing assumed
by the learner on the authority of the teacher only,
and the learner himself may have no opinion of his
own on the subject, or the postulate may actually
be rather contrary than otherwise to his opinion if
he has one. This description seems to fit the Postu-

lates of Euclid accurately enough; they are demands
made without any assent on the part of the learner,
but they are necessary as a basis if a science is to get
under way at all.
The first three Postulates are generally called the

144 - NOTES
Postulates of construction. But they are more than
that. Aristotle had remarked that, before we can
have a demonstrative science dealing with points
and we must be satisfied that points and lines
lines,
exist. The existence, then, of straight lines and

circles is postulated by Euclid in the form of a
categorical assumption that such lines can be con-
structed^ the actual construction of a thing (if it can
be effected) being a suifident proof that it exists.
There is a further point. No straight line or circle
that we can draw with the imperfect instruments

at our disposal is a perfect (mathematical) straight
line or drcle satisfying the respective definitions.
Euclid was aware of this, and yet he deliberately

assumes that mathematical straight lines and drcles
can be drawn. By this he implies that the imper-
fection of our instruments and the inaccuracy of the
figure actually drawn do not matter. We are entitled
to suppose that the true straight line or circle is
there, although we have not been able to draw it
accurately; the validity of the proof is not affected

by the imperfection of the figure, because the proof
is not concerned with the particular imperfect
straight lines and drcles which we have drawn but
with the real straight lines and cirdes of which
those that we draw are only illustrations. As
Aristotle says, a geometer may quite legitimately
call a line a foot long when it is not, or straight when
it is not straight; he is not on that account using
hypotheses which are false.
POSTULATE I H5
With these general observations, we pass to the
individual Postulates.
POSTULATE I.
d-TTo TravTos anfjfJiCLOV cttI Trdv uqiJiôVj literally

‘from every point to every point.’ In such cases as
this the Greeks speak of every where we say any.

The meaning of ‘ every ’ here is of course ‘ any
whatever’ or ‘any we please.’ Cf. the enunciation
of T, 1 8, ‘In every triangle the greater side subtends
the greater angle.’
dyayeiv, aorist infinitive of dyeiv, ‘to draw.’
The full translation is ‘Let the following be pos-
tulated: to draw a straight line from any point to
any point ’ (i.e. that we can draw, etc.).
The statement that we can draw- a straight line
from any one point to another implies that such a

straight line exists; but the Postulate has yet
another implication, namely that the operation is
one, and that the straight line joining the two

points is unique', we cannot draw two (or more)
different straight lines joining two points, and this
again implies that two straight lines cannot enclose a

space. Consequently the axiom to the latter effect
which was interpolated among the genuine Eu-

clidean axioms is unnecessary. That Euclid had no
axiom to this effect is clear from i, 4, in the proof
of which he tacitly assumes that two straight lines
cannot enclose a space without referring to any
such axiom.
H. E. 10
146 NOTES
POSTULATE 2.
'jre.nrepao'ixhnii'y ^limited’ or ‘terminated,’ tliat

is to say, a straight line which has ends. It is possible
also to translate TreTr^paa-fieyrjy by ‘finite,’ just as
a^retpo?, ‘unlimited,’ may also be translated by

‘infinite.’
Kara to ouvex^^y literally ‘according to the con-

tinuous’ or ‘continuity’; i.e. ‘continuously.’
CTT euOeiag, literally^ ‘on a straight line,’ i.e. ‘in
a straight line’; iKô^Xciv, slot, infin. of iKjSaXXeLVy

‘to produce,’ as usual.
‘To produce a terminated straight line continu-
ously in a straight line,’
The Postulate implies that the produced portion
of the straight line is uniquey or, in other words, that
tuoo straight lines cannot have a segment common to
both.
Attempts were made in ancient as weU as in

modem times to prove this latter proposition.
Proclus and Simplicius attempted it unsuccessfully.
Another ancient proof was based upon Eucl. r, ii;
Simson also used this proposition for the purpose,
but his proof is in reality a fetitio ^rincipii. The
proposition is really required as early as i, i and
I, 4. It is best therefore to mate it a postulate.
POSTULATE 3.
iravrl K€f/rp(o Kal SLaoTTjpxLTLy ‘with every centre
and distance,’ meaning ‘with any centre and dis-
tance’ (cf. Note on Post. i). Sido-Trjpa means
2
POSTULATES 2—4 147
‘distance’ quite generally, as well as distance in tlie

sense of the three dimensions. The word was regu-
larly used, in describing the construction of a circle,
to express the length of the radius, the idea bdng

that all the points on the circumference of the circle
are at the same distance from the centre. The
Greeks had no word for radius', if they had to speak
of radii, they called them ' the (straight lines drawn)

from the centre,’ ai l/c tou Korrpov (sc. dyo/icyat
ypap,ju.at).
KVKkov ypdL<^€crBajL. In this Postulate the verb is

in the passive; ‘that a circle can be (literally ‘is’)
drawn.’ Proclus however has the active aorist

{ypdxj/cu) here also.
POSTULATE 4.
‘That all right angles are equal to one another’

(acc. and infin. after 'Q'n^crOw).
In other words, a right angle is an angle of a

definite size which is always the same wherever it is
placed, is by which
in fact a determinate magnitude
other magnitudes of the same kind (acute and obtuse
angles) can be measured. The statement implies the
invariability of figures. we attempted to prove
If
the Postulate, we should have to use in some form
or other the method of application of one figure to
another; that is, we should have to assume the
invariability of figures as an antecedent postulate.
Euclid preferred to assert directly, as a postulate,
that all right angles are equal; moreover it was
10 —
148 NOTES
absolutely necessary from his point of view to put
it before Post. 5, because the condition in that
Postulate about two angles being together less than
two right angles would be useless unless it were first
made clear that right angles are angles of deter-
minate and invariable size.
As above indicated, the truth of the Postulate

cannot at this stage be proved except by assuming
other postulates in place of it. One method is that
of the modern Italian school, who deduce the fact
that all right angles are equal from the equivalent
fact that aU angles (so-called) are equal, which
is either itself assumed as a postulate or deduced
from other postulates. (A ^Jiaf angle is of course
equal to two right angles, being the ‘angle’ which
two parts A C, CB of one straight line make with one
C
A B
another. If we turn CA about C as centre until it
takes the position of CB, this represents a certain
definite amount of turning; a ‘flat’ angle therefore

represents a definite amount of turning, and a right
angle represents exactly half that amount of turn-
ing.)
POSTULATE 5.
This is the famous Postidate (often inappropri-

ately called the Axmn) of Parallels, It is well known
that the Euclidean theory of parallels cannot be
established except by assuming this or some other
POSTULATES 4, 5 149
equivalent postulate, it is clear from passages of Aris-

totle that in his time the theory of parallels
was not
grounded on a recognised postulate, but contained
some petitio principii. Hence we may conclude that
Euclid’s epoch-making Postulate was first formulated
by Euclid himself; and when we consider the count-
less attempts made through more than twenty cen-
turies to prove the Postulate, many of them by
geometers of great abihty, we cannot but admire

the genius of the man who first definitely concluded
that a hypothesis such as this, which he found
necessary to the validity of his whole system of
geometry, was really indemonstrable.
The long sentence in which the Postulate is stated
consists of the following parts: (i) a conditional
clause €ay...7rot^, (2) the accusative and infinitive
€KpakXofji.€uas, ..(TVfjLTrL'uT^Lv governed by the original

^
rjnqaOiiiy let it be postulated,’ (3) a relative clause
€<!> d ... eXda-a-oves qualifying (jvjjiTruirTeLV.
‘meet,’ and showing in which of the two directions
the two straight lines referred to in the first words

of the Postulate must meet.
The order of words in the conditional clause is
lav evOeia (the subject of the clause) liLTruirrovaa

eh (‘falling on’) Svo ÔeCas (‘two straight lines’)
TTOiy (‘make’) rds evrbs Kal hrl to. avrd ym/iwi
(‘the angles [which are] interior and on the same
side,’ lit. ‘towards the same parts,’ i.e. regions or
direction) âcr<rovas Svo opOihv (‘less than two
right angles’).
150 NOTES
We next come to the accusative and infinitive,
where the words are to be translated in this order:
Tas 8uo €vO€La>s (‘that the two straight lines’) ck-

âXkofjLo/as hr aireipov (‘ [when] produced without
limit,’ cf. €ts oTreipov in Def. 23 and note thereon)
(rvyifiruirreiv (‘meet,’ lit. ‘fall together’).
Lastly, k<j> a pîpnf} = hr\ ra pieprj a, ‘on that
side on which’ or ‘in the direction in which,’ the
whole clause a. . . I^do-o-oves meaning ‘on that
side on which the angles less than two right angles
are.’
The meaning of the Postulate will easily be seen

by means of a figure. AB]z the straight line which
meets two other straight lines CZ), EF. Then the

angles marked x are ‘interior’ angles ‘on (one and)
the same side of’ ABi the angles marked 0 are
‘interior’ angles on the other side of AB; both

these pairs of angles are pairs of angles which AB
‘makes’ with CD, EF'y ‘interior’ angles are angles
‘inside’ (i.e. in the space between CD and EF).
Now suppose that, as in the figure, the sum of the
two angles marked x is less than two right angles.
Then the Postulate asserts that CD, EF wiU meet
POSTULATE 5
on that side of AB on which the two angles marked

X are situated; that is, CZ), EF -vinll meet if produced
far enough on that side, i.e. if produced far enough
beyond D and F respectively.
The Postulate is used in i, 29 as a means of proving
that, iftwo straight lines* (as CD, EF) are parallel,
and AB meets them both, then the sum of the two
angles marked x is equal to two right angles (and so
is the sum of the two angles marked 0).
THE AXIOMS OR COMMON

NOTIONS
AFTER the Postulates come the Axioms or, as
Euclid calls them, ‘Common Notions’ (kolvoI
hn/oicu). These are certain self-evident truths,

which cannot be proved but nevertheless have (as
Aristotle says) to be assumed by any one who is to

learn anything at all. They are not confined to
geometry or to any one science, but are true for all
alike.
Only five of the ‘Common Notions’ appearing in

the MSS. are certainly genuine. Others were inter-
polated after Euclid’s time.
COMMON NOTION I.
TO. Tio avrw L<ra (sc. ovra), ‘Things (which are)

equal to the same (thing)’; xai, ‘also.’
Although Aristotle had emphasised the futility

152 NOTES
of any attempt to prove the axioms, it appears from
Proclns that no less a person than Apollonius of
Perga^ ‘
the great geometer,’ as he was called,
actually made the attempt. Proclus gives as an
example his attempted proof of Axiom i. “^Let A
be equal to B, and B to C ;1 say (says Apollonius)
that A is also equal to C. For since A is equal to
it occupies the same space with iti and, since B is
equal to (7, it occupies the same siâce with it. There-

fore A occupies the same space with C [and is there-
fore equal to it].’ The " proof ’ (it will be observed)
assumes two things (i) that things which ôccupy
the same space’ are equal to one another, and (2)
that things which occupy the same space with one
and the same thing occupy the same space with one
another; in fact Apollonius explains the obvious
by something much more obscure. Moreover the
‘proof’ is partial, and not general, for it is not all
equal things that ‘occupy space’ at all.
COMMON NOTION 2.
TrpoarreOrjy aorist subjunctive passive of irpocr-

TtBhfaL, the regular word for ‘
to add.’ ‘
If equals be
added to equals, the wholes are equal.’
COMMON NOTION 3.
aor. subjunctive passive of aâipeivy
‘to take away,’ ‘to subtract.’

Ttt KaTak€U2r6p.€va, ‘the remainders’ (literally ‘the
things left behind’).
COMMON NOTIONS (AXIOMS) 1-4 153
This axiom, differing from the preceding in that

the equals are subtracted from equals instead of
being added to equals, is the axiom which is so
favourite an illustration with Aristotle.
The MSS. have, after the first three Common
Notions, the following four, the genuineness of
which is more than doubtful.
‘4. If equals be added to unequals, the wholes
are unequal.
5. If equals be subtracted from unequals, the

remainders are unequal.
6. Things which are double of the same thing
are equal to one another.
7. Things which are halves of the same thing are

equal to one another.’
They are really unnecessary and, in view of the
principle that axioms ought not to be multiplied,
should be omitted.
COMMON NOTION 4 [7].

The word c<âp/jLO^€«/, as a geometrical term,
has a different meaning according as it is used in the
active or in the passive. In the passive, itjiapfjLOêa-dai,
it means ‘ to be applied to,’ vdthout any implication
that the applied figure will exactly fit, or coincide
with, the figure to which it is applied. On the other
hand, the active 6^€Lv
i<j>apfji is used intransitively
and means ‘to fit exactly,’ ‘to coincide with.’ In
Euclid and Archimedes €âp/jtd^€tv is constructed
with hrC and the accusative.
154 NOTES
Ta i<j>apfJi6ôvrcL Itt aXXnfjXa, ‘things which coin-
cide with one another’ (sc. when one is applied to,
or laid on top of the other so as, if possible, to fit
all over).
This axiom is distinctly geometrical in character
and therefore hardly proper to be included among

Common Notions. It is in fact a definition of geo-
metrical equality more or less sufficient, but not a
real axiom. This criterion of equality is required
in I, 4, but the axiom is not there referred to;
Euclid says simply ‘The base BC will coincide with
the base EF, and will be equal to it.^ That is to say,
the axiom is tacitly assumed. I am inclined to think
therefore that it is more likely than not to be an
interpolation.
Whoever formulated this Common Notion, it
seems to be intended to assert that superposition is
a legitimate way of proving the equality of two

figures which have the necessary parts respectively
equal, or, in other words, to serve as an axiom oj
congruence.
It seems clear, however, that Euclid disliked the
method of superposition, and avoided it wherever

he could. He probably found the method handed
down by tradition (if Thales proved that a circle is
bisectedby its diameter, he could hardly have done
so by any method other than that of superposition),
and he followed it, in the few cases where he does
so, only because he had not been able to see his way
to a satisfactory alternative.
COMMON NOTIONS (AXIOMS) 4, 5 155
COMMON NOTION 5 [8].
‘The whole is greater than the part.’

In I, 6, where this Common Notion might, if
genuine, have been referred to, Euclid uses a

different expression ; ‘
the triangle DCB will be
equal to the triangle ACBy the less to the greater:
which is absurd? It seems probable therefore that

this axiom too was interpolated. It seems to be an
abstraction or generalisation substituted for an
immediate inference from a particular geometrical
figure, but it takes the form of a sort of definition of
whole and part.
After this Common Notion the editions up to

Heiberg’s generally included the interpolated axiom
‘Two straight lines cannot enclose a space,’ and
after this the Postulates 4 and 5, which were classed
as Axioms ii and 12.
THE PROPOSITIONS
The
first
kinds,
propositions in Euclid’s Elements are of
known
three in Book
as
i
theorems and problems.
are problems, the fourth begins
two
The
a series of theorems. The Book contains in all 14

problems and 34 theorems.
In ancient times there was controversy as to the
proper meaning of the terms, some (e.g. Speusippus
156 NOTES
and Amphinomus) holding that both kinds of pro-
(among whom
position are alike theorems, and others
was Menaechmns, the discoverer of the conic sec-
tions) maintaining that both are alike problems. So

far as etymology is concerned, both views are
possible. $€<afyr}fjLa, theorem^ means that which is the
object of investigation ((9€<op€Lv)y and this would
equally cover a problem; Trpo^Xrjpia, problem, means
something propounded (n-popdXXecv), which again
would cover a theorem. Euclid himself does not
make any distinction except that at the end of
problems he has the words oirep eSei TroLrjcrat, mean-
ing ‘ (which is) what it was required to do ’ (our
O.E.F,, short for quod erat faciendum), and at the
end of theorems the different words oircp ISct
‘(which is) what it was required to prove’ (our

Q,E.D,, short for quod erat demonstrandutri). These
expressions indicate the distinction usually drawn.
The object of a theorem is to demonstrate something,
that of a problem to do or to construct something,
although even in problems a demonstration follows,
which is brought in for the purpose of confirming
the construction by showing that what was required
has actually been done.
In a translation from the Greek into English it is
convenient to use, for the points in the figures,

Roman capital letters instead of Greek, because not
only are the Greek capitals rather more troublesome
to write, but the parts of the figure are rather more
easily grasped by means of the letters which are more
THE PROPOSITIONS 157
familiar to us. The natural equivalents (following

the proper order) are as follows:
ABPAEZHeKAMTSfSOnPSTT^X'PO
=ABCDEFGHKLMNOPORST XJVW XY-,
and, except in quotations from the Greek text, I
shall use the Roman letters in the following notes,
in accordance with this fixed arrangement.
PROPOSITION I.
1 . I. hr\ ‘upon.’
TreTrepaoTfiarr)^ (perf. pass, participle of ir^paJivîv)
means ‘limited’ or ‘terminated’; that is, the given
straight line is the straight line included between
two terminal points. The word is usually translated
‘
finite,’ and there is no objection to this.
Observe that, where we say ‘on a given finite

straight line,’ the Greek usage has the more graphic
expression ‘on the given finite straight line,’ i.e. the
particular finite straight line which we have chosen
to take.
2. orucmô-aa'âi, aorist infinitive middle from
crwiardvaL, ‘ to construct^ (literally ‘ set up together ’),
is the regular word for setting up or constructing a
figure on a given base. It is best translated ‘con-
struct.’ Here we are asked to construct an equi-
lateral triangle on a given finite straight line as base.
3. €crrft)
^ SoO&xra Treir^paxTyiarq ^ AB. The
subject of this sentence is rj Soêtora evOeia xcircpacr-
the proper translation being therefore ‘Let

the given finite straight line be the (straight line)
158 NOTES
Nevertheless it is best to translate as if ^ AB
were the subject ^Let AB be the given finite
straight line.’ The reason will appear when we

come to such sentences as the following (in i, 2):
€<TTQ) TO crrjfjLCLOv TO A, §€ Soêto'a evOcXa
17 Br, literally ^Let the given point be A and the

given straight line BC,^ which is awkward in con-
sequence of the omission of the verb in the second
clause. It is better therefore to say ‘Let A be the
given point and BC the given straight line.’
4. Set ‘Thus it is required,’ introduces the

more specific statement of what it is required to do,
with reference to the particular data (here the given
straight line AB).
6. ‘With centre A and distance AB^ follows
the phraseology of Post. 3 about drawing a circle.
kvkXo^ yeypdfpdo) o BP A, literally ‘let a circle have
been described, (namely) the (circle) BCDJ I have
already (note on Postulates, p. 142) called attention
to this elegant and expressive use of the perfect im-
perative passive. In English there is no practicable
alternative to translating the present phrase thus,
‘let the circle BCD be described.’
9. aTTO Tov r (rrjfielovy KaO’ o rifivovcnv aXXi^Xovs
OL kvkXoi, ‘from the point C in (lit. at) which the
circles cut one another.’ It ought to be proved
that the two circles do in fact cut one another;
Euclid however assumes this without proof. It is
true that we can without difficulty satisfy ourselves
that the circles do meet, and in two points, but

1 '
PROPOSITION I 159
only by means of some postulate other than those

formulated by Euclid. There is therefore here a
real defect in Euclid’s exposition.
9-1 . aTTo rov T cr7)fL€LOv . . . i'ul ra A, B criqfxâ. hre-
êv)( 0 (jicrav cvOeiai al TA, TB, ‘from the point C...

let the straight lines CA, CB [literally ‘ straight lines
(namely) CAj be joined to the points A, B^;
a longer and more careful expression for drawing
the straight line joining two points, which gives way

later (i, 5 and afterwards) to the shortened ex-
pression ‘let the straight line FC be joined’ (hr-
€.^€oxOisi 'q Zr), the regular form in Greek geometry.
hr£^€6y(0<ii€rcLv, perfect imperative passive, as usual
(i'TTLêuyvvvai) .
12. /cat= ‘Then,’ introduces the proof of the

correctness of the construction.
16. eKaripa twv TA, TB. In English we have
to insert the words ‘straight lines’: ‘each of the
straight lines CA^ CB.^ So (18) al rpeL<s al FA,
AB, BT = ‘ the three straight lines CA, AB^ BCJ*
apa is the particle generally used in demonstra-
tions where we say ‘Therefore.’ The force of it is
‘
So, then, (we may infer that).’
This first proposition gives a convenient oppor-
tunity of distinguishing the recognised formal
divisions of a proposition in Greek geometry.
Though these are best illustrated from Euclid’s
propositions, they did not originate with him: on
the contrary, the form which his propositions took
was already traditional. Proclus, in his Commentary
i6o NOTES
on Book j, observes that every problem and every
theorem which is complete, with all its parts perfect,
purports to contain the following constituent parts.
(1) The TTporacTLs, ‘proposition/ i.e. ênunciation^
which states the problem to be solved, or the

theorem to be proved, in general terms. Cf. the
enunciation of this proposition ‘On a given finite
straight line to construct an equilateral triangle.’
(2) The €fc^€crts, ^setting-out^ (iKTiOevai). This

sets out the data in a concrete form ready for use
in the subsequent stages. In i, i the ‘setting-out’
consists of the words ‘Let AB be the given finite
straight line.’ The straight line AB is the one

datum in this case.
(3) The BLopLCTfjLOSi literally the ‘delimitation’

or ‘definition’ (but not definition in the ordinary
sense, the Greek word for which is opos). By this
is meant the specification or particular statement of
what the theorem or problem requires us to prove

or to construct, not in general terms (as in the
enunciation) but with reference to the particular
concrete data which axe shown in a diagram; the
object of the BiopicrpLo^ is then to fix the attention
better, by visualisation, as it were. In i, i, it consists
of the words ‘Thus it is required to construct an

equilateral triangle on the straight line AB.^ It is
regularly introduced by the words Set S77, ‘

Thus it
is required.’
(4) The KwraaKeuy, apparatus or construction. In

the particular case of i, i the construction consists of
DIVISIONS OF A PROPOSITION i6i
the drawing of the two circles and of two lines

joining pairs of points.
(5) The oLTroBeL^Ls, proof or demonstration.

(6) The <rvfjLTr 4pacrfjia, conclusion^ which reverts to
the enunciation and shows that the problem has
been solved or the theorem proved. In i, i it con-
sists of the words ‘Therefore the triangle ABC is
equilateral, and it has been constructed on the given

finite straight line AB^ and in other problems it
takes a similar form. In theorems the conclusion

generally repeats the enunciation word for word.
The conclusion is immediately followed in the case
of problems by the words OTrep cSct -n-ot^crat,
‘(which is) was required to do/ quod erat

what it
faciendum (O.B.F.), and in the case of theorems by

the words o^rcp ISci Seiiat, ‘(which is) what it was
required to prove/ quod erat demonstrandum {O.E.D.').
In particular cases some of the above divisions
are not found; but all propositions contain at least

three of them, the enunciation^ the proof and the
conclusion.
The word Siopicrpos (cf. 3 above) has another and
a more important sense in relation to a problem.
Problems may be possible or impossible of solution
according to the nature and relations of the data.
This is best explained by an example. We learn
from Eucl. I, 20 that any two sides of a triangle are
together greater than the third. Hence, when we

are asked to construct a triangle with sides equal to
three given straight lines respectively (i, 22), it is
H. E. II
i 6z NOTES
necessary to be satisfied beforeband that, with the
given data, a solution is possible or, in other words,
that the three given straight lines are such as £:an
form a triangle; and by i, 20 this can only be when

any two of the straight lines are together greater
than the third. Accordingly Euclid adds to the
enunciation of i, 22 the words ‘Thus it is necessary
(Set Sy) that tw'o of the straight lines taken together
in any manner should be greater than the remaining
straight line.’ This is a formal statement of the
necessary condition which must be fulfilled if the
solution of the problem is to be possible; and this
is what the Greeks meant by Stoptor/Aos. Stopt^ctv
= St -j- opLêiv, literally ‘to set boundaries (opos)
between,’ i.e. to make a demarcation between the
cases in which the solution is possible and those in
which it is not. It is difficult to find an English
word that is a proper equivalent. If we say de-
finition we have to remember that it is a different
thing from a definition in the ordinary sense; ‘de-
limitation’ might perhaps serve, but it is probably
better to abandon the attempt to be literal and to
say ‘limiting condition.’
PROPOSITION 2.
I. xpos t<3 (rrjpL€L(o, ‘at the given point,’

in accordance with the Greek usage, whereas we say
‘at a given point’ and should so translate,
2. ^€crât,‘toplace’(aor,inf.mid.o£Tt^ât). To
place a straight line at a point means to place it so
—
PROPOSITION 2 163
that one of its extremities is at the point. We may,

in order to make this clear, translate ‘
To place with
its extremity at a given point a straight line equal
to a given straight line.’
7. crwcoTaTO), perf. imperative of crvvLtrrdvat, with

passive sense, ‘let the equilateral triangle DAB be
constructed’ (lit. ‘have been constructed’).
8. €Kj8€jSX‘)7Gr0<rt<rav hr eu^ctas ratg AA, AB
€v 6 ^(ll al AE, BZ, ‘let the straight lines AE, BF be
produced in a straight line with DA^ DB.^ Observe
that the straight lines here said to be ‘produced’
are not the original straight lines DA, DB but the
produced portions of them. Euclid has this longer
description of the operation of producing a straight

line in i,
5 also; it is not till i, 16 that we find him
using the shorter form familiar to us, ‘let BC be
produced to 2).’
16. Sviq AA AB ten; ia-riv. It is better to begin

a new sentence here and to combine this clause with
the next in one sentence. ‘And in these (wv) DA
is equal to DB’, therefore the remainder AL is equal
to the remainder BG,^ The expression tJ AA
means Hterally ÂL remaining’ or ‘which remains’;
but it is hardly possible to say in English '’AL which
remains is equal to BG which remains’; it better is
to translate Xonr^ y AA by ‘the remainder AL^:

‘the remainder AL is equal to the remainder BG.^
The reason for the insertion of this proposition is
that Euclid does not regard it as legitimate to
suppose that we can take up a straight line of given
II 2 .
164 NOTES
length from one position and place it in another
(e.g. with its extremity at a given point). He would
regard it as ungeometrical to carry a distance by
means of a ruler with marks on it showing the dis-
tance, or by means of an opened pair of compasses.

As De Morgan says, we are not allowed to draw a
circle with a compass-carried distance; it is as if we

had to suppose that the compasses would close of
themselves the moment they cease to touch the

paper.
There are several possible cases of this proposition.
The relative positions of the point and the straight
line may vary, and we may describe the equilateral
triangle DAB on either side of AB, According to
these differences the figure of the proposition takes
slightly different forms; the proof is however,
mutatis mutandis, the same. Two other figures are
added to show this.
PROPOSITIONS 3 165
differences between AD^ DL and BD, DG instead

of tbe differences between DL, AD and DG, BD
respectively; while in tbe other figure AL and BG
are the sums instead of the differences of DL, AD and
DGy BD respectively. The pupil can work out the
proof in these or other cases for himself.
The Greek commentators revelled in distinctions
of cases ; the classical Greek geometers, and notably
Euclid, were in the habit of giving one case only, for
choice the most difficult, leaving the reader to think
out the rest for himself. Where there was a real
disparity between the cases, sufficient to involve a
substantial difference in the proofs, the practice was

to give separate enunciations and proofs altogether.
This is seen in the Conics and other treatises of
Apollonius.
PROPOSITION 3,
I. We may, if we please, translate Svo Soêicriov

evOeiiov avicrfov as if it were a genitive absolute,
‘given two unequal straight fines,’ although the
genitive may equally be taken as the partitive geni-
tive after anro ‘from the greater of
two given straight lines.’
2. The order of the words is evOelav unqv

ry iXdcrorovt, ‘to cut off a straight line equal to the
less.’
7. Keia-Ofo, KeLo-Oai is here, as usual, employed as
the passive of riOiyaL, ‘let (AD) he flaced^ (cf.
Prop 2).
i66 NOTES
PROPOSITION 4.
2. eKaripav eKCLTepa, The traditional translation

is êach to each,’ which however is unsatisfactory
because it might be inferred from it that all the four
sides mentioned (two in each triangle) are equal,
whereas what is meant is that one of the two sides
in the first triangle is equal to one of the two sides

in the second triangle, and that the second of the
two sides in the first triangle is equal to the second
of the two sides in the second. Translate by ‘re-
spectively’ here and always. ‘If two triangles have

two sides equal to two sides respectively,’ i.e. if two
sides in the one triangle be equal respectively to two
sides in the other. I have translated ras Sw TrXeupas
by ‘two sides’ (without the definite article) because
our English practice substitutes the indefinite article
for the definite in such a case.
2-4. ymviav rqv rnro T(ov tcriov evOcLKoy TrepL€)(0~
pihrqv, ‘the angle contained by the equal straight

lines.’ This means the angle included in the one tri-
angle by the two sides which are respectively equal to
the two sides in the other. With tq yusvia. we should,

strictly speaking, understand a similar expression
rp vTTo T(ov tarwv evOcmv '3r€pt€ô/xciqy. We may trans-
late ‘and have the angles contained by the equal
straight lines equal.’ Euclid no doubt says ‘straight
lines’ rather than ‘sides’ here in order to follow

scrupulously at the outset the language of Def. 9.
4. Kttl T^v pdcriv. The apodosis begins here.
‘They will also have the base equal to the base.’
PROPOSITION 4 167
7, 8. '
as at to-at TrXevpat inroreLvovaLV : as
qualifies at konral ywtat. worctvftv means literally
‘stretch underneath,’ and with wd and acc. it means
‘subtend ’ in the sense of ‘ being opposite to.’ ‘
The
remaining angles wiU be equal to the remaining
angles respectively, namely those which the equal
sides subtend.’ Cf. 17 xmoreLvovara, the regular word
for the hypotenuse of a right-angled triangle, which
subtends a particular angle in the right-angled
triangle, namely the right angle.
12. 77 -uTToBAr y<oi/La, the regular Greek expression
(with or without ym/Cd) for ‘the angle BAC.^ The
full expression w'ould be BA, AP Trcpt-
77 viro tS>v
exp/xhnfj ycovta, ‘
the angle contained by the (straight
lines) BAy AC.^ But it was a common practice of
Greek geometers, e.g. Archimedes and Apollonius
(though not apparently of Euclid himself), to write
at BAP for at BA, AP, ‘the straight lines BA, ACJ
Thus, when Trcptexofianj was dropped, the expression
for the angle BAC became, first vtto rwv BAP 77
ytui/ta, then 77 wd BAP ywvta, and finally xytro 77
BAP, without ycDvta, as regularly found in Euclid.

13. Observe that the ‘particular statement’ of
the proposition is, in the case of theorems, intro-
duced by the word Xeyia ‘I say’ (that. . .).
19 . €âppLOôpL€vov Tov TpLywvov, ‘if the triangle

be applied (to)’ : this is the regular meaning of the
passive l<^sappu6^€cr6aL.
21. ifjyapfxoar^L, ‘wiU coincide (with)’ or ‘exactly

fit’; this is the meaning of the active forms of l<âp-
:
i6S NOTES
fi 6^€iy, which are used intransitively (cf. note on
Common Notion 4 above, p. 153). This distinction
between the active and the passive of the verb
€<pap/Aof€tv is well brought out in this proposition.

28. ‘
Hence the base BC will coincide with the base
EFy (33) and will be equal to it/ The coincidence of
BC with EF is immediately inferred from the fact
that the ends of the two straight lines coincide re-
spectively. Euclid uses the fact, implied in Post, i,
that there is only one straight line joining two

points. Some commentator, not deeming the argu-
ment suiEciently clear, inserted an explanation in
support of the inference that BC coincides with EF
(29-33) For if, w’hen
‘ B coincides with E and C with
Fy the base BC does not coincide with the base EF,
two straight lines will enclose a space: which is im-
possible. Therefore the base BC will coincide with
ef; introducing in this way the interpolated and
unnecessan/- axiom that ‘two straight lines cannot
enclose a space.’
39-45. Where Euclid, in his conclusion, simply (as

here) repeats the enunciation word for word, we can
simply write ‘Therefore etc.’
PROPOSITION 5.
I. ai Trpos ry yioviaL, ‘the angles at the

meaning ‘adjacent to the base,’ i.e. the angles
base,’
formed with the base by the tw'o equal sides re-

spectively.
2. rrpocr€Kj3Xr}0€Lcr<ov t<ov Lir<ov evOemv, genitive
PROPOSITIONS 4, 5 169
absolute, ‘when’ or ‘i£ the equal straight Hues are

’
produced further ‘ in addition the ‘equal
straight lines’ meaning the equal sides.
10. €Lki]<liO<Ji yap i'JTi BA rv)(ov cnjpLCLOv to Z.

Tvxov is the aorist participle of Tvyxavew^ so that
rvxpv arrfp^Lov is ‘any point as it may happen,^ ‘a
chance point’: ‘For let a point F be taken at
random on BD.'
15. ytaviav koiv^v Trepiôvcri rrjv vtto ZAH, ‘they
(i.e. the two sides in the two triangles FAC^ GAB
respectively) contain a common angle, the angle
FAG 7 KOLv^v means ‘ common to both triangles,’ and

its force is here predicative, as if we were to say ‘the
angle FAG which they contain is common (to both
triangles),’ and therefore equal in both triangles.
26. /Bd(rLS avrSiv. avroiv here evidently means the

angles just mentioned: ‘the base BC is common to
them.’The triangles BFC, CGB again, having a
common base, have their bases equal.
3 5. €to'l TTpos ry jSdcret, ‘ are at the base.’ The sub-

ject is the angles ABC, ACB. The angles ABC,
ACB ‘are (the angles) at the base.’
This, being the first proposition which has to be
closely argued, is no doubt difficult for a beginner:
hence the name given to it, pms asinorum. The
points to remember are that the proposition i, 4
has to be applied, Jirri to the large triangles FAC,
GAB, and secondly to the small triangles beyond the
namely FBC, GCB. The second result of the
base,
proposition, the equality of the angles beyond or
170 NOTES
under tJie base, follows from the equality in all
respects of the latter pair of triangles. And, lastly,
the equality of the angles at the base is proved by

subtracting equals from equals (namely equal angles
in the triangles FBC^ GCB from equal angles in the
triangles FAC, GAB),
It is to be observed that the pairs of triangles as
they appear in the figure are not superfosalle as the
triangles in i,
4 are: they are symmetrically equal
rather than equal in the sense of i, 4, though they
would be equal in that sense if one of the two were
regarded as turned over or seen from the back.
A much shorter proof of the theorem of i,
5 was
given by Pappus. It requires no ‘construction’: it
suffices simply to draw the triangle ABC,

‘Let us,’ says Pappus, ‘conceive the one triangle
\ABC] as two [i.e. as two triangles ABC
triangles
and A CB respectively] and let us argue in this way.

Since AB equal to AC, and AC to AB,
is
the two sides AB, AC are equal to the two sides

AC, AB.
And the angle BAC is equal to the angle CAB,
for it is the same.
Therefore all the corresponding parts (in the
triangles) are equal, namely
BC to CB,
the triangle ABC to the triangle ACB,
and the angle A CB to the angle ABC
(for these are the angles subtended by the equal
sides AB, AC).
:
Therefore in isosceles triangles the angles at the

base are equal.’
If there is any difficulty in visualising the one
triangle as two, we may regard the second triangle
ACB as the first triangle ABC seen from the hack.
We are told by Proclus that the first to discover
the truth of this proposition w’as Thales, who how-
ever spoke of the equal angles as ‘
similar ’ angles.
Aristotle refers to a proof of the theorem of i,
5
which is quite different from Euclid’s. Aristotle no
doubt found this proof in the text-book of Elements
in use in his time, which was probably that of
Theudius of Magnesia. We infer that Euclid’s proof
was his owm, and in fact that he arranged his book
from the very beginning on a plan altogether
different from that followed by his predecessors.
PROPOSITION 6.
2- al VTTO ras teas y(0VLas viroreLvovcraL TrXeupal

as usual, ‘
the sides subtending [i.e. opposite to] the
equal angles.’
15. ‘The triangle DBC will be equal to the
triangle the less to the greaten which is absurd.’
Euclid assumes that, because D is between A and B^
the triangle DBC is less than the triangle ABC. To
prove this theoretically would require some postu-
late; but Euclid is satisfied to make the inference
directly from the look of the figure.
As regards this proposition we have to observe

(i) that it is what is called the converse (diArt<7Tpo</>os)
172 NOTES
of the preceding Prop. 5, and (2) that the method of
proof b}r reductio ad ahsurdum is here used by Euclid
for the first time.
(1) Conversion (in Greek avTLcrrpo^T^y ^turning

the opposite way,’ from avrC and <jTpi<j>€Lv) may be
either complete or partial. This proposition is a case
of complete conversion, conversion, that is, where the
hypothesis and the conclusion of a theorem change
places exactly. The hypothesis of i, 5, the equality
of the two sides, is the conclusion of i, 6, and the
conclusion of i, 5, the equality of the two angles, is
the hypothesis of i, 6. We shall find a case of partial

conversion when we come to i, 8, which is the
partial converse of i, 4.
(2) The method of proof by reductio ad ahsurdum

is not peculiar to geometry. It is described by
Aristotle as a logical method of proof (Analytica

priora. Book i); he calls it alternatively cts to
dSvvaTov aorayioyjj, ‘reduction to the impossible,’
or y Sta Tov aJSvvdrov or dxdScifts, ‘proof by
the impossible,’ or y eh to dSvvarov dyovaa dnro-
êt^vs, ‘proof leading to the impossible.’
‘Proof leading to the impossible,’ says Aristotle,

‘differs from the direct (SeiKriKy^, demonstrative)
in that it assumes what it desires to destroy [namely

the hypothesis that the proposition which we seek to
prove is false] , and then reduces it to something which
is admittedly false.’ If we arrive at a conclusion ad-
mittedly false, this shows that our hypothesis was
false, and therefore that the original proposition
which we desired to prove, and which that hypo-

thesis contradicted, cannot but be true. So in this
proposition, where we have to prove that the sides
are equal, we begin by assuming that they are un-

equal and show that this hypothesis leads to a con-
clusion which is obviously false. It follows that the
hypothesis is false, that is, that the sides are not un-
equal as we supposed, and therefore must be equal.
Produs also has a good description of reductio ad
absurdumi ‘Every reductio ad ahsurdum^ he says,
‘assumes what conflicts with the desired result,
then, using that as a basis, proceeds until it arrives at
an admitted absurdity and, by thus destroying the
hypothesis, establishes the result originally desired.’
PROPOSITION 7.
1-5. The enunciation of this proposition is very

difficult to render intelligibly, because the wording is
extremely condensed. The fact is that the phraseo-
logy is traditional and would convey more to the
contemporary Greek than is actually contained in
the words; for we find the identical phrase ov
(rvaraOTjorovTaL irpos oAAo Kal oAAo cnyftctov used
with the same meaning in Aristotle.
Just as avoTTâ-aorOcLL hrl is used of constructing a
triangle on a given base (cf. i, i), so it is used of
setting up or constructing two straight lines from
the ends of a straight line so as to meet in a point.
Let us now analyse the sentence, hri rrjs ttvrrj^
evOeCas means ‘on the same straight line [i.e. a given

174 NOTES
straight line] (as base).’ The subject of the sentence
is aXXxLL Svo evOetoLL, ‘ two other straight lines,’ and the
words qualifying it are l<tcu ovo tols avralî evOeiais
l/carcpa iKarepa^ ‘equal to the same two straight lines
(i.e. to two given straight lines) respectively.’ ov <rv-
(rraSricrovTaij ‘ cannot be constructed ’ (literally ‘ shall

not be constructed’), Trpos ^Xo) /cat oAXw (rrjfjLecio,
‘to different points’ (literally ‘to another and an-

other point ’), the meaning being that ‘ the same two
straight lines’ are constructed to one point and the
‘two other straight lines’ are supposed to be con-
structed to another point; hr! ra avra fJieprjy ‘on the
same side’ (sc. of the given base). The last expres-
sion ra avra Trepara tôvcrai rat? ^vOeCaus
qualifies aXk.ajL Svo evOeiaL as predicate and means

‘having (or ‘if they have’) the same extremities
respectively as the original straight lines’; i.e. the
lines which are supposed to be respectively equal are
those starting from the same end of the given base.
The nearest possible approach to. a literal trans-
lation serving to convey the full meaning seems to
be the following.
‘Given two straight lines [constructed on a given
straight line from its extremities and meeting at a
point], there cannot be constructed on the same
straight line two other straight lines meeting in
another point on the same side [of it] and equal to
the former two respectively, namely each to that
which has the same extremity with it.’
Simson altered the wording to the following
5
PROPOSITION 7 175
wHch expresses the same thing more simply: ‘On

the same base and on the same side of it there cannot
be two triangles that have their sides which are
terminated at one extremity of the base equal to
one another and likewise those terminated at the
other extremity.’
The whole thing is made intelligible by therîtog-
mty with reference to the figure of the proposition.
1 . fLCL^dyuapa 7 VTTo A AT
}
wo ATB, therefore ‘
the angle ADC is greater than the angle DCB.^ This

inference would be more obvious if, immediately
before it, we inserted, as an intermediate step, ‘But
the angle ACD is greater than the angle DCJBJ

16. TToXkQ
fji€L^(av3 ‘much greater’ (Hterally
by much’), the usual expression in Greek
‘greater
geometry for a fortiori greater.’
^
There are two possible cases of this proposition.

Euclid, in accordance with the usual habit of the
classical Greek geometers, only gives one case, leav-
ing the other to be worked out by the pupil. It is
supplied by Proclus, and is shown in the accompany-

ing figure. In this case Z), instead of being on the
side of CB ocwzj from A^ is inside the triangle.
176 NOTES
We join CD and produce AC^ AD to F.
Then, since AC^ÂD^ the triangle ACD is
isosceles, and therefore [second part of i,

5] the
angles ECD^ FDC^ under the base, are equal.
Therefore the angle FDC, being equal to the
angle ECD, is greater than the angle BCD;
therefore the angle BDC [which is greater than
the angle FDC\ is much greater than the angle
BCD.
But since, in the triangle BCD^ BC = BD (by-
hypothesis), the angle BDC is equal to the angle

BCD. [i, 5]
But it was proved much greater: which is im-
possible.
Prop. 7 is, actually, only used by Euclid for the

purpose of proving the following proposition, i, 8.
It has however an independent value of its own, as
is easily seen. For it proves that, if, in addition to

the base of a triangle, the length of the side termin-
ated at each extremity of the base is given, only one
triangle satisfying these conditions can be con-
structed on one and the same side of the given base.
It shows therefore that the constructions of i, I and
I, zz give only one triangle on one and the same side
of the base.
PROPOSITION 8,
12, 14 &c. i<l>CLpiJ, 6^€cr0aL hrl, ‘to be applied

to^; i<l>apfJi 6^€Lv hri, ‘to coincide with,’ as usual.
20. irapaXXdôvcriVy ‘will fall beside them’;

irapaWa/rreLV {irapd and dXXarrciv, ‘to change’)

means ‘to pass by without touching,’ ‘to miss,’ or
‘to fall awry.’
24. ov (TWLcrravTai 8e, ‘but they cannot be so

constructed’ [Prop. 7].
Prop. '8 is, as already mentioned, a partial con-
verse of Prop. 4. The meaning oi partial conversion
will be clear from this example. The enunciation
of Prop. 4 contains two hypotheses, (i) that two

two triangles are respectively equal, (2) that
sides in
the included angles are equal, and the conclusion

proved is (3) that the bases are equal. Prop. 8, the
partial converse, takes as its two hypotheses one of
the hypotheses of Prop. 4, namely (i), and the con-
clusion of that proposition (3), and proves as the
conclusion the second hypothesis of Prop. 4, namely
that the angles included by the equal sides are equal.
More generally, the partial converse takes as hypo-
theses the conclusion and all the hypotheses but one
of the original theorem, and deduces, as the con-
clusion, the remaining hypothesis of the original
theorem.
An ancient alternative proof of i, 8, which avoids
the use of i, 7, is due to Philo of Alexandria. It has
the disadvantage that it is necessary to consider
three cases.
Let ABC^ DEF be two triangles in which the
ABy AC are equal respectively to the two sides
sides
DE^ DF, and the base BC is equal to the base

178 NOTES
Let the triangle ABC be taken up and placed so
that BC coincides with EF but the point A falls at
A D
G on the opposite side of EF from D (i.e. the tri-
angle ABC is turned- over as well as moved to a new

position).
Then either DF^ FG are in a straight line or they
are not. If they are not, join Z)G.
T>G may in the latter case either meet EF between
2
E and F or may meet EF produced in either direc-

tion.
There are therefore three cases as shown in the
figures.
In the first case, FG are in a straight

where Z)F,
line, we have DEG in which the sides DE,
a triangle
EG are equal. Therefore (by 5) the angle EDG

i, is
equal to the angle EGD.

In the second and third cases we have two tri-
angles DEGy DFG which are isosceles, since, by

hypothesis, DE is equal to EGy and DF to GF.
Therefore, in the triangle DEGy the angles EDG,
EGD are equal, and, in the triangle DFG, the angles
FDG, FGD are equal. Therefore in the second
figure the sum of the angles EDG, FDG is equal to
the sum of the angles EGD, FGD, so that the whole
angle EDF is equal to the whole angle EGF. In the
third figure the difference between the angles EDG,
FDG is equal to the difference between the angles
EGD, FGD', that is, the angle EDF is again equal
to the angle EGF.
Therefore in all the cases the angle EDF is equal
to the angle EGF, which is, by hypothesis, equal to
the angle BAG.
PROPOSITION 9.
I. ^To bisect a given rectilineal angle.’

r€/L€iv (aor. inf. of Tcfivciv), literally ^to cut in two,’
means to cut into two equal parts, i.e. to bisect
(cf. Def. 17).
12—
i8o NOTES
4. 'n;;j(ov a-rjiJL^tov, ‘a chance point,’ as usual:
^let a point D be taken at random on JB’
8. Slxol rirfLTjTaiy ‘ has been bisected.’
Observe that Euclid describes the equilateral
triangle on the side of DE opposite to although
he does not note the fact. In general it would make
no difference on which side the triangle is drawn;
but, if it is drawn on the side of DE towards the

construction would fail in one particular case,
namely that in which the given angle is equal to the

angle of an equilateral triangle; for then F would
coincide with and there would be nothing to
show in which direction a straight line should be
drawn through A in order to bisect the angle
DAE.
PROPOSITION 10.
I. ‘To bisect a given finite straight line,’ i.e. a
straight line terminated at two given points, as dis-

tinct from an unlimited straight line.
Apollonius, we are told, bisected a straight line

by a construction like that of i, i. With centres
A.^ B and radii AB, BA respectively describe circles
PROPOSITIONS 9_n i8i
intersecting at C, D, Join CD meeting AB in E,

ABi% then bisected at E.
To prove this we join AC, CB, BD, DA and then

prove (i) that the triangles ACD, BCD are equal in
all respects (this follows from i, 8), and therefore

that the angle A CD is BCD, and
equal to the angle
then (2) that the triangles ACE, BCE are equal in all
respects (this follows from i, and therefore that
4),
AEis equal to EB.
It will be observed that Apollonius’s method is
equivalent to that in use in 'practical geometry.
PROPOSITION II.
I. BoOeLorj evOeta, dative after irpos opOa^

'ytavLas; ât right angles to a given straight line.’
aaro tov Trpos airry (r7]p.eCov, ‘from a given
point on it,’ i.e. on the given straight line.
9. K€La 6 <a TA Lcnq y PE, ‘let CE he made equal

to CZ).’ As usual, kô-Ool is used for the passive of
TLO&aL, which means ‘to place’ but is often used in
the sense of making (e.g. making one straight line
equal to another straight line).
18, 19. c^€^s, ‘adjacent,’ as in Def. 10.

iSz NOTES
If we regard CD as making with its continuation
CE what is often called a angle/ the construc-
tion of this proposition in effect bisects the flat
angle by means of the straight line CF, The con-
struction therefore is really a particular case of that
in I, 9. The equal halves of the ‘flat angle/ namely

the angles DCF, FCE are right angles in accordance
with Def. 10, and each of the right angles is seen to
be half a ‘flat angle/
If it is diJB&cult .to conceive of the ‘
flat angle’ as
an angle, the idea of the size of this angle can be
DCF
got by reflecting that, if CD be turned about C as
centre until it coincides with CE, the flat angle

represents twice the amount of turning that the
right angle DCF represents; for, in order to arrive
at CE, CD has to describe successively the two right
angles DCF, FCE.
PROPOSITION 12.
I. hr\ 7t)v hoOucrav eu^ctctv aveipov, ‘to (lit.
‘upon’) a given unlimited straight line,’ after

KaOerov . . . ayay€LV.
2. o iirj icmv hr avrijs, ‘which is not on it’ (sc.
the given straight line).

PROPOSITIONS II, 12 1S3
3. KaOerov evOeiav ypafjLfjn^v^ ‘a perpendicular

straight line.’ This is the full expression for a ^perpen-
dicular.’ Ka^CTOS (KaOtevaLt) means ‘let faU’ or ‘let
down,’ ‘ dropped,’ so that the expression corresponds
to OUT plumb-line, rj KaOeroq is constantly used alone
for a perpendicular, evdeia ypa/x-pTy being understood.
9. CTTt TO. €T€pcL pîpnfj Tijs AB evO€LCL<s, ‘on the other

side of the straight line (lit. ‘towards the
other parts of the straight line AB^), Cf. hri ra
avra. ‘on the same side’ in Post. 5 and Prop. 7;
piipTj,
eKorepa ra piiprj, ‘in both directions’ (Def. 23).

D is taken at random m the side of AB away from
C in order that it may be certain that the circle
drawn with C as centre and CD as radius wiU really
cut AB in two points. Just as in i, i, however, we
need some postulate to assure us of the fact.
Proclus tells us that this problem was first in-

vestigated by Oenopides of Chios (fifth century
B.C.), who thought it useful for astronomy. Oeno-
pides called the perpendicular Kara yvco/x-om, ‘(a
straight line drawn) in the manner of a gnomon’
(or ‘ gnomon-wise ’). The gnomon was (in its original
meaning) a stick or needle placed in a vertical

position for the purpose of casting shadows and so
serving as a means of measuring time; hence the use

of the term for a perpendicular. It is not to be
supposed that Oenopides was the first to draw per-
pendiculars in geometrical figures. What is meant
is doubtless that Oenopides was the first to give the
present theoretical construction; previously, no
184 NOTES
doubt, perpendiculars would be drawn by means of
a set-square or other equivalent mechanical means.
PROPOSITION 13.
I. (rradelora (aor. pass, participle of tcrramt),
‘setup ‘if a straight line set up on a straight line
make angles ’ (ywvtas Troiy), i.e. if the first straight
line makes (two) angles with the second straight line

on which it is set up. ‘It will make either two right
angles’ (i.e. each of the two angles will be a right
angle) ‘or angles equal to two right angles’ (i.e. it
will make unequal angles, the sum of which however

will be equal to two right angles): with 6p6as and
Laas understand ytovias.
II. KOLvy 7rpocrK€L(rB(ii'^vvo'EBA, As usual, TTpoc-
KCLcrOaL is used in place of the passive of xpocm^ât,
‘to add.’ The phrase is very characteristic in Greek
geometry, but a literal translation is hardly practic-
able. The meaning is ^let the angle EBD he added
(so as to he) common^ (to both equals). ‘Let the
common angle EBD be added’ would clearly be
an inaccurate rendering, since the angle is not
common before it is added ; the kolv^ is proleptic. In
the corresponding expression for subtraction, kolv^
d<f>rjp7jcr6(a (dâtpctv), ‘let the common angle be
subtracted’ would be less unsatisfactory as a trans-
lation; but, as it is desirable to use corresponding
words when translating the two expressions, it is
probably best to say ‘

let the angle EBD be added
to each^ (or ‘sub tract ed/row each^), though we may.
PROPOSITIONS 13-15 185
if we like, add ‘(in) common’ in order to render

KOLV^.
PROPOSITION 14.
1. In the conditional sentence Bvo cttl
TO. avra fi.ip’iQ KCLpLcvai is the subject and goes with

'iroiwai Tws yoivla^ Svalv SpOats tcra^Sy ‘make
the adjacent angles equal to two right angles.’
Observe the two meanings of -Trpd? in the enuncia-
tion. 'irpos TLVL evOeia— ^with any straight line,’
which is constructed with iroiGxTL ya>i/ms, ‘make

angles with^'^ koI (irpos;) Trpos avry ‘and
at a point on it’ (sc. the straight line).
2. pJfi i'lrl Tct aura fiipr) KeCpEvcu, qualifying Svo

€vO€LaL, ‘two straight lines not lying on the same
side’ (of the straight line first mentioned).
The whole enunciation is: ‘If with any straight

line, and at a point on it, two straight lines not
lying on the same side make the adjacent angles (to-
gether) equal to two right angles, the two straight
lines will be in a straight line with one another.’
1 9. o/xoio)5 Brj hdôp.€Vy ‘ similarly vre can prove
’
—
(literally ‘we shaU prove,’ i.e. if we please).
PROPOSITION 15.
2. Kopv<l>’iQ, vertex, so that Kara Kopvcjyi^v means

‘byway of vertex’ or ‘vertically related.’ We may
translate at Kara Kopv<f>^v yiavLOL (the angles which
are ‘by way of vertex,’ sc. to one another) by ‘the
vertical angles,’ by which are meant the vertically
opposite angles.
i86 NOTES
This theorem, according to Eudemus, was first
discovered hy Thales but found its scientific de-

monstration in Euclid.
In one ms. and in the margin of another (not the
best) a corollary is added. The Greek word for what
we call ‘corollary’ is rropicrfm, meaning something
‘provided’ or ‘found’ (yropL^fo). In the Elements the
word (which we may translate by ‘Porism’) always
has the meaning of corollary, which is an incidental
result springing from the proof of a theorem or the
solution of a problem, a result not directly sought
but appearing, as it were, by chance without any
additional labour and constituting, as Proclus says,
a sort of windfall (eppxiLov) or bonus (KcpSos).
The text of the Porism is Tl/c Bjj toutou <l>avepbv
oTt, iav Svo evOcLai ripiviocriv aXkTjka^, ras 7rpo9 ry
ytavCas rerpaariv opBcus tcra? rrouqa'ovciVy ‘From
this it is manifest that, if two straight lines cut one
another, they will make the angles at the (point oQ
section equal to four right angles.’ ‘
section,’
is regularly used of the ôint of section of two lines
and of the line of section of two surfaces.
Proclus adds notes (i) that it follows from the
Porism that, if any number of straight lines intersect
one another at one point, the sum of aU the angles
soformed about the point will still be equal to four
right angles (this corresponds to what is usually
given in the text-books as Cor. 2), and (2) that the
Porism forms the basis of the theorem which proves
that only the following three (regular) polygons can
fill up the whole space surrounding one point,

namely the equilateral triangle, the square, and the
equilateral and equiangular hexagon, a theorem
which, he says, was due to the Pythagoreans.
PROPOSITION 16.
I. TravTos Tpiyuivov. The Greek has, as usual,

‘every triangle’ where we say ‘any triangle.’ The
genitive is constructed with ^ c/cto? ytovta; but it is
best to translate iravro? Tpiywvov by ‘In any tri-
angle.’
2. 17 Iktos yayvLOL, ‘the exterior angle’ (literally

‘the outside angle’), is the angle formed by any side
and an adjacent side produced.
eKaripa^ rtov hn-bs koI aireyavTiov ytaruaVy ‘than

each of the interior and opposite angles.’ evros
and aTrevavTLov are adverbs but are used here, like
Iktos, adjectivally.
10. (perfect imperative passive of ^layîv),

a variation on the usual ckJ^cÂ-tJo-ô); ‘let be
drawn (or ‘carried’) through.’
PROPOSITION 17.
I. ‘jravTOs rpiytavov, ‘In any triangle,’ as usual.

at Svo ycovtat. As usual, the Greek idiom uses the
definite article where we say ‘two angles’ simply.
The force of the definite article is ‘the two angles
that may be taken (whichever they are).’
2. 'jrdvTYj peraXapupavoficvcuy ‘taken together in
any manner’; the meaning is ‘any two angles added
together.’
:
t88 NOTES
We may translate ‘In any triangle two angles
taken together in any manner are less than two right
angles.’
7. cKTos ioTL ytavta. cfcrosis again used adj ectivally
‘since the angle A CD is an exterior angle of the

triangle ABC.^
13. bfjLoiia^ ‘similarly we can prove,’
as usual.
PROPOSITION 18.
It is necessary to be dear as to the distinction

between this proposition and the next, to know, that
is, which comes first and is proved directly, and
which comes second and is proved by reductio ad
ahsurdum; and, also, in which of the two it is the
greater side that is given and in which the greater
angle is the datum. On the first point Todhunter
has the useful note ‘In order to assist the student
in remembering which of these two propositions
(i, 18, 19) is demonstrated directly and which in-
directly, it may be observed that the order is
similar to that in i, 5 and i, 6.’
On the second point it is to be noted that the

datum comes first in each enunciation. Thus in
i, 18 the enunciation ‘In any triangle the greater
side subtends the greater angle’ means that we are
given that one side is greater than another and have
to prove that, of the angles which the two sides
respectively subtend, the angle subtended by the
greater side is the greater.
j
2. In this enunciation we have vTroT€LVGLv (‘to
subtend’) used with the simple accusative instead

of the more usual vtto and accusative.
6. Keta-OiD ry AB tcrr)
^ AA: as usual, ‘let AD he
made equal to
PROPOSITION 19-
In order to get the datum (here the greater angle

not the greater side) first in the enunciation, we can
use the passive of the verb for ‘subtend’ in the
translation instead of the active. ‘In any triangle
the greater angle is subtended by the greater side.’
7, 8. tcrq yap av ‘would have been equal.’ ovk

eoTi Se, ‘but it is not,’
9. ovSc ‘nor yet’ or ‘nor again.’
PROPOSITION 20.
I, ‘In any triangle two sides taken together in

any manner are greater than the remaining side.’
The phraseology is similar to that of the enunciation

of Prop. 17: see the note on that proposition. The
meaning is that the sum of any two sides is greater
than the third side.
According to Proclus, it was the habit of the

Epicureans to ridicule this theorem as being evident
even to an ass and requiring no proof, the argument
being that, if fodder is placed at one angular point
of a triangle and the ass is at another, he does not,
in order to get to his food, traverse the two sides of
the triangle but only the one side: an argument

1 ^ NOTES
which made Savile retort that its authors deserved
to be put on the same diet as the ass.
PROPOSITION 21.
I. anrb tS>v êpdrwv^ ‘from its extremities.’
2. €vtos (TvarraOtiicnv. As we have seen (note on

Prop. 7), (jvyL(rra(r6aLy ‘to be constructed,’ is
specially used of ‘constructing’ two straight lines

from the extremities of a straight line (in this case
a side of the triangle) to meet at a point, cvtos
means ‘within the triangle.’
We may translate ‘If on one of the sides of a
triangle, from its extremities, there be constructed
two straight lines meeting within the triangle.’
2, 3. at (TvcrraOeLcraL^ ^the straight lines (so) con-
structed.’
3. Tcov Xonrlbv toO rpiydyov Bvo wA-cvpaiv, genitive

after iXarroves, ‘less than the remaining two sides
of the triangle.’
20. ‘were proved (to be)
greater.’
25. 8ia Toura roan/y, ‘For the same reason then.’
PROPOSITION 22.
I. at euriv urai rpiai rats So^cwrats. We may

translate ‘which are equal to three given straight
lines,’ although the force of the words is rather

‘which are equal to three straight lines, namely those
given’ or ‘to given straight lines three in number.’
3. Set S^, ‘Thus it is necessary that two of the

straight lines taken together in any manner should
.:
PROPOSITIONS 21 , 22 191
be greater than the remaining (straight line).’ What

is introduced by the words Set St) is the statement
of the condition which is necessary in order that a
solution of the problem may be possible. This, as we
have seen (p. 161-2 above), was called the hopia-fios.
The criterion of possibility is of course supplied by

the theorem in Prop. 20. If any two of the given
straight lines are not greater than the third, no

triangle can be constructed with sides equal to them
respectively.
The MSS. add the following words to the enuncia-

tion: Sta TO Kol iravro^ rpuy^vov ra^ Svo TrXevpas
ttJs A.oi'ttjJs peCôvas ctvox ‘irdvry peraXappavopeva^.

These words (Sta to . . . ‘because of the fact that’ . .
followed by the acc. and inf.) give a mere repetition

of the enunciation of Prop. 20 and are probably a
gloss; they are not found in Produs.
12. c#cK€to-0ft), for the passive of eKTiOei/cu, as usual
‘
let a straight Hne be set out.’
Treirepacrph/ri p}y Kara to A d-TTCtpog Sc fcaTa to

E, ‘terminated at D but unlimited (in length) at
(i.e. in the direction of) E.^

We have again, as in i, i, to satisfy ourselves that
the two circles drawn as in the figure will actually
meet in some point. They will in fact meet in two
points, one on each side of DE, if the condition is
fulfilled that any two of the straight lines A, B, C

are together greater than the third. Euclid assumes
this without proof.
192 NOTES
PROPOSITION 23 .
I- TTpos 8o0€L(ry evdeCa. In this proposition,

where an angle has to be constructed on a straight
line, i.e. so that the given straight line is one of the

straight lines containing it, it is best to translate
‘On a given straight line and at (irpog understood)
a point on it.’ The order of the rest of the enuncia-
tion is cn)(rrq<ra(r6<u y<mriav €vOijypafjLfiov LarrfV rfj
SoOeLoiTj yfsiVLa evdvypajxpM^ ‘to construct a recti-

lineal angle equal to a given rectilineal angle.’
Observe that we have to construct the triangle

so that AG lying along AB and equal to CE is the
base, and AF is equal to CD and GF to DE ;
that
is to say, we have to describe two circles (i) with A
as centre and radius equal to CD and (2) with G
as centre and radius equal to DE.
According to Proclus, Eudemus said that this
problem was ‘rather the discovery of Oenopides,’
from which we may probably infer, not that the
problem was the invention of Oenopides, but that
he was the first to propound the solution here given
by Eudid.
Apollonius, we are told, gave a different solution.
Although, according to Proclus, the language used
by Apollonius was inappropriate to Book i, the
solution amounts to this. Let CDE be the given
angle, AB the given straight line.
With A as centre and AB 2iZ radius describe a
circle BF. Cut off DC equal to AB and with D as
centre and DC as radius describe a circle meeting

DE in E.
Join CEy and with B as centre and radius equal to

CE draw a circle cutting the circle BF in F. Join
AF,
Then the angle BAF is equal to the angle CDE ;
for in the triangles DCE^ ABF two sides CZ), DE

are equal to two sides BA^ AF^ and the base CE is
equal to the base BF^ so that i, 8 applies.
This method was evidently intended to be more
practical than Euclid’s.
PROPOSITION 24.
The wording of the enunciation follows closely
that of the enunciation of i, 4, with fueCôva, in-
stead of Lo-j/v, qualifying the included angle and the

base in one of the triangles. We may translate ‘If
two triangles have two sides equal to two sides re-
spectively but have the one of the angles contained

by the equal straight lines greater than the other,
they wiU also have the base greater than the base.’
15. KCLo-Bio oTToripa r<ov AV, AZ tcnf r) AH, ‘let
Z)G be made equal to either of the (sides) J9T’;

since AC, DF are equal, it does not matter whether
DG is made equal to AC or to DF.
H- E. 5^3
194 NOTES
As the proposition is framed, there are three
possible cases, of which, as usual, Euclid gives only
one; the others are left for the student to supply
for himself, and they are not difficult. In the case
taken by Euclid EG falls within the angle DEF, The
other two possibilities are these:
(i) G may fall on EF produced. In this case it
is at once obvious that EG is greater than EF,
A D
(2) EG may fall below EF^ making the angle

DEG greater than the angle DEF.
In this case, by i, 21, DF, FE are together less
than DGy GE. But DF is equal to DG; therefore,

subtracting the equals, we have EF less than EG
or BC.
Simson, as is well known, reduced the number of
cases to one by inserting, at the beginning, the
PROPOSITION 24 195
words ‘ Of the two sides DE^ DF let DE be the side

which is not greater than the other.’
A modern proof covering both Euclid’s case and
the case numbered (2) above is worth giving.
Make the triangle DEG equal in all respects to the

ABC, as in the proposition.
triangle
Bisect the angle FDG by the straight line DH
meeting EG in H.
Then, in the triangles FDH, GDH,
the two sides FD, DH are equal to the two sides
GD, DH,
and the included angles FDH,GDH are equal.
Therefore the bases HF, HG are equal. [i, 4]
ThereforeEG is equal to the sum of EH, HF,
But EH, HF are together greater than EF [i, 20] ;
therefore EG, or BC, is greater than EF.

196 NOTES
PROPOSITION 25.
3. Kcu T^v y<i)VLav t^s yoivias fieiôva t^v

xnro Tiov Lcrtav evSciGtv Tr€pi€)(0]ji€vr}v. We translate
‘they will also have the one of the angles contained
by the equal straight lines greater than the other.’
This proposition is a partial converse of the pre-

ceding, and Euclid proves it by rediictio ad ahsurdum.
Two direct proofs are handed down in Proclus’s
Commentary. They are due to Menelaus and Heron
(both of Alexandria) respectively.
PROPOSITION 26.
This is the third theorem proving that two tri-
angles which have certain parts respectively equal

are congruent, that is, equal in all respects, i,
4 proves
that this is the case if two sides and the included
angle are respectively equal; i, 8 shows that the
two triangles are congruent if all three sides are
equal respectively. In the present proposition the
hypothesis is that the triangles have two angles and
one side respectively equal, and there are two cases
according to the particular sides which are assumed
to be equal. The equal side is tJtoi t^v irpos rais
urats ywviais ^ r^v xnroreivova-av vtto pXav roiv taoiv
yiovmv, êither (i) the side adjoining (or adjacent to,

ffpo5) the equal angles, or (2) the side subtending (i.e.
opposite to) one of the equal angles ’ in each triangle.

29. hroKîTOJL loyj, ‘is by hypothesis equal’
(literally ‘is supposed equal’), wroKeio-Oai being used
as the passive of vn-oTiOcyaL (cf. ‘hypothesis’).
PROPOSITIONS 26 197
Tlie proofs are by reductio ad dbsurdtm and,

though somewhat long, do not require particular
elucidation.
Another method in use in ancient times was based
on the application of one triangle to the other: this
is the method of superposition which there is reason
to believe that Euclid avoided wherever he could.
Proclus has the following interesting note.
‘Eudemus in his geometrical history refers this
theorem to Thales. For he says that in the method
by which they relate that Thales proved the distance
of ships on the sea it was necessary to make use of
this theorem.’ This may have been done in either
of two ways.
(i) The distance of the ship from the shore may
have been found thus. Let B be the ship, A a point

on the shore. Take any distance AC on the shore at
right angles to AB, and bisect AC at Z). From C
draw CE at right angles to CA in the direction away
198 NOTES
from and let E be the point on CE which is found
to be in a straight line with B and D.
Then in the two triangles BAD, ECD two angles
BAD, ADB are equal to two angles ECD', CDE re-
spectively, and one side AD is equal to one side CD ;
Therefore, by i, 26, the triangles are equal in aU

respects, and CE (which can be measured) is equal
to AB, the required distance.
(This was the method used by Marcus Junius
Nipsus, a Roman agrimensor, to find the distance
from a point A to an inaccessible point B.)
As however this method would require a certain
extent of free and level ground to admit of the con-
struction and measurements described, I suggest
that the method of Thales may rather have been

the following.
(2) If, as we may assume, he could reach an
elevation, e.g. the top of a tower, he had only to use
a rough instrument made of a straight stick with a
straight cross-piece fastened to it so as to be capable
of turning about the fastening (say a nail) so that it
could form any angle with the stick and would

remain where it was put. (Even a large pair of
compasses would serve the same purpose.) The first
thing then was to fix the stick upright (by means of
a plumb-line) and direct the cross-piece to the ship.
Next, leaving the cross-piece at the angle so found,
the observer would turn the stick round, while
keeping it always vertical, until the cross-piece
pointed to some visible and accessible object on the
shore, when the object could be mentally noted and

the distance from the bottom of the tower to it
could be measured at leisure. This would, hy i, 26,

give the distance from the bottom of the tower to
the ship.
Professor David Eugene Smith has confirmed the
probability of this suggestion. He observes (The
Teaching of Geometry^ p* 172) that this method
is found in so many practical geometries of the first
century of printing that it seems to have long been

a common expedient. He adds a story that one of
Napoleon’s engineers won the imperial favour by
quickly measuring the width of a stream that
blocked the progress of the army, using this very
method.
PROPOSITION 27.
I. els Svo evOelas efjuirLTrrovo-a, ‘ falling on two

straight lines.’ The phrase is the same as that used
in Post. 5, and means that the straight line lies
across the two straight lines, is, in fact, what we call
a transversal,
ras epaAAo^ ywws, ‘the alternate angles.’ evaAAaf,

an adverb, ‘alternately,’ is here used adjectivally.
The same word is used of the transformation of a
proportion in four terms by which the terms are
taken alternately^ i.e. if the proportion A h to B as
C is to Z> is altered to the proportion ^ is to C as 5
is to D. The use of ‘alternate’ to describe the angles
AEF^ EFD (or the angles BEF^ EFC) in the figure
.
200 NOTES
of the proposition is natural if we consider the four
internal angles at E and F and take them (say) in
the order of the motion of the hands of a watch, i.e.
in the order AEF, FEB^ EFD^ EFC, when it is seen

that AEF and EFD are alternate.
8. crvju.7r€<rowTat (future of crujw.7rwrT0)), ‘will
meet.’
9. hrl Ttt B, A jx€pi7, ‘in the direction oi B,
(literally, ‘towards the parts, or regions, B^ Z)’)
We must understand fiiprj after ‘A, P’ also: ^rot . .
^ ‘in the direction either of B, D or of A, C.’

,
^ Iktos ycox/ta, the exterior angle ’ ; y cktos Kal

‘
11
.
oLTrei/avTLov ycovLa, ‘the interior and opposite angle,’

as usual (cf. Prop. 16).
15, 16. €7rl firjSmpa ra p-epT/, ‘in neither direc-
tion,’ as in Post. 5.
This proposition begins the second section of

Book I. The first section has dealt mainly with
triangles, their construction and their properties in
the sense of the relation of their parts, the sides and

angles, to one another, and the comparison of
diiferent triangles with reference to their parts and
to their areas in the particular cases where they are
congruent.
This second section establishes the theory of
parallels and introduces the cognate matter of the
equality of the sum of the three angles of any tri-
angle to two right angles; then come two pro-
positions, namely i, 33, 34, which introduce the
parallelogram for the first time and form the tran-
sition to the third and final section. The third

section treats of the areas of triangles, parallelo-
grams and squares in relation to one another, the
special feature of the section being a new conception
of equality of areas, equality not dependent on
congruence.
PROPOSITION 28.
One criterion of the parallelism of two straight

lines, the equality of the alternate angles made
with any transversal, has been given in i, 27; this
proposition gives two alternative tests which are
equivalent to that of i, 27 and are in fact reduced
to it.
2. eyrog Kal aTrcvavrtov koi IttI ra avra fJiipr},
‘ (equal) to the interior and opposite (angle) on the

same side,^ i.e. on the same side of the fine (the
transversal) which ^ falls on’ (i.e. across) the two

others.
3. Tos cFTos Kol iTTL TO, ovTo, ‘the interior

angles on the same side.’
From remarks on the criteria of parallelism made

by Aristotle {Anal. Post, i,
5 and Afial. Prior, ii, 17)
it is clear that the two propositions i, 27, 28 were
familiar to him; but, as we gather that the theory of
parallels current in his time contained some vicious
circle (an assumption, presumably in a different
form, of the very thing required to be proved), it
seems clear that Euclid’s Postulate 5 and his pro-

position I, 29 based upon it were due to Euclid
202 NOTES
himself, who may therefore be taken to have placed
the theory on a proper basis for the first time.
PROPOSITION 29.
This important proposition is the converse of

I, 27, 28 and depends for its proof on the famous
Postulate 5. The three alternative criteria of paral-
lelism are shown to be satisfied by any two parallel
straight lines.
18. at Sc aTT IXacrcrovdiv ^ Bvo opOtov c/cjSaAAo/jtcvat
€ts wireipov avp/TTLirrovariv, literally ^ straight lines pro-

duced without limit from angles less than two right
angles meet.’ This is a variation from the more
explicit language of Post. 5 but means the same
thing. A good deal is left to be understood, namely
that the straight lines start from points at which
they meet a transversal, and that they make with
it interior angles on the same side the sum of which
is less than two right angles.
21 . Slot TO TrapaXX'i^Xovs auras wro/cetcrât, ‘ because
they are by hypothesis paraUel’ (lit. ‘arc supposed
parallel’).
In modem text-books Euclid’s Postulate 5 is often

replaced by another, known as Playfair’s Axiom. It
was however no new discovery of Playfair’s, since it
is explicitly stated in Proclus’s Commentary on

Book I. It is to the effect that ‘Through a given
point only one parallel can be drawn to a given
straight line’ or Tf two straight lines intersect one
another, they cannot both be parallel to one and the

same straight line.’
If this Axiom is used, the proof of i, 29 proceeds

thus.
Given two parallel straight lines AB^ CD and a
transversal EF, to prove that the alternate angles
AGH, GHD are equal.
If they are not equal, draw another straight line

KL through G making the angle KGH equal to the
angle GHD,
Then, since the alternate angles KGH^ GHD are
equal, KL is parallel to CD, [i, 27]
Therefore two straight lines KL, AB intersecting
at G are both parallel to CD : wliich is impossible (by
the Axiom).
Therefore the angle AGH cannot but be equal to
the angle GHD.
The rest of the proposition follows as in Euclid.
PROPOSITION 31.
This proposition implies, as Proclus observes,

that only one straight line can be drawn through a
given point parallel to a given straight line.
204 NOTES
PROPOSITION 32.
1-3. ‘In any triangle, if one of the sides be

produced, the exterior angle is equal to the two
interior and opposite angles,’ i.e. to the sum of the
two interior and opposite angles.
According to Proclus, Eudemus referred the dis-

covery of this theorem to the Pjâgoreans, and
gave what he affirmed to be their demonstration of
it (see below). It is a question whether Thales was
aw'are of the theorem, so far at least as the right-
angled triangle is concerned; for Pamphile (reign of

Nero, A.D. 54.-68) is quoted by Diogenes Laertius as
saying that Thales, who learnt geometry from the
Egyptians, was the first to inscribe a right-angled
triangle in a circle and sacrificed an ox on the
strength of it; in other words, that he proved that
the angle in a semicircle is right (whence it would
easily be seen that the sum of the three angles of a
right-angled triangle is equal to two right angles).
But other authorities said it was Pythagoras.
The Pythagorean proof of the equality of the sum
of the three angles of any triangle to two right
angles was, according to Eudemus, the following.
Let ABC be any triangle, and through A draw

DAE parallel to BC.
Then, since BC^ DE are parallel,
the alternate angles DAB, ABC are equal to one
another; and so are the alternate angles EAC,
ACB also.
Therefore the sum of the angles ABC, ACB is

equal to the sum DAB, EAC.
of the angles
Add to each the angle BAC;
therefore the sum of the angles ABC, ACB, BAC
is equal to the sum of the angles DAB, BAC, CAE,
that is, to two right angles.

We gather from a remark of Aristotle that the
proof with which he was familiar followed Eudid^s
method.
PROPOSITION 33.
I. at Tos l(Tas T€ #cat TrapaWi^Xovs hrl ra. avra

IJiipTf iTriêuyvuovaaL evOcica, literally ‘the straight
lines joining equal and parallel (straight lines) in
the same directions’; by this is meant, and we may

SO translate, ‘the straight lines joining (those ex-
tremities of) equal and parallel straight lines (which
are) in the same directions.’
2, 3. teat T€ /cat TrapdkXrjXoL eicriv, ‘are (them-
selves) equal and parallel.’

5. hrL^€vyvvr<i><rav avras: iwi ra avra fJiifyrj cvOeicu
at Ar, BA, 4et the straight lines AC, BD join them

(i.e. their extremities) in the same directions.’ The
meaning is that A, C and B, D are joined respec-
2o6 NOTES
tively, not B, C and J, jD, which pairs of points are
not ‘in the same directions.’
12. Svo St7 at AB, BP dvo rats BP, PA t^crat elariv,
‘the two sides AB, BC are equal to the two sides
BCyCD ’
; we should naturally say DC, CB instead of
BC, CD in order to put the equals in corresponding
order.
The effect of this proposition is to demonstrate
the existence of the figures which we call parallelo-
grams.
PROPOSITION 34.
The preceding proposition has shown a method

of constructing a four-sided figure with its opposite
sides parallel, and Euclid in this proposition uses the
terms ‘ parallelogrammic area’ (or, more exactly,
‘parallelogram area,’ TrapaXkTjXoypafifWv ^wpiov,
meaning an area formed by parallel lines, in pairs)
and ‘parallelogram’ (7rapaXkr^X6ypap,p^v alone) to

describe the said figure, without any definition or
further explanation. The word is tacitly appro-
priated to four-sided figures with opposite sides
parallel; and we gather from Proclus that Euclid
was the first to introduce the word TOpaAAiyA.o-
ypap.pjov in this restricted sense, which excludes
such figures with more than four sides, e.g. a regular
hexagon, as have their opposite sides parallel.
I. ^(coptov means any ‘place’ or ‘area,’ but is, in

Greek geometry, usually restricted to areas in the
shape of rectangles or parallelograms.
1. at aTTo/avTiOV 'jrX.evpaC re /cat ywvLoi, ‘tl3.e
opposite sides and angles (respectively).’

3. BidfLerpos, ‘diameter.’ We generally use the
word ‘diagonal’ in relation to a parallelogram; the
Greek usage preferred ‘diameter,’ and the word
should be so translated here.
avrd ‘bisects them’ (i.e. the paraUelo-
grammic areas or parallelograms).
17. /ttav TrXevpdv pud ‘TrXevpa tanrjv r^v -/rpos ratg

t(rat5 ycovtats kolv^v avrwvr^vBV, ‘one side equal to
one side, namely that which adjoins the equal angles
(and is) common to the triangles, that is, JSC.’
PROPOSITION 35.
Euclid here drops the longer expression

‘paraUelogrammic area’ and uses the word ‘paral-
lelogram’ exclusively.
2. €v reus avrous gropoAAoyAotg, ‘in the same
parallels,’ means included by the same two parallel
straight lines; that is, the side opposite the base

always lies alongoneandthe same straight line which
is parallel to the base.
la-a aXXiyXois loriv, ‘are equal to one another.’

We are here introduced to a new sense of the word
‘equal’ as applied to two figures. Hitherto we have
had equality in the sense of congruence or symmetry
only; the equal triangles, as weU as the equal straight
linesand the equal angles, which we have met with
up to now have aU been equal in this sense. But now,
without any explicit warning of a change in the
20S NOTES
meaning of the word, figures are inferred to be
‘equal* which are equal in area or content but need
not be of the same shape.
In this proposition the ‘ equality,’ in the new
sense, of the parallelograms ABCD, EBCF is in-
ferred by means of two steps. First, one and the
same area, the small triangle DGE, is subtracted
from each of the two triangles ABE, DCF which are
equal in the sense of congruent, and it is inferred that
the remainders, the trapezia ABCD, EGCF, which

are not of the same shape or congruent, are ‘equal’
in the new sense. Then, secondly, to each of these
‘equals’ is added one and the same area, the triangle
GBC, and it is inferred that the sums, which are the

two parallelograms respectively, are also ‘equal’ in
the new sense.
Simson (after Clairaut) slightly altered the proof

by substituting one step of subtracting congruent
areas (the triangles AEB, DFC) from one and the
same area (the large trapezium ABCF) for the tzvo
steps in Euclid’s proof.

While, in either case, nothing more is explicitly
used than the axioms that, ‘if equals be added to

PROPOSITION 35 209
equals, the wholes are equal’ and equals be

subtracted from equals, the remainders are equal,’
there is the further tacit assumption that it does
not matter to what part or from what part of the
same or equal areas the same or equal areas are
added or subtracted. De Morgan observed that the
postulate ‘an area taken from the same area leaves
the same area from whatever part it may be taken’
is particularly important as the key to equality of
rectilineal areas which could not be cut into parts
that are respectively coincident.
As Proclus points out, there are three cases of this
proposition according to the form which the figure
takes. Euclid, as usual, takes one case only and leaves

the others, shown in the accompanying figures, for
the student to investigate for himself.

Proclus observes that the present theorem and the
similar proposition (37) regarding triangles are
among the theorems in mathematics which were
deemed paradoxical, because the uninstructed might
well think it impossible that the area of the parallelo-
grams should remain the same while the length of
the sides other than the base and the side opposite
to it may increase indefinitely. In fact the sum of
H. £. X4
210 NOTES
the sides, or the ferimetery of parallelograms or tri-
angles is of itself no criterion as to their area. Mis-

conceptions on the subject were rife among non-
mathematicians; Proclus tells us that people used
to judge of the size of cities by their perimeters,
and certain members of communistic societies in his
own time cheated their fellow-members by giving

them land of greater perimeter but less area than
that which they took themselves, so that, on the one
hand, they got a reputation for greater honesty,
while, on the other, they took more than their share
of the produce. Literature furnishes other examples
of the same misconception; thus Thucydides esti-
mated the size of Sicily by the time required to
circumnavigate it.
PROPOSITION 37.
The equality of the triangles ABC^ DBC is in-

ferred from the facts (i) that the parallelograms
EBCAy DBCF are equal, and (2) that the triangle
ABC is half of the former parallelogram, while the
triangle D 5 C is half of the latter. The inference is, in
the Greek text, justified by the remark (18) ra 8c t^v
Lo-fov lora dAAiyXots lariv, ‘But the halves of
equals are equal to one another.’ This is a reference
to an axiom which was apparently interpolated
among Euclid’s Common Notions in very early times.
It is probable therefore that the reference to it here
was also interpolated, and that Euclid himself made
the inference without any explanation.
2
PROPOSITIONS 37, 39 zii
A similar remark applies to the final step in the

next proposition (38).
PROPOSITION 39.
We translate the enunciation thus, ‘Equal tri-
angles which are on the same base and on the same

side are also in the same parallels.’
After the words (5) hrl ra avra t^s BP in
the setting-mitj the text proceeds ort koX kv rat?
avTOL^ irapaXX.T^XoiS icrriv. ’ETre^cuxô) yap tj AA*
\€y<a OTL TrapaXXiyXos i<mv yj AA ry BP, ‘I say that
they are also in the same parallels. Eor let AD be
joined; I say that AD is parallel to BC,^ The two

statements of the fact to be proved, introduced by
X^<o, ‘I say,’ are suspicious; and Heiberg has proved
by means of an ancient papyrus-fragment that the
words same parallels’
say that they are also in the
are an interpolation by some one who did not ob-
serve that the words ‘And let AD be joined’ are
part of the setting-mt but took them to belong to
the construction^ the interpolator then altered ‘And’
into ‘For.’ We should therefore, after BP, read
as follows: koX A A- Xô) on irapaX-
rj
XrjXos ioTLv y AA ry BP, ‘And let AD be joined.

I say that AD is parallel to BC,^
The theorem is a partial converse of i, 37. In i, 37
we have triangles which are (i) on the same base,
(2) in the same parallels, and the theorem proves
(3) that the triangles are equal. The partial con-
14 —
212 NOTES
verse takes as hypotheses the hypothesis (i) and the
conclusion (3), and proves (2) as the conclusion.
PROPOSITION 40.
Heiberg has shown by means of the papyrus-

fragment mentioned in the last note that this pro-
position is an interpolation by some one who thought
that there should be a proposition following i,
39
and related to it in the same way as i, 38 is related
to I,
37 and 1, 36 to i, 35.
PROPOSITION 41.
I. €av TrapaXXyjXAypafifiov rptytovo) pdaiv re
^XO OLvrriv Kal iv rocs avrais TrapaXXTjXoLS The

dative rpLydvio is constructed with avrrjv and
rats avTOL^s TrapaXX’^Xois. ‘If a parallelogram
have the same base, and be in the same parallels,
with a triangle’ (or âs a triangle’).

Euclid infers that, because the parallelogram is
double of the triangle JBCy and the triangle ABC

is equal to the triangle BBC, the parallelogram is
also double of the triangle EBC. This can be de-
duced by means of Common Notions i and 2, in
this way. Triangle ABC = triangle EBC; therefore

(adding equals to equals, C. N. 2) twice triangle
ABC=^ twice triangle EBC. But parallelogram —
twice triangle ABC; therefore parallelogram =
twice triangle EBC (C. N. i).
PROPOSITIONS 40^.3 213
PROPOSITION 42.
I. The order of words in the enunciation is
crvoT'iqa'acrOaL 7rapaXX.rjX6ypafjLfiov urov rw SoOevTi

Tpi-ycovw, ^to construct a parallelogram equal to a
given triangle.’
2. iv ry SoOcLCTT} yiavCa evdvypdpupxa, ‘în a given
rectilineal angle’; this means having one of its
angles equal to a given rectilineal angle, so that the

parallelogram could be conceived to be placed in
the angle in the sense of having one of its angles
coincident with the angle.
PROPOSITION 43.
I. It is best to translate iravros nrapaXkyXo-

ypdpfjLOv by Tn any parallelogram,’ as usual.
rd 'TrapaTrXypdpxvra twv Trepl ryv Zidp^rpov TrapctX-

XyXoypdpLpLwv, ‘the complements of the parallelo-
grams about the diameter.’ The complements are
the parts which fill up the interstices (jrapd and
TrXrifyiqs, ‘full’), i.e., when added to the ‘parallelo-
grams about the diameter,’ they complete the
original parallelogram.
6. In the setting-out the complements are de-
scribed as ‘the so-called complements,’ ra Xeyoftcva
Trapd'irXypdpxLTa'. ‘let KD be the so-called com-
plements.’ This sufficiently explains what is meant;
the complements are the figures (parallelograms)
left on each side of the diameter when the ^para-
lelograms about the diameter,’ EH and GjP, are
taken away from the original parallelogram.

Z14. NOTES
TMs seemingly insignificant theorem is, in the
next proposition, made the basis of one of the most
important and far-reaching problems of construction
in the whole of theoretical geometry.
PROPOSITION 44.
We are here introduced to an elementary case

of the general method of application of areas which
is fundamental in Greek geometry.
I, 2. ‘irapapaXeLv irapa, literally ‘to lay along-side
of’ or, as we generally say, ‘to apply to.’ ‘To apply
to a given straight line a parallelogram equal to a
given triangle in a given rectilineal angle.’ The last
phrase, ‘in a given rectilineal angle,’ means, as in
Prop. 42, that the parallelogram must be such that
one of its angles is equal to a given rectilineal angle.
15. eveTTcorcv: we may translate ‘falls,’ although
Euclid uses the aoiist, ‘fell’ or ‘has fallen.’
17. al Se d-TTO cA.ao'crdvwv ^ Svo opOlav /c.t.A.. Post. 5

is here cited, as in Prop. 29, in shorter but equivalent
terms: ‘straight lines produced without limit from
angles less than two right angles meet.’
24. TrepL Se ryv ®K irapOLXXrjXoypapLpjOL pukv rd
AH, ME, rd Sc dcyo/ACva ‘Trapa^rA.iypoiâTa rd AB,
BZ. The words ir^pi r^v ®K, ‘ about HKy go with
both the expressions distinguished by and Sc:
‘and JG, ME are parallelograms, and LBy BF the
so-called complements, about HK,^
It is necessary to lay stress on the supreme im-
portance of this proposition. Prop. 42 has shown us
;
PROPOSITION 44 ^15
how to make a parallelogram equal to a given tri-

angle and having one angle equal to a given angle.
But the length of the sides of that parallelogram are
determined for us hy the construction, since one of

them is necessarily equal to half one of the sides of
the given triangle, and the construction determines
the length of the other; we cannot, by means of that
proposition, make either side of the parallelogram
oj any length we please. This is what the present
proposition enables us to do; v/hatever shape the
parallelogram in Prop. 42 has, we can now transform
it into another with the same angle and of equal
area but vdth one side of any required lengthy e.g. a
unit length, say an inch, a foot or a centimetre long
and the combined effect of the two propositions is
that we can construct a parallelogram having one

side of any length we please and one angle equal to
any angle we please, which shall be equal in area to
any given triangle. The marvellous ingenuity of the
solution of the problem by the simple application of
the property that the ‘complements of the parallelo-

grams about the diameter ’
are equal is indeed
worthy of the ‘godlike men of old,’ as Proclus calls
the discoverers of the method of application of areas.

We have the authority of Eudemus for the state-
ment that the discovery of the method of applica-
tion of areas was due to the Pythagoreans, and there
would seem to be no reason to doubt that the par-

ticular solution of the problem in i, 44 that g^ven
by the Pythagoreans,
2I6 NOTES
Tile present problem is however only the most
elementary case of the complete method of applica-
tion of areas discovered by the P3^hagoreans, which,
taken as a whole, is one of the most effective methods
employed by the Greeks in higher as well as elemen-
tary geometry. Not only did they apply parallelo-
grams of given area to a given straight line exactly,
but they applied parallelograms eqnal to any given
rectilineal figure in such a way that they overlapped
or fell short of the straight line to which they were
applied, and that the portion of the parallelogram
which overlapped or fell short had its sides in a given
ratio. The extended method requires propositions
in Book ii and Book vi of Euclid and therefore can-
not be described here. It must suffice to say that the
problems solved by the method are the geometrical
equivalent of the solution of mixed quadratic
equations in algebra.
Apollonius used the method as the foundation of
his treatment of conic sections. The overlapping of
the area was called exceeding, falling-
short was called IWeii/rtg, and the exact application
was rrapapoX-rl; and these terms (in English hyper-
bola, ellipse and parahold) were actually used for the
first time by ApoUonius to describe the three different
conic sections.
PROPOSITION 45.
Any rectilineal figure can be divided up into a

certain number of triangles; and this proposition
PROPOSITIONS 44^47 217
shows how, hy means of r, 42 and

44, we can make
one parallelogram with one side of given length and
one angle equal to a given angle which shall be equal
to the whole rectilineal figure.
2, 3. evOvypafifjLOVj literally ‘rectilineal,’ is used

in the enunciation and later on as a substantive to
mean a ‘rectilineal figure,’ just as irapaXkrjXo-

ypapLpLov is used for a parallelogram.
PROPOSITION 46.
I. avaypau^€a/ a^ro, literally ‘to draw up from^

is the technical expression for ‘ describe upon.’ ‘
On
a given straight line to describe a square.’ The ex-
pression is to be distinguished from (rvcrr^cracrOaL^
which is used of constructing a. triangle (or an angle).
The triangle is, so to speak, pieced together, while the
describing of a square on a given straight line is the
making of a figure ‘from’ one side.
As before remarked, the definition of a square
(Def. 22) says nothing as to whether a figure corre-

sponding to the definition exists or not. The proof
that such a thing exists is furnished by the actual
construction of it in this proposition; for the figure
drawn is proved to be both equilateral and right-
angled (in the sense of having all its angles right
angles).
PROPOSITION 47.
I. TO a-TTo 7*5? opO^v yiiiVLav vTroreLVovanfjs
rrXevpa^ Terpdyiavov, ‘the square on the side sub-
tending the right angle.’ As ‘to describe a square

2x8 NOTES
on’ a straight line is av(Lypd<jieLV rerparyaivov cxTro, so
TO a^ro evOeCas dvaypa(f>h/ or dvayeypafJifLci'ov

TCTpdymvov is the full expression for “^the square
described on the straight line.’ The participle is
generally omitted, as here; and rerpdyoivov itself is
commonly left to be understood, to d7ro (with gen.)

meaning invariably ^the square on.’
Tj r^v 6p6^v yoiVLav ujrorewovcra irXevpd is ‘the

side subtending the right angle,’ uTroTetvoucra being
here used with the simple acc. instead of with vô
and acc. V 7roT€Lvov(ra in connexion with a right-
angled triangle came to be used alone in the special
sense of hypotenuse^ meaning the particular side
which ‘subtends,’ or is opposite to, the right angle.
3. TOis aTTO Ttisv T^v opO^v yoiviav irepiexpvcrihv
wkeopihv rerpaydivoL% ‘to the squares on the sides
containing the right angle.’

The full enunciation then is ‘In right-angled tri-
angles the square on the side subtending the right
angle is equal to the squares on the sides containing

the right angle.’
11. oTToripa rwv BA, PE, ‘ to either of the straight
lines £D, CE’
16, 17. hr evOcias dpa icrrlv tj PA ry AH, ‘There-
fore CJ is in a straight line with
22. Svo ^ at AB, BA 8vo tcus ZB, BP urai elcriv
eKoripa eKaripa^ ‘the two sides DB^ BA are equal to

the two sides FB^ BC respectively.’ The sides are
not given in corresponding order; we should rather

say ‘the two sides ABy BD ’
PROPOSITION 47 219
33. ra 8k tS)v IcrtAV 8vir\aa-UL ura dXA.iyA.ois ioTcv,
^But the doubles of equals are equal to one another.’

These words are no doubt interpolated; the in-
ference easily follows from Common Notion 2.
This great theorem is universally associated with

the name of Pythagoras. Proclus says Îf we listen
to those who wish to recount ancient history, we

find some of them referring this theorem to Py-
thagoras and saying that he sacrificed an ox in cele-
bration of his discovery.’ Plutarch, Diogenes
Laertius and Athenaeus also agree in attributing
the proposition to Pythagoras. These are, it is true,
somewhat late witnesses, but Cicero (De natura

deorum)y while he disputes the story of the sacrifice
(which would have been inconsistent with Pytha-
gorean ritual), does not seem to question the fact of
the geometrical discovery. Cicero is apparently
commenting on the verses of Apollodorus ‘the cal-
culator’ (o XoyLCTTLKos), wMch allude to a discovery
of Pythagoras in the words ‘when Pythagoras dis-
covered the far-famed figure, that on account of

which he made the famous sacrifice of an ox.’
Plutarch, who quotes the same distich, is not certain
whether the sacrifice related to the discovery of the
proposition about the square on the hypotenuse or
to the solution of the problem about the ‘applica-
tion of an area’ (by which he means the problem
‘given two rectilineal figures, to describe a third
which shall be equal in area to one of the two given

figures and similar to the other’). But Plutarch
220 NOTES
says nothing to suggest that he had any hesitation
in attributing both propositions to Pythagoras.
The connexion o£ Pythagoras with i, 47 is further
confirmed by the tradition crediting him with a rule
for finding a series of square numbers which are the
sums of two squares, in other words, for finding any
number of right-angled triangles with all their sides
expressible in rational numbers. This theory is con-

nected with the use of the gnomon in its geometrical
and arithmetical sense. According to Aristotle the
gnomon is the figure which, when placed round two
sides of a square makes a larger square, while in
arithmetic the odd numbers are gnomons. This is
easily seen from the accompanying figures. Three
dots placed round one make 4 (= 2^); five more

placed round the 4 make 9, or 3^; seven more round
the 9 make 16, or 4^; and so on. The geometrical
gnomon shown in the second figure is regularly em-
ployed in Euclid Book ii, and we know that it was
a term well understood by the Pythagoreans.
The rule attributed to Pythagoras for finding
right-angled triangles in rational numbers connects
itself naturally with the use of the successive odd
numbers as gnomons.
PROPOSITION 47 Z2l
Since i =
1+ 3=2^
1+ 3+ 5= ^2+ 5= 32,
1+ 3+ 5+7=
and so on, we easily deduce the general formula
^2-}- (2^z-f l)= (fl-j- i)2.
If this formula is to give a sum of two squares

equal to a square, we have simply to make (2^ + i)
a square.
We write therefore
2 /z + I = «2 ^
whence i);
therefore ^4- i = J («2 q-

and the above formula becomes
[J (»2 — l)]2q- j-j odd)

which is the precise formula attributed to Pytha-
goras.
In view therefore of the unanimity of tradition on
the subject, there seems to be no sufficient reason to
question that, so far as Greek geometry is concerned,
Pythagoras was the £rst to introduce the theorem of
I,
47 and to give a general proof of it.
Not that particular cases of the theorem were not
known long before Pythagoras. The triangle 3> 4, 5
was apparently recognised as right-angled in Egypt

as far back as 2000 b.c. The ancient Babylonians
were also probably aware of this case; ihe Chinese
certainly knew it. The Indians too, long before the
222 NOTES
date of Pythagoras, had independently investigated,
at least empirically, many cases of right-angled
triangles. They even enunciated the theorem in all
its generality, although there is no trace of their
having given a scientific proof of it.
It is not possible to say with certainty how

Pythagoras came to discover the general proposition
or how he proved it. It is certain that the proof in
Eucl. 47 is Euclid’s own. A proof identical in sub-

I,
stance with Euclid’s can be derived from proposi-

tions in Book vi (Props. 8, 4 and 17), and it is likely
enough that the original proof used the method of

proportions, and that Euclid ingeniously trans-
formed this proof into that in i,

47 which uses the
methods of Book i only.
A most interesting and withal easy extension of

this proposition to the case of parallelograms drawn
on the sides of any triangle (not necessarily right-
angled) is given by Pappus.

Pappus enunciates the theorem as follows.
If .^^jSCbe a triangle, and any parallelograms what-

ever ABEDy BCFG be described on ABy BCy and if
Z)E,jFGbeproduced to meet atH, and^fjB be joined,
the parallelograms ABEDy BCFG are equal to the
parallelogram contained by ACy HB in an angle
which is equal to the sum of the angles BACy DHB.
Produce HB to K-y through Ay C draw ALy CM
parallel to HKy and join LMy meeting HK in N.
Then, since ALHB is a parallelogram, ALy HB
are equal and paralldL
PROPOSITION 47 223
Similarly MCy HB are equal and parallel.
Therefore AL^ MC are equal and parallel;

H
whence LM, AC are also equal and parallel,
and ALMC is a parallelogram.
Further, the angle LAC of this parallelogram is
equal to the sum of the angles BAC^ DHBy since

the angle DHB is equal to the angle LAB.
Now, since the parallelogram DABE is equal to
the parallelogram LABH (for they are on the same
base AB and in the same parallels AB, DH)y
and likewise LABH is equal to LAKN (for they
are on the same base LA and in the same parallels
LAy HK)y
the parallelogram DABE is equal to the parallelo-
gram LAKN.
For the same reason,
the parallelogram BGFC is equal to the parallelo-
gram NKCM.
Therefore the sum of the parallelograms DABEy
BGFC is equal to the parallelogram LA CM, that is,
to the parallelogram which is contained by ACy HB
224 NOTES
in an angle LAC which is equal to the sum of the
angles BAC^ BHD.
‘And this,’ says Pappus, ‘is far more general than
what is proved in the Elements about squares in the
case of right-angled triangles.’
The accompanying figure shows the adaptation of
Pappus’s proof to the particular case of i, 47.
BDy BF are the squares on AB, BC, and we have

first to prove that, if CM are drawn parallel to
AL,
BH to meet Z)E, FG in L, M, then ALMC a is
square.
By the construction, BEHG is a parallelogram, so
thati&jy==5G=^C.
And BE= AB.
Therefore in the two triangles BEH^ ABC^ two
sides BEy EH are equal to two sides AB, BCy while
the included angles are equal.
Therefore the triangles are equal in all respects,
so that the angle EHB is equal to the angle ACB.
(Andj5F=^C.)
PROPOSITION 47 225
Also, by construction, ALHB is a parallelogram,

so that the angle EHB is equal to the angle LAB.
Therefore the angle LAB is equal to the angle
ACB.
Add to each the angle BAC;
therefore the angle LAC is equal to the sum of
the angles BAC, ACBy and is therefore a right
angle.
And ALMC is a parallelogram, as in Pappus’s pro-
AL and CM are each equal to BHy
position, while
which is equal to AC.
Therefore ALMC is the square on AC.
We have now
square BD = parallelogram ALHB
= parallelogram ALNKy
and square BF = parallelogram CMHB
= parallelogram CMNKi
and, by addition, the sum of the squares BDy BF is
equal to ALMCy which is the square on A C.
The best known alternative proof of i, 47 is that
which is attributed to Thabit b. Qurra (826-901
A.D.). This proof has the effect of showing the
equivalence of the sum of the two squares to the one
square in a form which appeals directly to the eye,
Thabit proceeds in this way.
Let ABC be the given triangle, right-angled at A,
Produce AC to Fy making EF equal to AC.
Describe on AB the square ABDEy and on EF the
square EFGH.
H. E. 15
226 NOTES
JoinGC. Produce EH (whicli dearly lies along
ED) Ky so that DK may be equal to AC.
to
Then, since KD = EH, HK = ED or AB.
Therefore, in the four triangles BAC, CFG,
KHG, BDK,
the sides BA, CF, KH, BD are all equal,
and the sides AC, FG, HG, DK are all equal.

K
The equal sides also indude equal angles (namely

right angles) in all the triangles.
Therefore the four triangles are equal in all re-
spects.
Hence BC, CGy GK, KB are all equal.

Again the angles KBD, CBA are equal;
add to each the angle DBC;
therefore the angle KBC is equal to the angle
ABD, and is therefore a right angle.
Similarly the angle CGK is right.
Consequently BCGK is a square.
Now, taking the two squares FH, EB alongside
one another, and subtracting the two triangles FCGy

ABCy we have as the remainder the quadrilateral
GCLH and the triangle LBD^ Adding to this re-
mainder the two equal triangles DBK^ (these HKG
triangles being also equal to the triangles FCG^ ABC
which were before subtracted), we have as the sum
the square BG.
It follows that the latter square, the square on
BC^ is equal to the sum of the two other squares
FH^ which are the squares on AC^ AB re-
spectively.
To show the equivalence, we have only to take
the squares FH^ EB as placed, cut off the two tri-
angles CFGf BAC and replace them in the positions
KHG, BDK.
Although this proof is first found in the Arabic,
it may well have been derived from a Greek source.
PROPOSITION 48.
This theorem is i, 47 and presents

the converse of
no difficulty. It is however worthy of remark that,
by drawing AD at right angles to AC on the side of
it opposite to AB, Eudid avoids the appearance of
indirect demonstration, because he has only to
prove that the two right-angled triangles are equal
in aU respects. This he does by means of i, 8. If,
on the other hand, he had drawn AD on the same

side as AB, it would have been necessary to prove
that the two triangles coincide, and this would have
been done by reductio ad ahsurdum. For, if the two
15—2
22 $ NOTES
triangles did not coincide, we should have, on the
same side of CA^ two pairs of straight lines CB, AB
and CD, AD such that CD= CD and AB^ AD,
which is impossible by i,
7 (the very proposition on
which I, 8 depends).
INDICES
5
INDEX OP GREEK TERMS

EXPLAINED IN THE NOTES
to draw, 1325 ayayelv, 145
dSvvaToSy -oj/, impossible. 37 els ro ddvvarov diraycayri^
reduction to the impossible^ i.e. reductio ad absurdum^
172
alreLVj todemand or postulate, 1425 ^crBcd, 142
a postulate, 142
atrrjfia,
akXos, other. ^p 6 s dXXcd Kal a)\X^ to different
points, 174
dfi 0\.vyd)VLos, -01/, obtuse-angled, 137
dfipXvs, -€La, -V, obtuse, 127
dvaypd(j>€Lv diro, to describe on, 217
dviaos, unequal, 165
conversion, 172
dvTi<rrpo<j>rj^
dvriarpoôsy -oj/ (adj.), converse, 171
dTrayoayr], reduction (of one problem to another, or of
a hypothesis to an absurdity, reditaio ad absurdurri)^
aTreipos, -ovy unlimited, 140, 182, 191; ear* aTveipov, els

aneipov, ad infinitum^ 140, 150
dwevavrlov (adv. Used adjectivally), opposite, 139, 187,
207
dirXaTTjSy -€f, breadthless, 1 1
diro (c. gen.), on (constructed with dvaypd(j)€iv)y 217;
ro UTTO Tjjs BF (T€Tpdy<avov')= the square on SC
arrobeiîs^ proofs l6l
dwoKapi^dveiv^ to cut off, 134
dirrea-Sai, to meet (of lines), 124; dist. e^dirreo-ôt
apa, therefore, 1 59
dpxriy beginning, at ef dpx^s cvSeiai, the original
straight lines, 174
dcl>aip€Lv, to take away or subtract, 152;
dfj>jfpri(r6 a>j 184
Bdortr, base, i66j base of an angle^ 169

232 INDEX OF GREEK TERMS
TvafLoav, gnomon^ origmally an upright needle. Kara
'/vSfjLova, gnomon-wisey ancient term for perpen-
dicular, 127, 183
ypafifxq, line or curve, 117. 129
ypd<l>€LV, to drav^, 147; yeypd^Bco, 142, 158
ycovLCL, angle, 124-6; rj virb BAT
ycavia, the angle
167
AeLKvvvaL, to show or prove, birep eBa BcLâi, QJE,D,y

156, 161; opolcos B€Lôp,€v, iS^j 188
Beivj to be necessary, orrcp eBci Beiâi or iroi^trcu.
O.EJ). or F., 156, 161; Bel B^y 158, (in necessary
condition) 190-1
Bid (c. acc.), because of. Btd rb,..€ivaLy 191; Bid ravra
Toivvvy 190
BLayeiVy to draw through, 8Lrj)(6a>y 187
BiaydavLOSy diagonal, 132
BidpieTposy J7, diameter, 132, 207
StaoTardr, -or, extended: i<f) evy one way, eVt Bvo or
Bixfi, two ways, 122
BLdorrrjpLOy distance, 126, 146-7
BiBovcuy to give. Bodeis, -eLcra, -ev, given, 1 57-8
dtoptcTftdff,^particular statement’ (of a proposition),
160; ^ limiting condition’ (in a problem), 161
Blxa ripLv civy to cut into two equal parts or bisect, 133,
179, 180, 207
'EKCTepor, each (of two). €Ktvripa tS>v FA, FB, 159.

LcraL eKarepa eKaripa, 166, 174, 218
€icj8dXX«i^, to produce. eK^c^Xrja-daxrav, 163
€K 0 €(ris, setting-outy 160
€KK€L(rdaL, to be set out (used as pass, of iKTidevat).
€KK€LCr6G>y I 9 I
iKToSy outside. ^ cktos y<avia, the exterior angle, 187,
188, 200
fidling-‘sbort (in conics, the dlipse)y zx 6
ipTTLTTTeLVy to fall ou or across, 149? *995 iviTrô'ev, 214
cVaXXdf, itemately, adv. used as adjective, at eVaX-
Xa§ yavLaiy the ^temate angles, 199-200
ivTosy inside, within, 130, 190, ^ €vTds...y(av[ay the in-
terior angle, 187, 200, 201
1
EXPLAINED IN THE NOTES 233
eTTL (c. acc.), to (in ‘perpendicular to*=let fall on\
127, 182; (in ‘joined to’), 159; towards (see fieprj)

(c. acc.) on (in ‘set up on’), 127, 184
(c. gen.) on (in ‘construct on’), 157, 173, (= situ-

ated on), 182; in (in ‘in a straight line,’ evOelas),
146, 163, 218
enLêvyvvvaij to join, 205; i-ire^cyxOco, -oaorav (iirl and
acc.), 159, 21
eTTLTreboSy -ov, plane, 123; iir. iirK^dveicLy 1235 ya>vta,
124; arxnp^ 129
eTTL^dveia, surface, 121. iirl'iredos i^r.^ 123
€T€p 6fir)K€s, oblong, 138, 139
cvOcia (fern, of €v 6vs)-, sc. ypappf}^ straight line, 118-
121; eV’ evdcUtSj ‘in a straight line,* 146, 163, 218
cvdvypapposj -ov, rectilineal, 1265 with
ycov/a, 126;
135; alone (in neuter), 217
orxrjpcL,
€<pd 7rT€(rdai, to touch, 124
iipappoêLV {iirl and acc.), to coincide (with); 6<jE>apfio-
feo-dat, to be applied to, 153-4, 167-8, 176
i(p€^s, successively, 126; adv. used as adj., at
yoaviat^ the adjacent angles, 126, 1 81, 185
€(picrTdvaL, to set up. standing on (of a
straight line), 126
^HpiKVKXiov, semicircle, 134
Qecyprfpoy theorem, 156
’lo-dirXevpos, -ov, equilateral, 136

LOTOS, equaL la-ov, evenly, 118-19
l(ro(nce\-qs, -es, isosceles, 136
lardvai, to set up. crradcwra, 126, 184
Kddcroff, rj, perpendicular, 127, 183

KoBiivai, to let fall, 126, 183
K.ai, then, 159
Kara (c. acc.), according to. Kara yvwpova, gnomon-
wise, 127, 183; Kara Kopu(l>rjv, vertically opposite (of
angles), 185; jcctd* b {a-qpo7.ov), the point in (or at')
which, 158
KaraXoiireiv, to leave behind, ra KorakoLiropova^ 152
1 1
Z34 index of GREEK TERMS

KaTCLcricevr}^ construction^ i6o
Keia-Sac, to lie, ii8, 124, 130; to be placed or made,
used as pass, of TiSevm- 165, j 8 i, 189, 193
Kei(r6(i>,
KevTpovj centre, 132, 134; 17 ck row fccvrpou, radius, 147
KXdo-is, breaking or deflection, 125
icXto-tff, inclination, 124, 125
icoivdf, common, 169; koivcll %vvoiai^ Common Notions,
143, 151 ; Koivrj 7rpo(rKeia‘6G> or d^rjpfjcrS&j 184-5
#:opt;qf)]7, vertex. Kara Kopv<f)rfv^ 185
kvkXos, drde, 129
AapL^dv€iv, to take. 169: (7‘d) Xappavopevov^

assumption, 121
Xeyto, I say, 167, 21
XotTTOff, -rj, -01/, remaining, 163
Mcpjy, parts, side, direction. €<p* iKorepa rd /z., 133,
141 ;
pjjddrepa rd p., 141, 200; cttl rd avrd p., 149,
irrl
174, 185, 201, 205: €<!> d p., 150; iiri rd erepa p.,
183 ; €7ri rd B, p., 200A
peraKafi^dvw, to take together. Tcdvrri p.eraXap.âv 6-
pjcvcu, 187, 191
’OîryAvtos, -01/, acute-angled, 137

acute, 127, 128
d|tJs, -tio, -V,
oirorepos, either (of two), 193,218
6p6oyd»vios, -ov, right-angled, 137
opBos, tight. 6p$^ ytovicL, 126; srpdr 6p6ds (yoovias), at
right angles, 18
opos, boundary, 128; definition, ib.
lEapd (c. acc.), alongside of. TrapajSoXecv irapd, to

apply to, 214
TTflpajSdXXetv (TTopd), to apply (to), 2 14
TrapaôXfj, application (of areas), 216; (in conics, a
parabola), ib.
irapaKXdrreiv, to pass by or fall awry, 176-7
7rapd\X7}\6ypappx>£, -ov, formed by parallel lines,
TrapaXXT/Xdypappov (x®^P‘o*')j parallelogram, 140, 206
TrapoXXiyXoj, -ov, parallel, 140; dv tols avroLS irapctkXT}-
\ois, 207
EXPLAINED IN THE NOTES 235
‘irapairXrjpoaiia^ complement^ 213, 214

TrSff, TToo-a, Trap, ‘every’ in sense of ‘any,’ 145, 146,
187, 213
TTcpcdveiv^ to limit. TreTrepacrpUvos, limited, 146, 157,
Tripas, limit or extremity, 117, 128, 129, 174, 190

veparovvy to limit or terminate, 133
irepL (c. acc.), about, 213, 214
TrepiexciVy to contain or enclose, 126, 128, 166
7r€pL<l>€p€ia, circumference (or arc), 129-30, 134
Trepiipcpris, -er, round, circular, 130
Tr€pL<j)eparYpappoSy -ov, bounded by circular lines, 131
jrXevpdy side, 135
TTOLcivy to do or make, oirep edct TroLrja-aii OJE.F.y 156,
l6l; TTOLCLv ycovtasy 149, 184
ffoXX <5 pcLôav, much (= a fortiori) greater, 175
TToXviirXeupos, ~0Vy multilateral, 135
TTopurpaj porism (
= corollary), 186
TTpo^XqpLOy problem, 156
wpos (c. acc.), tOy towards, irpos dXkrjXaSy 124; (in
‘falling on’), 130
(c, dat.), on (a straight line), 181, 185, 192; with
(a straight line), 185; at {3. point), 162, 1855 irpos
rfj pdo-€i, at the base, 168, 169
TTpoaTTLiTTCLV (eVt), to fall on, 130
‘irpocrrtOivaij to add, 152; Trpoaredjjy 152
TTpoTaariSy enunciationy 160
‘Po/xj3o€tSes, rhomboid, 139

poftjSor, rhombus, 139
2rfp.€iovypoint, 113-14
o-KoXi^vdr, -dv, scalene, 137
OTLypd), point, 1 14
crup.Tripacrpay conclusiony i6t
arvpwLTrreiv, to meet, 15O3 200
avvex^Sy -er, continuous, 146
crvvLcrrdvaty to construct. orvarrjo-aa-daLy 157, 1925 tn/v-
LcrravToiy 1775 oweordro), 163; (rvcrradricrovTaLy 1735
avoTaB&a'LVy 190; (rvcrTa$€i<rat, 190
crxvp-O; figure, 128, 129
236 INDEX OF GREEK TERMS
Terpdycovov (neut. of rerpayaivos), a square, I 3 ^> ^39
reTpdrrXevpos, ~ov, four-sided. TerpdTrXevpov {orxVH'o)}
quadrilateral, 135-6, 139
TiBivaLy to place. dicrSoL (mid.), 162
Topffy (point of) section, 186
rpaTreîov, trapezium, 138
TpiycovoSy -oi', triangular, rpiyoavov^ triangle_, 136
TpLw\€vpos, -ovy ftree-sided, 135
Tvyxdvoi^ to happen or chance, rvxov (xrjpeTov, a point
(taken) at random, 169, 180
'Y irepôXrjj exceeding (in application of areas) ; (in conics,

a hyperbola)^ 216
VTTo (c. acc.), under or contained by. rj viro BAF ycavla,
the angle BAC, 166, 167
viroKeiarSaL, to be supposed or assumed (=pass. of
vTTOTiQivai). vTToiceiTaL, is by hypothesis, 196, 202
VTroTCLvetv, to be opposite to or subtend, c. vtto and
acc., 167, 171, 1965 c.acc. 189, 217-185 vTroTeipova-a,
the hypotenuse, 167, 218
viroridivaiy to suppose or assume (cf. hypothesis), 196
Xpoidj colour or skin, Pythagorean name for surface,

122
Xtopiov, space or area, 206
1.
index of proper names

‘Agajois/ 17
Campanus, Johannes, 28,
Amphinomus, 156
295 3 h 32
Amyclas of Heraclea, 1 Carpus of Antioch, i8
An-Naiiizij 16,
17, 22 Cassiodorius, 25
ApoUodorus (6 XoytoTt- Censorinus, 25
Ko?), 219 Cicero, 24-25, 219
Apollonius of Perga, Clairaut, A. C., 208
4, 8,
14515718, 116,125, 152, Clavius (Christoph Klau),
165, 167, 180-1, 192,216
33 .
Archibald, R. C., Cleonides, 10
7
Archimedes, 3, 4, 5, 15, Commandinus, F.,
29, 32, 34
pi, 124, 140, 167 Cunn, Samud, 34
Aristaeus, 8
Aristotle, ii, Dee, John, 32, 38
15, 22, 114,
115, 116, 123, 125, 126, De Morgan, A., 164, 209
128, 129, 131, 136, 137, Dinostrattts, ii
139 ) 153 ) m, 172) 201 ,
220 Epicureans, 15, 189-90
Athelhard of Bath, 28, 29, Euclid, 3-5 and
3^536 Eudemus, 18, 19, 192, 197,
Athenaeus of Cyzicus, 1 1
204, 215
Athenaeus (of Naucratis), Eudoxus of Cnidos, 4, 10,
219 11,12-13, 19
August, E. F., 36 Eutodius of Ascalon, 17
Autolycus of Pitane,
9
Friedlein, G., 25
Bacon, Roger, 28, 36
Balbus, 25 Gauss, C. F., 121
Ball, W. W. Rouse, 39 Geminus, iS
Barrow, Isaac, 33, 38 Gherard of Cremona, 27,
Besthom, R. 0 ., 26 36
Billingsley, Sir H., 31-32 Gow, James, 39
Bjornbo, A. A., 27 Gregory, David, 9, 34
Boetius, 25 Grymaeus, Simon, 31
Briggs, Henry, 38
Al-Hajjaj, 26
Camercr, J. G., 36 Halley, Edmund, 38
238 INDEX OF PROPER NAMES
Hariin ar-Rasbid, 26 Nicomachus of Gerasa, 25
Hauber, C. F., 36 Nipsus, Marcus Junius,
Heath, T. L., 36 198
Heiberg, J. L., 6, 9, 21, 26,
34, 36, 21 1, 212 Oenopides of Chios, 183,
Hermotimus of Colophon, 192
II Oughtred, W., 33
Heron of Alexandria, 16,
18,22, 121, 124, 140, 196 Pamphile, 204
Hipparchus, 5 Pappus, 7, 8, 17, 18, 170-
Hippocrates of Chios, 10 I, 222-$
Hudson, John, 34 Peyrard, F., 21, 35, 36
Hypsides, 14 Philo of Byzantium, 177
Plato, 10, 115, 1 18, 1 19,
lamblichus, 23 120, 123, 128, 129, 131,
Ishaq b. Hunain, 26, 27 i35i.»36,i37
Isidorus of Miletus, 14 Playfair, John, 202-3
Plutarch, 125, 219
Jan, C, 9 Porphyiy, 17
Posidonius, 18, 129, 141
Keill, John, 34 Proclus, 3, 7, 17-20 and
passim
Legendre, A. M., 121 Pseudo-Boetius, 25-26
Leibniz, 121 Ptolemy I, 4
Leon, 10 Ptolemy, Claudius, 9, 18,
Leonardo of Pisa (Fibon-
297 37
acci), 29, 30 Pythagoras, 5, 204, 219-
21
AI-Ma’mun, 26 Pythagoreans, 12, 13, 22,
al-Mansur, 26 139, 187, 204, 215
Marcellus, 3
Marinus, 10 Saccheri, Gerolamo, 121,
Martianus Capella, 25 141
Menaechmus, ii,
156 Savile, Sir H., 34, 38, 190
Menelaus of Alexandria, Scarburgh, Edmund, 134
196 Sextus Empiricus, 23
Menge, H., 6, 9, 34, 36 Simplicius, 17
Simson, Robert, 6, 20, 21,
An-Naiiizi, i6, 17, 22 325 34-35, i 33> ^74,
Nasiraddln at-fusi, 27 194, 208
Newton, 38 Skiver, Sir Thomas, 31
INDEX OF PROPER NAMES 239
Smith, David Eugene, 199 Theon of Sm5rnia, 124

Speusippus, 155 Theudius of Magnesia, ii,
Stobaeus, 4 171
Syrianus, 18 Thucydides, 210
Tonstall, Cuthbert, 31
Tacquet, Andre, 33, 34
Tannery, Paul, 9
Wallis, John, 38
Thabit b. Qurra, 26-27
Whiston, William, 34
Thales, 133, 154, 171, 197-
Williamson, James, 35
8, 204 Woepcke, F., 7,
Theaetetus, 14, 19 17
Theodoric, 25
Theodorus of Cyrene, 14 Zamberti, Bartolomeo, 30-
Theon of Alexandria, 9, 31
20-21, 23-24, 35 Zeno the Epicurean, 18
CAMBRIDGE : PRINTED BY
J. B.
PEACE, M.A.,
AT THE UNIVERSITY PRESS

.Euclid in Greek Book I With Introduction and Notes Text

Uploaded by

Copyright:

Available Formats

.Euclid in Greek Book I With Introduction and Notes Text

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

.Euclid in Greek Book I With Introduction and Notes Text

Uploaded by

Copyright:

Available Formats

EUCLID IN GREEK

CAMBRIDGE UNIVERSITY PRESS

LONDON : FETTER LANE, EC

ALL RIGHTS RESERVED

SIR THOMAS L. HEATH

might be thought) of one inutility on another. But,

complete education. Generation after generation of

Euclid in the original, the two elements will help

the particular tongue. The advantage to the student

Secondly, from the point of view of learning geo-

between propositions which are nearly allied, e.g, as

converses. I cannot but think that the formality

was in the days long ago when most people were

to see the actual language in which the old Alex-

than the fundamentals of geometry to make the

worth while, whatever the schoolboy’s after career

WORKS OTHER THAN THE ELE-

EARLIER WRITERS OF ELEMENTS lo

CONTENTS OF EUCLID’S ELE-

THE ELEMENTS IN GREECE . IS

CHANGES IN THE TEXT . . 20

LATER HISTORY OF THE ELE-

GREEK TEXT OF BOOK I . 41

INDEX OF GREEK TERMS . 231

INDEX OF PROPER NAMES . 237

lised nations of the West, can hardly have been

that can be assumed is that he received his mathe-

assigned to the reign of the first Ptolemy (306-283

a little earlier. He taught and founded a school at

and domiciled at Damascus. But the Arabian

WORKS OTHER THAN THE

where the authors, although basing themselves on

flaw in the argument, to refute it, and, by dint of

Greek, and the full text with the commentary by

straight lines be given, then shall each of them

expressly attributed to Euclid and corresponds to

other figures either like or unlike; thus a triangle is

divided into triangles or into a triangle and a quad-

restoration, but none that can be regarded as dis-

posing of the subject. This much is clear, that the

‘The four books of Euclid’s Conics were completed

appeared in Gregory’s edition o£ Euclid, but it has

Teubner, 1895). The Catoptrica (theory of mirrors)

of expressions and at aU events one proof which are

EARLIER WRITERS OF ELEMENTS

Hippocrates of Chios (fl. in the second half of the

From the account of these quadratures we can judge

numerous and more serviceable. The geometrical

naturally silent as to his relation to his predecessor;

CONTENTS OF EUCLID'S ELEMENTS

lines; and it contains a few simple problems of con-

The second section beginning with i, 27 establishes

the theory of parallels, leading to the proposition

equal to two right angles. The third section begins

Book II carries the subject up to the transformation

kinds, both, commensurable and incommensurable.

equal to one given rectilineal figure and similar to

irrationals, and is, in form, the most finished of all

the Books. Euclid here owed much to the original

its contents to Book i. Book xii to Books ii and vi,