V*1

A TEXTBOOK OF

GENERAL EMBRYOLOGY

WILLIAM ErKELLieOTT

PROFESSOR OF BIOLOGY, COLLEGE OF THE CITY OF NEW YORK

NEW YORK
HENRY HOLT AND COMPANY

COPYRIGHT, 1913.

BY
HENRY HOLT AND COMPANY

August, 1927

K

PREFACE

General embryology should occupy an important place in the
collegiate study of biology. In no other connection are the
essential phenomena of life better illustrated, in no other form
are they more readily appreciated. The facts of embryology
lead directly to the great problems of the science of biology as
it exists to-day, and many fundamental biological conceptions
either are directly connected with, or are illuminated by, the
study of the early phenomena of individual development.

The author's experience has clearly indicated that the subject
has this value as a collegiate study. Indeed, the book is the
direct outgrowth of such experience, and it has, in substance,
been in use as such a text for several years. In its present form
it is hoped that it will be found useful to the student who is
endeavoring to comprehend the general principles of the science
of life, as well as to the student preparing for the professional
study of some field of biology or of medicine.

Its design as a textbook, rather than as a handbook, accounts
for certain characteristics. The topics considered have through-
out been approached from the standpoint of their general
biological relations, and in the selection of the facts mentioned
and the topics discussed, as well as in the style and method of
presentation, the student has been first in mind. The arrange-
ment of the subject matter in two sizes of type may prove
useful for those undertaking a brief course. In a few instances
this has involved slight repetition, but repetition is not always
a pedagogic evil.

At the end of each chapter will be found a list of references to
literature. Usefulness to the student has been the only
criterion in determining the admission of titles to these lists.
Consequently there will be found titles of works of historical
importance, of recent works containing contributions of impor-

iii

iv PREFACE

tance or representing present tendencies in research, and of
papers containing extensive literature references, valuable
illustrations, or general summaries. As far as possible the lists
contain references to works presenting both, or several, sides
of mooted questions mentioned in the text. There will also be
found, in nearly every instance, the titles of papers from which
illustrations may have been taken.

To a large extent the figures have been redrawn, from the
original sources, for this work: it is a pleasure to notice the
uniform courtesy with which authors have granted permission to
make this use of their illustrations. The following special debts
are gratefully acknowledged: to Prof. Edmund B. Wilson and
The Macmillan Company, for cliches and for permission to
copy a considerable number of illustrations in their "The Cell
in Development and Inheritance"; to Prof. Gary N. Calkins,
The Macmillan Company, and Lea and Febiger, for cliches and
for permission to copy certain illustrations in their "The
Protozoa'7 and "Protozoology"; to Prof. Ulric Dahlgren, Prof.
William A. Kepner, and The Macmillan Company, for permis-
sion to copy certain illustrations in their "Principles of Animal
Histology"; to Prof. J. W. Jenkinson and the Delegates and
Secretary of the Clarendon Press, for cliches from their " Experi-
mental Embryology"; and finally to Herr Gustav Fischer and
to the several authors, for cliches and for permission to copy or
otherwise make use of illustrations from Korschelt and Heider's
"Lehrbuch," Oscar Hertwig's "Handbuch," Doflein's "Proto-
zooenkunde," and Ziegler's "Lehrbuch." In every instance
specific reference, both to the immediate and the ultimate
sources of the figures borrowed, is made in the legends. I
desire also to acknowledge my indebtedness to the authorities
of The Johns Hopkins University, for the use of valuable
library facilities.

W. E. K

BALTIMORE, MD.,
March, 1913.

CONTENTS

CHAPTER I

PAGE

ONTOGENY 1

CHAPTER II ^

I/ THE CELL AND CELL DIVISION 31 S/

CHAPTER III

/- THE GERM CELLS AND THEIR FORMATION 85

CHAPTER IV
MATURATION 131

CHAPTER V
FERTILIZATION 164 }C_

CHAPTER VI
CLEAVAGE 219

CHAPTER VII

THE GERM CELLS AND THE PROCESSES OF DIFFERENTIATION, HEREDITY,

AND SEX DETERMINATION 260

CHAPTER VIII

THE BLASTULA, GASTRULA, AND GERM LAYERS. MORPHOGENETIC
PROCESSES 329

INDEX . 367

TEXT-BOOK OF
GENERAL EMBRYOLOGY

CHAPTER I

ONTOGENY

LIVING organisms come into existence only as the offspring
of preexisting living organisms of the same kind or species.
Aristotle's belief that eels were generated from mud and slime
is represented to-day, in many youthful minds, by the firm
conviction that a horse hair, if only kept long enough in water,
will surely "turn to a worm." From the time of Redi the
belief that the living might be generated from the wholly
non-living gradually became restricted, in its application, to
lower and still lower groups of organisms. For a long time it
remained applied only to those forms at the lower limit of the
living — the bacteria. From this position the belief that " spon-
taneous generation" of organisms occurs nowadays, was
finally driven by the brilliant demonstrations of Pasteur and
Tyndall that even these simplest and smallest of organisms
arise only from preexisting living organisms of the same kind.

This property of producing new, specifically similar individ-
uals is one of the few really distinctive characteristics of living
things, and since the newly produced resemble closely the
parent form, we speak of the property as Reproduction. The
fact that at corresponding ages offspring resemble their parents,
is the fact of heredity. But when these offspring are first
distinguishable as separate and new individuals they bear little
or no visible resemblance to the adult organisms producing
them. This resemblance appears gradually, as the result of a
long series of processes, complex and often very special, in-
volving changes in structure, function, and form, only at the
conclusion of which has the new - organism reproduced, more

1

2 GENERAL EMBRYOLOGY

or less precisely, the form and other characteristics of its
parents. The facts regarding all of these processes of develop-
ment, of the external and internal changes in form and structure
of the new organism, of the complex chain of processes leading
to its first formation, and of the role of external factors through-
out all of these, constitute the science of Embryology. We
may define Embryology briefly then, a^_Jhe__gcience__pl_the
genesis of the adult organism.

As a general introduction to this subject we may first sketch
in outline the broad features of reproduction among the lower
animals, mentioning but a few of the almost infinitely varied
forms which this process assumes. It should be understood in

FIG. 1. — Simple fission in Amoeba vespertilio. After Doflein. A. Normal
vegetative form. B. Commencement of fission ("biscuit form"). C. Fission
nearly completed; separation of daughter cells, z, cells of an Alga, Zoochlorella.

advance that the series of reproductive processes, of increasing
complexity, to be outlined, has little if any phyletic significance;
this arrangement is made for comparative purposes alone.

The simplest and apparently the most primitive mode of re-
production is that known &s fission, characteristic of the single-
celled organisms, the Protozoa and Protophyta. In the case of
simple or Unary fission a separation of the nuclear material of
the cell into two separate masses is followed by a constriction

ONTOGENY 3

of the cell body into two bodies, each of which may then form
parts corresponding with those carried away by its sister
cell, and finally develop into a creature resembling the original
organism (Fig. 1). In many unicellular forms, particularly
among the Sporozoa, a process of multiple fission or brood
formation occurs. This is frequently preceded by growth of
the organism to an unusual size; then an uninterrupted series
of simple fissions, without intermediate growth or development
on the part of the daughter cells, results in the formation of a

cy

FIG. 2. — Multiple fission in a parasitic Infusorian, Holophrya multifiliis.
After Hatschek. A. Normal vegetative individual. B. Cyst containing the
products of repeated binary fission; some of the zoo spores are shown leaving the
cyst. C. One of the zoHspores, enlarged, cv, contractile vacuole; cy, cyst;
m, mouth; ma, macronucleus; mi, micro nu cleus ; t, zoospores.

large number of small organisms. These usually remain
associated, frequently within a cyst formed by the parent cell,
until the whole process of fission is completed (Fig. 2). In
other cases the nucleus, or nuclei, alone may divide, either
successively or simultaneously, into a large number of separate
nuclei; the cytoplasm immediately surrounding each nucleus is
then cut out as a separate cell, so that the organism appears to
fragment simultaneously into a large number of small daughter
organisms (Fig. 3) . The number of new individuals formed in
this way may vary from four, as in some Infusoria, up to as
many as several hundred in many different forms, especially
among the Sporozoa. When the number is large the process

4 GENERAL EMBRYOLOGY

is more frequently termed spore formation or sporulation and
the products are known as swarmspores or zodspores. Tech-
nically the term sporulation or sporogony (metagametic division)
is used only when this process of multiple fission occurs sub-
sequently to a process of cell fusion. If no such process of
cell fusion or conjugation has preceded, the process is termed
schizogony (agamogony), and the products of the division are
called schizonts (agamonts).

E

FIG. 3. — Multiple fission (schizogony) in the Sporozoan, Coccidium schubergi.
After Schaudinn. A. Entrance of organism ("sporozoite") into epithelial cell
of host. B, C. Two stages in growth. D. Multiple division of nucleus. E.
Daughter nuclei superficial, cytoplasm still undivided. F. Fission of the super-
ficial cytoplasm surrounding the nuclei, leaving a central undivided mass. G.
Free schizonts ("merozoites").

Reproduction by analogous processes also called fission,
simple or multiple, is only occasional among the many-celled
animals. In a few such forms the entire body separates as a
unit into two or more organisms. Organs and tissues in the
plane of separation are divided, and each of the daughter or-
ganisms then regenerates parts corresponding to those carried

ONTOGENY

off by the other, forming new individuals smaller than the
parent form, but otherwise similar to it (Fig. 4). This fission
in the Metazoa is not really comparable with the similarly
named process in the Protozoa; it represents a special acquire-
ment and is usually, though not always, associated with other
more complex modes of reproduction. Normal fission is
known to occur in many genera among the Ccelenterates, and
less frequently among the Porifera, Platyhelminthes, Annulata,
Bryozoa, and Echinoderms.

It is obvious that in reproduction by fis-
sion the individuality of the parent organism
is lost in the act of giving rise to the new
individuals, although none of the substance
of the original organism perishes in the act.

FIG. 4. — Fission in Metazoa. A, B. Two stages in the transverse fission of
the Actinian, Gonactinia prolifera. From Korschelt and Heider, after Blochmann
and Hilger. C. Successive transverse fissions in the Platy helminth, Micro-
stomum lineare. After L. von Graff. /, //, ///, mark the levels of the successive
fissions; a fourth fission also is indicated, p, p, pharynges.

In the fission of Metazoa usually many of the structures of the
parent are simply transferred to the new organisms with com-
paratively little differentiation of parts anew, out of a visibly
undifferentiated condition. In the Protozoa this may or
may not be the case. In some of the highly organized
Ciliata most of the structural differentiation disappears just
previous to fission, after which each daughter cell differ-
entiates a typical form and structure anew (Fig. 5). In
other Protozoa there is a considerable transference of char-
acteristics accompanied by a lesser amount of regeneration or

6

GENERAL EMBRYOLOGY

redifferentiation. After multiple fission or sporulation, the
end products of the process are commonly minute and visibly
quite unlike the adult form; this they come to resemble only
through growth and differentiation, that is, through processes
of true development.

A reproductive process closely
allied to fission is the familiar proc-
ess of budding. Here one or
several small outgrowths or ubuds"
are produced from some portion of
the parent organism, and develop
into forms resembling the parent,
either before or after becoming de-
tached. This process frequently ap-
pears as a sort of unequal fission
and may indeed rightly be regarded
as such; but usually so great is the
disparity in size, as well as in extent
of differentiation, between parent
and buds, that the processes are prop-
erly distinguished. Budding occurs
in many Protozoa (e.g., EpJielota
(Fig. 6, A), some Rhizopoda), and
is quite frequent among the lower
Metazoa, particularly in colony form-
FIG. 5.— Binary fission, in ing species. It occurs among the

the Infusorian, Euplotes harpa. pnrl'fpro fVplpntprfltfl Plfltvhplmin-
After Wallengren. Division rc 6ra> ^Q Grata, J aiyni

thes, Annulata, Bryozoa, and Tuni-
cata (Fig. 6, B). In budding there
is ordinarily very little direct transference of structures,
the development of the bud occurring after it has been
completely delimited as a comparatively undifferentiated mass;
its development may then be almost complete before the
separation of the two organisms. Nor does this process involve
necessarily the loss of identity of the parent, which may con-
tinue to live and produce buds for a considerable period.
In a few Protozoa, fresh water sponges, Trematodes and

Wallengren.
stage immediately before the
separation of the daughter cells.

ONTOGENY 7

Bryozoa, internal buds are formed within the parent cell or
body (Fig. 7). These become free and develop into new indi-
viduals ordinarily only after the death of the parent body.
Although the formation of these gemmules (Porifera), OYstato-
blasts (Bryozoa) suggests that form of reproduction character-
istic of the higher Metazoa, i.e., through internally formed germ
cells, it will appear later that the two processes are not at
all to be compared.

FIG. 6. — Reproduction by budding. A. In the Infusorian, Ephelota. From
Calkins, "Protozoa." N, branched macronucleus extending from the parent
cell into each bud. B. In the Tunicate, Doliolum. After Neumann. Ventral
outgrowth of a phorozooid. b, reproductive buds of various ages; g, germinal
knob; s, stalk.

In some of the shelled Rhizopods (Euglypha, Arcella) a
combination of the processes of budding and fission occurs.
In this process, called bud-fission, about one-half of the proto-
plasm flows outside the original shell, which these forms possess,
and secretes a new shell like that of the parent form; or the
rudiments of the new shell may be formed before the appear-
ance of the bud. The cell body then divides and the two similar
individuals may either separate completely, or they may
remain in contact, forming after repeated bud-fission, an
aggregate of related though distinct organisms.

In many of the Protozoa multiple fission may be incomplete,

8 GENERAL EMBRYOLOGY

or the daughter cells, not scattering as separate individuals,
may remain associated during division and growth, forming a
colony of organically related cells which as a whole is to be
regarded as the individual organism. Among nearly all of
these colonial or compound forms the entire colony, or cceno-

F

FIG. 7. — The formation and development of the statoblast, in the fresh water
Bryozoan, Cristatella. A. After Braem. Others, after Verworn. A. Longi-
tudinal section through the funiculus, showing the relations of the statoblast.
B-E. Optical sections of stages in the development of the statoblast. F. Sec-
tion showing rudiment of embryo, e, superficial ectoderm; i, inner layer of
funiculus (ectoderm) ; o, outer layer of funiculus (endoderm) ; s, rudiment of
statoblast.

bium as it is called, is not later involved as a unit in reproduc-
tion, but the component cells may individually undergo
fission leading to the formation of new colonies. In forms like
Pandorina or Platydorina all the cells are alike, and each repro-
duces, so that there may be as many new colonies formed as
there are cells in the original colony (Fig. 8). In other forms
the power of reproduction is limited to certain cells, termed

ONTOGENY 9

gonidia, and there results a sharp distinction between repro-
ductive and vegetative cells (Fig. 9). Thus in Pleodorina, of
the thirty-two cells forming the colony, four anterior cells are
purely vegetative, the remaining twenty-eight are gonidial;

FIG. 8. — Reproduction in the colonial Flagellate, Pandorina morum. From
Hertwig, after Pringsheim. /. Free-swimming vegetative colony. II. Daugh-
ter colonies formed by the fissions of each individual in a colony similar to the
preceding. ///. Gamete formation in a colony formed as above. IV, V, VI.
Stages in the conjugation of gametes to form a zygote (VI). VII. Zygote after
growth to full size. VIII, IX. Production of swarm spores by the zygote.
X. Vegetative colony formed by fissions of the swarm spore.

while in Volvox, where the fully developed colony may include
as many as twelve thousand or more individuals, only five to
fifty cells, scattered through the posterior half of the colony are
gonidial, the remainder being vegetative (Fig. 10).

10 GENERAL EMBRYOLOGY

Among most of these colonial Flagellates, and indeed in
many other Protozoan groups, at irregular intervals these
gonidial cells lose the property of directly giving rise to new
colonies, and become very highly differentiated in structure
and in behavior. These highly modified gonidia are termed
gametogonidia, the ordinary gonidia being then given the name

FIG. 9. — Colony of the Flagellate, Pleodorina illinoisensis. Lateral view.
After Kofoid. A, anterior; g, gonidial (reproductive) .cells, v, vegetative
cells.

of parthenogonidia. The gametogonidia form specialized cells
termed gametes which must meet and fuse in pairs, i.e., con-
jugate, before reproduction may proceed. In the simplest cases
these two gametes are nearly or quite alike. In most cases,
however, gametes of two very unlike forms are produced by
different gametogonidia. These must conjugate in pairs of
unlikes before reproduction is possible. A series of forms
illustrating the progressive differentiation of the gametes, or
germ cells, is described in Chapter V. For the present we may
merely notice that in such a colonial form as Volvox, under
certain conditions, the parthenogonidia cease reproducing;
certain of them (odgonidia or ovaries) enlarge, and each differ-
entiates a large non-motile cell, the ovum, or odsphere, or
macrogamete, while others (spermagonidia, or spermaries) after
enlarging similarly, divide repeatedly, each forming a large
number, often as many as one hundred and twenty-eight,

ONTOGENY

11

small actively motile cells termed sperm cells or microgametes.
A single sperm from one colony then meets and conjugates
with a single ovum from another colony, forming thus a single

C

FIG. 10. — Reproduction in the colonial Flagellate, Volvox. A. After Prings-
heim, B-E, after Klein and Schenck. A. Young colony showing distinction
between, s, somatic cells, and g, reproductive cells (parthenogonidia). B. Older
colony showing, p, parthenogonidia, o, oSgonidia, sp, spermagonidia in various
stages of formation. C. Spermagonidium consisting of thirty-two spermagam-
etes, seen from the side in D. E. Spermagametes more highly magnified.

cell or zygote, which through continued fission gives rise to the
individuals of a new colony.1

1 Useful diagrams of- these forms of reproduction and gamete formation
are given in HEGNER, "An Introduction to Zoology," New York, 1910, pp.
112-115.

GENERAL EMBRYOLOGY

Reproduction following gamete formation and fusion (syn-
gamy) is commonly known as "sexual" reproduction, while the
term "asexual" is applied to modes of reproduction which do
volve the fusion of gametes. It is now clear, however,
that several different and unrelated processes are included under
each of these heads. Thus as "asexual" must be included
such diverse modes of propagation as budding, fission, brood
formation, parthenogenesis, development from spores, and so
forth; and "sexual" reproduction would also follow many
forms of syngamic fusion. Two further conditions tend to
rob these terms of their precise significance; these are, the
existence of transitional conditions, some of which have been
mentioned above, and the doubtfully essential character of
the primary relation of syngamy to reproduction. These
useful terms are to be retained only as convenient though
inexact expressions, and are to be used in much the same way
that we still employ the convenient words "vertebrate" and
"invertebrate." Even so, we may avoid any unintentional
implications by substituting the more exact terms amphigony
and monogony for sexual and asexual respectively.

Without suggesting the idea of direct relationship among any
existing forms, we may say that it is but a short step from the
processes of gamete formation and reproduction among the
colonial Protozoa, to the mode of reproduction characteristic
of the Metazoa, which after all may be considered highly
organized cell colonies, y One of the more distinctive as well as
more obvious characteristics of the Metazoa is the structural
and functional differentiation of large groups of cells as tissues,
each variety of tissue performing, in the animal economy,
chiefly one function, such as conduction, support, or excretion.
Among these various tissues is the reproductive or germinal
tissue^_which, in all save a few of the lower Metazoa, is always
in the form of definite organs, the gonads^ The distinction,
suggested by some of the colonial Protozoa, between the repro-
ductive and the vegetative tissues was probably the earliest of
those "physiological divisions of labor" which involved tissue
differentiations in the Metazoa. The essential cells of the

I

ONTOGENY 13

gonads, which correspond functionally to gametogonidia, divide
repeatedly, the products of their multiplication being, with few
exceptions, ultimately thrown outside the body as the highly
modified germ cells or gametes. In all the Metazoa these germ
cells are of two quite unlike forms; the ova or "eggs" are large
and ordinarily non-motile, while the spermatozoa or sperm cells,
are small and actively motile. These Metazoan germ cells
clearly correspond with the ova or macrogametes and the sperms
or microgametes of certain of the Protozoa. The gonads
forming the ova and Spermatozoa are known as the ovaries
and testes respectively. In most of the Metazoa germ cells of
only a single kind are formed by a single organism, a condition
which leads to the primary distinction of sex; individuals
forming ova are called females, those forming spermatozoa,
males. *

& In comparatively few kinds of animals does a single individual
normally possess gonads of both types, and thus become capable
of forming both kinds of germ cells, either simultaneously or
successively. Such a condition is known as hermaphroditism;
it occurs chiefly among the lower Metazoa,/ such as the Platy-
helminthes, Nemertea, some Annulates and Tunicates, and
less frequently among the Molluscs, Echinoderms, Bryozoa,
Brachiopoda, and Crustacea. Among the Chordata normal
hermaphroditism is found only rarely (some Teleosts) , though it
may occur as an abnormality in any group. In a few animals
a special form of hermaphroditism occurs, where a single gonad
may produce first sperm and later ova, a condition known as
protandry and found in some Nematode worms and in the
Cyclostome, Myxine, for example. The process of forming
first ova and later spermatozoa, known as protogony, is very
rare among animals.

In the reproduction of all except a very few of the Metazoa,
the initial phase is the union of an egg cell and a sperm cell;
this process is known as fertilization or syngamyj and the double
cell thus formed, which is called the'^gofe or odsperm, then
gives rise directly to the new individual. Not all of the germ
cells formed by an organism actually happen to give origin to

14 GENERAL EMBRYOLOGY

new organisms, although any may do so. But with infrequent
exceptions, where unusual methods of reproduction occur at
times, a few of which have been noted above, new Metazoan
individuals arise only from the union of two germ cells.
C2P The substance which forms the reproductive cells or gametes of
an organism is called the germinal substance, or briefly, the germ.
This is visibly distinguishable at a very early period in the
existence of the new organism, from that material which is to
form all of the remainder of the organism, in distinction known
as the somatic tissue or soma, or simply as the body. Among
the higher forms these two kinds of substance — germ and soma
— have very different histories and fates: According to the
theory of Gejinjnjil^C^ elaborated by Weismann, the

germ represents or contains an organic substance which has
Ibeenlnjr living^ state since the beginning of lite, andjwhlcTi
must continue in this state,in some form, as long as living
thmgsjshall be produced. The soma, on the contrary, is
thought to be built up around the germ, anew and under its
influence, in each generation of organisms. Upon the death of
the individual it is destroyed completely as living substance;
somatic cells finally leave no descendants. Thus in species
which reproduce by this method the soma or body is wholly
temporary, while the germ may properly be said to be poten-
tially ever enduring. For while actually the greater part of the
germ substance formed in an organism is destined to perish,
either before the body or with it, or at any rate with the race,
some germ must always remain, producing the generations of
the future. The essentials of this idea are expressed in the
accompanying diagram (Fig. ll)/

On account of its usefulness the value and significance of the
distinction between germ and soma are frequently over-
emphasized. In many organisms the distinction can scarcely
be drawn at all, for under certain conditions, either normal
or unusual, cells which are evidently " somatic" may take
on reproductive characteristics and function as germ cells.
Many such -cases are known among animals, and among the
plants reproduction from somatic tissues and cells is very

ONTOGENY

15

common, indeed in some it appears to be the normal method
of reproduction.

Among the Protozoa this familiar distinction between germ
and soma cannot be drawn at all. In these simple forms the

o o o o\A

FIG. 11. — Diagram illustrating the theory of Germinal Continuity. A, B, C,
represent successive generations; g, gamete (sperm cell) produced by another
organism; z, zygote. White circles indicate successive cell divisions within the
somatic tissues, the existence of which terminates with the organism of the given
generation. Solid black indicates the germ. Dotted circles indicate gametes
which may perish, or may unite with those of another organism.

process of reproduction is not so directly associated with the
processes of gamete formation and fusion. Nearly all non-
colonial Protozoa, while in the so-called vegetative state, have
the power to reproduce by fission, so that the plasm of these
cells is both germinal and somatic in the Metazoan sense. It is

16 GENERAL EMBRYOLOGY

only at irregular intervals that the individuals ordinarily
multiplying by fission, lose this property as well as their
vegetative characteristics, become specialized as gametes, and
require to undergo syngamy, later resuming their duplex
vegetative and reproductive character.

While it might be misleading to say that the reproductive
cells of Metazoa are, like the Protozoa, both germ and soma, yet
it is quite true that in these germ cells we have a substance
which produces both germinal and somatic tissues. In a
sense we are hardly justified in saying that the soma is built up
anew in each generation, while only the germ has a continuous
existence. ^The germ cell is potentially soma as well as germ,
and for a time during the early development of the organism
there is no visible distinction; this distinction occurs very early
in the development of a few forms, but in most organisms not
until a considerable number of cells has been formed. In
development the germ cells give rise to other cells like them-
selves (germ) and to cells unlike themselves (soma) and we
may regard the " unlike" as "new."

The common conception of the life of a species as a succession
;of generations of individuals linked together by the germ,
/while superficially true, leads to a fundamentally erroneous
/ point of view. The fertilized germ cell is just as much the
individual organism as the matured individual is. The species
is no more a succession of somas than it is a continuous germ.
It is not the function of the germ to provide links between
successive generations of "organisms" or somas, any more than
it is the function of the soma to insure the continuity of the
germ, and to provide materials for its increase and means of its
dispersal. We should recognize that the essential continuity
between successive generations is, after all, not continuity of
plasm but "continuity of organization."

The term reproduction, strictly speaking, does not mean
quite the same thing among Metazoa and simple Protozoa.
Among the Protozoa the formation of free daughter cells, by
fissions of the zygote or its descendants, constitutes repro-
duction. Among the Metazoa the corresponding fissions of the

ONTOGENY 17

zygote and its daughter cells are not considered in themselves
reproductive processes, but as steps (cell divisions) in the
building up of the whole new individual, as but one phase in
the general process of reproduction.

Some physiological reorganization of substance, such as
ordinarily results from the intermingling of the plasmas of two
individuals or lines, seems a necessity for the continued exist-
ence and reproductive activity of most organisms. In some
Insects (Aphids, etc.), Rotifers, Crustacea, and other forms,
reproduction occurs normally, through long periods, without
any such syngamic fusion, the new organisms developing from
single unfertilized ova (parthenogenesis). While this condition
is, in these cases, clearly derived from the normal, yet it seems
to illustrate the non-essential relation of syngamy and repro-
duction. In such forms syngamy does occur under certain
conditions or during certain periods in the life cycle of the
organism. For example, the difficult conditions of winter or
drought may be successfully withstood by the organism while
in the form of the zygote.

So too in many, perhaps most, Protozoa, reproduction or
fission proceeds normally and for long periods without fertili-
zation or conjugation. The process of conjugation is opposed
to reproduction and may actually inhibit it for a time. Here
it appears frequently to be associated with the onset of con-
ditions unfavorable to the existence of the organism in its
vegetative condition.

It seems, therefore, that the processes of fertilization and
reproduction may not be essentially related, and that the
intermingling of the plasmas of two individuals is related
directly to phenomena other than reproduction. Such a
modification of substance as results from fertilization, however,
may be essential to continued existence, and it is certainly true
of most Metazoa that such a plasmic fusion is an organic
necessity. In the simple Protozoa this may be accomplished
at any time by the fusion of two individuals in the form of
gametes. In the Metazoa, however, it is obvious that this
necessary intermingling of substance occurs only when the

18 GENERAL EMBRYOLOGY

organisms are in the form of single cells, i.e., gametes. And
thus it comes about that in the many-celled animals, reproduc-
tion and syngamy are so uniformly associated, and while these
processes may not have been related primarily, in some in-
stances are not even at present, yet now they have come to
be so related in the vast majority of Metazoa, that fertilization
actually appears as the first and most important step in the
whole chain of reproductive events.

The actual processes involved in' the formation, from the
zygote, of the mature Metazoan individual are extremely
complicated and diverse, but they are for the most part reducible
to three fundamental general processes. We must leave aside,
for the present, the causal or directive processes which, though
probably the essentials of development, are still obscure and
little known. * The grosser external phenomena of development
are, essentially, growth, cell division, and differentiation. The
living germ is contained within the limits of a single cell, often
of minute dimensions and only slightly differentiated visibly;
the mature organism consists of an enormous number of cells,
comprising a considerable mass, and exhibiting various degrees
of differentiation in diverse directions. The transition from
one of these states to the other is a gradual process, proceeding
by minute steps; yet it is convenient to consider the whole life
history of an organism as a succession of phases, each with
some chief characteristic.

First, complex processes occur within the gametes or germ
cells themselves, concerning chiefly their nuclei, as a result of
which they come to have a constitution quite unlike that of the
somatic nuclei. These preliminary events we group under the
term gametogenesis, or odgenesis and spermatogenesis in the ova
and sperm cells respectively. Then normally follows fertiliza-
tion or syngamy, the fusion of the two gametes, derived from two
different organisms, into a single cell which is the "new"
organism. Through fertilization a typical nucleus is recon-
stituted in the zygote, and there follows a period of rapid cell
multiplication which is called the period of segmentation or
cleavage. During cleavage are formed the cellular elements

ONTOGENY 19

which are to be built into the structures of the simple embryo,
and various differentiated substances of the egg are segregated
among different groups of cells. Following this are the phases
of blastula formation, when the cells become arranged in a
definite layer, and then gastrula formation when the cells are
rearranged into two definite layers. Then comes the period of
embryo formation, when the cells of the layers are moulded into
the earliest beginnings of the chief systems and organs, blocking
these out in the simplest manner. During this last phase
growth becomes very rapid, accompanied by continued cell
division, no longer termed cleavage, and as the formation of
organs becomes more complete and more particular, the embryo
increases in bulk and dimensions. This period of embryonic
development may occupy a long time, and usually leads to
the formation of an organism which is capable of leading an
-independent life, either as a larva or as a form closely resem-
bling the adult, except in size. Finally, accompanied by con-
tinued growth, the last phases of development appear as cellular
differentiation becomes more complete, and the organism begins
to assume more fully the characteristics of its parents. When
the reproductive tissues become functional as such, the animal
is considered mature and its development complete, although
in a true sense development is never entirely completed, for
the form of the organism never becomes definitely fixed, and
cellular differentiation seems never to cease during the life of
the organism.

It is important to remember that all of these phases of
development are continuous and more or less overlapping, and
in all of them, excepting perhaps the earlier, where the impor-
tant changes concern chiefly the structure and the composition
of the germ nuclei, the three processes — growth, cell division,
and differentiation — are going on together. Yet in general it is
clear that in the early stages of development, after the gametic
nuclei are differentiated and fused, cell division is the process
of greatest activity; then follow stages during which develop-
ment is characterized chiefly by growth; and lastly the final
aspects are chiefly the result of cellular or tissue differentia-

20 GENERAL EMBRYOLOGY

tion, processes often described separately under the term
Histogenesis.

This brief outline reflects the fundamental character of the
relation of Embryology, as of all biological science, to the Cell
Theory. The recognition of the ovum (Schwann, 1839;
Gegenbaur, 1861) and spermatozoon (Schweigger-Seidel, 1865)
as modified cells, of the basic importance of cell continuity in
development (Virchow), and of the processes of fertilization,
cleavage, growth, and differentiation as essentially cell pro-
cesses, marked noteworthy and fundamental steps in the
history of the science of development. But our recognition
of the importance of this relation, and of the especial importance
of the cell as the descriptive unit in development,*should not
obscure the fact that in many developmental processes we
cannot recognize the cell as the actual unit of physiological
activity. Many important steps in development concern ele-
ments which are distinct and individual intra-cellular elements.
And later, during the cleavage period, the boundaries of
specific materials behaving as units in development do not
always coincide with cell boundaries or distributions. We
must regard the view that the cells are the ultimate units in
development as a stage in the history of opinion, and for the
present recognize certain intra-cellular elements as the " ulti-
mate" structures in development.

But the province of Embryology is not merely thus to de-
scribe the upbuilding and unfolding of the structure and form
of the new organism through these successive stages of develop-
ment; it is, further, to describe the more fundamental pro-
cesses involved in this development, and still further, to
summarize these descriptions of both kinds in the form of simple
general statements or laws. In the historical development of
the science of Embryology, as of any natural science, the descrip-
tion and comparison of visible forms and conditions came
first. /XThis morphological account of development, concerned
chiefly with the description of what happens, what is produced
in development, has now been accomplished to such an extent

ONTOGENY 21

as to furnish a basis of this kind sufficient for immediate
necessity. Next comes the study of the real processes leading
to the production of one condition out of another, processes
which underlie the externally visible form changes. This
physiological aspect of Embryology is concerned more with
how development occurs, how, and through the operation of
what factors or mechanisms, one condition leads to another.
In a way this is also the why of development — not "wjiy" in
the philosophical sense of course, but in the sense of "how
does it happen that" these things occur in development.
Here the two methods of observation and experiment are com-
bined and by the artificial modification or the elimination of
one condition of development after another, the essential
factors are discovered and their modes of operation determined.
The science of Embryology has now fairly entered upon this
stage and the dominant note of the subject to-day is this
search for underlying processes and modes of action. But as
yet it is impossible to say that we have reached the final
period of the formulation of the broad fundamental generali-
zations which give unity to the infinitely diverse phenomena of
development, and which are expressed in the form of laws.
While something has been accomplished in this direction, the
basis of fact is not as yet sufficiently broad, and the necessity
of frequent restatement of such "laws" shows their formulations
to be premature, save as guides in investigation.

These steps in the development of the science of Embryology
do not so nearly represent the course of thought and hypothesis
as that of actual knowledge and achievement. For even in
the eighteenth century the earliest embryologists had their
hypotheses as to the causes of this mysterious process of develop-
ment. They offered first what seemed to them an explanation
of the facts of development which came to be termed the idea
of "evolution" or ^reformation. This idea was that within
the germ, either in the egg ("ovists") or in the spermatozoon
("spermists, " "animalculists") there was contained a miniature
organism resembling, in a general, 'or even in a precise way,
the adult form (Fig. 12). And this miniature had merely to

22

GENERAL EMBRYOLOGY

1

expand, or to unfold and grow, to produce the individual of
the next generation. The relation between the germ and the
adult seemed much like that between the bud and the branch —
all the parts present in minute rudiments, ready to come forth
and expand. We may recognize in this idea a morphological
conception of development such as we should expect to appear
first. This conception, with which are associated such great
names as Malpighi, Bonnet, and Haller,
proves in reality an attempt to explain
development by denying its' occurrence. For
the assumed formation of the original indi-
viduals of a species by the Creator involved
at the same time the creation, within them,
of the preformed germs of all the other later
individuals of the species. The belief that
the germ cells of an organism contained in
miniature the members of the second genera-
tion necessitated the further belief that in
these latter must be contained, within still
smaller limits, the individuals of the third
generation, and thus ad infinitum. And so it
was estimated that some two hundred mil-
lions of human beings were actually contained

FIG. 12. — Draw- . ° J

ing of a human in this preformed condition within the ovaries
of Eve- This conception of infinite encase-
ment or " emboitement" proved to be the

ganism enclosed in
a thin membrane.

After o. Hertwig, Teductio ad oosurdum of the theory of pre-
formation in this its first and crudest form.

from
(1694).

Hartsoeker

Those who actually observed the chick ap-
pear within the egg could not accept this naive explanation of
development, but believed that there occurred a true formation
of parts anew out of unformed material not possessing at all
the characters of the adult organism. This was Wolff's idea of
epigewsis, clearly a physiological conception of development,
following quite naturally the earlier morphological conception.
In its original form epigenesis was chiefly a dissent from the
idea of preformation rather than an explanation of develop-

ONTOGENY 23

ment. Indeed it seems now to have been merely a restate-
ment of the fact that development occurs, leaving this fact to
be explained through the operation of some supernatural or
miraculous process, for the spontaneous generation of the
embryo within the egg was at first definitely assumed.

Thus we have almost from the beginning of embryological
study, two opposing explanations of the visible phenomena
of development, preformation explaining development by de-
nying it, epigenesis explaining development by reaffirming it.
Since this early conflict of opinions, the crudity of which we un-
derstand when we think of the means then at hand for observing
such minute objects as are many eggs and embryos, there has
been constant opposition of morphological and physiological
interpretations of development. The modern understanding
of preformation is better termed predelineation, or better still,
predetermination, less crude, less complete and particular than
preformation. What is preformed or predetermined in the
germ in some way represents the embryo without being at
all like it. The idea of epigenesis, too, is to-day less complete;
a certain structural organization is admittedly present in the
germ as a heritage from previous generations, and real develop-
ment occurs as a physiological process directed by this rudi-
mentary structure already present. The history of these opin-
ions indicates that neither conception is exclusively true, but
that development must involve both predetermination and
epigenesis; and the present endeavor is to find out not which,
but to what extent each, is true.

The present understanding of development seems to be an
extremely refined predetermination strongly tinged with epi-
genesis, using these words in their modern sense. A more
extended statement of this modern view and the facts upon
which it is based is reserved for a later chapter (Chapter VII).
Briefly stated, we believe that while the embryo, not to say
adult, is by no means preformed nor even fully predelineated
in the germ, yet there is a certain degree of protoplasmic
structure or regional differentiation in the germ cells. This is
spoken of now as the organization of the germ, and it may be

24 GENERAL EMBRYOLOGY

both material and dynamic (i.e., energetic). And further this
organization is definitely related to the structure of the future
embryo and adult, having reference, but not resemblance, to the
adult. The organization of the cytoplasmic part of the germ
is itself a condition which develops (epigenesis) under the
influence of the primary structure, or organization, of the
nucleus. At present this inherited organization or predelinea-
tion of the nucleus seems primary and fixed, and to represent
the only strictly predelineated portion of the germ, controlling
and directing the later and epigenetic developmental processes,
which may be said often to have commenced in the germ cells
even before syngamy has occurred. But history warns us
against believing that this organization of the nucleus will
prove the ultimate organization. As knowledge becomes more
complete this will be thrown farther back to restricted elements
of the nucleus; indeed it seems probable now that the primary
organization concerns, not the entire nucleus nor perhaps its
chromosomal elements alone, but some, as yet invisible,
problematic, chemical and physical configurations of its
structure.

But we must not search for an explanation of the whole
process of development, alone in the structure of the germ
cells. We must look upon development as upon other forms
of activity in living things, as a succession of reactions on the
part of the organism to the normal stimuli of its surroundings.
The things that an adult organism does are obviously reactions;
it reacts to the conditions of its environment by making certain
movements, forming certain substances, undergoing certain
structural modifications; in short, by doing certain things
collectively termed its behavior. The precise character of an
animal's behavior is determined not alone by its structure,
by the organs it has to react with, nor alone by its physiological
condition at the time, nor alone by the nature of the external
conditions acting; but by all of these combined. What the
adult organism does at any particular moment is therefore
determined by two interacting sets of conditions, one within
the organism— its organization, the other without the organism

ONTOGENY 25

— its environment. Either set of conditions alone can lead to
no action; for organismal activity is reaction.

Just so the developing organism, at whatever stage it be
considered, reacts to the stimuli of its environment in a manner
determined for the moment, on the one hand by its own state
or "organization," both morphological and physiological, and
on the other by the character of the stimuli acting. The ovum
is not to be regarded as a mechanism wound up, ready upon
receipt of a single stimulus, to go through its development
into an adult organism. It is rather to be regarded as an
organism which reacts to its surroundings by undergoing certain
changes. This changed organism then reacts further by under-
going certain other changes. One reaction of the fertilized
ovum is to cleave, of the blastula to gastrulate, and so on.
Step by step, one condition succeeding another and leading to
still another, the organism gradually alters its morphological
and physiological characteristics. Throughout its whole exist-
ence the organism shows transformations of substance, energy,
and form; we agree to set apart certain of these transformations
occurring at a very early period, and to refer to them as pro-
cesses of "development." The normal " behavior" of the egg
or of the embryo is to develop. The processes of development
are neither easier nor more difficult to explain than the phe-
nomena of adult behavior, and they have just the same basis
in the relation between the internal conditions, within the
organism, and the external conditions, without the organism,
at the time.

From this point of view the question why the egg develops
is a problem not different, in its essentials, from why the
)rganism grows, or why it seeks or avoids the light. None of

iese is to be solved by consideration of the organism alone,
whether egg or adult, apart from the conditions acting upon
the organism; both must be studied together.

With this conception of development in mind, we should here
mention briefly one of the great generalizations that has come
from the study of organic development, namely, the_Biogenetic_
Law_or_Jhe^ieory of Recapitulation. Briefly stated this

26 GENERAL EMBRYOLOGY

familiar theory is, that the organism, in its individual devel-
opmentaljiistory, tends to. repeat in outline the evolutionary
history of its species. This repetition is seldom particular, or
detailed, never complete, yet so many of the phenomena of
development can be satisfactorily interpreted from this his-
torical point of view, seeming to have this historical significance
rather than an immediately adaptive relation, that as a general
statement the law remains fundamentally true.

This law is not so much an attempt to resume the facts of
embryology as to apply these facts in the interpretation of racial
history (Evolution). This application is in many instances
difficult because of the fact that there has been an evolution of
the egg, the embryo, and the larva, just as of the adult. The
fact that the organism is specific at all stages of its existence,
includes the parallel evolution of ova, and of all succeeding

Species Zygote Stages in Development Adult

D AD BD CD D

E AE BE CE DE E

F AF BF CP DF EF F

G A0 BG CG DG EG FG G

H AH BH CH DH EH FH GH H

FIG. 13. — Diagram to illustrate the essentials of the Biogenetic Law. Modified

from O. Hertwig.

developmental stages, if there is to be any evolution of adult
structures, else diversity of adult organization would depend
upon external conditions of development, rather than upon egg
organization. But we know that the eggs of any species of
sea-urchin and star-fish will develop, respectively, into adult
sea-urchins and star-fish of those species, although in the same
dish, with identical environing stimuli.

Many important points concerning the relation between
ontogeny and phytogeny may be represented schematically,
as in the accompanying diagram (Fig. 13). Here we com-

ONTOGENY 27

pare the ontogenies of five related species, the adults of which
represent an evolutionary series; species E has evolved beyond
D, F beyond E, and so forth. The first stage (i.e., the
fertilized ovum or zygote) of D is not merely a zygote (A),
but it is the zygote of species D, and consequently indicated in
our diagram by AD; in its development this passes through
the specific stages BD, and CD, to the adult D. Species E
is more highly evolved than species D, but it begins its existence
as a fertilized ovum which again is specific, this time AE. In
its development to the adult form this may pass through stages
B and C similar to those of species D, but merely similar, not
identical, else the result would be D and not E. Therefore, we
call these intermediate stages BE and CE. Further, species E
may pass through a stage in some particulars resembling D;
this, however, does not exactly resemble D and is therefore
designated DE. Similarly for species F, G, and H; each is
more highly evolved than the preceding. Each passes through
stages which resemble stages in the development of the less
highly evolved species, yet each stage is really specific.

Conditions of life change for the embryo as well as for the
adult, and if these younger organisms are to remain in existence
they must evolve to meet the changed conditions. The process
of evolution concerns not merely the adult, but the organism
at every stage of its existence. Stages such as BG or CH may
finally become so highly modified that they are no longer recog-
nizable as related to B and C, and might as well be termed
XG and YH. It is then said that these traits are " coenogenetic
modifications" in distinction from " palingenetic characteris-
tics," which are obvious similarities to previous racial condi-
tions. But recognition of the idea that the entire life history
is undergoing evolution, at every point, very largely minimizes
the value of this very common distinction between coenogenetic
and palingenetic traits in development, for in a very true sense
all the traits of the developing organisms are in varying degrees
both coenogenetic and palingenetic.

Finally, we see that the problem why the egg develops into
a form resembling its progenitors, rather than organisms of

28 GENERAL EMBRYOLOGY

another kind, that is to say, the problem of heredity, may be
more clearly understood by recognizing that the characteristics
of the organism are specific at all stages of its existence. The
egg of the star-fish is just as much a star-fish as the adult is. The
germinal substance of successive generations of star-fish is
directly continuous. This continuity of specific organization
through the germ, combined with essentially uniform conditions
of development, determines the essential uniformity of each
series of interactions leading to the formation of a new adult
organism. In a real sense the problem of heredity thus becomes
the same as the problem of development. And the problem
why the egg of the star-fish develops into a star-fish and not
into a sea-urchin, is fundamentally the same as the problem
why the star-fish is not a sea-urchin; it is the general problem
of the evolution of organic diversity.

REFERENCES TO LITERATURE

In the "references to literature," given at the end of each chapter,
the author's name and the title of the work, are followed by the reference
to the journal in which the work appeared, or to the place of publication,
in case the work is a separate publication. The number of the volume
(Band, tome, etc.) is printed in black-face Arabic numerals, followed
by the year of appearance. References to pages, parts, etc., are omitted
except in a few necessary instances.

The abbreviations of the more common references are as follows :
Amer. Jour. Anat. American Journal of Anatomy. Baltimore and

Philadelphia.

Amer. Jour. Physiol. American Journal of Physiology. Boston.
Amer. Nat. American Naturalist. Boston and New York.
Anat. Anz. Anatomischer Anzeiger. Jena.
Anat. Hefte. Anatomische Hefte. Wiesbaden.

Arch. Anat. u. Entw. Archiv fur Anatomie und Entwicklungsgeschichte.
Arch. Anat. u. Phys. Archiv filr Anatomie und Physiologic. Leipzig.
Arch. Biol. Archives de Biologie. Leipzig and Paris.
Arch . Entw .-Mech . A rchiv fur Entwickelungsmechanik der Organismen.

Leipzig.
Arch . gesamte Physiol . A rchiv fur die gesamte Physiologic des Menschen

und der Tiere. Bonn.
Arch. mikr. Anat. Archiv fur mikroscopische Anatomie und Entwicke-

lungsgeschichte. Bonn.

ONTOGENY 29

Arch. Protist. Archivfur Protistenkunde. Jena.
Arch. Zellf. Archivfur Zellforschung. Leipzig.

Arch. Zool. Exp. Archives de zoologie experimental et general. Paris.
Biochem. Bull. Biochemical Bulletin. New York.
Biol. Bull. Biological Bulletin. Woods Hole, Mass.
Biol. Centr. Biologisches Centralblatt. Leipzig.
Botan. Gaz. Botanical Gazette. Chicago.

Bull. Mus. Comp. Zool. Harvard Coll. Bulletin of the Museum of Com-
parative Zoology at Harvard College. Cambridge, Mass.
Ergebnisse und Fortschr. Zool. Ergebnisse und Fortschritte der Zoologie.

Jena.
Ergebnisse Anat. u. Entw. Ergebnisse der Anatomie und Entwickelungs-

geschichte. Wiesbaden.

Jena. Zeit. Jenaische Zeitschriftfur Naturwissenschaft. Jena.
Jour. Anat. Phys. Paris. Journal de I'anatomie et de la physiologic

normales et pathologiques de I'homme et des animaux. Paris.
Jour. Coll. Sci. Imp. Univ. Tokyo. Journal of the College of Science,

Imperial University of Tokyo.
Jour. Exp. Zool. Journal of Experimental Zoology. Baltimore and

Philadelphia.

Jour. Morph. Journal of Morphology. Boston and Philadelphia.
Mitt. Stat. Neapel. Mitteilungen aus der zoologischen Station zu Neapel.

Berlin.

Morph. Jahrb. Morphologisches Jahrbuch. Leipzig.
Phil. Trans. Roy. Soc. Philosophical Transactions of the Royal Society

of London.

Pop. Sci. Mo. Popular Science Monthly. New York.
Proc. Am. Phil. Soc. Proceedings of the American Philosophical Society.

Philadelphia.

Q. J. M. S. Quarterly Journal of Microscopical Science. London.
Sitz.-Ber. Acad. Wiss. Berlin. Sitzungsberichte der koniglich preussis-

chen Akademie der Wissenschaften zu Berlin.
Sitz.-Ber. Phys.-Med. Ges. Wlirzburg. Sitzungsberichte der Physicalisch-

medizinisch Gesellschaft zu Wurzburg.
Sitz.-Ber. Ges. Morph. Phys. Sitzungsberichte der Gesellschaft fur

Morphologie und Physiologic in Miinchen.
Trans. Am. Phil. Soc. Transactions of the American Philosophical

Society. Philadelphia.
Zeit. Indukt. Abstamm. Vererbungslehre. Zeitschrift fur induktive

Abstammungs- und Vererbungslehre. Berlin.

Zeit. wiss. Zool. Zeitschriftfur wissenschaftliche Zoologie. Leipzig.
Zool. Jahrb. Zoologische Jahrbucher. (Abteilung fur Anatomie und

Ontogenie der Tiere, unless otherwise specified.) Jena.

30 GENERAL EMBRYOLOGY

REFERENCES TO LITERATURE— CHAPTER I

CALKINS, G. N., The Protozoa. Columbia Univ. Biol. Ser. VI. New
York. 1901.
Protozoology. New York. 1909.

DOFLEIN, F., Lehrbuch der Protozoenkunde/ Jena. (3 Aufl.) 1911.

HARTSOEKER, N., Essay de dioptrique. Paris. 1694.

HERTWIG, O., Zeit- und Streitfragen der Biologic. I. Praformation
oder Epigenese. Grundziige einer Entwickelungstheorie der
Organismen. Jena. 1894. English translation by P. C. Mitchell,
"The Biological Problem of To-day: Preformation or Epigenesis,"
etc. London. 1896. Die Entwickelungslehre im 16. bis 18.
Jahrhundert. Handbuch der vergl. u. exp. Entwick. d. Wirbelt.
I, 1, 1. Jena. 1906. (1901.) Ueber die Stellung der vergleich-
enden Entwickelungslehre zur vergleichenden Anatomie, zur
Systematik und Descendenztheorie. (Das biogenetische Grund-
gesetz, Palingenese und Cenogenese.) Id. Ill, 3. 1906.

HERTWIG, R., Mit welchem Rechte unterscheidet man geschlechtliche
und ungeschlechtliche Fortpflanzung? Sitz.-Ber. Ges. Morph.
Phys. 15. 1899. English translation by Curtis, W. C., Science.
12. 1900.

KOFOID, C. A., On Pleodorina illinoisensis, a new Species from the
Plankton of the Illinois River. Bull. 111. State 'Lab. Nat. Hist. 5.
1898.

KORSCHELT und HEIDER, Lehrbuch der vergl. Entwickelungsgeschichte
der wirbellosen Thiere. IV Abschnitt. Ungeschlechtliche Fort-
pflanzung and Regeneration. Jena. 1910.

LILLIE, F. R., The Theory of Individual Development. Pop. Sci. Mo.
75. 1909.

LOCY, W. A., Biology and its Makers. New York. 1908.

MORGAN, T. H., The Problem of Development. International Monthly.
Burlington, Vt. 1901.

SCHULTZ, E., Prinzipien der rationellen vergleichenden Embryologie.
Leipzig. 1910.

WEISMANN, A., Essays on Heredity. English Translation. I and II
Series. Oxford. 1891, 1892. The Germ Plasm. English Trans-
lation. New York. 1893. The Evolution Theory. English
Translation. New York. 1904.

WHITMAN, C. 0., T^e Inadequacy of the Cell-theory of Development.
Woods Holl Biol. Lect. 1893. Bonnet's Theory of Evolution.
Id. 1894. Evolution and Epigenesis. Id. 1894.

WILSON, E. B., The Problem of Development. Science. 21. 1905.

CHAPTER II
THE CELL AND CELL DIVISION

Two universal characteristics of living things are the posses-
sion of protoplasm and a cellular composition. Recognition of
these fundamental facts was dependent upon the use of the
compound microscope, and so we find them comparatively
late acquirements in the history of biology. The cell-unit
structure of an organic tissue was first described in plants
(cork tissue), by. Robert Hooke in 1665, and quite naturally,
therefore, emphasis was laid upon what we now know as the
"cell walls." As a consequence the term "cell" was applied
to these small box-like units which seemed to resemble the
cells of a honey comb. When, about the middle of the nine-
teenth century, it became apparent that the cell content, and
not the cell wall, was the important thing, the word "cell" had
become so definitely fixed that it could not but be retained,
although its utter inaptness was fully recognized.

This is not the place to discuss the general importance and
significance of the Cell Theory of Schleiden and Schwann (1839)
and their successors. It will become clear as we proceed that
the cell, in structure and in action, is the basis of modern
Embryology. Most of the early processes of organic develop-
ment are strictly cell processes and must be studied from the
standpoints of both Cytology and Embryology, from neither
alone, and throughout development constant reference must
be had to cellular phenomena.

As known to-day cells of different organisms and different
tissues exhibit an unending variety in size, frfrm, structure, and
function (Figs. 14, 15), but throughout there are two essentials
of structure expressed by the definition of a cell given by Leydig
(1852) and by Schultze (1861) as "a mass of protoplasm con-

31

32

GENERAL EMBRYOLOGY

VIII

FIG. 14.— Various forms of cells. IV-IX, from Dahlgren and Kepner, X,
after Prenant and Bouin. /, II. Human leucocyte, X 350. ///. Human red
blood-corpuscle, X 350. IV. Cell from root cap of calla lily, X 350; p, plastids.
V. Epidermis of earthworm, showing four mucous cells in various stages of
secretion, and CM, cuticle, X 550. VI. Fat cells in skin of chicken; n, nucleus.
X 435. VII. Ovarian ovum (oocyte) of cat; cm, cell membrane; mi, micro-
somes; nm, nuclear membrane; o, nucleolus; y, yolk alveoli. VIII. Connective
tissue cells from the lobster; cy, cytoplasmic mass; pr, cytoplasmic processes;
per, peripheral layer of cytoplasm upon which the rigid material of the tissue is
laid down. IX. Pigment cell from the peritoneum of the fish, Ammodytes.
Fully extended; the processes can be completely retracted. Two nuclei, X 90.
X. Stratified epithelium from human pharynx, showing intercellular connections
or bridges, X 375.

THE CELL AND CELL DIVISION

XII

FIG. 15. — Various forms of cells, continued. XI-XIII, after Prenant and
min, XIV-XVI, from Dahlgren and Kepner. XI. Ganglion cell from human
spinal cord; n, nucleus, X 250. XII. Sensory (receptor) cells from human
olfactory epithelium, X 175. XIII. Multipolar ganglion cell from optic ganglion
of horse. (The processes of the right side are cut off.) a. Axon, X 63. XIV.
Part of muscle cell from the fish, Catostomus; c, capillary containing blood cells
and platelets, X 500. XV. Ciliated cells from the digestive tract of the Mollusc,
Cyclas. XVI. Gland cell from the leech, Piscicola; ch, cytoplasmic channels
containing the secretion granules; dt, discharging tubes; n, nucleus; o, nucleolus;
s, secreted materials in various stages of elaboration.

34 GENERAL EMBRYOLOGY

taining a nucleus." To-day we modify this but slightly and
define a cell as a limited mass of protoplasm containing nuclear
material.

~The mass of the cell is usually very small. The smallest
cells known are the Bacteria, some of which may be only
0.001 mm. (1 micron, or 1/25000 inch) in length (Streptococci),
or even less (Staphylococci). But among the Metazoa, cells
are never so minute. The human white blood corpuscle or
leucocyte (Fig. 14, 7), is perhaps of average size, measuring
something less than 0.01 mm. (8-10 micro) in diameter.
Tissue cells having a diameter of 0.05 mm. (50 micro) are
considered large, although a few specialized cells may far exceed
this (e.g., muscle or nerve cells). The egg cells of animals are
usually larger than tissue cells, but this is a special condition,
and is frequently due to the accumulation of stored food sub-
stance, rather than to the possession of a larger amount of
protoplasm. Within the species the sizes of specific varieties
of cells are very constant. The size of an organ or of an
individual is related to the number of its component cells
rather than to their size (Amelung, Conklin).

We may proceed now to describe the essentials of structure
exhibited by a typical cell — an imaginary thing which has no
V more real existence than the " average man." Such a cell
would consist of a spheroidal or irregular mass of protoplasm,
limited by a definite cell membrane or cell wall. The wall may
be a surface condensation of the protoplasm or, more frequently,
a true secretion of the cell body, either membranous as in most
animal tissues, or thick and rigid, like the cellulose walls of
most plants. In many cells the viscid transparent protoplasm
just within and in contact with the cell wall forms a thin layer,
the ectosarc or ectoplasm (Fig. 16), clearer than the granular
and more refractive central endosarc or endoplasm, which
contains, besides the granules, many cell organs and inclusions.

The protoplasm itself is made up of a combination of two
forms, perhaps two kinds, of material plainly differing in density
and arrangement (Fig. 17). The denser material called the
mitome, spongioplasm, reticulum, or filar substance, forms a sort

THE CELL AND CELL DIVISION

35

of complex framework or fine network of irregularly woven
paths along which are scattered minute granules called micro-
somes. The spaces or meshes of this spongioplasmic network
are filled with the less dense ground substance or cell sap, called
also the hyaloplasm, paraplasm, or inter filar substance. The

nm

ch

m

ex

en

d -

FIG. 16. — Diagram of a typical cell, a, aster; c, centre-some (centriole) ;
ch, chromatin; cr, chromidia; cs, centrosphere; d, deutoplasmic granules; en,
endoplasm; ex, exoplasm (cortical plasm); hy, hyaloplasm; k, karyosome; I, linin
network; m, cell membrane; n, nucleus; nm, nuclear membrane; o, nucleolus;
p, plastids; sp, spongioplasm ; v, fluid vacuoles (metaplasm).

actual relation of these two kinds of substances varies in
different kinds of cells or even at different times in the same
cell. A frequent arrangement is that of a reticulum just
described, in which the spongioplasm is definitely fibrous,
forming a felt-work holding the more fluid hyaloplasm. In
other cells, or at other times, protoplasm has a distinctly alveolar
structure resembling a fine emulsion. Here the hyaloplasm is

30

GENERAL EMBRYOLOGY

in the form of minute drops or alveoli, while their walls or the
irregular interalveolar spaces are of the denser material.
Occasionally other structural relations are seen, such as the
granular, where the fibrous reticulum is represented by rows of
excessively minute granules, and the fibrillar, where the fibers

v

FIG. 17. — Alveolar protoplasmic structure in the egg of the sea-urchin, Toxo-
pneustes, one and one-half minutes after the entrance of the spermatozoon.
From Wilson, "Cell," X about 2000. The protoplasm consists of alveoli sur-
rounded by microsomes. In the middle is the centriole, surrounded by the
centrosphere, while radiating from it are the rays of the aster. The large and
small black masses are the sperm head and middle-piece.

of the reticulum are larger, longer, and less branched than in
the ordinary reticulum. It is still uncertain how exactly the
real structure of living protoplasm is represented by its appear-
ance after it has been killed, in preparing it for examination.
It should be remembered that in living protoplasm these
fibrils, reticula, etc., are in all probability fluid structures of
greater density than the ground substance.
The cell is far from being a simple unit, for it contains a

THE CELL AND CELL DIVISION 37

variety of structures and materials differing chemically and
functionally; these may not all be directly visible as organized
structures. The only constantly differentiated substance is the
nuclear material which is usually contained in a definitely
formed body, the nucleus, though it may be scattered through
the protoplasm. There are many reasons for believing that
primitively the nuclear substance was not thus organized into
a definite nucleus, but that it was distributed through the
cytoplasm in the form of small granules, as it is still in many
of the simplest organisms, and that gradually these became
aggregated into fewer larger masses. / The Protozoa show many
stages in the gradual enlargement and numerical reduction of
the nuclear elements, but in the Metazoa the nuclear material
is nearly always collected into a single body./ The nucleus is
to be regarded as a specialized portion of the protoplasm of
the cell, highly differentiated in structure, chemical compo-
sition, function, and behavior. All cell activities seem to
involve mutual interaction between the nucleus and the
remainder of the cell and neither is able long to function
normally without the other. But the action of the nucleus is
primary and directive, to a large extent controlling and regu-
lating cell activities and cell life as a whole. In_most cases
the nucleus is a spherical or ovoid body of fixed form; in some
very active cells it may be elongated or of irregular form, or
even branched, ramifying all through the cell; in a few rare
instances the nucleus may be amoeboid (Figs. 14, 15).

Typically this complex center of cell activity shows much the
same fundamental structure as the remainder of the protoplasm,
which in distinction from the nucleus is called the cytoplasm.
he nucleus is limited by a definite nuclear wall or membrane
formed either from the cytoplasm or from the nucleus itself.
The substance of the nucleus as a whole is termed the karyo-
plasm. The equivalent of the spongioplasmic reticulum of the
cytoplasm is here termed the linin network, and the hyalo-
plasmic ground substance is known as the nuclear sap, or
karyolymph, or paralinin. The chief distinction of the nucleus
is the presence of a very special nucleo-protein substance called

38 GENERAL EMBRYOLOGY

chromatin, a name given it on account of the ease
it is colored with such dyes as hsematoxylin (logwood
carmine. This chromatin is ordinarily in the form of snial
flakes or granules of variable size, termed chromioles (Kison).
These are always distributed along the linin fibers, which stain
less readily and are therefore described as achromatic. Some-
times a few large masses of chromatin called karyosomes may
be seen within the nucleus. Apparently the chromatin may
be rapidly dissolved or condensed, or even formed anew, so
that its visible amount and condition vary considerably from
time to time.

The structural differentiation of the nucleus is frequently
complicated further by the presence of one or more bodies
called plasmosomes or nucleoli, which often superficially resem-
ble the karyosomes, at least morphologically, although chemic-
ally and physiologically they are quite unlike (Fig. 16). The
nucleoli are spheroidal bodies, staining densely, though not of
the same material as the true chromatin. The karyosomes are
sometimes called chromatin nucleoli. The nucleoli vary con-
siderably in number, size, and form, and their significance in
the nucleus is not altogether understood; probably several
unlike bodies having different functions have been included
under this single term.

Another organ of the cell is the centrosome. This is not
always to be seen for it may be lost from the cell at certain
times and reappear later. While commonly a cytoplasmic
structure lying just outside the nucleus, in some forms it i
an intra-nuclear organ and it is quite possible that primitively
it was an essential part of the nucleus, as it still is in many
Protozoa (Figs. 29, 30). The_centrosome is a minute densely
staining granule or pair of granules, sometimes hardly larger
than the granules or microsomes of the cytoplasmic reticulum
Occasionally it consists of several separate granules closely
associated. The cytoplasm in the neighborhood of the centro-
some and directly under its influence is ordinarily differentiatec
as the archoplasm. This substance consists typically of two
portions. A medullary region forming a small spheroida

THE CELL AND CELL DIVISION 39

„

lass called the centrosphere or attraction sphere immediately
surrounds the centrosome, and peripherally, radiating from
this out into the cytoplasm, there is at times a collection of
diverging rays or fibers called collectively the aster. In the
ordinary vegetative cell the centrosome and archoplasmic
structures are usually reduced in size and perhaps even absent
in some cells, but in the dividing cell they may be very large
and prominent organs extending nearly throughout the cell,
for their chief function is in connection with the process of
cell division. Similar dense granules and fibers are found as-
sociated with organs of the cell which are motile, for example,
the granules at the bases of cilia and flagella, the axial fila-
ments of some flagella, undulating membranes, and some
pseudopodia, the fibrilta of muscle cells, etc. All such modi-
fications of protoplasm connected particularly with the pro-
duction or regulation of motion are included under the general

rm kinoplasm^ (Strasburger). It has been suggested that
lie kinoplasm of the cell is of the nature of a definite and

rmanent cell organ. In many cells the centrosome is such

permanent organ, but many other kinoplasmic structures
seem to be more or less temporary and may disappear or
be formed anew at the time of cell division or at other times.

e rays of the aster, for example, in many cases apparently
formed from the enlargement and rearrangement of the
cytoplasmic reticulum resulting from the activity (probably
hemical) of the centrosome.

In addition to these nearly constant cell structures of com-
paratively uniform characteristics, there are various other
organs or bodies which are peculiar to certain special kinds
of cells. Among these are the bodies called in general jd(istids_
(Fig. 14, IV), of which the more familiar are the pigment bodies,
such as cMoroplastids or chlorophyl bodies, the chromoplastids
or colored bodies not containing chlorophyl, amyloplastids or
starch-forming bodies, protein-forming plastids, and others.
K^ Vacuoles of various sizes and kinds are very common, such as
the digestive, excretory, food, water, and food-storage vacuoles;

casionally one or more are specialized as contractile or pul-

40

GENERAL EMBRYOLOGY

sating vacuoles. Nutritive substances are often stored tem-
porarily in cells. These materials are collectively known as
deutoplasm, metaplasm, or paraplasm, and may be starch or
protein granules, yolk plates, oil drops, etc. There are also

Fio. 18. — "Trophospongien" in cells of the hepatic duct of the snail, Helix
vomatia. After Erhard. g, basal granules; n, nucleus; t, trophospongien.

granules of various other kinds — of materials being secretec
or excreted, of food in various stages of ingestion or digestion
pigment granules, crystals, and formed substances of many
kinds, usually specific for the organism or tissue.
In many cells true chromatic structures are found in the

THE CELL AND CELL DIVISION 41

cytoplasm outside the definite nucleus. These are usually
small granules and bits of chromatin of varied form and
significance in different cells. Collectively they are termed
chromidia; for the most part they are not formed in situ, but
are derived directly from the nucleus (Fig. 18). They are most
frequent in very active cells such as gland cells, the rapidly
growing germ cells, and many others. Many forms of chro-
midia, functionally as well as structurally distinct, have been
given special names such as chondriosomes, " Chondromiten,"
idiochromidia, mitochondria, pseudochromosomes} " Nebenkern,"
"Trophospongien," etc.

It has been said already that no single cell shows typically
all the structures described above and illustrated in Fig. 16.
Without stopping for further description many of the details
of structure truly characteristic of a few varied types of cells
are shown for comparison in Figs. 14, 15. (For descriptions and
further details the student should consult the standard texts
of Cytology and Histology, e.g., Dahlgren and Kepner's
"Principles of Animal Histology," New York, 1908.)

One further aspect of cell structure remains to be mentioned.
This is the important fact that there is in most tissues a fairly
definite cell form combined with a nearly constant arrangement
of the cell organs and contents (Van Beneden, Rabl, Heidenhain).
In any epithelial cell the different physiological conditions at
the free and the attached surfaces lead to this definite relation
of cell structures which is called polarity. In such cells the
nucleus lies toward the basal end of the cell, the centrosome
either toward the free end or on that side of the nucleus. Thus
an imaginary axis may be passed through the centrosome and
nucleus perpendicular to the free surface, about which the
cell organs are arranged symmetrically, either bilaterally or
radially (rotatorially). This polarity extends also to other
less constant structures and even occasionally to non-epithelial
cells. It may be seen for example in the arrangement of cilia/
cuticle, conductile and contractile fibers, granules or drops of
substances being excreted or secreted, and yolk or other stored
materials (Figs. 14, 15, 16, etc.). In the germ cells, as we shall

42 GENERAL EMBRYOLOGY

sec later, this fact of polarity becomes of the greatest importance,
for it is frequently related closely to the symmetry of the mature
organism.

While we may thus describe the Metazoan tissue cell as a
separate morphological unit, complete in itself, it is also true
that in a great many tissues the cells are in direct material
continuity with one another through minute protoplasmic
connections or bridges which pass through fine perforations in
the cell walls. These have been observed in a great variety
of tissues in many different forms (Fig. 14, X) ; whether or not
they are present in all tissues it is yet impossible to say definitely,
and we must recognize clearly that in the physiology of the
organism the cells do not behave as completely autonomic
units. While each represents a localized field of activity, the
life of the cell is subordinated to the life of the organism as a
whole — a fact that comes out with especial clearness in the
"development of the organism. In some way not yet understood
the cell, or groups of cells, influence the activities of other cells,
and are in turn influenced by them. The activities of the
Metazoan organism of course equal the sum of the activities of
its component cells; but these combined activities are organized
and- unified into a whole in such a way that this represents
more than the unit activities when considered separately, just
as the action of a community represents something beyond
the sum of the actions of its members taken individually.

From the embryological point of view one of the most im-
portant phases of cell activity is cell reproduction or cell
division. For cells arise only from preexisting cells. The
history of opinion regarding the genesis of cells parallels roughly
that regarding the genesis of organisms. It was Virchow who
finally demonstrated convincingly (1855, 1858) the universality
of the fact of cell continuity in the tissues of a single organism,
and further the fact that in a succession of generations of
organisms the process of the formation of cells from preexisting
cells is not interrupted. We know now that in this process of
cell division all the essential organs of the cell take an active

THE CELL AND CELL DIVISION

43

part — the cytoplasm, nucleus, the centrosome, arid even many
of the plastids. So that the final result of the process is typically
the formation of two daughter cells similar to each other and
also to the parent cell in all essential respects save in size.

The division of cells occurs in two quite dissimilar ways. The
simpler method, and the less frequent, is termed direct division
or amitosis (Flemming). Here the first step is sometimes the
elongation and constriction of the nucleolus, when this is
present, into two separate daughter nucleoli, or in other cases
the appearance of a new second nucleolus (Fig. 19). Next the
whole nucleus divides into two, sometimes by simple constric-
tion into two separate elements, sometimes by the ingrowth of

FIG. 19. — Amitosis in tendon cells of a new-born mouse. After Nowikoff, X 800.

nc, nucleolus.

a partition wall, or in still other cases, by the formation of two
new nuclear membranes within the original membrane, the
disappearance of the latter freeing the two daughter nuclei.
This division of a nucleus is typically followed by the division
of the cytoplasmic portion of the cell which is ordinarily
accomplished by the development of a cell wall between the
two daughter nuclei. Very frequently, however, division of
the cell body does not follow and the cell remains binucleate; or
this process of nuclear fission may be repeated/a multinucleate
cell resulting, such as a striated muscle cell. In such cases of
incomplete cell division the essence of the process seems to be
the rapid increase of nuclear surface and then volume; it is

44 GENERAL EMBRYOLOGY

usually associated with special forms of cell activity. Conditions
in some of the Protozoa suggest that primitively division of the
nucleus or the multiplication of the nuclear bodies might not
have been associated with a corresponding division of the
cytoplasmic body, but that these originally independent
divisions have gradually come to be uniformly associated.

It has commonly been supposed that the direct form of cell
division occurs but rarely and then usually in cells which are
moribund. It is becoming clear, however, that amitosis is in
reality not particularly infrequent. It seems to occur normally
hi many tissues (mesenchyme), and often where there is an un-
usual or a sudden increase in nuclear activity and energy
expenditure on the part of the whole cell, as in ovarian follicle
cells and tapetal cells, muscle cells, rapidly growing or rediffer-
entiating cells in regenerating tissues; it is also true that amitosis
is frequent in such tissues as stratified epithelia whose cells
are nearing the end of their life or activity.

The second and more usual method of cell fission is that
termed indirect cell division, or mitosis (Flemming), or karyo-
kinesis (Schleicher). This is a complicated process involving
the establishment and operation of an intricate mechanism
within the cell, shared in by nearly all its living parts. The
essential result of the division of the cell by the action of this
complex mechanism concerns in particular the chromatic sub-
stance of the nucleus, for in nearly all known instances the
chromatin sharing in the process is very precisely divided into
two equal portions, each of which goes to one of the daughter
cells. We may give here only a brief outline of the essentials of
this process of mitosis, again describing an imaginary schema
with which to compare later some of the variations in detail
shown by actual cells.

As the first step in mitosis we should consider the division of
the centrosome into two, which remain lying together within
the undivided kinoplasmic centrosphere. This division of the
centrosome is usually quite removed in point of time from the
other phenomena of mitosis, for it occurs normally during the
reconstruction of a daughter cell immediately after its formation,

THE CELL AND CELL DIVISION 45

and so is separated by a considerable intervening vegetative
period from the other events of mitosis or the doubling of
other parts (Fig. 20, A). This vegetative phase of cell life is
frequently referred to as the "resting" period or interkinesis;
a state of inaction is not implied by the term " resting," for
during this period the cell is performing its normal and char-
acteristic functions as a tissue cell; the word merely indicates
that the cell is not undergoing any active phase of division.
The termination of the vegetative phase and the immediate
inauguration of mitosis is ordinarily first distinguishable in
the structure of the nucleus. The chromatin granules become
more distinct, enlarge rapidly, and undergo some change in
chemical constitution indicated by an increase in staining
capacity (Fig. 20, A; 22, A). As the chromatin increases some
of the granules or flakes come to be arranged in a linear, or
sometimes bilinear series, still upon some of the linin threads
which share in this arrangement. Thus the chromatin and
linin form a tangled thread or ribbon called the skein or spireme
(Figs. 20, B; 21, B; 22, B).

We should note here that at this time the chromatin of the
nucleus which is not included in the spireme, often indeed the
greater part of the whole amount of this material, is thrown out
into the cytoplasm and dissolves (Fig. 32) ; the more fluid parts
of the nucleus are also thrown into the cytoplasm by the
dissolution of the nuclear membrane. It may thus be only a
comparatively small part of the whole nuclear structure that is
formed into the spireme proper.

, The spireme may be quite continuous throughout the nucleus,
or it may appear from the first as a fragmented thread com-
posed of several short pieces; when in this latter condition it is
spoken of as a segmented spireme. In a few cases the spireme
stage is largely suppressed and the chromatin granules collect
immediately into compact groups without indication of a
skein stage. Tke_ljnin.jaetwQrk,in- part becomes a sort of fine
core throughout the spireme and in the extra-chromatic region
remains as a network of naked fibers. The latter portion soon
becomes polarized so that its fibers converge, more or less

46

GENERAL EMBRYOLOGY

FIG. 20. — Mitosis in cells of Salamandra maculosa. After Prenant and Bouin.
D, H. Primary spermatocytes, others, spermatogonia. A, B^C, X 1000, others,
X 800. A. Interkinesis or resting stage. B. Early prophase; spireme continu-

THE CELL AND CELL DIVISION

47

K

ous. Centrosomes omitted. C. Prophase; spireme segmented into chromo-
somes. Centrosomes commencing to diverge; spindle forming. D. Longi-
tudinal splitting of chromosomes. E. Disappearance of nuclear membrane;
continued divergence of centrosomes and asters. F. Mesophase; formation of
equatorial plate. Polar view. Chromosomes V-shaped. G. Same in side view.
Only a few of the chromosomes are shown. H. Anaphase; daughter chromo-
somes diverging, still united at ends. /. Anaphase; continued divergence of
chromosomes, now entirely separated. J. Late anaphase; complete divergence
of chromosomes. Spindle breaking down, asters disappearing. K. Telo-
phase; beginning of reconstruction of daughter nuclei. Chromosomes disin-
tegrating. L Late telophase; division completed. Nuclei reconstructed;
centriole divided; cell walls completed. Nuclear membrane forming, c, cen-
tiioles; cs, centrosphere; n, nucleus; s, spindle remains; sf, spindle fibers cut across.

48

GENERAL EMBRYOLOGY

FIG. 21. — Diagrams of the process of mitosis. From Wilson, "Cell," slightly
modified. A. Resting-cell with reticular nucletfs and true nucleolus; at c the
attraction-sphere containing two centrosomes. B. Early prophase; the chroma-
tin forming a continuous spireme, nucleolus still present; above, the amphiaster
(a). C.D. Two different types of later prophases; C. Disappearance of the pri-
mary spindle, divergence of the centrosomes to opposite poles of the nucleus
(examples, many plant-cells, cleavage-stages of many eggs). D. Persistence of
the primary spindle (to form in some cases the "central spindle"), fading of the

THE CELL AND CELL DIVISION 49

distinctly, toward the region of the centrosphere; in many
colls such a polarization of the linin toward the centrosome
exists throughout the vegetative phase. The two centrosomes
now begin to diverge and the surrounding centrosphere pulls
in two, one portion accompanying each centrosome (Figs. 20, E;
21, D). As the centrospheres diverge they enlarge, and within
each appear fibers radiating from the centrosome as a center
and producing the asters. While the chromatic and achromatic
parts of the nucleus have been passing through these early
stages of mitosis, the nucleolus when present becomes vacuo-
lated, commences to dissolve and finally disappears. Soon
the nuclear membrane also commences to break down and
dissolve, first in the region of the asters, leaving the nuclear
substance free in the cytoplasm. Next the chromatin thread
shortens and thickens, breaking in the case of the continuous
spireme, into a number of separate segments or rods; or if the
spireme itself is of the segmented type, its elements now shorten
and thicken. When the spireme segments, the linin thread
upon which the chromatin granules are strung may remain
continuous between as well as through the chromatic rods.
These chromatic segments now become quite homogeneous,
clearly differentiated structures called the chromosomes (Figs.
20 C, E; 21, D; 22, C). Strictly speaking, each chromosome
consists of a dense mass of fused chromatin granules with a
portion of linin embedded.

In practically all organisms in which the nucleus is a definitely
formed structure, the number of chromosomes appearing during
mitosis is fixed, and is constant throughout all divisions of

nuclear membrane, ingrowth of the astral rays, segmentation of the spireme-
thread to form the chromosomes (examples, epidermal cells of salamander, forma-
tion of the polar bodies). E. Later prophase of type C; fading of the nuclear
membrane at the poles, formation of a new spindle inside the nucleus; precocious
splitting of the chromosomes (the latter not characteristic of this type alone).
F. The mitotic figure established. G. Metaphase; splitting of the chromo-
somes (e.p.) ; n, the cast-off nucleolus. H . Four stages in the divergence of the
two halves of a chromosome. /. Anaphase; the daughter-chromosomes diverg-
ing, between them the interzonal fibers (£./.), or central spindle; centrosomes al-
ready doubled in anticipation of the ensuing division. J. Late anaphase or
telophase, showing division of the cell-body, mid-body at the equator of the
spindle and beginning reconstruction of the daughter-nuclei. K. Division
completed.

FIG. 22. — Mitosis in the segmenting egg of the clam, Unio. From Dahlgren
and Kepner. A. Prophase of the fourth cleavage. Chromatin reticulum;
centrosomes on opposite sides of nucleus. B. Prophase. Spireme beginning to
segment into chromosomes; nuclear membrane disappearing; spindle forming.
C. Late prophase. Chromosomes formed; spindle becoming completed; nucle-
olus nearly disappeared. D. Mesophase. Chromosomes in equatorial plate.
E. Early anaphase. Divergence of the daughter chromosome groups. F.
Telophase. Nuclear division completed and daughter nuclei reformed; cyto-
plasmic division commencing.

THE CELL AND CELL DIVISION 51

somatic _ceUs... In all but a few groups the chromosomes appear
in an even number, ordinarily between twelve and thirty-six,
although these limits are frequently passed.. The chromosomes
are at present considered the most important elements in the
cell, and interest in the whole process of mitosis centers in their
behavior. The details of the process of mitosis seem directed
toward the exactly equal division and distribution of these
elements, the importance of which justifies the more detailed
consideration which we give them after the general description
of mitosis is completed.

While the chromosomes are forming, the centrosomes and
asters continue to diverge, passing around toward opposite sides
of the nucleus. The linin fibers of the nucleus tend throughout
to remain polarized toward the centrosomes and the separation
of these_bodies from one another draws out the linin fibers into
an elongated bundle converging at each end toward the centro-
some. Finally the centrosomes come to lie on exactly opposite
sides of the nuclear structures and as the nuclear membrane
disappears completely we find the rays of the asters penetrating
into the nuclear region and forming, together with the linin,
a swindle-shaped structure lying between the centrosomes, its
component fibers passing among the chromosomes (Figs. 20, G;
21, F; 22, D). In many cells the centrosomes do not thus
migrate to opposite sides of the nucleus, but separate directly;
the nucleus in this case is simply drawn up to lie between them
(Fig. 21, D). The result is the same, but the difference in
relative behavior of the centrosomes and nucleus is real and
must be taken into account in some cases. The spindle and
asters now form a figure resembling a diagram of the Tines of
force within a simple bipolar magnetic field. This figure is
called the amphiaster, sometimes the achromatic figure, empha-
sizing its distinctness from the chromatic portion of the nucleus,
now all or largely included in the chromosomes. The parts of
the linin network directly continuous with those upon which
the chromosomes were formed originally, seem to have a some-
what different history from the remainder of the linin. They
remain attached to the chromosomes and extend thence toward

52 . GENERAL EMBRYOLOGY

the centrosomes, forming in this stage a superficial sheatl:
around a central portion of the spindle, and are hence term<
the mantle fibers. The central core of the spindle seems in
many cases to be formed largely from the remainder of the
nuclear linin, though in other cases this is formed in the same
way that the asters are, and from cytoplasmic materials. Thus
the amphiaster is usually of mixed origin, nuclear and cyto-
plasmic, though in some cases the spindle at least seems to be
wholly nuclear (linin); the asters are always cytoplasmic in
origin. All these fibers form definite threads, enlarged as
compared with the original linin reticulum.

The definite formation of the chromatic portion of th<
nucleus into chromosomes and the achromatic substance inl
the amphiaster marks the termination of the first phase of
mitosis which is known as the prophase. During the prophase
there has occurred the actual division of only the centrosome
and centrosphere; the other important changes have been
preparatory to further divisions — the dissolution of the nuclear
membrane, the enlargement and rearrangement of the chro-
matin granules, the formation of definite chromosomes, and the
establishment of the achromatic figure. We should remember
that the nucleolus meanwhile has fragmented and, together
with a large or small amount of chromatin which is not formed
into chromosomes, has passed out into the cytoplasm and
disappeared.

The arrangement of the materials forming the achromatic
figure is evidently the result of certain tensions within the cell,
the effect of which is first to draw the chromosomes, until now
distributed irregularly, into a circle about the equator of the
spindle. When in this position the chromosomes are said to
form the equatorial plate (Figs. 20, F, G; 21, F). This phase of
mitosis is also in general preparatory to actual division but it
is carried on after the division mechanism is completely
established. This period of division is known as the mesophase
(Lillie).

Following the mesophase is the metftphase. The chief event
of this phase is the longitudinal splitting or division of each

THE CELL AND CELL DIVISION 53

chromosome into two parallel halves. This forms two equal
and similar groups of daughter chromosomes, each group
similar to the original group except in size. As a matter of
fact this longitudinal splitting of the chromosomes is by no
means always deferred until this time, for frequently it occurs
during the pro phase of division, even in the spireme stage; or
rarely the chromatin granules may divide even before a definite
spireme is constituted (Fig. 23) . In such cases the chromosomes
are in the form of double rods throughout the entire prophase
and mesophase; the metaphase is then present only virtually.

The mantle fibers from the oppo-
site poles of the spindle are now
attached to the daughter chromo-
somes, usually in their middles.
Next the mantle fibers begin to
shorten as the result of some proc-
ess centering in or about the cen-
trosomes at the poles of the spindle.
As the poles are relatively fixed,
perhaps by the anchoring asters or
through the rigidity of the central
part of the spindle, the result is the FIG 23._Longitudinal fission

Separation Of the tWO groups Of of the spireme in the division of
j -i , T i • -i spore-mother-cell in Lilium can-

daughter chromosomes Which move didum. After Farmer and Moore.

along the central spindle fibers to-
ward the regions of the centrosomes. If the mantle fibers
are attached to the middle of the chromosome it is first
drawn out into a 3- or > -shape; frequently this form is
assumed during the mesophase, the apex of the > pointing
centrally in the equatorial plate (Figs. 20, F, H; 21). The
period of mitosis occupied by the divergence of the two chro-

"" " '"* " ' TT """ " —- ' .-^^_ ^B^,,— — '

mosome groups is called the anaphase; this is usually very brief
as the chromosomes diverge rapidly. Their divergence exposes
the central fibers of the spindle which are then called the
interzonal or connecting fibers, and which frequently come to
have an important share in the formation of later structures
(Figs. 21, H; 22, E). In their divergence the chromosomes

54 GENERAL EMBRYOLOGY

themselves seem to be entirely passive, except in a very few
isolated instances where they are definitely amoeboid (Opalina)
(Fig. 31). The two chromosome groups are finally drawn
completely to the opposite poles of the spindle and the process
of mitosis then enters upon its final period, the telophase.

During this phase the cytoplasmic portion of the cell becomes
divided into two parts, usually equal, though occasionally
extremely unequal. Sometimes, as in many animal cells, this
division of the cytoplasm results from its peripheral constriction
in a plane corresponding with that of the equatorial plate, the
constriction deepening until the cell body is completely severed
and the two daughter cells formed (Figs. 20, L; 21, /, J; 22, F).
More frequently, in some animal and in practically all plant cells,
the division of the cytoplasm results from the formation of a
partitioning cell wall, in the formation of which the interzonal
fibers of the spindle usually take an important part. These seem
to increase in number and to thicken in the middle, ultimately
fusing and forming a transverse plate which is the rudiment of
the future cell wall ; the remainder of the wall forms as a distinct
secretion of the cytoplasm in that region.

As the diverging chromosomes approach the poles of the
spindle they lose their distinct outlines, become vesicular, and
gradually lose their visible identity and separateness to a large
extent; with a few exceptions they finally seem to disintegrate
completely and form scattered granules, and a new typical
nucleus is constituted in each daughter cell. Meanwhile the
centrospheres and asters have diminished in extent and clear-
ness and have returned again to the condition which is char-
acteristic of the interkinesis. About the new nucleus a mem-
brane is formed, either from the nucleus or the -cytoplasm, and
the mitosis is accomplished (Figs. 20, L; 21, H-J). In most
cases just about this time the centrosome divides in prepa-
ration for the next mitosis. During the interkinesis the nucleus
and cytoplasm increase in size and soon the process of division
is repeated.

The length of time occupied by the whole process of mitosis
varies greatly. In the division of some egg cells it may be

THE CELL AND CELL DIVISION 55

completed in fifteen minutes, or it may occupy one or even two
hours, and in some special cases a much longer period.

This account of mitosis, although brief and including only
some of the essentials, brings out clearly the unity of the
nucleus as an organ; it behaves as a more or less separate
unit of cell organization throughout all this intricate process.
And we see clearly this extremely important fact, that the
nucleus of a cell is formed from a preexisting nucleus of
the same constitution. Nuclei arise only from preexisting
nuclei; there is a nuclear continuity quite parallel with cell
continuity. And going one step farther, it is probable that
chromosomes are derived only from preexisting chromosomes.
This idea of genetic continuity is not completely applic-
able to all cell organs, however, for occasionally the centrosomes
are not derived from preexisting centrosomes, but may arise
de novo, and in the development of the new organism the
centrosome is typically derived from the sperm cell alone.
Among the plants many of the plastids seem to be genetically
related and to be formed by the division of preexisting plastids.
The other less living cell structures are usually distributed
passively to the daughter cells, and such structures may be
formed anew in the new cells.

The relation between the direction of the plane of cell division
and the general morphology of the cell body demands a word.
From the preceding account it is obvious that the plane of
division is at right angles to the longitudinal axis of the spindle,
but the position which the spindle assumes is itself prede-
termined. The position of the spindle axis is fixed by the
location of the centrosomes. When the single centrosome
divides and the daughter centrosomes pass to opposite sides of
the nucleus, they usually migrate equal distances from the ori-
ginal position of the centrosome; it follows, therefore, that this
region falls within the plane of the equator of the spindle and
consequently in the plane of the new division. When the
centrosome does not alter materially its relative position in the
cell, the next division, again being through the plane occupied
by the centrosome and the center of the nucleusf will, be in

56 GENERAL EMBRYOLOGY

general at right angles to the preceding division (Fig. 24) . Thus
the planes of succeeding divisions tend to alternate, each per-
pendicular to the preceding. Any other relation involves the
migration of the centrosome from the' position originally
occupied by it in the daughter cell; or the spindle may change
its position during or after its formation, and this regular
relation thus be disturbed. Typically the spindle takes such
a position that its long axis lies in the direction of the greatest
protoplasmic extent of the cell (0. Hertwig), a position which
would result from the tensions between a comparatively
elongated body and a fluid medium in which it is suspended

III

FIG. 24. — Diagram illustrating the relation between the position of the centro-
aome and the plane of cell division. Symmetrical motion of the daughter centro-
somes results in the regular alternation, at right angles, of successive division
planes.

and free to move in any direction. There are many ex-
ceptions to these two general rules in special conditions, such
as simple columnar epithelia, stratified epithelia, the maturing
germ cells, etc.; these indicate perhaps that a more funda-
mental cause of the direction of cell division remains to be
discovered.

Some important modifications of this schema of cell division given
above will be noted in other connections, but a few special conditions
are conveniently mentioned here. The most important and fundamental
modifications are doubtless those which occur during the forming and
maturing of the germ cells ; these are to be described in detail in Chapter
IV, and here we should only note that the behavior of the chromosomes

THE CELL AND CELL DIVISION 57

in these divisions is very complex ; that there is here only one-half the
number of these bodies formed in the cells of the somatic tissues; that
mitoses may occur without an intervening interkinesis ; that in the
case of the egg cell the division may be of extreme inequality, so that
one of the daughter cells is like an extremely small bud, although with
respect to nuclear structure the two cells are alike ; and that in certain
special mitoses one or more of the chromosomes may fail to divide and
therefore may be unrepresented in one of the daughter cells.

Among the higher plants an important characteristic is the absence
of definite centrosomes and asters, although these structures are
normally present among the lower plants. The absence of centrosomes
results in the formation of a characteristically blunt or truncated

A B

FIG. 25. — A. Multipolar spindle in spore-mother-cell of Equisetum. From
Wilson, "Cell," after Osterhout. B. Intranuclear spindle in the oocyte of the
Copepod, Canthocamptus staphylinus. From Hegner. X 850.

spindle in most of the higher plants. In some animals the spindle is
rather truncated also, but this is usually found to be in reality multi-
polar, composed of many small bundles of spindle fibers terminating
in a row of centrosomes or centrosome-like bodies (Fig. 25, A). In the
tissue cells of most animals the asters are relatively small, though the
spindle remains large and distinct ; in a few cases it seems that even in
animal cells division may be effected in the absence of centrosomes.

Many significant modifications of mitosis occur among the Protozoa,
where we find certain conditions which seem to offer suggestions as to
the evolution of the mitotic figure and process, as well as of some of the
chief cell organs themselves. Among these forms the process of divi-
sion is often complicated through its being at the same time the essential
step in reproduction, rather than merely a step in or condition of differ-
entiation, as in the Metazoan. Thus the process of budding or gemma-
tion is essentially an unequal cell division; and in brood ("spore") for-
mation we see a form of cell division in which the nucleus divides many

58

GENERAL EMBRYOLOGY

times without any corresponding cytoplasmic divisions, until finally, all
at once, the cytoplasm is cut into a large number of small cells. In
these forms of division no mitotic figure is formed ordinarily, and one
of the modifications due to association with reproduction is seen in the
fact that the resulting bud, or brood-cell, may have a form and structure
entirely unlike that of the mother cell. In budding, especially in that
form called bud-fission, the nucleus does not ordinarily divide until

A B

FIG. 26. — Nuclear division in the Ciliate, Dileptus. From Calkins, "Proto-
zoology." A. Vegetative form. Nucleus in the form of chromatin granules
scattered through the greater part of the cell ("distributed nucleus")- B.
During division. Each chromatin granule elongates and divides into two.

after the bud is practically completed and ready to be cut off, i.e.,
cytoplasmic division tends to precede nuclear division. Another com-
plication due to the same association is the frequent differentiation of
two forms of chromatic substance in the nucleus. These are the repro-
ductive chromatin, or idiochromatin, and the vegetative, or tropho-
chromatin (Dobell) ; in some Protozoa these may come to be organized
into two separate nuclei which are sometimes equivalent to what are
called the micro- and macronucleus respectively. We have already
mentioned the distributed nucleus of many unicellular organisms in
which the chromatin is not organized into a definite nuclear organ,
but is in the form of scattered granules or collections of granules

THE CELL AND CELL DIVISION

59

throughout the cell (Fig. 26, .4). Such a condition indicates strongly
that the nuclear and cytoplasmic parts of the cell have arisen through
the gradual differentiation of a common protoplasmic basis. In cell
division each of these chromatic bodies may first divide into two
(Dileptus), though the members of the resulting pair are not distributed
to different daughter cells, for the accompanying division of the cell
is completed without any rearrangement of the chromatin granules
(Fig. 26, B). In other forms with distributed nuclei (Tetramitus}, the
scattered granules collect about an active kinoplasmic organ termed
the division center (Fig. 27) ; this divides, and the two products separate,

FIG. 27. — Cell division in the Flagellate, Tetramitus. After Calkins. A
Vegetative condition showing scattered chromatin granules (distributed nucleus)
and division center. B. Collection of chromatin granules preparatory to divi-
sion. C. Fission of the division center. D. Separation of the division centers
accompanied by the daughter groups of granules (nuclei),

each accompanied by a group of chromatic granules which are then
redistributed equally to the daughter cells, although no definite mitotic
figure is formed. Even when a definite nucleus does come to be estab-
lished, much of the chromatin of the cell may not be contained within
it but may remain distributed (Heliozoa, Radiolaria) . Finally, of course,
all of the chromatin becomes contained within one or more definite
nuclear structures which may be simple spherical bodies, or they may
show considerable complication and variation in form. A definite
nuclear membrane may be absent at first, as in Chilomonas and
Trachelomonas, though it is formed in practically all cases where the
chromatin granules form definite nuclear groups. Within the nucleus
the chromatic substance may not be definitely organized into chromo-
somes, or these bodies may appear only in certain divisions associated
with gametic reproduction (Paramcecium) ; in some Protozoa, however,
definite chromosomes are typically established and become clearly
marked during each mitosis (Fig. 28).

00 GENERAL EMBRYOLOGY

Probably the most interesting modifications of mitosis among the
Protozoa are those connected with the formation and behavior of the
centrosome and mitotic spindle, upon the origin of which they may

Fio. 28. — Nuclear division in the Rhizopod, Euglypha. After Schewiakoff.
X 800. A. Coarse network of chromatin. B. Contraction of chromatin fibers
and beginning of formation of loops (chromosomes). C. Arrangement of chro-
mosomes in equatorial plate. Polar view. Z>. Equatorial plate. Lateral view.
Spindle forming. E. Splitting of chromosomes and beginning of divergence.
F. Continued divergence of chromosomes. G. Division nearly completed.

perhaps throw some light. In some species there is no indication of
a specialized organ concerned particularly with division. In a few
forms, as mentioned above, a division center may be formed, although

Fio. 29. — Nuclear division in the Rhizopod, Centropyxis aculeata. After
Schaudinn (Doflein). A. Nucleus in vegetative stage. B. Appearance of
centriole. C. Equatorial plate. Spindle with centrosomes; plasma radiations.
D. Beginning of reconstruction of daughter nuclei.

no definite nucleus is present, the distributed chromatin granules
collecting into a single group only at the time of cell fission. In some
species of Amceba, and in a few other forms, one of the large chromatin

THE CELL AND CELL DIVISION

61

bodies, or karyosowies, within the nucleus is specialized as an organ of
division, called the central body and functionally equivalent to a centro-
some (Fig. 29). This body does not lose its chromatic character, may
be surrounded by a definite membrane, and appears to have functions
other than those of a centrosome which it exercises during the intervals
between divisions. In cell division this central body remains wholly
or in part intranuclear. Apparently this represents a very early stage

r;X")-y /"G'?<;isrtSiK5?wV:''i-X!

FIG. 30. — Nuclear division in the Rhizopod, Chlamydophrys stercorea. After
Schaudinn (Doflein). A. Nucleus with central body and chromatin threads.

B. Elongation of central body and beginning of formation of equatorial plate.

C. Equatorial plate. Central body spindle-shaped with polar centrosome-like
thickenings. D. Equatorial plate divided into two. Plasma radiations from
the "centrosomes." E. Fission of central body and chromatin masses. F.
Division completed. Daughter nuclei being reconstructed.

in the differentiation from a chromatic body of the centrosome which
later becomes wholly achromatic and typically extranuclear. Several
other forms have a dynamic division center equivalent to the centrosome
but intranuclear; a nucleus of this type is known as a centronucleus (Fig.
30; see also Fig. 25, B} . In the division of such nuclei the nuclear mem-
brane may remain entirely intact (Euglena) or, as in Noctiluca, the
nuclear membrane may partly break down during mitosis. In several
forms possessing a definite extranuclear centrosome this body remains
undivided in cell fission, passing to one daughter cell alone, while a
new centrosome develops in the other cell ; but this forms first as an

62 GENERAL EMBRYOLOGY

intranuclear structure which later moves out into the cytoplasm along-
side the nucleus. These conditions indicate strongly the nuclear origin
of the centrosome. And there is some reason for believing the spindle
also originally a nuclear structure, as it still is, in part at least. The
spindle is a less constant organ than the centrosome, compared with
which it is of secondary importance. In several Protozoa and simple
plants the spindle is entirely absent, usually where the centrosome is
intranuclear, so that no definite mitotic figure is formed. In other forms
the spindle is intranuclear, and then the centrosomes or their equivalents
may be absent, as in Opalina (Fig. 31). This form is further remark-
able for the chromatic character of some of its spindle fibers, and in

A/T ir Er division in the Infusorian, Opalina intestinalis. After

1200. A. Nucleus in anaphase showing chromatic reticulum,
which may be regarded as equivalent to the spindle, and branched, amoeboid
chromosomes. B. Late telophase.

that, in the absence of centrosomes, the chromosomes separate by an
active amoeboid movement, suggesting a possibly primitive behavior
of the chromosomes in cell division (Metcalf).

It is of course impossible now to get definite information regarding
ie evolution of the cell organs and the process of mitosis, and in these
connections the conditions found at present among the Protozoa are
nly suggestive. Many of these conditions do suggest strongly, how-
ever, that the nucleus has been developed from the gradual aggregation
scattered, chemically differentiated particles; that within the nucleus
sun chromatic elements became specialized into division centers
ich finally became extranuclear, achromatic bodies— the centre-

THE CELL AND CELL DIVISION 63

somes ; and that certain achromatic elements became specialized into a
spindle, also at first intranuclear (linin), and at present in nearly all
forms both intra- and extranuclear in origin. The chromosomes we
must now consider more particularly.

Interest in the process of mitosis centers in the chromosomes
and their behavior, for, as we have said, this whole process seems
to be directed toward the equal distribution of the daughter
chromosomes to the daughter cells, while the cytoplasm may
or may not be equally divided at the same time. We do not
know how the constituents of the nucleus other than the
chromosomes are distributed in cell division. There is little
reason for supposing that these are distributed with exact
equality.

We may recognize two chief aspects of chromosomal behavior
in mitosis. The first is the division of the chromatin granules
or chromioles composing the chromosomes; these granules all
divide in the same direction so that the total result appears as
an exactly equal longitudinal division of each chromosome, or
of the spireme in those cases where the division of the granules
occurs very early. This is the essential act of chromosome
reproduction and it is obviously a process concerning the
chromatin alone, independent of the remainder of the mitotic
mechanism. The second important fact is the distribution of
the two chromosome halves or daughter chromosomes to the
two daughter cells. This is accomplished by the extra-chromo-
somal elements of the mitotic figure, the chromosomes apparently
taking a purely passive part in the process. We see in the
mitotic figure, not a mechanism for cell division merely, for this
is frequently accomplished in the absence of mitosis, nor for
chromosome division, for this frequently precedes the formation
of the mitotic figure; but essentially the mitotic figure is a
mechanism effecting the equal distribution to the daughter
cells of the products of chromosomal division within the nucleus
of the parent cell, so that each new cell has a complete group of
chromosomes similar to those of the parent cell. The precision
and wide occurrence of this equal distribution, through mitosis,
of the chromosomes and of these cell organs alone, leads to

54 GENERAL EMBRYOLOGY

the assumption that the chromosomes constitute the essential
physiological elements of the nucleus and therefore of the cell.
There are many subsidiary facts indicating the great importance
of these bodies. Consideration of the relation of the chromo-
somes to the special problems of heredity and sex will be deferred
to a later chapter (Chapter VII) for fuller consideration, but
we should mention here a few of the important facts and hypo-
theses regarding these bodies.

In the nuclei of many of the Protozoa definite chromosomes
are already present, but in some unicellular organsims there are
conditions suggesting a possible mode of evolution of these
structures. We have mentioned the collection of the distrib-
uted chromatin granules into small groups through the cell;
these groups have been regarded as the rudiments of chromo-
somes. After a definite nucleus is established the chromatin
granules remain as definite bodies, and each divides in cell
fission. Even when the chromatin granules merely become
rearranged about a division center, as in Chilomonas and
Trachelomonas, although definite chromosomes may not be
formed, the division of the granules occurs as the essential step
in fission, just as later when the granules collect and fuse into
chromosomes. In Paramcecium definite chromosomes are
formed only in certain divisions, namely, those immediately
preceding conjugation, that is during garnet ogenesis (Calkins
and Cull, Fig. 82), and it is but a short step from this condition
to the regular formation of chromosomes in all mitoses.

In the reconstruction of the daughter nucleus the chromo-
somes become vacuolated and finally break up into scattered
granules whose distribution through the nucleus is so irregular
that in nearly all cases no trace of chromosomal structure is
apparent. The nucleus then grows rapidly, the chromatin
content often increasing to many times that in the original
chromosome group, until it soon reaches a quite definite size,
varying widely in different cells, but fairly constant for cells of
a single kind in a given species. The nucleus and chromatin
then remain without much further change, quantitative at
any rate, until toward the close of the vegetative or normally

THE CELL AND CELL DIVISION

65

functional period of cell life. At the time of the next division
much of the chromatin is usually eliminated from the nucleus,
is cast out into the cytoplasm and disappears along with the
nucleolus (Fig. 32); the chromosomes which then appear,

FIG. 32. — Early cleavages of the egg of the Nematode, Ascaris. Origin of
primordial germ cells and casting out of chromatin in the somatic cells. From
Wilson, "Cell," after Boveri. A. Two-cell stage dividing; polar view, s, stem-
cell, from which the germ cells are derived. B. Later stage of same division;
lateral view, c, chromatin in somatic cell being thrown out into the cytoplasm.
C. Completion of four-cell stage showing eliminated chromatin. D. Division of
four-cell -stage showing continued chromatin elimination in the third somatic
cell.

or reappear, are similar in every respect to those in the preceding
division (Van Beneden). The chromosomes, therefore, show
the same specific constancy as any other characteristic of the
organism.

66 GENERAL EMBRYOLOGY

The most obvious characteristic of the chromosomes is that
of numerical constancy. In different species of organisms the
number varies greatly but there is in general little if any
relation between the grade or relative complexity of the
organism and the number of chromosomes in its nuclei; closely
related species of a single genus may differ widely, e.g., Ascaris
megalocephala has four, A. lumbricoides forty-eight. In general
the number seems highest in some of the Protozoa. Where
there are very many minute chromosomes the difficulty of
counting them exactly is very great and it cannot then be
said precisely what or how constant the number is. In
Mastigella there are about forty, in Actinosphcerium one hundred
and thirty to one hundred and fifty, in Paramcecium about two
hundred. Among the Metazoa the smallest number is two, in a
variety of Ascaris, the largest known is one hundred and sixty-
eight, in Artemia. Frequent numbers are twelve, sixteen, and
twenty-four, but any number may be found within these known
limits. The number is practically always an even one in
somatic cells, even or odd in the germ cells or their immediate
predecessors. The numbers found in the tissue cells of some
of the familiar organisms are the following: rat, guinea-pig,
ox, sixteen; Amphioxus, salmon, salamander, frog, mouse,
man (female), twenty-four; earthworm, thirty-two; shark,
thirty-six; sea-urchin, eighteen in one species, thirty-six or
thirty-eight, in another; pine, onion, sixteen; lily, peony,
twenty-four.

While the number of chromosomes is thus practically constant
it is not absolutely invariable and deviations from the normal
are now known to occur in several forms. Of course the most
frequent variation is the typical reduction of the number to

one-half the somatic (|j, during certain phases in the for-
mation of the germ cells (Van Beneden), or throughout the
gametophytic generation of many, perhaps most, plants.
But as we shall see later, this should hardly be called a variation
from normal. A deviation of an entirely different kind is seen

in a few cases where the chromosome number is -- in cleavage

2

THE CELL AND CELL DIVISION 67

or tissue cells of certain individuals; thus in the cleavage cells of
Ascaris megalocephala the number is two or four, in the tissues of
Helix pomatia, twenty-four or forty-eight, in Strongylocentrotus,
eighteen or thirty-six. In such cases each of the lesser number
is said to be bivalent, and it is supposed, not without reason,
that each is actually composed of two ordinary or univalent
chromosomes. In a few instances the number may be even
less than one-half the normal and each is then said to be pluri-
valent; thus in the formation of the embryo-sac in the lily a
variation in certain nuclei has been found, the number varying
by fours from twelve to twenty-four (s = 24). The significance
of these unusual cases is varied and sometimes doubtful.
Again, constant differences in chromosome number, in both
somatic and germ cells, are associated with sex differences in a
large number of species of several widely divergent phyla. In
such cases the females have a somatic group from one to five,
or even more, in excess of that of the male; in such cases the
specific number is fixed, though some variable species are
known.

None of these unusual deviations from the normal is of the
character of a " normal variability." Indeed very few in-
stances of this kind of variability in chromosome number are
known. One instance is that of the salamander larva, in
certain tissue cells of which the number is said to vary (Delia
Valle) in different individuals from nineteen to forty, the normal
being twenty-four, and in a single specimen limits of nineteen
and twenty-seven have been described.

This very high degree of specific numerical constancy of the
chromosomes indicates strongly that the appearance of a
chromosome in mitosis is not determined by a large number of
causes, but that it is the result of the operation of a single and
simple factor; what this may be is a matter of conjecture only.

Another important characteristic of the specific chromosome
group is the constancy of the form and size differentiations
among the members of the group. In many organisms it can
be seen clearly that the chromosomes of a single nucleus are
not all of the same size and form; they may differ in shape,

G8 GENERAL EMBRYOLOGY

dimensions, and proportions (Fig. 33) (Montgomery). More-
over these differences are constant from one cell generation to
the next, so that similar chromosomes may be identified in
successive mitoses. The form is somewhat more variable than
the volume, which is remarkably uniform. It should be said
that the size of a given chromosome is not fixed throughout the
entire cell history, for at certain periods, particularly in the

FIG. 33. — Various chromosome groups illustrating variation in size and form,
and coupling of chromosomes. A,B, from Sutton, C, after Wilson, D, after Agar.
A,B. Spermatogonia of the grasshopper, Brachystola magna. C. Spermato-
gonium of the squash-bug, Anasa tristis. D. Spermatogonium of the lung-fish,
Lepidosiren.

germinal tissues, the chromosomes may be many times larger
than at other periods (Fig. 34). But at corresponding cell ages
the corresponding chromosomes are practically equal in volume,
and in somatic cells such volume changes of single chromosomes
are relatively infrequent.

One most significant and very important fact in this con-
nection is that in the somatic cell the chromosomes are pres-
ent in couples of similar elements; there are two of each size or
form (Montgomery) (Figs. 33, 72, A; 142, A). The exceptions

THE CELL AND CELL DIVISION

69

are found chiefly in those forms where sex differences are found;
in such cases one or more chromosomes are unpaired, or the
members of the pair may be dissimilar in size. And further

o

in the germ cells with - chromosomes, none is paired — all are

single, each somatic pair is represented, and the groups in eggs
and sperm are alike.

FIG. 34. — Changes in the volume of chromosomes in the egg of the Elasmo-
branch, Pristiurus. All drawn to same scale. From Wilson, "Cell," after
Riickert. A. In egg of 3 mm. diameter. Chromosomes at maximal size and
minimal staining capacity. B. In egg of 13 mm. diameter. C. In fully grown
ovarian egg. Minimal size and maximal staining capacity.

Such facts as those given above taken in connection with the
precision with which each chromosome is halved in mitosis, lead
almost irresistibly to the supposition that the chromosomes
must be qualitatively unlike. Such qualitative differences
cannot be observed directly, and can only be inferred, but as
we shall see in connection with the relation of the chromosomes
to heredity, this inference seems to be justified from the results
of the experimental or accidental modification of the chromo-
somal content of the nuclei and the character of the resulting
cells or cell groups (Chapter VII).

70 GENERAL EMBRYOLOGY

In connection with the chromosomes there remain to be mentioned
two important hypotheses. The first is the hypothesis of the speci-
ficity of the chromosomes. Stated in its barest form the essence of
this idea is that each chromosome functions, in cell life, in its own
particular way, representing a center for reactions of a specific kind
only ; that the chromosomes are cell " organs," functionally differentiated
and representing a division of labor roughly analogous to the functional
differentiations of the whole animal body. It is impossible to discuss
this hypothesis satisfactorily here and it is deferred to Chapter VII,
where it occupies a natural place in our account of the mechanism of
differentiation.

The second hypothesis is that of the genetic continuity of the chromo-
somes. The essential of this idea is that the chromosomes which appear
during the preparation for a mitosis, are definitely related in a precise
way, to the chromosomes entering that nucleus at the close of the
preceding division. In its first form this hypothesis was called that
of the individuality of the chromosomes (Rabl, Boveri), and it was
held that the chromosomes actually, though not visibly, preserve
their structural identity during the period of interkinesis, that the
chromosomes of one mitosis are not related to those, of the preceding
division, but are actually the same chromosomes. The fact that little
or no direct evidence of chromosomal identity during the interkinesis,
is to be had, has led to the remodeling of the idea of individuality,
into the hypothesis of genetic continuity.

The nature of the evidence bearing upon this hypothesis, while not
scanty, is largely circumstantial, and hardly affords definite proof, either
affirmatively or negatively. We may suggest some of this evidence
without pretending to give a detailed account of the facts.

At the very beginning we must recognize that the chromosomes are
actually visible, as differentiated structures, only during that com-
paratively brief period of cell life occupied by the process of mitosis.
Considering first some of the facts opposed to this hypothesis, we should
say that those who deny the fact of continuity, maintain that during
the vegetative period of cell life the dissolution and disappearance of the
chromosomes is not only apparent but real. There is truly little visible
and direct evidence of chromosomes during interkinesis. And further,
much new chromatin is formed during this period which cannot be dis-
tinguished from the chromosomal chromatin; in the early stages of
mitosis much chromatin is again thrown out of the nucleus and takes
no part in the formation of chromosomes. It is impossible to say
whether the chromatin forming the chromosomes is or is not then, the
same as that previously derived from them, or that resulting from growth.
Many instances are known where the chromatin of the vegetative
nucleus is nearly all contained within a single homogeneous chromatin

THE CELL AND CELL DIVISION

71

nucleolus or karyosome (e.g., Asterias), and in preparation for mitosis
some granules pass out of the karyosome and form into distinct chro-
mosomes while the greater part of the karyosome dissolves (Fig. 35) ;
it is difficult to understand how the chromosomes could have preserved
their integrity through such a history. In amitosis no chromosomes
are visible yet the nuclei of cells thus formed seem to function normally
and such cells are capable of typical differentiations and in a few in-
stances are said, with some question, however, to be subsequently
capable of division by mitosis, and of normal chromosome formation.

FIG. 35. — Primary ob'cyte of the star-fish, Asterias forbesii, at beginning of
division. From Dahlgren and Kepner (Jordan). Chromatin leaving the "chro-
matin reservoir" or chromatin nucleolus, and being added to the chromosomes.

Such an interruption of a series of mitotic divisions by a period of
amitotic division would seem to exclude the possibility of both genetic
relation and specificity of the chromosomes. In those forms where
chromosomes are not normally formed, the chromatin granules are the
units in nuclear division ; and even when these are formed into chromo-
somes the essential step in nuclear division is the fission of these granules,
which thus seem to be the real units of the chromatic substance. Yet
one could hardly maintain the genetic continuity of these granules upon
other than logical grounds, and to many there seems no stopping place

72 GENERAL EMBRYOLOGY

short of this, if the fact of continuity of organized chromatic structure
be accepted to begin with.

On the other hand, those who adhere to the continuity hypothesis
find many supporting facts in the phenomena of fertilization and
maturation, the importance of which can be appreciated more fully
after our consideration of these subjects. They assume, from the con-
sideration of the behavior of the chromosomes, that they only apparently
lose their structural and functional identity during the interkinesis,
and that something directly representative of the chromosomal structure

FIG. 36. — Indications of the individuality of the chromosomes in the cleavage
of the egg of Ascaris. From Wilson, "Cell," after Boveri. E. Anaphase of first
cleavage. F. Two-cell stage with lobed nuclei, the lobes formed by the ends of the
chromosomes. G. Early prophase of next division. Chromosomes reforming,
centrosomes dividing. H. Late prophases of same division, the chromosomes
lying with their ends in the same position as at the close of the preceding division.

is present in the resting nucleus, invisible directly and known only from
its consequences. There are, it is true, a few instances in which the
chromosomal arrangement is said really to be visible in the interkinesis
(Figs. 36, 37), but these cases are not very clear, except in the maturation
divisions leading directly to the formation of the specialized germ cells,
where the chromosomes are definitely known to be directly continuous.
The remarkable constancy of form, volume, and number of chromosomes
throughout the cells of a given organism and species, is important evi-
dence favoring the hypothesis under consideration. The probability
is so high as to amount almost to certainty, that if the chromosomes

THE CELL AND CELL DIVISION 73

were new and independent formations in each mitosis, they would show
a normal variability in number throughout long series of mitoses.
However, numerical variability is so rare as to be practically absent,
although the few exceptions known are emphasized by those who do
not accept the idea of continuity. Constancy of form seems to be less
precise than constancy of volume, but both are sufficiently marked to
be noteworthy. It is difficult to get precise observations here on account
of the liability to shrinkage or deformation of the chromosomes in the
preparation of the material for study. There is often undoubtedly a
definite pairing of chromosomes in the somatic nuclei (Fig. 33), while in

the germ nuclei, with - chromosomes, the same categories of chromo-

£

somal form and size found in somatic nuclei are distinguishable, but
these are no longer paired — there are
only single representatives of each.
This is one of the strongest points
favoring this hypothesis. And no less
significant is the fact that the odd or
unpaired chromosomes associated with
sex differentiation, mentioned above,
remain constant in size and form and
number, throughout the tissue cells of
certain individuals, and are easily rec-
ognizable, not only through their pe-
culiar morphology but on account of
their peculiar behavior as well. It is
important in this connection, to no-
tice that these particular chromosomes
may remain undissolved even in the early prophase of division of a sper-
resting nucleus, where they had often matogonium of Brachystola magna.
v i MI i ,• 11. From Button. Spiremes forming in

been described as chromatm nucleoh lobes of the nucieus corresponding

or other bodies, before their signifi- with the chromosomes which en-
cance was appreciated or their history tered the nucleus at the close of the

J preceding division,
known.

Another group of facts of quite a different character has an important
bearing upon these hypotheses. It sometimes happens in mitosis that
one or more chromosomes belonging to one daughter group, accidentally
become included with the other group so that one of the daughter
nuclei has fewer, the other more, than the normal somatic number.
In subsequent divisions of these cells the number of chromosomes
appearing is not the normal, but the increased or diminished number,
the sum of the two, however, always being 2s. Or in fertilization of the

egg by the sperm, each of which has ~- chromosomes, various abnor-

74 GENERAL EMBRYOLOGY

malities occur in the distribution of the chromosomes, and it is always
the case that the number of chromosomes forming out of a nucleus is the
same as the number passing into it, no matter how that deviates from the
normal (Boveri) (Fig. 38). There seems to be no regulation within the
nucleus in this respect, such as would result were it a unified structure
tending always to maintain its own normal. In many instances of

B

0

FIG. 38. — Indications of the individuality of the chromosomes in the fertiliza-
tion of Ascaris. From Wilson, "Cell," after Boveri. A. The two chromosomes
of the egg-nucleus, accidentally separated, have given rise each to a reticular nu-
cleus (9 , 9 ) ; the sperm-nucleus below (cT). B. Later stage of the same, a single
chromosome in each egg-nucleus, two in the sperm-nucleus. C. An egg in which
the second polar body has been retained; p.b.2 the two chromosomes arising from
it, 9 the egg-chromosomes, cT the sperm-chromosomes. D. Resulting equator-
ial plate with six chromosomes.

alternation of generations, the agameticaily produced generation is
formed from a cell with -- chromosomes; in all the cells of such an

organism, the nuclei show the same reduced number, even in so
complex an organism as the fern prothallus. In some forms, the

THE CELL AND CELL DIVISION 75

chromosome groups derived from the egg and sperm nuclei, each of S

st

chromosomes, remain distinguishable through a considerable series of
the early divisions of the zygote; two separate spiremes, each forming

s chromosomes, may even be distinguished sometimes (Ruckert).
And in some hybrids where the chromosomes are unlike in number, or

ill •«

.•it'-'V.^'v^';

Hi ?: .<*

^ ../.\>
^T.

.;

H

FIG. 39. — The chromosome group in the hybrids of the Teleosts, Fundulus and
Menidia, showing the distinctness of the paternal and maternal elements. From
Moenkhaus. A. Late anaphase of first cleavage of normally fertilized Fundulus.
All chromosomes of the long type. B. Anaphase of first cleavage of normally
fertilized egg of Menidia. All chromosomes of the short type. C. Anaphase of
first cleavage of Fundulus egg fertilized with Menidia sperm. To the left long
(Fundulus) chromosomes only, to the right short (Menidia) chromosomes only.

D. Anaphase of first cleavage of Menidia egg fertilized with Fundulus sperm.

E. Metaphase of first cleavage of Fundulus egg fertilized with Menidia sperm.
d\ chromosomes of paternal origin. 9 , chromosomes of maternal origin.

form, or size, the two groups derived from the male and female parents
remain distinct (Fig. 39), often for a very long time, perhaps even
throughout life (Moenkhaus, Herbst, Baltzer).

While evidence of the kind suggested above does not constitute
definite proof of the genetic continuity of the chromosomes, it is very
difficult to explain the facts upon any other basis. In the absence of any
other satisfactory hypothesis in this field it seems wise to accept a
certain modification of the essential idea, as a working basis, while
admitting the difficulties of demonstration and the existence of some
apparent contradictions. We may recognize in this difference of opinion

76 GENERAL EMBRYOLOGY

regarding the continuity of the chromosomes, an outcropping of the
opposed preformational and epigenetic conceptions, which pervade all
descriptions of developmental phenomena. The hypothesis of chromo-
somal continuity is essentially a preformational view; those who deny
continuity assume the epigenetic view.

Perhaps the immediate solution of the difficulty here may not be un-
like the solution of the greater problem and more fundamental difference
of opinion. It seems likely that what is directly continuous from
nucleus to nucleus, i.e., what is preformed, is some sort of fundamental
organization determining the chromosomal structure of the nucleus,
just as the "organization" of the egg determines the structure of the
embryo developed from it. Yet what appears is a new structure, formed
epigenetically under or through the influence of the "organizational"
factor, by the material present, and subject to the modifying influences
of changing conditions external to the nucleus. That is to say, the
formation of the chromosomes out of the chromatin of the vegetative
nucleus is to be regarded as a true process of development. The reac-
tion between the fundamental organized structure of the nucleus, and
the stimuli acting upon it, consists in the formation of the chromosomes
and other structures, not to be seen in the nucleus, previous to this
reaction. The chromosomes are thus no more genetically continuous
than the organs of adults, and yet there is a real continuity of
organization underlying.

In many respects the nucleus is analogous to an organism, the chro-
mosomes and other nuclear structures representing the organismal
organs; both are functionally specific, are constant in number, form,
and size throughout the species; both reproduce and exhibit develop-
ment as a form of response. As Wilson points out, the analogy is far
from complete — no complete analogy is known, but that there is an
underlying organization of some kind, continuous and specific, seems
clear although we remain entirely ignorant of what it really is, and just
how it operates and is affected by new conditions.

Before leaving the subjects of the cell and cell division we must con-
sider briefly two other questions which are of great importance, but
which are also still in a hypothetical state. These are, the nature of
the causes leading to cell division, and the nature of the fundamental
mechanism of the process. The interactions between the nucleus and
cytoplasm, and between these and the external medium, which consti-
tute the life of the cell, are largely, probably wholly, of a physico-chem-
ical nature. As such their normal procedure is dependent upon certain
mass relations of the interacting substances, and upon the maintenance
of adequate pathways of interaction between them. For these reasons
we look quite naturally, in searching for a possible cause of cell division,
to the volumetric relations of the nucleus and cytoplasm, and to the

THE CELL AND CELL DIVISION 77

extent of the surface of these parts of the cells in relation to the two
masses.

Immediately after mitosis is completed the nucleus grows very rapidly
for a brief period, and then much more slowly or not at all. The cyto-
plasm, however, does not show this rapid initial growth, but maintains
a fairly uniform and continuous growth rate. As a result, in a newly
formed cell the ratio of nuclear mass to cytoplasmic mass becomes quite
high, but soon diminishes, and in an older cell diminishes rapidly, since
the cytoplasm continues to grow after the nucleus nearly ceases. Con-
sidering first the relation of the surface to the mass, and assuming for
illustration that the cell and nucleus are both spherical, we can see how
this increase in size alters the relation of mass to path of interchange,
since the area of the surface of an enlarging sphere increases relatively
slower than the volume. Should the cell double its diameter through
growth, it increases its volume eight times and its surface only four times.
A cubical cell doubling the length of its sides reduces its ratio of area to
volume as 2 : 1. Or expressing a similar relation somewhat differently,
a spherical cell doubling its volume, lessens its ratio of surface to volume
approximately as 5 : 4, while a cubical cell doubling its volume reduces
the same relation approximately as 6 : 4.75.

These relations are probably important for both of the chief interac-
tions of cell life, those between nucleus and cytoplasm, and between
cytoplasm and surrounding medium. The only pathway between cyto-
plasm and nucleus is the nuclear membrane, while the surface of the
cytoplasm or cell wall forms the pathway between cell and medium.
There is of course a limit to the capacity, so to speak, of these surfaces,
and as the cell increases in volume this limit tends to be reached. The
ratios of these surfaces to the masses are raised by the division of the
cell, which reduces volume more than surface, and thus restores the
efficiency of the surfaces as pathways of interaction. The tendency for
the cytoplasmic mass to increase more rapidly than the mass of the
nucleus in older cells seems to be of even greater importance than these
surface to mass relations. There seems to be a fairly definite specific
limit to the ratio of nuclear mass to cytoplasmic mass, although it is
difficult to say whether after all the nuclear surface relation is not even
here an equally important factor. This mass relation is called the
"kern-plasma" or nucleo- cytoplasmic relation, the importance of which
is emphasized particularly by Richard Hertwig. There is a definite
average cell size in a given tissue and species ; a large or a small organ
or organism does not possess respectively larger or smaller cells, but
larger or smaller numbers of cells of the same average dimensions.
The limiting factor, however, seems to be not the actual bulk of the cell,
but the proportion between the volume of the nucleus and that of the

78

GENERAL EMBRYOLOGY

cytoplasm, i.e., (y^) . In many cells during, and immediately after,
division the nucleus grows at the expense of the cytoplasm, and the
ratio (^) is raised (Fig. 40). The cell then enters upon its vegetative
phase, during which the cytoplasm grows more rapidly than the nucleus,
and the ratio diminishes toward the lower functional limit; as the ratio
approaches this limit the functional activities of the cell change, and the
normal vegetative processes give place to that form of action which we
call cell division, during which the ratio again rises to a value permitting
normal vegetative functioning. It is not yet possible to state in
precise quantitative terms what the limits of these ratios of volume

2

1.8

1 b

C

1.9

1,8

1.4

16 17

FIG. 40. — Curve showing the increase in volume of nucleus (a-d) and cytoplasm
(6-c) during inter kinesis, in the Infusorian; Frontonia leucas, at temperature of
25° C. After Popoff. Ordinates, volume; abcissas, time in hours. Each curve
shows a doubling of volume (1:2) during the seventeen hour period of the inter-
kinesis. Each curve is based upon its own units of measurement, which are
different for the two curves. The nucleo-cytoplasmic relation is identical at the
beginning and end of the period. The size of the nucleus is relatively smallest at
fifteen hours; then the nucleus begins to grow very rapidly, so that at the time
of the division of the whole cell, the original relation is restored.

and surface are in specific instances, nor even to say whether the volume
or surface, or volume and surface relations are those essentially involved.
But so far this nucleo-cytoplasmic hypothesis is the most plausible ex-
planation of the nature of the immediate conditions of cell division.
It should be said, however, that the applicability of Hertwig's " Kern-
plasma Relation" is still chiefly limited to Protozoan cells, and that
even here there are many contradictions. Some of the more obvious
exceptions to the definition of such a limiting ratio are to be ex-
plained as special adaptations. Such are the very great cytoplas-
mic bulk of many egg colls or the relatively large size of the nucleus
in the sperm cell. Many other exceptions of special character are to

THE CELL AND CELL DIVISION

79

be found, usually associated with reproductive processes, e.g., brood
formation.

Regarding the real nature of the mechanism of mitosis, even less
can be said to be known than with respect to its causes. The arrange-
ment of the achromatic amphiaster and the behavior of the chromosomes
show that in the mitotically dividing cell the forces or tensions are

FIG. 41. — Diagrams of the arrangement of the spongioplasmic network
(mitome) in the cell. A, B, C, from Korschelt and Heider, after Heidenhain.
A. Schema of the arrangement of the spongioplasm as it would appear in the
absence of a nucleus. Symmetrical monocentric system. B. Arrangement in
the presence of a nucleus. Asymmetrical monocentric system. C. Arrange-
ment at the beginning of mitosis. Commencement of dicentric system. D.
Arrangement during the process of mitosis. Symmetrical dicentric system.
a,6, cell axis; A;, nucleus.

arranged in a dicentric. system, whereas in the vegetative cell the system
is monocentric (Fig. 41). As the result of the action of the forces in this
dicentric system the chromosome halves are separated and the cyto-
plasm divided. The problem here is to discover the nature of the
forces and the cause of the formation of the amphiaster in the form
which it has. In explanation of the first part of the problem, i.e., the

80 GENERAL EMBRYOLOGY

divergence of the chromosomes, it was formerly believed that the fibers
in the amphiaster were the active elements and that different groups of
these fibers had different functions. Thus the mantle fibers attached
to the chromosomes were supposed to be contractile and by shortening
to draw the chromosomes toward the ends of the spindle, the central
spindle remaining rigid and resisting any tendency for the ends of the
spindle to approach as the result of the contraction of the mantle fibers,
and at the same time serving as a sort of track upon which the chromo-
somes would slide along. The asters then, either as anchors helped to
fix the ends of the spindle, or by contraction served to draw the fully
diverged chromosomes further into the daughter cells (Fol, Van Bene-
den, Heidenhain). This naive explanation cannot be applied in toto
to any known instance of division, although certain features may be
correct descriptions of the events in certain forms. And there are many
facts opposing such an account of the action of the forces of mitosis, such
as the absence of mantle fibers or of asters in many mitoses. Another
hypothesis, in greater favor at present, and apparently well founded, is
that the centrosomes and centrospheres rather than the achromatic
fibers are the active elements, and further, that their activity is primarily
of a chemical nature. Thus the chromosomes are believed to be chem-
ically attracted toward the centrosomes, the achromatic fibers being
passive and formed merely as the result of the rearrangement of cyto-
plasmic granules along the paths of the chemical transformations which
have their seat in the centrosomes and extend thence through the cyto-
plasm (Strasburger) . This hypothesis goes farther and makes it possible
to explain the division of the cytoplasm on the same basis. For the
chemical transformations centering in the centrosomes might easily
influence the tensions of the comparatively impermeable surface film
of the cell so that in the region near the centrosomes the tensions would
be increased while that region farthest from the centrosomes and sym-
metrically related to them both, namely, the plane of the equatorial
plate, would be a region of lower tension; the result of this would, of
course, be a constriction in this plane. It is quite likely too that dif-
ferences in electric tension accompany these chemical transformations,
and these might assist in the alteration of the surface tensions in such a
way as to contribute to the same end (R. S. Lillie). It is known that
there are differences in the electric potential in different regions of the
dividing cell, in some cases at least.

A further development of the chemical hypothesis attempts to explain
the formation of the amphiaster itself. Thus, increase in the ratio of
the volume of the cytoplasm to volume or surface of the nucleus beyond
a certain point leads to a chemical alteration of the centrosomes such
that they become the centers of two equivalent series of chemical reac-
tions with the cytoplasm, the result of which is the formation of the

THE CELL AND CELL DIVISION 81

dicentric system. It is suggested, reasonably, that the chemical altera-
tion is such that the centrosomes absorb or condense the more liquid
parts of the cytoplasm, leaving this considerably more dense than in the
vegetative condition (Butschli, Rhumbler). The existence of two such
centers withdrawing the more liquid parts of the cytoplasm would
lead to the radiations seen in the asters and spindle, which would thus re-
sult from a physical alteration of the structure of the cytoplasm induced
by the chemical changes within the centrosomes. This is all extremely
hypothetical of course, but there are many inorganic phenomena as
well as processes to be seen in the cleavage of the egg which lend con-
siderable support here. At any rate these hypotheses represent the
state of our present ignorance of the nature and origin of the mitotic
figure and process. There are many reasons for believing that the chem-
ical differentiations within the cell are of fundamental importance here,
such as the fact that cells can be made to divide artificially by altering
their chemical structure. And that interactions of the nucleus and
cytoplasm are involved is indicated by the important observation that
while the amphiaster may be formed in the absence of a nucleus, no
real division of the cell may occur without the presence of nuclear
material (Boveri, Ziegler).

REFERENCES TO LITERATURE

VON BAER, K. E., Die Metamorphose des Eies der Batrachier vor der
Erscheinung des Embryo und Folgerungen aus ihr fur die Theorie
der Erzeugung. Arch. Anat. u. Phys. 1834.

BALTZER, F., Zur Kenntnis der Mechanik der Kerntheilungsfiguren.
Arch. Entw.-Mech. 32. 1911.

BENDA, C., Die Mitochondria. Ergebnisse Anat. u. Entw. 12. 1903
(1902).

VAN BENEDEN, E., Recherches sur la composition et la signification de
1'ceuf, etc. Mem. Acad. roy. Belgique. 34. 1870. Recherches
sur la maturation de Fceuf et la fecondation. Arch. Biol. 4.
1883.

BONNEVIE, K., Chromosomenstudien. Arch. Zellf. 1. 1908.

BOVERI, T., Zellenstudien II. Die Befruchtung und Teilung des Eies
von Ascaris megalocephala. Jena. Zeit. 22 (15). 1888. Zel-
lenstudien III. Ueber das Verhalten der chromatischen Kern-
substance bei der Bildung der Richtungskorper und bei der
Befruchtung. Jena. Zeit. 24 (17). 1890. Ueber die Befruch-
tungs- und Entwickelungsfahigkeit kernloser Seeigel-Eier, etc.
Arch. Entw.-Mech. 2. 1895. Zur Physiologic der Kern- und
Zelltheilung. Sitz.-Ber. Phys.-Med. Ges. Wiirzburg. 1897.

82 GENERAL EMBRYOLOGY

BIJTSCHLI, 0., Untersuchungen iiber mikroskopische Schaume und das

Protoplasma. Leipzig. 1892.
CALKINS, G. N., Nuclear Division in Protozoa. Woods Holl Biol. Lect.

1899. (See also ref. Ch. I.)
CALKINS, G. N., and CULL, S. W., The Conjugation of Paramaecium au-

relia (caudatum). Arch. Protist. 10. 1907.

CONKLIN, E. G., Karyokinesis and Cytokinesis in the Maturation, Ferti-
lization and Cleavage of Crepidula and other Gasteropoda. Jour.

Acad. Nat. Sci. Philadelphia. 12. 1902. Cell Size and Nuclear

Size. Jour. Exp. Zool. 12. 1912.
DAHLGREN, U., and KEPNER, W. A., Text-book of the Principles of

Animal Histology. New York. 1908.
DELAGE, Y., La structure du protoplasma et les theories sur 1'heredite et

les grands problems de la Biologic generale. Paris. 1895.
DELLA VALLE, P., L'Organizzazione della ehromatina studiata mediante

il numero dei chromosomi. Archiv. Zoologico. (Napoli.) 4.

1909.
DOBELL, C. C., Chromidia and the Binuclearity Hypotheses: a Review

and a Criticism. Q. J. M. S. 53. 1909.
DOFLEIN, F. (See ref. Ch. I.)
DUESBERG, J., Nouvelles recherches sur 1'appareil mitochondrial des

cellules seminales. Arch. Zellf. 6. 1910.
ERDMANN, R., Kern- und Protoplasmawachsthum in ihren Beziehungen

zueinander. Anat. Hefte. 18. 1910.
ERHARD, H., Studien iiber "Trophospongien." Zugleich ein Beitrag

zur Kenntnis der Secretion. Festschr. f. R. Hertwig. Jena.

1910.
FARMER, J. B., The Structure of Animal and Vegetable Cells. Lan-

kester's Treatise on Zoology. I, 2. London. 1902.
FICK, R., Vererbungsfragen, Reduktions- und Chromosomenhypothesen

Bastardregeln. Ergebnisse Anat. u. Entw. 16. 1906 (1907).
FLEMMING, W., Zellsubstanz, Kern und Zellteilung. Leipzig. 1882.
GURWITSCH, A., Morphologie und Biologie der Zelle. Jena. 1904.
HACKER, V., Praxis und Theorie der Zellen- und Befruchtungslehre.

Jena. 1899.
HARTMANN, M., Die Konstitution derProtistenkerne und ihre Bedeutung

fur die Zellenlehre. Jena. 1911.
HEIDENHAIN, M., Cytomechanische Studien. Arch. Entw.-Mech. 1.

1895. Plasma und Zelle. Jena. 1907.

HERTWIG, 0., Die Zelle und die Gewebe. Jena. I, 1893. II, 1898.
HERTWIG, R., Ueber neue Probleme der Zellenlehre. Arch. Zellf. L

1908.
JORGENSEN, M., Zur Entwk'kluiigsgeschichte des Eierstockeies von

THE CELL AND CELL DIVISION 83

Proteus anguineus (Grottenholm). Festschr. f. 11. Hertwig.

Jena. 1910.

LILLIE, F. R., Karyokinetic Figures of Centrifuged Eggs. An Experi-
mental Test of the Center of Force Hypothesis. Biol. Bull. 17.

1909.
LILLIE, R. S., The Physiology of Cell-division. IV. The Action of

Salt Solutions followed by hypertonic Sea-water on unfertilized

Sea-urchin Eggs, and the Role of Membranes in Mitosis. Jour.

Morph. 22. 1911.
METCALF, M. M., Opalina. Its Anatomy and Reproduction, with a

Description of Infection Experiments and a Chronological Review

of the Literature. Arch. Protist. 13. 1909.
MOENKHAUS, W. J., The Development of the Hybrids between Fundulus

heteroditus and Menidia notata, with especial Reference to the

Behavior of the Maternal and Paternal Chromatin. Am. Jour.

Anat. 3. 1904.
MONTGOMERY, T. H., JR., A Study of the Chromosomes of the Germ

Cells of the Metazoa. Trans. Amer. Phil. Soc. 20. 1901. On

the Dimegalous Sperm and Chromosomal Variation of Euschistus,

with Reference to Chromosomal Continuity. Arch. Zellf. 5.

1910.
NOWIKOFF, M., Zur Frage liber die Bedeutung der Amitose. Arch.

Zellf. 6. 1910.

POPOFF, M., Experimented Zellstudien. Arch. Zellf. 1. 1908.
PRENANT, A., Theories et interpretations physiques de la mitose. Jour.

Anat. Phys. Paris. 46. 1910. Les mitochondries et 1'ergasto-

plasme. Id. 46. 1910.
PRENANT, A., BOUIN, P., et MAILLARD, L., Traite d'Histologie. Paris.

1910.
PRZIBRAM, H., Experimental Zoology. I. Embryogeny. Cambridge.

1908. Anwendung elementarer Mathematik auf biologische

Probleme. Leipzig. 1908.
SCHAXEL, J., Die Morphologic des Eiwachstums und der Follikel-

bildungen bei den Ascidien — Ein Beitrag zur Frage der Chromidien

bei Metazoen. Arch. Zellf. 4. 1910.
SCHEWIAKOFF, W., Ueber die karyokinetische Kerntheilung der

Euglypha alveolata. Morph. Jahrb. 13. 1887.
SCHNEIDER, K. C., Histologisches Praktikum der Tiere. Jena. 1908.

Histologische Mitteilungen. III. Chromosomengenese. Fest-
schr. f. R. Hertwig. Jena. 1910.

STRASBURGER, E., Zellbildung und Zellteilung. 3 Auf. Jena. 1880.
BUTTON, W. S., On the Morphology of the Chromosome Group in

Brachystola magna. Biol. Bull. 4. 1902.

84 GENERAL EMBRYOLOGY

VIRCHOW, R., Die Cellularpathologie in ihrer Begrundung auf physiolog-
ische und pathologische Geweblehre. Berlin. 1858.

WILSON, E. B., Atlas of Fertilization and Karyokinesis. New York.
1895. On Protoplasmic Structure in the Eggs of Echinoderms and
some other Animals. Jour. Morph. 16, Suppl. 1899. The Cell
in Development and Inheritance. Columbia Univ. Biol. Ser. IV.
2ed. New York. 1900.

ZIEGLER, H. E., Experimentelle Studien iiber die Zelltheilung. II.
Furchung ohne Chromosomen. Arch. Entw.-Mech. 6. 1898.

CHAPTER III
THE GERM CELLS AND THEIR FORMATION

reproductive elements of the Metazoa^are single cells,
often greatly specialized in form, and always highly differen-
tiated in internal structure and in function. They differ from
the reproductive cells of several groups of the Metaphyta, in
that they cannot function, i.e., develop, until two single cells,
usually derived from two different individuals, shall have met
and fused, or conjugated. In many plants single reproductive
cells (spores) are formed which develop directly without any
such conjugation, and which are therefore to be distinguished
from the true germ cells, or gametes, which develop only after
conjugation. We shall describe the process of conjugation, or
fertilization, in a following chapter, but in order to appreciate
the significance of many of the details of germ-cell form and
structure, we must remember that they are adapted toward
ensuring the conjugation of two unlike cells, egg and sper-
matozoon.

Reproductive cells are set apart from vegetative cells in
many of the colonial Protozoa. In some cases they are dis-
tinguishable only at certain times, when cells usually vegetative
may give up such characteristics and become reproductive; in
others the reproductive and vegetative cells remain perma-
nently distinct though only slightly differentiated structurally.
Finally in a few colonial Protozoa the reproductive cells are
considerably modified from the vegetative condition, and in
form and composition, as well as in function and behavior, are
readily distinguished as germ cells. Many of the details
regarding these cells have already been mentioned, others can
be more conveniently and more significantly considered later,
in connection with the process of fertilization (Chapter V). In

85

86 GENERAL EMBRYOLOGY

xthis chapter we shall first describe the form and structure of the
typical germ cells of the Metazoa, as they appear when fully
formed and ready to function, that is, just prior to fertilization.
We shall then mention some of the more important modifica-
tions of structure shown in different groups, and finally give a
brief account of the formation and history of the germ cells up
to the time when they are ready actually to enter upon the real
process of development. The details of certain phases in the
history of the germ-cell nuclei, namely, the maturation processes,
are of such importance that we shall refer to them only briefly
in this chapter and devote that following to a more extended
account of these events.

Among the Metazoa the fully formed germ cells are always
of two very unlike types, the ova or eggs and the spermatozoa
or sperm cells. These cells are alike only with respect to their
nuclear structure and composition. Their form differences
are associated with fundamental differences in function. The
egg cell contains by far the greater share of the substance which
is to form the material basis of the new individual. The sperm,
on the other hand, contributes little substance, and that chiefly
nuclear, to the new individual. One of its important functions
seems to be that of ensuring to the comparatively passive egg,
a stimulus which leads it to react, i.e., to commence develop-
ment; and the sperm's nuclear configuration, with that of the
egg, together appear to determine the course of development
to a large extent, if not wholly.

The substance of which the ovum is composed is not a homo-
geneous protoplasm. The cytoplasm is differentiated and
organized into a definite structural and chemical (energetic)
configuration. The details of this configuration are uniform
in the eggs of each animal kind, i.e., it is specific. This cyto-
plasmic structure of the ovum, although itself apparently
determined primarily by nuclear activity, is of great importance
in maintaining the continuity and uniformity of organismal
characteristics through successive generations (heredity).

The ova are less modified in external form than the sperma-
tozoa, and often approach the form of a typical cell, except

GERM CELLS AND THEIR FORMATION

87

that they are nearly always larger than ordinary cells (Fig. 42).
The size of the ovum is not related to the size of the organism
producing it, but is in general related to the amount of food
substance stored in it, the actually living protoplasm showing
much less variation in amount than the deutoplasm. The
smallest eggs are those of the Mammals (Figs. 43, 14, 7/7), in
man only 0.25 mm. (250 micra or 0.01 inch) in diameter, others
being still smaller — 0.07-0.10 mm. (70-100 micro) in the deer,

FIG. 42. — A. Section through the egg of the lamprey, Petromyzon flumatilis.
After Herfort. B. Spermatozoon, drawn to scale, d.en., dense endoplasm;
i.m., inner membrane (? vitelline) ; a.m., outer membrane (? chorion) ; p, granular
"polar plasm;" v.en., vacuolated endoplasm; v.ex., vacuolated exoplasm; I, first
polar body; //, second polar spindle.

and only 0.065 mm. (65 micra) in the mouse. The largest eggs
are, in volume, the largest known cells; such are the " yolks" of
birds' eggs, the largest of which are several inches in diameter
and equalled in other groups only by the eggs of one of the
sharks (Heterodontus) which are nearly 2 inches (4.0 to 5.0 cm.)
in diameter. In a very few cases (some Ccelenterates and
Porifera, and a few worms) the ova may be capable of loco-
motion, performing amoeboid movements (Fig. 44), but in
nearly all cases they are quiescent, passive structures although

88

GENERAL EMBRYOLOGY

Fia. 43. — Fully grown human oocyte freshly removed from the ovary. Out-
side the oocyte are the clear zona pellucida and the follicular epithelium (corona
radiata). The central part of the oocyte contains deutoplasmic bodies and the
eccentric nucleus (germinal vesicle) ; superficially is a well marked exoplasmic or
cortical layer. From Waldeyer-Hertwig.

Fio. 44. — Fully grown egg of Hydra viridis, containing nuclei of ingested cells.
gr>, nucleus of egg. The egg is amoeboid. From Waldeyer-Hertwig, after
Kleinenberg.

GERM CELLS AND THEIR FORMATION

89

containing a great deal of potential energy which becomes
kinetic after fertilization, during the early stages of develop-
ment.

Upon examining the internal structure of the egg we find
that the nucleus is unusually large in most cases, spherical or
ovoid, and with or without a nuclear membrane (Figs. 43, 45).
It is located centrally or eccentrically, usually the latter if the
egg contains an appreciable amount of food material. The
chromatic network of the nucleus may be either dense, or so
open as to give, even after staining, the appearance of a lighter

FIG. 45. — Axial section through the oocyte of the Annulate, Nereis. After
Lillie, slightly modified, c, cortical protoplasmic layer (exoplasm) ; n, nucleus;
no, nucleolus; o, oil vacuoles; v, vitelline membrane; y, yolk bodies.

area in the cytoplasm (Figs. 43, 45), often known as the germinal
vesicle. In some eggs the nucleus is in the process of division at
the time of egg-deposition (Figs. 42, 46); this condition will
be explained later. The actual morphological composition of
the nucleus is one of the chief characteristics of the egg,
and this too is reserved for consideration in Chapter IV.
There is commonly a large plasmosome or nucleolus of varying
form and size (Figs. 43, 45). Centrosome and centrosphere are
typically present at this time, though often minute and difficult
to observe; later these structures disappear entirely. Fre-
quently the cytoplasm contains unusual bodies termed "yolk-
nuclei." The term yolk-nucleus includes organs of several
different types, in some way related apparently to the formation

90 GENERAL EMBRYOLOGY

and deposition of the yolk and certain other substances within
the egg cytoplasm. Sometimes the cytoplasm appears homo-
geneous, in other cases it shows considerable differentiation.
Often the peripheral layer of the cytoplasm is more vacuolated
and less granular than the central portion; the former is then
spoken of as exoplasm or as the cortical layer, the latter as
endoplasm (Figs. 42, 43, 45, 46). There may also be various
materials in the cytoplasm which have been laid down during
the formation of the egg, under the influence of the nucleus, or
yolk-nucleus, and deposited in different regions (Figs. 42-48).

FIG. 46. — Section through the ovarian egg (oocyte) of Amphioxus. After
Sobotta. X 525. c, vacuolated cortical layer (exoplasm) ; e, endoplasm con-
taining deutoplasmic bodies; v, vitelline membrane; /, first polar body; //,
second polar spindle.

The extent of the cytoplasmic differentiation varies greatly
in eggs of different species. In many forms it can hardly be
demonstrated in the egg at the time it is fully formed; in such
eggs this differentiation appears later, during or after the proc-
esses of fertilization, or even still later, during cleavage. We
shall have to return to this subject in connection with the sub-
ject of cleavage, after we have described the fertilization proc-
ess. But there are two or three fundamental aspects of this
fact that should be mentioned here. There are quite commonly
three, sometimes more, distinct forms of cytoplasmic material
arranged as definite regions of the ovum, occasionally as zones,
or layers, or as localized masses (Figs. 42, 45). These may be
distinguished by the more or less vacuolated character of the

GERM CELLS AND THEIR FORMATION

91

protoplasm, or by the collection of various pigments and dif-
ferently colored granules, or by forms
of deutoplasmic materials other than
yolk, or in various other ways. The
disposition of these substances usu-
ally expresses, incompletely, how-
ever, an underlying organization or
morphology of the egg substance as ,.
a whole, which is considered a fun-
damental structure of the egg as a
specific organism. This organiza-
tion is practically always polar, i.e.,
disposed symmetrically with refer-
ence to one chief axis (Von Baer),
and in the eggs of most bilateral d
animals examined, it is bilateral also
(Roux, Van Beneden). In some
Iway this morphology of the egg is
related to the morphology of the
embryo developed from the egg, and
hence is called its promorphology.

This promorphology is better
termed organization, for it is not only
grossly material, but also dynamic,
i.e., energetic, depending upon chem-
ical and physical arrangements not

often visible directly. The extent drawing of a median section

and nature of this organization are

FIG. 47. — Semidiagrammatic

Often Obscure, but this, and the and Heider, after Henking and

, Blochmann. en, chonon; d,

nuclear Structure OI the OVUm, are flattened dorsal side of the egg;

probably its most important char-

acteristicS, for together these deter- truded through themicropyle;

,, » . . , , A;, cortical layer; m, micropyle;

mine the COUrse Of its development this also marks the anterior end

as a specific creature. of «"» e««: ?> e^ a«d ?Perm

. . t pronuclei in process of fusion; r,

Polarity is One expression Of this polar bodies; v, ventral side of

organization. The polarity of the egg'

fully formed ovum is related to the polarity of the egg cell

92 GENERAL EMBRYOLOGY

as it was placed in the epithelium of the gonad, the chief
egg axis corresponding with the axis passing through the
attached and free surfaces of the epithelial cell. This polarity
apparently determines the primary position of the egg nucleus
and centrosomes, and thus secondarily determines also the
arrangement of the cytoplasmic substances which develop
through the interactions between the nucleus and cytoplasm,
processes which may frequently be observed during the growth
period of the ovum. The two poles of the egg are commonly
unlike (Figs. 42, 48), so that we distinguish an animal and a
vegetal pole, corresponding in most cases with the originally
free and attached surfaces respectively, although this relation
may occasionally be reversed (Echinoderms). In general
the animal pole is that toward which the nucleus is eccentrically
displaced, and nearer which the centrosome, or similar body, is
located; it is the more protoplasmic, and therefore the more
active region of the egg. The vegetal pole is frequently occu-
pied largely by the relatively inert food substance, the material^
in general related with the vegetative organs of the developing
embryo.

As regards the nature of the organization, or promorphological
relations, of the egg, two views have been taken and will be discussed
in Chapter VII. The first is that the differentiated substances visible
within the cytoplasm are genuinely " organ-forming substances," or at
any rate tissue forming, in their potentiality. Thus they represent the
organs of the embryo in an intracellular form. The second view is that
these substances are only secondarily related to the real morphology
of the embryo, and that both embryonic structure and the differentiated
substances of the egg, are the result of an underlying, invisible, and as
yet little-known organization of the ground substance of the egg cyto-
plasm. According to this view the correspondence between the ' 'organ-
forming substances" of the ovum, and the organs and tissues of the
embryo, is not in itself direct, but results from their common relation to
the primary underlying arrangement or organization of the substance
of the egg. In some cases the arrangement and position of these sub-
stances may be considerably altered experimentally without disturbing
the normal course of development. It should be added, however, that
in some other cases such a disarrangement does effect a corresponding
disarrangement of the organs or tissues of the embryo.

GERM CELLS AND THEIR FORMATION 93

In addition to the formative substances mentioned above
eggs may contain varying amounts of nutritive substance of
many different kinds, collectively termed yolk oT_deutoplasm.
The yolk may be in the form of granules, small spherical bodies,
large plates, fluid drops of various sizes, or in compact masses
(Figs. 45, 48). These substances may be of different chemical
compositions and staining reactions in a single egg. They
may be formed within the egg by its own activity, or they may
be contributed indirectly by cells associated with the egg during
its formation. The arrangement of the food substances in the
egg has an important bearing upon its later development, espe-
cially upon the form of its cleavage (Balfour). Eggs in which
the yolk is distributed quite uniformly through the cytoplasm,
and in which the protoplasm is therefore more or less com-
pletely intermingled with the yolk granules, or plates, are
termed homolecithal or isolecithal eggs. Some eggs have been
described as alecithal, i.e., without yolk, but many of these
have been found really to contain a small amount of quite
uniformly distributed deutoplasm, and a truly alecithal egg is
rarely if ever found. Eggs of some species among nearly all
the large groups are of this homolecithal type, for example,
the star-fish, sea-urchin, and also the Mammals, which were
formerly thought to be alecithal (Fig. 43). More frequently
the yolk and cytoplasm are not uniformly mingled but are
chiefly accumulated in different parts of the cell. Ordinarily
these materials occupy opposite poles of the egg so that this
retains a radial or rotatorial symmetry; the yolk is accumulated
toward the vegetative pole, the protoplasm toward the animal
pole|(Fig. 48). Such eggs are termed telokcithal. They show
great variation in the relative amount of yolk contained. On
the one hand it is often difficult to distinguish the telolecithal
egg from the homolecithal type, for the tendency toward polar
accumulation of the yolk may be very slight. The egg of
Amphioxus illustrates such a transitional condition. At the
opposite extreme we find eggs such as those of the Reptiles and
Birds, which are relatively immense cells, in .which it is difficult
to distinguish, before development begins, any definite region

94

GENERAL EMBRYOLOGY

which is entirely free from yolk. Between these extremes all
intermediate conditions are found. This telolecithal type of
egg is very common among the Invertebrates, and is charac-
teristic of all the Craniata except the true Mammals. Among
the Chordata successive stages in the accumulation of yolk are
represented by _Amphioxus, Lampreys, Ganoids, Dipnoans,,
Amphibians, Reptiles, and Birds. In the last two groups the

d

FIG. 48. — Egg of the Teleost, Fundulus heteroclitus. Total view, about an
hour after fertilization, c, chorion; d, protoplasmic germ disc or blastodisc;
o, oil vacuoles; p, perivitelline space; v, vitelline membrane; y, yolk.

protoplasm is extremely limited in amount, and is found only
as a small disc or layer on the surface of the spherical yolk-
mass, at the animal pole. A third and less common arrange-
ment of yolk is that seen in the centroledthal eggs of many
Arthropods, chiefly Insects. Here the yolk occupies a greater
or lesser portion of the center of the egg while the protoplasm
forms a superficial layer all around it (Figs. 47, 117, 118).

GERM CELLS AND THEIR FORMATION

95

The various constituent materials of the egg differ, often
very considerably, in density, and since they are definitely
disposed with reference to the chief egg axis, the eggs tend to
assume a definite position with respect to gravity when they
are free to move. Usually, in such cases, the yolk is heavier
than the protoplasm, and the animal pole is therefore directed
upward; this position is reversed occasionally, particularly
when the deutoplasm is in the form of oil drops, e.g.. Nereis and
most Teleosts.

In some forms the egg cells are naked, without cell coverings
or membranes, as in many Ccelenterates and some Molluscs.
Or the egg may be naked at first, but soon after becoming free
from the parental body it may acquire a thin membrane over
its surface, as in Echinoderms. Ordinarily the egg is surrounded
by definite membranes of varying nature and origin. The pri-
mary egg membrane is the vitelline membrane,
or true egg membrane. Typically this is a
thin membrane secreted by the superficial
protoplasm of the egg and closely applied to
its surface (Figs. 42, 45, 46, 47). In most
cases it is quite structureless; sometimes it
is thicker and may be perforated radially
by minute canals or pores, when it is termed
the zona radiata. Occasionally the vitelline
membrane may appear double, showing an
inner zona radiata and an outer structureless ziegier, after Baifour.

i /TV Af\ i-r\\ T i *j • fe, Follicular epithe-

layer (Figs. 49, 50). In such cases it is iium;^ outer portion
possible that the membrane is not wholly of the vitelline mem-

. . J brane (zona pellu-

vitelhne, or it may be the case more fre- cida); yk, surface of
quently that the pores originally passed the ovum; Zn' inner

FlG- 49.— Section

through the egg mem-

branes of the Bias-

portion of the vitel-
line membrane (zona
radiata) .

completely through the membrane and dis-
appeared from the outer portion of it upon
contact with an external medium.

The vitelline membrane may envelop the egg completely, or
there may be left a minute, funnel-shaped perforation through
it at the point where the egg was attached or otherwise espe-
cially related to the epithelium of the ovary. This aperture

96 GENERAL EMBRYOLOGY

is the micropyle, (Fig. 50), and, as we shall see, this sometimes
affords an aperture for the entrance of the sperm cell.

A secondary membrane called the chorion often surrounds the
egg outside the vitelline (Figs. 42-50). This is a secretion
formed while the egg is still contained within the ovary, from
the cells there surrounding the egg, and its presence depends
upon the arrangement of these ovarian cells in the form of a
definite layer or epithelium surrounding the egg, termed the

FIG. 50. — A. Animal pole of the egg of the Cephalopod, Argonauta. From
Wilson, "Cell," after Ussow. Surrounding the egg is the chorionic membrane
perforated by the funnel-shaped micropyle, m. Beneath the micropyle lies the
egg nucleus in the cortical protoplasmic layer, p.b, polar bodies. B, C. Sec-
tions through the egg membranes and micropyle of the egg of the Teleosts Esox
(B) (ovarian egg) and Pygosteus (C) (ovarian egg, 0.4 mm. in diameter). After
Eigenmann. X 375. ez, zona radiata externa; /, egg follicle; iz, zona radiata
interna; m, micropylar cell; p, egg protoplasm; pp, protoplasmic processes;
y, yolk in egg; z, zona radiata.

follicle. The chorion may be a thin flexible membrane, or a
tough resistant shell, as in Insects and Teleosts. Penetrating
the chorion there is nearly always a continuation of the micro-
pyle. The formation of this results from the fact that one of
the cells of the follicle usually acquires a very intimate relation
with the ovum, through a fine pseudopodial process, so that
at this point no membranes are laid down (Fig. 50). When the

GERM CELLS AND THEIR FORMATION 97

egg is fully formed and leaves the follicle, this process is with-
drawn, leaving a funnel-shaped canal. In a few instances
(some Insects) there are several micropylar perforations
through the egg membranes. Such openings are to be regarded
as specializations of the minute canals, mentioned above, which
give the appearance of the zona radiata to the membranes.

Finally there is a great variety of tertiary membranes formed
by the walls of the oviducts, or by special glands in connection
with the reproductive system. These are applied outside the
chorion, or if this is absent, directly upon the vitelline mem-
brane. These envelopes may be of slime or jelly of an albu-
minous character, fibrous, or shelly coverings of chitin, lime, or
other substance. In forms depositing the eggs in the water this
is sometimes a thick jelly, holding the eggs together in strings
or masses, or serving to attach them, either singly or in masses
to plants, sticks, or other solid objects (e.g., Amphibia). These
tertiary membranes serve also, in special instances, as protec-
tion against drying, temperature changes, pressure, or mechani-
cal injury, and against the attacks of food-seeking organisms,
or infection by bacteria or other parasitic organisms. Fre-
quently they are nutritive in character, as the albumin or
" white" of the birds' eggs or the dense oily substance surround-
ing the eggs of the snails.

Eggs may possess none or all three of these classes of mem-
branes; sometimes only primary and secondary, or primary and
tertiary membranes are present. This usually depends upon
the nature of the egg-laying habits, method and duration of
development, and various other conditions.

The spermatozoa, when fully formed, bear little resemblance
to ordinary cells, yet their individual history clearly shows them
to be such. In a few forms, chiefly among the Crustacea, the
sperms do resemble ordinary cells, and are often provided with
long radiating processes, sometimes, though rarely, pseudopo-
dial. But by far the most common form is that known as the
flagellate spermatozoon, found in all groups of animals from Pro-
tozoa to man. These are minute thread-like cells in which three
general regions can usually be made out (Fig. 51). One end is

GENERAL EMBRYOLOGY

n
ac
pc

- 8

mi

- a

• af

e

ABC

FIG. 51. — Flagellate sper-
matozoon. A , B. Two views
of human sperm cell. After
Retzius. X 2000. C. Dia-
gram of the structure of a
generalized type of flagellate
spermatozoon. After
Meves. a, annulus; ac,
anterior centrosome; af, ax-
ial filament; c, centrosomes
(end knobs) ; e, protoplasmic
envelope; h, head; m, middle
piece; mi, mitochondria; n,
nucleus; ne, neck; p, perfora-
torium (acrosome) ; pc, pos-
terior centrosome; s, spiral
filament; /, tail piece; if,
terminal filament.

enlarged forming the so-called hea<L
This is in reality chiefly made up of the
nucleus of the cell, and it stains densely
with all nuclear dyes. The chromatin
of the nucleus is solidly packed, and
though the head of the sperm is much
smaller than the egg nucleus, the two
contain practically, perhaps precisely,
equal amounts of chromatin. Just
behind the head is a smaller middle
piece which is the chief cytoplasmic
portion of the cell; the cytoplasm is
really continued as a very thin envelope
over the head, at the anterior end of
which it is usually produced as a sharp-
ened perforating or attaching organ
called the acrosome or perforatorium.
In some spermatozoa (e.g., some Mam-
malia) the head is connected with the
middle piece by an intermediate section
called the neck (Fig. 51). The middle
piece is quite highly differentiated.. It
contains the centrosomal structures of
the spermatozoon, and its center is oc-
cupied by the proximal portion of a
kinoplasmic structure, the axial fila-
ment. Surrounding this are frequently
one or more differentiated layers, and
often a spirally wound thread; the
whole is covered with a dense outer
sheath. In some instances (toad) the
centrosome is said to be included in the
region of the head piece. Extending
posteriorly from ;the middle piece is a
long flagellum or tail, in some species
flattened and provided with a fin-like
undulatory membrane (Fig. 52, L).

GERM CELLS AND THEIR FORMATION

0

99

FIG. 52. — Various types of spermatozoa. A, B. The Teleost, Leuciscus
(Ballowitz). C, D. The birds, Phyllopneuste and Tadorna (Ballowitz). E, F.
Two forms of the sperm of the snail, Paludina (von Brunn). G. The Nematode,
Ascaris (Van Beneden). H. The Annulate, Myzostoma (Wheeler). I. The
bat, Vesperugo (Ballowitz). J. The opossum, Didelphys (Wilson). K. The
rat (Wilson). L. The Urodele, Amphiuma (McGregor). M. The Crustacean,

100 GENERAL EMBRYOLOGY

Through the middle of the tail is an axial filament connect-
ing with the centrosome of the middle piece, a relation
which is very common in flagellate or ciliate structures.
Proximally the axial filament passes through a ring-like
structure, the annulus, in the end of the middle piece.
Distally the filament may continue beyond the cytoplasmic
envelope of the tail as a terminal filament or end piece. There
is the greatest variety in details of form of the sperm, affecting
chiefly the form of the head and acrosome, length of tail, and
total size; a few of the more common or more striking forms
are illustrated in Fig. 52. (In cases where a neck region is
distinguished, the middle piece is often regarded as the proximal
part of the tail.)

The smallest spermatozoa are found in Amphioxus and
are only 0.016-0.020 mm. (16-20 micro) in length; the largest
are found in some of the Amphibia where in Salamandra they
are about 0.7 mm. (700 micra) in length, and in Discoglossus
2.0 mm. (2000 micra) , the maximum length known. The sperm
cells of the Crustacean Cypris, are also of this gigantic size
(2 mm.). The spermatozoa of most of the Vertebrates are
25-75 micra in length. The human spermatozoon (Fig. 51)
represents about the average size; the dimensions of this are
(Krause): total length, 52-62 micra (1/400-1/500 inch); length
of head, about 4.5 micra (between 1/5000 and 1/6000 inch) ;
length of middle piece, about 6 micra (about 1/4200 inch);
length of tail piece, 41-52 micra (1/500 to 1/600 inch); width
of head 2-3 micra (1/12000 to 1/8000 inch); thickness of head,
(this is one of the few instances where the head is flattened),
about 1 micron (1/25000 inch).

The number of sperm formed by a single individual is very
large in most organisms and can be only roughly estimated.
It has been computed (Lode) that the total number formed in
man may average about three hundred and forty billion, or

Ethusa (Grobben). N. The Crustacean, Inachus (Grobben). O. The Crusta-
cean Stda (Wjnsniann) . P. The Crustacean, Bythotrephes (Weismann). k, end
knc ; m, middle piece; n, nucleus; p, perforatorium; u, undulatory membrane.
Not drawn to same scale. A-F, I-K, from Wilson.

GERM CELLS AND THEIR FORMATION 101

approximately eight hundred and fifty million sperm for each
one of the four hundred ova matured during the reproductive
period of the female. The volume of the human sperm is
roughly only about 1/195000 that of the egg (egg =0.25 mm. in
diameter). The sperm of the sea-urchin contains only about
1/400000 to 1/500000 the material in the egg (Wilson).

In several forms, both Vertebrate and Invertebrate, atypical
"giant" spermatozoa are occasionally found (Fig. 52, E). In
most instances these are abnormalities resulting from some
deviation from the usual course of events during sperm for-
mation. But in a few instances (Euschistus) a dimorphism of
the sperm seems entirely normal (Montgomery). In such cases
the larger sperm heads contain the normal amount of chro-
matin, but an excessive amount of linin and karyolymph.

In marked • contrast to the egg, the sperm cells are in very
active movement on account of the rapid vibration of the tail.
They are always contained within a fluid medium, the seminal
fluid, which is either the fluid of the cavities of the body, or a
special secretion of certain glands in connection with the
reproductive system. In the latter case this fluid is equivalent
to a tertiary egg envelope, and like this is sometimes of nutritive
value to the germ cells.

The form differences between ova and sperm give a nice
illustration of the modification of structure accompanying a
physiological division of labor, which is so well marked here.
Both cells contain equal amounts of nuclear substance, but the
ovum possesses in addition a large amount of cytoplasm, and
often a much larger amount of food substance, while the sperm
contains an amount of cytoplasm which represents practically
an irreducible minimum, and no deutoplasmic material what- $
ever. The egg, therefore, provides practically the whole of
the extranuclear substance of the developing organism. At
the same time the ovum is a passive, non-motile structure.
The spermatozoa, on the other hand, are not only extremely
motile, but they are produced in very large numbers, conditions
correlated with their function of finding the inactive ova and of
ensuring the initial stimulus to activity (development) of the

102 GENERAL EMBRYOLOGY

passive egg material. This differentiation affords the material
basis for development while at the same time it ensures the
fertilization of practically every egg produced, though the eggs
and sperm may be shed freely into the water some distance apart,
a distance often very great as compared with the size of the
cells. The relatively very large amount of cytoplasm in the
egg, and small amount in the sperm, constitute the most
marked exceptions to the nucleo-cytoplasmic (kern-plasma)
relation, mentioned in the preceding chapter; these conditions
are entirely special and are to be regarded as adaptations to
the very unusual functions of these cells.

The details concerning the form and number of the sperm and egg
cells, the amount of yolk in the eggs, the character of their membranes,
etc., are significant only from the viewpoint of adaptedness to the con-
ditions under which they must function. This adaptedness of the re-
productive phenomena toward ensuring the final . bringing to maturity
of a number of organisms sufficient to maintain the specific group
in undiminished numbers is a general biological topic of especial interest.
In strictness this lies outside our province, but to omit entirely
any reference to this subject, leaves without significance many of
the details of structure and behavior mentioned in the preceding
paragraphs. We may therefore suggest briefly a few of these relations,
not only as regards the germ cells, but also the general processes of
spawning, etc., which are all concerned in finally bringing together an
ovum and a spermatozoon.

All these varied, and often complex, phenomena of habit and morpho-
logical specialization of the reproductive cells are correlated with the
special conditions of life which affect the chances that a single egg shall
finally become a mature organism. They are conveniently grouped
under three chief heads : (a) the ensurance of mating, (b) the ensurance
of the actual meeting and fusion of the germ cells, (c) the chances of death
before maturity, involving such factors as abundance of food, enemies,
adverse conditions in the inorganic surroundings, necessity for reaching
special conditions of development, food, etc., duration of the period of
development, and the like.

A few forms, especially in the warmer climates, appear to breed quite
continuously throughout the year (many Ccelenterates, Mollusca, etc.),
but commonly the germ cells are produced at regular periods, which may
have a duration of only a few days or hours, or they may extend over
several months. Breeding or spawning periods are nearly always sea-
sonal and usually annual, but a few forms, particularly the Mammals,

GERM CELLS AND THEIR FORMATION 103

breed or spawn at shorter intervals. In many Mammals it is true that
there is only a single annual breeding season or period of oestrus; this
condition, known as moncestrous, is characteristic of most Carnivora
and is found also in the Chiroptera and Marsupials. Others, however,
are polycestrous, and exhibit two or three annual breeding seasons
(Insectivors), and in still others the period of oestrus may occur at inter-
vals of a few weeks (man), or it may be quite continuous, as in most
Rodents and some Carnivors. (See Marshall, Physiology of Reproduc-
tion, 1910.)

Among the higher animals the breeding season is often preceded by a
"nuptial season" during which, especially among the males, there may
develop various special morphological and physiological peculiarities.
The Fishes, Birds, and Mammals exhibit the frequent development of
special external markings or colorings, special secretions, and unusual
modes of behavior. Both these and the breeding habits proper, are to
be regarded as responses to stimuli, frequently climatic in origin,
resulting from changes in temperature, light, moisture, food characters,
etc. These phenomena are commonly regarded as indications of an
increased metabolism that affects not only the organs of reproduction,
but secondarily the whole body.

In a few rare instances, chiefly among the segmented worms, the an-
nual spawning season is very definitely fixed and varies within limits
of only a few calendar days. More usually the time of spawning is
subject to wide variation and is dependent upon temperature and other
seasonal conditions. The species of the Palolo worm afford one of the
best marked instances of a fixed spawning season. It is not quite as
regular and limited as tradition would have it, but in the Tortugas,
most of the individuals of the Atlantic Palolo (Eunice fucata) swarm and
spawn during one or two mornings which fall within three days of the
moon's last quarter between the latter part of June and the end of July
(Mayer). The Pacific species (E. viridis) spawns similarly on and near
the last quarter in October and November. Somewhat similar relations
have been determined for other Annulata, such as Amphitrite (Scott)
and Ceratocephale. The determining factor in these and similar cases
seems to be the character of the tides, combined with factors of tempera-
ture and light. Other organisms spawn at a definite time of day,
individuals coming to maturity at any time during a longer breeding
season. Thus Amphioxus and some Hydroids spawn only about sun
down or shortly thereafter.

During the intervals between the breeding periods the formation of
the germ cells may almost or quite cease, to recommence shortly prior
to the next period. In some creatures, however, the eggs are formed
continuously and are stored in secondary reproductive cavities pending
the time of their production. This is more likely to be the case in sperm

104 GENERAL EMBRYOLOGY

production. But in all such cases the rate of formation of the germ
cells is rhythmic, increasing just before the breeding period.

The sperm cells are always passed outside the body of the organism
forming them (save in self -fertilizing hermaphrodites); the eggs may
or may not be thus extruded. Animals used to be described (even clas-
sified) as "oviparous" or "viviparous," according to whether the female
extruded undeveloped eggs or living "young," but these terms have now
lost all precise meaning, for in any case eggs are formed, and in different
species the developing organisms may leave the body or reproductive
cavity of the parent at almost any stage.

The unfertilized eggs may be simply thrown outside the body of the
female, as in most aquatic animals, the sperm being thrown out at the
same time and in approximately the same place ; in such cases fertiliza-
tion is ensured chiefly by the production of immense numbers of sperma-
tozoa. Such a process is very common among the Sponges, Coelen-
terates, Echinoderms, Annulata, Mollusca, Fishes, and many Amphibia.
Eggs thus thrown off into the water may float at or near the surface, as
pelagic eggs, or they may sink to the bottom among >the debris (demersal) .
Or the extruded eggs may be deposited with reference to definite and
often very special conditions affording, to the new organisms, protec-
tion, food, etc. Among land animals which deposit the eggs outside the
body, these are usually very definitely placed wfth reference to such
conditions ; the Insects afford a great variety of excellent illustrations of
relations of this kind. In some cases, among both aquatic and terres-
trial forms, definite "nests" are constructed in which the eggs are
deposited, and where the newly hatched organisms may remain for
some time. The eggs and young then may or may not be guarded
or fed, by either or both of the parents. The nests may vary from
simple depressions or pockets in the mud or sand, like those of many
fresh water Fishes, to the structures of very complex architecture of
many Birds.

Among the forms which do not liberate their eggs at an early stage
in their development, there is a great variety of habit. In some Crus-
tacea and Amphibia, for example, the eggs are first extruded, but are
immediately placed upon the surface of the body, of either the male
or female parent, and develop there. Or they may become embedded
in the skin (many Amphibia) or may be deposited in some cavity not
primarily a reproductive cavity, such as the pharyngeal cavity, in some
of the Siluroid and Cichlid Fishes. In one of the Cyprinoid Fishes
(Rhodeus) the eggs are placed in the mantle cavity of a clam, where
they are fertilized and develop on the gills. In most cases where the
eggs are retained in "brood cavities," these are modified portions of
some part of the reproductive system proper ; here the eggs may remain
until a comparatively late period in their development. In such cases

GERM CELLS AND THEIR FORMATION

105

fertilization must be internal, and the sperm are then definitely intro-
duced into some reproductive cavity of the female.

An interesting series of relations may be traced illustrating the grad-
ual increase in the certainty with which fertilization shall be accom-

B

E

FIG. 53. — Forms of spermatophores. A, B, C. The Insects, Loricera, Locusta,
and an Ichneumonid. From Korschelt and Heider, after Gilson. D, E. The
Urodeles, Amblystoma and Diemyctylus. From Bertram Smith. X 3. F. Peri-
patus edwardsii (incompletely developed). After von Kennel^CKorschelt and
Heider).

plished, whether it be external or strictly internal. First, among many
of the Crustacea, Annulata, Gasteropods, Cephalopods, and Amphibia,
there is a process of pseudo-copulation or amplexus, where, sometimes

106 ^ GENERAL EMBRYOLOGY

after a complicated "courtship" (Urodeles, Jordan), the sperm are
received by the-female in or near the reproductive cavities or openings.
Usually in such instances the sperm are not scattered freely, but are
contained within definite packets or cases called spermatophores (Fig.
53) . In the Urodeles these are simple masses of spermatozoa enclosed
in a thin envelope; they are discharged by the male and then picked up
by the cloacal margins of the female and stored until the eggs are ready
to be fertilized. In the amplexus of the earthworm and many Gastero-
pods, there is a mutual exchange of spermatozoa between two herma-
phroditic individuals, the sperm being received into storage cavities
and retained until the eggs are deposited. Such a receipt of sperm or
sperm packets into storage cavities is quite common, and the sperm
may in these cases remain alive for long periods, even for three or four
years (honey bee, some snails). The spermatophores are sometimes
very elaborate affairs containing a complex mechanism arranged so as
to discharge the sperm just at the time the female is depositing the eggs
(Fig. 53, C). Among the more complex are those formed by the male

FIG. 54. — The Trematode, Bilharzia hcematobia. Two individuals living in
copula. After Fritsch. X 14. o, ova in oviduct, tf, male; 9 , female.

squid (Loligo) and transferred to the buccal membrane of the female,
where they remain attached pending the time of spawning (Drew) .

Among forms where internal fertilization is the rule, this is more
frequently the result of a definite act of copulation, by which the sperm
are transferred directly to a reproductive cavity of the female through a
male intromittent organ . This occurs in most Insects, many Turbellaria,
Crustacea, Molluscs, and in most of the higher Vertebrates. This gen-
eral relation, carried to an extreme may result in symbiosis, or even in
the parasitic character of the male upon or within the body of the female.
Thus in some of the Cirripedia, degenerate " complemental " males are
found living semiparasitically within the body of the female. Several
of the Trematoda live in pairs within a single cyst. In Bilharzia (Tre-
matoda) the female lives permanently in a groove on the body of the

GERM CELLS AND THEIR FORMATION 107

male (Fritsch) (Fig. 54). In Diplozoon (Trematoda) two hermaphrodite
individuals, at first entirely separate, become permanently fused so that
the openings of the reproductive ducts are in apposition. In Syngamus
(Nemathelminthes) also, the male lives permanently attached to the
female. The climax of this relation is afforded by Trichosomum (Nema-
thelminthes) where several (2-5) dwarfed males live within the uterus
of the female (Leuckart). Or it might be said that the climax is to be
seen in the cases of self-fertilizing hermaphrodites; these are usually
internal parasites, where the chances of the meeting of males and
females would be practically nil (e.g., many of the Trematodes and
Cestodes) .

The retention of the eggs during their development, within a brood
cavity is primarily a protective arrangement, but it often leads to the
establishment of an organic nutritive relation between the embryos and
the wall of the cavity. This is the case in some Tunicates, Elasmo-
branchs, etc., and of course it reaches its climax in the intimate and
extensive organic relation between embryo and oviducal (uterine) wall
among the Eutherian Mammalia.

The number of eggs produced during each reproductive or spawning
period varies enormously, and is related to a variety of conditions of
development. In general the number of eggs is larger when there are
few or .no other means of ensuring the complete development of a
number of organisms sufficient to maintain the species numerically.
Any structure of the egg or habit of deposition, adapted to ensure
development, is likely to be associated with a reduction in the number
of eggs formed. The number is largest among forms which discharge
the eggs at random or where they are subject to unfavorable external
conditions, to liability to the attacks of parasites, or use as food by other
organisms. For example, the marine fishes produce very large numbers
of pelagic ova, the codfish is said to form 8 to 10 millions in one season ;
and in a species of sea-urchin (Echinus') a single female is said to have
discharged, in a single season, as many as 20 million ova. Where very
special conditions of development are essential, as in the complicated
life histories of many internally parasitic worms, the number of eggs is
very large and fertilization is ensured by hermaphroditism.

The number of eggs is smallest where they develop within a brood
cavity, or where some degree of parental care is exercised. In a few
Mammals and Birds only a single egg is formed at each breeding period,
and in these groups the number rarely exceeds eight or ten. Further
the number of eggs produced, in general varies inversely with the
amount of food substance contained, or with the chances of the young
.finding food for themselves by the time they become free living.

The number of sperm cells formed is always larger than the number
of eggs, and often reaches many millions. The number is likely to be

108 GENERAL EMBRYOLOGY

smaller among forms in which the sperm are directly or indirectly intro-
duced into the reproductive cavities of the female. Some of the Crus-
tacea afford interesting illustrations of this. In some Ostracods only a
few hundred very active spermatozoa are formed; these are inserted
directly into the seminal receptacles of the female. They are most
remarkable for their size— 2 mm. in length in a few, or more than twice
as long as the body of the male. In Daphnia the number of sperm may
be only twenty, or even less, six to eight in some species. These too are
liberated directly into the brood cavity of the female, which forms only
two eggs at a time; these sperm are very adherent, and are said to be
somewhat amoeboid. Indeed the spermatozoa of many of the Crustacea
are unusually interesting on account of their atypical form and behavior
(Fig. 52). Some (Bythotrephes) are quite like ova, large (0.1 mm.),
rounded, and quiescent, depending upon a peculiar viscid or adherent
quality for their likelihood of attachment to the egg. Others (some
Decapods) have a number of stiff radiating processes which seem to
function by catching in the hair-like bristles surrounding the openings
of the oviducts, where they are placed in amplexus.

The amount of food yolk contained in the egg is related also to the
duration of the embryonic period of development, or to the rate of
development, a prolonged embryonic life requiring an abundant supply
of food materials, and an unusually rapid rate of development depending
upon a supply of easily assimilable nutritive substance. Such a relation
is illustrated by the difference between summer and winter eggs, formed
by Rotifers, and many Insects and Crustacea; the winter eggs, subject
to unfavorable conditions and passing a longer period in development,
contain considerably more yolk, and are covered with much tougher
and more resistant membranes, than the summer eggs which develop
rapidly and under favorable surroundings, indeed often within the brood
cavity of the female. Thus in Daphnia the small summer eggs are
formed by only three nurse cells (see below), while the large winter eggs
are supplied with food by forty or more nurse cells. When the developing
embryo acquires special nutritive relations with the parental tissues,
the eggs are of course practically yolkless.

The provision of egg membranes is associated not only with a reduction
in the number of eggs formed, but also with the duration of the em-
bryonic period, liability to unfavorable external conditions, prevalence
of food-seeking enemies, etc. The membranes which are functional
under such conditions are of the secondary and tertiary classes described
above. The nature of these varies from the common thin fibrous
coverings, to tough and impervious membranes capable of resisting
extreme dryness, or the leathery or calcareous " shells " of the Sauropsids.
Among the most complex and perfectly adapted membranes or shells,
are the egg cases of many Elasmobranchs, and particularly those of

GERM CELLS AND THEIR FORMATION 109

the Holocephali (Dean) (Fig. 55), which often remain intact and func-
tional, in their passive way, for more than a year. In some cases the
egg membranes may have a nutritive value and may augment or re-
place the food supply in the form of yolk ; the eggs
of birds and snails are good illustrations of this
relation.

We must now trace briefly the steps leading
to the formation of the ova and spermatozoa
as the highly specialized cells we have de-
scribed. In the lowest Metazoa, Sponges and
some Hydroids, the germ cells are scattered
through the tissues of the organism as sepa-
rate, free cells, which may migrate from place
to place, feeding and growing, often at the
expense of the other cells (Fig. 56). But in
other Hydroids, and in all forms above these,
the germ cells are localized in a definite re-
productive tissue and organ, or series of
organs, the gonads — ovaries and testes. The
simplest gonads are merely masses of rapidly
proliferating cells (Fig. 57), usually bordering
a cavity which is the ccelom, and which is
supposed to be primarily this reproductive
cavity simply. In most of the higher forms
the ccelom comes to have many secondary re-
lations, and forms in addition to the repro-
ductive and other smaller cavities, the very
extensive body cavity. In the embryos of
the Craniates there is a pair of longitudinal
ridges, either side of the attachment of the ^^eof thT^o"
dorsal mesentery, through a considerable ex- cephaian, Chimcera

r , -, T_i •, ,1 ,1 .. 7 monstrosa. Ventral

tent of the body cavity; these are the genital aspect. Two-thirds
ridges, and the peritoneum covering these be- j^1™1 size> After
comes thickened by the enlargement and pro-
liferation of the cells (Fig. 58). These are the rudiments of
the gonads. The cells composing these rudiments are often of
two kinds. Some of them, indifferent cells composing in gen-

110

GENERAL EMBRYOLOGY

FIG. 56. — Origin of the germ cells in the Hydro-medusa, Cladonema. From
Wilson, "Cell," after Weismann. A. Young stage:' section through the wall of
the manubrium. Ova developing in the ectoderm, ec; en, endoderm. B. Older
stage, showing ova, o, and nutritive cells, n. The ova contain small nuclei prob-
ably derived from ingested nutritive cells.

Fio. 57. — Diagram of the structure of the developing ovary of the Annulate,
Amphitrite rubra. From Korschelt and Heider, after E. Meyer, g.dr., rudiment
of ovary; Q.C., germinal epithelium; g.z., fully formed ova, scattering; pm., perit-
oneum; V.v., vas ventrale.

GERM CELLS AND THEIR FORMATION

111

eral the frame-work of the gonad, have been formed in situ
from the proliferating peritoneal cells. Others, the primordial
germ cells, are often first distinguishable in some other region of
the developing embryo (Fig. 59, A); they then make their way
into this germinal epithelium as development proceeds.

The gonad may consist almost entirely of the reproductive
cells proper, and may be then a more or less periodic structure,

FIG. 58. — A. Part of a section through the body of a young lizard, Lacerta
agilis, showing the genital ridges and associated structures. B. Genital ridge,
enlarged, showing young follicles containing ova. From Korschelt and Heider,
after Braun. ao, dorsal aorta; v, cardinal vein; ms, mesentery.

almost if not quite disappearing between the periods of repro-
ductive activity. Or it may be a permanent organ of con-
siderable though varying size, consisting of a complex stroma
of a variety of cells — nutritive, vascular, nervous, connective
tissue, and other cells, in addition to the true germ cells. When
highly developed the gonad may also contain various cavities,
of coelomic character, into which the germ cells are passed when
ripe. Thence they may pass directly into the body cavity
from which exit is made to the outside through simple perfora-

112

GENERAL EMBRYOLOGY

tions in the body wall or through special tubes or ducts. In
the testes these ducts may lead directly to the outside from the
cavities of the gonads. The structure of the gonad can nearly
always be reduced to that of a complexly folded epithelium
the essential elements in which are the germ cells; the coelomic
surface is the free surface of the germinal epithelium. This
relation becomes important in describing the fundamental
morphology of the germ cell.

It is a question whether the germ cells are to be considered as
originally undifferentiated cells, which become modified during
the life of the organism for the reproductive function, or
whether they are set apart from the beginning of the organism's
multicellular existence as reproductive cells, and become visibly
modified only in later stages. In those few forms where the

B

FIG. 59. — A. Section through an early embryo of the Teleost, Micrometrus
aggregatus, showing the distinct germ cells. After Eigenmann. ec, ectoderm;
en, endoderm; g, germ cells; so, somatic layer of mesoderm; sp, splanchnic layer
of mesoderm. B. Section through forty-cell stage of the Crustacean, Cyclops
brevicornis, showing, g, the cell that gives rise to the germ cells, en, the primi-
tive endoderm cell in the process of its first division. After Hacker.

germ cells are diffused it seems that any tissue cell which is not
completely specialized in some other direction may assume
reproductive characters. In forms which develop special
gonads, however, there are many reasons for believing that the
germ cells are always to be distinguished as such very early in
the history of the individual organism/7 In Ascaris the history
of the primordial germ cells has been traced back to one of the
two cells resulting from the very first division of the fertilized
ovum; not all of the descendants of this cell are germinal how-

GERM CELLS AND THEIR FORMATION

ever (Fig. 32). In some of the bony fishes germinal cells are
recognizable in the fifth cell generation, i.e., in the thirty-two
cell stage. And in many other forms, including some of the
Mammalia, the germinal cells can be distinguished from the
somatic cells very early, even in the blastula (Fig. 59). This
may indicate that, although not visibly distinct, the germ
tissue is, after all, in reality distinct from the somatic, in most
if not in all forms. It would seem more consistent with the
present conception of development, however, to say that this
distinction exists only potentially and comes about as a real
differentiation in the developing organism, for however early
this differentiation may occur, a stage is always found where
germinal and somatic substances are contained undifferentiated
within a single cell and are then indistinguishable.

The visible distinction between the gonads of different sexes
may occur very early. In some forms this distinction between
sexes can be made out in the fertilized ovum. And in many
forms the two kinds of gonads can be distinguished soon after
they are first marked out, though there is reason even here for
supposing that the distinction is really, though not visibly,
present in the fertilized egg.

The processes involved in the later differentiation or histo-
genesis of the eggs and spermatozoa are collectively termed
odgenesis and spermatogenesis respectively. They are con-
veniently divided for description into three periods or phases.
These are (1) the period of cell multiplication, during which the
simple epithelial cells, or primordial germ cells, divide more or
less continuously, increasing the bulk of the gonad; (2) the
period of growth, when cell division is less rapid or altogether
inhibited, and the cells enlarge rapidly, the egg-forming cells
much more considerably than those forming the spermatozoa;
(3) the period of maturation, when the germ cell nuclei undergo
profound modifications during their last two divisions as germ
cells. Sometimes the terms oogenesis and spermatogenesis are
used to indicate only the events of this third period, which are
of such importance that we shall make them the subject of the

114

GENERAL EMBRYOLOGY

next entire chapter. In the history of the spermatozoa a fourth
period is to be distinguished, namely, the period of transforma-
tion or metamorphosis; for the highly differentiated structure of
the spermatozoon is rapidly assumed after the process of
maturation is completed. This period is not marked in the
history of the ovum, for this, with the exception of its unusual
size, is not so markedly differentiated in structure.

During the first of these periods, that of multiplication, the
cells of the reproductive tissue are termed oogonia and sper-

Primordial Germ

Cell ("Primitive

Ovum").

Period of Multiplica-
tion . « c h r o m o-
somes. (The num-
ber of cell genera-
tions is much
greater than indi-
cated here.)

Period of Growth,
a chromosomes.

Period of Matura-

tion .

c h r o m o-

Oogonia.

Primary Oocyte.

Secondary Oocyte
and First Polar Body.

Mature Ovum and
Three Polar Bodies.

FIG. 60. — Diagram of the chief events of 06 genesis. Adapted from Boveri.
Compare with Fig. 61.

and of these there may be a great many generations
during this period, before growth commences. As the oogonia
and spermatogonia become older, division becomes slower and
ceases as the cells enter upon their growth period. At the close
of the growth period, while still contained within the ovary or
testis, the cells are known respectively as the primary ooeytes,
or ovarian eggs, and the primary spermatocytes. From this
point onward the histories of the eggs and sperm are not quite
identical, although entirely equivalent (Platner, O. Hertwig).

GERM CELLS AND THEIR FORMATION

115

'As said above, the chief events concern the nuclear structure
and we can only point out here that during the period of matu-
ration two more cell divisions occur.

In the testis each primary spermatocyte divides once, form-
ing two cells called the secondary spermatocytes, and then each
of these divides again, forming altogether four cells called the

P r i m o rdial Germ
Cell.

Period of Multiplica-
tion, s chromo-
somes. (The number
of cell generations
much greater than is
indicated here.)

Spermatogonia.

Period of Growth.
s chromosomes.

Period of Matura-
tion. — chromo-

Period of Metamor-
phosis. Tr

Primary Spermato-
cyte.

Secondary Sper-
matocytes.

Spermatids.

Spermatozoa.

FIG. 61. — Diagram of the chief events of spermatogenesis. Adapted from

Boveri.

spermatids. These are all alike and each becomes metamor-
phosed into a spermatozoon without further division. Thus
are formed, from each primary spermatocyte, four similar
spermatozoa. In the ovary, or sometimes after the primary
oocyte has left the ovary, this too divides, but here the division
is very unequal, resulting in the formation of one large cell,
the secondary oocyte ^ and one very small cell, called the first
polar body. Typically each of these then divides again. The
secondary oocyte divides unequally as before, forming a large

110

GENERAL EMBRYOLOGY

cell the mature ovum, and another small cell, the second polar
body. Meanwhile the first polar body divides equally forming
two similar polar bodies. In some cases the division of the
first polar body is suppressed. Thus each primary oocyte
typically gives rise, like the primary spermatocyte, to four cells,
but these are not all alike in form and size, although they are
fundamentally equivalent, i.e., homologous, to each other, and
to the four spermatids. Of these four cells, however, only one,
the ovum, is functional; the_£olar^bodies degenerate without
functioning. The parallel eventsoT^spermatogenesis and
oogenesis^are shown diagrammatically in Figs. 60-61. (See
also Figs. 76, 90, 94.)

Ogrt.

6v

FIG. 62. — Longitudinal section through the ovary of the Copepod, Cantho-
camptus. From Wilson, "Cell," after Hacker, og, the youngest germ-cells or
oogonia (dividing at og.2) ; a, upper part of the growth-zone; oc, oocyte.^or growing
ovarian egg; ov, fully formed egg, with double chromatin-rods.

All these processes may be going on in the ovary or testis at
the same time, occurring progressively from the basement
membrane of the germinal epithelium toward its free surface,
so that a section through such an epithelium shows practically
every step in the history of a single cell (Figs. 62, 68, 69).
Before considering in detail the nuclear changes involved in
the maturation processes we must consider the more important
facts concerning the history of the cytoplasmic parts during
this phase of the genesis of the germ cells.

GERM CELLS AND THEIR FORMATION

117

FIG. 63. — Diagrams of the egg-tubes (ovaries) of Insects. After Korschelt
and Heider. A. Orthoptera, without groups of nutritive cells. B. Coleoptera,
with many such groups. C. Hemiptera, with terminal nutritive group and
nutritive channels extending to the ova. c, nutritive channel; /, egg follicle;
m, zone of multiplication; n, nutritive cells; o, ova.

118

GENERAL EMBRYOLOGY

In the growth of the egg the chief aspects are those associated
with nutritive relations of the developing ovum to the adjacent
cells, especially in those forms whose eggs contain a considerable
amount of yolk. In the non-localized ovary such as that of the
Sponges and some Hydroids, the ovum grows at the expense
of whatever cells happen to be adjacent to it (Fig. 56). In the
common Hydra, as the ovum grows to be
considerably larger than these tissue cells,
it becomes amoeboid and actually ingests
these neighboring cells, digesting their sub-
stance and growing rapidly (Fig. 44) . The
nuclei of these ingested cells are relatively
indigestible and remain for some time scat-
tered through the egg cytoplasm. Among
all those forms with definitely localized
ovaries growth of the ova is accomplished
very differently. When the eggs are small
and contain relatively little food, no
special nutritive mechanism is developed,
the egg forming the food substances in its
own cytoplasm from materials drawn from
the circulating fluids in the cavities of
the ovary. Such eggs develop independ-
ently of the .neighboring cells intermingled
with the ova.

FIG. 64. — Longitudi- TST-U J.T r n i* i i

nai section through VYnen the tully lormed eggs contain large
^be^T amounts of food substance this is usually
marginaiis. obtained by one of two chief methods. In
Ko7Chnttritive the simplest cases certain of the ovarian
ceils; o, ovum contain- 'cells adjoining the egg take on the charac-

mg amoeboid nucleus .... ,.

partly surrounded by tenstics ot nurse cells. These may either
substance contribute their own substance directly to
the ovum or they may become intensely
active, forming deutoplasm which is then drawn from them by
the growing ovum (Figs. 63, 64). There may be a single
nurse cell for each ovum (Fig. 65), or the nurse cells may
be scattered irregularly through the ovary so that several

GERM CELLS AND THEIR FORMATION

119

may be related to each ovum. The nurse cella in many or
even in most cases, are cells which were potentially germ
cells, but which have lost their germinal potentiality and

FIG. 65. — Sections through ovarian ova and nurse cells in the Annulate,
Ophryotrocha. From Wilson, "Cell," after Korschelt. A. Young stage, the
nurse cell, n, larger than the egg. B. Growth of the ovum, o. C. Late stage,
the nurse cell degenerating.

become wholly nutritive, contributing to the formation of
the true germ cell, degenerating and disappearing completely
after the ovum is grown and has left the ovary. In the other

A B

FIG. 66. — Sections through ovarian eggs and their follicles in, A, young magpie;
B, newly born child. From Wilson, "Cell," after Mertens.

cases, which are commonest among the Vertebrates, almost uni-
versal among them in fact, the ovarian cells adjoining the ovum
— themselves potentially germ cells originally, form around the
ovum a definite layer termed the follicle (Fig. 66). The follicle

120 GENERAL EMBRYOLOGY

cells have the arrangement of an epithelium; they may form
either a single layered, simple epithelium or in other cases a
many layered stratified epithelium (Fig. 43). They not only
provide for the nutrition of the enlarging ovum, with ^vhich
they are frequently connected by definite intercellular proto-
plasmic strands, but toward the close of the growth period of

the egg they may become secre-
tory and form certain egg envel-
opes of the secondary type, i.e.,
chorionic. We have already men-
tioned how the micropyle is
formed through the chorion and
vitelline membrane by the inser-
tion of one of these follicle cells
with a long process, preventing the
membranes from forming at that
point. When the eggs are fully
formed and ready to be laid the
follicle ruptures 'allowing the eggs
to escape freely. Often there de-
velops in the follicle a definite
region along which it bursts; this
weakened region is called the
cicatrix (e.g., common fowl). In
a few instances some of the follicle
cells are actually taken into the
egg and absorbed, much as in the case of those ova which in-
gest adjacent interstitial cells. This is the case in many of
the Tunicata where some of the so-called "test-cells" lose
their cell outlines and are directly taken into the cytoplasm
of the egg; their nuclei remain distinct for some time (Fig. 67).
The remaining test cells then form a distinct follicle outside
the whole structure. These nuclei no longer function as
nuclei of course; as the growth of the egg is completed they
are extruded again, along with a portion of the superficial pro-
toplasm forming then a thick vitelline membrane resem-
bling a chorion and often so called.

FIG. 67. — Section through the
egg of the Tunicate, Cynthia
partita. After Conklin. In the
periphery of the egg are the nuclei
of the test cells which have been
ingested by the ovum (for their
later history see Fig. 91). X 357.
e, exoplasm or cortical layer; n,
egg nucleus or germinal vesicle,
with large nucleolus; t, nuclei of
test cells or follicle cells; y, yolk.

GERM CELLS AND THEIR FORMATION 121

Turning now to the formation of the sperm we find, as we
should expect, processes on the whole entirely comparable
with "those of egg formation. The chief differences result from
the fact that in the ovary conditions are associated with the
fomjajj^^ while in the testis

small !JeTIs~aref ormed, but in very large numbers. In Sponges
and Hydroids there is the same non-localized formation of the
sperm as of the ova, the germ cells being distinguished not so
much by position as by size. Apparently any ectoderm cell
may enlarge and become reproductive. In all forms above
these there are definite testes. Among many Ccelenterates
and Echinoderms the testis is composed purely of germ cells, but
usually the testis, like the ovary, contains other interstitial or
accessory cells, and frequently these are directly nutritive in
function. The general structure of the testis differs from that
of the ovary in that its epithelium is thrown into folds, forming
either simple columns or acini, each with an efferent duct or
pathway which is to be considered ccelomic in origin. ' In the
testicular epithelium, which is ordinarily reducible to the strati-
fied type, we find sperm cells in all stages of formation, from
spermatogonia to fully formed spermatozoa./ As we have
already said, the process of sperm formation is usually continu-
ous, though frequently periodic (seasonal) in its rate or inten-
sity. When the testis is composed of acini, each is usually
surrounded by a follicle, equivalent in function to the egg folli-
cle. But when arranged in lobules and columns, each of which
may be derived from a single primordial cell or prespermatogon-
ium, the nutritive cells do not show this follicular arrangement,
but are fewer in number and scattered along the basement
membrane of the epithelium, or along the connective tissue
partitions bounding the spermatic columns of the lobule, and to
some extent among the germ cells proper.

Among the Craniates the typical arrangement is that shown
in Figs. 68, 69. Here, along the basement membrane, are several
generations of spermatogonia with the scattered_jiatritive^
basal cells, sometimes called also the SejrtoU_cells, usually larger
than the spermatogonia. As the spermatogonia increase in

122 GENERAL EMBRYOLOGY

number, through continued mitosis, they begin to increase in
size, though not nearly to the same extent that the oogonia do.
There is not always the same distinctness between the phases
of multiplication and growth here, and the two final divisions
of the full grown spermatogonial cell, then known as the

FIG. 68. — Diagram of a section through part of the testis of the rat, showing
some stages in spermatogenesis. From Korschelt and Heider, after Lenhossek.
bz, basal cells; spc, spermatocytes; spg, spermatogonia; spt, spermatids; spz,
spermatozoa.

primary spermatocyte, are the reducing or maturation divisions.
These lead, as we have seen, to the formation, from each primary
spermatocyte, first of two secondary spermatocytes, both alike,
and then to four spermatids, all alike. The cells of the column
then become arranged so that groups of spermatids become
related with each of the basal cells, which often leave their

GERM CELLS AND THEIR FORMATION

123

1

original position and move out
toward the free surface of the
epithelium. Through their at-
tachment to the basal cells the
spermatids draw a supply of
food and energy during their
rapid and extensive metamor-
phosis into spermatozoa. In
some cases a very close relation
is established by the actual 7
embedding of one end of the
spermatid in the substance of
the basal cell. It should be
noticed that the function of the 6
follicle or basal cells of the
testis is to supply nutrition to
the germ cells, not so much
during their period of growth as
after that is completed, during 5
the period of metamorphosis;
while in the ovary the corre-
sponding cells function during
the growth period; this is cor- 4
related with the smaller size
of the spermatocyte and with

FIG. 69. — Diagrammatic outline of
the spermatogenesis of the rat in thirty-
two stages. After v. Ebner. Base-
ment membrane toward the left. 1-8.
Period of multiplication (the number
of cell generations is actually very large).
9-18. Period of growth. 19-24. Period
of maturation. 25-32. Period of meta-
morphosis. 6, basal cells or Sertoli
cells. 1-16. Spermatogonia. 17, 18.
Primary spermatocytes preparing for
division. 19. First spermatocyte divi-
sion. 20. Secondary spermatocytes.
21. Secondary spermatocyte division.
22-25. Spermatids. 26-31. Trans-
formation of spermatids. 32. Fully
formed spermatozoa.

32

124

GENERAL EMBRYOLOGY

the need for an easily available food supply ^for the large
number of sperm cells during their metamorphosis.

Probably the most interesting phase of the cytoplasmic as-
pects of spermatogenesis is this metamorphosis of the sper-

FIG. 70. — Formation of the spermatozoon in Urodeles. From
"Cell," A-E, Salamandra, after Meves; F-K, Amphiuma, after McGregor.
A. Spermatid with peripheral pair of centrosomes (c) lying outside the sphere,
and axial filament. B. Centrosomes near the nucleus, outer one ring-shaped;
a, acrosome. C. Inner centrosome inside the nucleus, enlarging to form middle
piece; n, nucleus. D. Portion of much older spermatid, showing divergence of
two halves of the ring, r. E. Portion of mature spermatozoon, showing upper
half of ring at r, and the axial filament proceeding from it. F. Spermatid of
Amphiuma, showing sphere-bridges and ring-shaped mid-bodies. G. Later
stage; outer centrosome ring-shaped, inner one double; sphere, s, converted into
the acrosome, a. H. Migration of the centrosomes. /. Middle-piece at base
of nucleus. J. Inner centrosome forms the end-knob within the middle-piece,
which is now inside the nucleus. K. Enlargement of middle-piece, m, end-knob
within it; elongation of the ring.

matids into spermatozoa. After growth and maturation the
sjDgrmatids have much the same external appearance as any
typical cell; they are more or less spherical cells with a pair of

GERM CELLS AND THEIR FORMATION 125

centrosomes or centrioles, and a large spherical nucleus with
a dense chromatin network. Internally we know that the
nuclei are unlike those of the somatic cells on account of the

o

presence of only - chromosomes. Without any further divi-
sion each is converted into the special form of the spermatozoon
typical of the species. While the details of this metamorphosis
vary considerably in different groups, the essentials of the
process are everywhere the same. The spermatid (Figs. 70, A, F;
71, A) contains, in addition to the typical cell organs just men-
tioned, a modified region of the cytoplasm which is sometimes
a centrosphere or idiosome, sometimes of rather doubtful charac-
ter and origin, which for convenience may be termed the sper-
matosphere. Close by lie the remains of the last division spindle.
The spermatids are further characterized by the presence of a
collection of chromidial structures termed the mitochondria
(Fig. 71).

The details of the metamorphosis of these structures into the
parts of the spermatozoon are subject to wide variation; the
following account is based upon the history of the mammalian
spermatid (Fig. 71)ij}The first step in the process is the migra-
tion of the centrosomes to the surface of the cell, and at the
same time the migration of the nucleus to the opposite side of
the cell. In most cases it is difficult to determine the relation
of the axis thus marked out, but in many instances this is per-
pendicular to the basement membrane of the germinal epithe-
lium, thus expressing a polarity which coincides with that of
the ovum; the nucleus lies toward the membrane, i.e.j the
attached surface of the cell, the centrosomes toward the free
surface. As the centrosomes and nucleus are taking these new
positions the spermatosphere moves up to the nucleus and
around it to the side opposite the centrosomes, quite to the

I surface of the spermatid. The two centrosomes now separate,
me approaching the nucleus, the other remaining peripherally.
Following these changes in the relative positions of the parts
;ome the real modifications of structure. The nucleus becomes
elongated or ovoid, and the chromatin condenses, first into a

126

GENERAL EMBRYOLOGY

FIG. 71. — Metamorphosis of the spermatic! of the guinea pig, Cavia cobaya.
After Meves. L. Side view, showing flattening of head, a, axial filament;
c, centrosomes (centrioles) ; e, neck, containing end knobs (proximal centriole) ;
k, chromidial "nebenkorper;" ra, middle piece; n, nucleus; s, centrosphere (aero-
some) ; u, annulus (posterior portion distal centrosome) ; 1), mitochondria (partly
becoming a portion of the tail envelope) ; y, cytoplasmic portion of spermatid,
being thrown off in K and L; in J, K, L containing mitochondrial remains.

GERM CELLS AND THEIR FORMATION 127

heavy reticulum, and then into a dense mass in which no visible
structure remains; finally it acquires the form of the head of
the mature spermatozoon. The spermatosphere meanwhile is
largely converted into the acrosome or perforatorium, at the
tip of the elongated nucleus; a smaller portion is transformed
into a very delicate envelope covering a part of the head, in
many cases scarcely distinguishable on account of its thinness.
At the other end of the cell a fine flagellum begins to grow out
in connection with the peripheral centrosome, either from the
substance of the centrosome itself, or from the cytoplasm under
its influence. Then the two centrosomes move farther in
toward the nucleus. In the simpler cases the distal centro-
some now divides into anterior (toward the head) and posterior
portions. The posterior part grows out peripherally into a
rapidly elongating fiber which becomes the axial filament of the
flagellum, while at its base it becomes ring-like; then through
this ring or annulus the axial filament grows in, finally connect-
ing with the anterior portion of the centrosome. The anterior
portion itself remains in the middle piece as a posterior centro-
some of the spermatozoon. The proximal or anterior centro-
some partly disappears, and partly is converted into that part
of the middle piece which connects with the head (the neck).

The cytoplasmic part of the spermatid seems to be largely
consumed in providing the energy for this transformation, in
addition to that drawn from the nurse or Sertoli cells. But the
cytoplasmic membranes of the middle piece and tail, including
the undulatory membrane when this is present, are derived
directly from the cytoplasm of the spermatid. The mitochondria
of. the spermatid seem to be transformed into the spiral layer
of the middle piece. In many instances, particularly among
the Mammals, the larger part of the cytoplasm remains for a
time connected with the middle piece of the developing sper-
matozoon and then is cast off, and finally degenerates without
taking any further part in the structural formation of the
functional sperm cell. The chief structural correspondences
between spermatid and spermatozoon are shown in the accom-
panying table.

128

GENERAL EMBRYOLOGY

COMPARISON OF THE STRUCTURES OF THE SPERMATID AND
SPERMATOZOON

With particular reference to the mammalian condition. (Partly from

Gegenbaur-Fiirbrmger, Lehrbuch der Anatomic des Menschen,

Leipzig, 1909)

SPERMATID
Nucleus.

Spermatosphere (centro-

sphere) .
Proximal centrosome (cen-

triole) .

Distal centrosome (cen-
triole) .

(a) Anterior portion.

(b) Posterior portion.

Cell body.

Mitochondria.

SPERMATOZOON

Head.

Acrosome (perforaijorium) and sheath
covering the anterior part of the head.

Forms an undifferentiated part of the
middle piece; in Mammals, the neck.
In part may disappear.

Centrosome of middle piece. •
Annulus and axial filament of middle
piece and tail.

Partly used as source of energy during
metamorphosis. Partly thrown off.
Remainder forms cytoplasmic envel-
opes of middle piece and tail (includ-
ing undulatory membrane) .

Spiral membrane of middle piece.

Whatever the details of the metamorphosis of the spermatid
may be, the facts of essential importance are always identical.
These are, that the nucleus of the spermatid is directly trans-
formed into the head of the spermatozoon; the centrosomes
of the spermatid become the centrosomes and kinoplasmic
structures of the spermatozoon and are contained within the
middle piece, or partly in the tail; the cytoplasm of the sper-
matid in part goes to form a thin cytoplasmic investment of the
spermatozoon or is in part cast off.

The fully formed spermatozoa now lose connection with the

GERM CELLS AND THEIR FORMATION 129

nurse cells and pass by way of the canals or ducts of the testis
into the efferent reproductive ducts, vasa efferentia and vasa
deferentia, to the outside, either directly, or after being stored
for a time in special cavities such as the seminal vesicles. The
sperm may remain alive in these storage cavities for a long
time, awaiting the period of extrusion, upon the approach of
which they become very active.

REFERENCES TO LITERATURE

AMMA, K., Ueber die Differenzierung der Keimbahnzellen bei den

Copepoden. Arch. Zellf. 6. 1911.
BALFOUR, F. M., Treatise on Comparative Embryology. London.

1880, 1881.
BALLOWITZ, E., Untersuchungen iiber die Structur der Spermatozoen,

etc. Arch. mikr. Anat. 32. 1888. 36. 1890. Also Zeit.

wiss. Zool. 50. 1890. 62. 1891.
BARDELEBEN, K. v., Dimorphismus der mannlichen Geschlechtszellen

bei Saugetieren. Anat. Anz. 13. 1897. Weitere Beitrage zur

Spermatogenese beim Mensch. Jena. Zeit. 31 (24). 1898.
BEARD, J., The Germ Cells. I. Raja batis. Zool. Jahrb. 16. 1902.
BOVERI, T., Zellenstudien. I. Die Bildung der Richtungskorper bei

Ascaris megalocephala und A scaris lumbricoides. Jena. Zeit. 21

(14). 1887. Die Entwicklung von Ascaris megalocephala mit

besonderer Rucksicht auf die Kernverhaltnisse. Festschr. f.

von Kupffer. Jena. 1899.
CONKLIN, E. G., The Organization and Cell-Lineage of the Ascidian Egg.

Jour. Acad. Nat. Sci. Philadelphia. 13. 1905.
DEAN, B., The Chimseroid Fishes and their Development. Publ.

Carnegie Inst. No. 32. Washington. 1906.
EIGENMANN, C., On the Egg Membranes and Micropyle of some Osseous

Fishes. Bull. Mus. Comp. Zool. Harvard Coll. 19. 1890. On

the Precocious Segregation of the Sex-Cells in Micrometrus aggrega-

tus. Jour. Morph. 6. 1891.
FRITSCH, G., Zur Anatomic der Bilharzia hcematobia, Cobb. Arch.

mikr. Anat. 31. 1888.
HACKER, V., Die Eibildung bei Cyclops und Canthocamptus. Zool.

Jahrb. 6. 1892.
HERFORT, K., Die Reifung und Befruchtung des Eies von Petromyzon

fluviatilis. Arch. mikr. Anat. 67. 1901.
HERTWIG, 0., Experimentclle Studien am tierischen Ei vor, wahrend

und nach Befruchtung. Jena. Zeit. 24 (17). 1890. Vergleich

der Ei- und Samenbildung bei Nematoden. Eine Grundlage fur

cellulare Streitfragen. Arch. mikr. Anat. 36. 1890.

130 GENERAL EMBRYOLOGY

HOEFER, P. A., Beitrag zur Histologie der menschlichen Spermien und

zur Lehre von der Entstehung menschlicher Doppel (miss) bildung.

Arch. mikr. Anat. 74. 1909.
JORDAN, E. O., The Spermatophores of Diemyctylus. Jour. Morph.

6. 1891.
KORSCHELT und HEIDER, Lehrbuch der vergleich. Entwick. der wir-

bellosen Thiere. II. Abschnitt. Die Geschlechtszellen, ihre

Entstehung, Reifung und Vereinigung. Jena. 1902-1903.
LENHOSSEK, M. v., Untersuchungen liber Spermatogenese. Arch.

mikr. Anat. 51. 1898.
LILLIE, F. R., Studies of Fertilization in Nereis. I. The Cortical

Changes in the Egg. II. Partial Fertilization. Jour. Morph. 22.

1911.
MCGREGOR, J. H., The Spermatogenesis of Amphiuma. Jour. Morph,

16. Suppl. 1899.

MARSHALL, F. H. A., The Physiology of Reproduction. London. 1910.
MAYER, A. G., The Atlantic Palolo (Eunice fucata) . Sci. Bull. Mus.

Brooklyn Inst. Arts and Sci. 1. 1902.
MEISENHEIMER, J., Entwicklungsgeschichte von Limax maximus.

I. Th. Furchung und Keimblatterbildung. Zeit. wiss. Zool. 62.

1896.
MEVES, F., Ueber die Entwicklung der mannlicher' Geschlechtszellen

von Salamandra maculosa. Arch. mikr. Anat. 48. 1897. Ueber

Struktur und Histogenese der Samenfaden des Meerschweinchens.

Arch. mikr. Anat. 64. 1899.
MONTGOMERY, T. H., JR., The Spermatogenesis of Peripatus (Peripa-

topsis) balfouri up to the Formation of the Spermatid. Zool.

Jahrb. 14. 1900. On the Dimegalous Sperm and Chromosomal

Variation of Euschistus with Reference to Chromosomal Continuity.

Arch. Zellf. 5. 1910.
PLATNER, G., Ueber die Bedeutung der Richtungskorperchen. Biol.

Centr. 8. 1889.
RETZIUS, G., Weitere Beitrage zur Kenntnis der Spermien des Menschen

und einiger Saugetiere. Biol. Untersuch. NeueFolge. 10. 1902.
SCHWEIGGER-SEIDEL, F., Ueber die Samenkorperchen und ihre Ent-

wickelung. Arch. mikr. Anat. 1. 1865.
WALDEYER, W., Die Geschlechtszellen. Hertwig's Handbuch, etc., I,

1, 1. 1901-1903. (1906.)

WEISMANN, A., The Evolution Theory. New York. 1904.
WILSON, E. B., The Cell, etc. New York. 1900.
ZELLER, E., Ueber den Geschlechtsapparat des Diplozoon paradoxum.

Zeit. wiss. Zool. 46. 1888.
ZIEGLER, H. E., Lehrbuch der vergl. Entw. der niederen Wirbelt,

Jena. 1902.

CHAPTER IV
MATURATION

IN this chapter we shall describe certain events which are in
reality essential steps in the processes of oogenesis and sper-
matogenesis, namely, the maturing of the nuclei of the definitive
germ cells. In animals these maturation processes are the
final steps in the complete specialization of the germ cells, and
must be accomplished before the two gametes can fuse com-
pletely and thus begin the life of the "new" organism as an
individual. As a matter of fact, the processes of maturation
may be inaugurated before the growth and differentiation of
the germ cells are entirely completed, and these processes may
then all go on together. They are considered here separately,
and without complete regard for their normal time relations,
partly as a matter of convenience, looking toward simplicity
of description, and partly to emphasize their great importance
as a period in the development of the organism. Such a sepa-
ration is easy because the maturation processes are not visibly
connected with the genesis of the germ cells as such, for, mor-
phologically at any rate, they concern only or chiefly the
nuclei alone; the accompanying cytoplasmic modifications of
structure have already been described.

That morphological characteristic chiefly distinguishing the
fully matured germ cells is the possession of but one-half the
number of chromosomes, and of but a smaller fraction of the
amount of chromatic material, possessed by the somatic cells
(Van Beneden) . We should include under the term maturation,
the whole series of events leading to this reduction in number
of chromosomes and amount of chromatin. It should be noted,
however, that the terms " oogenesis" and " spermatogenesis "
have sometimes been used in a restricted sense to mean what
we here term " maturation," but we have understood the former

131

132 GENERAL EMBRYOLOGY

terms to include the whole history of the germ cells up to
the time of their formation as completely specialized struc-
tures, and maturation therefore becomes a phase in 06- and
spermatogenesis.

It is a familiar fact that in fertilization the union of an egg
nucleus and a sperm nucleus is an essential step. The repeated
union of nuclei of the usual type, in this way, would result in
rapid and limitless increase in chromatic elements and material,
but for the operation of some mechanism preventing such an
accumulation, and yet permitting the fusion of germ-cell nuclei.
Maturation is such a process; but it is much more than this.
Consideration of all the phenomena of maturation raises many
questions, important, even fundamental, in their biological
significance. The full meaning of the phenomena can be
appreciated only in connection with the fertilization processes
to which they are introductory ; we may most profitably, there-
fore, postpone much of our discussion of the general significance
of maturation until we have become acquainted with the
process of fertilization.

At this time, then, we shall describe the essential facts of
maturation as we find them in typical, in some respects perhaps
schematized, form, together with a brief comparative account
of some maturation processes in a few special instances. Then
after a similar account of fertilization in the next chapter, we
shall be in position to consider some of the general aspects of
both these processes taken together. The events of maturation
and fertilization are really closely related in time, as well as in
significance. While the spermatozoa are always fully mature
before they enter the egg cells, the entrance of the sperm may
occur either before, during, or after the maturation of the ovum,
although of course the essential step in fertilization, namely, the
union of the nuclei, does not occur (excepting in some Protozoa
and plants) until after the maturation of the egg nucleus is
completed,

The maturation of the germ cells is accomplished, in the Meta-
zoa, by a modification of a mechanism common to all cells and
already familiar, namely, mitosis. But the cell divisions which

MATURATION 133

occur here are of a very special form, not found in the history
of other kinds of cells. These unusual mitoses are known as
the meiotic (maiotiCj Farmer and Moore) , or reducing divisions,
and their chief peculiarity consists in the fact that they result
in the formation of daughter nuclei containing the reduced
or haploid number of chromosomes.

We shalTreview first the maturation of the spermatozoon, as
this is less modified than the ovum, from those conditions which
we regard as typical. Throughout the multiplication divisions
of the spermatogonia, the mitoses are all of the usual character,
except that the mitotic figures are relatively larger than in the
somatic divisions. The number of the chromosomes is the
same as in the somatic cells of the same organism; this is spoken
of as the diploid number. In many, perhaps most, organisms
the chromosomes differ from those of the somatic cells in form
and size characters, so that the germinal tissue can usually be
identified; the germinal tissue nuclei are in general larger and
richer in chromatin than those of somatic tissues, a difference
which, as previously noted, in a few forms (Ascaris, Cyclops,
some Teleosts) can be traced from very early cleavage stages.
But the constitution of the nuclei which pass into the inter-
kinesis after the last spermatogonial division, i.e., into the
primary spermatocyte nucleus, is essentially normal. Some-
times certain peculiarities become noticeable during the growth
period of these cells; the nucleus frequently does not remain in
the typical "resting" condition, but forms a more or less dis-
tinct spireme (leptonema, Winiwarter), and sometimes, even at
the beginning of this stage, there may be a fission of the chro-
matin granules, forming a sort of double spireme (Fig. 73).
At the close of this growth period, when the primary spermato-
cyte prepares to divide, the nucleus begins to show a very
unusual condition. The nucleus itself remains large, but the
chromatin, as it begins to form a spireme, in those cases where
this lias not formed previously, condenses at one side of the
nucleus, in the vicinity of the nucleolus, usually in the region
near the centrosomes and perhaps through their influence
(Schonfeld), into a dense mass in which little structure can be

134

GENERAL EMBRYOLOGY

G

FIG. 72. — Early stages in the maturation of the Dipnoan, Lepidosiren paradoxa.
After Agar. X 933. A. Polar view of equatorial plate of late spermatogonium,
showing size and form differentiation and pairing of chromosomes. B. Spireme
stage (leptonema) of primary spermatocyte. Only a few of the threads are
shown. C. Nearly polar view, showing beginning of longitudinal fusion of
chromatin threads; beginning of synapsis (zygonema). D. Nearly polar view
of "bouquet stage" (pachynema). The threads are fused and condensed.
E. Polar view showing beginning of contraction (synizesis) and splitting of the
chromosomes (diplonema). Most of the thickened threads have split apart
except terminally, where they remain fused, forming rings. In some, the

MATURATION 135

made out (Figs. 72, 73). This stage is called the contraction
phase, or synizesis (McClung) (pachynema, Winiwarter). In
some cases synizesis may occur near the beginning of the growth
period, throughout , which the nucleus then remains in this"
condition.

When growth of the spermatocyte is completed this knot of
chromatin begins to disentangle and the spireme again becomes
visible. This later spireme is not continuous, however, but is
of the segmented type (Fig. 72), and the number of segments,
i.e., of chromosomes, is but one-half the number of chromosomes
that went into tJw nucleus at the close of the preceding division. As
regards the chromatic structures, this is the essential point in
the whole maturation process; the number of chromosomes
formed in the prophase 6f the first spermatocyte division is
reduced to one-half the somatic number. This numerical
reduction of the chromosomes is brought about in most, if not
in all cases so far known, by an actual fusion, by twos, of the
chromosomes contained in the last spermatogonial nucleus.
This fusion of pairs of chromosomes is termed synapsis (Moore,
McClung) or synopsis (Hacker) (zygonema, Winiwarter), and
the resulting units are thus double or bivalent chromosomes. It
seems entirely likely, if not definitely established, that the pairs
of chromosomes which come together in synaptic fusion are
each composed of one chromosome derived from the male
parent, and the corresponding chromosome derived from the
female parent, similar in size, and form, and also in function if
we assume the fact of chromosomal specificity (Montgomery).
These two groups of similar elements came into a single nucleus
during the fertilization process which was the beginning of the
new individual, whose cells are now preparing for reproduction;
they have remained separate throughout the life of this organ-
ism until this event, through all the divisions of the ancestral
germ-forming cells (Fig. 80). In a certain sense, therefore,_this_-
process of synapsis represents the real climax of the whole

splitting is less complete. One ring is cut through showing two free ends. F.
Advanced synizesis. G. Chromosomes appearing after synizesis, shortened and
thickened. The ex-conjugant chromosomes (univalent) have separated and show
transverse constrictions, preparatory to the second maturation division.

136 GENERAL EMBRYOLOGY

series of developmental processes, and it is at the same time the
starting point of the life cycle of a new organism of another
generation.

~~In a few forms (e.g., some Insects, Lepidosiren, Fig. 72), the
double nature of these bivalent chromosomes is distinctly visible
and is indicated by a split through the long axis of the chromo-
some, showing that the pair of univalent elements have fused
side by side, a condition known as parasynapsis. In most
cases observed (other Insects, Amphibia) the fusion is end to
end, a condition known as telos^na^sis (Wilson's terms). In
many instances, however, the fusion seems to have occurred
between the granules composing the chromosomes, so that in
the bivalent body there is no visible indication of the duplex
nature at this time; this is then only to be inferred from the
fact of numerical reduction. It is very important to notice
that the time relations between synizesis and synapsis may
sometimes be just the reverse of that described above, and the
synapsis stage may occur first, so that the numerical reduction
of the chromosomes occurs at the close of the last spermatogo-
nial division (some plants, Strasburger, Overton).

following the period of synapsis the nucleus and cell may
proceed at once to divide, or there may ensue another resting
period, during which the chromosomes again become indistinct.
In either case, when the new mitotic figure forms, always after
an unusually long prophase which is characteristic of this divi-
sion, the reduced (haploid) and bivalent chromosomes often
show an unusual condition, in that they may prepare at once,
not for a single ensuing division, but for two divisions which are
to follow immediately, without an intervening resting period.

From this stage onward in the history of the sperm and egg
nuclei, two general types of chromosome behavior are some-
times distinguished, although they are connected by transi-
tional conditions and so are regarded as modifications of a
single process. As one extreme condition we find a form of
chromosome behavior called tetrad formation, which we may
describe, not because it is a typical method of chromosome
reduction, but because the facts of reduction come out more

MATURATION

137

clearly in this form of maturation division (Boveri). In cases
oPteTfad formation, when^the chromosomes appear in the
primary spermatocyte, after the resting stage, each of the newly
organized bivalent elements comes out in the form of four
small bodies, the tetrads, arranged approximately in a square
(Fig. 73, E). These bodies result from two successive splittings
of each chromosome into four columns of granules, each of

FIG. 73. — Tetrad formation in the spermatogenesis of Ascaris megalocephala
bivalens. After Brauer. X 795. A-G. Stages in the division of the primary
spermatocyte. A, B, splitting, and C, condensation of chromatin thread, seen
in side view. D shows, in end view, that the splitting is double. Centrosome
divided. E. Migration of centrosomes and formation of spindle. F, G. Separa-
tion of the two groups of dyads and division of the cell body. H. Secondary
spermatocyte containing two dyads. /. Division of secondary spermatocyte.
/. Two of the spermatids, each with two "monads" or single, univalent, chromo-

which is then condensed into a single element (Fig. 73, A, B, C,
D). We may recognize here a precocious division of the chro-
mosomes, which in these cases precedes considerably the
division of the nucleus and cell as a whole. Not only this, but
there are two chromosomal divisions, corresponding with two

138 GENERAL EMBRYOLOGY

cell divisions, and these occur simultaneously, while the
nuclear and cytoplasmic divisions occur consecutively at a
later period, during which these chromosomes do not divide

again. The number of tetrad groups is thus the same as the

AA
haploid number of chromosomes ( ^ K and the total number of

elements composing the tetrads is four times the haploid or
two times the diploid number (2s).

The nucleus and cell now enter upon a mitosis in which each
tetrad behaves as a typical chromosome. The division and
migration of the centrosomes to opposite sides of the nucleus,
the formation of the spindle and asters, and other details of
this mitosis, have no unusual features and need not detain us.
The tetrads, containing all told 2s elements, become arranged
about the equator of the spindle and each separates into two
pairs of elements called the dyads (Fig. 73, F, G). The groups
of dyads then move to opposite poles of the spindle and the cell
divides into the two secondary spermatocytes. Since the rest-
ing stage is now omitted, the dyads do not dissolve after this
division, nor do they divide again in anticipation of the next
mitosis — the division of the chromatic elements for this cell
division has already occurred in the nucleus of the primary
spermatocyte, as we have seen. The dyads, containing all told
s elements, then move at once to the equator of the new spindle,
and each separates into two monads (Fig. 73, 7, «/). The two

groups of monads, each now containing - elements, diverge to

2i

opposite poles of the spindle, and the division of the cell
(secondary spermatocyte) results in the formation of two sper-

o

matids, each with - chromosomes (Fig. 74). Each nucleus

then reforms into a typical resting condition, and passes through
the metamorphosis into the head of the spermatozoon, as
described in the preceding chapter. The essential characteris-
tic, therefore, of the nucleus of the spermatid and spermatozoon
is that, while each of the bivalent chromosomes of the spermatocyte
is represented, yet, as the result of the process of reduction

MATURATION

139

through synapsis, the chromosomes are now only half as numer-
ous as in all the other cells of the organism. But while each
bivalent chromosome is thus represented, it may be a question

FIG. 74. — Diagram of reduction with tetrad formation in sperm atogenesis.
From Wilson, "Cell." The somatic number of chromosomes is supposed to be
four. A,B. Division of one of the spermatogonia, showing the full number
(four) of chromosomes. C. Primary spermatocyte preparing for division; the
chromatin forms two tetrads. D,E,F. First division to form two secondary
spermatocytes each of which receives two dyads. G,H> Division of the two
secondary spermatocytes to form four spermatids. Each of the latter receives
two single chromosomes and a centrosome which persists in the middle-piece of
the spermatozoon.

whether each univalent chromosome is also somehow repre-
sented. To this we shall return later.

But the formation of tetrads is by no means of universal,
even of common, occurrence in reducing divisions. Tetrads are
commonly found only among the Nematodes, Annelids, and

140 GENERAL' EM BRYOLOGY

Arthropods. In the great majority of animals the reduction
divisions proceed without the formation of actual tetrads in
their typical form, and when the bivalent chromosomes appear
in the nuclei of the primary spermatocytes, as the result of
synapsis, they have no unusual form. As they pass into the
equatorial plate, however, it is seen that the two longitudinal
halves, composing each bivalent chromosome, are united at
their ends (Figs. 72, E;7S). When the anaphase begins each
chromosome-half is drawn out first from its middle, so that the

FIG. 75. — Maturation divisions in certain Insects, showing forms of chromo-
somes and their relation to tetrads. After de Sinety. X 1125. A,B. Two
stages in anaphase of primary spermatocyte division in Stenobothrus parallelus.
Rings opening into Vs, which diverge. C. Anaphase of spermatogonial division
in Orphania denticauda, showing differentiated chromosome, x. D, E. Prepara-
tion for first spermatocyte division in Orphania, showing "tetrads" in various
stages of formation from rings and crosses.

whole original chromosome may appear as a ring or cross, or
some related figure. As the halves separate each may assume
a p- or >-form, the limbs of which may come to lie parallel as
the chromosome approaches the pole of the spindle, and thus
may appear to be double (Figs. 75, 20). On account of this
peculiar form assumed by the chromosomes in this division,
it is known as the heterotype division. And it is to be -noted
that the separation of the halves of the bivalent chromosome

MATURATION 141

here, is along the line of a split which is usually visible in the
chromosomes when they appear out of the resting nucleus.
As the result of this heterotype division each secondary
spermatocyte receives the haploid number of chromosomes.
The second maturation division commonly shows none of these
rings or crosses, or other figures, and is known consequently
as the homotype division (Flemming's terms). The homotype
division of the secondary spermatocyte follows the heterotype
either immediately, or after a considerable pause, during which
the chromosomes sometimes lose their definite outlines to some
extent. This pause, which does not occur when tetrads are
formed, is probably related to the fact that while in the tetrads
both divisions of the chromosomes occur at the commencement
of the process, in this form of reducing division the second split-
ting does not occur until after its first actual division. During
the homotype division the chromosomes behave, then, essen-
tially as in the divisions of the usual type, and the resulting
spermatids receive, just as in tetrad formation, the haploid
number of chromosomes, just as do the secondary spermato-
cytes. Numerically, the most important difference in reduc-
tion with and without tetrad formation is that in tetrad forma-
tion the secondary spermatocytes have the diploid number,
and in the absence of tetrads, the haploid number of chromo-
jjomes*. This is the result of the fact that when tetrads are
formed, the division of the chromosomes actually belonging to
the secondary spermatocytes (second maturation division)
really occurs in the nucleus of the primary spermatocyte (first
maturation division), while in the absence of tetrads the divi-
sion of the chromosomes has the normal relation to cell division,
and the haploid number persists from the primary spermatocyte,
after synapsis, to the spermatid and spermatozoon.

Before mentioning any further details of chromosome behav-
ior during maturation, we must compare the process of matura-
tion as it occurs in the ovum with that in the sperm. We may
say at the outset that in all essentials the two histories are
identical, so that this comparison may be brief, but there are a
few differences to be noted.

142 GENERAL EMBRYOLOGY

The divisions of the oogonia are normal; the diploid number
of chromosomes appear, and the details of spindle, aster, and
centrosome, call for no special mention. Aside from the
chromosomal behavior, the divergences of the later maturation
divisions from the normal are partly the result of the enormous
growth of the egg cell, and partly in the nature of adaptation
toward ensuring the practically undiminished size of the ovum
at the end of the process; that is, an equal subdivision of the
chromatic elements is accompanied by an unequal subdivision
of the cytoplasm and deutoplasm. The general formation of
the large primary oocyte has been described. We should
emphasize again the fact that the maturation of this cell
frequently is not completed until after the sperm cell has
actually entered its substance. If we were describing the events
of the maturation of the egg in strict accordance with their
usual, though not invariable, time relations we should next
describe the ensemination of the egg — the first step in fertiliza-
tion. For the sake of clearness, however, we shall describe
maturation as if it occurred before the entrance of the sperm;
as a matter of fact, there are a few forms in which this is really
the normal course of events, as in the sea-urchin and most
Echinoderms.

The nucleus of the oogonium is very large and lies toward one
side of the cell — practically always toward the animal pole of the
egg (Fig. 76, A). The first steps in the maturation of the ovum
closely resemble those in the sperm. During the brief synizesis
stage the chromatin condenses near the centrosome, closely
around the large "nucleolus" which is commonly found in most
oocytes; the emerging spireme shows that synapsis has occurred
for the spireme is segmented into the haploid number of ele-
ments, representing the bivalent chromosomes. That is, the
actual reduction occurs in the primary oocyte as in the primary
spermatocyte. The oocyte nucleus then passes through a
condition not represented in the spermatocyte in that a large
amount of chromatin leaves the chromosomes (spireme seg-
ments), either dissolving in the nuclear sap, or passing in the
form of granules or small masses to some region of the nucleus

MATURATION

143

quite apart from the chromosomes proper (Fig. 76). It is
important to note that no chromosomes are lost in this way;
the full haploid number of these bodies remains grouped at one,
usually the distal, side of the nucleus. During the active
preparation for the first maturation or heterotype division
the nuclear membrane disappears, liberating the dissolved

FIG. 76. — Maturation in the egg of the Nemertean, Cerebratulus. After Coe.
C, Z>, X 375, others X 250. A. Primary oocyte. Part of the chromatin has
been condensed into chromosomes, only five of which are shown (the number
present is sixteen). The remainder of the chromatin is thrown out into the cyto-
plasm. The centrosomes, each with a small aster, are diverging, and the nuclear
membrane is commencing to disappear. B. First polar spindle fully formed and
rotated into radial position. Chromosomes in equatorial plate. C. First
oocyte division; anaphase. D. First polar body nearly separated. E. First
polar body completely cut off; second polar spindle formed and rotating into
radial position. Spermatozoon within the egg. F. Second polar body com-
pletely separated. Egg pronucleus forming, surrounded by large aster. Sperm
pronucleus, also with a large aster, enlarged and approaching the egg pro-
nucleus, c, chromosomes; o, nucleolus, vacuolated and commencing to disap-
pear; s, spermatozoon just within the egg;z>, germinal vesicle; vc, contents (extra
chromosomal) of germinal vesicle. /, II, first and second polar bodies; cT, sperm
pronucleus; 9 , egg pronucleus.

chromatin, or the extra-chromosomal masses, which then
disappear gradually into the cytoplasm while the small chromo-
somes pass into the division figure. Sometimes this loss of
chromatin is effected by a shrinkage of the chromosomes them-

144 GENERAL EMBRYOLOGY

selves, as in some Elasmobranchs, where the chromosomes at this
time shrink to about one-fortieth of their previous length and
one-tenth their previous diameter (Ruckert, Fig. 34). Or the
"nucleolus" of the oocyte may be a karyosome or chromatin
nucleolus, and in such cases (Echinoderms, for example) during
synizesis the chromatin may be nearly all contained within this
body. Then the chromosomes are formed singly or in groups
out of this " chromatin reservoir." After they have all been
given off, much the greater part of the chromatin still remains
in the karyosome, which then may fragment before dissolu-
tion, or it may be dissolved directly (Fig. 35). The subsequent
behavior of the chromosomes is closely similar to that of the
epermatocyte chromosomes; tetrads may or may not be
formed, according to the species, as the chromosomes pass into
the division figure (Figs. 77, 78). The centrosome divides and
the spindle and asters form typically in most respects save in
size and position. The spindle is very small and in most eggs
is close to the surface of the cell at its animal pole (Fig. 76). In
alecithal and isolecithal eggs the nucleus and spindle are at
first located centrally and then later move to the periphery.
At first the spindle lies parallel to a tangential plane, but during
the mesophase it rotates through ninety degrees, putting its
axis in a radial direction (Figs. 76, 77). In many cases the
division of the oocyte is inhibited at this stage, until after the
entrance of the spermatozoon, when it proceeds to completion;
or this heterotype division may proceed without any inter-
ruption and the primary oocyte cut at once into two cells* The
extremely eccentric position of the nucleus in this stage leads to
one of the most characteristic features of odgenesis, namely, the
very unequal division of the cell body. One of the products of
division, the secondary oocyte, is of practically the same size as
the primary oocyte; the other cell — the first polar body — is
very much smaller, indeed usually minute (Figs. 76, 77, 78).
In essentials these two daughters of the primary oocyte are
equivalent; their nuclei are alike in size and composition, each
contains a daughter centrosome, but with the polar body there
is only the smallest amount of cytoplasm and practically none

MATURATION

145

FIG. 77. — Maturation in the egg of Ascaris megalocephala bivalens. From
Wilson, "Cell," after Boveri. A. Egg with spermatozoon just entering at d\
The germinal vesicle contains two rod-shaped tetrads (only one clearly shown).
B, C. Tetrads seen in profile and end views. D. First polar spindle forming (in
this case within the germinal vesicle). E. First polar spindle in definitive posi-
tion. F. Tetrads dividing. G. First polar body formed, containing, like the
egg, two dyads. H, I. Dyads rotating into position for the second division.
J. Dyads dividing. K, Each dyad has divided into two single chromosomes,
as the second polar division approaches. (For final stages, see Fig. 94.)

146

GENERAL EMBRYOLOGY

H

FIG. 78. — Diagram of reduction, with tetrad formation, in oogenesis. From
Wilson, "Cell." The somatic number of chromosomes is supposed to be four.
A. Initial phase; two tetrads have been formed in the germinal vesicle. B. The
two tetrads have been drawn up about the spindle to form the equatorial plate
of the first polar mitotic figure. C. The mitotic figure has rotated into position,
leaving the remains of the germinal vesicle at g.v. D. Formation of the first
polar body; each tetrad divides into two dyads. E. First polar body formed;
two dyads in it and in the egg. F. Preparation for the second division. G.
Second polar body forming and the first dividing; each dyad divides into two
single chromosomes. H. Final result; three polar bodies and the egg-nucleus
(9), each containing two single chromosomes (half the somatic number); c, the
egg-centrosome which now degenerates and is lost.

MATURATION 147

of the deutoplasmic substance. The equal division of the
nucleus is thus accomplished without appreciable loss from the
oocyte of any of the cytoplasm and food reserve.

The second maturation or homotype division follows, either
at once, or after a long pause in cases where the sperm normally
enters at this time. The figure for the second maturation
division is either a new figure or the reorganized remains of the
preceding. In either case it appears in the region occupied by
its predecessor, often in a radial position from its first appear-
ance (Figs. 76, 77). Again the division is very unequal and the
secondary oocyte gives rise to the large mature ovum and a
small second polar body, again alike as regards nuclear com-
position. The first polar body may or may not divide into
two at the same time; we may assume that such a division is
normal, but on account of the degeneration of the polar bodies
such a division tends to disappear; in different forms various
stages of this division are reached. In a few rare instances,
as in some Rotifers and Insects, the second polar body may also
divide.

The chromosomes remaining in the ovum then re-form a
reticular nucleus, smaller than the original oocyte nucleus, and
with the haploid number of chromosomes; after forming a thin
membrane the nucleus moves toward the cytoplasmic center
of the ovum and there awaits union with the nucleus of the
fertilizing spermatozoon (Figs. 76, 77). With few if any ex-
ceptions, the centrosome of the ovum is entirely lost as the
nucleus is reconstructed; the absence of the centrosome is one
of the peculiarities of the egg cell. It should be added that
when the maturation is not completed until after the entrance
of the sperm, the egg does not ordinarily re-form a typical
nucleus but proceeds at once to unite with the sperm nucleus.

The final result of the two maturation divisions of the
primary oocyte is the formation of four cells whose nuclei are
similar, and which are morphologically exactly equivalent to
one another (Figs. 74, 78). Physiologically, however, there is
the greatest difference among them, since of the four only one,
the ovum, is able to function, or indeed even to remain alive,

148

GENERAL EMBRYOLOGY

and share in the development of a new organism. The other
three (or two), the polar bodies, after a brief time degenerate
and disappear.

The view that the ovum and the polar bodies are equivalent
cells morphologically, and that the latter are in reality to be
looked upon as degenerate egg cells (Mark) is now familiar.
Their identity in nuclear structure and history is of course a
decisive similarity.

FIG. 79. — Variations in the size of polar bodies. A, B. Sections through
segmenting ovum of the Gasteropod, Limax maximus, showing polar bodies of
very different sizes. After Meisenheimer. C-F. Ascaris megalocephala. C, D,
E, after Sala, show influence of low temperature; F, after Boveri. C. Large
first polar body which has divided. D. Very large first polar body. E. Equal
division of egg during "first polar division." F. Equal division of egg during
second polar division." 7, //. First and second polar bodies; 9, egg pro-
nucleus; cf , sperm pronucleus.

The actual size of the polar bodies varies enormously. Those
of the Echinoderms are among the smallest, some being only
5-8 micro, (1/5000-1/3000 inch) in diameter; in Amphioxus
they are about 7 micra, or about one-fourteenth the diameter
of the egg, while in the mouse they are relatively much larger—
13 micra, or one-fifth the diameter of the egg. An interesting
series of forms can be arranged bridging over the size differences,
and to some extent also the physiological differences between
the ovum and the polar bodies. In a few Annelids, some

MATURATION 149

Turbellaria, and most Molluscs (Fig. 79) the polar bodies are
very large, sometimes even one-fourth the diameter of the
egg itself. In occasional instances, abnormal " giant" polar
bodies are formed, which really approach the ovum in size
(e.g., Amphioxus, and the Turbellarian, Prosthecerceus) . Some
of these large polar bodies form a definite membrane, like the
vitelline membrane of the ovum; and in rare instances they are
actually fertilizable, although their development never proceeds
beyond an incomplete cleavage. The last step in the transition
from polar body to egg cell is represented by an interesting
condition occasionally found in Ascaris and some other forms,
(e.g., mouse) where an abnormally placed polar spindle may
result in the division of the oocyte into two equal cells, one of
which should be called a polar body (Fig. 79, E, F). The last
step in the opposite direction — the dissimilarity between egg
and polar body — reaches a climax in some of the Insects where
polar bodies are not really formed as such; the oocyte nucleus
divides as in polar body formation and daughter nuclei are
formed but these remain in the periphery of the egg cell, for
no cytoplasmic division whatever is accomplished. The
nuclear phase is the only part of the maturation division
remaining. The "polar nuclei" formed in this way degenerate
without sharing in development, just as if they had been cast
out of the cell.

We have already suggested that this dissimilarity in size
between the polar bodies and the ovum is in the nature of an
adaptation such that maturation of the egg nucleus may take
place without reducing the amount of cytoplasm and food
materials which are to form the chief portion of the
material basis of the new organism. In accomplishing this,
three of every four 'potential egg cells are totally deprived of
these substances and lose all possibility of developing. In
some few instances, the polar bodies may for a time remain
alive and during the early cleavage show some signs of activity,
such as the performance of amoeboid movement, "spinning
activity," etc. (Andrews).

The location of the polar bodies within or without the vitel-

150 GENERAL EMBRYOLOGY

line membrane depends upon the time relation between mem-
brane formation and maturation. The polar bodies may
form before the membrane, in which case they usually are lost
from the egg soon after their formation. More frequently
they form after, and therefore within, the membrane and so
can be seen for some time after development begins, when
they form useful points for orientation, for in nearly all cases
they mark the animal pole of the egg. The only exceptions to
this location are among the Insects and Copepods, in which
their position is variable.

Morphologically there is also complete correspondence
between the ovum and the three polar bodies formed from the
primary oocyte, and the four spermatids and spermatozoa
formed from the primary spermatocyte (Platner, O.Hertwig).
But again there is physiological divergence, in that all of the
derivatives of the spermatocyte are capable of functioning as
germ cells, while only one of the oocyte derivatives may do so.
This physiological divergence is an expression, in another form,
of the physiological division of labor between the egg and the
sperm already referred to.

The chief points of similarity and difference between ovum
and spermatozoon are summarized in the accompanying table.

Little is known regarding the nature of the stimuli which
lead to the process of maturation, but it is clear that they are
quite varied in different eggs. In some of those cases where the
eggs are discharged freely into the water, contact with the
water seems to initiate the process. But maturation may be
begun previously to such a discharge. In some cases of this
kind, as well as in others where the eggs are not thus freed, the
rupture of the egg follicle seems to start the maturation process.
In many cases the entrance of the sperm into the ovum is the
effective stimulus to maturation, or to the completion of
maturation in many of those instances where it has been begun
previously and has been inhibited, either just before or just
after the formation of the first polar body.

While we must postpone a part of our brief discussion of the
theoretical significance of maturation, for reasons stated

MATURATION

151

COMPARISON OF TYPICAL OVUM AND SPERMATOZOON
Similarities:

Nuclei contain the haploid number of chromosomes, the result of

two similar meiotic divisions.
Chromosomes alike in form, size, and, with a few exceptions of

special significance, in number.
Can function only after syngamy.

Differences:

SPERMATOZOON

Little cytoplasm.
No deutoplasm.

Actively motile.

Centrosome present.

One of four similar products
of the division of the sper-
matocyte, all of which are
functional.

Usually completely formed and
matured in the gonad.

OVUM

Much cytoplasm.

Nearly always contains deuto-
plasm, often very large amounts.

Non-motile.

Centrosome absent.

The functional one of four dis-
similar products of the division
of the oocyte, the other three
of which are alike and not func-
tional.

Usually formed but rarely com-
pletely matured in the gonad.

above, we should indicate at this time that the process has
significance from at least two points of view. As a preliminary
we should note that two different processes are involved in
maturation; first, a reduction in the number of chromosomes,
second, a reduction in the amount of chromatin. The earlier
idea that maturation is merely a process by which the germ
cells are rid of a part of their chromatin, and one-half of their
chromosomes, as a preparation for the restoring union of
chromatin and chromosomes during fertilization, is only one
and probably the least important aspect. Many cells other
than germ cells gain and lose large amounts of chromatin, and
without going through any such complex process as that out-
lined above. Frequently much more than half, sometimes
fully nine-tenths, of the chromatin is lost from the oocyte
nucleus during maturation, while during spermatogenesis com-
paratively little may be lost. And the mere numerical reduc-

152 GENERAL EMBRYOLOGY

tion of chromosomes is fully accomplished in s>napsis, before
the actual maturation divisions occur. For these and many
other reasons it seems that the chief importance of maturation
is from the standpoint of inheritance. This is true particularly
of most of the details regarding chromosome reduction, which
become significant only when correlated with the facts, first,
that the germ cells are the simplest phases in the life cycle of
the organism, alternating with the mature phases, the complex
characteristics of which are related to the simpler characters
of the germ, and second, that in some way, as yet unknown,
the structural and physiological characteristics of the new
organism are, at least in part, primarily determined by the
chromosomal structure of both of the germ nuclei, i.e., to the
fact of biparental inheritance. From the standpoint of in-
heritance then the details of the behavior of the chromosomes
during the maturation divisions take on the greatest importance.
One of the more important matters is the precise plane of
division of the chromosomes. It seems necessary to assume
that each chromosome is not entirely homogeneous, but that
its qualities differ in different parts. Consequently, in chromo-
somal composition the four nuclei derived from each primary
oocyte or spermatocyte nucleus may be all alike or may be of
different kinds, according to whether the original chromosomes
separated into similar or dissimilar parts in one or both of the
maturation divisions (Fig. 80) . In those cases where the chromo-
somes separate into qualitatively similar halves the division
is said to be equational, and when into qualitatively unlike
halves the division is reducing. And it is commonly believed
that one of the maturation divisions is equational and one re-
ducing. When the equational division precedes, post-
reduction is said to occur; when the reducing division precedes
it is described as prereduction (Korschelt and Heider). It is
by no means a simple matter to determine whether a given
chromosome division is equational or reducing, since there is
externally visible no indication, in a chromosome itself, of its
qualitative differentiation; and further because the processes
of rearrangement and redistribution of the chromatin granules

MATURATION 153

making up the chromosome are usually very obscure, particu-
larly when synizesis is pronounced.

This whole subject is in a rather more hypothetical state
than one might wish, considering the importance of the con
elusions to be drawn. Upon the assumption that the qualities
of a chromosome differ from end to end, a longitudinal fission
of the chromosome would divide it into exactly similar halves,
while a transverse fission would of course divide it dissimilarly.
In many forms it is clear that the first or heterotype division
is longitudinal and the second or homotype division is trans-
verse so that the four resulting nuclei are of two categories. In
other cases it appears superficially that both divisions are
longitudinal, while in still others it is really impossible to say
definitely whether a given division is longitudinal or transverse.
A transverse division would be reducing, however, only upon
the assumption of an end to end differentiation of the chromo-
some, and upon the further assumption, which should be clearly
apprehended, that no rearrangement of the chromatin granules
occurs during the maturation divisions. And the additional
fact must be taken into consideration that the chromosomes
of the primary spermatocyte or oocyte are bivalent, that they
represent two chromosomes, which as wholes have fused in
either parasynapsis or telosynapsis, and the actual chromatin
granules composing them may have an arrangement in fusion
which is not indicated by the behavior of the whole chromosome.
The question whether a given longitudinal or transverse
division of a chromosome is equational or reducing then can be
determined only by taking all of these preliminary arrange-
ments into account, and in most cases this is extremely difficult
or even at present impossible. The small size of the chromo-
somes themselves and the minuteness, often invisibility, of the
chromatin granules, often put the facts of their arrangement
and behavior beyond the possibility of observation, and we
can only infer their arrangement and history from subsequent
events. In most cases we know only that reduction occurs.
And from the few cases in which the course of events seems
clear, we infer that in all maturation divisions, one is equational,

154

GENERAL EMBRYOLOGY

A

FIG. 80. — Diagrams representing the behavior of the chromosomes during
fertilization and maturation. The differentiation of the three kinds of chromo-
somes is indicated by the number of small circles in each. d\ chromosomes
derived from the spermatozoon (sperm pronucleus) (black circles) ; ? , those from
the egg (egg pronucleus) (white circles). The somatic number of chromosomes
is six. A. Entrance of spermatozoon. B. Fusion of egg and sperm pronuclei,
forming the first cleavage nucleus. C. Splitting of chromosomes in equatorial
plate, during the division of any somatic, oogonial, or spermatogonial cell. D.
Primary oocyte or spermatocyte in synapsis (telosynapsis). Fusion of similar
chromosomes of maternal and paternal origin. E. Longitudinal splitting of
bivalent chromosomes during first maturation division. F. First division
completed forming the two secondary oocytes or spermatocytes. The nuclei
are alike in composition. G. Transverse division of chromosomes during second
maturation division (reducing division — postreduction). H. The resulting
four cells. With respect to each chromosome the cells are of two kinds, numer-
ically equal.

MATURATION 155

one reducing, resulting in the formation of germ cells of two
more or less unlike kinds, in equal numbers (Fig. 80).

Aside from its relation to the phenomena of heredity, the
meaning of the maturation process is very problematic. In
the Protozoa where definite chromosomes are formed only
infrequently, maturation frequently involves a separation
between reproductive and vegetative chromatin, as already
suggested. Among the Metazoa there is usually a loss of
chromatin from the nucleus, but it is very doubtful whether
it has a similar meaning. In a great many cells, particularly
those which are very active, e.g., gland cells, oocytes, etc.,
the cytoplasm is constantly receiving substance from the
nucleus. This material is frequently chromatic, and the
granules of this kind have received a variety of special names,
but collectively may be included under the term chromidia.
(See Chapter II.) It may be that through some such process
as this the nucleus exercises those forms of control and regula-
tion of cell life that are its chief function. The loss and
degeneration of the chromatin distributed to the polar bodies
can have no significance here, for that process is involved in the
degeneration of the entire polar bodies, which has an entirely
different meaning. But during the growth period of most ova,
just after synizesis, a relatively large amount of the chromatin
is thrown out into the cytoplasm, and during the later stages
of spermatogenesis a somewhat similar loss may be observed.
And in the very early history of the germ cells of the organism,
when this may consist of only a few cells, the primordial germ
cells may often be distinguished by just this process of chro-
matin discharge from the nucleus. Such cells are often
characterized by unusually large nuclei, and a large fraction of
their chromatin content may be liberated into the cytoplasm
at each mitosis. It may very well be, therefore, that this is a
regular and highly significant process in the formation and
maturation of the germ cells, having to do with the unusual
activity of the sperm or with the development of various formed
substances, both protoplasmic and deutoplasmic, present in
the cytoplasm of the ovum. Indeed it may not be too much

156

GENERAL EMBRYOLOGY

to suppose that the all-important " organization" of the ovum
may in some way be related to this process of chromatin
distribution.

The fact of maturation has been determined for all groups of many-
celled animals and plants, and among the unicellular forms it is by no
means uncommon. Among the Protozoa the phenomena of maturation
are of considerable theoretical interest. In those forms in which the
chromatin is not formed into definite chromosomes, but remains unor-

Fio. 81. — Maturation phenomena accompanying conjugation in the Rhizopod,
Actinophrys sol. From Calkins, "Protozoa," after Schaudinn. A. Parts of
two individuals, fused; the axial filaments terminate in granules on the surface
of the nucleus. B, Nuclei in prophase. C. Formation of first polar spindle.
D. Reconstruction of nuclei. E. Fusion of nuclei. F. First division spindle.
p, polar body.

ganized, grossly, there seems to be a kind of division of labor between
vegetative and kinetic (reproductive) forms of chromatin. The repro-
ductive nuclei (idiochromidia) are frequently distinctly separate from
the somatic nuclei (chromidia), and just before fertilization the former
may divide twice in rapid succession. This process bears the greater
resemblance to the maturation of the ovum since after each division one
of the two products degenerates, often without actually being thrown
out of the cell, leaving functional only one of the four products of the
original idiochromidium. Actinosphcerium is a typical example (R.

MATURATION

157

Hertwig), and there are several other forms where much the same thing
occurs, e.g., Actinophrys (Fig. 81), Entamceba. In all cases such as
these, it is impossible to say whether reduction, in a strict sense, is
accomplished, or whether this is merely an elimination of chromatin, for
the chromatin is not organized into chromosomes whose precise behavior
may be traced ; it is quite likely that there is here no true reduction in
the Metazoan sense. In some of the Infusoria, however, definite chro-
mosomes are formed in the nucleus during these divisions and a definite
chromosomal history may be made out. In Paramcecium, for instance,
as described by Calkins and Cull, where the idiochromidia are known as
the micronuclei, these alone are concerned in the " maturation " divisions.
The micronucleus forms a fairly typical division figure consisting

FIG. 82. — Maturation divisions in Paramcecium aurelia (caudatum). From
Calkins and Cull. Only a few of the chromosomes are represented in each case.
A. Late anaphase of first maturation division of micronucleus; some chromo-
somes incompletely divided. X 1000. B. Early anaphase of second maturation
division. X 633. C. Telophase of second maturation division. X 900.

of a spindle and more than 200 separate chromosomes. During the
first maturation division each of these divides longitudinally, the
resultants passing in each case to the separate daughter nuclei, without
any corresponding division of the cell body (Fig. 82) . A second matura-
tion division follows immediately and is precisely like the first giving
four daughter nuclei (micronuclei, idiochromidia), three of which then
degenerate as in polar body formation. The one remaining nucleus
divides again, this time the chromosomes dividing transversely (reducing
division?). This third division is not strictly comparable with anything
to be found in the Metazoa and is apparently correlated with the charac-
ter of the fertilization process in this form, for both parts share in repro-
duction. One-half remains in situ as the equivalent (analog) of the egg
nucleus, and the other half migrates, as the equivalent (analog) of the
sperm nucleus, to the body of another organism, fusing with (fertilizing)
the stationary nucleus of that individual. In the majority of the
Protozoa the so-called "maturation" or "reduction" divisions are not

158 GENERAL EMBRYOLOGY

equivalent to these processes in the Metazoa, but are merely divisions
by which a separation is effected between the reproductive and nutri-
tive chromatin, i.e., idiochromatin and trophochromatin; in nearly all
known forms only the former takes any active part in the subsequent
reproductive processes, while the trophochromatin usually dissolves and

disappears.

Two very special modifications of the maturation process deserve
just a word. The first is in connection with those few eggs which
normally develop without fertilization (parthenogenesis), i.e., without
the union of equivalent egg and sperm nuclei. In such cases, which are

FIG. 83. — Maturation in the parthenogenetic egg of the brine-shrimp, Artemia.
After Brauer. A, X 795, others, X 368. A. Second polar body incompletely
cut off. B. Second polar nucleus reSntering the egg and approaching the egg
pronucleus. C, D. Fusion of second polar body nucleus with egg pronucleus.
E. First cleavage spindle with two groups of chromosomes derived from the two
nuclei. II. Second polar body or nucleus; ?, egg pronucleus.

known in the Aphids, many Crustacea, and Rotifers, for example, the
normal course of maturation would lead to the formation of an organism

with the haploid number of chromosomes l^\ in all of its cells. In

most, if not in all, such cases which have been studied, it is now known
that as a matter of fact the egg is not left with the reduced number of
chromosomes. Thus in the brine-shrimp, Artemia (Brauer), which
illustrates the usual course of events in parthenogenesis, the first matu-
ration division proceeds as usual and is equational (reducing), leaving

2 bivalent chromosomes in the secondary oocyte nucleus. Then one
of two courses may be followed (Fig. 83). In most normally partheno-

MATURATION

159

genetic eggs a second polar body is formed typically, leaving the reduced
number of univalent chromosomes in the egg nucleus, but then the
second polar body immediately reenters the egg, apparently taking the
place of an equivalent sperm nucleus and restoring the chromosomal
characters to the normal somatic condition, after which development
proceeds. The polar body need not be actually extruded from the egg
cell in order to give the same history, as long as the nuclear events are
equivalent (Fig. 84). In some eggs, even of species in which the history
is at times similar to that just described, a different method gives the
same result. Thus, while the second polar spindle may form typically,
the chromosomes upon it do not divide, and the equivalent of the second

D ^(J;-£Y:W

FIG. 84. — Maturation in the parthenogenetic egg of the Echinoderm, Astro-
pecten. Afyer O. Hertwig. A. First polar body formed but not extruded;
second polar division in early anaphase. B. First polar body extruded; second
polar division completed, the polar nucleus near the periphery. C, D, E. Stages
in the gradual approach and fusion of the second polar nucleus and egg pronu-
cleus, to form the cleavage nucleus. /. First polar body; II, second polar nu-
cleus; 9 , egg pronucleus.

polar nucleus is never formed. The egg nucleus then re-forms with its

o

2 chromosomes, but these are bivalent as shown by the character of

the first maturation division, so that in effect the egg nucleus contains
the somatic number of chromosomes, which actually appears in subse-
quent divisions. It is therefore clear that while such parthenogenetic
eggs fail to receive a sperm nucleus they retain or receive back the
equivalent of such a nucleus in the form of the second polar body
nucleus, which is not lost as it is in eggs requiring to be fertilized;

100 GENERAL EMBRYOLOGY

The second unusual modification of maturation is to be seen in the
spermatogenesis of many Arthropods, chiefly Insects. These species
have already been mentioned as showing a numerical difference between
the chromosome groups of the male and female individuals, the female
having, in different species, one or several chromosomes more than the
male. In these forms the first maturation division is typical and the
two secondary spermatocytes are similar. But the second maturation
division is asymmetrical in that one or more chromosomes known as the
accessory or idiochromosomes fail to divide and are therefore distributed
to only one-half of the spermatids and spermatozoa. Half of the sperm

cells then have ~ chromosomes, the other half ~ plus one, or more as
2 £

the case may be in different species. These striking phenomena and
their relation to the question of sex determination are described more
fully in Chapter VII.

In conclusion we should mention briefly the place of the maturation
divisions in the life histories of different organisms. In any many-celled
organism the life cycle as a whole may be said to consist of two phases,
one characterized by the possession of the diploid chromosome group,
the other by the haploid group. Among all of the Metazoa, and many
of the Protozoa, there are invariably only two cell generations with the
haploid number, and further, these always are the tw.o final generations
in the process of gametogenesis. Here they seem bound up with the
process of fertilization and are to be understood only from the point of
view of what is involved in this process. Considering these forms alone
it is difficult to understand how it should have come about that numerical
reduction of the chromosomes should occur in advance of the condition
out of which arose the necessity for reduction, namely, the fusion of the
germ nuclei. But this arrangement is by no means invariable. In
Amoeba diploidea (Hartmann and Nagler) reduction does actually occur
after conjugation. And in some of the lower plants, such as many of
the green Algae (Chlorophycese), the relation between fertilization and
numerical reduction is that which apparently must have been the more
primitive. In these forms the gametic nuclei contain the same number
of chromosomes (s) as do the somatic or vegetative cells; these fuse
forming a zygote,with double this number (2s). This fusion is then
followed immediately by two maturation divisions, the first of which is
usually heterotypic, which result in the formation of four cells, each
again with the original vegetative number (s). Certain or all of these
four cells then produce the body of the new organism, all the cells of
which, including the germ cells when these form, have this same chro-
mosome number (s). That is, numerical reduction of the chromosomes
follows syngamy, a relation which seems more understandable than the
more common precedence of reduction. In describing cases like these

MATURATION 161

we might say that the somatic or vegetative cells and the germ cells all
have the haploid chromosome group. In fertilization the diploid group
is formed, but is then retained through only two generations, after which
the haploid condition is restored. In other words the predominating
stage in the life cycle is that with the haploid chromosome group, the
diploid group occurring only in the divisions following fertilization;
"haploid" is here synonymous with "somatic."

In many other plants, such as the ferns and mosses among others, the
life cycle is more equally divided into two distinct periods, one carried
on with the haploid (in the usual sense of the word), one with the diploid
chromosome group. The cells of the fern while it is in the typical
"fern-plant" stage have the diploid group, but during the formation
of spores by this plant, reduction occurs, the reduced number appearing
in the spore mother cell. And in all of the cells of the prothallus,
derived from the spore, the haploid number remains; no further reduc-
tion occurs when the prothallus forms gametes. The diploid number is
only restored by the union of two gametes in the formation of the new
fern plant, throughout the existence of which it is retained. It is a
matter of considerable theoretical interest that the familiar alternation
between the sporophyte and gametophyte generations, between fern
plant and prothallus, for example, should be accompanied by a corre-
sponding alternation between the diploid and haploid chromosome
groups. We may relate this to the condition in the green Algae by saying
that the number of cell generations following fertilization, in which the
diploid chromosomes are retained, is greatly increased and the number
with the haploid group correspondingly diminished, indeed in most
cases here, the diploid stage is of greater duration than the haploid. In
the higher plants (Gymnosperms and Angiosperms) it is agreed that the
prothallus, i.e., the stage with the haploid chromosome group, is repre-
sented only by certain vestiges — the pollen tube and embryo sac, and it
is significant that here, after the two maturation divisions leading to the
formation of the germ cells, two or more (but never more than a few)
additional divisions occur giving rise to these vestiges; the haploid
chromosome number is found in all of these divisions. Here then the
phase with the reduced number of chromosomes is still more limited —
practically to the extent found in animals. And whereas in the lower
plants the diploid stage is restricted to two cell generations, in the higher
plants it is the haploid stage which comes to be so limited.

Many consider the gametophyte generation, i.e., the prothallus, or
its equivalent in other forms, as the primary form or phase ; consequently
they regard the number of chromosomes in the cells of this phase, the
haploid number, as primitive or normal, and not as a reduced number.
Correspondingly the diploid number would result from a doubling, not
from a restoring of the normal. The development of this point of view

162 GENERAL EMBRYOLOGY

in connection with the conditions in the higher plants (Strasburger) has
led to the suggestion (Whitman) that even in animals the number of
chromosomes in the secondary 06- and spermatocytes and mature germ
cells, i.e., the haploid number, is again in reality the normal, that this
is doubled in fertilization, and remains doubled throughout the somatic
divisions, only to be again reduced to normal by the subsequent matura-
tion divisions. Upon this hypothesis, which also explains the present
precedent relation of maturation to fertilization, the two cell generations
immediately preceding fertilization are all that remain of the primary
phase of the animal life cycle.

The alternation between the sporophyte and gametophyte in ferns
and mosses is truly an alternation of generations and we may thus see
an alternation of generations even in the higher plants where there may
be a total of only four divisions with the haploid number — the normal
according to some. If this is allowed, it is possible that in animals
where there are but two of these corresponding cell divisions, we might
still speak of an alternation of generations as well ; the equivalent of the
gametophyte would then be represented, vestigially, only by the cells
with the haploid chromosome group, i.e., primary and secondary 06-
and spermatocytes which form the gametes proper — and the equivalent
of the sporophyte generation would be represented by all the remaining
generations of cells which we commonly think of as the true organism,
and which forms "asexually" the 06- and spermatogonia — the equiva-
lents then of the spore mother cells. Of course in animals the matter
is complicated greatly by the separation of the two sexes as two sepa-
rate individuals. Such a comparison as this must remain, at least for
the present, as an interesting speculation merely, for none of the
Metazoa offers any variations in the maturation process which shed any
light upon the comparison.

REFERENCES TO LITERATURE

AGAR, W. E., The Spermatogenesis of Lepidosiren paradoxa. Q. J. M. S.

67. 1911.

VAN BENEDEN, E. (See ref. Ch. II.)
BOVERI, T., Zellenstudien I. 1887. (See ref. Ch. III.)
BRAUER, A., Zur Kenntniss der Spermatogenese von A scaris megalo-

cephala. Arch. mikr. Anat. 42. 1893. Zur Kenntniss der Rei-

fung des parthenogenetisch sich entwickelnden Eies von Artemia

salina. Arch. mikr. Anat. 43. 1894.
CALKINS, G. N., and CULL, S. W. (See ref. Ch. II.)
CARDIFF, I. D., A Study of Synapsis and Reduction. Bull. Torrey

Botan. Club. 33. 1906.
COE, W. R., The Maturation and Fertilization of the Egg of Cerebratulus.

Zool. Jahrb. 12. 1899.

MATURATION 163

DAVIS, B. M., Nuclear Phenomena of Sexual Reproduction in Algae.

Amer. Nat. 44. 1910.
FARMER, J. B., and MOORE, J. E. S., On the Maiotic Phase (Reduction

Divisions) in Animals and Plants. Q. J. M. S. 48. 1904.
FICK, R., Ueber die Vererbungssubstance. Arch. Anat. u. Physiol.

(Anat. Abth.). 1907.
GATES, R. R., The Mode of Chromosome Reduction. Botan. Gaz. 51.

1911.
GREGOIRE, V., Les Cineses de maturation dans les deux regnes. L'unite*

essentielle du processus meiotique. Cellule. 26. 1910.
HARTMANN, M., und NAGLER, K., Copulation der Amceba diploidea,

n. sp., etc. Sitz.-Ber. Ges. Nat. Freunde. Berlin. 4. 1908.
HERTWIG, 0., Beitrage zur Kenntniss der Bildung, Befruchtung und

Teilung des Tierischen Eies. I. Morph. Jahrb. 1. 1875.
JORDAN, H. E., The Spermatogenesis of the Opossum (Didelphys vir-

giniana) with special Reference to the Accessory Chromosome and

the Chondriosomes. Arch. Zellf. 7. 1911.
KORSCHELT UND HEiDER, Lehrbuch, etc. (See ref. Ch. III.)
McCLUNG, C. E., The Chromosome Complex of Orthopteran Spermato-

cytes. Biol. Bull. 9. 1905.
MONTGOMERY, T. H., JR., The Heterotypic Maturation Mitosis in

Amphibia and its General Significance. Biol. Bull. 4. 1903.

(Seealsoref.Ch.il, III.)
MORSE, M. W., The Nuclear Components of the Sex Cells of four Species

of Cockroaches. Arch. Zellf. 3. 1909.
PLATNER, G., (See ref. Ch. III.)

SCHAFFNER, J. H., Synapsis and Synizesis. Ohio Naturalist. 7. 1907.
SCHONFELD, H., La Spermatoge'nese chez le taureau et chez le mam-

miferes en general. Arch. Biol. 18. 1901.
DE SINETY, R., Recherches sur la Biologic et 1'anatomie des Phasmes.

Cellule. 19. 1901.
STRASBURGER, E., Ueber periodische Reduktion der Chromosomenzahle

im Entwicklungsgang der Organismen. Biol. Cent. 14. 1894.

English Translation in Annals Botany. 8. 1895. Ueber Reduk-

tionsteilung. Sitz.-Ber. Akad. Wiss. Berlin. 1904.
WILSON, E. B., (See ref. Ch. II.)
WINIWARTER, H. v., Recherches sur Povogen&se et organogenese de

1'ovaire des Mammiferes. (Lapin et Homme.) Arch. Biol. 17.

1900. (See also Anat. Anz. 21. 1902.)
WINIWARTER, H. v., and SAINMONT, G., Nouvelles recherches sur

Povogenese et Torganogenese de 1'ovaire des Mammiferes. Arch.

Biol. 24. 1909.

CHAPTER V
FERTILIZATION

THE complex processes of the formation, differentiation, and
maturation of the germ cells, described in the two preceding
chapters, are to be regarded as preliminaries to the process of
fertilization. They can be understood only as preparatory
steps leading to the final meeting of an ovum and a spermato-
zoon, and their fusion into a single cell. The cell thus formed
is a "new" organism, which immediately commences a long
series of reactions, collectively termed development, leading
finally to the establishment of a form resembling that of the
individuals from which the fusing germ cells were themselves
derived.

Among animals, with the probable exception of a few of the
simplest, an almost invariable condition of the continued ex-
istence of any specific form of protoplasm is such a periodic
mingling of the living substance of two individuals of the same
species. The few exceptions among the Metazoa are found in
the rare self-fertilizing hermaphroditic creatures, and even
here the mingled plasms may be said to have had somewhat
separate histories, although formed within the body of a single
organism. Animals which reproduce parthenogenetically, or
by such methods as budding or fission, sooner or later in their
life history exhibit these processes of germ cell formation and
fusion.

Among the plants, on the other hand, while the union of
germ cells may be a frequent, and in some cases a necessary
preliminary to the formation of a new organism, yet other
tissues and living masses composed of several kinds of tissue,
taken from almost any part of the organism, may give rise to a
new individual. Many species of plants are thus normally
propagated by cuttings of leaf, stem, or root, or by runners,

164

FERTILIZATION 165

buds, etc. This is particularly true of the Begonias, where
even a few cells, from almost any part of the growing plant,
may be removed and, under proper conditions, be made to
form a new complete organism capable of producing typical
germ cells.

The questions why fertilization should be necessary, and how
fertilization actually accomplishes the results which obviously
follow it, are not easy to answer. But the essential facts
regarding the process of syngamic fusion, and the visible
results of it, are clear. We shall, therefore, confine our atten-
tion first to these; then, having described the phenomena of
fertilization we may consider briefly some of the more theoret-
ical aspects of the process.

The word "fertilization" is a general, inclusive term, used to
denote all of the various phenomena concerned in the meeting
and fusion of the germ cells (gametes) or germ plasms, and even
some of the results of such a fusion. The simpler fact of the
mere fusion of gametes is more precisely termed syngamy. In
all of the Metazoa syngamy may be defined as the meeting of
two completely specialized, unicellular gametes, an ovum and
a spermatozoon, derived in most cases from two individuals,
and their subsequent fusion, nucleus with nucleus, and cyto-
plasm with cytoplasm, into a single, uninucleate cell, the zygote.
This definition is not completely applicable to the unicellular
organisms, for in these the gametes are usually not completely
specialized, sometimes indeed not especially differentiated at all.
Such forms may offer some suggestions as to the history and
significance of the fertilization process, and we shall return to
consider this subject later in this chapter.

We have seen in Chapter III some of the methods by which
it is ensured that eggs and sperm shall be brought into the same
general region or into fairly close proximity, but it remains to
be seen how the ovum is actually encountered by the sperm
cell. Taken altogether, the processes leading to this result often
become very complicated and special, and in most species the
probability is very high that practically every normal egg
produced will be fertilized. The gametes are completely

166 GENERAL EMBRYOLOGY

specialized cells, and if they cannot conjugate and develop,
they soon perish, not able even to remain alive long, except
in a few special instances. Spermatozoa discharged freely
into the water, as in external fertilization, are usually able to
remain active only a few minutes or hours. But when fertili-
zation is internal and the spermatozoa are received into some
reproductive cavity of the female, or into some storage cavity,
they may remain alive and able to function under appropriate
conditions for a much longer period: various observers give
the following specific instances: dog and rabbit, eight days;
man, seven to twenty days; fowl, twenty days; the bats and
some snakes, from autumn to the following spring; Salamandra
maculosa, from summer to the following spring; snails, three
years; bees, four to five years.

Whatever the particular circumstances connected with and
ensuring the meeting of sperm and ovum, the medium in which
it occurs is a fluid. In this the sperm cells are in active, though
apparently random, movement, due to rapid vibration of the
flagellum or tail. In many instances their movement follows a
spiral path (Buller, Adolphi), either close or open, such as is
common among flagellated unicellular organisms. In a few
rare instances (some Crustacea) the spermatozoa are amoeboid.

Fertilization becomes more likely when direction is given to
the active random movements of the sperm cells. In some
instances where fertilization is internal, the movements of the
sperm seem to be directed by ciliary currents of the oviduct or
other passage. Spermatozoa tend to swim against such a
current, and thus to ascend toward the eggs which are being
carried down the passage toward the exterior (Lott). Accord-
ing to the interesting observations of Lott, human spermato-
zoa swim at the rate of 27 mm. (i.e., about 550 times their
own length) in 7.5 minutes. At the corresponding rate of
progression a man 5.8 feet in stature would walk a mile in
12.4 minutes.

There appear to be two chief methods by which spermatozoa
are finally brought to the surface of the ovum. In some few
forms the egg is said to give off a chemical substance to which

FERTILIZATION 167

the active sperm are attracted; when the random movements
of the sperm bring them within the sphere of chemical influence
of the egg, their movements immediately become directed
toward the unfertilized egg. Among some of the lower plants
it is known that weak solutions of malic acid and its compounds
attract spermatozoids; in others, solutions of cane sugar act
similarly (Pfeffer). It is at present doubtful, however,
whether in many animal eggs the control is also of a chemical
nature (Buller). In some forms the stimulus is certainly not
of a chemical sort, but is a contact stimulus. The sperm of
many fishes, for example, swim at random until they touch
some solid object, egg or other body, and from this they are
apparently unable to escape. According to the observations
of Drago the collection of the spermatozoa about an ovum is
unusual, and when it occurs, it is the result of agglutination.
It should be said that in most cases it is doubtful whether the
movements of the sperm really are given direction toward the
ovum, and what the nature of the stimulus may be, when such
is the case. * As we have seen, the contact between sperm and
egg results chiefly from the large numbers of sperm produced,
and from their general proximity to the egg resulting from
special habits of spawning, copulation, etc.

When the ovum is naked, or surrounded by only a thin
vitelline membrane, the sperm apparently may enter at almost
any point on the surface of the egg. This is true of many forms
among the Medusa3, Turbellaria, Nemertea, Annelida, Echino-
dermata, Gasteropoda, Cephalochorda, Amphibia, and Mam-
malia. Entrance is usually said to be effected by the active
swimming movements of the spermatozoon, which force the
sharp acrosome, adapted to this purpose, through the limiting
surface of the ovum, into its superficial cytoplasm. But here
again extended evidence is lacking, and in many forms the egg
is known not to be a wholly passive recipient of the sperm, but
to take a considerable share in accomplishing its entrance.
Thus in the sea-urchin, when a sperm head approaches the egg
closely, the superficial cytoplasm, at the point nearest the sperm,
is elevated into a small cone or papilla called the attraction

168

GENERAL EMBRYOLOGY

cone (Wilson). This not only rises to meet the spermatozoon,
but seems to aid in drawing it into the egg (Fig. 85). In some
instances (e.g., Julus) this attraction cone may be quite high
and may contain a part of the chromatic substance of the egg
nucleus. According to the observations of Lillie, the sper-
matozoon of Nereis is clearly drawn into the egg through
the activity of the latter, the sperm itself taking no active
part in the process (Fig. 86).

FIG. 85. — Entrance of the spermatozoon into the egg. From Wilson, " Cell,"
H, after Metschnikoff; /, after Fol. A. Spermatozoon of Toxopneustes, X 2000;
a, the apical body; n, nucleus; m, middle-piece; /, flagellum. B. Contact with
the egg-periphery. C,D. Entrance of the head, formation of the entrance-cone
and of the vitelline membrane («), leaving the tail outside. In some other
Echinoderms, the tail may enter the ovum. E,F. Later stages. G. Appearance
of the sperm-aster (s) about three to five minutes after first contact; entrance-
cone breaking up. H. Entrance of the spermatozoon into a preformed depres-
sion. /. Approach of the spermatozoon, showing the attraction-cone.

When the egg is surrounded by membranes of some thickness
or density, the spermatozoa are usually unable to penetrate
them and the only path of entrance is then through the micro-
pyle, the existence of which is an adaptation for this event.
There is apparently no agent directing the sperm toward this
perforation in the membranes; the finding of it is a matter of
chance. The chance is not small, however, that some sperma-
tozoon will enter the micropyle, for ordinarily the fluids around
the egg are filled with a swarm of sperm cells. As already
noted the micropyle is commonly at the animal pole of the
egg, though at the vegetal pole in a few instances (some

FERTILIZATION

169

Molluscs). Thus the region which is to receive the sperma-
tozoon is already determined, and frequently the cytoplasm is
considerably modified, in the region just beneath the micropyle,
into a special substance concerned in the receipt of the sperm;
this is known as the entrance disc (Fig. 86).

FIG. 86. — Entrance of the spermatozoon in the fertilization of .the Annulate,
Nereis limbata. After Lillie. A. Spermatozoon. B. Perforatorium has pene-
trated egg membrane; entrance cone well developed. Fifteen minutes after
insemination. C. Thirty-seven minutes after insemination. D. Entrance cone
sinking in and drawing the head of the spermatozoon after it. Forty-eight and
one-half minutes after insemination. E. Head drawn in still further. Forty-
eight and one-half minutes after insemination. F. Entrance completed. First
maturation division in anaphase. Fifty-four minutes after insemination. The
middle piece, as well as the tail, remains outside, c, head cap; e, entrance cone;
h, head of spermatozoon (nucleus); m, middle piece; p, perf oratorium ; v, vitelline
membrane; 7, first polar division figure.

*" With but comparatively few exceptions, only one sperm cell
normally enters a single egg (monospermy) . This sperm is the
first one to reach the egg or micropyle, and there are various
methods of excluding additional sperm, and thus of preventing

170

GENERAL EMBRYOLOGY

abnormal multiple fertilization. In some of the lower plants,
after one sperm cell has entered, the egg gives off immediately
a chemical substance which actually repels the other sperm
congregated about the egg. A frequent method among animals
is the secretion of an impenetrable membrane, or a layer of
jelly, immediately upon the entrance of one spermatozoon.
Or, if a membrane was previously present, its density may be
suddenly increased, or an additional membrane formed

I

sp f sp

FIG. 87. — Polyspermy in the egg of the Elasmobranch, Torpedo ocellata.
From Ziegler, after Ruckert. Germ disc with first cleavage spindle, /, and acces-
sory sperm nuclei, sp.

(Amphioxus), or the micropyle may be closed by the rapid
swelling of the egg membranes. Although this process of
membrane formation may really have this effect of excluding
supernumerary spermatozoa, the general significance of the
process renders it doubtful whether this is to be regarded
primarily as an event adapted toward this1 end.
' Occasionally two or more spermatozoa succeed in gaining
entrance into the ovum (polyspermy). This ordinarily results
in an abnormal course of development, which does not proceed
very far before the egg ceases to develop and dies. V A few
forms, however, are adapted to the receipt of more than one
sperm and polyspermy occurs normally (physiological poly-
spermy). Such eggs (Fig. 87) are usually yolk-filled, for

FERTILIZATION 171

example those of some Insects, Petromyzon, Selachians, Uro-
deles, Reptiles, Birds, and perhaps the toad and some Teleosts.
Sometimes a few, sometimes many, sperm thus enter the
ovum, but in any case only one of them ever takes any real
part in the actual processes of fertilization. The others, known
as accessory spermatozoa, may either remain quite inactive and
soon degenerate, or they may give rise to "vegetative" nuclei,
and perish after a brief period of activity. While active they
seem chiefly to be concerned in the preparation of the yolk for
ready absorption; they are then called merocytes. Rarely, if
ever, do the nuclei derived from accessory spermatozoa con-
tribute directly to the formation of any part of the embryo
proper.

Apparently there is little specific adaptedness in the behavior
of the germ cells such that an egg and a sperm of the same
species tend to unite much more readily than do those of differ-
ent species. With some eggs any spermotozoon that is morpho-
logically capable of gaining entrance, can do so, apparently
about as readily as the specific sperm. The limitations here
are frequently due to the size of the sperm head as compared
with the micropyle, or to the necessity for special perforating
mechanisms or powerful swimming movements in order to
penetrate the egg membranes, or the performance of appro-
priate reactions upon the part of the egg itself. As a rule the
eggs and sperm of a single species unite, because, as. the result
of the breeding or spawning habits, only the sperm and ova
of a single species are associated in time and space in any con-
siderable numbers. When eggs are placed in a mixture of
equal quantities of two or more kinds of sperm, there seems to
be no appreciable selective fertilization, provided, as said above,
that both or all kinds of the sperm are able to enter the egg
at all.

The ease with which "foreign" sperm may enter an egg is
affected in many instances by chemical treatment of the eggs
and sperm; treatment with alkalies or with specific salts often
renders penetration of the sperm readily possible in cases
where normally it is difficult or impossible (Loeb, Godlewski).

172 GENERAL EMBRYOLOGY

And once within the ovum, a " foreign " sperm seems to act
almost as efficiently as the proper sperm in inaugurating
development. After the entrance of a foreign sperm the two
germ nuclei may not, usually do not, fuse, and other internal
developmental processes may not be entirely normal, but the
external processes of cleavage and differentiation may proceed
normally for some time, even to the formation of a free-swrniming
larva, as in many species of Echinoderm eggs fertilized by the
sperm of other species, genera, or even of other classes of Echino-
derms (Baltzer), or of other phyla (Mollusca, Kupelwieser).

In many species the entire spermatozoon enters the cyto-
plasm of the ovum (some of the Turbellaria, Annelids, Insects,
Molluscs, and many Vertebrates) while in others the tail piece
separates from the remainder of the sperm cell and is left
outside of, or embedded within, the vitelline membrane, so
that only the head and middle piece actually share in the
formation of the zygote. In some instances the middle piece,
too, fails to enter the egg (e.g., Nereis, Fig. 86). Once within
the egg, the sperm continues its inward course for a short
distance only. If the entire sperm cell has entered, one of the
first events is a sharp bending or flexure between the tail and
middle piece, often followed by a separation of the two, after
which the tail piece is left behind as the remainder continues
its migration (Fig. 88).

The entrance of the spermatozoon within the egg cytoplasm
is the event which inaugurates a whole series of fertilization
processes culminating in the formation of a typical mitotic
division figure within the zygote. The precise character of the
stimuli which start this chain -of actions is still in doubt, but it
seems likely that in many instances it acts, first, by bringing
about the formation of a permeable membrane over the surface
of the egg, through which may occur rapid and extensive
osmotic interchanges leading to marked oxidations; and second,
by reducing the amount of fluid in the egg cytoplasm, either
actually, by its loss through the permeable membrane, or
relatively by the addition of the much denser substance of the
spermatozoon itself (Loeb). At any rate, whether it be a

FERTILIZATION

173

FIG. 88. — Fertilization in the sea-urchin, Toxopneustes. From Wilson, "Cell."
A-F, X 1067 -,G, X 533 ;#, /, X 667. A. Sperm-head before entrance; n, nucleus;
m, middle-piece and part of the flagellum. B,C. Immediately after entrance,
showing entrance-cone. D. Rotation of the sperm-head, formation of the
sperm-aster about the middle-piece. E. Casting off of middle-piece; centro-
some at focus of rays. F,G. Approach of the pronuclei; growth of the aster.
H. Union of pronuclei. /. Flattening of the sperm pronucleus against the egg
pronucleus; division of the aster.

174 GENERAL EMBRYOLOGY

Tprimary or a secondary process, this loss of water following
' the entrance of the spermatozoon, appears as one of the
I important aspects of fertilization.

In the eggs of many species there is a peripheral layer of

FIG. 89. — Changes in the structure of the ovum in Nereis, upon fertilization.
After Lillie. A. Unfertilized oocyte. Large germinal vesicle; cytoplasm con-
tains oil drops and yolk spheres, and shows well marked cortical layer (exoplasm).
B. Fifteen minutes after ensemination (the spermatozoon is not shown). The
cortical layer has chiefly gone to form a jelly-like layer outside the ovum, and is
not shown. C. Egg after centrifuging to show component substances, c,
cortical layer (exoplasm); n, germinal vesicle or nucleus; no, nucleolus; o, oil
drops; p, perivitelline space; v, vitelline membrane; y, yolk spheres; 1, layer of
oil drops; 2, hyaline cytoplasm with small basophile granules; 3, yolk spheres;
4, hyaline cytoplasm with large basophile granules.

cytoplasm (exoplasm) which is comparatively clear, free from
granules, and characterized by the presence of fluid vacuoles
(Echinoderms, Nereis (Fig. 89), Amphioxus (Fig. 90), Teleosts;
see Chapter III). Entrance of the spermatozoon leads to

FERTILIZATION

175

the breaking down of these vacuoles and the discharge of their
substance from the surface of the ovum (Figs. 89, 90). This
substance may be in part transformed into, or may carry
before it a modified surface layer of material which appears then

FIG. 90. — Fertilization in the egg of Amphioxus. C, after Cerfontaine, others
after Sobotta. A. Ovarian egg showing cortical plasm. B. Cortical layer form-
ing a membrane on the surface of the egg, within the vitelline membrane. C.
Egg membrane fully formed but still attached to surface of egg. D. Extruded,
fertilized egg. Membrane fully formed and beginning to leave the surface of
the egg. c. Cortical layer; e, endoplasm; m, egg membrane; externally vitelline,
internally a product of the exoplasm; p, perivitelline space; s, spermatozoon;
v, vitelline membrane; /, first polar body; II, second polar spindle.

as & fertilization membrane; this may be the vitelline membrane
or it may be an addition to a previously present vitelline
membrane (Fig. 90) . Then either by the shrinkage of the egg,
or the expansion of the membrane, or both, or by the rapid ab-

GENERAL EMBRYOLOGY

sorptionof water by the substance between the egg and the mem-
brane, a space of widely varying dimensions in different species,
is left between the egg and its membranes; this is the perivi-
telline space (Fig. 90). Within this space the egg is free to

FIG. 91. — Sections through the egg of the Tunicate, Cynthia partita. After
Conklin. X about 350. A. Ovarian egg fully formed. Germinal vesicle
surrounded by yolk bodies; peripheral layer of protoplasm containing test cells
and yellow granules (small circles). B. After extrusion of the test cells. Nuclear
membrane still intact with chromosomes at periphery of nucleus (germinal
vesicle). C. After laying (before fertilization, the egg remains in this condition
•until fertilized). Chromosomes and granular substance, from which the spindle
is formed, lie in the center of the karyoplasm, now free in the cell, e, exoplasm
or cortical layer; g, granules of yellow pigment; n, egg nucleus or germinal vesicle;
>,, nuclei of ingested test cells or follicle cells; y, yolk.

move or rotate, although the superficial membrane may be
fixed to some foreign body.

Frequently this phenomenon of membrane formation is but

FERTILIZATION 177

one phase of a general physical and chemical reorganization
of the whole substance of the egg, following the entrance of the
spermatozoon. The egg may exhibit more or less amoeboid
movement, or waves of contraction may pass over it. A
frequent result is seen in the rapid streaming of differentiated
cytoplasmic substances into certain regions, where these specific
substances collect. Thus in many yolk-filled eggs like those of
the Teleosts, the protoplasm, which before the entrance of the
sperm is quite uniformly distributed over the surface of the
egg as a very thin layer, now collects at the animal pole into
a thick and fairly circumscribed disc called the qerrn^disc (Fig.
48). The Ascidian egg, as described by Conklin, offers one of
the most marked examples of this rapid transformation and
redistribution of the substances of the egg cytoplasm (Figs.
91, 92). In the secondary oocyte of Cynthia (Styela) the
greater part of the cell is composed of a gray "endoplasm";
superficially there is a thin but complete layer of yellowish
" mesoplasm" ; while the large nucleus or germinal vesicle
contains a clear "ectoplasm." During maturation, which here
precedes sperm entrance, the ectoplasm collects at the upper
pole of the oocyte. Immediately upon entrance of the sperm
the yellow meso plasm streams from all directions toward the
lower pole; this is followed by the clear ectoplasm which forms
a stratum just above the mesoplasm, and leaves the upper
half or more of the egg cytoplasm composed entirely of the gray
endoplasm. Then this radial or rotatory symmetry gives
place to a bilateral symmetry, for the mesoplasm and ecto-
plasm move up on one side (the posterior) of the egg, appearing
on the surface in the form of a crescent just below the equator.
Meanwhile the yellow mesoplasm and gray endoplasm have
each become differentiated into two distinct substances, so
that altogether five forms of protoplasm are distinguishable
in the cytoplasm of the zygote (Fig. 92).

Very few eggs exhibit such marked differentiation as this,
but the corresponding phenomena are of widespread occurrence,
and it is quite likely that they have frequently been overlooked
because the various substances are not often marked by

178

GENERAL EMBRYOLOGY

cr

cr

E

FIG. 92. — Total views of the egg of the Tunicate Cynthia partita, showing the
changes in the arrangement of the materials of the egg subsequent to fertiliza-
tion. After Conklin. X 200. A. Unfertilized egg, before the fading out of the
germinal vesicle. Centrally is the mass of gray yolk; peripherally is the proto-
plasmic layer with yellow pigment, and surrounding the egg, the test cells and
chorion. B. About five minutes after fertilization, showing the streaming of the
superficial layer of protoplasm toward the lower pole, where the spermatozoon
enters, and the consequent exposure of the gray yolk of the upper hemisphere.
The test cells are also carried toward the lower pole. C. Side view of egg showing
the yellow protoplasm at the lower pole; at the upper pole a small clear region
where the polar bodies are forming. The location of the sperm pronucleus is also
indicated. D. Side view of egg shortly before the first cleavage, showing the
posterior collection of the pigmented protoplasm (yellow crescent) and the clearer
area above it. E. Posterior view of egg during the first cleavage, showing its

FERTILIZATION 179

characters so easily observed in the living egg. It is also note-
worthy that frequently a radial or rotatory symmetry of
the egg is changed to a bilateral symmetry by the entrance
of the spermatozoon, and that usually the position of the
plane of bilateral symmetry is determined by the point at
which the sperm enters or by the path which the sperm takes
through the cytoplasm. And further this new plane of sym-
metry of the zygote coincides closely with the plane of the first
division of the zygote and with the median plane of the embryo
and adult.

These aspects of organization and reorganization of the egg
are among the highly important aspects of development, and
largely determine many of the phenomena of subsequent
differentiation. They also illustrate the statement made in
the introductory chapter, that some of the most important
aspects of development are w^ra-cellular processes. We shall
return to this subject in a later chapter (Chapter VII).

After the entrance of the spermatozoon and the consequent
redisposition of the substances of the cytoplasm, the course
of the immediately subsequent events is determined largely
by the state of the egg nucleus as regards its maturation process.
For in most cases maturation proceeds only to a certain point,
varying greatly in different forms, when it pauses, and is com-
pleted only after receiving the stimulus caused by the entering
sperm. Ordinarily the state of the egg is such that sperm can
gain admission only when this pause has been reached.

In some cases, such as Nereis (Fig. 86), Ascaris (Fig. 94),
Cerebratulus (Fig. 95), and many Molluscs, it is true that the
sperm does enter the ovum before the maturation process has
even begun — while the egg nucleus (germinal vesicle) is still
in the final resting stage preceding the first oocyte division.
In other cases — most Molluscs, Thalassema, Sagitta, Teleosts,
the first polar spindle has formed and the first maturation
division may have reached the metaphase or anaphase, when

relation \o the symmetry of the egg. a, anterior; c, clear protoplasm; cr, yellow
crescent; e, exoplasm or cortical layer, with yellow pigment; g.v, germinal vesicle
k, chorion; p, posterior; p. b, polar bodies; t, test cells; y, yolk (central gray mate-
rial); y.h, yellow hemisphere; d\ sperm pronucleus.

ISO GENERAL EMBRYOLOGY

it pauses to await the sperm entrance. In Sycandra, Lepas,
Amphioxus (Fig. 90), and many Amphibia and Mammalia the
first polar division is completed and the second polar spindle
formed before the pause. And finally in some Ccelenterata
and most Echinodermata and Ascidians, the maturation of
the egg is entirely completed before the entrance of the sperma-
tozoon. In such cases as these, contact with sea water seems

FIG. 93. — Diagrams of the two most frequent relations between the events of
maturation and fertilization. From Wilson, "Cell." /. Polar bodies formed
after the entrance of the spermatozoon (Annulates, Molluscs, Turbellaria). II.
Polar bodies formed before the entrance of the spermatozoon (Echinoderms). A.
Sperm pronucleus and centrosome at d\ first polar body forming at 9 . B.
Polar bodies formed; approach of the pronuclei. C. Union of the pronuclei.
D. Approach of the pronuclei. E. Union of pronuclei. F. Cleavage nucleus.

to furnish the stimulus to complete the maturation process,
which may be begun before the eggs are produced.

We may distinguish two general types of behavior on the
part of the germ nuclei according to whether maturation has or
has not been completed at the entrance of the sperm (Fig. 93).
We shall consider first, and at greater length, those cases in
which maturation is not yet completed, for this would seem
the more usual course of events. Otherwise maturation occurs
precociously, apparently before the necessity for it has arisen.

FERTILIZATION 181

We left the sperm head and middle piece lying a short distance
below the surface of the egg. We may disregard the tail piece
now, for even in those cases in which it enters the egg it is left
behind the head and takes no active part in subsequent proc-
esses. The head and middle piece now move more slowly,
along a path which is, for a short distance at any rate, a
radius of the ovum. Then they separate slightly, and the two
rotate through approximately 180°, so that the middle piece is
placed in advance of the head (Figs. 94, 95). There ensues a
considerable metamorphosis of these elements. The sperm
head loses its sharp outline and gradually enlarges; its outline
soon becomes very irregular and indistinct, and vacuoles
appear. Soon it has expanded into an organ of considerable
size and has acquired a typical nuclear structure with linin
network, chromatin granules, and nuclear membrane. In the
meantime the middle piece has undergone an even more exten-
sive transformation (Fig. 94). Before the sperm halts in its
inward progress, even before the rotation in some cases, the
middle piece has begun to dissolve and in connection with it
appears a centrosome, surrounding which a small aster appears.
During the pause of the sperm nucleus, the centrosome and
aster each divide into two, and the daughters diverge slightly
while the asters grow somewhat larger, y

It will be remembered that during the metamorphosis of the
spermatid into the spermatozoon, one or both of the centrosomes
of the former either passed into the middle piece, or actually
formed the larger part of it, although frequently no centrosome
is actually visible in the middle piece of the fully formed
spermatozoon.! When the centrosome appears in the egg, in
connection with the middle piece of the entering spermatozoon,
it is possible, though not likely, that this is really the same
centrosome that was present in the spermatid, and that there
is consequently a genetic continuity of centrosomes, from gen-
eration to generation, as well as of nuclear components. Such
a continuity has, however, not been definitely observed. On
the other hand, it may be that the middle piece forms, either
from its owrn substance, or from that of the egg, a new centro-

182

GENERAL EMBRYOLOGY

P6

FIG. 94. — Fertilization in Ascaris megalocephala bivalens. From Wilson,
"Cell," after Boveri. (Later stages are shown in Fig. 36.) A. The spermato-
zoon has entered the egg, its nucleus is shown at d1; beside it lies the granular
mass of "archoplasm" (attraction-sphere); above are the closing phases in the
formation of the second polar body (two chromosomes in each nucleus). B.
Germ-nuclei (9 , <?) in the reticular stage; the attraction-sphere (a) contains the
dividing centrosome. C. Chromosomes forming in the germ-nuclei; the centro-
some divided. D. Each germ-nucleus resolved into two chromosomes; attrac-
tion-sphere (a) double. E. Mitotic figure forming for the first cleavage; the
chromosomes (c) already split. F. First cleavage in progress, showing divergence
of the daughter-chromosomes toward the spindle-poles (only three chromosomes
shown).

FERTILIZATION 183

some. The aster doubtless is formed out of the egg cytoplasm,
by the influence of the centrosome or centrosome-forming
substance of the spermatozoon; and it is quite possible that the
centrosome itself may be similarly formed from the cytoplasm
through the action of some chemical substance introduced by
the sperm. For it is known that the cytoplasm of the egg
does possess the property of forming asters with typical centro-
somes, under the influence of appropriate "artificial" stimulus
(Yatsu). And recently Lillie has shown, in Nereis, where the
middle piece does not enter the egg at all, that the centrosome
forms in association with the sperm nucleus, even when only a
small portion of this is allowed to enter the egg. This indicates
that the centrosome, as well as the aster, results from the
redisposition of substances of the egg cytoplasm following the
entrance of the spermatozoon. A conclusion as to whether or
not the law of genetic continuity applies to the centrosome in
fertilization is less important than recognition of the uniformity
of its chemical and physical actions, in either case. And
although the centrosome as an organized body may disappear
in the spermatozoon, this still contains kinoplasmic substance
of an equivalent function. Here, as elsewhere, the essential con-
tinuity may be chemical rather than morphological, but for
that reason it is not to be regarded as any less actual or

portant.

While the spermatic structures have been thus active, the
egg nucleus has been completing its maturation, at the con-
clusion of which the egg centrosomes and asters have disap-
peared (Figs. 94, 95). The egg nucleus is left near the surface
of the animal pole, either near the sperm nucleus or at some
distance from it. There are now present in the ovum all of
the chief elements which are to take part in the essentials of
fertilization and development. These are (1) the egg nucleus

o

with its ^ chromosomes, either distinct or formed into a
characteristic nuclear reticulum, and with or without a nuclear

o

membrane; (2) the sperm nucleus, also known to contain -=

184

GENERAL EMBRYOLOGY

FIG. 95.— Fertilization in the Nemertean, Cerebratulus. After Coe. C, D,
X 375, others X 250. A. Primary oocyte. Part of the chromatin has been
condensed into chromosomes, only five of which are shown (the number present
is sixteen). The remainder of the chromatin is thrown out into the cytoplasm.
The centrosomes, each with a small aster, are diverging, and the nuclear mem-
brane is commencing to disappear. B. First polar spindle fully formed and ro-
tated into radial position. Chromosomes in equatorial plate. C. First oocyte
division; anaphase. D. First polar body nearly separated. E. First polar body
completely cut off; second polar spindle formed and rotating into radial position.
Spermatozoon within the egg. F. Second polar body completely separated.
Egg pronucleus forming, surrounded by large aster. Sperm pronucleus, also
with a large aster, enlarged and approaching the egg pronucleus. G. Approach
of the two pronuclei. Egg aster reduced, sperm aster greatly enlarged and
centrosome divided. H. Fusion of pronuclei; divergence of the sperm
centrosomes. I. First cleavage figure in early anaphase. Chromosomes divided
and beginning to diverge; centrospheres enlarged, c, chromosomes; o, nucleolus,
vacuolated and commencing to disappear; s, spermatozoon just within the egg;
D, germinal vesicle; vc, contents (extra-chromosomal) of germinal vesicle; /, //,
first and second polar bodies; d\ sperm pronucleus; 9 , egg pronucleus.

FERTILIZATION 185

chromosomes; (3) the centrosomes and asters derived in some
way, either directly or indirectly, from the spermatozo6n.&;In
syngamy, therefore, the ovum supplies the great bulk of the
cytoplasmic basis of the zygote, together with one-half the
nuclear material, while the spermatozoon furnishes the other
half of the nuclear substance, and produces the centrosomes,
which here as elsewhere are to be regarded as the dynamic
centers for division. It should not be overlooked that a small
amount of cytoplasm, including certain mitochondrial structures,
does accompany the sperm, particularly when the tail piece
enters the egg; it is by no means impossible, though not at all
demonstrated, that this cytoplasm from the sperm may con-
tain substances of great importance in later development and
differentiation. In brief, however, it is true that in this union
of gametes the ovum is the material factor, the Spermatozoon
the dynamic, and each contributes equally to the nuclear or
controlling mechanism.

' But these structures are as yet distributed in different parts
of the cell. The association of the scattered elements into a
typical mitotic figure now follows and constitutes the final
step in fertilization and the formation of the new organism.
Maturation completed and the sperm nucleus dissolved, the
two germ nuclei commence to approach one another, the sperm
nucleus following the centrosomes and asters. l The paths of
their approach are seldom directly toward one another, as
they are in some of the Nematodes, but are more or less
curved (Fig. 96), and seem in a way determined by some factors
other than mere mutual attraction, though this is doubtless the
essential factor in their movement (Wilson).

The entrance path of the spermatozoon is frequently marked
by cytoplasmic modifications, often of a very pronounced
character, giving evidence of intense metabolic (katabolic)
activity ; thus we might note the frequency of the accompanying
formation of pigment, which is usually regarded as a by-product
of protoplasmic decomposition.

We have already seen that the path of the sperm nucleus
may be an important factor, either in determining or in making

186

GENERAL EMBRYOLOGY

evident, the position of the plane of symmetry of the zygote,
and hence of the embryo. In a few cases the nuclei are some-
what amoeboid, in others they seem to be carried by proto-
plasmic currents, and in still others they seem directed by the

FIG. 96. — Diagrams showing the paths of the germ-nuclei in four different eggs
of the sea-urchin Toxopneustes. From camera drawings of the transparent
living eggs. From Wilson, "Cell." In all the figures the original position of
the egg-nucleus (reticulated) is shown at 9 ; the point at which the spermatozoon
enters at E (entrance-cone). Arrows indicate the paths traversed by the nuclei.
At the meeting-point (M) the egg-nucleus is dotted. The cleavage-nucleus in
its final position is ruled in parallel lines, and through it is drawn the axis of the
resulting cleavage-figure. The axis of the egg is indicated by an arrow, the
point of which is turned away from the micromere-pole. Plane of first cleavage,
passing near the entrance-point, shown by the curved dotted line.

location of the asters. These have grown to a considerable
size now, and seem mechanically forced into that position where
the tensions between the cytoplasm and the more or less rigid
asters reach an equilibrium. Most frequently this is the center
of the cytoplasmic mass, so that ultimately the two nuclei

FERTILIZATION 187

M

approach and meet in this region.^As the nuclei come into
contact the two asters diverge in such a way as to lie at opposite
ends of a tangent drawn through the point of nuclear contact
(Figs. 94, 95). The nuclear walls then dissolve; a spireme

o

forms in each nucleus and segments, in each, into ^ chromo-

somes, and these, as a typical spindle forms from the egg
cytoplasm, become arranged at its equator. The result is the
formation of a typical mitotic figure with s chromosomes. This
is the first cleavage figure, and here, for the first time in the
existence of the new organism, substances of paternal and
maternal origin are associated, on equal terms, in a common

Before we mention any of the further details in the history
of this cleavage figureyVe must return to consider briefly the
course of events in the fertilization of those eggs which are
already fully mature when the sperm cell enters (Echinoderms,
Ascidians). In such cases (Figs. 88, 93) the chief divergences
from the account just given result from the absence of any pause
of the sperm nucleus and middle piece after their entrance.
The egg nucleus is ready for fusion, and immediately upon the
entrance of the sperm the two nuclei proceed toward one
another as described above. The sperm nucleus thus does not
have time to be dissolved to any considerable extent, so that
when the two nuclei meet they are by no means of equal size
(Fig. 88), for the egg nucleus nearly always returns to its " rest-
ing" state after its maturation is completed. Nevertheless it is
known from their history that the two nuclei are equivalent in
chromosomal composition. Frequently, too, the centrosome
does not divide until just as the two nuclei meet, or even after
they have begun to fuse. The two centrosomes accompanied
by asters then move to opposite poles of the combined nuclei
and there establish the mitotic figure. The sperm nucleus in
these cases does not become resolved into a typical nuclear
condition until after its fusion with the egg nucleus. It often
results from this that the two nuclear substances seem to mingle
quite completely before the spindle is formed and it is not so

188 GENERAL EMBRYOLOGY

easy, as in the cases previously described, to distinguish at
this time between the elements derived from the egg and sperm
nuclei. When this duplex nucleus forms Jits spireme, this
segments into the somatic number of chromosomes immedi-
ately, and the mitotic figure for the first cleavage forms
typically.

In the formation of the first cleavage figure we see the net
result of all the complex processes of the formation and matura-
tion of the germ cells, and the union of the two gametes. In a
word, what has been accomplished is the reestablishment
of a single typical cell with specific organismal characteristics.
But this cell now has a nucleus derived in equal parts from two
separate individuals of the species. Intp__this nucleus the
events of maturation have made it possible that there should
have been brought a complete and equivalent series of chromo-
somes from each parent, for the haploid group is composed of
one of each pair of chromosomes of the diploid or somatic
series. And from this nucleus are derived all of the nuclei of
the developing organism; hence every cell of the adult body may,
probably does, contain substance derived from both its parents.

We may regard the organization of these two haploid series
into a single nucleus as the culmination of the whole process of
fertilization. Or, on the other hand as previously suggested,
we may consider the final step in fertilization as not occurring
until these pairs of chromosomes actually fuse in synapsis
during the maturation of the succeeding generation of germ
cells. From this point of view the actual union of maternal
and paternal structures and substance never occurs in the
somatic cells, for in these synapsis is not known. There is
involved in this process of fertilization much more than these
simple morphological facts express, and to this subject we
shall return presently.

We shall find it profitable to consider now, as briefly as may be, the
phenomena of fertilization and accompanying gamete formation in a
series of unicellular organisms of increasing complexity and resemblance
to the Metazoa. This subject has been in part postponed from Chapters
I and III, and it should be stated again that the series to be described

FERTILIZATION 189

is not supposed to represent a phyletic relation. It is now too late to
state with any considerable degree of probability the course of evolution
of the germ cells and the process of syngamy. Apart from this, how-
ever, the consideration of such a series as this brings out many important
and interesting facts regarding the general process of fertilization, and
emphasizes the idea that this complicated process as we see it in the
higher organisms to-day, is all the product of an evolutionary history.

As a preliminary distinction of a general and underlying character we
should note that among the unicellular forms the cells which meet or
fuse may be of the same race or family, that is, closely related by descent
from a comparatively recent common ancestor; or they may be only
distantly related, so distantly as to be regarded as unrelated, coming
from races or families that have long been distinct, and that have had
different histories. The former condition is termed endogamy, the latter
exogamy. These two relations may be distinguished in all forms of
syngamy or conjugation; exogamy is much the more usual, and
involves the more complicated reproductive processes among the
Protozoa, but no such relation seems really necessary, for conjugation
may occur with equal facility between any two different individual
protoplasms, whether closely related or not.

In a general way we may arrange the varied phenomena of conjuga-
tion and syngamy in three classes with reference to the nature and
extent of the fusion which occurs. In its simplest form this fusion is
not morphological, but is expressed by the congregation of cells in
groups; this is to be regarded as a form of cytotropy. Occasionally
large collections of cells result and the elements come into close and
extensive contact without really fusing or losing cell limits. Whatever
exchange of substance there may be occurs through an osmotic process.
After a temporary association of this kind the cells scatter and resume
vegetative and reproductive processes. Such a process of cytotropy
has been observed in Amoeba (Rhumbler) .

The simplest form in which a real fusion of plasmas occurs is that
known as plastogamy. Here two or sometimes more (2-30 in Actino-
phrys) vegetative cells meet and flow together so that the cytoplasms
mingle completely; the cell nuclei remain separate, though osmotically
they may affect one another and the fused cytoplasms. The result of
this is the formation of a physiologically bi- or multinucleate cell.
Plastogamy may be only temporary; in such a case the cells come into
relation only through comparatively limited contact surfaces and the
original cell outlines are not lost. Then after a brief period, during
which chemical interchanges may occur, the cells separate again.
More frequently, however, plastogamy is permanent, and the fusion of
the cells is so complete that the original cell outlines are completely lost.
Then, following plastogamy, the nuclei of the combined cells usually

190

GENERAL EMBRYOLOGY

divide several times forming a considerable number of smaller nuclei;
finally the cytoplasm divides correspondingly, producing thus a group of
zoospores (brood formation). Such processes as these occur most
typically in the Mycetozoa (Myxomycetes) and also in such forms as
Arcella, Actinophrys, and some Foraminifera (Fig. 97).

Finally we come to a third general form of conjugation known as
karyogamy, which as the word indicates involves primarily a process of
nuclear fusion of the conjugating cells, although accompanied, of course,
by cytoplasmic fusion which may be of hardly secondary importance.
For instance, in some of the Infusoria the achromatic spindles fuse, as
well as the nuclei and undifferentiated parts of the cytoplasm; in some

FIG. 97. — Plastogamy in the Rhizopod, Arcella vulgaris. After Elpatiewsky.
A. Plastogamic union of about five individuals, apparently preparatory to the
formation of zoospores ("pseudopodiospores")- -B. Reproduction (formation
of " macroamoebaB ") following plastogamy. c, chromidia; n, nuclei; d, degen-
erating nuclei.

of the lower plants, even the plastids of the gametes, perhaps also their
centrosomes, fuse together during conjugation.

Most of the more familiar fertilization processes of the Protozoa are
essentially karyogamic, but, as a rule (as in the Metazoa), not all of the
nuclear substance of the cell is involved in the process. For usually,
as a preliminary to conjugation, the vegetative nucleus gives off, into the
cytoplasm, portions of its substance (chromidia) . These may be formed
as a result of general nuclear disintegration, or the nucleus may remain
quite intact and extrude chromidia, either directly through its mem-
brane, or by a process of nuclear budding. Some of these chromidia
are concerned in reproduction; such are termed idiochromidia. Karyo-
gamy, consequently, involves only a portion of the nuclear substance
ordinarily, and the remaining chromidia and vegetative nuclear struc-
tures may even break down and disappear during the process of ferti-
lization. Altogether these processes of chromidia formation are diverse

FERTILIZATION

191

and often very complicated and the details cannot be given here.
(Many of these details, and references to the literature of the subject,
are given by Calkins, " Protozoology," New York, 1909).

As a preliminary to the description of typical karyogamic union we
may refer to the very special form known as autogamy, which occurs
in many of the simplest Protista (e.g., Entamceba, Amoeba, some
Myxosporidia) . In autogamy there is really no fusion of cells at all;
the characteristic event is the separation of the nuclear chromatic

FIG. 98. — Autogamy in the Flagellate, Trichomastix lacertce. After Prowazek.
A. First nuclear division in the encysted form. B. The two nuclei completely
separated. C. First "reducing" division. D. Second "reducing" division.
E. Approach of the "reduced" nuclei. F. Fusion of the nuclei to form a single
nucleus (synkaryon).

substance of a single cell into a number of separate bodies (Figs. 98, 99),
which become scattered through the cytoplasm as chromidia, or rather
as idiochromidia, for they are concerned in reproduction. After their
formation is completed these idiochromidia fuse, by twos, or sometimes
in larger groups, forming in effect "new" nuclei, or at any rate new
combinations of chromatic substance. These fused chromatin masses
then commonly move to the surface of the cell and are budded off
with small bits of cytoplasm, as small cells or spores (Fig. 99) (Schau-
dinn, Calkins). Here then the nuclei which fuse together are the direct
derivatives of a single nucleus, and they remain within the same cyto-
plasmic mass throughout their formation and fusion. This might
readily be regarded as an extreme form of endogamy; it suggests the
roughly analogous process of the reentrance of the nucleus of one of
the polar bodies which occurs in a few of the parthenogenetic Arthropods.
Fertilization by autogamy is considered by some as a primitive method
of fertilization preceding all processes of gamete formation or cell fusions ;
others regard it as a derived condition in which the nuclei act preco-

192

GENERAL EMBRYOLOGY

ciously. However this may be it seems more instructive to classify the
process as karyogamic.

Coming now to the consideration of typical karyogamic fusion we
find that all of the fertilization processes common to the Metazoa, as
well as those of most of the Protozoa, belong here. And here again
fertilization may be either endogamous or exogamous.

Two general forms of karyogamy proper are usually distinguished,
although they are so clearly connected by transitional conditions that
they must be regarded as merely convenient groupings. These are
isogamy, where the pairing cells are similar in size, form and behavior;
and anisogamy, where the pairing cells are markedly dissimilar in size,

FIG. 99. — Autogamy in the Rhizopod, Entamceba histolytica. From Calkins,
"Protozoology," after Craig. A. Organism showing rods and granules of
chromatin in the nucleus, vacuole with some stained substance, and dense ecto-
plasm. B. Chromatin of the nucleus passing into the cytoplasm, as chromidia,
shown in C. D. Aggregation of chromidia to form secondary nuclei. E.
"Spore formation" by budding. F. Spores formed from buds.

form, and behavior. That is, this distinction is not based upon differ-
ences in nuclear structure, or behavior during union, indeed, these are
essentially the same throughout karyogamy, but upon external characters
of the conjugating cells.

Considering first isogamy, as the simpler and less modified process,
we find it restricted to the unicellular forms. Isogamic union frequently
occurs between two individuals of the usual vegetative type which do
not show, externally at least, any structural modifications usually
associated with gametic behavior (Fig. 100). This is most common
among the Flagellates, such as the familiar Copromonas, and Noctiluca,
but it occurs also in Actinophrys and in some species of Amceba. In other
cases the conjugating individuals show some modification in form as

FERTILIZATION

193

compared with vegetative individuals (they are modified similarly of
course), but there is no reduction in size. Thus in Cercomonas and
Tetramitus, the flagellate bodies of the conjugants become more or less
amoeboid and distinctly plastic and viscous. Other genera exhibit
various stages of reduction in size (Fig. 101), although in form they still
resemble vegetative cells. This is the case in most of the Foraminifera

FIG. 100. — Conjugation in the Flagellate, Copromonas subtilis. From Calkins,
"Protozoology," after Dobell. A. Vegetative form. B. Beginning of conjuga-
tion of two organisms; one flagellum withdrawn. C. Continued fusion. First
stage in nuclear "reduction." D. Second "reducing" division (heteropolar).
E. Conjugation completed; "reduced" nuclei fusing. F. Zygote within cyst.

and many other Rhizopoda ; it is rare among Flagellates (Stephanosphcera,
Chlamydomonas) and Ciliates. Finally, we find this reduction in size
accompanied by a modification in form, as in other Rhizopoda whose
gametes become flagellated, or where they are amoeboid in forms usually
flagellate or motionless.

We have then among isogamous organisms a series of forms, at one
extreme of which the gametes are morphologically unmodified, at the
other they are diminutive and structurally modified, usually in connec-
tion with the motor apparatus, in such a way as to render more likely
the accident of their meeting, likelihood of which is largely increased

194

GENERAL EMBRYOLOGY

through the fact that reduction in size is usually the result of multiple
fission or brood formation, which increases the number as well as the
activity of the gametes. In all of these forms of isogamy the union
of the gametes is permanent, the conjugants fusing completely and
thereby losing their identity as individuals. It need hardly be added
that in these cell conjugations the essential step seems to be the fusion
of the gametic nuclei into a single zygote nucleus.

FIG. 101. — Gamete formation and fusion (isogamy) in the Flagellate, Chlamy-
domonas steinii. After Goroschankin. A. Vegetative form, n, nucleus.
B. Group of gametes formed by multiple fission. C. Single gamete. D, E, F.
Stages in the fusion of gametes (isogametes). G. Zygote.

A special form of isogamy needs particular notice on account of its
frequency among the most familiar Protozoa— the Ciliata. This is a
temporary form of isogamy which involves, not the fusion of two gametes
to form a zygote, but the mutual fertilization of the two gametes through
the exchange of nuclear substance and perhaps also a small amount of
cytoplasm. The details of nuclear behavior in the conjugation of Para-
moecium, for example, are probably familiar but will bear brief restate-
ment here. This outline refers particularly to that form of Paramoecium
having only a single micronucleus ; one should recall that in these forms
which have both micronucleus and macronucleus, the former is, or
represents, the idiochromidia, the latter the vegetative nuclear struc-
tures. Two Paramcecia of normal vegetative size and external form
meet side by side, oral surfaces in contact, in a sort of plastogamic union,
The further course of events is exactly similar in each individual of

FERTILIZATION

195

the pair (Fig. 102). In each cell the micrenucleus divides and the
daughter micronuclei immediately divide again forming four. Of

FIG. 102. — Nuclear history during conjugation in Paramcecium putrinum.
After Doflein. The macro nuclear structures are omitted in all except A. A.
Formation of spindle for first division of micronucleus. B. Telophase of first
division. C. Second division of the micronucleus. D. Degeneration of three
of the micronuclei; the fourth, or permanent micronucleus, preparing for another
division. E. Division of the permanent micronucleus into the stationary and
migratory micronuclei. The spindle is greatly elongated and has a characteristic
mid-body. F. Grouping of micronuclei and exchange of migratory micronuclei.
G. Fertilization (mutual). Fusion of migratory with stationary micronuclei.
H. Formation of spindle for first division of the fusion nucleus in each conjugant.
/. Late phase of second division of fusion nucleus. J. Third division of fusion
nucleus. K. Exconjugant with eight micronuclei, derived from the fusion
nucleus. (The degenerating micronuclei are omitted from G-K.) d, degener-
ating micronuclei; ra, migratory micronucleus; ma, macronucleus; mi, micro-
nucleus; p, permanent micronucleus; s, stationary micronucleus; sp, spindle.

these, three degenerate (cf. maturation), while the one remaining
divides once more, this time unequally, forming a larger and a smaller

190 GENERAL EMBRYOLOGY

micronucleus in each organism. Each larger or "stationary" micro-
nucleus remains passive, but the smaller or " migratory" nucleus becomes
active and moves through the bridge of fused cytoplasms to the sta-
tionary micronucleus of the other individual, with which it fuses forming
a single compound or zygotic nucleus in each individual. The two
Paramcecia now separate each with a nucleus of modified composition.
The macronucleus, which has taken no share in the events of fertilization,
now fragments and dissolves leaving the fusion nucleus as the only
nuclear structure present. Then by three successive divisions the
fusion nucleus gives rise to eight small nuclei ; four of these in the pos-
terior end of the cell, remain small, as micronuclei, while the other four,
in the anterior end, enlarge, forming macronuclei. During the first
fission of this cell each daughter cell receives two nuclei of each kind,
and at the next division each of the four granddaughters of the " zygote "
receives one micronucleus and one macronucleus, and the normal
vegetative condition is restored. This form of karyogamy is peculiar
for at least three reasons; the nuclei alone fuse, both of the gametes
undergo nuclear reconstruction, and the individuality of the gametes is
not lost.

The sessile Ciliates show an adaptive modification of this process which
is anisogamic in character. In the common Vorticella, for example,
while one of the conjugants or gametes retains its normal vegetative
form, the other is small and one of a brood of four, which become free-
swimming. A small individual upon meeting a large one, is actually
absorbed by it. The early nuclear history of each organism is much the
same as in Paramcceium, save that the final micronucleus of the mega-
gamete is one of four, that of the microgamete one of eight, the remainder
in each gamete having degenerated. But after the equivalents of the
stationary and migratory micronuclei (idiochromidia) are formed
in each gamete, the process changes somewhat, for now one of the two
micronuclei (the equivalents of one stationary and one migratory body)
degenerates in each organism, while those remaining fuse together
forming thus only a single fusion nucleus in the single but duplex zygote.
Thus there is no mutual .fertilization, and while strictly this is anisog-
amous, it is mentioned here because it is clearly derived from the more
typical Ciliate condition as an adaptation to the sessile life of the vege-
tative form.

Coming now to anisogamous karyogamy (Fig. 103), we should note
that transitional conditions between isogamy and anisogamy are not
infrequent. Thus in the Flagellate, Bodo, the conjugants may be
either of equal or unequal size, apparently in an accidental fashion.
The colonial Pandorina forms gametes of three sizes, small, medium,
and large, and conjugation may occur between any smaller and any
larger individuals anisogamically, or the small or the medium organisms

FERTILIZATION

197

may conjugate together isogamically. Here then anisogamy is not
obligatory, but facultative or accidental. In this case exogamy is the
rule since a single colony forms gametes of one size only, though in
isogamy endogamy may occur.

—^ H

FIG. 103. — Formation of gametes and syngamy in the Sporozoan, Klossia
octopiana. From Calkins, "Protozoa," after Siedlecki. The chromatin of the
nucleus is distributed throughout the cell, A, B, finally forming nuclei of the
future gametes, C, D, E. The mature gametes, s, swim about, and join a macro-
gamete, F. The nuclei mingle, G, and then the cleavage nucleus divides repeat-
edly by mitosis, to form the spores, H. cT, microgametic nucleus.

In true anisogamy conjugation is practically always exogamous for
as a rule a single organism forms gametes of only one size at a time.
The essential difference between the gametes is probably that of behav-
ior, i.e., degree of activity, associated with which are constant differences
in size. The larger gametes or megagametes, are less numerous and less

198

GENERAL EMBRYOLOGY

active than the smaller microgametes, which may be formed in very
considerable numbers. Occasionally the gametes differ in form only,
as in Dallingeria, where one of the gametes has three flagella, like the
vegetative cells, while the other gamete has but one flagellum. In a

FIG. 104. — Micro- and macrogametes of various Protozoa, illustrating various
degrees of differentiation. After Doflein, from various authors. /, F, x 562,
others X 1125. a, A. Urospora lagidis (Brasil). 6, B. Collozoum inerme
(Brandt), c, C. Chlamydomonas braunii (Goroschankin) . d, D. Volvox aureus
(Klein), e, E. Cyclosporia caryolytica (with two polar bodies in cytoplasm of E)
(Schaudinn). /, F. Orcheobius herpobdella (Kunze).

few forms such as Monas and many of the Gregarines, the only morpho-
logical difference between the gametes is that of size, the microgamete
being smaller and somewhat the more active, but not otherwise unlike

FERTILIZATION 199

the megagamete. In other forms the microgametes are considerably
modified structurally, usually in connection with an increase in loco-
motor activity. At the same time the megagamete may increase con-
siderably beyond the ordinary vegetative size and may then lose motility
more or less completely. So it finally comes about that gametes of two
wholly different types are formed, both quite unlike vegetative cells,
and the typical Metazoan condition is reached.

Several groups of Protozoa, e.g., Gregarines, colonial Flagellates,
afford interesting series showing stages in this differentiation of the
gametes (Fig. 104). We may outline one such series selecting examples
from the Volvocine group of Flagellates.

In Stephanosphcera all the individuals of the colony are, or may be,
reproductive, and conjugation is isogamous and endogamous. There
is no differentiation of gametes. In Pandorina (Fig. 8) all of the indi-
viduals may be reproductive, but some of the gametes may be differ-
entiated in size, and conjugation may be either isogamous or anisogamous
as described above. Here dissimilarity of the gametes is facultative.
In Eudorina two kinds of colonies are found. In one, all the cells may
become reproductive, the individuals forming megagametes only slightly
larger than vegetative cells ; in another only four cells of the colony are
reproductive and each of these forms sixty-four very small and active
microgametes. Fertilization is here strictly anisogamous and exo-
gamous. The last step is represented by Volvox (Fig. 10), where the
number of gamete forming cells is always limited. Here too differentia-
tion of the gametes reaches its climax among the Protozoa, and the
Metazoan condition is reached. The reproductive cells lose their motor
organs and begin to enlarge. A few of them grow to a relatively
enormous size and become the passive megagametes. The others
grow to lesser extent and then divide rapidly, each forming, probably
?28 microgametes. These are very small flagellated, extremely active
cells, with an elongated rostrum or penetrating organ at one end (Fig.
104, d). The microgametes are liberated in large numbers and swim
about until one reaches a megagamete which it then enters and their
nuclei fuse forming a typical zygote which then reproduces a new
colony. The resemblance to the gametes of the Metazoa is so com-
plete that they are here termed the oosphere, or ovum, or oogamete
(megagamete) and the spermatozoon or spermagamete (microgamete) .

It should perhaps be noted here that the process of conjugation or
fertilization is not always associated with the reproduction of the
Protozoa mentioned above, not even in the colonial forms. For the
usual reproductive processes are carried out by the simple fission of
ordinary vegetative cells. In the simpler colonial forms, such as
Pandorina and Eudorina, as many new colonies may be formed as there
are individuals forming the original colony, in Volvox, however, the

200 GENERAL EMBRYOLOGY

number of cells which may reproduce in this way is limited, and there
seems to be a real distinction between soma and germ, much like that
of the Metazoan organism, the mother cell which divides to form the
multiple spermatozooids, has even been compared with the testis of a
Metazoan.

The principal forms of fertilization and gamete formation are sum-
marized in the accompanying table.

Several facts of prime importance are to be drawn from this account,
(a) Among the Protozoa as well as the Metazoa, the process of fertiliza-
tion is widespread. (6) Out of a variety of forms comes that form of
fertilization characteristic of the Metazoa, namely, karyogamy. (c)
Accompanying this karyogamy is a gradual and finally complete
differentiation of gametes, which differ morphologically and physio-
logically, both from vegetative cells and from each other, (d) The
Metazoa show much less diversity than the Protozoa respecting the
process of fertilization and the form of the gametes.

Such a series of stages as that outlined above, of the gradual differ-
entiation and specialization of gametes, cannot fail to suggest the
general subject of sex. It does indeed indicate the nature of the
original distinction between the sexes. Among the Metazoa the pri-
mary and familiar facts upon which the definition of sex is based, are,
that spermatozoa-producing individuals are males, - ova-producers are
females. In all cases of isogamic conjugation no distinction between
the gametes, and therefore between sexes, obtains. In those instances
transitional between isogamy and anisogamy, we may see the begin-
nings of sex distinction, often facultative. True anisogamy involves true
sex distinction; at first relatively slight (Pandorina), in such forms as
Volvox or Coccidium and many other Sporozoa, the fundamental
differentiation of sex seems to be completely established, i.e., the
gametes are markedly unlike and conjugation occurs only between two
dissimilar cells.

The essential processes of fertilization are entirely equivalent in
isogamy and anisogamy, so that the fundamental distinction of sex is
based only upon the external form and behavior of the gametes, not
upon any differences in the nature of the conjugation processes in sexual
and non-sexual forms, for none exist.

Most of the colonial Protozoa are monoecious or hermaphroditic,
producing gametes of both kinds, but cross-fertilization (exogamy) is
the rule here, as it is among the hermaphroditic Metazoa, and frequently
for the same reason in both groups, namely, a difference in the times of
ripening of the two forms of gametes of a single colony or individual.
Many of the Sporozoa are clearly dioecious or unisexual, and in some of
these there are also secondary sexual characters, usually size differences,
such that macrogamete-forming or female individuals can be distin-

FERTILIZATION

201

202 GENERAL EMBRYOLOGY

guished from microgamete-f orming or male individuals, some time before
the gametes are actually formed ; in a few rare instances this distinction
can be made throughout the life cycle, and individuals can be identified
at any time as males or females.

We now come to a consideration of the meaning and theo-
retical significance of these processes of fertilization or syngamy.
Probably there is, in the whole field of Biology, no process of
such widespread occurrence and obvious importance, where
the phenomena are so well known, which at the same time is so
little understood. Why fertilization should occur, what is
effected by it, and how syngamy brings about the results which
do follow it, are questions to which to-day, after decades of
speculation and research, no sure answers can be given.
v Although we may be on uncertain ground, it will be profitable
to review some of the suggestions and hypotheses that have
been proposed in this connection, even if we accomplish little
more than to point out the possibilities and difficulties of this
fascinating subject. And furthermore, while no thoroughly
demonstrated solutions of the problems of fertilization have
been reached, there are several carefully worked out hypotheses
in the field, some of which are certainly to be regarded as close
approximations toward the correct explanation of some of the
problems of fertilization. We may conveniently arrange these
current ideas as to the results and primary " purpose" of fertili-
zation in four groups. The results of fertilization may be
connected with (a) reproduction, (b) rejuvenation, (c) the
process of variation, (d) the process of heredity. In considering
each of these we shall state as briefly as possible the essentials of
the evidence for and against the central idea.

That fertilization is primarily a reproductive process was the
original view held by Harvey and his successors. There are
now two forms of the hypothesis. In one form we find the
idea that the ovum quite obviously seems to contain only a
part (i.e., the cytoplasm and one-half a nucleus) of the mechan-
ism necessary for development, and that the spermatozoon
brings into the ovum those parts (i.e., one-half a nucleus and
the centrosome, or the stimulus to its formation) which com-

FERTILIZATION 203

plete this mechanism and enable development to proceed. In
its other form this idea is that the ovum is a quiescent, passive
body which needs to be stimulated to its normal activity
(development) by the entrance of the spermatozoon, the kino-
plasm (centrosome) of which is the part chiefly acting as the
stimulus. In short, fertilization is to be regarded normally as
the necessary antecedent, as the cause of development.

There are many facts opposed to this view. Of these we shall
discuss two chief classes, first, those of parthenogenesis, both
normal and "artificial," and second, those drawn from the
relation between fertilization and reproduction among the
Protozoa.

For present purposes we may extend the definition of partheno-
genesis to include all those cases where single cells, specialized
for the purpose, develop without undergoing syngamy. In
the plant kingdom parthenogenesis, in this broad sense, is very
widespread. Development from spores is very common, even
among the higher (vascular) plants, and in some instances
(some of the Fungi) reproduction by single unfertilized cells is
the exclusive method. And the development of ova, typical
in every respect save that of needing to be fertilized, is not
uncommon. Among the single-celled animals phenomena
equivalent to development from spores are frequent, and among
the multicellular animals normal parthenogenesis is known in
the Rotifera, some of the Crustacea, and in several orders of
Insecta. In most of these forms fertilization does occur at
some period in the life cycle, after a widely variable number
of parthenogenetically produced generations, but there are a
few Metazoa, e.g., the wasp, Rhoditis, and the Crustacea, Cypris,
Limnadia, and sometimes Apus, in which males never develop
and fertilization is therefore entirely unknown (Weismann).
In some of the parthenogenetic Crustacea the nucleus of one
of the polar bodies seems to act as a fertilizing nucleus (see
Chapter IV), so that a sort of autogamic process occurs,
recalling that of some of the Protozoa and perhaps analogous
with it. In all of these Metazoa the form and history of the
parthenogenetic ova, the occasional presence of vestigial

204 GENERAL EMBRYOLOGY

spermathecse, and other similar conditions, clearly indicate that
this is a secondary or derived condition, a special adaptation,
which therefore throws but little light upon the fundamental
significance of the fertilization process; this proves only that
fertilization is not in all cases a necessary antecedent to develop-
ment, and that the ova may, in certain cases, contain in them-
selves complete developmental mechanisms, and need neither
the addition of sperm structures nor the special form of stimu-
lation afforded by the entrance of the sperm cell.

Instances of merogony, or "male parthenogenesis," where
egg fragments containing no nuclear substance develop after
penetration by a spermatozoon (Echinoderms), also show
that the addition of egg and sperm structures, each incom-
plete in itself, is not a necessary feature of fertilization.

It may truly be said that the obviously secondary character
of normal parthenogenesis renders the phenomenon of little
value as evidence regarding the real meaning of fertiliza-
tion in the vast majority of instances. But such an objec-
tion cannot be brought against the evidence from experimental
parthenogenesis, and probably the clearest evidence upon this
phase of the subject is that of "artificial" or induced partheno-
genesis. In view of this fact, and of the great general impor-
tance of the subject, we may consider this matter rather fully.

The eggs of many animals, belonging to many different
classes and phyla, normally requiring to be fertilized, may be
stimulated to begin their development by chemical or physical
treatment. Thus the ova of many Ccelenterates, Echinoderms,
Annulates, Molluscs, and even some Chordata (Teleosts,
Amphibia), may be induced to commence their development
parthenogenetically by being subjected to the action of a
great variety of organic and inorganic substances in solution,
to unusually high or low temperatures, to physical shock, or to
various other conditions.

In this process of artificial parthenogenesis two phases
must be kept separate, first, the phenomenon of maturation, in
those cases where this is not completed at the time fertilization
normally occurs; and second, the phenomena of cleavage and

FERTILIZATION 205

differentiation, occurring subsequently to maturation. Cer-
tain acids and some other substances seem to have, accord-
ing to Morse, a specific effect in bringing about maturation,
either not producing cleavage or actually inhibiting it, in
which latter case it may then be induced by other treatment.
The precise actions upon the egg of those chemicals inaugu-
rating cleavage are varied, but for the most part they appear
to effect certain changes in the egg which are similar in nearly
every instance. For example, according to Loeb, who is the
pioneer in this important work, in the sea-urchin, the best
result, that is, the closest imitation of natural fertilization, is
secured by treating the eggs, first with a solution of a monobasic
fatty acid, such as butyric acid, for one or two minutes. In
many cases the butyric acid can be replaced by an alkaline
solution of equivalent strength or by a solution of almost any
fat solvent. This treatment results in the formation of an
apparently typical fertilization membrane. Then second, the
eggs are treated for some minutes or hours (sea-urchin eggs,
thirty to fifty minutes at 15° C.) with a hypertonic sea water,
that is, sea water whose osmotic pressure has been raised about
50 per cent, above normal by the addition of salts, such as
sodium chloride. Finally the eggs are returned to normal sea
water and cleavage then follows in quite the usual fashion.

What is actually accomplished within the egg by such treat-
ment as this is largely conjectural. Loeb suggests that the
process may be as follows. The mature ovum is surrounded by
a relatively impermeable surface film which prevents the
oxidations necessary to development. The butyric acid or
similarly acting substance, by dissolving certain fatty constitu-
ents near the egg surface, frees from association with these,
certain other osmotically active materials which then form the
permeable fertilization membrane, and thus rapid oxidations
are permitted. Loeb believes that the nuclear substance
possesses a catalyzer which, in the presence of oxygen, brings
about a synthesis of nuclein, one of the chief constituents of
chromatin, and that this synthesis of nuclein is the chief
chemical action of the segmenting ovum. This process of

200 GENERAL EMBRYOLOGY

cleavage once started in the right direction, leads then in a
perfectly natural and normal way to the later processes of de-
velopment. Loeb suggests farther that the spermatozoon may
contain certain substances, enzymes, which form within the
egg, materials capable of producing effects similar to these,
and that herein lies the natural stimulating effect of the
spermatozoon.

Many facts regarding this hypothesis, both pro and con, have
been forthcoming in recent years, but it is still too early to say
how closely this approximates the truth. We might add, how-
ever, that Masing and others have not been able to detect any
marked synthesis of nuclein such a Loeb describes during
cleavage. And quite recently Conklin has determined that the
synthesis of nuclear substance is, in Crepidula, at least no
greater than the synthesis of cytoplasm.

The eggs of other forms are more successfully treated by other
methods; and each may have a particular treatment which is
most effective. But in very many of the instances of artificial
parthenogenesis the essential result of the treatment seems to be
a process of membrane formation accompanied by the with-
drawal of fluids from the egg. This has led to the suggestion
(Loeb) that the spermatozoon acts in this fashion, for it is
relatively very deficient in fluids, and upon entering the egg
reduces to some extent the relative fluidity of its cytoplasm,
thus acting as a stimulus.

That the action of the spermatozoon is not specific, and that
fusion of the two germ nuclei is really not necessary to inaugu-
rate development, is clearly shown by the fact that almost any
spermatozoon, of whatever species, that can gain entrance to
an ovum, is capable of initiating development, and of effecting
the apparently normal cleavage of the ovum; to what extent
the internal processes of fertilization and cleavage are entirely
normal in such a case, we shall see later; suffice it to say here
that frequently a foreign sperm nucleus remains quiescent and
takes no part in the formation of the mitotic figure.

It is true that the development of artificially fertilized ova
seldom proceeds farther than the cleavage stages. As a rule

FERTILIZATION 207

the lower the organism in the evolutionary series, the farther
its development may proceed. And while some artificially
fertilized Echinoderm eggs have been carried past the larval
stage (Delage), the Chordate ovum (Teleost, Cyclostome,
Urodele) will cleave only a few times. This indicates clearly
that the parallelism between natural and artificial fertilization
is not complete, although it is not unlikely that ultimately a
form of treatment may be found which will produce just the
same result as normal fertilization, save in so far as this is con-
cerned in the inheritance of individual characteristics.
In all cases of artificial parthenogenesis the cleavage figures

o

are essentially normal except that the reduced or • number

of chromosomes is present (exceptions to this have been
reported by Tennent and Hogue, and others) ; the poles of the
spindle are occupied by typical centrosomes, formed anew by
the substance of the egg after the disappearance of the oocyte
centrosomes of the maturation spindle. This is also true re-
garding the centrosomes of normally parthenogenetic eggs.
And there are several instances known where the specific effect
of certain reagents or external conditions is the formation, out
of the cytoplasm of the ovum, of numerous centrosomes,
apparently of normal structure and each with a small aster
(Yatsu).

To summarize the evidence from parthenogenesis, both nor-
mal and artificial, we may say that, among the Metazoa, the
ovum contains within itself a mechanism sufficiently complete
to function for a time at least, although the spermatozoon,
when it enters, does add to this mechanism and supplies some
parts not present in the egg; these parts either are not abso-
lutely necessary or they may, under certain conditions, be
supplied from the structure of the ovum in the absence of the
sperm. And further, while the egg may be stimulated to
develop by means other than the entrance of the sperm, this
is normally the form of stimulus which inaugurates the series
of reactions we call development. Taking this view of fertiliza-
tion the formation of the spermatozoon is a means of insur-

208 GENERAL EMBRYOLOGY

ing the properly effective form of stimulus, which might other-
wise be lacking in the environment of the egg.

Turning now to the evidence which the Protozoa offer
regarding the relation of fertilization and reproduction, we may
approach more closely the problem of the fundamental signifi-
cance of syngamy. Nearly all the known life histories of Pro-
tozoa are cyclic in character. The process of reproduction by
simple fission may proceed uninterruptedly for a longer or
shorter period, but finally this is interrupted by some form of
syngamy. Formerly this seemed obviously to mean that the
ordinary reproductive processes depended ultimately upon a
process of conjugation or fertilization, and that the life cycle in
the Protozoa was essentially similar to that of the Metazoa, the
divisions of the somatic cells of the latter being equivalent to the
simple fissions of the former, and the fertilization processes of
both being essentially similar (homologous). It was overlooked
at first that the process of fertilization might just as well be
considered the result of vegetative divisions as the cause of
them; to this phase of the relation we shall return shortly.

It is true that in some Protozoa, e.g., Noctiluca, Trichosphcer-
ium, some Gregarines, fertilization is really followed by a
marked increase in reproductive activity. And it is often true
that multiple fission tends to follow conjugation. But in
other cases reproductive activity seems not to be affected by
fertilization. And in many, probably most Protozoa, fertiliza-
tion tends to inhibit reproduction. In many Rhizopods,
Flagellates, and Ciliates, a pause in the succession of fissions
may be quite marked after conjugation. In many of the Sporo-
zoa a period of encystment follows, and the same is true of
many Algae. In such cases, therefore, fertilization seems
opposed to reproduction, or at least to any immediately ensuing
processes of multiplication; it may still be true that the ultimate
effect of fertilization may be increased rate or duration of fission.
Conjugation may occur without reproduction; reproduction
may occur without conjugation. And that conjugation is
frequently to be regarded as determined by external rather than
internal conditions is indicated by the occurrence of so-called

FERTILIZATION 209

"epidemics" of conjugation which may often be observed in
Protozoan cultures. In such cases conjugation may often be
artificially induced by regulating the character and amount of
the food supply (Jennings).

It may be concluded, therefore, that among the Protozoa the
processes of reproduction and fertilization are not funda-
mentally related, and the primary significance of fertilization
must be sought in some other relation. This view is widely
accepted to-day and it consequently becomes necessary to
explain the practically universal association of the two proc-
esses among the Metazoa, the only exceptions being, as we
have seen above, the secondary and obviously derived instances
of normal parthenogenesis. The commonly suggested explana-
tion is the following.

Whatever the real significance of fertilization may be, it
seems, for reasons which will appear later, a condition for
continued existence of specific forms of protoplasm that
occasionally some disturbance of its inner structure should
occur, such as would result from the mingling of the substances
of two distinct individuals. Among the single-celled organisms
this may occur at any time, whenever that action would form a
natural response to internal conditions of the organisms.
Among the Metazoa, on the other hand, such a complete fusion
of cells, and particularly of nuclei, can occur only when the
organisms are in the form of single cells, i.e., gametes, and
differentiation of the organism is at a minimum. Thus whereas
the two processes are originally distinct and unrelated in their
origin in the Metazoa, they have come to be related, and now
fertilization appears as the first step in reproduction.

Another general hypothesis regarding the function and signifi-
cance of fertilization is the rejuvenation hypothesis, associated
chiefly with the names of Btitschli, Maupas, and Richard Hert-
wig. This is based to a large extent upon the phenomena of
the Protozoan life cycle. It involves as a starting point the
assumption, partly based upon observation, that protoplasmic
activity tends gradually to diminish in intensity, and that
associated with this diminution are certain morphological altera-

210 GENERAL EMBRYOLOGY

tions in the structure and composition of the cell. . Altogether
these modifications are known as senescence. A frequent
characteristic of a senescent cell, in both Protozoa and Metazoa,
is the relatively large proportion of cytoplasm as compared
with nuclear substance. It is further assumed that syngamy
and the consequent admixture of nuclear and cytoplasmic
materials of two individuals, perhaps representing different
races, causes the restoration of the senescent protoplasm to a
condition of vigor, in a word, brings about rejuvenation. It
would follow from this, that protoplasmic activity is cyclic,
and that periods of senescence would be followed by death un-
less fertilization, or an equivalent process, should occur.

Precisely what is involved in the process of rejuvenation
cannot be stated definitely. Richard Hertwig suggests that
senescence is due chiefly to changed nuclear-cytoplasmic rela-
tions resulting from repeated cell-fissions, and that in rejuve-
nation there is essentially a restoration of the normal nuclear-
cytoplasmic ratio, as well as a certain chemical and physical
reorganization of the protoplasm through the combination of
materials from two more or less unlike individuals. Loeb and
others, as we have seen, also regard the rapid synthesis of
nuclein as the most important consequence of fertilization.
Still others (Minot, Bernstein) believe that rejuvenation is not
only a nuclear-cytoplasmic phenomenon involving or resulting
from an increase in the relative amount of nuclear substance,
but that it further includes an increase in the property of growth,
i.e., the formation of new protoplasm, both nuclear and cyto-
plasmic, out of non-living substance.

The real evidence for the cyclic character of the life processes
of the Protozoa is chiefly that of Maupas and Calkins, who
showed that in Paramcecium and some other Ciliates, when
conjugation is prevented, there occur, under laboratory condi-
tions, periods of depression in vital activity, accompanied by
changes in structure, i.e., periods of senescence. This depres-
sion leads finally to death unless conjugation is permitted, or
unless the organisms are subjected to some form of stimulus.
If stimulated by chemical or physical means, or naturally

FERTILIZATION 211

through conjugation, the organisms may in some cases recover
their original vigor and begin a new cycle with youth renewed.

But rejuvenation is by no means always the result of conjuga-
tion, for frequently the senescent organisms perish in spite of
conjugation; and it may even be the case that the descendants
of cells which have conjugated before the signs of senescence
have appeared, perish sooner than their immediate relatives
of approximately the same age, which have been prevented from
conjugating. Moreover, Jennings has shown that in certain
races of Paramoecium aurelia-caudatum conjugation may occur
at intervals of only one or two weeks, while in other races of
the same species conjugation occurs only at intervals of a year
or longer, and in still a third race no conjugation was observed
during a period of three years, although during this time
observation was not so continuous as to preclude the possibility
of conjugation having occurred.

Valuable evidence upon the question of the cyclic character
of the Protozoan life history is afforded by the work of Woodruff,
who has shown that if more natural conditions are substituted
for the artificial and more uniform conditions of the laboratory,
no cyclic relation appears, in some strains of Paramcecium at
least. By continually altering the character of the food, and
by imitating in other ways the naturally variable conditions
of pond life, he has been able to continue a single race of
Paramoecium for over five years. During this period more
than 3000 generations were formed by simple fission, and in
all this time conjugation did not occur, and no periods of
depression or signs of structural modification could be ob-
served.1 Finally, Woodruff has been able to carry a culture of
Paramoecium on a uniform diet of beef extract, which is sup-
posed to contain all of the materials necessary for their life,
for ten months (about 450 generations) without any indication
of senescence.

Such facts as the foregoing show, first, that protoplasmic
activity among the Ciliates may not be cyclic in character under

1 On Sept. 27th, 1912, Professor Woodruff writes that this culture is
in its 3265th generation, and still normal.

212 GENERAL EMBRYOLOGY

certain conditions, and second, that when cyclic periods of
protoplasmic depression do occur the protoplasm may be
restored to a condition of normal vigor, either by physical or
chemical stimuli, or by fertilization. Supposing, and the sup-
position is highly probable though not completely demon-
strated, as a fact, that the living processes do tend, in the absence
of continued stimulation, to diminish in intensity or otherwise
to deviate from the normal, then we find in the process of fer-
tilization a natural means of insuring the receipt of stimuli
which might otherwise be lacking. The onset of those struc-
tural and physiological modifications called senescence, leads
to a modification in the behavior of the organisms, i.e., they
form gametes and conjugate.

This becomes clearer when we recall that life itself is re-
sponse— reaction to the stimuli resulting from changed relations.
Such a changed relation may result (a) from changes in both the
environment and the cell or organism, or (b) from changes in
the environment alone while conditions within the organism
remain comparatively uniform, or (c) from changes within the
organism while the external conditions remain comparatively
uniform. Of these three possibilities the first two are certainly
the most frequent in the lives of most free-living Protozoa.
But we may interpret fertilization as fundamentally a means
of ensuring a changed relation through the realization of the
third possibility in the absence of the other two. In a way the
Ciliates act so as to ensure automatically a changed relation
between organism and environment; when external conditions
become too uniform to bring forth the normal vegetative ac-
tivities, the form of reaction actually changes and is modified
into gamete formation and fertilization, which immediately
leads to an internal disturbance and the condition of uniformity
is corrected, whenever it may occur.

Among the Protozoa we find this division of labor between
vegetative and conjugative or fertilizing cells occurring when-
ever internal-external relations demand it. Among the Meta-
zoa, however, such a division of labor must occur at a certain
period in the life history, on account of the impossibility of the

FERTILIZATION 213

complete fusion of two whole organisms at any time other
than when they are in the form of single cells; consequently we
find vegetative and gamete-forming tissues differentiated side
by side, and since these are, as components of a single organism,
in a fairly constant environment removed from continuously
rejuvenating stimuli, it is the function of the gamete-forming
tissues to form single cells which can fuse with cells of other
individuals and thus, by altering the composition of the organ-
ism, alter its relation to external conditions. And the almost
universal association of reproduction and fertilization consid-
ered as a rejuvenative process among the Metazoa may have an
added significance; the two occur together, not because they
are directly related to one another, but because they are both
occasioned by the same condition, or rather by the same limi-
tation of opportunity, for the complete fusion of multicellular
organisms can occur only when these are in the form of single
cells, or gametes.

Reference to the third possible significance of fertilization
may be more brief because of its extremely hypothetical char-
acter. The idea that the process of fertilization is primarily
related to the phenomena of variation is associated chiefly with
the names of Weismann and Oscar Hertwig. The scanty and
uncertain character of the evidence here is indicated by the
fact that there are two exactly opposed views as to the nature
of the relation. Hertwig maintains that the effect of fertiliza-
tion is to limit variation within a species, by tending to bring
back to the normal, through the process of heredity, the progeny
of extreme fluctuations and unusual or abnormal variations
(mutations), because the likelihood of their mating with the
much more common mediocre or average individuals is so
much greater than that of their mating with their likes. Weis-
mann, on the other hand, maintains that the effect of syngamy
or " amphimixis" is to cause or promote variation, which would
result from the new organic combinations in the continued
admixture of the gametic nuclei of different individuals. In
this way the process of fertilization becomes of great evolu-
tionary significance in that it accounts, in part at any rate, for

214 GENERAL EMBRYOLOGY

the presence, indeed the origin, of variations and fluctuations,
the "raw materials" of evolution.

Both of these views are based upon the more fundamental
and underlying hypothesis of the representative particle nature
of the elements of the chromosomes, or perhaps of other portions
of the germ cells, which themselves vary in their structure or
their combinations. Here again the occurrence, among the
Metazoa, of fertilization only when the organisms are in the
form of single cells, grows out of the fact that complete nuclear
fusion can occur only when in this state.

There is little direct factual evidence for or against these
views, either one of which can be maintained upon theoretical
grounds. In a few cases it is known that the amount of
variability is not significantly different among sexually (gamet-
ically) and asexually (parthenogenetically) produced indi-
viduals of the same species. And from the standpoint of more
recent studies upon heredity and variation the evidence is
chiefly either negative or opposed to the idea that this relation
constitutes an important element in the origin or present func-
tion of fertilization. The present aspects of this relation
between fertilization and variation merge in the larger question
of the relation with heredity which we may refer to next.

Whatever the significance of fertilization may prove to have
been originally, its relation to the phenomena of heredity is
to-day undoubtedly its most important aspect, at any rate
among the Metazoa. The general subject of the relation of the
structure of the germ cells to the main facts of heredity is
reserved for consideration in Chapter VII, but we should point
out here some of the underlying conditions involved in the
fundamental fact of the union of the two germ cells derived in
nearly all cases from two different individuals of the same group
or species.

As pointed out in the introductory chapter, the germ cells
are not to be regarded as the material links between successive
generations of specific organisms, for organismal specificity is
not discontinuous, but continuous, and the germ cells are no
less specific, no less the organism, than are the mature indi-

FERTILIZATION 215

viduals producing the germ cells or produced by them. From
the standpoint of the fertilization process and of heredity, the
essential fact is not that the zygote develops into an individual
of the same species to which belonged the organisms producing
the gametes, for in parthenogenesis, for example, specific organ-
isms are produced in the absence of fertilization. The signi-
ficant fact here is that offspring may possess some of those
characteristics which are the individual possessions of either
of the parents. On the whole, offspring inherit, or may inherit,
equally from both parents, and such a possibility must depend
upon the fact that the zygote is composed of substances or
structures derived from both the parent organisms.

Of course the only substance of the zygote which is derived
in equal or approximately equal parts from the two parents is
the chromatic portion of its nucleus, and it is frequently said
that therefore it must be the nuclear structures of the germ cells
which are involved in this fact of equal biparental inheritance.
And yet the fact should not be disregarded that the sperm does
bring into the egg a certain though indeed a small amount of
cytoplasm. The fact that the individual parental characters
are inherited equally does not necessarily mean that all non-
individual characteristics are thus inherited, for all of the
more general species characters are common to both parents,
and the offspring might conceivably inherit these wholly from
either parent. From this point of view the contribution by the
ovum of practically the whole of the cytoplasm of the zygote
might have at least two meanings. It might mean that the
general species characters of the offspring are determined by
the structure of the cytoplasm, and only the individual traits
by the nuclear structures; it would follow from this that the
spermatozoon takes a relatively subsidiary part in heredity.
Or it might mean that the two gametic nuclei are from the
beginning equally involved in the determination or direction of
development, while the cytoplasm of the ovum merely affords
the great bulk of the material basis for this development, and
is itself in no wise involved in the qualitative determination of
either specific or individual characters of the offspring. It

216 GENERAL EMBRYOLOGY

would follow from this that the two germ cells are of more
nearly equal importance in the process of heredity.

We may finally conclude from all of the foregoing discussion
that little can be very definitely asserted regarding the real
function of fertilization now, and still less regarding the
original significance of the process. Some of the questions
involved here are to-day the most interesting and important of
the unsolved problems in the fields of Embryology and Biology.

It is reasonably clear that fertilization is not, at the present
time, a simple process, although it may have been so originally.
Doubtless there has been an evolution both of the process and
of the consequences of fertilization, just as there has been of all
organ structure and organ physiology. Furthermore, it
seems clear that the various possibilities described above, as to
the significance of fertilization are not mutually exclusive:
fertilization may be important for several of these reasons, even
in a single case, and probably it has no one meaning that is
exclusively true. It is quite possible that normally, among
the Metazoa to-day, the spermatozoon may bring about in the
ovum the formation of centrosomes which do, as a matter of
fact, take an important part in the succeeding cleavages of
the zygote, it may also chemically and physically stimulate
the ovum to develop by bringing about initial changes in its
chemical or physical structure or organization, it may at the
same time introduce substances, the effect of which is " rejuve-
nation " of the specific protoplasm apart from the reproductive
phenomena, and finally the structure of the spermatozoon which
does these things may also affect the course of development so
that individual characteristics of the male parent, as well as
of the female parent, may appear. And to say that the result
of fertilization is, for example, rejuvenation, need not mean
that it is not also a stimulus to reproduction, a controlling factor
in variation, and a means of heredity.

REFERENCES TO LITERATURE

ADOLPHI, H., Ueber das Verhalten von Wirbeltierspermatozoen in

stromenden Fliissigkeiten. Anat. Anz. 28. 1906.
VAN BENEDEN, E., (Ref. in Ch. II.)

FERTILIZATION 217

BOVERI, T., Zellenstudien. I, II, III. (Ref. in Ch. II, III.) Zellen-

studien. IV. Ueber die Natur der Centrosomen. Jena. Zeit.

36 (28). 1901. Die Polaritat von Ovocyte, Ei und Larve des

Strongylocentrotuslividus. Zool. Jahrb. 14. 1901. Das Problem

der Befruchtung. Jena. 1902.
BULLER, A. H. R., Is Chemotaxis a Factor in the Fertilisation of the

Eggs of Animals? Q. J. M. S. 46. 1902.

CALKINS, G. N., Studies of the Life-history of Protozoa. I. The Life-
cycle of Paramcecium caudatum. Arch. Entw.-Mech. 16. 1902.

IV. Death of the A Series. Conclusions. Jour. Exp. Zool. 1.

1904. (See also, ref. Ch. I, II.)
COB, W. R., (See ref. Ch. IV.)
CONKLIN, E. G., (See ref. Ch. II, III.)
DELAGE, Y.,^ Etudes sur la Merogonie. Arch. Zool. Exp. (Ill), 7.

1899. Elevage des larves parthenogenetiques d' Asterias glacialis.

Arch. Zool. Exp. (IV), 2. 1904.
DOBELL, C. C., (Ref. in Ch. II.)
DOFLEIN, F., (Ref. in Ch. I.)
DRAGO, U., Nuove ricerche sull* "attrazione" delle cellule sessuali.

Arch. Entw.-Mech. 26. 1908.
ELPATIEWSKY, W., Zur Fortpflanzung von Arcella vulgaris, Ehrb.

Arch. Protist. 10. 1907.
GODLEWSKI, E., Untersuchungen iiber die Bastardierung der Echiniden-

und Crinoidenfamilie. Arch. Entw.-Mech. 20. 1906.
HARTMANN, M., (Ref. in Ch. II.)
HARVEY, E. N., Methods of Artificial Parthenogenesis. Biol . Bull. 18.

1910.
HERTWIG, O., Beitrage zur Kenntniss der Bildung, Befruchtung und

Theilung des thierischen Eis. Morph. Jahrb. 1-4. 1875-1878.

(See also ref. Ch. III.)
HERTWIG, R., Ueber Kerntheilung, Richtungskorperbildung und

Befruchtung von Actinosphoerium Eichhorni. Abh . Acad . Munch en

(II Cl.) 19. 1899. Was veranlasst die Befruchtung der Protozoen?

Sitz.-Ber. Ges. Morph. Phys. Mtinchen. 15. 1899. Eireife und

Befruchtung. Hertwig's Handbuch, etc. I, 1, 1. 1903. (1906).

Ueber den Chromidialapparat und der Dualismus der Kernsub-

stanzen. Sitz.-Ber. Ges. Naturf. Miinchen. 1907.
JENNINGS, H. S., What Conditions induce Conjugation in Paramecium?

Jour. Exp. Zool. 9. 1910.

KORSCHELT UND HEiDER, Lehrb., etc. (See ref. Ch. III.)
KOSTANECKI, K., Ueber parthenogenetische Entwicklung der Eier von

Mactra mit vorausgegangener oder unterbleibener Ausstossung der

Richtungskorper. Arch. mikr. Anat. 78, II. 1911.
KUPELWIESER, H., Entwicklungserregung bei Seeigeleiern durch

Molluskensperma. Arch. Entw.-Mech. 27. 1909.

218 GENERAL EMBRYOLOGY

LILLIE, F. R., Studies of fertilization in Nereis. III. The morphology
of the normal fertilization of Nereis. IV. The fertilizing power
of portions of the spermatozoon. Jour. Exp. Zool. 12. 1912.
(See also ref. Ch. III.)

LOEB, J., Ueber die Befruchtung von Seeigeleiern durch Seesternsamen.
2 Mitth. Arch, gesamte Physiol. 99. 1903. Artificial Mem-
brane-formation and Chemical Fertilization in the Starfish
(Asterina). Also, On an Improved Method of Artificial Partheno-
genesis. Univ. Calif. Publ. Physiol. 2. 1905. Die Chemische
Entwicklungserregung des thierischen Eies. Berlin. 1909.

MARK, E. L., Maturation, Fecundation, and Segmentation of Limax
campestris. Bull. Mus. Comp. Zool. Cambridge. 6. 1881.

MASING, E., Ueber das Verhalten Nucleinsaure bei der Seeigeleies.
Zeit. Physiol. (Hoppe-Seyler.) 67. 1910. See Arch. Entw.-
Mech. 31. 1911.

MAUPAS, E., Le rejeunissement karyogamique chez les Cilies. Arch.
Zool. Exp. (II), 7. 1889.

MINOT, C. S., The Problem of Age, Growth, and Death. New York.
1907.

MORSE, M. W., Maturing Reagents and those inducing Segmentation in
Artificial Parthenogenesis. Science. (Proc. Amer. Soc. Zoologists.)
33. 1911.

NEMEC, B., Das Problem der Bef ruchtungsvorgange' und andere cyto-
logische Frage. Berlin. 1910.

PHILLIPS, E. F., A Review of Parthenogenesis. Proc. Am. Phil. Soc.
42. 1903.

RtJCKERT, J., Ueber Polyspermie. Anat. Anz. 37. 1910.

SCHAUDINN, F., Ueber die Copulation von Actinophrys sol. Sitz.-Ber.
Acad. Wiss. Berlin. 6. 1896.

SILVESTRI, F., Ricerche sulle fecondazione di un animale a spermatozoi
immobili. Rich. Lab. Anat. Roma e altri Lab. Biol. 6. 1902.-

SOBOTTA, J., Die Reifung und Befruchtung des Eis von Amphioxus
lanceolatus. Arch. mikr. Anat. 50. 1897.

TENNENT, D. H., and HOGUE, M. J., Studies on the Development of
the Starfish Egg. Jour. Exp. Zool. 3. 1906.

WEISMANN, A., (See ref. Ch. I.)

WILSON, E. B., (See ref. Ch. II.)

WOODRUFF, L. L., The Life Cycle of Paramecium when subjected to a
Varied Environment. Amer. Nat. 42. 1908. Two Thousand
Generations of Paramaecium. Arch. Protist. 21. 1911. A Sum-
mary of the Results of certain Physiological Studies on a Pedigreed
Race of ParanuEcium. Biochem. Bull. 1. 1912.

YATSU, N., The Formation of Centrosomes in Enucleated Egg-Frag-
ments. Jour. Exp. Zool. 2. 1905.

CHAPTER VI
CLEAVAGE

THE grosser and externally visible processes of development
begin with the cleavage of the fertilized ovum, or zygote. The
period of cleavage therefore may be regarded as the second of
the " grand periods" in individual history. During the first
general period occur all the events leading up to and including
the final establishment of the zygote, a single cell, but a new
organism. During this second period the Metazoan really
becomes made up of many cells.

The essential process underlying many of the varied phenom-
ena of cleavage is a process already familiar, mitotic cell
division; but it is true that cell division continues long after the
cleavage period proper is terminated, in some tissues through-
out the life of the organism. And as we shall soon see, the
process of cleavage involves a great deal more than merely a
succession of cell divisions.

Certain general characteristics of the mitoses of the period
of cleavage, or segmentation, of the zygote, may be observed,
but it is difficult to state precisely wherein these cell divisions
differ from those of later development. Probably the most sig-
nificant characteristic of the divisions of this period is that they
are rarely at random, but nearly always occur in an orderly
fashion, according to a definite schema or plan, which is quite
fixed for each species or larger group, and which involves the
entire cell community. The mitoses of cleavage are fre-
quently very unequal and the daughter cells may be very unlike,
not only in size, but further as regards cytoplasmic character
and the nature of various cell inclusions, which may be distrib-
uted dissimilarly during these divisions. After the first few
mitoses the blastomeres may not divide synchronously, so that
the regular and rhythmic geometric increase in their number

219

220

GENERAL EMBRYOLOGY

to 2, 4, 8, 16, etc., is very rarely continued after eight or sixteen
cells have been formed; the regularity of division may be dis-
turbed as early as the second cleavage. In some forms (e.g.,
Echinoderms, Godlewski) the nuclei of the daughter cells
enlarge considerably after each division, in some cases perhaps
even to the original size: the cell bodies fail to do so. The
result is the constant increase in the relative size of the cell
nuclei; in other words, while the amount of cytoplasm increases
only slightly or not at all during cleavage, the amount of nucleo-

c

FIG. 105. — Types of blastulae. A. Amphioxus (coeloblastula). B. Petro-
myzon. After von Kupffer. C. Noturus (Teleost) (discoblastula). D. Clava
(Hydroid). After Hargitt. (Solid type.) a, animal pole; c, segmentation
cavity or blastoccel ; p, periblast (a non-cellular protoplasmic layer resting upon
the yolk mass) ; v, vegetal pole.

plasm increases considerably, so that at the close of this period
the organism contains an appreciably greater proportion of
nuclear material than did the zygote. This, however, may not
be regarded as a general characteristic of the cleavage mitoses
in all organisms (Conklin).

After a number of cells, varying in different species, have been
formed they become arranged so as to limit an internal cavity
filled with a fluid. In its simplest and apparently most primi-

CLEAVAGE 221

tive or typical form, the figure thus established is a hollow
sphere, the wall of which is composed of a single layer of cells
or blastomeres. This structure, however its actual form may
deviate from this type, is termed the blastula, and the cavity
within is the blastoccel, or segmentation cavity (Fig. 105). The
diverse forms of the blastula depend immediately upon the
arrangement of the preceding cell divisions; the blastula may in
some cases be almost or quite solid, so that the blastocoel exists
only virtually.

About the time the blastula is formed the successive cleav-
ages have reduced the cell size to a physiological minimum and
thereafter the daughter cells increase in size subsequently to
each division, and there is no further reduction in the size of
the blastomeres; the volume of the cytoplasm, as well as of the
nucleoplasm, commences to increase, in other words the organ-
ism begins to grow. The relative time of the appearance of this
growth phase is widely different in different forms; in the Echi-
noids it appears when about sixty-four cells have been formed
(Godlewski). While there is no general and externally visible
indication marking a definite close of the cleavage period, the
formation of some type of blastula, or the initiation of cyto-
plasmic growth, is more or less arbitrarily assumed to mark its
termination, although many of the processes characteristic of
this phase of development, including of course cell division,
may continue for some time longer. We may now define cleav-
age as that early period of development characterized externally
by a rapid and orderly succession of mitoses, which results in
the formation, from the zygote, of a regularly arranged group
of small blastomeres possessing relatively large nuclei.

In most cases there is a marked tendency for the blastomeres
to assume a spheroidal form, more or less modified by the
tension with which the cells are held together, by the pressure
of the egg membranes, etc. Sometimes the cells round up so
as to become practically spheres, in contact with one another
by greatly restricted surfaces (Amphioxus, Echinoderms, Ccelen-
terates). In other instances (frog, chick, Arthropods) the
separations between the blastomeres are mere grooves or fur-

222 GENERAL EMBRYOLOGY

rows on the surface of the mass; here the cells are broadly in
contact and in some instances remain connected by very
delicate protoplasmic bridges similar to those connecting tissue
cells previously mentioned. In a few instances (some Arthro-
pods, a few Coelenterates) these early cell divisions may be
imperfect, the nuclei alone dividing and forming a syncytium;
later the cytoplasm also divides simultaneously into a number
of complete cells. Or cells once formed may fuse into syncytial
masses, as in some Crustacea.

The internal processes of development occurring during this
period are of greater importance than the external phenomena.
As stated above (Chapter V) one of the underlying processes
of great importance seems to be the synthesis of chromatin
which occurs at this time.

In the opinion of many this is a highly characteristic chemical process
of early development. It results from rapid oxidations within the egg
which were made possible by the transformation of the egg membrane
from a condition of relative impermeability, to a state of high permea-
bility, oxygen thus being readily admitted from without. This trans-
formation is thought to result from the chemical reactions of the
cytoplasm following fertilization, during which there occurs also the
activation, or perhaps the introduction, of specific enzymes which bring
about this characteristic oxidative synthesis.

This view as to the chemical process most essential in cleavage
agrees well with the "kern-plasma" theory of Richard Hertwig, accord-
ing to which, as already mentioned, the ovum and zygote are to be
regarded as instancing abnormal or especially adapted relations between
nucleus and cytoplasm ; for here the relative amount of cytoplasm is far
in excess of the common proportion. In cleavage, with its proportional
increase in nuclear substance, we should see a restoration to or toward

Volume

the normal of the kern-plasma ratio nucleus (VA

Volume \FC/

cytoplasm

Another internal process is of prime importance. We
have already become familiar with certain facts — that one
result of maturation and fertilization is the presence in the
zygote of two similar chromosome groups, derived respectively
from the male and female parents; that while the egg nucleus

CLEAVAGE 223

nearly always returns to a "resting" stage before its fusion
with the sperm nucleus, the latter may or may not do likewise
before the fusion of the two germ nuclei into the cleavage nu-
cleus; that the egg and sperm nuclei may or may not form

o

separate spiremes each with - chromosomes during the pro-
phase of the first cleavage figure of the zygote; and finally that
whatever the preliminary details may have been, the constant
and essential facts regarding this first cleavage figure are, (1)
that the chromosome group now consists of the full somatic
number of univalent elements, (2) that these are present in
pairs, and (3) that they have been derived equally from the
two parents.

As cleavage begins the first important step is the longitudinal
division of each chromosome; the halves diverge during the
anaphase of the first mitosis, and into the nucleus of each
daughter cell there passes a precisely similar group of s chromo-
somes, paired as before and derived in equal numbers from each
of the two parents. In each succeeding mitosis the same thing
happens. So that in every cell of the blastula, and probably
even of the fully matured organism, the nucleus is composed of
substance derived in equal parts from the male and the female
parents (Fig. 106). In some forms (Copepods, Ascaris) the
two parental chromosome groups appear to remain fairly
distinct up to a late stage in cleavage (Fig. 107). And in
some cases of hybridization, when the chromosomes of the
two parents are easily distinguished by differences in size and
form (e.g., the hybrids of Fundulus and Menidia described by
Moenkhaus), such a process of equal distribution of the chro-
mosomes can be clearly followed into the blastula stage (Fig. 39).
It may fairly be assumed, therefore, that Huxley's comparison
of the body of an organism with a web, of which the warp comes
from one parent, the woof from another, has been justified by
the subsequently discovered facts of development. This idea
has been spoken of as the "autonomy of the male and female
chromosome groups." It follows from this that the parental
chromosome groups of the primordial germ cells of the new

224

GENERAL EMBRYOLOGY

D

FIG. 106. — Diagrams illustrating the distribution of the paternal and maternal
chromosomes during cleavage. A. Zygote containing sperm, d\ and egg, Q ,
pronuclei, with similar chromosome groups. B. First cleavage figure. All the
chromosomes on the spindle, and each divided. C. Two-cell stage, each nucleus
containing equivalent chromosome groups of paternal and maternal origin.
D. Second cleavage figure; the first figure is repeated in each cell. E. Four-cell
stage. Nuclei all alike, and each composed of similar contributions from each
parent.

CLEAVAGE

225

organism are also separate and remain so through their descend-
ants, the oogonia and oocytes, or spermatogonia and spermato-
cytes, of the mature individuals, until their period of synapsis,
when the members of each pair of chromosomes, similar but of
diverse ancestry, unite forming a single bivalent chromosome
which is represented in the mature ovum or spermatozoon
finally formed. It has already been suggested that this process
of synapsis, the ultimate fusion of paternal and maternal chro-

FIG. 107. — A. Cleavage figure in one of the first two blastomeres of the egg of
the Crustacean, Cyclops strenuus, showing the independence of the paternal and
maternal chromosome groups. After Riickert. B, C, D. Primitive germ cells
from embryos of the skate, Raja, showing the duplex character of the nuclei.
B and C are from a stage about the close of gastrulation ; D from a larva of 10 mm.
After Beard.

mosomes, may be regarded as the final step in syngamy, and
that it is at the same time the first step in the beginning of a
new organism.

With these briefly stated introductory facts in mind we may
proceed to a more precise description of the events of the
cleavage period. The process of cleavage may be described in
several different ways, or rather from several different view-
points. We may group these all under two heads and consider
cleavage, first, as a morphological process, emphasizing pri-
marily the forms of cleavage and describing the relation of the
cleavage planes and the blastomeres (a) to the entire zygote, and

226 GENERAL EMBRYOLOGY

(6) to each other. Then second, we may emphasize chiefly
the physiological aspects of cleavage describing the relation of
the cleavage processes (a) to the structure or organization of
the ovum and zygote, (6) to the later stages in the development
of the mature organism. As a matter of fact these aspects of
cleavage are not really separate, for all the particular phenom-
ena of cleavage, as of development in general, are to be referred
to a single fundamental condition, namely, the organization
of the ovum or zygote as it is related to external conditions; and
if our knowledge were complete here we should be able to
describe all the phenomena of cleavage from a single viewpoint.
But for the present we shall find it more convenient as well as
more instructive to separate more or less arbitrarily and to
consider apart, the chiefly morphological and the chiefly physio-
logical aspects of this process.

In considering the relation of cleavage to the grosser struc-
ture of the zygote we find that one of the primary factors in
determining the form of cleavage is the relative amount and the
form of distribution of the yolk and other deutoplasmic sub-
stances contained in the ovum. In Chapter III, three types of
eggs were described on this basis: (1) homolecithal or isolecithal
(alecithal), containing little deutoplasm, distributed with
considerable uniformity throughout nearly the entire ovum:
(2) telolecithal, containing varying, often considerable amounts
of deutoplasm chiefly localized toward the vegetative pole of
the ovum; (3) centrolecithal, really a form of telolecithal ova
in which the deutoplasm has a central rather than a polar
localization.

Corresponding in a general way with these variations in yolk
distribution we may distinguish certain types of cleavage, each
however with certain variations which may sometimes appear
as connecting intergradations. First we may distinguish
compkte and incomplete cleavage. When the eggs are com-
paratively small and of the homolecithal type, the earlier
cleavage planes pass completely through them in meridional
and latitudinal planes. Theoretically the simplest form of
complete cleavage is that known as equal cleavage, where the

CLEAVAGE 227

egg and the blastomeres formed from it are always divided
equally, so that the constant result is a group of similar cells.
This is rarely if ever completely realized; the nearest approach
to it is seen in the Holothurian, Synapta (Fig. 108). In most
examples of so-called equal cleavage slight inequalities may be
detected even as early as the two-cell stage (Amphioxus), and
quite frequently in the four-, or eight-cell stages. This modifi-
cation of equal cleavage is known as adequal. Amphioxus
and some of the Echinoderms (Fig. 109) illustrate the fact that
no sharp distinction can be drawn between equal and unequal
cleavage, for equal cleavage soon becomes unequal, and the
transition appears gradually.

More frequently the cleavage though still total is distinctly
unequal, at least by the time eight cells are formed, and often
from the very beginning of cleavage. This is characteristic
of those telolecithal eggs in which the accumulation of yolk is
slightly or moderately marked, as in most of the Platyhelmin-
thes, Nemathelminthes, Annulata, Trochelminthes, Mollusca,
Ganoids, and Amphibia (Figs. 110, 111). This leads to a second
general type of cleavage, the incomplete type, where a portion
of the ovum remains uncut by the cleavage planes (Fig. 113).
Such eggs are known in general as meroblastic, in distinction
from the holoblastic ova whose cleavage is complete. In
telolecithal eggs with very large accumulations of deutoplasm,
the cleavage planes are nearly restricted to the protoplasmic
region and extend only a short distance out into the yolk;
this is known as partial cleavage. When the protoplasmic part
is quite definitely restricted to the animal pole, cleavage is of
an extremely incomplete type known as discoid (Fig. 116), and
the result is the formation of a small cap or disc of cells on
the surface of the yolk mass (Teleosts, Reptiles, Birds). Or,
if the egg is of the centrolecithal type, cleavage is limited to the
peripheral protoplasmic layer and is known as superficial
(most Arthropods) (Figs. 117, 118).

Since there are all intermediate conditions between homo-,
telo-, and centrolecithal types of ova, we find, as we should
expect, all corresponding intermediate conditions between these

228 GENERAL EMBRYOLOGY

.various forms of cleavage; the details are not particularly in-
structive and may be omitted.

We may now turn to the morphological description of the
relations of the blastomeres among themselves. Before
describing the various relations which these may exhibit, it will
be useful to describe briefly a simple form of total and equal
cleavage, which may be regarded as a typical form. Such a
form of cleavage is indeed rare but it is found in the homoleci-
thal and holoblastic egg of the sea-cucumber, Synapta, as

I> E

FIG. 108. — Cleavage in the Holothurian, Synapta, Slightly schematized.
From Wilson, "Cell," after Selenka. A-E. Two-, four-, eight-, sixteen-, and
thirty-two cell stages. F. Blastula of 128 cells. B, in polar view, others in side
view.

described by Selenka (Fig. 108). The earlier cleavage planes
always appear in a definite relation to the polar structure of the
ovum, and they are described as if the main axis of the egg were
in a vertical position. The first cleavage plane passes through
both poles and the chief axis of the egg, dividing it equally and
appearing on the surface as a complete meridian. The second
plane is similarly meridional or vertical, is at right angles to the
first, and divides the egg into four equal quadrants. The third
division plane is at right angles to the first two and is therefore
horizontal. In Synapta it is practically midway between the

CLEAVAGE 229

poles of the egg and is therefore described as equatorial. In
most ova cleaving according to this general rule, the third plane
is displaced a variable distance above the equator and is then
termed latitudinal. When equatorial this cleavage divides the
four equal cells into eight, again equal, arranged in upper and
lower groups of four, known as the upper and lower quartets.
The fourth cleavage is again meridional and is really double,
for two planes appear simultaneously dividing each pair of
opposites in each quartet similarly; this results in the formation
of upper and lower octets. The fifth cleavage is horizontal or
latitudinal, and is again double for it divides simultaneously
the upper and lower octets each into two horizontal groups of
eight cells, so that the ovum is now divided into eight vertical
rows of four cells each. The cleavages continue to alternate
meridionally and vertically, until about the ninth cleavage when
512 cells are formed. After this, and in fact usually before this
time, the synchronism of cleavage begins to be disturbed, some
of the cells dividing more rapidly.

One of the very frequent causes of departure from this simple
schema is the telolecithal character of the ovum. Here the
upper quartet is usually smaller than the lower and the fourth
cleavage appears earlier in the upper quartet or cells of the
animal pole. This leads very soon to an irregularity in the
rhythm of cleavage, which may be entirely lost after eight or
sixteen cells are formed.

This typical outline of cleavage serves as a basis to illustrate
certain "laws" of cleavage which may be referred to briefly
at this point, although their applicability is now known to be
Very limited. The first of these is the Sachs-Hertwig law
describing the geometric relations of the successive cleavage
planes. This law really consists of two parts which may be
stated as follows: (1) The nucleus of a blastomere (or of any
cell) tends to assume a position near the center of the proto-
plasmic mass. From this results the equal division of the cell,
provided it is free from deutoplasm, or its unequal division
if the cell contains deutoplasm not uniformly distributed, for
in the latter case the center of the protoplasmic mass does not

230 GENERAL EMBRYOLOGY

correspond with the center of the entire cell. (2) The chief
axis of the mitotic figure tends to lie in the longest axis of the
protoplasmic mass. The result of this is that in cells that are
approximately spherical and homogeneous with respect to yolk
content, successive cleavage planes tend to alternate at right
angles with one another, for it would always be the longest axis
of the cell that is divided, and in most cases any other axis
would be greater than one-half the longest and no two successive
spindles would be parallel. The regular alternation of cleav-
age planes probably depends, as a matter of fact, upon a more
fundamental relation, namely, the position of the centrosome.
At the conclusion of a mitosis the centrosome lies at one end
of the axis passing perpendicularly to the plane of division;
when the centrosome divides, its halves usually migrate sym-
metrically to opposite sides of the nucleus, occupying the poles
of an axis lying parallel with the plane of the preceding division,
and since division always occurs at right angles to the axis
connecting the centrosomes, the plane of one division will be
at right angles to that of the preceding or succeeding cleavage
(Fig. 24). Any other relation between successive cleavage
planes involves either a change in the relative position of the
centrosomes, or a rotation of the cleavage spindle after its
formation. Thus in the formation of a simple epithelium,
where successive cleavages are nearly parallel, the centrosomes
migrate through approximately 90° at some time during the
inter kinesis; and in the cleavage of some ova encased in com-
paratively rigid shells, the position of the spindle may change
(Lepas, Bigelow).

Balfour's law of cleavage, which is really a corollary of
the first part of the Sachs- Hertwig laws, concerns the rate
rather than the geometrical relations of cleavage. This law
states that the rate of cleavage is inversely proportional to the
amount of deutoplasm contained within the cell. It follows
from the fact that the nucleus tends to lie in the center of the
protoplasmic mass, that in the unequal division of cells con-
taining localized deutoplasm, the smaller cell will contain
relatively a smaller proportion of yolk than the larger cell, and

l

CLEAVAGE 231

consequently, being free from the influence of the dead and
inert deutoplasm, will be able to divide sooner. •

While these laws are often applicable to the processes of cleav-
age in a general way, the exact study of cleavage in a great
variety of forms has disclosed very numerous exceptions and
contradictions. On the whole we may say that such laws, though
still retaining a limited applicability, are chiefly interesting as
indicating the attempt to refer the phenomena of cleavage to
the grosser mechanical relations of cell structures. It is now
clear, as we shall see later, that other factors are of greater im-
portance in determining the form and rhythms of cleavage.
The fundamental "organization" of the ovum, which is not only
morphological but physiological as well, is the primary factor in
determining the characteristics of cleavage. The numerous
"exceptions" to these laws of cleavage are definitely related to
both the organization of the ovum and also to the structural
and functional characters of the later stages of development,
since these too are primarily determined by the same organiza-
tion factor.

Most of the conditions which form exceptions to these rules,
and are therefore deviations from the simple and regular
form of cleavage like that of Synapta, may, following Wilson
("The Cell," etc.), be grouped under three heads. (1) Unequal
Division. While this is usually related to differences in deuto-
plasmic content, there are many instances where no such rela-
tion can be made out and the inequalities must be explained
upon other grounds (e.g., the micromeres of the Echinoids, the
teloblasts of certain Annelids and Molluscs). (2) Cell Dis-
placement. This may result from the atypical position of the
spindle or from the shifting of blastomeres after they have been
formed. Often the individual blastomeres are only lightly
held together since they normally show a tendency, often very
marked, to assume a spherical form. Under these conditions
they might tend to assume a position described by the law
(Plateau's) of "least surfaces" or "minimal contact," according
to which a group of elastic spheres, like bubbles, held together
and yet free to move, tend to become arranged in such a way

232 GENERAL EMBRYOLOGY

as to reduce their exposed surfaces to a minimum. But there
are frequent Exceptions here as in the case of the other "laws"
of cleavage. Furthermore, the active migration of blastomeres
is not infrequent, so that cells may ultimately be found in
regions considerably removed from the place of their formation
(Rotifers, Molluscs). (3) Rhythm. The rate of division fre-
quently does not correspond with the relative amount of
deutoplasm. The factors regulating the rhythm of division
still remain largely unknown. It is true here as in many other
" exceptional" cleavage phenomena, that the deviation is
related to the future morphological or functional character
of the developing organism or of parts of it. We shall return
to this aspect of cleavage later.

With these general considerations in mind we may proceed
now to a more exact description and classification of the geo-
metric forms of cleavage. Here we shall find illustrations of
many of the preceding statements. Considering first the various
forms of complete cleavage (holoblastic ova), we may distin-
guish rather roughly, four types, radial, 'spiral, bilateral,
irregular.

(1) Radial. — This is the form exemplified by Synapta (Fig.
108), already described as being geometrically the simplest.
This should perhaps better be termed rotatorial than radial, for
while the blastomeres are arranged in symmetrical fashion in
any single plane perpendicular to the main axis of the egg,
there are usually considerable differences in size between the
cells of the animal and vegetal poles. Cleavage of this type
is found in the sponges, jelly-fishes, and many Echinoderms,
in some Nematodes and Rotifers. In the sea-urchins (Fig. 109)
the third cleavage is meridional in the upper quartet, in the
lower latitudinal and very unequal, cutting off a quartet of very
small cells or micromeres which curiously are found at the lower
or vegetative pole.

(2) Spiral — This may be regarded as a modification of the
radial type resulting from the displacement of cells so that the
blastomeres above and below any horizontal cleavage furrow
tend to alternate with one another in a vertical direction, some-

CLEAVAGE

233

FIG. 109. — Cleavage in the sea-urchin, Strongylocentrotus limdus. From
Jenkinson, after Boveri. Animal pole uppermost in all cases, a. Primary
oocyte surrounded by jelly, and containing large germinal vesicle with nucleolus.
Pigment uniformly distributed over surface, b. Ovum after formation of polar
bodies. Pigment forms a band below the equator, c, d. First cleavage, e.
Eight-cells. Pigment almost wholly in lower quartet (vegetative blastomeres) .
/. Sixteen-cells. The lower quartet has divided latitudinally and unequally,
forming four micromeres at the vegetal pole; the upper quartet has divided me-
ridionally forming a plate of eight cells, g. Section through blastula. h. Later
blastula, showing formation of mesenchyme at lower pole, i, j, k. Three stages
in gastrulation, showing the infolding of the pigmented cells to form the endoderm
(archenteron) . In j the primary mesenchyme is separated into two masses, in
each of which a spicule is formed (fc). In k the secondary, or pigmented, mesen-
chyme is being budded off from the inner end of the archenteron.

234

GENERAL EMBRYOLOGY

what like the bricks in a wall or the bones of the wrist. They
may be cut off in this way on account of the obliquity of the
spindle in the parent cell, or they may shift to this position
after having been formed according to the radial plan. This
spiral arrangement may be foreshadowed in the four-cell stage
by the meeting of the first two cleavage planes at the poles of
the egg in the form of a zig-zag line instead of at a common
point.

D

FIG. 110.— Cleavage in the Annulate, Polygordius. From Wilson, "Cell."
A, B. Four- and eight-cell stages, from the animal pole. C. Side view of eight-
cell stage. D. Side view of sixteen-cell stage.

One of the simplest illustrations of this type is the adequal
cleavage of Polygordius (Fig. 110) but it is also well represented
by the markedly unequal cleavage of many Platyhelminthes,
Nemertines, Annelids, and Molluscs (Figs. Ill, 112, 119).
These forms illustrate at the same time a graduated series in the
inequality of the blastomeres. This inequality may appear in

CLEAVAGE

235

different stages; in the very first division of the egg (Nereis);
at the second (Clavelina)] third (Cerebratulus, Fig. 112); fourth
(sea-urchin), or still later (Synapta). The direction which the
spiral takes is fixed in each species; it is described as dextral

FIG. 111. — The eight-cell stage of four animals showing gradations in the
inequality of the third cleavage, and in the extent of the spiral rotation of the
micromeres. From Wilson, "Cell." All viewed from the animal pole. A. The
leech Clepsine (Whitman,). B. The chsetopod Rhynchelmis (Vejdovsk^),
C. The lamellibranch Unio (Lillie). D. Amphioxus.

(dexiotropic) or sinistral (Iceotropic) when the upper cells are
rotated clockwise or counter-clockwise respectively, as viewed
from the animal pole.

(3) Bilateral. — In this form we see a second modification
of the radial type, which is first indicated by the fact that the
third of the meridional cleavages fails to reach precisely the
poles of the egg but meets either the first or second plane at
some distance from the pole. In this way it comes about that

GENERAL EMBRYOLOGY

FIG. 112. — Cleavage in the Nemertean, Cerebratulus marginatus. From
Korschelt and Heider, after Zeleny. X 216. A, B. Two- and four-cell stages
in side view. C. Four-cells, from animal pole. D. Eight-cells from vegetal
pole. E. Eight-cells, side view. F. Sixteen-cells, side view. G. Twenty-eight
cells, side view. H. Twenty-eight-cells from vegetal pole. For explanation of
lettering, see p. 248.

CLEAVAGE

237

one of the first two cleavage planes, usually the first, becomes
the plane of a bilateral symmetry which may remain quite pro-
nounced for some time, and indeed corresponds with the median
plane of the bilaterally symmetrical adult. The bilateral type

FIG. 113. — Meroblastic cleavage in the squid, Loligo pealii. A, B. Egg viewed
obliquely, showing animal pole. X 45. After Watase. C, D. Surface views of
animal pole, more highly magnified, to show bilateral arrangement of blasto-
meres. From Wilson, "Cell," after Watase. A. Four-cell stage. B. About
sixty-cells. Cells at animal pole very small, lowermost cells incomplete, cell
walls extending down toward the uncleaved lower pole. C. Eight-cell stage.
D. The fifth cleavage (sixteen to thirty-two cells), a-p, marks the plane of
the first cleavage and the median plane of the organism; l-r, marks the second
cleavage, and the transverse plane of the organism.

of cleavage is found in the Cephalopods (Fig. 113), a few
Rotifers and Nematodes, in Amphioxus and the Ascidians, and
perhaps in most of the Craniates, but in these last forms varia-
tions are more frequent, especially in those forms with discoid
cleavage.

238

GENERAL EMBRYOLOGY

A rather special form of bilateral cleavage known as the
disymmetrical type is found in the Ctenophores. In these
Coelenterates the first and second furrows are meridional, the
planes are complete, and divide the egg adequally. The
third, a double cleavage, is oblique to the first, passing in the
same general direction as this, on each side of it, but approach-

D

FIG. 114. — Diagrammatic representation of the cleavage in the Ctenophore
(based upon Beroe). After Ziegler. A. Four-cells, from side and above.
B. Eight-cells in side view (the animal pole is downward here, and in Z>). C.
Eight-cells, from animal pole. D. Sixteen-cells, from side. E. Sixteen-cells,
from animal pole, m, m. Median plane; t, t, transverse plane.

ing the vegetal pole more closely than the animal. The descend-
ants of each of the first two cells then become symmetrically
arranged about the second plane so that two similar groups of
cells are formed, each group bilaterally symmetrical with refer-
ence to a plane perpendicular to the plane of symmetry of the
entire cell group (Fig. 114).

CLEAVAGE

239

'(4) Irregular. — In many phyla, scattered forms are known
in which cleavage adheres to no single or simple type and may
truly be said to be irregular. This does not mean that no
definite plan is followed, for each species follows a fixed rule;
these cleavage forms are in this sense regular, but cannot

C

FIG. 115. — Irregular cleavage in the Turbellarian, Mesostomum ehrenbergi.
After Bresslau. X 700. A. Three-cell stage, in section. B. Four-cells
becoming five. Side view. C. Seven-cell stage, from animal pole. D. Twelve-
cells. A. Macromere, giving rise to A\ and Az in the seven-cell stage. B. First
micromere, forming Bi and Bz in the five-cell stage. C. Second micromere
formed in three-cell stage, and giving rise to Ci and Ci in the seven-cell stage.

be described in general terms. We cannot stop to describe
any of these instances in detail. Irregular cleavage may be
found among the Porifera, Coelenterates, Platyhelminthes,
Molluscoids, Enteropneusta, and Teleosts; a typical example
igi illustrated in Fig. 115.

The remaining forms of cleavage are grouped as incomplete,
and are found among those species with markedly telolecithal

240 GENERAL EMBRYOLOGY

or centrolecithal ova (meroblastic) . Here little or no geometric
regularity of cleavage pattern can be made out. We may add
a few details concerning discoid and superficial cleavage to the
brief statements made on a preceding page.

Discoid.— This is chiefly characteristic of the Craniata but it
is found occasionally in the Arthropods (Scorpions). In the

c •,•:,:?

s.c.

FIG. 116. — Cleavage in the sea-bass, Serranus atrarius. From H. V. Wilson.
A. Surface view of blastodisc in two-cell stage. B. Vertical section through
four-cell stage. C. Surface view of blastodisc of sixteen cells. D. Vertical
section through sixteen-cell stage. E. Vertical section through late cleavage
stage. c.p.t central periblast; ra.p., marginal periblast; s.c., segmentation cavity
(blastocoel).

ova of these forms there is often a fairly definite demarcation
between the protoplasmic and deutoplasmic portions (Elasmo-
branchs, Teleosts, Reptiles, Birds) and the cleavage planes are
practically limited to the former region, known as the blasto-
disc (Figs. 48, 157-159). The early cleavages may be fairly
regular, and approximate either radial or bilateral arrange-
ments, and as far as the protoplasm alone is concerned, the

CLEAVAGE 241

cleavage is frequently equal. When the protoplasm forms
merely a disc resting upon a large mass of yolk, it is obviously
impossible to speak of meridional or latitudinal cleavages;
hence the cleavages are described as vertical, either radial or
circular, and horizontal. The vertical cleavages soon become
connected below the surface of the ovum by horizontal planes,
separating the lower surface of the protoplasm from the
underlying yolk, and the peripheral circular cleavages similarly
separate the protoplasm from the outlying yolk (Figs. 116,
158, A). This is seen in the Teleosts, and in many Elasmo-
branchs, Reptiles, and Birds. After the protoplasmic blasto-
disc is divided into a number of cells, that is after it becomes a
blastoderm, other cleavages may occur parallel with the surface,
forming internal cells not visible upon the surface (Fig. 116),
and the blastoderm may thus come to be many cells in thick-
ness (Figs. 158, 105, D).

Some interesting transitions are to be found between total
unequal cleavage and discoid cleavage, in those telolecithal
eggs where the accumulation of yolk is not as great as it is in
the Elasmobranchs, Teleosts, and some of the higher Craniates.
Thus in the ganoid, Amia, and some of the Urodeles, as well as
in the squid (Loligo), while cleavage is at first limited to the
upper or animal pole, the earlier cleavages gradually extend
down through the yolk mass and may finally divide it into a
few large cells. Here the more peripheral circular cleavages
(latitudinal) do not form any sharp separation between' proto-
plasm and deutoplasm, and the yolk mass is for a long time
only partially divided by meridional cleavages alone (Fig. 113).

Superficial. — This type of cleavage is characteristic of Arthro-
pods in general and occurs elsewhere only in a few Coelenterates.
The central accumulation of the yolk, as it occurs in these forms,
is an unusual condition, and correlated with this we find
several unusual features in cleavage. Of course in such an
arrangement the most obvious distinction between animal
and vegetal poles is entirely lacking, and usually the position
of the polar bodies and the external form of the egg are the
only outward indications of the polarity of the ovum. Before

242 GENERAL EMBRYOLOGY

cleavage begins the nuclear structures are located centrally,
together with a small amount of protoplasm, and surrounding
this is a dense mass of yolk, interpenetrated by a very fine
network of protoplasm. At first nuclear division is not followed
by division of the inert remainder of the egg mass. But after
a varying number of nuclear divisions, the daughter nuclei
separate, each accompanied by a small mass of protoplasm,
the "protoplasmic island" and migrate to the surface of the
egg, continuing to multiply as they go (Figs. 117, 118). In

D

FIG. 117. — Superficial cleavage in the Decapod, Dromia. (Sections.) After
Cano. A, B. Intravitelline divisions of the nucleus. C, D, E. Formation of
yolk-pyramids. F. Blastula; a superficial layer of cells enclosing a mass of
yolk, n, nuclei; p, yolk pyramids; y, yolk bodies.

this way the nuclear and cytoplasmic portions form a kind of
superficial syncytium leaving the condensed and undivided
yolk centrally.

Now either of two things may occur. In some forms (Deca-
pods, Copepods, Ostracods, Amphipods) cleavages appear
almost simultaneously, dividing the egg completely into a
number of cone-shaped cells with the apices directed centrally;
these cells are known as yolk pyramids (Fig. 117). In some
cases the formation of the yolk pyramids does not occur
simultaneously throughout the egg, but occurs earlier on that
side of the egg which corresponds with the ventral surface of

CLEAVAGE

243

the embryo and larva. After a variable but considerable
number of yolk pyramids are formed the planes of separation
gradually disappear except in the superficial protoplasmic
layer, which alone remains cellular, while the yolk again
becomes a solid mass. Such a process as this is taken to mean
that this type of cleavage may have been derived from the
total adequal type.

FIG. 118. — Cleavage in the beetle, Hydrophilus. From Korschelt and Heider,
after Heider. A, E. Intravitelline divisions of the nucleus. C. Beginning of
the formation of a superficial layer of cells. D. Later stage in formation of
"blastoderm," or superficial cell layer. 6, blastoderm, or superficial layer of
cells; d, yolk;/, nuclei surrounded by protoplasm (protoplasmic islands) ; z, nuclei
remaining in yolk (merocytes).

In other examples of nearly all groups of Arthropods the
yolk pyramids are incompletely developed or even entirely
absent and the cleavage is strictly superficial (Fig. 118).
Here, as in the preceding, blastomeres may be formed either
wholly or only partially around the ovum. The preliminary
divisions of the nucleus and the formation of protoplasmic
islands occur as described above. In many of the Insects,
whose cleavage is typically superficial, the substance of the
ovum ultimately becomes completely divided internally.

The preceding classifications and descriptions of cleavage
have been almost wholly morphological in their basis. But

244 GENERAL EMBRYOLOGY

development is a process, not a succession of morphological
stages, and there remains to be described the most important
aspect of cleavage as a developmental process. We come then
to still another classification of cleavage types as (1) deter-
minate and (2) indeterminate.

Cleavage is said to be determinate when exact morphological
and physiological relations exist between the individual blasto-
meres and, (a) specific structures in the embryo and fully
developed organism, and also (6) specific regions or substances
in the ovum. Each of the products of cleavage is here a true
organ, of particular and known value in development: blasto-
meres are not interchangeable, and their removal or destruction
may lead to specific, related defects or abnormalities in later
development (e.g., the Ascidians). Cleavage is described as
indeterminate when the blastomeres seem to have no specific
relation to the structure of either the egg or embryo and adult.
Here all the blastomeres may have equal value as factors in
development; they are more or less interchangeable, and
removal or destruction leads only to the loss of a corresponding
amount of substance, not to the absence of any specific or
related parts (e.g., the Echinoderms).

In order that this classification should not be misleading, it
should be said at once that this is a purely artificial distinction,
based rather upon historical grounds than upon the facts of
development, for these two types are completely connected by
transitional conditions, and soon or late in development, all
cells come to have specific, determined values.

Such a grouping as this, of the varieties of cleavage, obviously
rests primarily upon a physiological rather than a morphological
basis, for cleavage here is regarded as a process of development.
In all cases of determinate cleavage the essential fact is that
cleavage is not the mere division of the zygote into separate
masses and units which can be moved about and moulded into
the form of an embryo; cleavage is not merely a series of cell
divisions, not the mere " vegetative reduplication of parts"
occurring in accordance with certain mechanical rules like
those of Balfour, Sachs-Hertwig, or Plateau, mentioned above.

CLEAVAGE 245

We have already seen that the exceptions to these rules are so
numerous and so fundamental that they must have some real
significance. It is now clear that in determinate cleavage al1
the details have a significance that is prospective, looking toward
the structural and physiological characteristics of the larva or
fully formed organism. It has been said regarding determinate

cleavage that "One can go over every detail of

cleavage, and knowing the fate of the cells, can explain all the
irregularities and peculiarities exhibited" (Lillie).

Why this should be true is partly explained when we remem-
ber that the characters of cleavage and of the fully developed
organism are both the primary result of the underlying structure
of the ovum. Cleavage stands as an intermediate process
between egg organization and adult structure; it is one of the
processes through which the primary organization of the ovum
gains expression in adult form.

This view of the cleavage process is by no means the only,
or the original view, but it serves to bring out clearly the fact
that the problems as to the nature and causes of the differ-
entiations occurring during the cleavage process are related to
the problems of the nature and causes of the differentiations of
adult structure. Indeed 4hese differentiations have a common
cause in the structure and reactions of the ovum, and are there-
fore fundamentally equivalent. In our introductory chapter
we said that the organism is specific at every stage, the zygote,
the group of blastomeres, the embryo, the adult, are all the
same specific organism, and the question why the cleavage
group is what it is, is the same as the question why the mature
organism has its own specific and individual characteristics.
The problems and processes of development are fundamentally
alike throughout.

In continuing our discussion of this determinative aspect of
cleavage we shall make little further attempt to distinguish the
determinate and indeterminate forms since this separation is
clearly artificial. The apparent differences between deter-
minate and indeterminate cleavage may arise from the fact
that one of the determining factors contains a variable. That is.

246 GENERAL EMBRYOLOGY

the time at which the organization of the egg becomes suffi-
ciently complete to be effective in determining the course of
differentiation of the blastomeres, may vary. Thus cleavage
would be completely determinate if the egg organization were
entirely or largely completed before the cleavage process begins,
incompletely determinate if the organization is only partial,
and indeterminate if the organization is only slightly
marked during the earlier cleavages. For egg organization is
progressive, it develops. So the determinate or indeterminate
character of cleavage may depend, partly at least, upon the
relative time during cleavage at which the organization becomes
marked to such an extent as to determine the fate of particular
blastomeres. Other factors obviously enter into the process
and we shall review the subject from another point of view in
the next chapter.

We have seen above that in nearly all species the earliest
cleavage planes are definitely related to the polar axis of the
ovum. The polarity of the egg is one of the fundamental
aspects of its organization. We have seen also 'that the ovum
often contains formed substances of various kinds, both proto-
plasmic and deutoplasmic, distributed in the cytoplasm in a
definite and usually specific manner/ It is a common feature
of cleavage that the first plane symmetrically divides the egg
or that part of it which takes part in the process of cleavage.
And furthermore, with very few exceptions, this first cleavage
plane coincides either precisely or approximately with the
median plane of the embryo and adult. Nereis is one of the
few exceptions to this rule; in this Annelid, as in some of the
Urodeles, the second plane marks the future median plane.

The factors which appear-immediately to determine the loca-
tion of the first cleavage plane are mainly two. First and most
important is the structure of the ovum itself, which in many
cases, even in the unfertilized condition, is obviously bilaterally
symmetrical; the first cleavage plane corresponds closely with
this plane of symmetry, which is therefore determined by the
same " organizational " factor that determines the polarity and
other structural features of the ovum. In other cases the

CLEAVAGE 247

symmetry of the egg appears to be radial or rotatorial before
fertilization and is converted into a bilaterally symmetrical
structure by the entrance of the spermatozoon, the entrance
path of which marks the plane of symmetry of the egg and
developing organism, and determines the location of the first
cleavage. It is quite possible, though hardly demonstrated
as yet, that even in such cases there is really an invisible bilateral
structure of the ovum which underlies the radial symmetry
and really determines the point at which the sperm shall enter.
In such a case the entrance path of the spermatozoon would
itself be predetermined and could not be regarded as a primary
factor in fixing the position of the first cleavage. This would
obviously be the case in many of those eggs possessing micro-
pyles. But in some eggs whose cleavage is indeterminate, even
though they possess micropyles (Teleosts), there seems to be
no regularity in the position of the first cleavage plane and no
correspondence between this and any morphological char-
acteristic of either ovum or adult.

The second cleavage, usually at right angles to the first,
ordinarily corresponds with the median transverse axis of the
egg, embryo, and adult. The third cleavage is usually hori-
zontal and separates animal and vegetal poles and corresponds
most frequently with the separation, in the embryo and adult,
of the more active animal, and less active vegetative tissues.

The facts that in all eggs of a given species or genus cleavage
occurs according to a definite pattern, and that there may be
an exact relation between the individual blastomeres of the
cleaving ovum and the tissues and organs of the later organism,
make it possible to speak of the "cell lineage" (Wilson) of an
organism. In forms with determinate cleavage it becomes
possible to identify, even in a comparatively late embryonic
stage, various groups of cells as the real lineal descendants of
certain individual cells of the earlier cleavage group. In other
words, it is in such cases possible to trace the structures of the
embryo and adult back to single cells or parts of cells.

In order to illustrate the nature of the facts of cell lineage,
and the completeness and exactness of the correspondence be-

248 GENERAL EMBRYOLOGY

tween blastomcres and differentiated groups of cells in the later
embryo, we may describe briefly a typical instance, using as the
subject one of the simpler and more regularly cleaving types,
the Turbellarian, Planocera, as described by Surface. This
account of the cleavage of this form should be read with the
expectation of finding frequent "exceptions" to the laws of
cleavage mentioned above.

In Planocera (Figs. 119, 120) cleavage is total, unequal, and
spiral (dexiotropic). The first plane is meridional and divides
the egg into two adequal blast omeres known as AB and CD.
The division of each of these is also meridional but is unequal,
each forming a smaller and a larger cell, and dividing the entire
ovum vinto two larger, and slightly unequal, cells known as B
and D, and two smaller cells, also slightly unequal, known as
A and C. Of these D is the largest, and from later develop-
ment is known to be posterior in position; B is anterior, A on the
left, and C on the right, as viewed from the animal pole and
with reference to later structure.

The third cleavage is horizontal, unequal^ and strongly
spiral (dexiotropic). As usual among the Turbellaria the
larger cells divide shortly before the smaller. On account of the
size differences between the cells of the upper and lower
quartets, we may describe the upper quartet of smaller cells
(micromeres) as budded off from the lower quartet of macro-
meres. The quartets of micromeres are designated by small
letters, the macromeres by capitals. Thus in the eight-cell
stage we have the first quartet of micromeres, la, Ib, Ic, Id,
and the first quartet of macromeres, 1A, IB, 1C, ID. Successive
quartets are designated by numerical coefficients, products of
the division of the quartet cells by exponents. Thus in passing
from eight cells to sixteen the macromeres bud off, this time
in a Iseotropic direction, a second quartet of micromeres, 2a, 2b,
2c, 2d, the macromeres themselves remaining known now as 2A,
2B, 2C, 2D. Shortly thereafter the first quartet of micromeres
divide, also Iseo tropic ally, forming two groups of four cells each
known as la1, Ib1, Ic1, Id1, and la2, Ib2, Ic2, Id2; the cells lying
toward the animal pole are designated by the lower exponent.

CLEAVAGE 249

The fifth cleavage, dividing the sixteen cells into thirty-two,
in general resembles the preceding but is dexiotropic through-
out. As the result we have the macromeres, 3A, 3B, 3C, 3D;
a third quartet of micromeres, 3a, 3b, 3c, 3d; the second quartet
of micromeres divides into 2a*, 2b\ 2c\ 2d\ and 2a2, 2b2, 2c2,
2d2, while the eight cells previously derived from the first
quartet of micromeres now form la11, la12, la21, la22, Ib11, lb12,
lb21, lb22, Ic11, Ic12, Ic21, Ic22, Id11, Id12, Id21, Id22.

After thirty-two cells have been formed in this fairly regular
fashion, the rhythm of cleavage becomes modified so that there
follow stages of 40, 44, 45, 53, 61 cells, etc.

It is unnecessary for us to go farther with the details of
these cleavages save in one particular. After the thirty-two-
cell stage the macromeres divide again unequally giving off a
group of large cells which contain most of the deutoplasm of
the original ovum. In spite of their size relations these large
cells are known as the fourth quartet of micromeres, 4a, 4b, 4c,
4d, and the remaining smaller cells as the fourth quartet of
macromeres, 4A, 4B, 4C, 4D. This contradiction in termin-
ology is justified by the later history of these 'cells.

We may now consider the fates of these various groups of
blastomeres. Quoting from Surface, "From the first quartet
[of micromeres] arises the ectoderm, covering the anterior
and dorsal portions of the body. From cells of this quartet four
strings of cells bud into the interior of the embryo and form the
ganglion. The eyes arise in ectodermal cells of this quartet.
The second quartet gives rise to the larger portion of the ecto-
derm on the ventral and posterior regions of the body. From
cells of this quartet is formed most of the ectodermal pharynx.
A portion of the second quartet is budded into the embryo and
forms mesoderm. From this source arises probably only that
mesoderm formed around the blastopore and which is later
concerned in the structures of the pharynx.

"The third quartet consists of small cells from which appar-
ently only ectoderm is derived. The individual divisions of
these cells have not been traced very far, but there is every
reason to believe that they form ectoderm only.

250

GENERAL EMBRYOLOGY

FIG. 119. — Cleavage and cell lineage in the Polyclad Turbellarian, Planocera
inquilina. From Surface. A. Egg during the first cleavage; side view. The
cell C-D is slightly larger than A-B. B. Four-cell stage, from animal pole. C.
Formation of first quartet, from right side, showing spiral cleavage (dexio tropic).
D. Eight-cell stage, from animal pole. E. Eight-cells dividing into sixteen,
showing laeotropic division. The division of the cells of the D quadrant is in
advance of the others. F. Sixteen-cells, from animal pole. G. Sixteen-cells,
from vegetal pole.

CLEAVAGE

251

FIG. 120. — Continuation of Fig. 119. H. Dexiotropic division of la1-!^1 and
of 2a-2d. From animal pole. /. Thirty-two-cells, from animal pole. J.
Thirty-two-cells from vegetal pole. K. Thirty-two-cells from right side. L.
Late cleavage showing the history of cells 4a-4d and 4A-4D. M. Optical section
of a much later stage, viewed from near the vegetal pole. The mesoderm bands
are stippled.

252

GENERAL EMBRYOLOGY

"The history of the fourth quartet is peculiar .....
The posterior cell 4d is the mesentoblast, from which the ali-
mentary canal and a portion of the mesoderm arise. The
other three cells of the fourth quartet, 4a, 4b, 4c, do not divide
as long as their history can be traced. They, however, break
up into a large number of homogeneous yolk spheres which are
absorbed by the endoderm cells. The large nuclei of these
three cells can be traced until the alimentary canal is partly
formed.

TABLE OF THE CELL LINEAGE OF PLANOCERA
Condensed from Surface.

la11
| ia- i

la

v la2
A (left)

2a

AB

B (ant.)

Zygotc

C (rt.)

CD

D (post.)

CLEAVAGE

253

la111 ] 1

Ib111 ]

lciii j> apical cells, j

Id111 j 1

4a, 4A

4b, 4B

yolk (no cell

4c, 4C

descendants) .

4D

12212
Ib112212
10112212
1^112212

4d

primary ganglion
cells.

mesoblast.

4d5

4d* — alimentary canal (endoderm).

4d211 — probably endo-
derm.

4d212 — mesoderm (right
"mesoblast band").

4d221 — probably endo-
derm.

4d222 — mesoderm (left
" mesoblast band").

The remaining cells form covering ectoderm.
Ectoderm of first quartet — anterior and dorsal, including eyes.
Ectoderm of second quartet — posterior and ventral, including pharyngeal.
Mesoderm of second quartet — blastoporal (pharyngeal).

"The nuclei of the small macromeres [4A, 4B, 4C, 4D] show
evidences of degeneration. These do not divide as long as

they can be followed and it seems probable that

they degenerate without giving rise to any morphological
structure."

This cell lineage of Planocera is summarized incompletely in
the accompanying table.

It is interesting to compare with this lineage of Planocera
that of Ascaris, described by Zur Strassen, which is somewhat
less regular. This is particularly interesting as it shows
clearly the history of the germ cells, which become wholly
separate from somatic cells in the sixteen-cell stage. Cleavage
of Ascaris (Fig. 121) is bilateral but more or less irregular/
particularly in its rhythms, so that without attempting to
apply the ordinary terminology completely we may summarize
the early cell history in the table accompanying (p. 255).

The cell lineage of a considerable number of organisms has
been definitely traced, often in much greater detail than we
have indicated. The histories best known are found among
the Platyhelminthes, Nemathelminthes, Nemertinea, Annulata,
Trochelminthes, Mollusca, and Tunicata. Of course the eggs of
many classes and phyla show no such regularity, for as we have

254

GENERAL EMBRYOLOGY

pointed out, cleavage may be irregular as well as indeterminate.
And in many of the groups named above, normal development

(CMSt)

FIG. 121. — Cleavage in Ascaris megalocephala bivalens. From Jenkinson, after
Boveri. 1. Division of the two-cell stage. Elimination of chromatin in the
somatic cell Si(AB). la. Chromosomes of the cell Si(AB). 2. Four-cell stage
(T-form). In A and B can be seen the eliminated chromatin. The cell Pi has
divided into a somatic cell Sz(EMSt), in the descendants of which chromatin
elimination occurs, and the cell Pz. 3. Four-cell stage (lozenge-form). A is
anterior, A and B, dorsal. 4. Continued chromatin elimination in somatic
cells. Pz has divided into Ps, and Ss(C) — secondary ectoderm, a, b, primary
ectoderm of right side, a, /?, of left side. 5. The endoderm cell has been formed
and has divided (Ei, Ez). Ps has divided into P-i, the primordial germ cell, and
S*(d), tertiary ectoderm. 6. Ventral view at the beginning of invagination.
Elimination of chromatin in S*(D). The four endoderm cells (E) beginning to
invaginate. On each side two mesoderm cells (M) in which granular chromo-
somes may be seen, and two stomodseal cells (Sf).

may occur even though interrupted by the removal of parts,
by pressure, etc.

CLEAVAGE

255

TABLE OF THE CELL LINEAGE OF ASCARIS

Modified from Zur Strassen
(Letters in parenthesis are the notation of Zur Strassen, Boveri and others)

a1

(al)

f '

a2

A

(a)

(all)

(A)

a1

a

(al)

- 9

AB

(0° [ (all)

fbl

^ \)

J (bl)

b ] b2

B

(b) [ (bll)

(B)

6'

b J (£1)

c

f (MSt)

' c1 (rt.)
(mst)
c2(left)

c11 mesoblast I (posterior).
c12 stomodseal cells (anterior).
c21 mesoblast I (posterior).
c22 stomodaeal cells (anterior).

C

(EMSt) *

(E)

c1 (rt.)
(El)
c2(left)
(EII)

c11 (rt.)

c12 (left) T
c21 (rt>) endoderm I.

1^ c22 (left)
r d11 }

CD <

d1

,12 > ectoderm II.

1 (P,)

[ (S8)

(c)
d2 <
/•jA

' d21 1
,22 > mesoderm III.

((c))

d1 <

, > mesoderm II.

D \

I rfl2 /

( (P2)

d
(P.)

(D4)
d2*

f d21 , ,. f d211 (left)
( (ant.) | rf212 v(rt ' prime

^ , . . / d221 (left) ge.r.m
(QI) (post.) | d222 (rt0 cells.

* Primordial germ cell.

The study of cleavage from this point of view discloses the
fact, of the utmost importance in development, that blasto-
meres may be individually and specifically recognizable as
morphological and morphogenetic units. They bear much the
same relation to the whole cell group that the organs and tissues

256

GENERAL EMBRYOLOGY

bear to the embryo or later organism. As distinct morpho-
logical and physiological units they represent real differentia-
tions at a very early stage of development, and may truly be
said to form embryonic rudiments of structures appearing later
in the form of germ layers or derivatives of these.

FIG. 122. — Diagrams illustrating the value of the quartets in three animals.
From Wilson, "Cell." Ectoplasm is unshaded; mesoplasm is dotted; endoplasm
is vertically ruled. A. The Poly clad, Leptoplana, showing mesoplasm formation
in second quartet. (Compare Planocera, Fig. 120, where Surface finds meso-
plasm in cell 4d descendants.) B. The Gasteropod, Crepidula. C. The
Pelecypod, Unio.

Not only this but comparison of the cell lineages of different
classes and phyla often brings out the fact that particular
cells can be identified and compared in diverse groups of ani-
mals, making it possible to apply the idea of homology to blas-
tomeres and groups of blastomeres in the early embryo, as well

CLEAVAGE

257

as to the organs and parts of the fully formed organism. Cells
may be vestigial, rudimentary, and the like, in the same way
that organs may be. The three or four successively formed
quartets of micromeres or even an individual cell, for example
that known as 4d, can be identified and homologized both in
origin and in fate, in the phyla Platyhelminthes, Annulata,
and Mollusca (Fig. 122). Wilson has written ("The Cell,"
etc., page 416): "Thus we find that the cleavage of polyclades,
annelids and gasteropods shows a really wonderful argeement
in form, yet the individual cells differ markedly in prospective
value. In all of these forms three quartets of micromeres are
successively formed according to exactly the same remarkable
law of alternation of the spirals; and, in all, the posterior cell
of a fourth quartet lies at the hinder end of the embryo in
precisely the same geometrical relation to the remainder of the
embryo; yet in the gasteropods and annelids this cell gives rise
to the mesoblast-bands and their products, in the polyclade to a
part of the archenteron, while important differences also exist
in the value of the other quartets." (It should be added that
in the Polyclad, Planocera, the particular cell mentioned gives
rise to mesoblast also.)

Such conditions also illustrate how the facts of embryology
may have a certain value, often very great, as evidence upon
phylogenetic problems.

Often these similarities of structure can be carried back into
the pre-cleavage stage, and in the uncleaved zygote or ovum
before fertilization, substances can be identified which later
become contained within restricted groups of similar cells.
So that cleavage is in part to be regarded as a process by which
specific substances or regions of the egg become segregated in
different regions of the embryo, where each continues its
normal differentiation during later developmental stages and
gives rise to specific tissues or organs. In other words the
process of differentiation is not limited to the later stages of
development following cleavage; it occurs during and even
preceding cleavage. In AscariSj for example, each of the two
cells resulting from the first cleavage is specific; in other

258 GENERAL EMBRYOLOGY

forms each of the four or eight cells has already become differ-
entiated. In forms like the Ascidians (Conklin) true differ-
entiation has commenced in the uncleaved ova and has been
carried to a very pronounced degree. And it is not at all
unlikely that these differentiations of the undivided egg may
represent the most essential and most fundamental differen-
tiations of the organsim. This aspect of cleavage cannot be
discussed satisfactorily here without encroaching widely upon
the subject of the next chapter and it is therefore left at this
point.

It is to be noted, however, in conclusion that while cleavage
may have an important chemical significance of general char-
acter, and a general physiological significance, yet the process
is primarily a specific process of development and not mere
cell multiplication. The process is closely related in all its
details to the structure of the egg and also to the structure of
the adult, since this too is similarly related to egg structure.
Many of the details of cleavage do not occur according to phys-
ical laws based upon space and time relations of the parts,
but departures from what we should expect on the basis of
such laws may result, (a) from the historical factor of the
relationships of organisms and the process of descent, (6) from
the teleological factor, for cleavage has a prospective signifi-
cance, looking forward, as well as a retrospective significance —
cleavage has its promorphology as well as its morphology.

REFERENCES TO LITERATURE

BALFOUR, F. M., (See ref. Ch. III.)

BEARD, J., (See ref. Ch. III.)

BOVERI, T., Die Polaritat von Ovocyte, Ei und Larve des Strongylocen-
trotus lividus. Zool. Jahrb. 14. 1901.

BRESSLAU, E., Beitrage zur Entwicklungsgeschichte der Turbellarien.
I. Die Entwicklung der Rhabdocolen und Alloicolen. Zeit. wiss.
Zool. 76. 1901.

CONKLIN, E. G., (See ref. Ch. II, III.)

DRIESCH, H., Entwickhmgsmechanische Studien. VIII. Ueber Varia-
tion der Mikromerenbildung. (Wirkung von Verdiinnung des
Meerwasser). Mitt. Stat. Neapel. 11. 1893.

CLEAVAGE

259

GODLEWSKI, E., Plasma und Kernsubstanz in der normalen und der

durch aussere Factoren veranderten Entwicklung der Echiniden.

Arch. Entw.-Mech. 26. 1908.
HARGITT, C. W., Some Problems of Coelenterate Ontogeny. Jour.

Morph. 22. 1911.
HERTWIG, 0., Das Problem der Befruchtung und der Isotropie des Eies,

eine Theorie der Vererbung. Jena. Zeit. 18 (11). 1885.
KORSCHELT UND HEiDER, Lehrbuch, etc. . Ill Abschnitt. Furchung

und Keimblatterbildung. Jena. 1909.

LILLIE, F. R., Adaptation in Cleavage. Woods Holl Biol. Lect. 1899.
MASING, E., (See ref. Ch. V.)
MOENKHAUS, W. J., (See ref. Ch. II.)
ROBERT, A., Recherches sur le developpement des Troques. Arch.

Zool. Exp. (III). 10. 1903.

RUCKERT, J., Ueber das Selbstandigbleiben der vaterlichen und mutter-
lichen Kernsubstanz wahrend der ersten Entwicklung des befruch-

teten Cyclops-Eies. Arch. mikr. Anat. 45. 1895.
SACHS, J., Ueber die Anordnung der Zellen in jiingsten Pfianzentheilen.

Arbeiten Bot. Inst. Wurzburg. 2. 1882.
SELENKA, E., Die Keimblatter der Echinodermen. Stud, liber Entw.

II. Wiesbaden. 1883.
SURFACE, F. M., The Early Development of a Polyclad, Planocera

inquilina, Wh. Proc. Acad. Nat. Sci. Philadelphia. 1907.
WATASE", S., Studies on Cephalopods. I. Cleavage of the Ovum.

Jour. Morph. 4. 1891.
WILSON, E. B., The Cell-Lineage of Nereis. Jour. Morph. 6. 1892.

Cell Lineage and Ancestral Reminiscence. Woods Holl Biol.

Lect. 1899. (See also ref. Ch. II.)
WILSON, H. V., The Embryology of the Sea-Bass. (Serranus atrarius.)

Bull. U. S. Fish Com. 9. 1889. (1891).
ZELENY, C., Experiments on the Localization of Developmental Factors

in the Nemertine Egg. Jour. Exp. Zool. 1. 1904.
ZIEGLER, H. E., Experimented Studien uber Zelltheilung. III. Die

Furchungszellen von Beroe ovata. Arch. Entw.-Mech. 7. 1898.
ZUR STRASSEN, 0., Embryonalentwickelung der Ascaris megalocephala.

Arch. Entw.-Mech. 3. 1896. Ueber die Lage der Centrosomen

in ruhenden Zellen. Arch. Entw.-Mech. 12. 1901.

CHAPTER VII

THE GERM CELLS AND THE PROCESSES OF

DIFFERENTIATION, HEREDITY, AND

SEX DETERMINATION

THE_ problem^ of heredity is^La^prcJaknL-Qf development.
tnde^^ concerned primarily with the com-

parison of ^hejtrajts of adult organisms as they appear in suc-
cessive_generationst jsnid with the methods of the distribution
of distinctively individual parental characteri«tif*a ^rn^g
successive generations of offspring. The student of develop-
ment is concerned primarily with the genesis of the traits of
the individual, with that continuous and orderly sequence of
changes that gives to the single-celled zygote the final form of
the fully matured animal.

It might seem, therefore, that any consideration of the
problems of heredity is somewhat out of place in an account
of the processes of development. At one time this might
justly have been urged. But to-day the students of heredity
and of embryology have in common much that is fundamental.
Their interests meet in the ideas that the organism is a specific
creature at every stage of its existence, from zygote to adult;
that the qualities of each later stage are conditioned by those
of an earlier; and so ultimately the structural and functional
differentiations of the adult must be traced back to correspond-
ing differentiations of the zygote, or even to pre-conjuga-
tion phases of the gametes. It is the common endeavor of
the students of embryology and of genetics to answer the
question why the egg of a star-fish develops into a star-fish,
with the characteristics of its parents, rather than into a sea-
urchin, although it may develop in the same dish of water
with other eggs that do develop into sea-urchins.

260

DIFFERENTIATION, HEREDITY, SEX 261

Heredity is the fact of resemblance between offspring and
parents, not the resemblance of adult stages alone, but the
likeness at all corresponding ages. That the individual ova of
Asterias vulgaris are alike, that the cleavage processes, blastulas,
gastrulas, larvae, and adolescent stages of all the individuals
of this species are essentially alike, in structure and in behavior
— all this is similarly the fact of heredity. A specific kind
of protoplasm is never, whatever its form, anything other than
that specific kind.

In other words, the interest of the embryologist in the prob-
lems as to why the ovum develops as it does, passes from one
condition to another as it does, and finally produces the kind
of adult that it does, is essentially an interest in the problem
of heredity — the problem of organismal specificity. This is
the central point of embryological study. Consequently we
are fully justified in considering in this place, these general
problems of the relation of the facts of cytology and embryology
to the facts of parental and specific likeness of organisms.
Failure to do so would mean the omission of the most vital
topic around which much, perhaps it would not be going too
far to say most, of recent embryological investigation has
centered, and upon which it is to-day focussed.

The answers which the facts of embryology have to offer to
these fundamental questions are still rather vague and uncer-
tain. Most of them are stated in the potential mood and must
still be framed as hypotheses. But although the facts here may
be much clearer than their significance, we must attempt a state-
ment of both ; and it goes almost without saying that our treat-
ment of both must be as brief and as elementary as possible;
this is not the place for extended consideration of hypothetical
views, however great the importance of the central ideas.

We may address ourselves, therefore, to a survey and brief
analysis of the answers which have been given to the question
why the organism develops in the way it does. First, let us
recall the idea, stated briefly in the introductory chapter, that
development is a form of behavior — a series of reactions. In
any organic reaction the two factors of external and internal

262 GENERAL EMBRYOLOGY

conditions are involved. The reactions of the ovum in cleaving,
of the blastula in gastrulating, and the like, are of a general
nature, i.e., the external conditions involved may be consider-
ably varied, within limits, and yet produce the same response
on the part of the organism. Emphasis thus is placed upon the
internal factors in the reactions of development, for on account
of the definite character of habits of life and of spawning, the
normal external conditions of development are sufficiently
uniform to produce a series of reactions, a development, which
is also uniform, i.e., normal. Of course it is easily possible to
alter, artificially, the external as well as the internal conditions
of development and the results of such alteration often lead,
as we shall see, to important ideas regarding normal or char-
acteristic embryonic behavior.

Development is, then, a series of reactions, one condition
leading to the next; and the primary factor in determining the
quality of each reaction is the internal condition or structure,
both morphological and physiological, of the organism, whether
it be ovum, zygote, blastula, larva, or adolescent individual.

Throughout the efforts to solve the problem of individual
development, the attempt has always been to explain existing
differentiations as being dependent upon some preexisting
differentiation, related but of a different kind. Thus at one
time the earliest differentiations that became visible in the
developing organism were the germ layers, and these were
consequently regarded as the fundamental differentiations
of the embryo, determining its subsequent history. Next,
differentiations among the cleavage cells were noted and
emphasized as primary. Then as technique improved, and
the subject of cytology developed, attention became focussed
upon the nucleus and its organs not only as the centers of cell
life, but as the structures primarily concerned in the differen-
tiations of the developing egg cell. And lately, chiefly as a
result of experimental analysis of the processes of development,
rather than as the consequence of observation alone, structural
differentiations of the cytoplasmic portion of the ovum have
occupied the center of interest as determining factors in

DIFFERENTIATION, HEREDITY, SEX 263

development. This succession of views represents rather the
order of their general acceptance as working bases, than of
their discovery and individual promulgation.

In accordance with this historical succession of ideas regarding
the nature of the underlying differentiations in development,
we may outline briefly three general hypotheses of the causes of
differentiation. We may omit, as being now of historical
interest chiefly, any further reference to the germ layers as the
primary determiners of development, and may begin with the
idea of Pfliiger, that the egg and the blastomere group are
homogeneous or isotropic throughout, and that the early
developmental processes of cleavage are nothing more than
indifferent multiplications of similar units, resulting in the
formation of blocks out of which later differentiating structures
may be built. During the early J90's this view was quite
prevalent, and especially favored by such embryologists as
Oscar Hertwig and Driesch, who developed it somewhat farther
into what has been termed the "cell interaction" hypothesis.
According to this hypothesis, while the cells of the blastomere
group are essentially similar and equivalent in their poten-
tialities ("prospective potency/' Driesch), differentiation exists
among them by virtue of their relative position in the cell
group — not through any actual, individual, intracellular
differentiation. Stated in the words of Wilson (The Cell, etc.,
page 415), two sentences of Driesch summarize this view as
follows : " The blastomeres of the sea-urchin are to be regarded
as forming a uniform material, and they may be thrown about,
like balls in a pile, without in the least degree impairing thereby
the normal power of development." "The relative position
of a blastomere in the whole determines in general what develops
from it; if its position be changed, it gives rise to something
different; in other words, its prospective value ["prospective
significance," Driesch] is a function of its position."

In itself, then, the cell interaction hypothesis offers no
explanation of differentiation or development, for it throws
back upon some unknown factor the real cause of differentiation
through position. Later investigation, chiefly experimental, has

264 GENERAL EMBRYOLOGY

shown clearly that cell interactions alone play but a small
part indeed in the process of differentiation, and has led to the
search for that underlying factor or group of factors. And
while Driesch himself concludes that no explanation is possible
in known terms of matter or energy, and relies upon an unknown,
and therefore metaphysical, factor, the great majority of em-
bryologists believe that the question is still susceptible of
further scientific analysis. We find two general hypotheses
regarding the nature of the causes or conditions of differentia-
tion, and since differentiations are always specific we may speak
of these as also hypotheses as to the causes of those resemblances
among generations of organisms which we call in a word,
heredity.

The first of these is the hypothesis of "germinal localization"
or "germinal, organ-forming regions" associated primarily
with the names of His, Lankester, and Whitman. The essen-
tials of this hypothesis in its present form may be stated as
follows. The cytoplasm of the ovum before development (i.e.,
cleavage) begins, has a definite structure or morphology of its
own, such that particular regions or substances, by effecting
specific developmental reactions, are seen to correspond with,
or to lead to the formation of, particular tissues -or structures
of later stages and of the fully developed organism. The
cytoplasm is conceived as a mosaic-work of physiological units
which have not only a definite morphology but a definite pro-
morphology, looking toward the structure of the mature individ-
ual. This immediately suggests the old idea of "preforma-
tion" but it omits, of course, the naive crudities of this con-
ception and rests upon the idea that certain structures and
regions of the egg cytoplasm are in direct genetic relation with
corresponding structures and regions differentiating later by a
true process of development or epigenesis. These germinal
structures have specific reference, but not resemblance, to the
parts of the mature organism. This predetermination may be
only general to begin with, but it becomes more complete and
specific as one condition succeeds another, the cytoplasmic
structure of the ovum representing in the beginning all there

DIFFERENTIATION, HEREDITY, SEX 2G5

is of definite, specific, organismal structure. This cytoplasmic
structure of the germ is regarded as continuous from one
generation of ova to the next, through the germ within the
organism, and it thus serves as the physical basis of heredity.

As more or less opposed to this conception we have a group
of hypotheses which may collectively be termed the hypothesis
of " nuclear analysis." Here the nucleus alone is regarded as
that part of the germ or zygote which bears a specific relation to
later differentiations, and within the nucleus, the chromosomes
are the elements chiefly concerned. Chromosomes are supposed
to possess a structural predetermination that is promorpho-
logical; they are unlike and individually behave specifically
in determining the characteristics of developmental reactions.
This hypothesis, in its various forms, is associated chiefly with
the names of Nageli, Roux, Weismann, DeVries, and Oscar
Hertwig. It will be recognized as also preformational in its
essentials;- specific configurations of chromatin represent
potentially, corresponding embryonic and adult traits, which
become actual by a truly epigenetic series of developmental
reactions.

It has been pointed out frequently that this hypothesis really
transfers the idea of germinal localization from the cytoplasm
to the nucleus; the structure of the cytoplasm is regarded as
real, but as secondary and dependent upon the primary
structure of the nuclear elements. These nuclear organs, the
chromosomes, are thought to maintain a specific physiological
continuity from one generation of ova to the next and thus
to constitute a real physical basis of heredity.

These two general hypotheses have in common the idea of a
fixed promorphological structure within the germ which becomes
expressed epigenetically. They differ as regards the particular
part of the germ whose promorphology is to be regarded as
primary, and yet it is quite possible that both hypotheses
contain elements of truth. It remains now for us to review
some of the more significant facts of development bearing
directly upon these ideas in order to determine, perhaps which,
perhaps how much of each, is justified. In doing this we shall

266 GENERAL EMBRYOLOGY

not attempt to relate the evidence directly to either hypothesis,
leaving that for the reader; and in conclusion we shall attempt a
summary which may serve to bring the two hypotheses
together.

We may first describe certain facts associated chiefly with the
hypothesis of germinal cytoplasmic localization.

In the first place it is perfectly clear that the ovum does
possess a marked structure and organization, indicated in
several ways. Occasionally the ovum may show external
differentiations of form; such an egg as that of the squid
(Loligo, Fig. 113) or the fly (Musca, Fig. 47), is obviously not
only bilaterally symmetrical but it exhibits definite antero-
posterior and dorso-ventral differentiation. In a few instances
the eggs of a species are dimorphic, and while apparently the
nuclei of both kinds are identical in structure, the total
volume of one form may be three times that of the other. One
of the very important and highly significant factors in the
organization of the ovum is that of polarity, already described
(e.g., Figs. 42, 93). In many cases the polarity of the ovum
can be traced back into oogonial stages, where it is seen to
correspond with the polarity of the cells of the germinal epithe-
lium (Mark). Polarity pervades the whole structure of the
mature ovum and is expressed in & variety of ways, by the
eccentric position of the nucleus, by the point at which the
polar bodies are formed, by the disposition of the deutoplasm,
and by the arrangement and distribution of a variety of formed
substances, such as pigments, granules, and vacuoles of many
kinds.

Living protoplasm, as we have seen, consists of a fundamental
matrix or ground substance of rather uncertain form and com-
position, and suspended within this, particles and granules
of many sizes, forms, and materials, the nature of which gives
character to a particular region of protoplasm. These different
kinds of substance are not distributed at random through the
ovum, but they are localized in certain regions, as zones or
layers, either horizontal or concentric.

There are at present two views as to the relation of these

DIFFERENTIATION, HEREDITY, SEX 267

substances to the fundamental polarity and general organization
of the egg. In the opinion of some these substances are truly
to be regarded as the initial developmental differentiations of
the ovum. Each of these substances has a specific function
in development and leads to the formation of a certain tissue or
organ alone. Consequently these materials are known as
" organ-forming substances. " Before cleavage they usually
assume a very definite symmetry of distribution, closely
related to that of the later developing organism, and during
cleavage they become distributed among certain specific groups
of cells whose lineage can be traced directly to the rudiments
of certain organs.

Thus in the egg of the Ascidian, Cynthia (Styela), fully
described by Conklin and one of the best examples of this type
of structure, at the close of the first cleavage there are five
regions of protoplasm, present in amounts roughly proportional
to the size of the parts to which they later give rise, and
distinguishable by the character of their granular contents
(Fig. 92). At the animal pole is a superficial region of com-
paratively clear protoplasm, the ectoplasm, from which the
ectoderm develops; in the vegetal pole there is a dark gray
region, the endoplasm, rich in yolk and later forming the
endoderm; the mesoderm is formed from a crescentic region,
the mesoplasrrij located just below the equator, on the posterior
side; this is characterized by its content of yellow pigment, and
is divided into lighter and darker areas, forming respectively
the mesenchyme and the tail muscles (myoplasm); a light
gray crescent around the anterior border forms later the neural
plate and notochord (neuroplasm, chordaplasm).

These substances are arranged symmetrically with reference
to the first cleavage plane and this corresponds also to the
median plane of the larva and adult, and thus from the first
separates right and left sides of the body. The second cleavage
plane, at right angles to the first, separates the yellow meso-
plasm from the light gray neuroplasm and chordaplasm, and
the third cleavage, horizontal, separates the clear ectoplasm
from the other substances. The later cell lineage of this form

268

GENERAL EMBRYOLOGY

D

FIG. 123. — Changes in the structure of the egg of the Annulate, Chcetopterus,
during and after fertilization. From Lillie. A. Axial section through fully
grown oocyte, still within ovarian epithelium. Ectoplasm in upper two-thirds of
egg only. B. Axial section through primary oocyte, ten minutes after extrusion,
three minutes after fertilization. Ectoplasm has already flowed to the vegetal
pole, leaving an exposed area of endoplasm at the animal pole. Part of the o

DIFFERENTIATION, HEREDITY, SEX 269

is fully described by the same author and it is clear that each
of these substances becomes contained within the cells forming
the rudiment of the particular organ or tissue mentioned.

There are few other instances known where it is quite so
easily possible to distinguish the specific formed substances in
the egg. But in many ova, before cleavage it is possible to
distinguish several kinds of substance; thus there are known
in the eggs of certain Echinoderms four differentiated materials
(Lyon), three in Hydatina (Whitney), Dentalium (Wilson),
Physa, Lymncea (Conklin), four in Cumingia (Morgan), three
in the frog (McClendon), etc. (Figs. 42, 45, 86, 91, 123, 126,
129).

The term " organ forming" as applied to these materials does
not rest alone upon the observation of normal development,
but upon experimental grounds as well. For an example of this
kind of evidence we may return to the work of Conklin on the
egg of Cynthia. If one, say the right, of the blastomeres of
the two-cell stage is injured so as to prevent its further develop-
ment, the remaining blastomere develops as it would normally,
i.e. , into the left half of an embryo and larva, containing approxi-
mately one-half the number of cells found in the normal
organism at the corresponding stage (Fig. 124). Embryos
derived from the two anterior cells of the four-cell stage never
possess a tail or any muscle cells, while chorda cells and
neural plate cells differentiate normally, and the endoderm and
covering ectoderm are also typically formed (Fig. 125). Corre-
spondingly, embryos derived from the two posterior cells have
no chorda, nerve, or sensory cells, or gastral endoderm, but

endoplasm has also passed to the vegetal pole. The germinal vesicle has broken
down, and the maturation spindle is in the process of formation, between the
two asters. The residual substance of the germinal vesicle is clearly seen.

C. Axial section through secondary oocyte, thirty-two minutes after fertilization.

D. Longitudinal section, first cleavage; late anaphase. Posterior end toward the
left, anterior toward right. The ectoplasm of the polar lobe has been separated
from the remainder. E. Sagittal section through stage of about sixty-four cells.
The small upper cells are the apical cells. The ectoplasmic defect will be noted
in the posterior apical cell to the observer's right. A, animal pole; c, chromatin;
c.v., chromosomal vesicles of daughter nuclei; E, ectoplasm; e.a., e.b., e.c., endo-
plasm a, 6, and c. E.d., ectoplasmic defect; En, endoderm cells; m, mesoblast
cell; n, nucleolus; p.L, polar lobe; r.s., residual substance of germinal vesicle;
s.a., sperm aster; s.n., sperm nucleus; V, vegetal pole; X, derivatives of first
somatoblast.

270

GENERAL EMBRYOLOGY

E

FIG. 124. — The development of one of the blastomeres of the two-cell stage of
the Tunicate, Cynthia. From Conklin. The yellow material (see Figs. 91, 92)
is stippled; the boundary between clear protoplasm and yolk is indicated by a
crenated line.

A. Right half of eight-cell stage; posterior view. J5. Right half of thirty-cell
stage; dorsal view. C. Right half of forty-eight-cell stage; dorsal view. D.
Right half of sixty-four- to seventy-six-cell stage; dorsal view. E. Right half
of gastrula of about 220 cells derived from four-cell stage. The neural plate,
chorda, and mesoderm cells, are present only on the right side, and in their
normal positions and numbers. F. Right half of young tadpole; dorsal view.
Derived from four-cell stage. The notochord consists of a small number of cells
which are interdigitating ; muscle-cells and mesenchyme lie on the right side of
the chorda, but not on the left side, though the muscle cells have begun to grow
around to the left side. The neural plate is normal in position but not in form.
m'ch., mesenchyme; ms, muscle cells.

The cell nomenclature in this and the following figure, differs from that de-
scribed in Chapter VI. The right and left halves of the embryo are designated

DIFFERENTIATION, HEREDITY, SEX 271

consist of a mass of muscle and mesenchyme cells with a
double row of caudal endoderm cells, as in the corresponding
region of a normal larva. Equivalent results may be obtained
by injuring one or three cells of the four-cell stage (Fig. 125).

The work of Roux, Fischel, Wilson, and many others has
demonstrated similar localizations in the eggs of many forms —
the frog, other Ascidians, several Molluscs, Annulates, and the
Ctenophores, but we must limit ourselves to the mention of
only a few interesting details of the experiments on these
forms.

The Mollusca afford several very striking illustrations of
the effects of the removal of parts of the egg or of blastomeres.
The egg of Dentalium, as described by Wilson, has an upper
clear area which normally forms the ectoderm, a middle reddish
or brownish pigmented zone forming endoderm, and a lower
clear area which during cleavage forms a peculiar "yolk lobe"
or "polar lobe" (Fig. 126). When this yolk lobe is entirely
removed from the segmenting egg the development of the
remainder proceeds as if it were present, and a larva is formed
which lacks the apical organ and the entire post-trochal region
(for explanation of terms see Fig. 126), and which develops
later into an organism lacking those structures which would
normally have been formed from this part of the egg and larva,
namely, the foot, mantle, shell glands and shell, pedal ganglion,
and apparently also ccelomic mesoblast. Other Mollusca give
essentially similar results although of course not all possess a
yolk lobe; but removal of blastomeres is always followed by
absence of specific parts in later development (e.g., Wilson,
Crampton, Conklin) (Fig. 126).

The blastomeres of several species of animals fall apart, or
may be shaken apart easily, after a brief treatment with calcium-
free sea water, a fact discovered by Herbst and applied by him

by the same letters, those referring to the right side being underscored. A and B
refer, respectively, to the anterior and posterior hemispheres. After the third
cleavage, all cells lying on the polar body side of that cleavage plane are desig-
nated by lower case letters, while those on the opposite side of that plane continue
to be designated by capitals. The first exponent following a letter indicates the*
generation to which the cell belongs. The second exponent refers to the position
of the cell relative to the vegetal pole.

272

GENERAL EMBRYOLOGY

and many others to an analysis of this problem of localization.
The blastomeres of many Echinoderms, Molluscs, etc., can be
thus separated, and it is a remarkable fact that one of two, four,

ms.

v.end/

FIG. 125. — The development of blastomeres of the four-cell stage of Cynthia-
From Conklin. A. Anterior half-embryo derived from two anterior blastemeres-
The yellow crescent remains visible in the posterior, uninjured cells (B3). Sense
spots are present but the neural plate never forms a tube. The chorda cells lie
in a heap at the left side. There is no trace of muscle substance or of a tail.
B. Posterior half-embryo from the two posterior blastomeres. Dorsal view,
focussed deeply upon the double row of ventral endoderm cells in the mid-line,
a mass of mesenchyme cells on each side. No neural or chorda cells. C. Left
anterior quarter embryo from cell A; dorsal view. An invagination of the
ectoderm cells has the appearance of a gastrula, but is probably the invagination
of the neural plate. £>. Left anterior, and right posterior quarter-embryos,
from cells A and B; dorsal view. The former shows thickened ectoderm cells,
probably neural plate, around the endoderm cells; in the latter are eight muscle
cells and three caudal endoderm cells, m'ch, mesenchyme; ms, muscle cells;
n.p., neural plate; v.end., ventral endoderm.

eight, or even one of sixteen cells, continues to develop for
^ome time and forms those parts, and only those, which it
would have formed, had development of the entire cell group

DIFFERENTIATION, HEREDITY, SEX

273

Q

FIG. 126. — Development of the Mollusc, Dentalium, after removal of the
"polar lobe." From Wilson. A. Egg twenty minutes after extrusion, and
before maturation is completed, showing regional differentiation. B. Section
through egg one hour after fertilization, showing the beginning of the formation
of the polar lobe. C. Normal eight-cell stage, viewed from lower pole. The
polar lobe is the light part of cell D. D. Normal sixteen-cell stage viewed from
lower pole. The materials of the polar lobe are now contained in the cell marked
X. E. Sixteen-cell stage of egg from which the polar lobe was removed during
the first cleavage period. F. Normal trochophore of twenty-four hours. G.
Trochophore of twenty-four hours, developed from "lobeless" egg. H. Normal
larva of seventy-two hours, showing foot and shell. /. Seventy-two-hour larva
from "lobeless" egg. p, polar lobe.

274

GENERAL EMBRYOLOGY

FIG. 127. — Cleavage of isolated blastomeres in the egg of the Mollusc, Patella.
From Wilson. A-D, X 167; G, H, X 242; others X 208. A. Normal eight-cell
stage, viewed from upper pole. Fourth cleavage in progress. B. Normal thirty-
two cell stage, from side. C. The so-called "ctenophore stage" (normal) viewed
from upper pole. The primary trochoblasts are ciliated. Z>. Normal trocho-
phore of thirty hours, from left side. Body wall in section, prototrochal cells in
surface view. E. Second cleavage of an isolated micromere of the first quartet
(one of eight cells). F. Entire quadrant — products of first and second quartet
cells, being formed much as in the normal egg. G. Larva of twenty-four hours
from one of eight cells (micromere). From side, showing trochoblasts below,
apical cells above. H. Product of primary trochoblast isolated from sixteen-
cell stage. /. First division of isolated first quartet. J. Division of isolated
basal cell of eight-cell stage, showing typical arrangement of these four cells as
in the normal group of thirty-two. K. Larva of twenty-four hours, developed
from group like I, showing two secondary trochoblasts and two feebly ciliated
cells (? pre-anal cells), m, primary mesoblast cell; s.g, shell gland.

DIFFERENTIATION, HEREDITY, SEX 275

been occurring normally, although ultimately a normal larva
may be formed (Fig. 127).

The idea that these differentiated materials of the cytoplasm
really play the role of organ-forming substances in develop-
ment, is opposed by some (Lillie, Morgan, and others). The
opposed idea rests upon the experimental evidence that,
briefly stated, the really primary and fundamental organization
of the egg cytoplasm concerns the ground substance of the
protoplasm; the arrangement of the various formed stuffs
coincides with a similar and primary polarity and organization
of this fundamental protoplasmic matrix. This structure is
less manifest, but is really the factor which determines the
arrangement of the suspended cytoplasmic and deutoplasmic
granules and vacuoles. The correspondence between the
arrangement of these stuffs and the organ-forming substances
proper, is thus unessential, for the localization of the germ is
primarily a localization of the ground substance. In other
words, the varieties of material described by Conklin in Cynthia,
for example, are only secondarily related to the later differen-
tiation of particular organs or tissues, and their arrangement
is dependent upon the same primary factor that determines the
arrangement of the organs and tissues.

The evidence for this view is found chiefly in the results of
certain experiments upon the eggs of Choetopterus (Lillie) and
Arbacia (Morgan and Spooner). The granules which give
character to the various regions of the cytoplasm differ in
specific density, and consequently can be thrown, by centrifu-
gal force, into abnormal regions of the ovum. When this
is done normal cleavage and development may proceed, normal
with respect to the original polarity of the ovum and not with
respect to the new polarity as indicated by the altered arrange-
ment of the plasmas.

To illustrate, the egg of the sea-urchin Arbacia, contains four
different kinds of substance; one of these is distinguished by
the presence of bright orange or reddish pigment. In normal
development this substance lies toward the lower pole and
becomes localized in the lower quartet, so that when the micro-

27G

GENERAL EMBRYOLOGY

meres form here, they are composed of this material. The
micromeres, which later form the mesenchyme, always appear
at the pole opposite the micropyle (Fig. 109), which marks the
point of attachment of the ovum in the ovarian germinal
epithelium; this is also the point at which gastrulation com-
mences. The centrifuge brings about a stratification of these
substances which is independent of the polarity of the ovum,
since the ovum may assume any position with reference to the

FIG. 128. — Normal cleavage in the sea-urchin, Arbacia, following abnormal
distribution of egg substances by centrifuging. From Morgan and Spooner.
The figures are turned so that the pigment (dotted area) is downward. The
location of the cleavage planes, and the position of the micromeres, which always
mark the invaginating pole also, are independent of the induced stratification of
the egg substances.

axis of rotation of the machine. The pigmented protoplasm
may be thrown to any part of the cell. But Morgan has found
that the cleavage of eggs with abnormally distributed sub-
stances proceeds normally with reference to the original
polarity of the ovum and not according to the induced arrange-
ment. The micromeres, for example, continue to form
opposite the micropyle, and gastrulation occurs here, as usual,
although the pigmented protoplasm may occupy some remote
and unusual position in the cell (Fig. 128). Perfectly typical
larva? develop from such eggs, normal save in the distribution
of pigment. Development and differentiation thus seem to be
quite independent of the so-called "formative stuffs/' which
are, in such instances, evidently not " organ forming."

DIFFERENTIATION, HEREDITY, SEX

277

Several other forms are known to give similar results. One
of the clearest instances is to be seen in the Lamellibranch,
Cumingia, also described by Morgan. The egg -of Cumingia
contains, besides the clear protoplasm, three kinds of formed
substance, yolk, pigment, and oil. With the centrifuge these
can be thrown to any part of the cell whatever, and yet cleavage
and development proceed normally with reference to the

FIG. 129. — Normal development of the Pelecypod, Cumingia, following ab-
normal arrangement of the egg substances by centrifuging. After Morgan. The
pigment is indicated by stipples, the oil by small circles. A. Two-cell stage with
oil in small cell. B. Same with oil in large cell. C. Same with oil in both cells.
D. Normal trochophore, showing usual distribution of pigment and oil. E.
Trochophore with oil on oral side, and yet normal. F. Normal trochophore with
oil aboral and interior.

original polarity and not at all to the actual distribution of
these substances (Fig. 129). It should be noted that, although
this has not been definitely determined for Cumingia, the
Mollusca in general are excellent examples of determinately
cleaving eggs and the removal of parts of ova is followed by
definitely corresponding defects in embryo and larva.

Such experiments as these seem to indicate clearly that the

278 GENERAL EMBRYOLOGY

determining structure of the ovum is really that of the under-
lying protoplasmic ground substance, and that the arrange-
ment of the various formed substances coincides with this, is
determined by it in the first place, but is not always or neces-
sarily concerned directly in the later differentiations of the
ovum. The defects following removal of parts of the unseg-
mented ovum, or of blastomeres result therefore from the loss
of parts of this underlying structure, and not from the loss of
the formed materials or "formative stuffs/7 which in such
cases at least turn out not to be " formative."

Lillie has shown that carefully graduated centrifuging
reveals the existence in the egg of Chcetopterus, of certain
regional differentiations of the ground substance, indicated by
the differences in the ease with which the granules of various
sizes, and other structures, such as parts of the mitotic figure,
pass through it. These regions are not otherwise visible but
Lillie suggests that since they are undoubtedly real they may
represent or mark in some way the primary organization of the
cytoplasm.

According to this view of organization the term "organ-
forming substances" for the visibly differentiated substances
is a misnomer. For if these are removed to abnormal positions
within the cell they are then not related to the formation of the
same structures that they are in normal development.

At present it seems difficult, though not impossible as we
shall see later, to reconcile these two views as to the real seat
of the primary organization of the cytoplasm of the ovum; the
balance of evidence appears to favor the conception of organi-
zation as a condition of the fundamental ground substance of
protoplasm. But in any event it is perfectly clear that the
cytoplasm is organized definitely.

We should call attention in passing to the fact that many of
the results described above indicate that cleavage is not to be
regarded always as a developmental process of primary impor-
tance. Conklin has called attention to the fact that in Cynthia
the early cell boundaries do not always coincide with the limits
of the various kinds of cytoplasm. The determination of the

DIFFERENTIATION, HEREDITY, SEX

279

structure of the egg and the localization of these materials,
precede cleavage and are independent of it. Certain pressure
experiments to be mentioned shortly, illustrate the independ-
ence of cleavage and determination, and the centrifuging
experiments similarly demonstrate the independence of
cleavage and the distribution of the cytoplasmic stuffs. Indeed
Lillie describes the formation of a trochophore-like larva from
the egg of Chcetopterus in which cleavage had been artificially
prevented; this embryo formed external cilia and certain other

FIG. 130. — Development and differentiation in the absence of cell division, in
Chcetopterus. From Lillie. A, B, C. Ciliated, uninucleated unsegmented eggs,
about twenty-three hours old. The vacuoles are about in the position of the
prototroch of the larva. D. Ciliated unsegmented egg about twenty-eight hours
old; most of the endoplasm has been consumed, e, endoplasm.

differentiated structures in the complete absence of cell divisions
(Fig. 130). The normal processes of development are varied
and more or less independent of each other, while having
common reference to some general underlying condition.

We must now consider the facts of development in a consider-
able group of eggs which do not show any such results as those
described in the foregoing pages. Although these eggs possess
more or less differentiated regions of cytoplasm, yet removal of

280 GENERAL EMBRYOLOGY

parts or of blastomeres is not followed by any structural defect.
For example, the blastomeres of many Coelenterates (Haeckel,
Zoja, Maas, Wilson) may be separated when in the two-, four-,
eight-, or even, in some cases, in the sixteen-cell stage, and
from such isolated blastomeres typically formed embryos and
even free-swimming Iarva3 develop, normal in every respect
save that of size, being respectively approximately one-half, one

FIG. 131. — Normal development of one of the blastomeres of the two-cell
stage of the Hydroid, Clytia flavidula. After Zoja. A. Two-cells. B. Four-
cells. C. Eight-cells. D. Blastula. E. Young polype.

fourth, one-eighth, or one-sixteenth the normal size (Fig. 131).
This is true to a certain extent also of some of the Teleosts and
of Amphibians, the Nemerteans, and Echinoderms (Figs. 132,
133); in the last named forms even portions of the blastula or
gastrula (Driesch) may give rise to normal but diminutive
Iarva3 (Fig. 134). It seems very apparent that if cytoplasmic
localization occurs at all in such cases, it must be of a very
different kind from that described above.

This is an appropriate place to mention certain experiments
of a different kind bearing upon this same problem. Eggs

DIFFERENTIATION, HEREDITY, SEX

281

may be subjected, during their early cleavages, to deforming
pressure so that the planes of cell division appear in abnormal
relations to one another and to the egg as a whole (Hertwig,
Born). Thus in the sea-urchin (Driesch) the blastomeres of
the eight-cell stage instead of forming a spheroidal group may
be forced into the form of a flat plate (Fig. 135). When
released from the pressure such eggs form perfectly typical

FIG. 132. — Gastrulae and plutei from isolated blastomeres of the sea-urchins,
Echinus (A-D), and Sphcerechinus (E-G). After Driesch. A. Gastrula from
entire egg. B. Gastrula from one blastomere of the two-cell stage. C. Gastrula
from one blastomere of the four-cell stage. D. Gastrula from one blastomere of
the eight-cell stage. E. Normal pluteus. F. Pluteus from one blastomere of
the two-cell stage. G. Pluteus from one blastomere of the four-cell stage.

larvae. And even in a form like the Annulate, Nereis, whose
cleavage is determinate and whose blastomeres are highly
differentiated, Wilson has found that when the egg, subjected
to pressure, became divided by vertical planes into a flat plate
of eight cells, each one contained substance normally found only
in the macromeres of the lower pole; when released these
eight cells divided into sixteen, eight micromeres and eight
macromeres, instead of into the normal twelve and four

282

GENERAL EMBRYOLOGY

respectively. And from these, normal larvae developed; the
eight macromeres developed as the normal four would have
done, although under normal conditions four of the eight cells
and nuclei would have formed the first quartet, giving rise to
the apical nerve cells and anterior band of ciliated cells.

Furthermore, the experiment of bringing about the coales-
cence of parts of two eggs, or even of two complete eggs, has

FIG. 133. — Four normal but diminutive plutei from the isolated blastomeres
of the four-cell stage of the sea-urchin, Strongylocentrotus. After Boveri. A, B.
in oral view. C, D, lateral view.

been accomplished with Ascaris (Sala, Zur Strassen) and with
the sea-urchin (Driesch). The result is again the development
of a normal larva, of very large size when two entire eggs are
fused (Fig, 136). Even when two blastulas coalesce the final
result may be a single larva, though with some doubling of
parts. One especially interesting point is that normal develop-
ment may result even though the parts of the two blastulas

DIFFERENTIATION, HEREDITY, SEX

283

may be of different species, in somewhat different stages of
development, and no micromeres included in the mass
(Garbowsky).

FIG. 134. — Normal but diminutive larvae of Echinoderms, derived from por-
tions of gastrulse. From Jenkinson, after Driesch. a. Normal pluteus of
Sphcer echinus, b. Pluteus of same from portion of gastrula. c,e. Normal
bipennaria of Asterias glacialis. dj. Bipennaria of same, from vegetative half
of gastrula. g. Larva of Asterias with typical three-parted gut, but no ccelom,
from vegetative half of gastrula, removed after development of the coelomic sacs.

While such results as these are at first sight opposed to the
hypothesis of germinal localization, yet it is quite possible to
reconcile the differences between such extreme forms as the
Echinoderms, where one of eight or sixteen cells finally forms a
typical larva one-eighth or one-sixteenth normal size, and the

GENERAL EMBRYOLOGY

Ascidian, where one of four, eight, or sixteen cells gives rise, not
to a complete diminutive larva, but to a group of differentiated
tissue cells of the same kind that would normally have been
formed from the particular cell, had it remained in situ in the
normal group.

The discordance of these results may have one of two mean-
ings. First, it may mean that in such eggs as those of the
Echinoderms and Amphioxus a process of regeneration or

C D E

FIG. 135. — Cleavage in the egg of the sea-urchin, Echinus micro-tuberculatus,
under pressure. From O. Hertwig, after Ziegler. A, B. Eight- and sixteen-cell
stages. C. Sixteen-cell stage preparing for division. D. Thirty-two-cell stage,
in the form of a flat plate. E. Thirty-two-cells preparing for next division.
Crosses mark cells in which the spindle is vertical or oblique, to the plane of the
cell group.

regulation goes on. And, just as many adult organisms are
easily capable of restoring or regenerating lost parts, so the
embryo or even the ovum may have the property of reforming
parts artificially removed. This process has received the
special term of post-generation (Roux). Such a possibility is
indicated by the classic experiment of Roux upon the egg of the
frog. Here, if one of the two blastomeres is destroyed the

DIFFERENTIATION, HEREDITY, SEX

285

remaining one, if undisturbed, develops into a half-embryo;
but if the egg is inverted after the injury of one blastomere,
then during the consequent rearrangement of the substances of
-the uninjured cell, through the action of gravity, the organi-
zation is restored to the normal and a small normal embryo
subsequently develops. This shows that the uninjured half of

FIG. 136. — Fusion of Echinoderm larvae. A, B. Sphcerechinus. After Driesch.
C, D. Psammechinus miliaris. After Garbowski. A. Normal gastrula. B.
Single gastrula derived from the fusion of two normal blastulse, showing single,
large gut and doubled spicule. C. Normal stage of thirty-two-cells. D. Organ-
ism formed by the coalescence of parts of two organisms in different stages. The
cells ruled obliquely were part of an eight-cell stage, stained intravitally with
neutral red. The remaining cells were part of a normal thirty-two-cell stage.
In both C and D the stippling marks the cells derived from the vegetative half
of the egg.

the egg does possess the potentiality of developing as a com-
plete egg.

Or second, the contrast between the two extreme cases
mentioned may mean that localization results from a progres-
sive process of true development. Of course, in all organisms,
sooner or later, groups of cells become specifically differentiated
as particular tissues and organs or parts of organs. And
similarly there comes a time in the history of any cell group

286 GENERAL EMBRYOLOGY

when, once started on its course of differentiation, return or
^differentiation in another direction is impossible.

The formation of localized germinal areas of cytoplasm is to
be regarded as a process of development, and in the eggs of
different species this process may be carried forward at rela-
tively different times with respect to fertilization, cleavage, and
other early developmental phases. The most important steps
in cytoplasmic localization of the germ may be completed while
maturation and fertilization are going on, prior to the first
cleavage (Ascidians); or localization may be accomplished
during cleavage (Cerebratulus), or not until the gastrula or post-
gastrula stages (Echinoderms).

This idea is not essentially different from that of post-genera-
tion in certain respects, for regeneration and regulation are
after all essentially processes of development, deferred develop-
ment. The two differ however in that, according to the former
view localization is really present throughout the early stages
and disturbances are followed by an active process of regula-
tion; according to the latter, localization is not determined dur-
ing the earlier stages and when it does appear, the parts of the
egg remaining after the removal or injury of parts, behave as a
complete and normal unit, no regulation being necessary.

There is evidence for both of these views, and both may be
true at the same time. The second appears to be the more
widely applicable. Regulation seems more likely to occur
during comparatively late phases of localization. Evidence of
regulation following the separation of blastomeres is afforded
by such eggs as those of the Echinoderms, where the isolated
blastomere continues to segment for a time as if it were part
of a normal cell group, but gradually its cell products assume
the characters of a typical whole group and finally give rise
to a normal embryo and larva. In other cases (Amphioxus)
separated blastomeres develop from the beginning like whole
eggs, and no regulation is necessary. The results of deforma-
tion by pressure also indicate that localization is subject to a
regulatory process which may occur even in a comparatively
late stage in cleavage.

DIFFERENTIATION, HEREDITY, SEX 287

That the " organization " of the cytoplasm results from a
progressive developmental process is clearly evidenced by the
experiments of Wilson, Yatsu, and Zeleny • on the egg of
Cerebratulus before cleavage. If portions of this egg are
removed before maturation has begun, while the egg nucleus
is still in the form of an intact germinal vesicle, no defects are
seen in the resulting larva. Entrance of the spermatozoon is
followed by maturation and a general rearrangement of the
substances of the cytoplasm, one result of which is the forma-
tion of a cap of clear protoplasm at the animal pole. Removal
of this substance prior to or during the first cleavage, often pro-
duces no later abnormality. The separated blastomeres of the
two-cell stage, however, while for a time cleaving like halves,
soon assume the character of wholes. Those of the four-cell
stage continue longer to behave like parts, even through the
blastula stage, although ultimately they may form typical
free-swimming larvae. The degree of defect corresponds in a
general way with the stage to which cleavage has progressed
at the time of separation. Larvse developed from eggs with-
out the upper quartet, which contains the clear protoplasm
mentioned, have typically formed enteron, but lack the apical
organ. Larvse from this upper quartet have the apical organ
but are without enteron. And the same is true when, in the
sixteen-cell stage, the upper and lower octets develop sep-
arately. Parts of the blastula continue to develop for a time
and form only the restricted cell groups to which they give rise
in normal development.

-Such facts seem clearly to mean that cytoplasmic germinal
localization may be complete in later stages, but incomplete or
absent in the earlier, that it is truly a process or result of devel-
opment and not a primary determiner of the course of develop-
ment, not a fixed thing persisting from generation to generation,
which might be regarded as the physical basis of heredity.

The conception of cytoplasmic localization as a progressive
process, i.e., as one factor or link in the chain of developmental
events, immediately raises the question as to what condition

288 GENERAL EMBRYOLOGY

then lies back of this, and determines the character of the pro-
gressive steps or reactions. This leads us directly to tha second
chief view as to the fundamental character of the specific
organization of the ovum, that is, to the hypothesis of "nuclear
analysis" or nuclear determination, and to this we may now
give our attention. To state them again, the essentials of this
hypothesis are, that the real germinal localization of the ovum
is to be sought in the nucleus, that the organization of the cyto-
plasm is preceded and its character determined primarily, by
the organization of the nucleus, that this organization is
continuous from one generation to the next and is so to be
regarded as representing the physical basis of heredity. Polarity
and other cytoplasmic differentiations, certainly exist in the
ovum, even before fertilization or cleavage, but the only struc-
tural differentiation of the ovum which is invariably marked
out at all stages of the organism's existence, is the differentiation
between nucleus and cytoplasm. And while not alone develop-
ment, but all the normal life processes of the cell are the results
of interaction between nucleus and cytoplasm, both being
essential, yet the action of the nucleus is primary and seems to
determine the particularity of the cell actions.

This general subject of nuclear determination is enormously
complex and has been the occasion for whole volumes; our
account of it must perforce be brief and therefore more or less
fragmentary and dogmatic.

The search for the underlying causes of development is in
part a search for elements or conditions that are comparatively
fixed and that remain continuous from generation to generation
through the individual waves of species life. Specificity is
continuous; are there structural elements or conditions corre-
spondingly fixed and constant, not having to develop anew in
each individual ontogeny? Are there structures in the germ
cells which determine the direction of development and thus
represent (using this word in a very broad sense) the organs and
parts of the developing embryo?

In the endeavor to answer these questions the nuclei of the
germ cells at once compel attention as containing organs whose

DIFFERENTIATION, HEREDITY, SEX 289

morphology appears to be constant and specific at all stages of •
the individual life history, and through successive generations.
Thus the chromosomes at once become the foci of observation
and discussion, and the hypothesis of nuclear determination
becomes, to a considerable extent, the hypothesis of the specific-
ity of the chromosomes.

This conception has already been outlined in Chapter II; the
chromosomes are believed to be differentiated functionally, in
a specific manner so that each chromosome of the nucleus
represents a center of activity of a particular character. That
is to say each chromosome, either individually or as a compo-
nent of a unified group, determines a specific form of reaction
with the cytoplasm, or rather influences in a particular way
certain of the reactions constantly occurring between nucleus
and cytoplasm. And the final result of these reactions is the
production of certain structural and physiological characteris-
tics of the embryo or mature organism. Thus, leaving out all
the intermediate chain of processes or reactions, there is an
actual correspondence between certain traits of the mature
organism and certain chromosomal characters of the gametic
nuclei. Of course the chromosomal characters determine only
the first step in the development of the corresponding trait; but
this in turn determines the next, and so on. And since the
quality of one step or reaction in development is determined by
the preceding, we are correct in relating directly the character
of the final steps in development with the factor that first
determined the trend of reaction.

Emphasis is thus placed upon the physiological character of
the relation between chromosome and later structure, and care
must be exercised constantly, in the discussion of this subject,
to guard against a conception of this relation which is too
strictly morphological, and which might suggest too strongly the
conception of development as preformational. A wrong inter-
pretation of the modern view of the chromosome relation leads
to a rather strict preformational view; but such an idea does
not to-day represent the hypothesis fairly. What is formed,
or preformed, in the germ is a certain arrangement or configura-

290 GENERAL EMBRYOLOGY

tion of the chromatic substance, which in its reactions with the
cytoplasm produces new and specific conditions, these lead to
others, and so on through development.

The conception of the determinative character of the chro-
mosomes must now be modified to include the idea that each
chromosome is not a simple unit, homogeneous either morpho-
logically or physiologically. Each chromosome is to be regarded
as made up of a series or group of elements which singly are
simple and homogeneous, and behave as physiological units or

determiners." These may or may not correspond with the
chromioles, or granules of chromatin, of which the chromosome
is composed; and while it is true that they have never been
positively identified as units or determiners, some such bodies
must be present in the chromosome according to this hypothesis.
Such determiners, although apparently necessary hypothetical
units cannot be described; they may prove not to be definite
particles at all, but rather dynamic relations, or configurations
of substance. So for practical and descriptive purposes we are
nearly limited to the chromosomes.

It is quite likely that the chromosomes may not be the only
factors in the determination of development, there may be a
whole series of factors back of these, and we know that a whole
series of factors follows after. But if they are proved to be
necessary links in a chain of determining factors, then they are
causes of differentiation, and if they are found to be the earliest
visible differentiations with which later differentiations some-
how correspond, then we may refer to them as the causes of
specific differentiation. At some future time it may indeed be
possible to push the analysis of the factors of differentiation still
farther back; such a possibility is in no wise excluded by the
chromosome hypothesis as it stands to-day.

One of the obvious requirements of any hypothesis of differ-
entiation and heredity is that it must readily allow interpreta-
tion, in cytological terms, of the enormously complex phe-
nomena of alternative or Mendelian heredity. Most of the traits
of an organism are the property of the species, common to all
the individuals of a specific group. But there are other charac-

DIFFERENTIATION, HEREDITY, SEX

291

ters that are family possessions and may or may not be inherited
by individuals. These individual characteristics are, in many
cases, comparatively late developments. The early characters
are those of the larger group; those of the species appear later,
and finally the family and individual traits. The whole sub-
ject of Mendelism has developed into an extremely complicated
system, in directions largely unforeseen. And yet it is hardly
too much to say that the cytology of the germ cells and their
nuclei has on the whole fairly kept pace, and it is in most in-
stances quite possible to parallel the facts of Mendelism with the
facts of chromosome behavior. We shall return briefly to this
subject in a more appropriate connection.

FIG. 137. — The structure of chromosomes. A, after K. C. Schneider, others
after Bonne vie. A. Nucleus from epidermis of Salamander larva, in telophase.
B. Prophase of first cleavage of Ascaris megalocephala bivalens. C. Nucleus
from cleavage stage of same. D. Interkinesis in Amphiuma.

Let us now repeat, from this particular point of view the
general ideas regarding the chromosomes mentioned in Chapter
II, emphasizing certain topics and adding a few details which
bear directly upon the relation of the chromosomes to the proc-
esses of development and heredity.

The chromosomes are not structurally homogeneous masses,
but are built up of certain granules which often have a definite
arrangement giving the chromosome as a whole a general
structure. This structure has been variously described (Fig.
137); in some cases it seems to be a cylinder of chromatin

292 GENERAL EMBRYOLOGY

granules with a core of differentiated substance, probably linin;
in other cases the granules are arranged in a linear series wound
in a definite spiral; but in still other cases the granules seem
to have little if any regular disposition. These granules are
very close to the limit of vision, indeed often they are invisible,
and in most cases it is difficult to make confident assertions
regarding then- arrangement. Of course their invisibility on
account of minuteness does not prove that they are not present.
Indeed many, following Weismann, postulate an elaborate series
of representative particles within the nucleus : the chromosomes
or "idants," are divided into "ids" or chromatin granules;
the ids are then assumed to be formed of groups of "determi-
nants/7 and these in turn are thought to be composed of the
really elementary, self-propagating protoplasmic units, the
"biophores." Of this series, the idants and ids are visibly known;
the determinants and biophores are invisible hypothetical
bodies, postulated to aid in relating many of the complicated
facts of heredity to certain cytological facts. The assumption
of a final determining unit that is, and seemingly must remain,
invisible has proved fortunate as affording a convenient shelter
against criticism, for such an assumption partly removes the
question from scientific treatment. We shall lose little and gain
much by considering here merely those elements of the nucleus
that can be identified and whose behavior can be traced to
some extent, i.e., the chromosomes and chromatin granules.

Our ideas in this field are, to a remarkable degree, the out-
growth of the pioneer work of Weismann. Although based
upon the properties of hypothetical units whose behavior was
outlined upon purely hypothetical grounds, his conceptions
of the relation between chromosome behavior and the facts of
development and heredity, formulated more than thirty years
ago, before the science of cytology was established, have a
distinctly modern aspect. The remarkable convergence of the
facts of heredity, of development, and of cytology, which have
become known subsequently to the formulation of Weismann's
hypotheses, constitutes splendid evidence of the keenness of
this great embryologist.

DIFFERENTIATION, HEREDITY, SEX 293

We may here suggest again the character of the evidence for
regarding the chromosomes of the germ nuclei of the very
greatest importance as factors controlling or directing the proc-
ess of development (i.e., the process of heredity), facts of such
constancy and universality that they must have some meaning.

First of all comes the fact of the very high degree of morpho-
logical constancy of these organs throughout the tissues of the
species, not merely of the individual. This constancy, always
considering corresponding ages of course, concerns their number,
size, and form, and proves them to be specific organs in a real
sense. They are, with a few easily explained exceptions, pres-
ent in pairs of similar elements, whose history can be traced
back to their derivation in groups of unpaired elements in the
male and female gametes. During mitosis the distribution of
the chromosomes to the daughter cells is never a haphazard proc-
ess, but the whole process of mitosis appears to be an adapta-
tion toward securing their equal division, and the distribution
to the daughter nuclei of groups of similar morphological
composition.

It is unnecessary to repeat here any of the evidence out-
lined in Chapter II for the idea of the genetic continuity
of the chromosomes from cell to cell. We saw there that it
is the chromatin granules which may be regarded as actually
morphologically continuous, and that these may seem to
become similarly associated every time that the chromosomes
visibly appear. What determines their constant association
in the reforming chromosomes of each cell generation is to be
answered only hypothetically if at all. But in spite of the
contrary belief held by some, the chromosomes may be regarded
as genetically continuous individual elements, although the
details of their composition may vary slightly from generation
to generation. And after all it may be that the most important
continuity is that of the chromatin granules, supposing them
to be qualitatively unlike and to play the role of specific de-
terminers.

The validity of the chromosome hypothesis has always been
strongly indicated by the essential facts of syngamy, namely,

294 GENERAL EMBRYOLOGY

the formation of the zygotic nucleus from equal contributions
of chromosomes from the male and female parents. This is
the only portion of the new organism which is derived equally
from both parents, and while it is true that a small amount of
cytoplasm accompanies the sperm nucleus in its entrance into
thfe ovum, this varies considerably in amount in different forms.
This latter fact together with the general fact of the primary
importance of the nucleus in all aspects of cell activity, com-
bine to enhance the significance of the equal derivation of the
chromosomes (Fig. 106), especially in view of the further fact
that on the whole the family and individual traits of organisms
'are inherited with equal likelihood from either parent.

The behavior of the chromosomes throughout the matura-
tion process affords many highly interesting and significant
parallels between chromosome behavior and the facts of hered-
ity. Interest centers here in the phenomena of synapsis
and the " reducing'7 divisions.

Any precise interpretation of these two phenomena seems
impossible until more is known with certainty regarding the
behavior here of the chromatin granules, but the phenomena
themselves are readily interpretable in the light of the facts of
alternative, or Mendelian heredity.

In synapsis we see the final union of pairs of chromosomes
introduced into a single nucleus at the time of fertilization, but
remaining distinct throughout the life of the hybrid generation,
until the time when the hybrid organism forms its gametes.
Synapsis is not a haphazard junction of chromosomes, but an
orderly union of elements of paternal and maternal origin,
similar in size, in details of form, and probably also in function.
The bivalent chromosomes thus formed are, in consequence of
their derivation from two individuals, not quite homogeneous
throughout. Following synapsis come two divisions of each
chromosome, and in most organisms one of these apparently
divides the chromosome equally, into two similar parts (equa-
tion division), while the other divides each of the daughter
chromosomes dissimilarly (reducing division), the dissimilarity
resulting from the relation of the plane of division to the plan

DIFFERENTIATION, HEREDITY, SEX 295

of arrangement of its dissimilar component granules. The
relation of the reducing division to the chromosome depends
upon the character of the synapsis, whether telosynapsis or
parasynapsis, and also upon the behavior of the chromatin
granules in all these events, and it is difficult to be certain of
this. It is safe to say, however, that in most cases each biva-
lent chromosome, composed in equal parts of substance from
each parent, clearly separates into four elements, two having
one composition, two another. These elements are then dis-
tributed to separate gametes, so that with respect to the com-
position of each separate chromosome, the gametes produced by
an organism are of two kinds, approximately equal numerically.
This accords perfectly with the facts of Mendelian heredity,
upon the supposition that there is a correspondence between
chromatic elements and organismal traits. This may be made
somewhat clearer with the aid of a diagram : see Fig. 80.

In the process of maturation, therefore, it is easily possible
to find a mechanism which permits the segregation of charac-
teristics in the germ cells and their distribution to separate
organisms in regular Mendelian ratios. One important corre-
spondence should not be overlooked. In Mendelian heredity
the individual qualities of the parents may not appear sepa-
rately until the first generation after the hybrids. This is
possibly related to the fact that the parental chromosomes
undergo synapsis and subsequent redistribution first in the
germ cells formed by the hybrid, and the segregated elements
are, therefore, distributed separately first in the organisms
formed from these hybrids, i.e., in the Ft generation.

The conclusion resulting from the study of Mendelian hered-
ity, that the organism is a sum of "unit characters" which
in the organism interact with one another, so as to produce
a physiological whole, but which in heredity are more or less
clearly separable units, affords strong evidence for the general
hypothesis of the representative particle composition of the
germ nuclei. Chromosomes might thus represent groups of
such " units" or in occasional instances perhaps, single units,
although this must be the case only rarely, for the total number

296 GENERAL EMBRYOLOGY

of unit characters is far in excess of the number of chromosomes.

That the chromosome of the Metazoan is really made up of a
group of unit determiners, is also indicated by the behavior of
the Protozoan nucleus in maturation. In most of the simpler
Protozoa where the maturation phenomena appear, there is no
indication of definite elements like chromosomes in the nucleus.
But in many of the Ciliates, in -which vegetative chromatin and
reproductive chromatin become sharply separated, the latter,
or idiochromatin, is seen to be formed into definite bodies. Thus
in Raram.cecium, as observed by Calkins and Cull (Fig. 82), the
micronucleus (idiochromatin) becomes resolved into a large
number — more than 200 — chromatin granules (idiochromidia)
whose definite behavior can be traced. Their behavior is com-
plex, but the result is that each idiochromidium is divided
longitudinally and transversely, and the resulting daughter-
bodies may, therefore, be dissimilar. After fertilization the
division of the zygotic nucleus brings about the division of each
chromidium and the distribution of the halves to the two
daughter cells.

The very large number of separate elements in these "gam-
etic" nuclei may indicate that each corresponds with a single
character, or with a smaller group of characters than in the
Metazoan, and that therefore the chromosome of the Metazoan
must be an enormously complex affair. All of this lends weight
to the idea that " chromosomes, the characteristic structures of
the nucleus in mitosis, have had an evolution no less surely
than has the nervous system, digestive system, or supporting
system of the higher animals, and that the chromosomes of the
protozoa have the same relation to the chromosomes of the
metazoa that the organization of the protozoan body has to
that of the metazoan, i.e., a unit structure." (Calkins, "Proto-
zoology," page 171). Admitting the representative particle
composition of the chromosomes, it must of course follow that
their evolution in the Metazoa, parallels the evolution of adult
form and structure.

If this is true, then the chromosomes of the Metazoan germ
cells must each represent a congeries of determiners, the form

DIFFERENTIATION, HEREDITY, SEX 297

of association of which might differ in different species, as widely
as the groups of characteristics of the adults differ.

The question as to just how the chromatic determiners (as-
suming their existence) really do affect the quality of the
reactions of the developing organism, is still practically un-
touched. To some it seems necessary to postulate the asym-
metrical distribution of the chromatin granules through suc-
cessive mitoses, so that certain kinds of granules or " deter-
miners" become distributed to certain cells and regions, directly
effecting there specific reactions. No such form of distribution
has been1 observed, though indeed it has not been sought in a
thorough fashion. In tissues whose differentiation is fairly
advanced there are certainly characteristic and specific nuclear
appearances which indicate that the nuclei as well as the
cytoplasms have undergone a real differentiation, but whether
this is related to chromosome or granule structure remains
undemonstrated.

If any such sorting out of determiners occurs it must be at
widely divergent stages of development in the various groups,
on account of the variety of the results in the way of specific
embryonic defect following the removal and pressure experi-
ments described above. Indeed the results of the pressure
experiments referred to, become highly significant from this
point of view, for it will be remembered that the presence of a
completely " foreign" nucleus may in some -cases not influence
the particular form of differentiation of the cytoplasm. To
say, in such cases as these and in the removal experiments, that
regeneration may occur and the proper determiners be reformed,
does not offer much that is helpful in the way of a solution of
this particular problem, for it would necessitate the assumption
of some mechanism back of the " determining " particles, by
which they themselves are formed and determined.

The fact that parts, even small1 bits, of a fully developed and
differentiated organism may finally, through a process of regu-
lation, give rise to a complete organism again, or that in many
plants, buds, bits of stem or leaf, may similarly give rise to a
completely formed organism capable of developing typical

298 GENERAL EMBRYOLOGY

germ cells, renders extremely unlikely any strictly morpho-
logical conception of the relation between strictly differentiated
germinal determiners and the formation of certain tissues or
organs. The ideas cannot be overemphasized or repeated too
often that, while the thing or the relation that we call a deter-
miner may sometimes have a morphological expression in the
germ, essentially the relation is physiological — functional —
probably chemical or energetic (dynamic), and that the reac-
tions or interactions of whole groups and masses of particles or
systems are involved in determining the intermediate and final
results of development.

One further group of observations must be considered in con-
nection with the possibility of the primary character of the
nuclear control of development and heredity. During the
process of development there occurs a constant giving off of
substance from the nucleus to the cytoplasm (Figs. 138, 32).
At every mitosis only a part of the chromatic substance is
formed into chromosomes, while the remainder passes into the
cytoplasm and dissolves, and of course the whole fluid content
of the nucleus, the nuclear sap, is discharged into the cytoplasm.
Herbst has emphasized the importance of the nuclear sap as an
important determining factor in development and heredity.

In many instances, substances discharged from the nucleus
into the cytoplasm of the oogonia, especially during their
growth period, are directly concerned in the formation of specific
materials and bodies of the mature ovum. And later in develop-
ment Conklin has observed that the cilia of the superficial cells
of Crepidula develop only when certain chromatic granules
reach that region, a single cilium then differentiating opposite
each granule. We have already mentioned the fact, described
by Wilson, that in Cerebratulus the effects of the removal of
parts of the egg cytoplasm before the germinal vesicle has
broken down, are very different from the effects of the removal
of similar portions after the contents of the vesicle have been
discharged into the surrounding cytoplasm during the initial
stages of maturation.

But the evidence for the hypothesis of nuclear determination

DIFFERENTIATION, HEREDITY, SEX 299

J5

—•V^ajJ"

FIG. 138. — A. Chromatin extrusion from the nucleus into the cytoplasm in
the oocyte of the Medusa, Pelagia noctiluca. After Schaxel. B. Extrusion of
chromatin into the cytoplasm during the maturation of the oocyte of Proteus
anguineus. After Jorgensen. X 1080.

300

GENERAL EMBRYOLOGY

is not altogether purely observational; there is some experi-
mental evidence as well, although largely indirect and possibly
none is positively conclusive.

For example, Boveri has described the results of dispermy in
the sea-urchin egg. When two spermatozoa enter the ovum the
result is frequently the formation of three or four centrosomes
and asters connected with one another by spindles, upon which
the chromosome groups are usually drawn in abnormal com-

FIG. 139. — Larva of Sphcer echinus, derived from a dispermic egg, showing
differences in nuclear size, distribution of pigment, etc. The dashed line marks
the separation of the two portions of the larva. After Boveri (reconstructed
from two figures), n, small nuclei; N, large nuclei; p, pigment.

binations, so that when such an ovum cleaves it separates
into three or four cells containing nuclei whose composition is
therefore abnormal. In many such cases Boveri finds that
each of the three or four cells forms a group of cell descendants
which can be identified by the presence or absence of certain
characters or by unusual combinations of characters (Fig. 139),
so that the entire embryo may be said to consist of three or
four regions, each with certain distinctive characteristics.
Furthermore, in some instances, the cells of the various fractions

DIFFERENTIATION, HEREDITY, SEX 301

may be characterized by unusually large or small nuclei, indicat-
ing the presence of larger or smaller amounts of chromatin
(numbers of chromosomes) than usual ; the microscopic examin-
ation of these multipolar spindles shows that the chromosomes
may be distributed with great irregularity in the first division.

More striking are the results following the separation of the
blastomeres of such dispermic eggs. The isolated cells of a
four-cell stage resulting from normal fertilization, develop
normally, producing four similar but small normal larva?
(Fig. 133). But the isolated cells of one of these three-cell
stages develop dissimilarly, each with certain defects; and just
as any possible combination of chromosomes may have occurred
in each of the three original cells, so all possible combinations
of characters are found in the larvae developing from such cells
when isolated. Boveri believes that this warrants the conclu-
sion that, while the presence or absence of certain chromosomes
may not result in the presence or absence of specific traits, yet
a certain combination of chromosomes is essential for normal
development, a fact which would mean only the physiological
specificity of the individual chromosomes.

Perhaps the most striking experimental results are those
obtained by fertilizing the eggs of one species, with the sperm
of another species, genus, or even phylum. In the first place,
Boveri in 1889 reported that non-nucleated egg fragments of
Sphcerechinus (one of the sea-urchins), fertilized with the sperm
of Echinus, developed into larvae exhibiting only paternal
characters. This appeared to afford strong evidence that the
characteristics of the nucleus rather than of the cytoplasm
determine the course of development. Later attempts (See-
liger, Morgan, Boveri) to confirm these facts led to incon-
clusive results. Indeed exactly opposed results were obtained
by several investigators (Driesch, Loeb, Godlewski, Hage-
doorn). Eggs of one species of Echinoderm fertilized with the
sperm of another species, genus or class, of Echinoderm, or
even with Molluscan sperm, resulted in the development of
larvse possessing wholly or largely the maternal characters.
These results indicated just as strongly that the nuclear com-

302

GENERAL EMBRYOLOGY

FIG. 140. — History of the paternal chromatin during the first cleavage in the
pseudohybrid sea-urchin, Sphcerechinus ? X Strongylocentrotus d\ After Herbst.
A. First cleavage figure. Sperm and egg pronuclei associated, but not fused.
Chromosomes beginning to form in the egg pronucleus. B. Chromosomes in
both pronuclei; in separate groups with separate spindles. C. Anaphase of
first cleavage. Maternal chromosomes reaching the poles. Paternal chromatin
(chromosomes no longer) forming an irregular mass, spun out on the spindle
between the maternal chromosome groups. D. Division completed. Daughter
nuclei reconstructed and consisting entirely of maternal chromatin. One of
the cells contains a small vesicle consisting of the paternal chromatin, which takes
no further share in cleavage. <?, chromatin of the sperm pronucleus.

DIFFERENTIATION, HEREDITY, SEX 303

position is of the lesser importance in determining the devel-
opment of specific traits, and of course seriously affected the
validity of the hypothesis of nuclear determination.

These experiments, however, curiously turned out to afford
very strong evidence in support of this hypothesis. For other
workers (Herbst, Kupelwieser, Bataillon) showed that in many
of these, and in other instances, the nucleus of the spermatozoon
did not actually fuse with the egg nucleus, but remained either
partly or wholly inactive, taking little or no share in the forma-
tion of the mitotic figures of the first and subsequent cleavages
(Fig. 140). The resulting larvaB therefore were not truly
hybrids; the spermatozoon had merely stimulated the egg to
develop, as in artificial parthenogenesis, but itself took no part
in the formation of the nuclear structures of the larva. In the
absence of microscopic examination of the embryo, therefore,
it is impossible to place any emphasis upon the development
of purely maternal or paternal characters under such conditions.

Fortunately such cytological evidence is now provided
extensively through the work of Baltzer, who has traced the
nuclear history of many forms of Echinoderm hybrids. It
appears that part or all of the paternal chromatin, never the
maternal, may be thrown put of the nuclei of such " hybrids "
(pseudohybrids). Such an elimination of paternal chromatin
may occur during the very first cleavage, or it may be delayed
until the blast ula or even early gastrula stage (Fig. 141). The
examination of a long series of hybrids, showing all degrees
of purity of the maternal characters, leads Baltzer to the con-
clusion that the degree to which paternal characters appear in
the resulting hybrids, is closely parallel to the relative amount
of paternal chromatin which is retained within the nuclei of the
organism. Where the fusion of the sperm and egg nuclei
remains complete, the hybrids have intermediate characters;
where little or no chromatin from the spermatozoon is retained
in the nuclei, there appear, chiefly or alone, maternal characters.
Only in the case of the fertilization of the eggs of a sea-urchin
(Strongylocentrotus) with the sperm of a Crinoid (Antedori) has
it been shown that the fusion of the germ nuclei really occurred

304

GENERAL EMBRYOLOGY

(Godlewski, Baltzer) while the larvae resulting from this cross
exhibited certain characters which were purely maternal; but
this result is wholly inconclusive as evidence opposing the hy-
pothesis of nuclear control.

C

FIG. 141. — History of the paternal chromatin in the pseudohybrids of the sea-
urchins, Strongylocentrotus $ X Sphcerechinus c?. After Baltzer. A. Cleavage
cell showing paternal chromatin (cT) outside the division figure. B. Early
blastula. C. Late blastula, showing the elimination of the paternal chromatin
in the irregular cells and spaces within the blastoccel (for normal blastula see
Fig. 109).

For it has been found, in the first place, that external condi-
tions often determine whether certain characters shall be
paternal or maternal in their qualities (Vernon, Tennent).
Under certain conditions of temperature, alkalinity, etc., the
larva may exhibit paternal resemblances, while under other
conditions maternal resemblances may appear in the same

DIFFERENTIATION, HEREDITY, SEX 305

cross. And in the second place, the phenomenon of " domi-
nance" appears even in these early stages of development, and
a hybrid may show certain clearly maternal characters and yet
in other respects closely resemble the paternal type (Stein-
bruck, Driesch, Boveri, Loeb and Moore). Great variability is
often the rule and frequently it is impossible to say whether
either parental trait really appears purely. It should be pointed
out, first, that it frequently happens in Mendelian inheritance
that true hybrids are either purely maternal or purely paternal
with respect to single traits, and second, that only after synap-
sis, which occurs in the germ cells of the mature hybrid organ-
ism, are the paternal and maternal chromosomes really brought
into complete relation.

On the whole, then, while there are some few results difficult
of favorable interpretation, we may say that the evidence from
hybridization, though at first distinctly opposed to the hypothe-
sis of nuclear determination, at present affords the strongest
support of this hypothesis, and indicates that normally the
characters of a hybrid are determined by both of the germ
nuclei, and that when nuclear material from only one parent is
functional the characters of the so-called hybrid are deter-
mined thereby.

We might mention one further possible interpretation of
some of the results opposed to this conclusion, and emphasized
by Conklin and others, namely, that in some cases, at least, the
fundamental or general traits of an organism may be deter-
mined immediately by the cytoplasmic structure of the ovum
alone or chiefly, while the nuclei are equally concerned in the
determination of the more particular specific or individual
traits, often appearing relatively late in development. Such
a possibility seems to be indicated by many of the facts of
germinal localization already described, and it may be that some
of the results indicated above, non-conformable with the
hypothesis of nuclear determination, point in the same direc-
tion. Conklin writes (Science, XXVII, 89-99) : " At the time of
fertilization the hereditary potencies of the two germ cells are
not equal, all the early development, including the polarity,

300 GENERAL EMBRYOLOGY

symmetry, type of cleavage, and the relative positions and
properties of future organs being predetermined in the cyto-
plasm of the egg cell, while only the differentiations of later
development are influenced by the sperm. In short, the egg
cytoplasm fixes the type of development and the sperm and egg
nuclei supply only the details." And yet we should not over-
look this fact, of basic importance, that these fundamental
cytoplasmic differentiations have resulted from interactions
between the cytoplasm and the nucleus of the oogonial cell,
and that the nucleus of the oogonial cell, and egg cell, is itself
originally derived in equal parts, from paternal and maternal
ancestry. And further many of the conditions of polarity,
symmetry, and the like, may in some cases be determined or
altered by the entrance and subsequent activity of the spermat-
ozoon within the ovum.

Probably altogether the most striking evidence in support
of the hypothesis under consideration is to be found in some
of the recent work upon the association of sex with chromosome
characters. The nature of this association is so particular and
significant that certain chromosomes are actually regarded by
many as representing sex " determiners." This relation,
besides affording striking evidence in this connection, is of very
great importance in itself, and we may therefore consider it at
somewhat greater length than this connection alone would
justify.

During recent years many instances have come to light, of a
variation in the number of chromosomes in different individuals
of a single species. With but very few exceptions these
numerical differences are associated with difference in sex, and
when any such difference exists it is usually found that the cells
of the female contain one or more chromosomes in excess of the
number found in the male. In some species then, the chromo-
some number may be uneven in one sex, and therefore not all
the chromosomes are paired structures. In other cases the
equivalent diversity of the chromosome groups is indicated by
size differences between the members of a certain pair.

DIFFERENTIATION, HEREDITY, SEX 307

Differences of these kinds are now known in many scores of
species of many groups, from the lower worms .to man. It is
clearly impossible to include here any extended history or
survey of this fascinating subject and we can do little more
than describe a typical instance or two, and then mention
some comparisons which may throw some light, from this point
of view, upon the general interpretation of the chromosome
problem.

Since the larger number of the known instances of this rela-
tion are found among the Arthropoda, particularly the Insecta,
we may select our first illustration from this group. The num-
ber of chromosomes in the somatic cells of the common squash
bug, Anasa tristis (Henking, Paulmier), is twenty-two in the
female, and twenty-one in the male. How does this difference
come about?

For an answer to this question we must observe the behavior
of the chromosomes during the process of maturation of the
germ cells. In the process of oogenesis, preparatory to the first
oocyte division, synapsis occurs normally, and eleven bivalent
chromosomes are formed. The succeeding steps in oogenesis
are not unusual and the result is the formation, in each ovum,
of a group of eleven univalent chromosomes representing every
pair of the original somatic or oogonial group.

The events of spermatogenesis do not run quite parallel,
however. In the somatic and spermatogonial cells, twenty-one
chromosomes are present, i.e., ten pairs plus one, and in the
division of these cells every chromosome divides in the usual
way (Fig. 142). In synapsis the paired elements fuse, forming
ten bivalent chromosomes, but the odd chromosome, or X-
chromosome remains free, and usually quite apart from the
other chromosomes. This X-element, or idiochromosome, may
be distinct, even throughout the growth period of the spermato-
gonia, and during the two spermatocyte divisions it can be
identified in many species as a nucleolus-like body, indeed
formerly it was described as a chromatin nucleolus. The behav-
ior of this body during the maturation divisions is entirely un-
usual. During the first spermatocyte division the idiochro-

308

GENERAL EMBRYOLOGY

mosome, although univalent, divides just as the ten bivalent
elements do, and eleven chromosomes consequently pass into
the nuclei of the secondary spermatocytes. But during the

FIG. 142. — Maturation during the spermatogenesis of the squash-bug, Anasa
tristis, showing the behavior of the X-chromosome or idiochromosome. A, after
Wilson, others after Paulmier. A. Spermatogonium. Polar view of equatorial
plate showing twenty-one chromosomes (ten pairs, plus one). The X-chromo-
some is not distinguishable at this time. B. Primary spermatocyte. Tetrads
formed. C. Equatorial plate of first spermatocyte division. X-chromosome
divided. D. Anaphase of same division. The daughter X-chromosomes have
also diverged. E. Equatorial plate of second spermatocyte division. F. Meta-
phase of same division. The X-chromosome lies, undivided, between the two
groups of daughter chromosomes. G. Anaphase of same division. The undi-
vided X-chromosome' has passed to the upper pole, lagging behind the others.
H. Telophase of same division. X-chromosome still distinct.

mitosis of these secondary spermatocytes the idiochromosome
fails to divide and passes as a whole to one pole of the spindle
(Fig. 142, F, G, H). The result is that the nuclei of one-half
of the spermatids, and therefore of one-half of the spermatozoa,

DIFFERENTIATION, HEREDITY, SEX 309

contain eleven chromosomes, while the other half contain but
ten, lacking the idiochromosome.

Since the nuclei of all the ova contain eleven chromosomea

FIG. 143. — Diagram illustrating the behavior of the X-chromosome during
maturation. The X-element is Shown in solid black. The essentials are the
same in cases where X is a multiple element, or where it is paired with a Y-ele-
ment (see text).

there are but two possibilities in fertilization. The egg with its
eleven chromosomes may be fertilized by a sperm vrth ten,

310 GENERAL EMBRYOLOGY

giving a somatic group of twenty-one; or the egg with its
eleven chromosomes may be fertilized by a sperm with eleven,
giving a somatic group of twenty-two. And since there are
equal numbers of ten- and eleven-chromosome spermatozoa,
there will be approximately equal numbers of zygotes with
twenty-one and twenty-two chromosomes. These relations
are shown in diagrammatic form in Figs. 143, 144.

Since this numerical difference between the somatic chromo-
some groups is constantly associated with sex-difference, males
possessing twenty-one, females twenty-two chromosomes, it
may be said that the presence of the idiochromosome is in

Mature ova. All Spermatozoa. Zygotes. Sex of

of one class Two classes Two classes adults

— i Female

FIG. 144. — Diagram of the relations of chromosome number and sex during
fertilization in Anasa. The essentials are the same in cases where X is a multiple
element, or where it is paired with a Y-element.

some way connected with the determination of the female, or
the lack of it the male, sex.

Since Paulmier's description of these events in Anasa, in
1899, a great many similar instances have come to light, and
more recently quite a variety of conditions related to this but
more or less dissimilar in details, have become known. The
X-chromosome or idiochromosome has been described under
many different names such as accessory chromosome, allosome,
heterotropic chromosome, heterochromosome, monosome, etc. Such
an unequal distribution of the chromosomes was first observed
by Henking in 1890, and in 1902 McClung described similar
processes in several of the locusts and grasshoppers (Orthop-
tera), and first suggested the possible relation between this

DIFFERENTIATION, HEREDITY, SEX 311

" accessory chromosome" and the determination of sex. The
work of Wilson, Stevens, Montgomery, Payne, Guyer, Morgan,
and many others, has made known the presence of these elements
in a whole host of Insects, including most of the orders, in
Myriopods, Arachnids, and Copepods. Among the lower
forms, Nematodes (Boveri, Edwards, Boring, Gulick), Sagitta
(Stevens), and Echinoderms (Baltzer) are now known to pos-
sess idiochromosomes. And more recently some of the Chor-
dates have been added to the ever increasing list, for idiochro-
mosomes have been described in the common fowl, guinea fowl,
and rat (Guyer), the guinea pig (Stevens) and even in man,
where Guyer has reported two idiochromosomes, half the sperm
containing twelve, and half ten, chromosomes; the number of
chromosomes in the human somatic cells is, therefore, twenty-
two in the male, and twenty-four in the female.

Not all of the forms included in the above list exhibit this
phenomenon as simply as it occurs in the case of Anasa; we may
mention two general modifications of this typical condition.
In many Coleoptera, Diptera, and Hemiptera, the idiochrbmo-
some is not strictly an unpaired element for during the sperma-
tocyte divisions, and in the spermatids, it is paired with a very
small chromosome called the Y-element; together with X this
makes up an XY bivalent chromosome which behaves like any
bivalent chromosome in the preliminaries to the first spermato-
cyte division (Fig. 145). In the spermatozoa, therefore, half
the nuclei contain the large idiochromosome (X), and half the
small one (Y). The relation to sex is what might be expected,
namely, the females contain the large X, the males the small Y.
In Metapodius, one of the Orthoptera, this small Y-element may
be either present or absent apparently, although it is possible
that it may be present and fused with another chromosome
when it is said to be absent.

In Ascaris megalocephala it seems clear that a small X-
element, no Y-element being present, may thus appear either as
a separate body, or fused with one of the other chromosomes
(Boring, Boveri, Edwards). Such an attachment of the idio-
chromosome to a certain one of the ordinary chromosomes is

312

GENERAL EMBRYOLOGY

Protenor Syromasfas
Anasa Homo

Y
X

AscAris
lumbricoidfS

t f t

£t(jtftit&ts Nezara T/iynnfa
Yiricfttfa Coenua Jiitaris catceata

Y
X

&

Fttctt in

Prionirfus Geiastocoris
5ine&

A.cholla
ultisjbin osa

Fio. 145. — Diagrams illustrating some of the variations in the X- and Y-
chromosomes. From Wilson. X, either as a simple or as a multiple element,
may or may not be paired with a Y-chromosome. •

I

a

A

I

e

FIG. 146. — Compound chromosome-groups, formed by the union of the X-
rhromosome with other chromosomes, in the Orthoptera. From Wilson, a, 6,
after de Sinety, others after McClung. a. Triad group, first spermatocyte
division of Leptynia, metaphase. 6. Division of similar triad in Dixippus.
c. Triad group formed by union of the X-chromosome with one of the bivalent
chromosomes, first spermatocyte prophase, Hesperotettix. d. The same element
from a metaphase group, e. The same element in the ensuing interkinesis.
/. The compound element of Mermiria, from a first spermatocyte prophase.
g. The same element in the metaphase (now, according to McClung, united to
a second bivalent chromosome, to form a pentad element). A. The same element
after its division, in the ensuing telophase.

DIFFERENTIATION, HEREDITY, SEX 313

well known as the constant relation in some Insects (Sinety,
McClung), and among these forms various degrees of the inti-
macy of the association occur (Fig. 146).

Several stages can be found in the gradual increase in the
relative size of the Y-element, until in such forms as Nezara
hilaris, one of the Hemiptera-Heteroptera described by Wilson,
X and Y are nearly equal (Fig. 145), and finally in some of the
Lepidoptera and other forms, X and Y are quite equal and
indistinguishable from one another, although the XY pair may
be distinguished from the other chromosomes by staining prop-
erties and behavior.

We thus reach through gradual transitions a condition where
the spermatozoa are no longer dimorphic with respect to chro-
mosome content. The conditions of such a series suggest, how-
ever, the possibility that spermatozoa that visibly appear
morphologically alike, may after all be physiologically dimor-
phic as regards chromosome characters; such an assumption
must of course be made with respect to traits other than sex,
which are inherited in an alternative fashion.

Another series of modifications of the Anasa-type is illus-
trated by various genera of Hemiptera, where the X-chromo-
some is represented by more than a single element. Such a
series has been described by Payne, and is readily derived
from the condition in such a form as Euschistus (Fig. 145),
with unequal X- and Y-elements. Thus in Fitchia and several
others, Y is a fairly large chromosome while X is represented
by two somewhat smaller chromosomes; in Prionidus, Sinea, etc.
there are three X-elements to one Y; in Gelastocoris there are
four X and one Y, and in Acholla five X and one Y. In still
another series (Fig. 145) Y is entirely absent and X is repre-
sented by several chromosomes — two in Phylloxera (Morgan),
Syromastes (Wilson), Agakna (Wallace), and man (Guyer),
five in Ascaris lumbricoides according to Edwards.

In all of these cases where X is a multiple element,
different species show greatly varying relations among the
members of the X-group; they may be approximately equal
in size or very unequal, but it is important that the size

314 GENERAL EMBRYOLOGY

relations, whatever they are, are constant within a given
species.

One further general condition of the idiochromosomes must
be noted. In all of the instances mentioned above the sper-
matozoa are the dimorphic gametes, i.e., the males are "diga-
metic" (Wilson's term). In a few species, however, it is the
female that is digametic, and while the spermatozoa are all
alike, the ova are of two classes with respect to certain modified
chromosomes which may properly be regarded as homologous
with the X- and Y-elements of the spermatozoa. Thus in two
of the sea-urchins, Strongylocentrolus, and Echinus, Baltzer

FIG. 147. — Chromosomes in the sea-urchin, Strongylocentrotus lividus. After
Baltzer. X 2610. A. First cleavage spindle (reconstructed from two draw-
ings). No small hooks present. B. First cleavage spindle with small hooks.
The modified chromosomes, both long and short hooks, are shown in solid black.

finds the chromosome groups of the spermatids all alike, with
eighteen components uniformly differentiated; the nuclei of
some of the ova, however, are characterized by the presence
of a single modified component which is absent from the
remainder (Fig. 147). This obviously corresponds with the
X-element of the dimorphic sperm. There is also very good
reason for believing the female digametic in some of the
Lepidoptera; this is of considerable interest in view of the fact
that repeated investigation has thus far failed to disclose
idiochromosomes in the spermatocytes of these Insects.

DIFFERENTIATION, HEREDITY, SEX 315

The idea that the relation between the chromosomes and
sex characters parallels that between the chromosomes and
any other traits involves the conclusion that sex is a character,
or group of characters, inherited in the same way that other
bodily traits are. And this conclusion may now be accepted.
Indeed there are in the field several hypotheses as to the precise
statement, of a Mendelian formula according to which sex is
inherited, and while no one of them has a preponderance of
evidence in its favor, the fundamental fact of sex heredity is
clear.

There are extant scores of hypotheses regarding the factors
and processes involved in sex determination, depending upon
the action of conditions outside of the germ itself. These must
be abandoned when the facts now known to be true of the
germinal structure of a comparatively limited number of
species, gain a wider applicability. For the sex of an organism,
as well as other fundamental characters, appears to be already
determined in the zygote, and all that external conditions can
do toward determining sex is to alter sex ratios by affecting
differentially (selectively) the gametes or immature organisms
of a certain sex.

There is some evidence of other kinds that sex is determined
in the gamete and not by external conditions. In certain
cases a single egg indirectly gives rise to a number of embryos
or larva? (multiple embryo formation) which are all of one sex,
either male or female. Silvestri describes such a case in the
development of a Hymenopter, Litomastix, parasitic in the
larva of a Lepidopter, Plusia, where as many as one thousand
embryos, all of one sex, are thus formed. And there is good
reason for believing that the embryos of one of the armadillos,
described by Newman and Patterson, are all derived from a
single ovum, and these are always of one sex only, either male
or female. There is also the familiar example of the bees
(Dzierzon), where unfertilized eggs develop parthenogenetically
into males (drones), while the fertilized ova produce females
(queen and workers) ; the same thing is apparently true of most
ants. And in one of the rotifers, Hydatina, a certain kind of

316 GENERAL EMBRYOLOGY

female produces eggs which if fertilized produce females, if
unfertilized produce males.

One further point is to be mentioned in connection with
the general hypothesis of nuclear determination. This is in
connection with that curious form of Mendelian heredity
known as " sex-limited " heredity, where certain characters
are exhibited by the individuals of one sex only,, although
transmitted by the individuals of the other sex without being
exhibited by them. Such a form of heredity can readily be
explained upon the chromosome hypothesis, upon the simple
assumption of a close relation between the " determiner" for
the sex-limited character and the sex-determining element.

Any further consideration of the problems of sex determina-
tion and heredity would lead us too far afield; more extended
treatment must be had from other sources. In conclusion
then, it is hardly necessary to point out that the constancy of
the form and of the complicated behavior of these idiochromo-
somes affords very striking confirmation of the hypothesis of
the specificity and genetic continuity of the chromosomes.
While it is possible that their form and behavior are determined
by underlying conditions, such conditions cannot be directly
observed and can only be postulated. Taken in connection
with the facts mentioned in Chapter II, and with the results
drawn from the development of dispermic eggs, and from
hybridization, they amount to practical demonstration of
some form of chromosomal specificity in development.

As to the question whether the idiochromosomes are in
particular the sex determinants, several views may be held in
the absence of conclusive experimental demonstration of the
precise relation. It has been held in some quarters that sex
is determined by the relative amount of chromatin received
into the nucleus of the zygote, irrespective of its content in
certain chromosomal elements. This is hardly tenable how-
ever in view of many contradictory conditions. Others have
suggested that the dimorphism of the gametes is merely asso-
ciated with other more fundamental diversities, and that sex
differentiation and gamete differentiation are related only be-

DIFFERENTIATION, HEREDITY, SEX 317

cause both are related to some primary differentiation. Still
others hold to the idea that the idiochromosomes actually
determine by their presence or absence, the nature of the
reactions of development, so that finally organisms with female
or male characteristics are formed. The most adequately justi-
fied and most conservative view seems to be that the nature
of the interrelations of the components of the whole chromo-
some group, among themselves and to the cytoplasm, is modi-
fied by the presence or absence of certain elements so that in
one case the primary and secondary female characters develop,
in the other case the male characters.

Returning now to the general subject of this chapter, namely,
the factors determining the course of development and the
process of heredity, we come to another extremely important
subject. We have thus far emphasized the importance of the
internal factors of development. But we have defined develop-
ment as a series of reactions between internal and external
factors. The omission hitherto of specific reference to the
external conditions of development is not because these are of
lesser general importance. Alterations in the conditions of
gravity, pressure, temperature, light, moisture, and chemical
composition of the surrounding medium may, each or all affect
the course of development, either in a general or in a specific
way. A great deal is known of the results of modifying such
conditions and a rather full discussion of these effects would be
in order, were the results susceptible of more definite and more
uniformly applicable statement. For normal development,
normal environing conditions are necessary. However, slight
variations in external conditions rarely produce effects com-
parable with those following slight variations in the internal
conditions. That is to say, slight variations in external condi-
tions are "normal." When the modification of external condi-
tions is sufficiently marked to produce visible effects upon
development, these are frequently so marked as to be regarded
as distinct abnormalities, and the organism so affected is rarely
able to complete its development to maturity.

318 GENERAL EMBRYOLOGY

The natural environment ordinarily varies within rather
narrow limits, frequently on account of the ovipositing habits
of the adult, and changes within these limits rarely affect the
course of development, so that for the subjects of heredity and
differentiation we should inquire here only into the question to
what extent external conditions are necessary factors in carry-
ing on the life of the organism as it exists in the form of an
egg or embryo. The particulars of development and heredity
are referable to internal characteristics which determine the
specific or individual quality of the reactions between organism
and environment.

We may proceed, therefore, to mention a few illustrations of
the effects of alterations in external conditions of development,
not attempting to do more than to suggest the nature of the
work accomplished in this field; an adequate survey falls outside
the scope of such a text as this. (For a convenient summary,
see,e.</., Jenkinson, "Experimental Embryology," Oxford, 1909.)

More is known regarding the effects upon development, of
chemical substances than of other conditions. • While a few
forms, such as the minnow, Fundulus, are able to develop
normally in media so widely unlike, physically and chemically,
as sea water and distilled water, this and other forms show
specific effects of the presence or absence of certain salts alone.
Thus in Fundulus Stockard has shown that the presence of cer-
tain amounts of magnesium salts brings about the fusion of the
optic vesicle regions, so that one-eyed monsters develop, appar-
ently normal in other respects (Fig. 148). The eggs and em-
bryos of the Echinoderms offer many striking facts in this
connection. We have already noted that the alkalinity of the
sea water may determine the appearance of paternal or mater-
nal characters in hybrid Echinoderm larvae. Herbst and others
have shown that the absence of potassium salts is fatal or very
harmful to Echinoderm larvaB apparently on account of the
resulting diminution in the process of water absorption; the
absence of calcium causes a tendency for the blastomeres to fall
apart; magnesium and the sulphates are necessary for the nor-
mal differentiation of the alimentary tract; the production of

DIFFERENTIATION, HEREDITY, SEX

319

ciliary movement depends upon the presence of magnesium,
and an excess of calcium results in the hypertrophy of the cilia;
sulphates are necessary also for the establishment of the funda-
mental structure of the embryo and for the formation of
pigment; magnesium, sulphates, and calcium carbonate are

FIG. 148. — Effects of magnesium chloride upon the development of the Teleost,
Fundulus. From Stockard. A. Normal fish, eight days after hatching.
M, mouth. B, C, Two views of fish, showing the fusion of the optic vesicles
as the result of treatment with MgCh.

necessary for the development of a normal skeleton (Fig. 149).

Certain optima exist for moisture, density, pressure, light and
temperature; in development as in later life, deviations from
the optimum condition, in either direction, affect the rate of
development rather than its character. The direction of
gravity takes an essential part in determining normal develop-
ment in a few cases, but ordinarily development is independent
of this factor.

In general all of these conditions are involved not so much in
the regulation of development in specific and particular direc-
tions, as in determining whether it shall proceed at all or not.
Modifications of development produced by effective variations

GENERAL EMBRYOLOGY

in these conditions are often so extreme that the phenomena of
heredity are scarcely apparent and usually the modified organ-
ism does not come to maturity.

We may now attempt to summarize a conservative concep-
tion of the relation of the structure of the germ cells to the pro-
cesses of development and heredity. The zygote is an organ-
ism, morphologically and physiologically specific. It possesses

h

FIG. 149. — Effects of chemical alteration of the surrounding medium, upon the
development of the sea-urchin. From Jenkinson. a. Without OH; ciliated
solid blastula of Sphcer echinus, b. KOH has been added, c. Normal blastula
of Sphcer echinus, d. Blastula in a K-free medium, e. Reared in K-free medium
and replaced in normal sea- water (Sphcerechinus) . f. Sphcerechinus larva from a
medium devoid of Mg. g. Echinus pluteus with three-parted gut, mouth, and
ccelomic sacs, but neither skeleton nor arms; reared without CaCOs or CaSO4.
h. Normal pluteus of Echinus.

polarity, symmetry, various forms of differentiated substance,
even organs, composed of subsidiary elements and capable of
performing definite and highly varied and specialized functions.
This organism in its parts and as a whole does certain things,
makes certain reactions, in a word, develops. The quality of
the developmental reactions is determined primarily by the

DIFFERENTIATION, HEREDITY, SEX 321

conditions within the organism itself, and as it reacts, as the
organism develops step by step, these internal conditions rapidly
change. These reactions on the part of the organism fall into
two groups. (1) Reactions between the organism (i.e., cyto-
plasm and nucleus, whether the organism consists of one cell or
many) and its environing stimuli. (2) Reactions between the
nucleus and cytoplasm of each cell. The idea of reaction must
involve two factors, but while equally necessary for reaction,
they are not necessarily of equal value in determining or con-
trolling the quality of the reaction. A great many organisms
react to light; but the quality of the reaction is determined
primarily by the organism.

The whole structure of the cytoplasm may play a large part
in determining the quality of the reactions of the egg, but this
cytoplasmic structure is itself the result of a series of interac-
tions between cytoplasm and nucleus, and the action of the
latter is of primary importance in affecting the quality of the
result. Going one step further, what the nucleus does is deter-
mined by its structure, and this is also the result of interactions
of its parts with one another, and with the cytoplasm, which is
its environment; and here again certain elements of the nucleus,
namely, the chromosomes, seem to be of primary importance
in determining the quality of the interaction. The most impor-
tant of the concrete, visible organs of the nucleus are the chro-
mosomes. And when we attempt to analyze the behavior of
these components we are met by the same problem — what
determines the structure and behavior of these? Two answers
have been offered. First, that here we reach the limit of analy-
sis, that the chromosomes are autonomic, self-perpetuating,
self-regulating bodies, whose morphology and behavior are the
determining factors in all that happens in the life of the organ-
ism. Second, that the chromosomes are themselves made up
of still more fundamental units, the chromatin granules; that
these are the autonomic, self-perpetuating, finally determina-
tive units in development, and chromosome structure is the
result of the primary activity of these bodies.

Logically there is no reason why we must stop with the chro-

322 GENERAL EMBRYOLOGY

matin granules. And history, enumerating germ layers, cleav-
age cells, cytoplasmic organ-forming substances, chromosomes,
and chromioles (chromatin granules), warns us against the idea
that we must seek or hope to find ultimate particles, concrete,
definable, and representatively determinative in function. In
fact a few students of the problem frankly declare their belief
that the idea of any sort of representative particle mechanism
is futile, that the regulation of the processes of development
and heredity depends upon interrelations which are not sus-
ceptible of interpretation in terms of any material basis.

Scientifically, however, we can to-day go no further back
than the chromosomes, for here we find the most fundamental
units whose actual behavior can be correlated with the facts
of development. To say scientifically, that the chromosomes
are (to-day known to be) the determining elements in develop-
ment and heredity, is not to deny the existence of other bodies
or conditions which may determine the existence and qualities
of the chromosomes. Granules we know, but of their behavior
we know little, and this little cannot at present be correlated
with the facts of development. Of the real existence of elements
underlying the granules we know nothing whatever. Assump-
tion of the reality of such bodies or conditions may be a logical
necessity, but to-day it carries us beyond the boundaries of
observed fact.

To repeat a statement made on an earlier page, if the exist-
ence and activity of the chromosomes can be shown to be a
necessary link in the processes of development and heredity,
and if these can be shown to be the simplest and most nearly
primary factors whose behavior can be correlated with these
processes, then we shall be justified in saying that the chromo-
somes are to-day the determining factors in development and
heredity.

REFERENCES TO LITERATURE

BALTZER, F., Die Chromosomen von Strongylocentrotus lividus und
Echinus microtuberculatus. Arch. Zellf. 2. 1909. Ueber die
Beziehung zwischen dem Chromatin und der Entwicklung und

DIFFERENTIATION, HEREDITY, SEX 323

Vererbungsrichtung bei Echinodermenbastarden. Arch. Zellf. 6.
1910.

BONNEVIE, K., (See ref. Ch. II.)

BORN, G., Ueber Druckversuche an Froscheiern. Anat. Anz. 8.
1893.

BOVERI, T., Ein geschlechtlich erzeugter Organismus ohne mutterliche
Eigenschaften. Sitz.-Ber. Ges. Morph. Phys. Miinchen. 6.
1889. English translation in Amer. Nat. 27. 1893. Ueber die
Befruchtungs- und Entwickelungsfahigkeit kernloser Seeigel-Eier
und liber die Moglichkeit ihrer Bastardirung. Arch. Entw.-Mech.
2. 1895. Die Entwicklung von Ascaris megalocephala, mit
besonderer Rticksicht auf die Kernverhaltnisse. Festschr. f. v.
Kupffer. Jena. 1899. Ueber mehrpolige Mitosen als Mittel zur
Analyse des Zellkerns. Verh. phys.-med. Ges. Wiirzburg. 36.
1902. Das Problem der Befruchtung. Jena. 1902. Protoplas-
madifferenzierung als auslosender Faktor fur Kernverschiedenheit.
Sitz.-Ber. phys.-med. Ges. Wiirzburg. 37. 1904. Ergebnisse tiber
die Konstitution der chromatischen Substanz des Zellkerns. Jena.
1904. Zellenstudien V. Ueber die Abhangigkeit der Kerngrosse
und Zellenzahl der Seeigel-Larven von der Chromosomenzahl der
Ausgangszellen. Jena. Zeit. 39 (32). 1905. Zellenstudien VI.
Die Entwickelung dispermer Seeigel-Eier. Ein Beitrag zur
Befruchtungslehre und zur Theorie des Kernes. Jena. Zeit.
43 (36). 1908. Ueber " Geschlechtschromosomen " bei Nema-
toden. Arch. Zellf. 4. 1909. Die Potenzen der Ascaris- Blasto-
meren bei abgeanderter Furchung. Zugleich ein Beitrag zur
Frage qualitativ-ungleicher Chromosomen-Theilung. Festschr. f.
R. Hertwig. Jena. 1910.

BUCHNER, P., Das accessorische Chromosom in Spermatogenese und
Ovogenese der Orthopteren zugleich ein Beitrag zur Kenntnis der
Reduktion. Arch. Zellf. 3. 1909.

CASTLE, W. E., A Mendelian View of Sex-Heredity. Science. 29.
1909.

CHILD, C. M., Some Considerations regarding so-called Formative
Substances. Biol. Bull. 11. 1906.

CONKLIN, E. G. Mosaic Development in Ascidian Eggs. Jour. Exp.
Zool. 2. 1905. Organ Forming Substances in the Eggs of Asci-
dians. Biol. Bull. 8. 1905. Does Half of an Ascidian Egg
give Rise to a whole Larva? Arch. Entw.-Mech. 21. 1906. The
Mechanism of Heredity. Science. 27. 1908. The Effects of
Centrifugal Force upon the Organization and Development of the
Eggs of Fresh Water Pulmonates. Jour. Exp. Zool. 9. 1910.
The Organization of the Egg and the Development of Single Blasto-

324 GENERAL EMBRYOLOGY

meres of Phallusia mamillata. Jour. Exp. Zool. 10. 1911.

(See also Ref. Ch. II, III.)
CRAMPTON, H. E., Experimental Studies on Gasteropod Development.

Arch. Entw.-Mech. 3. 1896.
DRIESCH, H., Entwicklungsmechanische Studien. IV. Experimen-

telle Veranderungen des Typus der Furchung und ihre Folgen.

(Wirkung von Warmezufuhr und von Druck). Zeit. wiss. Zool.

55. 1892. Zur Analysis der Potenzen embryonaler Organzellen.

Arch. Entw.-Mech. 2. 1895. Resultate und Probleme der

Entwickelungsphysiologie der Tiere. Ergebnisse Anat. u. Entw.

8. 1898 (1899). Die Localisation morphogenetischer Vorgange.

Ein Beweis vitalistischen Geschehens. Arch. Entw.-Mech. 8.

1899. Die isolirten Blastomeren des Echinidenkeimes. Arch.

Entw.-Mech. 10. 1900. Studien liber das Regulationsver-

mogen der Organismen. IV. Die Verschmelzung der Individu-

alitat bei Echinidenkeimen. Id. 1900. Zur Cytologie partheno-

genetischer Larven von Strongylocentrotus. Id. 19. 1905. Altes

und Neues zur Entwicklungsphysiologie des jungen Asteridenkei-

mes. Id. 20. 1905. Die Entwickelungsphysiologie. 1905-

1908. Ergebnisse Anat. u. Entw. 17. 1907 (1909). The

Science and Philosophy of the Organism. London. 1908.
EDWARDS C. L., The Idiochromosomes in Ascaris megalocephala and

Ascaris lumbricoides. Arch. Zellf. 6. 1910.
FICK, R., (See ref. Ch. II.)
FISCHEL, A., Experimented Untersuchungen am Ctenophorenei. I.

Von der Entwickelung isolirter Eitheile. Arch. Entw.-Mech. 6.

1897. Id. IV. Ueber den Entwickelungsgang und Organisa-

tionsstufe des Ctenophoreneies. Id. 7. 1898.
GARBOWSKI, T., Ueber Blastomerentransplantation bei Seeigeln. Bull.

Internat. Acad. Sci. Cracovie. 1904 (1905).
GODLEWSKI, E., Das Vererbungsproblem im Lichte der Entwicklungs-

mechanik betrachtet. Leipzig. 1909. (See ref. Ch. V.)
GROSS, J., Heterochromosomen und Geschlechtsbestimmung bei

Insecten. Zool. Jahrb. (Abth. f. Allgem. Zool. u. PhysioL). 32.

1912.
GUYER, M. F., Deficiencies of the Chromosome Theory of Heredity.

Univ. Cincinnati. (II). 5. 1909. Accessory Chromosomes in Man.

Biol. Bull. 19. 1910. Nucleus and Cytoplasm in Heredity.

Amer. Nat. 45. 1911.
HACKER, V., Die Chromosomen als angenommene Vererbungstrager.

Ergebnisse und Fortschritte der Zoologie. 1. 1907. Ergebnisse

und Ausblicke in der Keimzellenforschung. Zeit. Indukt.

Abstamm. Vererbungslehre. 3. 1910. Allgemeine Vererbungs-

lehre. 2 Auf. Braunnschweig. 1912.

DIFFERENTIATION, HEREDITY, SEX 325

HEPFNER, B., A Study of Chromosomes of Toxopneustes variegalus
which show individual Peculiarities of Form. Biol. Bull. 19.
1910.

HEGNER, R. W., Germ-Cell Determinants and their Significance.
Amer. Nat. 45. 1911.

HENKING, H., Untersuchungen iiber die ersten Entwicklungsvorgange
in den Eiern der Insekten. II. Ueber Spermatogenese und
deren Beziehung zur Eientwicklung bei Pyrrhocoris apterus.
Zeit. wiss. Zool. 61. 1891.

HERBST, C., Ueber das Auseinandergehen von Furchungs- und Gewebe-
zellen in kalkfreiem Medium. Arch. Entw.-Mech. 9. 1900.
Formative Reizen in der thierischen Ontogenese. Leipzig. 1901.
Vererbungsstudien. VI. Die cytologischen Grundlagen der Ver-
schiebung der Vererbungsrichtung nach der rmitterlichen Seite.
Arch. Entw.-Mech. 27. 1909. Vererbungsstudien. VII. Same
title, 2. Mitteilung. Id. 34. 1912. -

HERTWIG, O., Das Problem der Befruchtung und der Isotropie des
Eies, eine Theorie der Vererbung. Jena. Zeit. 18 (11). 1885.
Altere und neurere Entwicklungstheorien. Berlin. 1892. Ueber
den Werth der ersten Furchungszellen fur die Organbildung des
Embryo. Experimentelle Studien am Frosch- und Tritonei.
Arch. mikr. Anat. 42. 1893. Der Kampf um Kernfragen der
Entwicklungs- und Vererbungslehre. Jena. 1909.

HERTWIG, R., Ueber neue Probleme der Zellenlehre. Arch. Zellf. 1.
1908.

His, W., Unsere Korperform und das physiologische Problem ihrer
Entstehung. Leipzig. 1874.

JENKINSON, J. W., Experimental Embryology. Oxford. 1909.

KORSCHELT UND HEiDER, Lehrbuch, etc. I. Abschnitt. Experi-
mentelle Entwicklungsgeschichte. Jena. 1902.

KUPELWIESER, H., (See ref. Ch. V.)

LANKESTER, E. R., Notes on Embryology and Classification. London.
1877.

LILLIE, F. R., Differentiation without Cleavage in the Egg of the An-
nelid Chcetopterus pergamentaceous. Arch. Entw.-Mech. 14. 1902.
Observations and Experiments concerning the elementary Phe-
nomena of Embryonic Development in Chcetopterus. Jour. Exp.
Zool. 3. 1906. Polarity and Bilaterality of the Annelid Egg.
Experiments with Centrifugal Force. Biol. Bull. 16. 1909.

LOEB, J., KING, W. 0. R., and MOORE, A. R., Ueber Dominanzerschei-
nungen bei den hybriden Pluteen des Seeigels. Arch. Entw.-
Mech. 29. 1910.

LYON, E. P., Some Results of centrifugalizing the Eggs of Arbacia.
Proc. Amer. Physiol. Soc. in Amer. Jour. Physiol. 15. 1905.

326 GENERAL EMBRYOLOGY

McCLENDON, J. F., Cytological and Chemical Studies of Centrifugalized
Frog Eggs. Arch. Entw.-Mech. 27. 1909.

McCLUNG, C. E., A peculiar Nuclear Element in the Male Reproductive
Cells of Insects. Zool. Bull. 2. 1899. The Accessory Chromo-
some—Sex Determinant? Biol. Bull. 3. 1902. The Chromo-
some Complex of Orthopteran Spermatocytes. Id. 9. 1905.

MAAS, O., Experimentelle Untersuchungen iiber die Eifurchung. Sitz.-
Ber. Ges. Morph. Physiol. Mlinchen. 17. 1902. Einfuhrung
in die experimentelle Entwickelungsgeschichte (Entwickelungs-
mechanik). Wiesbaden. 1903.

MEVES, F., Die Chondriosomen als Trager erblicher Anlagen. Cyto-
logische Studien amHuhnerembryo. Arch. mikr. Anat. 72.1908.

MONTGOMERY, T. H. JR., Chromosomes in the Spermatogenesis of the
Hemiptera-Heteroptera. Trans. Am. Phil. Soc. 21. 1906. The
Terminology of Aberrant Chromosomes and their Behavior in
certain Hemiptera. Science. 23. 1906. Are Particular Chromo-
somes Sex Determinants? Biol. Bull. 19. 1910. (See also
ref. Ch. II.)

MORGAN, T. H., A Biological and Cytological Study of Sex Determina-
tion in Phylloxerans and Aphids. Jour. Exp. Zool. 7. 1909.
The Effects of Altering the Position of the Cleavage Planes in
Eggs with Precocious Specification. Arch. Entw.-Mech. 29.
1910. Cytological Studies of Centrifuged Eggs. Jour. Exp. Zool.
9. 1910. Chromosomes and Heredity. Amer. Nat. 44. 1910.

MORGAN, T. H., and SPOONER, G. B., The Polarity of the Centrifuged
Egg. Arch. Entw.-Mech. 28. 1909.

NAGELI, C., Mechanisch-physiologische Theorien der Abstammungs-
lehre. Miinchen. 1884.

PAULMIER, F. C., The Spermatogenesis of Anasa tristis. Jour. Morph.
16. Suppl. 1899. (For important corrections and additions,
see Wilson, E. B., Studies. II. 1905— Full title below.)

PAYNE, F., Some New Types of Chromosome Distribution and their
Relation to Sex. Biol. Bull. 16. 1909.

PFLUGER, E., Ueber den Einfluss der Schwerkraft auf die Theilung der
Zellen. Arch. Gesam. Physiol. 31, 32. 1883.

PRZIBRAM, H., Experimentelle Biologic der Seeigel. (In Bronn's
Klass. u. Ordn. II. 3). Leipzig. 1902. Experimental Zoology.
I. Embryogeny. Cambridge. 1908.

Roux, W., Ueber die Bedeutung der Kernteilungsfiguren. Leipzig.
1883, 1895. Ueber die verschiedene Entwickelung isolirter
erster Blastomeren. Arch. Entw.-Mech. 1. 1895. Einleitung.
Arch. Entw.-Mech. 1. 1894. English Translation by W. M.
WHEELER, "The Problems, Methods, and Scope of Developmental
Mechanics." Woods Holl Biol. Lect. 1895.

DIFFERENTIATION, HEREDITY, SEX 327

SALA, L., Experimentelle Untersuchungen iiber die Reifung und
Befruchtung der Eier bei A scaris megalocephala. Arch. mikr.
Anat. 44. 1895.

SCHAXEL, J., Die Eibildung der Meduse Pelagia noctiluca, etc. Festsch.
f. R. Hertwig. Jena. 1910. Versuch einer cytologischen
Analysis der Entwicklungsvorgange. I. Die Geschlechtszellen-
bildung und die normale Entwicklung von Aricia foetida, Clap.
Zool. Jahrb. 34. 1912.

SCHLEIP, W., Geschlechtsbestimmende Ursachen im Tierreich. Ergeb.
und Fortschr. Zool. 3. 1912.

SCHNEIDER, K. C., (See ref. Ch. II.)

SILVESTRI, F., Un nuovo interessantissimo caso di germinogonia (poli-
embrionia specifica) in un Imenottero parassita endofago con
particolare destine dei globuli polari e dimorphismo larvale.
Atti Accad. Lincei Rend. (V.) 14. 1905.

SINETY, R. DE, (Ref. in Ch. IV.)

STEVENS, N. M., A Study of the Germ Cells of Certain Diptera, with
Reference to the Heterochromosomes and the Phenomena of
Synapsis. Jour. Exp. Zool. 6. 1908. Further Studies on the
Chromosomes of the Coleoptera. Id. 6. 1909.

STOCKARD, C. R., The Development of artificially produced Cyclopean
Fish— "The Magnesium Embryo/' Jour. Exp. Zool. 6. 1909.

SUTTON, W. S., On the Morphology of the Chromosome Group in
Brachystola magna. Biol. Bull. 4. 1902. The Chromosomes in
Heredity. Id. 1903.

TENNENT, D. H., The Dominance of Maternal or of Paternal Characters
in Echinoderm Hybrids. Arch. Entw.-Mech. 29. 1910. Echi-
noderm Hybridization. Publ. Carnegie Inst. No. 132. Washing-
ton. 1911.

VERNON, H. M., The Effect of Environment on the Development of
Echinoderm Larvae. An Experimental Inquiry into the Causes of
Variation. Phil. Trans. Roy. Soc. 186, B. 1895. Cross Fertili-
sation among Echinoids. Arch. Entw.-Mech. 9. 1900.

DEVRIES, H., Intracellulare Pangenesis. Jena. 1889.

WEISMANN, A., (See ref. Ch. I.)

WHITMAN, C. 0., The Embryology of Clepsine. Q.J.M.S. 18. 1878.
Evolution and Epigenesis. Woods Holl Biol. Lect. 1894.

WILSON, E. B., Experimental Studies on Germinal Localization. I.
The Germ-Regions in the Eggs of Dentalium. II. Experiments on
the Cleavage Mosaic in Patella and Dentalium. Jour. Exp. Zool.
1. 1904. Experiments on Cleavage and Localization in the
Nemertine Egg. Arch. Entw.-Mech. 16. 1903. Studies on
Chromosomes. I. The Behavior of the Idiochromosomes in
Hemiptera. Jour. Exp. Zool. 2. 1905. II. The paired Micro-

328 GENERAL EMBRYOLOGY

chromosomes, Idiochromosomes and Heterotropic Chromosomes
in the Hemiptera. Id. 1905. III. The sexual Differences of
the Chromosome Groups in Hemiptera, with some Considerations
on the Determination and Heredity of Sex. Id. 3. 1906. IV.
The " Accessory" Chromosome in Syromastes and Pyrrhocoris^
with a Comparative Review of the Types of Sexual Differences
of the Chromosome Groups. Id. 6. 1909. V. The Chromo-
somes of Metapodius. A Contribution to the Hypothesis of the
Genetic Continuity of Chromosomes. Id. 6. 1909. VI. A
new Type of Chromosome-Combination in Metapodius. Id. 9.
1910. VII. A Review of the Chromosomes of Nezara. with some
more general Considerations. Jour. Morph. 22. 1911. The
Chromosomes in Relation to the Determination of Sex. Science
Progress. 4. 1910. The Sex-Chromosomes. Arch mikr. Anat.
77. 1911.

YATSU, N., Experiments on Germinal Localization in the Egg of
Cerebratulus. Jour. Coll. Sci. Imp. Univ. Tokyo. 27. 1910.
Observations and Experiments on the Ctenophore Egg. II.
Notes on Early Cleavage Stages and Experiments on Cleavage.
Ann. Zool. Japonensis. 7. 1911.

ZELENY, C., (Ref. Ch. VI.)

ZOJA, R., Sullo sviluppo dei blastomeri isolati dalle uova di alcune
meduse (e di altri organismi). Arch. Entw.-Mech. 1 and 2.
1895.

ZUR STRASSEN, 0., Ueber die Riesenbildung bei Ascaris-Eiern. Arch.
Entw.-Mech. 7. 1898.

CHAPTER VIII
THE BLASTULA, GASTRULA, AND GERM LAYERS

MORPHOGENETIC PROCESSES

IN this chapter we shall endeavor to summarize the more
general processes of early development which lead to the
formation, out of the group of cleavage cells, of an embryo
possessing the beginnings of the chief organs or systems. This
will carry us from the formation of the blastula, through the
important events of gastrulation and germ layer formation,
and the varied processes by which tissue and organ rudiments
are set apart and differentiated.

We shall give particular attention, indeed practically shall
limit ourselves, to a descriptive morphological account of
these events. This is done, not as a matter of choice, but
because the experimental results of the functional analysis
of these processes are so fragmentary and so scattered, that the
attempt at their summary, in a text of this character, seems
unwise. This is partly because the efforts to analyze these
processes experimentally have been delayed until the similar
problems of cell organization and cleavage should have been
more satisfactorily "solved. These topics we have already
considered. We should say, however, that what has already been
accomplished in the way of describing these later phenomena
of the mechanics of development (Entwicklungsmechanik,
Roux) bears out the general conclusions indicated by the earlier
processes, namely, that while the action of external conditions
as stimuli is essential, their place normally is chiefly that of
affording the general conditions of life and development. The
actual quality and the really significant details of the later,
as of the earlier phenomena of development and differentia-
tion, depend primarily upon internal conditions and relations.

329

330 GENERAL EMBRYOLOGY

And further, there is little reason for supposing that here too
the essential determinative conditions are other than nuclear,
although from the nature of the case, . the evidence must be
less direct.

We shall limit ourselves in another direction also. It is
obviously impossible to give a brief, and at the same time an
adequate, account of the extremely diverse methods of gas-
trulation, germ layer formation, etc., in the Metazoa as a whole.
We shall, therefore, confine ourselves here largely to the con-
sideration of these events among the Chordata. This will
enable us to give a more adequate consideration to the topics
selected.

We have seen that while no definite termination can be
placed to the period of cleavage, there is rather general, though
arbitrary, agreement that cleavage may be said to have termi-
nated when the blastomeres become arranged as a more or
less definite epithelium or layer, bounding an internal space of
some sort; and further that this may, in some cases, also be
marked by the attainment of a certain nuclear-cytoplasmic
relation. The organism exhibiting these characteristics is
known as the blastula. In this stage the organism is essentially
a monodermic structure, that is, it is composed of a single tissue
or layer of somewhat similar cells. In the simplest form of
blastula this layer is but one cell thick (Amphioxus, Fig. 150, A),
but in most of the Chordate blastulas the wall is many cells in
thickness.

Like the cleavage pattern, though to a much greater extent,
the general form of the blastula is largely determined by the
amount of yolk or deutoplasm contained in the ovum; and the
form of the blastula, in turn, largely determines the form of the
gastrula, and the methods of gastrulation and germ layer
formation. We may for convenience, therefore, distinguish
three general forms of blastulas, although intergradations are
not infrequent. When the ova are nearly homolecithal, and
cleavage is total and adequal, as in Amphioxus (Fig. 150, A),
the blastula is practically a hollow sphere (coeloblastula) . Its
wall is a simple epithelium, one cell in thickness, and its cavity,

BLASTULA, GASTRULA, AND GERM LAYERS 331

the blastocoel or segmentation cavity, is large and nearly central,
though not quite, for nearly always the cells at the vegetal pole
are larger than those at the animal pole, and the wall conse-
quently thicker in the former region. This form of blast ula is
commonly regarded as primitive, though hardly typical of the
Chordates in general, for among these it is found only in
Amphioxus.

C

I)

en

FIG. 150. — Types of Chordate blastulse. A. Amphioxus (coeloblastula).
B. Petromyzon. After von Kupffer. C. Noturus (Teleost) (discoblastula).
D. Triton (Urodele). After Greil. a, animal pole; c, blastocoel; p, periblast;
v, vegetal pole.

More usually the Chordate ovum contains a considerable
amount of yolk, as in the Ganoids and Amphibia. And here
the cleavage, though nearly or quite complete in most cases, is
decidedly unequal and the form of the blastula is considerably
modified in consequence. Here the wall of the blastula is
several or many cells thick; the cells of the animal pole are
quite small, while the yolk-containing cells, of the vegetal
hemisphere, are very large. The blastocoel is therefore quite
eccentric, displaced toward the animal pole, and usually
much reduced in size (Fig. 150, B, D).

332 GENERAL EMBRYOLOGY

The most highly modified type of Chordate blastula is found
in those forms with extremely meroblastic, telolecithal eggs,
where cleavage is of the discoid type. This condition is com-
mon to the Elasmobranchs (Fig. 158), the true Teleosts (Fig.
150, C), and to the higher Craniates — the Reptiles, Birds. In
reality this is, in a modified way, characteristic of the Mam-
mals also, for although the Mammalian ovum is nearly alecithal,
it is clearly derived from the Reptilian condition, and many
features of its development show unmistakably the effects of a
large yolk content previously present, but now lost in correla-
tion with the newly acquired source of nourishment possessed
by the Mammalian embryo. The result of discoid cleavage is
the formation of a small mass of living active cells, the blasto-
derm, or blastodisc, or germ disc, lying upon the surface of the
yolk mass. The blastula instead of being spherical, has there-
fore the form of a circular disc, the cellular elements of which
can really be compared, at first, only with the cells of the animal
pole of the spherical blastula, the unsegmented yolk represent-
ing, in this stage, the large cells of the vegetal pole of such a
blastula. In comparing these two types of blastulas we may
imagine that the ordinary spherical blastula has been cut in two
horizontally, through or just above its equator, and the animal
hemisphere flattened out, its circumference being thereby
somewhat extended. This form of blastula (discoblastula) is
several cells in thickness and is usually separated from the
underlying yolk by a shallow space called the sub-germinal
cavity which represents the blastocoel (Fig. 150, C). While
the yolk mass is usually by no means wholly devoid of nuclei,
these may or may not be associated with true cellular structures,
and even when present at this stage these rarely give rise to
any structural parts of the definitive embryo when this forms.
The Mammalian blastula diverges widely from any of these
conditions and on account of its very special character further
reference to it may be omitted here.

The next important step in development consists essentially
in the conversion of this monodermic blastula into a didermic
organism, that is, one in which the cells are arranged in two,

BLASTULA, GASTRULA, AND GERM LAYERS 333

more or less distinct, tissues or layers. This process is known
as gastrulation, and the didermic embryo itself is called a
gastrula.

Among the lower forms the process of gastrulation often
remains a simple one, involving little more than the mere
rearrangement of the cells of the blastula into two nearly
homogeneous layers. But in the higher forms, such as the
Chordata, with which we are dealing, the process is greatly
complicated by the precocious formation of the rudiments of
the chief axial structures of the later developing embryo, as
well as by the differentiation of a third tissue, or intermediate
layer between the other two. The establishment of the rudi-
ments of the axial notochord and central nervous system,
characteristic structures of the Chordate embryo, is termed
notogenesis. These rudiments are formed out of the substance
of the two primary layers of the gastrula. But the formation
of a tissue between these converts the didermic embryo into a
tridermic organism.

It is possible to analyze the whole process of development
during this period into these three subsidiary processes, gas-
trulation, notogenesis, and middle layer formation, and in
some instances they may occur somewhat separately and
successively. But usually there is much overlapping, and
the attempt to describe the process of gastrulation by itself
would give a very incomplete and incorrect view of the events
of this period. Consequently we shall describe all three of these
processes together.

Three general types of gastrulas and modes of gastrulation
may be found, corresponding with the three types of blastulas
and ova mentioned above. Again the simplest and probably
the most primitive, though not the most typical, condition is
found in Amphioxus. On the posterior side of the blastula, in
the region just between animal and vegetal hemispheres,
Cerfontaine has described a small group of cells marked by a
tendency to rapid and continued multiplication (Fig. 151, A).
This region of active proliferation gradually extends laterally
around the blastula and ultimately forms a nearly complete

334 GENERAL EMBRYOLOGY

ring, though this is not until the blastula has become converted
into the gastrula. This specialized group of cells may be
termed the germ ring, for it is evidently equivalent to the
structure already known by that name in the Fishes, Amphibia,
and other forms. At the time this rapid proliferation com-
mences, the vegetal hemisphere becomes flattened. The large
cells of this region then arch up slightly into the blastocoel and
soon begin to fold, or swing, inward about their postero-ventral
margin as a relatively fixed point (Fig. 151, B, C, D). This
motion is made possible by the rapid extension of a sheet of
cells which come off from the germ ring and which are thus
drawn in, to line the inside of the animal hemisphere. Without
going into details here, we may say that finally the inturning
of the vegetal cells becomes complete (Fig. 151, E) and the
blastula is converted into a cup-like structure, widely open
toward one side (the posterior or postero-dorsal).

The wall of the organism is now composed of two layers or
epithelia, the original blastoccelic cavity is nearly or quite
obliterated, and a new cavity is formed, lined by the inturned
cells, and widely open to the outside. This structure is the
didermic gastrula. The two cell-layers composing its wall are
the primary germ layers. The newly established gastrular
cavity is the archenteron or primitive gut cavity. The super-
ficial layer of cells, including the. original animal hemisphere of
the blastula and also some cells derived from the proliferating
area, is known as the outer germ layer, or ectoderm, or ectoblast,
or epiblast; the layer lining the archenteron, partly the cells of
the vegetal hemisphere of the blastula and partly the cells
derived from the proliferating region, is known as the inner
germ layer, or endoderm, or entoblast, or hypoblast. The opening
of the archenteron to the outside is the blastopore; the periphery
of the blastopore is spoken of as its margin or lip, and we have
seen that it is the region largely occupied by the germ ring:
it is here that the two primary germ layers are directly continuous
with one another.

The process of infolding, such as is carried out here by the
vegetal cells which come to line the ventral region of the archen-

BLASTULA, GASTRULA, AND GERM LAYERS 335

FIG. 151. — Gastrulation in Amphioxus. After Cerfontaine. A. Blastula
showing flattening of the vegetal pole and the rapid proliferation of cells in the
postero-dorsal region (germ ring). B. Flattening more pronounced; mitoses
in cells of germ ring. C. Commencement of the infolding (invagination) of the
cells of the vegetal pole. D. Continued infolding and inflection, or involution,
of ectoderm cells in the dorsal lip of the blastopore. The blastoccel becoming
obliterated and the archenteron being established. E. Invagination complete.
Continued involution in the dorsal lip of blastopore. Mitoses in germ ring.
F. Constriction of blastopore and commencement of elongation of the gastrula.
Remnants of blastoccel in ventral lip of blastopore. G. Gastrulation completed.
Continued elongation, and narrowing of blastopore. H. Neurenteric canal

336 GENERAL EMBRYOLOGY

teron, is known as invagination, and in some of the lower forms
the endoderm is wholly formed by this process. The process
of inturning, such as results in the lining of the dorsal region
of the archenteron by the cells derived from the margin of the
blastopore, is known as involution. In some forms the endo-
derm is largely formed by this process. Usually this is accom-
panied by the growth of the margin of the blastopore, or germ
ring, on over and past the involuted region, so that a layer of
cells is continually being overgrown, leading to more extended
gastrulation; this process of overgrowth may be termed epiboly.
The blastopore of the Amphioxus gastrula, at first widely
open, soon closes rapidly on account of the growth and epiboly
of its lip. This process leads also to the rapid elongation of the
gastrula (Fig. 151, F-H).

Although we have included involution and epiboly as gas-
trulation processes they are here more properly to be regarded
as processes leading to the formation of the notochord and the
intermediate layer. For as the gastrula continues to elongate
these structures begin to differentiate in the anterior part of the
roof of the archenteron, from that part of the inner layer formed
by involution and epiboly. Along the dorso-lateral regions of
the archenteron appear a pair of folds out of the archenteron,
each containing a narrow groove (Fig. 152, B, C). These folds
are the rudiments of the intermediate layer, or mesoderm, or meso-
blast; and the grooves, known as the enteroccelic grooves, are the
rudiments of the ccelom, the cavity of the mesoderm. That
portion of the archenteric roof between the mesoderm folds,
later becomes folded outward and forms the rudiment of the
notochord. The remainder of the archenteric wall, namely the
ventral and ventro-lateral regions "formed by the invaginated
portion of the inner layer, forms the primitive gut or enter on.
Whether we choose to call the involuted region really endoderm
or not, is immaterial; if we do, then we must say that the chorda

established by overgrowth of neural folds. Continued mitosis in germ ring.
a, animal pole; ar, archenteron; b, blastoporal opening; ch, rudiment of noto-
chord; dl, dorsal lip of blastopore; ec, ectoderm; en, endoderm; gr, germ ring;
nc, neuenteric canal; nf, neural fold; up, neural plate; s, blastocoel or segmenta-
tion cavity; v, vegetal pole; vl, ventral lip of blastopore; //, second polar body.

BLASTULA, GASTRULA, AND GERM LAYERS 337

ch

D

FIG. 152. — Transverse sections through young embryos of Amphioxus, showing
formation of nerve cord, notochord, and mesoderm. -After Cerfontaine. A.
Commencement of the growth of the superficial ectoderm (neural folds) above
the neural plate (medullary plate). B. Continued growth of the ectoderm over
the neural plate. Differentiation of the notochord, and first indications of meso-
derm and enterocoelic cavities. C. Section through middle of larva with two
somites. Neural plate folding into a tube. D. Section through first pair of
mesodermal somites, now completely constricted off. E. Section through middle
of larva with nine pairs of somites. Neural plate folded into a tube. Notochord
completely separated. In the inner cells of the somites, muscle fibrillse are form-
ing (compare Fig. 153). or, archenteron; c, enteroccel; ch, notochord; ec,
ectoderm; en, endoderm;/, muscle fibrillse; g, gut cavity; ra, mesoderm (gastral);
ms, mesodermal somite; nc, neural canal; nf, neural fold; np, neural plate (med-
ullary plate) ; nt, neural tube.

338

GENERAL EMBRYOLOGY

and mesoderm are both derived from endoderm and the process
of involution is to be regarded as a true gastrulation process.

But only the lesser part of the mesoderm is formed in the way
just described. This part of the mesoderm is known as the
gastral or parachordal, or axial mesoderm. If we trace the
mesoderm folds, just described, posteriorly, we can follow them
into the region of the germ ring or blastopore lip which has now
become considerably thickened on ac-
count of its contraction, and consists of
a more or less undifTerentiated mass of
cells. This mass now passes almost en-
tirely around the blastopore, laterally
and toward the ventral side. The
rapid proliferation of the cells of the
germ ring has early led to the disap-
pearance of the original simple epithe-
lial arrangement of its cells (Fig. 153),
but as it moves backward, with the
elongation of the gastrula, it leaves
behind it (i.e., anteriorly from it) its cell
products, which rapidly become differ-

entiated ^° C6rtam larS- On the

FK,153.-Frontal section

through larva of Amphioxus, surface of the embryo a layer is left

with six pairs of mesodermal -u« r, • i- ,1 ,• •,-, .-,

somites, at the level of the wnlcn 1S directly continuous with the
notochord and somites, ectoderm of the original gastrula de-

After Cerfontaine. a, • , -,

archenteron; e, enterocoei; rived irom the animal hemisphere of

irmt°ode' ^'af ±t:snd th*. blastula. Another layer is differ-

entiated lining the archenteron and

continuous with the layer already there, and consisting of a ven-
tral region continuous with the invaginated layer or true gut
endoderm, a dorsal region continuous with the involuted layer
forming the rudiment of the notochord, while dorso-laterally,
between these two regions, the inner sheet is continuous with
the mesodermal rudiments described above. In other words,
put of the germ ring there are gradually differentiated, true cover-
ing ectoderm, gut endoderm, chordal cells, and mesoderm. The
mesoderm formed in this way, directly from the germ ring, is

BLASTULA, GASTRULA, AND GERM LAYERS 339

called peristomial mesoderm, to distinguish it from the gastral
mesoderm formed in connection with the enterocoelic grooves.
Very soon, as we have seen, the region which gives rise to the
peristomial mesoderm comes to extend nearly to the ventral side
of the blastopore region. The ccelomic spaces of this peristomial
mesoderm are not formed as derivations of the archenteron; they
result from rearrangements of the mesodermal cells, and are en-
tirely independent of the enterocoelic portions of the coelom in
their origin, although the two become continuous later (Fig.
153). These two forms of mesoderm are directly continuous
with one another and have indeed a common primary origin, the
germ ring or margin of the blastopore. If we recognize the
essential difference between them as that of time of formation,
the altered circumstances surrounding the formation of each
due to this time difference, become of secondary importance as
regards our real conception of the mesoderm and its relations
to the other germ layers. Thus the relation of both chorda
and mesoderm proper to the cells of the monodermic blastula
is the same as that of the endoderm proper.

Stated in a word then, the gastrulation of Amphioxus is a
combination of invagination and involution, accompanied by
epiboly, and the processes of notogenesis and mesoderm forma-
tion are intimately bound up with the formation of the inner
layer.

Having become familiar now with the general ideas of gastru-
lation and the terminology of the process we may consider the
remaining forms of this process in the Chordates much more
briefly.

Our second type of gastrulation, as it occurs in the Amphibia
and Ganoid Fishes, may be easily understood by comparison
with the preceding. The chief differences result from the accu-
mulation of yolk in the vegetal pole of the ovum and blastula,
and the consequent comparative inertness of this region. That
is, the chief modifications of the typical process of gastrulation
appear in respect to the behavior of the lower pole, destined to
form the inner layer of the gastrula.

In the Amphibian blastula, the form of which was described

340 GENERAL EMBRYOLOGY /

above, a region of more actively dividing cells can be distin-
guished extending around the fully formed blastula just above
its equator, that is, between animal and vegetal poles (Fig. 154) ;
this is termed the germ ring, although it is frequently not a
complete ring, being interrupted on the anterior side of the
blastula. Gastrulation commences by the true invagination of
cells just below the germ ring on the posterior side. The inert-
ness of the large yolk-filled cells of the vegetal hemisphere
greatly impedes the process of invagination and it is never
carried very far. Involution occurs extensively in the dorsal

FIG. 154. — Section through late blastula of frog, showing location of germ ring.
Later the germ ring is thickened and contracted, forcing the yolk cells upward
into the segmentation cavity. After O. Schultze. a, animal pole; gr, germ ring;
p, pigment; s, blastoccel; v, vegetal pole.

region and is accompanied by very pronounced epiboly (Fig.
155); this latter process is very marked in the lateral and
ventral regions where little or no invagination occurs. All of
these processes are relatively less extensive than in the ccelo-
gastrula of Amphioxus however, probably on account of the
larger amount of yolk contained in all of the cells. Their place
in development is taken, in a way, by a new process, namely,
delamination. This is a process of splitting, whereby an
extended mass or sheet of cells becomes cleaved apart by the
rearrangement of the cells into two more or less distinct layers,
separated by a definite space. This process of delamination

E

FIG. 155. — Series of diagrammatic drawings of sections showing the process
of gastrulation in the Urodele, Triton. After Greil and Ruffini. F. Transverse
section, others sagittal. A. Blastula. B. Commencement of gastrulation
(invagination). C. Continued invagination accompanied by epiboly and invo-
lution. Formation of archenteron. D. Continuation of all three processes
of gastrulation. Blastocoel nearly obliterated. E. Archenteron completely
formed. Rudiments of notochord and neural plate differentiated. F. Trans-
verse section through embryo shown in E, through the plane marked ff. a,
archenteron (gut cavity in F) lined with endoderm; b, blastopore; ch, rudiment
of notochord; ec, ectoderm; en, qpdoderm; ff, plane of section shown in F; gm,
gastral (axial) mesoderm ; n, neural plate (in F, bounded by neural folds) ; pm,
peristomial mesoderm; s, blastocoel or segmentation cavity; x, marks corre-
sponding points on the surface ectoderm, showing extent of epiboly; y, yolk cells.

342 GENERAL EMBRYOLOGY

begins where the processes of imagination and involution leave
off, and it is important to recognize that the didermic character
of the gastrula of the Amphibian results partly from all three
of these processes. The chief result of involution is the forma-
tion of the rudiments of the notochord and gastral mesoderm,
as in Amphioxus. Figure 155 shows how the blastocoel is
finally obliterated by the invaginated and involuted regions.
The germ ring finally completes its growth over the yolk cells
or endodermal floor of the archenteron, and closes together much
as in Amphioxus.

The formation of the mesoderm offers some points of differ-
ence when compared with Amphioxus. The peristomial meso-
derm forms typically in the margin of the blastopore, out of
the undifferentiated cell mass of the germ ring. Sometimes, in
the region just within the blast opore dorsally, traces of entero-
ccelic outgrowths can be seen (Fig. 156), but most of the gastral

FIG. 156. — Part of a section through the body of an embryo of the frog, Rana
fusca, showing traces of enteroccel formation. After O. Hertwig. a, archen-
teron; c, enterocoels; ec, ectoderm; en, endoderm; m, mesoderm; n, notochord;
p, neural plate; y, yolk cells.

mesoderm is formed either from involuted cells derived from
the germ ring, or later from the surface of the endoderm by a
process of delamination or splitting off of the superficial cells
lying next the ectoderm; these come off first as a solid sheet,
which much later itself splits into two layers leaving a coslomic
cavity between them.

Thus while invagination occurs to a slight extent, gastrula-
tion in these forms results largely from the processes of involu-

BLASTULA, GASTRULA, AND GERM LAYERS 343

tion and epiboly combined with delamination. The didermic
condition results, to a considerable extent, from the overgrowth
of the animal hemisphere cells (germ ring) which come to enclose
the yolk cells of the vegetal hemisphere. As in Amphioxus,
however, the yolk, although here so much more abundant, is
finally included in the floor of the gut cavity, and the yolk cells
take a direct share in the formation of the structures of the
later embryo. After the mesoderm and chorda have been
formed from the roof of the archenteron, this is left as a thin
layer, only one cell in thickness, quite in contrast with the thick
mass of cells forming its floor (Fig. 155).

Turning now to the third type of gastrula, that formed from
the discoid blastula, we find conditions which vary widely from
the Amphioxus type, but which after all may be interpreted
in the light of the processes just outlined. In the Ganoid or
Amphibian, both the animal and vegetal hemispheres of the
egg share directly in the processes of cleavage, and blastula and
gastrula formation; and the yolk, contained in typical cells, is
carried directly into the wall of the primitive gut. But in
the extremely merolplastic eggs of the Elasmobranchs, Teleosts,
Reptiles, and Birds, the large yolk mass, which is the equivalent

FIG. 157. — Sagittal section through early gastrula of the catfish, Ameiurus.
en, endoderm; gr, germ ring; p, periblast; s, segmentation cavity, or sub-ger-
minal cavity; y, yolk.

of the vegetal pole of the egg, does not cleave (Figs. 150, 48)
and takes no direct share in the formation of the cellular blastula
and gastrula. For comparative purposes, therefore, we have
already seen that we must recognize the germ disc or " blastula"
of this type as equivalent only to the animal hemisphere of such
a form as the frog or Amphioxus, flattened out and resting
upon the undivided yolk mass. In such a condition as this the
equivalent of the germ ring would be found forming the

344

GENERAL EMBRYOLOGY

pm

FIG. 158. — Semi-diagrammatic drawings of sections through Elasmobranch
embryos. After Greil, after Riickert and Ziegler. A-D. Sagittal sections;
E, F, transverse sections. A. Diagram of section to show relations of blasto-
derm to animal pole of blastula with less yolk. B. Commencement of gastrula-
tion by invagination, epiboly, and involution. C, D. Continuation of all three
processes of gastrulation. E. Transverse section showing relation of endoderm
to yolk, mesoderm, and the germ ring. F. Transverse section farther forward,
through stage resembling D. Anterior margin of blastoderm toward the left in
A-D. a, archenteron; c, vestiges of enterocoels; ch, rudiment of notochord;
ec, ectoderm; en, endoderm; gm, gastral (axial) mesoderm; gr, germ ring; m,
mesoderm; n, neural groove, bordered laterally by neural folds; pm, peristomial
mesoderm; s, blastoccel or sub-germinal cavity; y, yolk.

BLASTULA, GASTRULA, AND GERM LAYERS 345

periphery oftJw blastodisc. -The process of gastrulation concerns
only the blastodisc, for the yolk mass takes no more share in
this than in the later processes of development; the gastrula
forms separately from the yolk, which is left outside the embryo
and as we shall see, comes into relation with it only indirectly.
The blastula or blastodisc, once formed as a flat plate, many
cells in thickness, begins to extend over the surface of the yolk
mass. Its central part becomes quite thin in consequence, but

pm

en

FIG. 159. — Diagrammatic drawings of sagittal sections through embryos of
Sauropsids. After Greil, after Will and Schauinsland. A, B. Two stages of
Reptile. C. Bird, a, archenteron; ch, rudiment of notochord; en, gut endo-
derm; pm, peristomial mesoderm; s, blastoccel or sub-germinal cavity; y, yolk.

the margin of the disc, or germ ring, remains thickened, and as
it advances over the yolk its margin becomes slightly involuted,
forming a narrow shelf of cells on its inner surface, toward the
yolk (Fig. 157). This inner layer is the rudiment of the primary
inner layer. Gastrulation is thus primarily accomplished by
involution. The margin of the germ disc is clearly the germ
ring or rim of the blastopore, although its form and relation to
the yolk are quite unlike what we have seen heretofore.

The Elasmobranchs and Reptiles afford important transi-
tional conditions here, in that a definite process of invagination
is indicated (Figs. 158, 159). Invagination is here limited to

346 GENERAL EMBRYOLOGY

the posterior margin of the blastoderm, where the germ ring
becomes elevated above the yolk as the endoderm is folded
under the ectoderm. In the Teleosts little or no indication of
invagination can be found. In these forms the germ ring
extends rapidly over the yolk, reaches and passes an equator
of the egg, and then as it continues to advance, gradually nar-
rows, and finally closes completely, having passed over the
entire yolk mass (Fig. 160). This overgrowth of the blasto-
derm occurs more rapidly -in the anterior and lateral directions
than in the posterior direction, so that the blastopore finally
closes in nearly the same relative position as in the frog and in
Amphioxus, i.e., postero-ventrally. As the germ ring extends
around the yolk, only a single, and very thin, layer of cells is
left behind it as a covering layer. In the posterior and postero-
lateral regions alone, is the involution of an inner layer well
marked. It should be noted that in the Teleosts the endoderm
is largely replaced functionally by a specialized protoplasmic
region on the surface of the yolk, known as the periblast, which
contains free nuclei derived originally from those of the margin
of the blastodisc (Figs. 150, C; 157).

During the later stages of the overgrowth of the germ ring,
as it contracts after passing the equator of the egg, its sub-
stance is payed into its more slowly advancing posterior region,
where it forms a longitudinal median thickening (Fig. 160).
This thickened region of the blastoderm is the primitive streak,
the earliest rudiment of the essential parts of the embryo, which
gradually differentiate out of its anterior end.

In such a gastrula as this the endoderm forms a flat median
plate of cells lying directly upon the surface of the periblast
(yolk), and the archenteron is present only virtually as a narrow
space between the endoderm and periblast (Fig. 157). In such
a case the formation of a true gut cavity is independent of the
formation of the inner layer, and occurs later by a process of
folding.

The mesoderm is differentiated at a comparatively early stage,
and the distinction between peristomial and gastral mesoderm
is very clear. The peristomial mesoderm appears as a small

BLASTULA, GASTRULA, AND GERM LAYERS 347

mass of slowly differentiating cells lying in the germ ring,
between the superficial ectoderm of the blastoderm and the
involuted shelf of endoderm (Fig. 158, F). The gastral meso-
derm is seen budding off laterally from the primitive streak
region, also between ectoderm and endoderm or even beyond the
region where the endoderm is found (Fig. 158, F). Posteriorly

FIG. 160. — Diagrams of the formation of the Teleost embryo by confluence of
the germ ring, and the growth of the germ ring around the yolk. From Kopsch.
A. In half-profile. B. In profile.

of course the gastral mesoderm of each side becomes continuous
with the peristomial mesoderm, and as peripheral portions of the
germ ring, where the three layers are slowly differentiating,
are continually passing into the posterior end of the primitive
streak, it is clear that peristomial mesoderm is constantly
becoming gastral, merely through relative change of position,
not through any change in the mode of its formation or in its
relation to the other germ layers.

The Elasmobranchs and Reptiles are again transitional in
that vestiges of enteroccelic grooves may be seen as shallow
and narrow longitudinal depressions, either side of the mid-
line, in the region of which the formation of mesoderm is most
rapid (Fig. 158, F). In the Teleost, as in the Bird, no traces of
enteroccels are to be seen. As usual the notochord forms from
the cell mass lying between the rudiments of the gastral meso-
derm, and may be said to have been derived either from the
gastral mesoderm, or from the endoderm in the same way that

348 GENERAL EMBRYOLOGY

the mesoderm itself is. After the separation of the chorda
and mesoderm, the endoderm proper, or enteroderm, as it is
called, is left as a thin narrow strip of cells spread flat over the
periblast (yolk) surface, continuous posteriorly with the diverg-
ing limbs of the germ ring.

In the Sauropsids, where the accumulation of the yolk is most
pronounced, the blastoderm does not grow entirely around the
yolk until long after the gastrula is formed and the embryo
established. Correlatively we find no typical germ ring forma-
tion in the periphery of the blastoderm, save in that posterior
region which is to be concerned in embryo formation. Remem-
bering that in the Reptile both true invagination and enterocoel
formation occur, while these processes are not apparent in the
birds, we may describe (following Patterson's account) the
processes of gastrulation and embryo formation in the pigeon,
as illustrating these events in the development of the extremely
meroblastic ovum.

The blastoderm first becomes quite thin, particularly toward
its posterior side, where, at the same time, the margin thickens
forming a segment of a true germ ring (Fig. 161). The exten-
sion of this posterior part of the germ ring, however, involves
the usual processes of cell multiplication accompanied by
involution and epiboly; there is no true invagination here (Fig.
161). The formation of an inner layer is thus limited to the
posterior region of the blastoderm. Soon, as this whole region
extends posteriorly, this segment of a germ ring begins to
contract toward the mid-line, and the result is the formation
of a median thickening in the posterior half or third of the germ
disc. This thickening is the primitive streak (Fig. 161), and as
usual it is the seat of the formation of the chief embryonic rudi-
ments. As in the Teleost, the primitive streak, formed by the
gradual fusion of the lateral halves of the germ ring, is obviously
the equivalent of the blastoporal margin of the frog or of
Amphioxus. On its surface is a shallow longitudinal groove
marking the separation of the two halves; this is the primitive
groove (Fig. 161), which may be regarded as representing the
blastopore proper. The archenteron, in such a gastrula as this

BLASTULA, GASTRULA, AND GERM LAYERS 349

r-O

Hi

350 GENERAL EMBRYOLOGY

of the chick, may also be said to exist only virtually, for it is
represented only by a shallow space left between the endoderm
and the yolk.

The mesoderm and chorda are here more closely related with
the outer than with the inner layer. In the germ ring there is
little indication of separation of the germ layers, other than the
distinction of the endoderm, and when the primitive streak is
formed, it appears rather as a thickening of the ectoderm. The
mesoderm begins to be differentiated along the sides of the
primitive streak, and back as far as the region where this is
being formed by the fusion of the limbs of the germ ring. Hence
the mesoderm is more largely gastral, that is to say, it does not
become distinct, as a separate rudiment, until the establishment
of the primitive streak has occurred. In the germ ring there
is of course a region where ectoderm passes into endoderm, and
where the cells may be said to belong to either layer or neither
layer. This is the region where the primary " mesoderm"
forms and apparently special conditions may determine with
which of the primary layers it may seem to have the more inti-
mate relation. Little is gained by attempting to define germ
layers in the germ ring.

In the pigeon or chick the rudiment of the notochord appears
in the deeper part of the primitive streak after its lateral parts
are cut off as mesoderm. The endoderm has therefore the value
of an enteroderm from the beginning, and has the form of a very
thin flat sheet of cells widely spreading over the yolk surface.
Ultimately the yolk mass becomes entirely enclosed in a layer
of endoderm as well as by the other germ layers, but this does
not occur until a comparatively late stage in the development
of the embryo.

In Amphioxus and the frog we have seen that the embryo is
formed from the entire ovum, that is, the yolk-containing cells
become actually included within the wall of the gut. In the
Teleosts the yolk mass is so large, and so completely separated
from the embryogenic tissues, that the embryo may be said to
develop upon the surface of the yolk, which, enclosed within a
structure called the yolk sac, is only indirectly related to the

BLASTULA, GASTRULA, AND GERM LAYERS 351

embryo proper. In forms like the Elasmobranchs and
Sauropsids, the accumulation of yolk is still greater and the
embryo forms quite apart from the yolk, with which it later
acquires a secondary relation. In the Sauropsids, after the
rudiments of the embryo are well established, a process of fold-
ing begins and a series of infoldings of the cellular blastoderm,
anterior, posterior, and lateral, pinch off the embryo from the
yolk mass or yolk sac, with which it then remains only indirectly
connected by a narrow tube known as the yolk stalk which
includes a portion of the gut wall and a very abundant blood
supply.

In the Sauropsids and Mammals other folds of the blastoderm soon
appear, beyond the limits of the embryo proper, which result in the
formation of a very special and highly characteristic structure known
as the amnion. And from the wall of the hind-gut grows out another
special and extra-embryonic structure, the allantois. The formation
and function of these extra-embryonic structures, together called the
embryonic appendages, cannot be described here. They are of the
greatest importance in development and their presence has led to the
application of the term Amniota to all the forms possessing them
(Birds, Reptiles, Mammals) while the other Craniates, without these
embryonic appendages (Cyclostomes, Fish, Amphibia) are then known
as the Anamnia.

On account of the difficulties of comparison it seems wise to
omit reference here to the Mammalian gastrula and germ layer
formation. For in spite of the nearly alecithal condition of the
Mammalian ovum, its development shows marked yolk influ-
ence, and the whole course of early development is complicated,
not only through the one time presence and the subsequent loss
of yolk, but through the very special relations of the early
embryo, and particularly the embryonic appendages, with the
walls of the maternal cavity in which development proceeds.

CONCRESCENCE

We should consider here, in a particular way, a developmental
process which, besides being of great general importance in
Chordate development, is of considerable historical interest as

352

GENERAL EMBRYOLOGY

well. In the foregoing pages we have seen that where a germ
disc is formed, its margin, known as the germ ring, and recog-
nized as the homolog of the lip or margin of the blastopore, is of
great importance in the formation of the primary rudiments of
the embryo.

The His- Whitman theory of concrescence emphasizes the
general importance and significance of this relation. First
stated fully by His, in 1876, the essential idea of this theory was
that each side of the germ ring, not only forms, but really is,

FIG. 162. — Diagrams illustrating four stages in the formation of the Tel cost
embryo and the growth of the germ ring around the yolk mass. After O. Hert-
wig. e, embryo; gr, posterior margin of the germ ring; y, yolk mass; 1, 2, 3, 4,
successive positions occupied by the germ ring as it advances over the yolk.

the rudiment of the corresponding half of the embryo, which is
thus actually formed by the approach and gradual, continued
fusion posteriorly of the germ ring. In each half of the ring
the essential rudiments of the embryo were thought to be al-
ready formed, partly at least, and the process of embryo forma-
tion consisted merely or chiefly in the junction or addition of
these two originally separate halves. The anterior end of the
embryo would thus be formed first, and embryo formation

BLASTULA, GASTRULA, AND GERM LAYERS 353

would be complete when the germ ring became fully contracted
or closed.

With some modifications of a really fundamental kind, this

D

FIG. 163. — Diagrams of the formation of an embryo by confluence ("con-
crescence"). A. Germ ring before formation of the embryo is indicated. The
letters a-e, represent symmetrical portions of the germ ring. B. Beginning of
confluence. C. Embryo forming. A A, BB, represent regions of the embryo,
formed out of the materials of the germ ring at aa, bb. D, E, Later stages in the
formation of the embryo. The germ ring regions, cc, and dd, have been differ-
entiated into the embryonic regions, CC, DD.

conception is widely adopted to-day. Both observation and
experiment have shown, however, that definite halves of an
embryo cannot be said to exist preformed in the lateral portions
of the germ ring. These regions do contribute to the formation

354

GENERAL EMBRYOLOGY

of the median thickening, known as the primitive streak or
embryonic rudiment, by a process of gradual fusion posteriorly
(Figs. 162, 163). But in this process of coming together, which
may better be termed confluence (Sumner) than concrescence,
the materials from the two sides of the germ ring are fused into
a mass which is largely ^differentiated, and out of this the
rudiments of the embryo appear, by a process of differentiation
which occurs largely after confluence. One side of the germ
ring contains not a half of the embryo, but the substance out of
which, later, a half of an embryo forms. This process of
differentiation is progressive and commences of course in that

164. — Sagittal section through the hinder end of a fish embryo (Serranus),
showing the undifferentiated primitive streak, anterior to which the structures
of the embryo are being differentiated. From H. V. Wilson, a.p. (v.L), anterior
margin of blastoderm or ventral lip of blastopore, after having grown entirely
around the yolk mass. bl., blastopore; ec, ectoderm; en., endoderm; g.r., germ
ring; k.v., Kupffer's vesicle; nc., notochord; nr. ch., nerve cord; p., periblast;
pp. (d.l.), posterior margin of blastoderm (dorsal lip of blastopore); pr. str.,
primitive streak.

part of the primitive streak formed first, i.e., its morphological
anterior end (Fig. 163). Then as the primitive streak lengthens
posteriorly, the extent of the differentiated region at its anterior
end similarly increases posteriorly, roughly keeping pace with
the process of elongation (Fig. 164). The primitive streak
thus may be regarded as a region which moves backward,
receiving posteriorly the diverging limbs of the germ ring, and
leaving anteriorly the differentiated rudiments of the embryo.
Soon after the germ ring is completely closed or contracted,
the primitive streak becomes wholly differentiated, and the

BLASTULA, GASTRULA, AND GERM LAYERS 355

rudiments of the embryo may be said to be fully marked out.
The process of concrescence is seen most clearly and typically
in those forms with large amounts of yolk and with well-marked
germ ring, especially in the Teleosts and Elasmobranchs (Figs.
160, 165). In the Amphibia, where the amount of yolk is less,
and the Sauropsida, where the germ ring is less marked, the proc-
ess of concrescence, though somewhat modified and slightly
obscured, still takes an important part in embryo formation.

THE GERM LAYERS
While this is not the place to give an historical or critical

FIG. 165. — Blastoderms of the Elasmobranch, Torpedo, showing formation of
the embryo. After Ziegler. X 27. A. "Stage B." The postero-median
thickening is the "embryonic shield," the first indication of the real embryo.
B. "Stage C." Early embryo; nerve cord rising above the surface of the blasto-
derm. In both figures the embryonic portion of the blastoderm is directly
continuous postero-laterally, with the germ ring, which appears as the thickened
margin of the blastoderm.

account of the germ layer theory, it is important that the stu-
dent should have in mind, before taking up the study of the
development of particular organisms, certain fundamental
conceptions of the germ layers, and their relation to develop-
ment, particularly among the Chordata.

When the science of Embryology was itself in a very early
stage of its development, the earliest differentiations recognized
by the students of animal development, were the sheets or

356 GENERAL EMBRYOLOGY

layers of tissue, such as those of the chick, which seemed to give
rise to the chief organs of the embryo. These layers of sub-
stance were described and their significance recognized, by
such pioneers in embryology as C. F. Wolff (1768), Pander
(1817), and Von Baer (1828). The whole history of the forma-
tion of the systems and organs of the embryo could be read back
to these layers, but beyond these, few constant structural
features, susceptible of homolgy in different forms, could be
made out. The idea became very firmly fixed therefore, that
these layers were really the primary differentiations of the
embryo, and quite naturally their importance was strongly
emphasized.

Subsequent to the statement and establishment of the cell
theory, the genesis of the germ layers was traced, theblastula
and gastrula fully described in a great variety of forms, and it
was found that in spite of the greatest diversity in the earlier
processes of development, the general character and structure
of the germ layers remained remarkably uniform. And not
only were the relations of the germ layers to one another quite
constant, but their relation to the tissues and organs of the
later embryo were subject to but little variation. In all forms
inner and outer layers (endoderm and ectoderm) were present,
and in all forms above the Ccelenterates a definite intermediate
layer (mesoderm) was to be found. Moreover, in all of these
forms the outer layer gave rise to the whole nervous system,
central and peripheral, the essential parts of the sense organs,
the epidermis and its appendages; from the inner layer came the
lining of the digestive tract and its glandular appendages;
while the intermediate layer gave rise to the sustentative,
vascular, and muscular tissues throughout the body. All of
this was finally developed, notably by the Brothers Hertwig
(1879-1883), into a carefully and elaborately worked out Germ
Layer Theory, the essential points of which were that the three
germ layers are entirely homologous throughout the Metazoa,
excepting only the Porifera (the Ccelenterates of course lacking
a middle layer), and that these layers truly represent the pri-
mary and fundamental homologies in the structure of the

BLASTULA, GASTRULA, AND GERM LAYERS 357

Metazoan phyla. Exceptions and contradictions were indeed
occasionally noted, but their importance was minimized and
they were treated frankly as exceptions, and put down to the
account of " coenogenetic " modifications of " palingenetic "
characteristics (see Chapter I).

It is difficult to overestimate the influence of this theory upon
the history of Embryology, and upon fundamental embryo-
logical ideas. Perhaps no conception, other than the general
theory of evolution, has had greater influence in the field of
descriptive embryology.

More recently, however, the limitations in the general
applicability of this theory have been more fully recognized
and the exceptions to the validity of its essential ideas empha-
sized. At present we must recognize the germ layers as
representing a stage in development, just as do the blastula or
gastrula, and of no greater or lesser importance than these.
The germ layers are descriptive terms of the greatest .impor-
tance, as such they are indispensable. They are not, however,
starting points in any real sense; and to regard them as such is
to look forward merely, not backward. Looking both forward
and backward we see that the establishment of the germ layers
is only one step in the continuous process of development.
They represent no more 'essential homologies than many other
features held in common by many developing organisms.

While we cannot consider in extenso the facts which have led
to this change of opinion regarding the importance of the germ
layers, we are bound to state the nature of certain classes of
these facts. In the first place are to be noted the difficulties
of homologizing the layers of certain groups with their typical
condition. For example, in the Porifera that layer which seems
entitled to be termed the ectoderm, really gives rise to struc-
tures ordinarily derived from endoderm, while the "endoderm"
itself forms the covering tissues. In the Mammals the "ecto-
derm" may contribute little or nothing to the formation of the
real embryo and the inner, outer, and middle layers cannot
be exactly homologized with those of other Chordates, save by
the grace of terminology. In the earlier part of this chapter

358 GENERAL EMBRYOLOGY

the varied relations of the mesoderm to the other layers were
mentioned; in some cases the middle and inner layers arise
from a common rudiment, in others the middle and outer
layers. Among the Invertebrates there are many instances of
development where even the two primary layers are to be made
out only with considerable difficulty, as for example, in the
Trematodes, Cestodes, certain of the Bryozoa, etc.

In the second place the morphogenetic value of the individual
layers is subject to a considerable variation. Thus in the
Chordata, leaving aside the Mammals, the mesenchymal con-
nective-tissue cells may be occasionally of "ectodermal" or
"endodermal," as well as of "mesodermal" origin. The endo-
thelium of the heart may be "endodermal" or "mesodermal."
The notochord may with equal correctness be described as
endodermal, mesodermal, or even ectodermal, in various forms.

Single organs like the nephridia may be composites, ecto-
dermal and mesodermal, or, in some cases ectodermal, in others
mesodermal.

In the process of regeneration certain contradictions to the
germ layer theory become apparent. Organs and tissues nor-
mally derived during embryonic development from a certain
layer may, during regeneration, be produced from another
layer. In certain OligochaBtes new mesoderm is of ectodermal
origin, and the regenerated pharynx may be lined with endo-
dermal, rather than ectodermal cells.

Especially in the process of budding, as it occurs in a great
many groups, do we find abundant exceptions to this theory.
In some of the Polyzoa the gut may be of ectodermal origin;
the nervous system and pharynx are mesodermal in some of
the flatworms. Analogous conditions are very common among
the Tunicates; here the pharynx may be endodermal or ecto-
dermal; the atrium and even the nervous system may be ecto-
dermal, mesodermal, or endodermal, in different forms where
in egg development the relations of these structures to the
germ layers are typical.

Finally, the most important qualifications and limitations of
the germ layer the'ory grow out of the observed facts of normal

BLASTJJLA, GASTRULA, AND GERM LAYERS 359

development prior to the formation of the germ layers. These
structures are by no means the earliest constant embryonic
differentiations, and as we have seen in the chapter on cleavage,
it is just as easy to draw homologies between cell groups in the
blastula stage, or in an earlier cleavage stage, as it is between
the later appearing germ layers. It is not too much to say
that in some cases homologies may be drawn between various
formed substances in the undivided egg. Animal and vegetal
poles of the ovum, cleavage patterns, cell-groups, micromeres,
macromeres, upper and lower poles of the blastula, are all
constant and comparable features of development no less than
inner, outer and middle germ layers. We may recall that the
cell known as 4d may be identified and its history and fate
compared, in the cleavage of many groups, even in different
phyla. This cell whose form, position, and derivation are so
constant, may or may not form "mesoderm"; even when it
does form mesoderm this may go to form very different parts of
the embryonic and adult structure. Often the " mesoderm"
may be a cell, just as truly as a layer.

Summarizing we may say that while the arrangement of the
cells of the embryo in the form of definite layers is almost
universal, at the same time, in the comparison of different
groups or of different modes of development, these layers
exhibit great inconstancy in their relations to one another,
and to the structures forming them and formed from them.
The germ layers are valuable, indeed indispensable descriptive
units, but they do not represent primary differentiations, and
their homologies are no more, though probably no less, funda-
mental throughout groups larger than phyla, than are many
other structures of the developing organism.

MORPHOGENETIC PROCESSES

It remains now to describe some of the more general processes by
which the rudiments of the organs and tissues of the embryo may be
formed out of the layers or cell masses of the gastrula and post-gastrula
stages. We shall not attempt to describe here the actual formation of
any specific embryonic structure, but rather shall give a brief classifica-

360 GENERAL EMBRYOLOGY

tion of the more common and important processes, a few of which have
already been mentioned earlier in this chapter.

While the morphogenetic processes within the embryo show the
greatest diversity and vary almost infinitely in specific details, yet it is
possible to include them all under a few heads, when these matters of
detail are omitted. Again it should be recalled that we are limiting
our description to the Chordata.

The primary condition of morphogenesis is cell multiplication. After
each division the daughter cells increase to practically the size of the
parent cell; and numerical increase in cells, together with their growth,
i.e., cell proliferation, play either a primary or a secondary part in every
morphogenetic process. When the process of cell division is quite
general throughout the extent of the germ disc or layer, the result is an
increase in the thickness or in the extent of the sheet, respectively, when
the plane of the cell divisions is in general parallel with, or perpendicular
to, the plane of the whole layer. If there should be little or no regularity
in the positions of the division planes, the membrane would increase in
al) directions (Fig. 166).

Ordinarily, in embryogeny, cell multiplication and growth are more
intense in restricted areas of the blastoderm or germ layer. It is
convenient then to distinguish between (a) those processes in which the
multiplying cells tend to remain associated in the same -general region,
and (6) other processes where they become more or less separated, either
from one another or from their seat of origin. Under the former head
we must again distinguish between the results of increase in thickness
and in extent. A localized increase in thickness is frequently termed a
bud; buds may project either above (limb bud) or below (Teleostean lens)
the free surface where they are formed. If the thickening region is
elongated the result may be the formation of a strand or plate of cells,
again either a ridge-like structure above the surface of the layer (genital
ridge), or a keel-like thickening below the surface (Teleostean nerve
cord, in part).

Increase in extent of a localized area frequently involves the obstruc-
tive action of the region bounding the area. When this form of growth
occurs generally, so that the tendency to extension occurs in every
direction from the middle of the area concerned, the result is frequently
an arching, either outward or inward. This may take the form of a
simple arching as in the Teleostean blastula (Fig. 150, C), or the same
process may be carried farther and followed by a constriction near the
base. Such processes are very common indeed and are termed in-
vagination and evagination, according as the growth is below or above
the free surface. Simple illustrations of evagination are afforded by
the formation of intestinal villi, the rudiments of lung or thymus, and

BLASTULA, GASTRULA, AND GERM LAYERS 361

the like; typical invaginatiojis are seen in the formation of theoptit
cup out of the optic lobe, the auditory sac, etc. When this form of
growth in extent is limited to certain directions instead of occurring
radially, the result is often the formation of a fold which bears some-
what the same relation to the dilation that the strand does to the bud.
The fold also may be above the surface of the membrane, forming a
sort of arch or hollow ridge, usually bounded by lateral depressions
(frog's pronephric duct), or below the surface forming a groove or furrow
bordered by lateral elevations (medullary groove) (Fig. 167). In some
instances a solid strand may be formed in this way instead of by the
simpler process of direct increase in thickness (Teleostean nerve cord,

HS3

FIG. 166. — Diagrams of four stages in the formation of an epithelial thickening,
several layers of cells deep. From Korschelt and Heider.

in part). When the processes leading to the formation of a groove are
continued, the groove may be converted into a closed canal by the
approach, apposition, and fusion of the borders (neural tube) (Fig. 167).

In those conditions where the proliferating cells become, to some
extent, separated either from one another or from the proliferating
region itself, we may note first, instances of actual cejl migration.
This may be either emigration or immigration, according as to whether
we fix attention upon the source or the destination of the migratory
cells (mesenchyme cells) . Secondary processes of thickening or thinning
may accompany these processes. In other cases the movement of cells
may be described as rearrangement rather than migration; this may be
illustrated by the formation of mesodermal somites and blood islands
(chick).

One of the common morphogenetic processes is a combination of
increase in thickness and cell rearrangement, such as the usual forma-

362

GENERAL EMBRYOLOGY

tion of the notochord, or the process of delamination, which consists
in the splitting of a single thickened sheet into two separate layers,
either as a whole or in localized areas (formation of mesoderm in the
frog, or division of the mesoderm into somatic and splanchnic layers) ;
in some instances the initial thickening may not be very apparent.-

Occasionally cells of different layers, or of different rudiments, meet
and fuse, forming a continuous rudimentary mass (pituitary body) .

c

FIG. 167. FIG. 168.

FIG. 167. — Diagrams of the formation of the medullary canal or neural tube, ia
the Vertebrates. From Korschelt and Heider.

FIG. 168. — Diagrams of the formation of the mouth and stomodseum in a
typical form. From Korschelt and Heider. ec, ectoderm; md, pharynx;
td, stomodseum.

Finally we may mention certain morphogenetic processes of a wholly
different kind, namely, resorption, and changes in the form and size of
cells. From this point of view merely, the process of resorption may be
regarded as the reverse of proliferation. Definite rudiments may appear
first as spaces thus formed by the gradual dissolution and absorption of
cells in certain areas. Thus the oral and anal openings, gill-clefts, etc.,
are usually formed as "perforations" by the resorption of areas where
previously separate and continuous layers became united by a process
of fusion (Fig. 168). In other cases rudiments once established may

BLASTULA, GASTRULA, AND GERM LAYERS 363

gradually disappear, either wholly (tail of tadpole), or in part
(pronephros) remaining then as vestigial organs.

Changes in cell form and size chiefly fall under the head of tissue
differentiation, or histogenesis, but in a few instances such processes are
primarily involved in the formation of rudiments. The development
of the eye affords illustrations of these processes; the lens forms as a
thickening of the cells on one side, and the thinning on the other, of
a sac originally formed by invagination, and the optic lobe, at first
nearly spherical, flattens and invaginates, one layer becoming thickened
as the chief part of the retina (later complicated by cell proliferation)
while the other layer forms a thin pigmented layer.

These processes of morphogenetic value are tabulated in the accom-
panying summary; the arrangement here is merely a convenient one
and has no other significance.

It is extremely important to recognize that these morphogenetic
processes described above are not merely simple mechanical processes.
The arrangement and behavior of the cells in an invaginating or delami-
nating region are determined by other factors than those of physical
resistances, attractions, etc. These events are both in fundamentals
and in details to be regarded as active phenomena of a living organism.
They are frequently also to be understood from the historical or adaptive
points of view.

What the precise conditions are which determine the nature of
these events, may usually only be conjectured. In some cases they may
result from osmotic conditions, absorption of water, etc. But for the
most part the nature of the specific stimuli, and the conditions within
the layer or cell group, which lead to the definite reactions of rudiment
formation are unknown. And in this particular field of development we
can do little more than to describe what happens from the morphological
viewpoint.

364 GENERAL EMBRYOLOGY

SUMMARY OF THE CHIEF MORPHOGENETIC PROCESSES
I. Cell division and growth throughout the layer or disc, resulting in

(a) Increase in thickness.

(b) Increase in extent.

(c) Increase in both thickness and extent.

II. Cell division and growth localized in restricted areas of the layer

or disc.

A. Cells remain related and in continuity
(a) Increasing in thickness.

1. Radially— formation of buds I ™1^'

| hollow.

2. Chiefly in one axis — formation of strands I -^ ? "

(6) Increasing in extent.

Dilation.

1. Radially

Invagination.
Evagination.

Groove.

2. Chiefly in one axis

Folds.

Tube (strand secondarily).
B. Cells become separated to a varying degree.

f Immigration ) with corresponding

(a) Migration <«.-,. > .u. , . JAU- •

[ immigration J thickening and thinning.

(b) Rearrangement.

(c) Delamination (sometimes preceded by thickening).

(d) Fusion.

III. Resorption.

(a) Accompanying a process of perforation.
(6) Disappearance (degeneration).

IV. Changes in cell form and size.

(a) Thickening.
(6) Thinning.

REFERENCES TO LITERATURE

CERFONTAINE, P., Recherches sur le deVeloppement de PAmphioxus.

Arch. Biol. 22. 1906.
DAVENPORT, C. B., Studies in Morphogenesis. IV. A preliminary

Catalogue of the Processes concerned in Ontogeny. Bull. Mus.

Comp. Zool. Harvard Coll. 27. 1896.
EYCLESHYMER, A. C., The Formation of the Embryo of Necturus, with

Remarks on the Theory of Concrescence. Anat. Anz. 21. 1902.
GREIL, A., Ueber die erste Anlage der Gefasse und des Blutes bei

BLASTULA, GASTRULA, AND GERM LAYERS 365

Holo- und Meroblastiern. Verb. Anat. Gesell., in Anat. Anz. 32.

1908.
HERTWIG, 0., Die Lehre von den Keimblattern. Handbuch, etc.

I, 1, 1. 1903 (1906).
HERTWIG, 0., und R., Studies on the Germ Layers. Jena. Zeit.

13-16 (6-9). 1879-1883.
His, W., Untersuchungen tiber die Entwickelung von Knochenfischen,

besonders iiber diejenige des Salmens. Zeit. Anat. Entw. 1.

1876. Untersuchungen liber die Bildung des Knochenfischem-

bryo. Arch. Anat. Entw. 1878.
JENKINSON, J. W., (Ref. Ch. VII.)
KEIBEL, F., Die Gastrulation und die Keimblattbildung der Wirbeltiere,

Ergebnisse Anat. u. Entw. 10. 1900 (1901).
KOPSCH, F., Untersuchungen tiber Gastrulation und Embryobildung

bei den Chordaten. I. Die Morphologische Bedeutung des Keim-

hautrandes und die Embryobildung bei der Forelle. Leipzig. 1904.
KORSCHELT UND HEADER, Lehrbuch, etc. I Abschnitt. Experimented

Entwicklungsgeschichte. Jena. 1902. Ill Abschnitt. Furchung

und Keimblatterbildung. Jena. 1909-1910.
LILLIE, F. R., The Development of the Chick. New York. 1908.
PATTERSON, J. T., On Gastrulation and the Origin of the Primitive

Streak in the Pigeon's Egg. — Preliminary Notice. Biol. Bull.

13. 1907.
PRENANT, A., BOUIN, P., ET MAiLLARD, L., Traite d'Histologie. Paris.

1911.
SCHAUINSLAND, H., Beitrage zur Entwickelungsgeschichte und Anatomie

der Wirbeltiere. I, II, III. Bibliotheca Zoologica. 39. Stutt-
gart. 1901-1903.
SCHULTZE, 0., Ueber das erste Auftreten der bilateralen Symmetrie im

Verlauf der Entwicklung. Arch. mikr. Anat. 65. 1900.
SELYS-LONGCHAMPS, M. DE, Gastrulation et formation des feuillets chez

Petromyzon planer i. Arch. Biol. 26. 1910.
SUMNER, F. B., Kupffer's Vesicle and its Relation to Gastrulation and

Concrescence. Mem. N. Y. Acad. Sci. 2. 1900. A Study of

early Fish Development. Experimental and Morphological. Arch.

Entw.-Mech. 17. 1903.

WHITMAN, C. O., The Embryology of Clepsine. QJ.M.S. 18. 1878.
WILL, L., Beitrage zur Entwicklungsgeschichte der Reptilien. I.

Die Anlage der Keimblatter beim Gecko (Platydactylus facetanus).

Schreib. Zool. Jahrb. 6. 1892.
WILSON, H. V., (Ref. Ch. VI.)
ZIEGLER, H. E., (Ref. Ch. III.)
ZIEGLER, H. E., und F., Beitrage zur Entwickelungsgeschichte von

Torpedo. Arch. mikr. Anat. 39. 1892.

INDEX

(The numbers in black-faced type refer to pages with illustrations.)

Achromatic figure, 51

acrosome, of spermatozoon, 98

Actinophrys, conjugation, 192; matu-
ration, 166; plastogamy, 189

Acttnosphcerium, chromosome num-
ber, 66

Adolphi, 166

agglutination, of spermatozoa 167

alecithal ova, 93, 226

allantois, 351

allosome, 310

alveoli, of protoplasm, 36

Amblystoma, spermatophore, 105

Ameiurus, gastrula, 343

Amia, cleavage (incomplete), 241

amitosis, 43

amnipn, 351

Amniota, 351

Amceba, autogamy, 191; conjugation,
192; cytotropy, 189; fission, 2;
karyosomes, 60, 61

Amceba diploidea, maturation di-
visions, 160

amphiaster, 51; mechanism of for-
mation, 80

Amphibia, blastula, 331; gastrula-
tion, 339, ff.; 341

amphigony, 12

amphimixis, 213

Amphioxus, blastula, 220, 330, 331 ;
cleavage (radial), 235; fertiliza-
tion, 175; gastrulation, 333, ff.,
335 ; notogenesis and mesoderm
formation, 337, 338; oocyte, 90;
spawning time, 103

Amphitrite, ovary, diagram, 110 ;
spawning season, 103

Amphiuma, spermatogenesis, 124;
spermatozoon, 99

amplexus, 105

Anamnia, 351

anaphase, 53

Anasa, chromosomes, 68 ; - - in
relation to sex, 307, ff.; spermato-
genesis, 308

Andrews, 149

animal pole, of ovum, 92

anisogamy, 192, 196, ff.; facultative,
196, 197

annulus, of spermatozoon, 100

Apus, parthenogenesis, 203

Arbacia, egg structure. 275, 276

Arcella, plastogamy, 190

archenteron, 334

archoplasm, 38

Argonaut a, micropyle, 96

Aristotle, 1

armadillo, multiple embryos, 315

Artemia, maturation, 158

Ascaris, cell lineage, 253, 255;
chromatin elimination during
early cleavage, 65; chromosomal

continuity, 72, 74 ; variation

(numerical), 66, 67; chromosome
groups during cleavage, 223;
cleavage, 254, 257; fertilization,
182; fused ova, 282; "giant"
polar bodies, 148, 149; idiochro-
mosomes, 311; primordial germ
cells, 112; segregation of germ,
133; spermatozoon, 99; tetrad

formation, in oogenesis, 145 ;

in spermatogenesis, 137

asexual reproduction, 12

aster, 39, 49

Asterias, chromosomes from chro-
matin reservoir (nucleolus), 71;
development of parts of gastrula,
283

Astropecten, maturation in partho-
genesis, 159

attraction cone, 168; sphere, 39

autogamy, 191

autonomy of male and female chro-
mosome groups, 223

axial filament, of spermatozoon, 98

von Baer, 91, 355

Balfour, 93; Law of Cleavage,

230

Baltzer, 75, 172, 303, 304, 311, 314
basal cells, 121
Bataillon, 303
bee, relation of sex to fertilization,

315
Begonia, reproduction from cuttings,

165

van Beneden, 41, 65, 66, 80, 91, 131
Bernstein, 210
Beroe, cleavage (disymmetrical), 238

367

368

INDEX

Bigelow, 230

Bilharzia, in copula, 106

Biogenetic Law, 25, ff.

bipphores, 292

Birds, gastrula and gastrulation,
345, 349

blastocoel, 221, 331

blastoderm, 241, 332

blastodisc, 240, 332

blastomeres, homologies, 256, 257;
isolated, development of, 269, ff.,
280, ff.; nomenclature, 248, ff.

blastopore, 334

blastula, 19, 221, 330; types, 220,
221

Bodo, conjugation, 196

Bonnet, 22

Boring, 311

Born, 281

Boveri, 70, 74, 81, 137, 300, 301, 305,
311

Brachystola, chromosomal individu-
ality in spermatogonium, 73;
chromosomes, 68, 73

Brauer, 158

breeding habits, 102, ff.

bridges, intercellular, 32, 42

brood cavities, 104

brood formation, 3; in Mycetozoa,
etc., 190

budding, 6

bud-fission, 7

Buller, 166, 167

Butschli, 81, 209

Bythotrephes, spermazotoon, 108

Calcium, effect upon blastomeres,
318-319

calcium-free sea water, effects upon
cleavage stages, 271-272

Calkins, 191, 210; and Cull,

64, 157, 296

Canthocamplus, intranuclear spindle,
67 ; oogenesis, 116

Cavia, metamorphosis of spermatid,
126

cell, continuity, 42; definition, 31;
diagram of, 35; displacement, in

cleavage, 231 ; division, 43, jf.;

causes of, 76, ff.; - - indirect,

44, ff.; mechanism, 79, ff.;

plane, 55, 56 ; forms, 32, 33 ;

general account, 3 l,ff.; interaction
hypothesis, 263-264; lineage, 247;
migration, in morphogenesis, 361;
multiplication, in 06- and sper-
matogenesis, 113, 114; polarity,
41; proliferation, in morphogene-
sis, 360; structure, 34, ff.; Theory,
in Embryology, 20; wall, 34

central body, 61

centrifugal force, effects upon egg
structure, 275, ff.

centrolecithal ova, 94, 226

centronucleus, 61

Centropyxis, nuclear division, 60

centrosome, 38, ff.; " artificial," 183,
207; continuity, during fertiliza-
tion, 181, 183; formation by
spermatozoon, 181; in Protozoa,
60, ff.; origin, in Protozoa, 60-62

centrosphere, 39, 125

Ceratocephale, spawning season, 103

Cercomonas, conjugation, 193

Cerebratulus, cleavage, 236 ; develop-
ment of isolated blastomeres, 287 ;
fertilization, 184 ; maturation of
ovum, 143 ; organization of ovum,
development of, 287

Cerfontaine, 333

Chcetopterus, differentiation without
cleavage, 279, 279; organization
of ground substance of ovum, 278 ;

ovum, 268 ; polarity of

ovum, 275; structure of ovum
following fertilization, 268

chemicals, effects of, upon differen-
tiation, 318, ff.

Chilomonas, cell division, 59, 64

Chimcera, egg case, 109

Chlamydomonas, conjugation, 193,
194; gametes, 198

Chlamydophrys, nuclear division, 61

chondriosomes, 41

"chondromiten," 41

chordaplasm, 267

chorion, 96, 120

chromatin, 38; elimination, 65; ex-
trusion, 45; possible signifi-
cance, 298; granules, 293, 294;
nucleoli, 38; reservoir, 71

chromidia, 41, 156

chromioles, 38; in reduction, 153

chromosomes, 49-51; accessory, 310;
as determiners in differentiation,
289, ff.; as factors in heredity, 291,
ff.; behavior, as related to Men-

delian heredity, 294, 295;

during cleavage, 223, 224;

maturation and fertilization,

diagram, 154 ; mitosis,

63, ff.; bivalent, 67, 135; changes
in volume (growth), 69; con-
stancy of form and size, 67, 68;
during interkinesis, 64, 70, 72;
evolution, 296; genetic continuity,
70, ff.; heterotropic, 310; "hy-
pothesis," 293, ff.; individuality,
70; in heredity, 322; in Protozoa,

59, 60, 64; numbers, 66;

in

INDEX

369

artificial parthenogenesis, 207; nu-
merical constancy, 66; varia-
tion, 66-67; pairing, 68-69, 68;
plurivalent, 67; relation to sex,
306, ff.; specific constancy, 65;
specificity, 70; structure, 291-
292, 291 ; univalent, 67

cicatrix, 120

Ciliates, mutual fertilization, 194,
196

Cirripedia, complemental males, 106

Cladonema, germ cells, 110

Clava, blastula, 220

cleavage, 18, 219, ff.; adequal, 227;
bilateral, 235-237; complete, 226,
232; determinate, 244; deviations
from "laws" of, 231, ff.; dexio-
tropic, 235; discoid, 227, 240;
disymmetrical, 238; equal, 226;
forms, 226, ff.; holoblastic, 227;
incomplete, 226, 239; indetermi-
nate, 244; irregular, 239; Iseo-
tropic, 235; "laws" of, 229, ff.;
meroblastic, 227 ; nomenclature of
blastomeres, 248, ff., 270; partial,
227; plane, first, location of. 246,

ff.; meridional, vertical, etc.,

228-229; relation to position

of centrosome, 230, 56 ;

to symmetry of ovum and adult,
246; processes of, 244, ff.; radial,
228, 232; rate, 230; rhythms, 232;
rotatorial, 232; spiral, 232, 234;
superficial, 227, 241-243; termina-
tion of period of, 221; under pres-
sure, 281-282, 284; unequal, 227

Clepsine, cleavage (radial), 235

Clytia, development of isolated
blastomeres, 280

coalescence, development after, 282-
283

Coccidium, reproduction, 200; schizo-
gony, 4

cod fish, number of ova, 107

coeloblastula, 220, 330, 331

ccelom, 336

ccenobium, 8

coenogenetic traits, 27

Collozoum, gametes, 198

complemental males, 106

concrescence, 347, 351, ff., 352

confluence, 347, 347, 351, 352, 353,
354

conjugation, epidemics of, 209; re-
lation to reproduction, 208, ff.;
see also fertilization.

Conklin, 206, 220, 258, 267, 269,
271, 275, 278, 298, 305

connecting fibers, 53

contraction phase, 135

Copepod, chromosome groups dur-
ing cleavage, 223, 226

Copromonas, conjugation, 192, 193

copulation, 106

cortical layer, 90

Crampton, 271

Crepidula, blastomeres, 266; differ-
entiation of cilia, 298; synthesis
of nuclein, 206

Cristatella, statoblasts, 8

Crustacea, spermatozoa, 99, 108

Ctenophore, cleavage (disymmetri-
cal), 238, 238

Cumingia, development after cen-
trifuging, 277, 277

cyclopia, in Fundulus (artificial),
318, 319

Cyclops, blastula showing primitive
germ cells, 112; segregation of
germ, 133

Cyclosporia, gametes, 198

Cynthia, development of single blas-
tomeres, etc., 269, 270, 272; non-
correspondence of cell boundaries
and formative stuffs, 278; or-
ganization of ovum, 267; structure
of ovum preceding and following
fertilization, 176, 177, 178; test-
cell nuclei in ovum, 120

Cypris, parthenogenesis, 203; sper-
matozoon, 100

cytoplasm, 37; localization in, 264,
ff.; of ovum, differentiation in,
90, ff.

cytotropy, 189

Dallingeria, anisogamy, 198

Daphnia, spermatozoon, 108

Dean, 109

Delage, 207

delamination, 340, 362

Delia Valle, 67

Dentalium, development of isolated

blastomeres, 271, 273; yolk lobe

(polar lobe), 271
determinate cleavage, relation to

indeterminate, 284, ff.
"determiners," in differentiation,

290, 297; possible nature of, 298
duteoplasm, 40
development and differentiation, as

interaction, 320-321; as reaction

(behavior), 24-25, 261-262; phases

of, 18-19

dicentric system, 79
Didelphys, spermatozoon, 99
didermic organism, 332-333
Diemyctylus, spermatophore, 106
differentiation, conditions of, 262,

ff.; cytoplasm and nucleus in, 305;

370

INDEX

determination of, 321; role of
external factors in, 317, ff.; with-
out cleavage, in Chcetopterus, 279,
279

digametic, females, 314; males, 314
Dileptus, nuclear division, 58
diploid number of chromosomes, 133
Diplozoon, 107
direct cell division, 43
discoblastula, 220, 331, 332
Discoglossus, spermatozoon, 100
dispermy, effect upon differentiation,
300; evidence from, upon chromo-
some hypothesis, 301
distributed nucleus, 58-59
Dixippus, X-chromosome, 312
Dobell, 58

Doliolum, budding, 7
dominance, in embryo hybrids, 305
Drago, 167
Drew, 106

Driesch, 263, 264, 281, 301, 305
Dromia, cleavage (superficial), 242
dyads, 138
Dzierzon, 315

Echinoderm, polarity of ovum, 92
Echinus, cleavage under pressure,

284; development in chemically

altered media, 320 ; of

isolated blastomeres, 281 ; number

of ova, 107
ectoblast, 334
ectoderm, 334
ectoplasm, 34, 267
ectosarc, 34
Edwards, 311, 313
eggs, care of, 104; see also ovum.
egg membranes, 95, ff.; as related to

conditions of development, 108
egg tubes, of Insects, 117, 118
Elasmobranch, blastoderm, 365;

blastula, 344; gastrulation, 343,

ff., 344

emboitement, 22
embryo, formation from germ ring,

347, 352
Embryology, defined, 2; phases in

history of, 20-21
endoderm, 334
endogamy, 189
endoplasm, 34, 90, 267
endosarc, 34

end piece, of spermatozoon, 100
Entamceba, autogamy, 191, 192;

maturation, 157
enteroccelic grooves, 336
enteroccels, in Elasmobranchs and

Reptiles, 347; in frog, 342
enteroderm, 348, 350

enteron, 336

entoblast, 334

entrance disc, of Nereis, 169

Ephelota, budding, 7

epiblast, 334

epiboly, 336

epigenesis, 22, 23

equational division, in maturation,

152

equatorial plate, 52
Equisetum, multipolar spindle, 57
Esox, micropyle, 96
Ethusa, spermatozoon, 99
Eudorina, reproduction, 199
Euglena, nuclear division, 61
Euglypha, nuclear division, 60
Eunice, spawning season, 103
Euplotes, fission, 6
Euschistus, dimorphic spermatozoa,

101

ev agination, 360
exogamy, 189, 200
exoplasm, 90; of ovum, during

fertilization, 174, 175
external conditions, as factors in

differentiation, 318, ff.

Farmer and Moore, 133

fertilization, 13, 164, ff.; among
Protozoa, 189, ff.; by "foreign"
spermatozoon, 171, 172; chemical
processes of, 205-206; defined,
165; following maturation, 187-
188; membrane, 175; arti-
ficial formation, 205; mutual, in
Ciliates, 194; of enucleate egg-
fragments, 301-303; preceding or
during maturation, 180, ff.; re-
lation to heredity, 214, ff.;

rejuvenation, 209, ff.; 212;

reproduction, 17, 202,

'., 208, ff.', variation,

fi

!13; selective, 171; significance of,
202, ff.; time relation to matura-
tion, 179, ff., 180

filar substance, 34

fission, in Metazoa, 4-5; multiple, 3;
simple or binary, 2, 5

flagellum, of spermatozoon, 98

Flemming, 43, 44, 141

Fol, 80

follicle, of ovum, 96; ovarian, 119-
120, 119

formative stuffs, 278

frog, blastula, 340; development of
single blastomeres, 284-285; en-
teroccels, 342

Frontonia, nucleo-cytoplasmic rela-
tion during interkinesis, 78

Fundulus, chromosomes, hybrids in

INDEX

371

76, 223; monsters (cyclopean) in
presence of lithium, 318, 319;
ovum, 94

Gametes, of Metazoa, 13; of Pro-
tozoa, 189, ff.

gametogenesis, 18

gametogonidia, 10

gametophyte, chromosomes of, 161-
162

Ganoids, blastula, 331

Garbowsky, 283

gastrula, 19, 333, ff.

gastrulation, 333, ff.; in meroblastic
ova, 343, ff.

Gegenbaur, 20

gemmules, 7

genetic continuity, of cell organs,
55; of chromosomes, 70, ff.

genital ridge, 109

germ, 14; organization of, 23, 25;
predetermination of, 264, ff.

germ cells, 85, ff.; derivation of, 112-
113; of Protozoa, 189, ff.; relation
to breeding habits, 102, ff.

germ disc, 177, 332

germinal continuity, theory of, 14,
15

germinal localization, hypothesis,
264, ff.

germinal vesicle, 89

germ layers, in budding, 358; in
regeneration, 358; morphogenetic
value of, 356, 358; primary, 334;
theory of, 355, ff.; - - ex-

ceptions to, 357, ff.

germ ring, 348; of Amphioxus, 334;
of frog, 340, 343

"giant" polar bodies, 148, 149

Godlewski, 171, 220, 221, 301, 304

Gonactinia, fission, 5

gonads, 109; epithelial structure,
112; of Metazoa, 13

growth period, of 06- and spermato-
cyte, 113

guinea pig, metamorphosis of sper-
matid, 126

Gulick, 311

Guyer, 311, 313

Hacker, 135

Haeckel, 280

Hagedoorn, 301

Haller, 22

haploid number of chromosomes,

130, 133

Hartmann and Nagler, 160
Harvey, 202

head, of spermatozoon, 98
Heidenhain, 41, 80
Helix, chromosomal variation, 67;

" trophospongien " in hepatic duct
cells, 40

Henking, 307, 310

Herbst, 75, 271, 298, 303, 318

heredity, defined, 261; mechanism
of, 321-322; relation to develop-
ment, 260, ff.;- - fertiliza-
tion, 214, ff.; r61e of external
factors in, 317, ff.

hermaphroditism, 13

Hertwig, O., 56, 114, 150, 213, 263,
265, 281

Hertwig, O. and R., 356

Hertwig, R., 77, 157, 209, 210, 222

Hesperotettix, X-chromosome, 312

heterochromosomes, 310

Heterodontus, ovum, 87

heterotype division, 140

His, 264, 352

histogenesis, 20, 363

Holophrya, multiple fission, 3

homolecithal ova, 93, 226

homotype division, 141

Hooke, 31

human ovum, 88; spermatozoon,
98, 100

Huxley, 223

hyaloplasm, 35

hybridization, evidence from, upon
chromosome hypothesis, 301, ff.

Hydatina, relation of sex to fertiliza-
tion, 315

Hydra, ovum, 88, 118

Hydrophilus, superficial cleavage,
243

hypoblast, 334

Ids, idants, 292
idiochromatin, 58, 158
idiochromidia, 41, 156, 190
idiochromosomes, 307; variations

in, 311, 312, 313
idiosome, 125
Inachus, spermatozoon, 99
indirect cell division, 44, ff.
Insects, maturation, 140; ovaries

(egg tubes), 117, 118
interfilar substance, 35
interkinesis, 45
intermediate layer, 333
internal buds, 7
interzonal fibers, 53
invagination, 336, 360
involution, 336
islands, protoplasmic, 242
isogamy, 192, ff.
isolated blastomeres, development

of, 269, ff., 280, ff.
isolecithal ova, 93, 226
isotropic blastomere group, 263

372

INDEX

Jenkinson, 318

Jennings, 209

Jordan, 106

Julus, attraction cone, 168

Karyogamy, 190, 192, ff.

karyokinesis, 44, ff.

karyolymph, 37

karyoplasm, 37

karyosomes, 38

kern-plasma relation, 77; during

cleavage, 222
kinoplasm, 39
Klossia, syngamy, 197
Korschelt and Heider, 152
Kupelwieser, 172, 303

Lacerta, genital ridge, 111

Lankester, 264

Lepas, movement of spindle, 230

Lepidoptera, digametic females, 314

Lepidosiren, chromosomes, 68; mat-
uration, 134, 136

leptonema, 133

Leptoplana, blastomeres, 256

Leptynia, X-chromosome, 312

Leudscus, spermatozoon, 99

Leuckart, 107

Ley dig, 31

Lilium, spireme in spore-mother-
cell, 63

Lillie, F. R., 168, 183, 245, 275, 278,
279

Lillie, R. S., 80

Limax, polar bodies, 148

Limnadia, parthenogenesis, 203

linin, 37

L'thium, effects upon development of
Fundulus, 318, 319

Litomastix, multiple embryo for-
mation, 315

localization, development of, 286;
germinal, 264, 288; regulation of,
284, 286

LociLsta, sperm atophore, 105

Loeb, 171, 172, 205, 206, 210, 301

Loeb and Moore (read, Loeb, King
and Moore), 305

Loligo, cleavage (bilateral), 237, 241;
spermatophores, 106

Loricera, spermatophore, 105

Lott, 166

Lyon, 269

McClendon, 269
McClung, 135, 310, 313
Maas, 280
macrogamete, 10
macronucleus, 58

magnesium, effects of upon differ-
entiation, 318, 319

malic acid, effects of upon motion of

spermatozooids, 167
Malpighi, 22
Mammalia, ova, 87
mantle fibers, 52, 79
Mark, 148, 266
Marshall, 103

Mastigella, chromosome number, 66
maternal characters, in hybrids,
301, ff.; affected by external con-
ditions, 304, ff.

maturation, 131, ff.; induced ("arti-
ficial"), 205; in oogenesis, 142,
ff.; in parthenogenesis, 158, ff.;
in plants, place in life history, 161 ;
in Protozoa, 156, ff.; in spermato-
genesis, 133, ff.; period, of 06- and
spermatocyte, 113; place in life
history, 160, ff.; reducing divisions
in, 152; relation to heredity, 152;
results of, 151, ff.; stimuli leading
to, 150; time relation to fertiliza-
tion, 179, ff., 180

Maupas, 209, 210

megagamete, 197

meiotic (maiotic) division, 133

membrane, chorionic, 96; formation
by fertilized ovum, 175; nutritive,
protective, etc., 97; of ova, 95, ff.;
tertiary, 97; vitelline, 95

Mendelian heredity, relation of
chromosome structure, 294, 295

Menidia, chromosomes in hybrids,
75, 223

Mermiria, X-chromosome, 312

merocytes, 171

merogony, 204

mesoblast, 336

mesoderm, 336; axial and gastral,
338, 339

mesophase, 52

mesoplasm, 267

Mesostomum, cleavage (irregular),
239

metamorphosis, of spermatid, 114,
125, jf., 126

metaphase, 52

metaplasm, 40

Metapodius, idiochromosome, 311

Metcalf, 62

microgamete, 11, 198

Micrometrus, primitive germ cells in
embryo, 112

micronucleus, 58

micropylar cell, 96, 120

micropyle, 96

microsomes, 35

Microstomum, fission, 5

middle piece, of spermatozoon, 98

Minot, 210

INDEX

373

mitochondria, 41, 125, 127

mitome, 34

mitosis, 44, ff.; causes of, 76, ff.;
diagram, 48; duration, 54; in
Protozoa, 57, ff.; in Salamandra,
46-47; in Unio, 60; mechanism
of, 79, ff.; modifications, 57, ff.;
of cleavage, 219

Moenkhaus, 75, 223

monads, 138

Monas, anisogamy, 198

monocentric system, 79

monodermic organism, 330

moncestrous, 103

monogony, 12

monosome, 310

monospermy, 169

Montgomery, 68, 101, 135, 311

Moore, 135

.Morgan, 269, 275, 276, 277, 301,
311, 313

Morgan and Spooner, 275

morphogenetic processes, 359, ff.

Morse, 205

mouse, amitosis in tendon cells, 43

multiple embryo formation, 315

multiplication, period of, during 06-
and spermatogenesis, 113, 114

Mus; see mouse, rat.

Musca, ovum, 91

Myzostoma, spermatozoon, 99

Nageli, 265

"Nebenkern,"41

neck, of spermatozoon, 98

Nereis, development after subjec-
tion to pressure. 281-282; en-
trance of spermatozoon, 168, 169;
fertilization, 174; oocyte, 89;
ovum, 95

nests, formation of, 104

neuroplasm, 267

Newman and Patterson, 315

Nezara, idiochromosomes, 313

Noctiluca, conjugation, 192; nuclear
division, 61

notochord, 336

notogenesis, 333

Noturus, blastula, 220, 331

nuclear analysis, hypothesis of, 265,
288, ff.

nuclear determination, 288

nuclear sap, 37; as a factor in deter-
mination, 298

nuclear substance, synthesis in cleav-
age, 220

nuclein, synthesis in fertilization,
205-206

nucleo-cytoplasmic relation, 77, 78 ;
in senescence and rejuvenation, 210

nucleolus, 38
nucleus, structure, 37-38
nuptial season, 103
nurse cells, 118, 119
nutritive membranes, 97
nutritive relations of growing ova,
118, ff.

Octets, of blastomeres, 229

oestrus, 103

oocytes, primary and secondary,

114, 115, 144

oogenesis, 18, 113, ff.', diagram, 114
oogonia, 114
oogonidia, 10
Opalina, chromosomes (amoeboid),

54; conjugation, 193; nuclear

division, 62

Ophryotrocha, nurse cells, 119
Orcheobius, gametes, 198
organ-forming substances, 92, 267,

organization, of germ, 23, 25; of
nucleus, 288, ff. ; of ovum, 91;

during fertilization,

179; relation to cleav-
age, 246

Orthoptera, idiochromosomee, 312

ovary, 109

Overton, 136

oviparous, 104

ovum (or ova), amoeboid, 87, 88;
animal pole, 92; comparison with
spermatozoon, 101,. 150, 151;cyto-
plasmic differentiation, 90, ff.;
demersal, 104; deutoplasm in, 93;
follicle, 96, 119; human, 88; mem-
branes, 95, ff.; nucleus, 89;
numbers, 107, ff.; nutritive rela-
tions during growth, 118, ff.;
organization, 91, 266, ff.; pelagic,
104; polarity, 91, ff.; position with
reference to gravity, 95; promor-
phology, 91; reorganization fol-
lowing fertilization, 177; sizes, 87;
types, 87, ff.; vegetal pole, 92

Pachynema, 135

palingenetic traits, 27

Palolo, spawning period, 103

Paludina, spermatozoon, 99

Pander, 355

Pandorina, conjugation, 196; repro-
duction, 8-9, 9, 199, 200

paralinin, 37

Paramcecium, chromosomes, 59, 64,
296; - - number, 66; fertiliza-
tion (conjugation), 194, 195;
life cycle, 210, 211; maturation,
167 '

374

INDEX

paraplasm, 35, 40

parasynapsis, 136

parthenogenesis, 17, 203; "artificial
204, ff.

parthenogonidia, 10

Pasteur, 1

Patella, development of isolated
blastomeres, 274

paternal resemblances in hybrids,
affected by external conditions,
304, ff.

paths, of pronuclei, in fertilization,
185, 186

Patterson, 348

Paulmier, 307, 310

Payne, 311, 313

Pelagia, chromatin extrusion in
oocyte, 299

perforatorium, of spermatozoon, 98

periblast, 346

Peripatus, spermatophore, 105

perivitelline space, 176

Petromyzon, blastula, 220, 231;
ovum and spermatozoon, 87

Pfeffer, 167

Pfliiger, 263

Phyllopneuste, spermatozoon, 99

pigeon, gastrulation, 349

plane, of cell division, 55, 56

Planocera, cell lineage, 248, ff., 252-
253; cleavage, 248, ff., 250, 251

plasmosomes, 38

plastids, 39

plastogamy, 189, 190

Plateau's law, 231

Platner, 114, 150

Pleodorina, reproduction, 9, 10

polar bodies, 115, 116, 144, 147;
comparison with spermatids, 116,
147, 148; " giant," 148, 149; loca-
tion, 149, 150; size, 148, 149

polarity, of cell, 41; of ovum, 91, ff..
266

polar lobe, of Dentalium, 271

polar nuclei, 149

Polygordius, cleavage (radial), 234

polyoestrous, 103

polyspermy, 170; physiological, 170

postgeneration, 284

postreduction, 152

potassium, effects upon differentia-
tion, 318

predelineation, 23

predetermination, 23 ; of germ, 264, ff.

preformation, 21

prereduction, 152

prespermatogonium, 121

pressure, effects upon cleavage and
differentiation, 281-282, 284

primitive groove, 348

primitive gut cavity, 334

primitive streak, 346, 348; of Ser-
ranus, 354

primordial germ cells, 111

Pristiurus, chromosomes, 69

promorphology, of germ, 264, ff.;
of ovum, 91

pronuclei, behavior during fertiliza-
tion, 180, ff., 186, 187-188

prophase, 52

prospective potency, 263

prospective significance, 263

protandry, 13

Proteus, chromatin extrusion in
oocyte, 299

prothallus, chromosome number,
161

protogony, 13

protoplasm, structure, 34-36, 36

protoplasmic bridges, 32, 42

Protozoa, gametes, 198; mitosis, 57,
ff.; relation between fertilization
(conjugation) and reproduction,
208, ff.; reproduction, 2-11 ; senes-
cence and rejuvenation, 209, ff.

Psammechinus, development of
fused embryos, 285

pseudochromosomes, 41

pseudocopulation, 105

pseudohybrids, 303

Pygosteus, micropyle, 96

Quartets, of blastomeres, 229

Rabl, 41, 70

Raja, chromosome groups (auton-
omy), 225

Rana, see frog.

rat, spermatogenesis, diagram, 122,
123 ; spermatozoon, 99

recapitulation, theory of, 25-27, 26

Redi, 1

reducing divisions, 133; in matura-
tion, 152, relation to chromioles,
153

reduction, with tetrad formation, in
oogenesis and spermatogenesis,
139, 146

regulation, of localization, 284, 286

rejuvenation, relation to fertiliza-
tion (conjugation), 209, ff., 212

reproduction, relation to fertiliza-
tion (conjugation), 202, ff.

Reptiles, gastrula, 345

resorption, in morphogenesis, 362

resting period, 45

reticulum, protoplasmic, 34, 35

Rhodeus, ova, 104

Rhoditis, parthenogenesis, 203

Rhumbler, 81, 189

INDEX

375

Rhynchelmis, cleavage (radial), 235
Roux, 91, 265, 271, 284, 329
Riickert, 75, 144

Sachs-Hertwig, law of cleavage, 229

Sala, 282

Salamandra, mitosis, 46, 47; sper-
matogenesis, 124; spermatozoon,
100

salts, effects upon development,
318, f.

Sauropsids, gastrulation, 348, ff.

Schaudinn, 191

schizogony, 4

Schleicher, 44

Schleiden, 31

Schonfeld, 133

Schultze, 31

Schwann, 20, 31

Schweigger-Seidel, 20

Scott, 103

Scyllium, egg membranes, 96

sea-urchin, see Arbacia, Echinus,
Sphoerechinus, Strongylocentrotus,
Toxopneustes.

secondary sexual characters, in
Protozoa, 200

Seeliger, 301

segmentation (cleavage), 18

segmentation cavity, 221, 331

seminal fluid, 101

seminal vesicles, 129

senescence, 210

Serranus, cleavage (discoid), 240;
primitive streak, 354

Sertoli cells, 121

sex, determination of, 306, ff., 315;
heredity of, 315; limited heredity,
316

sexual and asexual reproduction, 12

shells, of ova, 96

Sida, spermatozoon, 99

Silvestri, 315

SinSty, 313

skein, 45

soma, 14

spawning habits, 102, ff.

spermagonidia, 10

spermatid, 115; comparison with
polar bodies, 116, 147, 148; com-
parison with spermatozoon, 128;
structure, 125

spermatocytes, 133, 135, ff.; pri-
mary and secondary, 114, 115

spermatogenesis, 18, 113, ff., 122, ff.;
diagram, 115

spermatogonia, 114, 133

spermatophores, 106, 106

spermatosphere, 125

spermatozoon (or spermatozoa), 97,

ff., 115; accessory, 171; agglutina-
tion of, 167; behavior upon enter-
ing ovum, 172, ff., 181; comparison
with ovum, 150, 151; direction of
swimming, 166; entrance into
ovum, 167, ff.; entrance path,
185, 186; exclusion of, by ova,
170; flagellate, structure, 97, 98;
forms, 97, ff. 99; "giant," 101;
head of, transformed into pronu-
cleus, 181; human, size of, 100; lon-
gevity, 166; number, 100, 107-108;
penetration into ovum, 167; rate
of swimming, 166; size, 100; spiral
path, in swimming, 166; volume,
101

Sphcer echinus, development in chem-
ically altered media, 320 ; of

fused blastulse, 286 ; iso-
lated blastomeres, 281 ;

parts of gastrula?, 283 ; larva from
dispermic ovum, 300; pseudo-
hybrids, history of chrpmatin, 302

spindle, 51; chromatic, in Opalina,
62 ; intranuclear, 67 ; multipolar,
67 ; origin of , in Protozoa, 62; posi-
tion of, in cell, 230

spinning activity, 149

spireme, 45

spongioplasm, 34; arrangement dur-
ing mitosis, 79

spontaneous generation, 1

spores, development from, 203; for-

• mation, 4

sporophyte, chromosomes, 161-162

squash-bug, see Anasa.

star-fish, see Asterias, Astropecten.

statoblasts, 7

Steinbruck, 305

Stephanosphcera, conjugation, 193;
reproduction, 199

Stevens, 311

Stockard, 318

Strasburger, 80, 136, 162

stroma, of ovary, 111

Strongylocentrotus, chromosomal va-
riation, 67; cleavage, 233; de-
velopment of isolated blastomeres,
282; history of chromatin in
pseudohybrids, 304 ; modified
chromosomes in ova, 314

structure of protoplasm, 34-36

Styela, see Cynthia.

sub-germinal cavity, 332

Sumner, 354

Surface, 248, 249, 252

symmetry, of ovum and adult,
relation to first cleavage, 246; of
ovum, changes during fertiliza-
tion, 179

376

INDEX

synapsis, 135, 294

Synapta, cleavage (regular), 227,

228,229
syndesis, 135
Syngamus, 107
syngamy, 12, 13, 165
synizesis, 135

Tadorna, spermatozoon, 99

tail, of spermatozoon, 98

Teleost, gastrulation, 343, 346, ff.,
347

telolecithal ova, 93, 226

telophase, 54

telosynapsis, 136

Tennent, 304

Tennent and Hogue, 207

terminal filament, of spermatozoon,
100

test cells, of Tunicates, 120, 120

testis, 109; structure, 121

tetrads, 136, ff., 139, 146

Tetramitus, conjugation, 193; divi-
sion, 69

Torpedo, blastoderm, 355; poly-
spermy, 170

Toxopneustes, entrance of sperma-
tozoon into ovum, 168; fertiliza-
tion, 173 ; paths of pronuclei, 186

Trachelomonas, cell division, 59, 64

Trichomastix, autogamy, 191

Trichosomum, 107

tridermic embryo, 333

Triton, blastula, 331 ; gastrulation,

trophochromatin, 58, 158
" trophospongien," 40, 41
Tyndall, 1

Unequal division, in cleavage, 231
Unio, blastomeres, 256 ; cleavage

(radial), 235; mitosis, 50
Urodeles, spermatogenesis, 124
Urospora, gametes, 198

Vacuoles, 39

variation, relation to fertilization,
213; relative amount in gameti-

cally and agametically produced
individuals, 214

vasa deferentia, 129

vasa efferentia, 129

vegetal pole, 92

Vernon, 304

Vesperugo, spermatozoon, 99

Virchow, 20, 42

vitelline membrane, 95

viviparous, 104

Volvox, gametes, 198 ; reproduction,
9-11, 11, 199, 200

Vorticella, fertilization (conjuga-
tion), 196

DeVries, 265

Wallace, 313

Weismann, 14, 203, 213, 265, 292

Whitman, 162, 264, 352

Whitney, 269

Wilson, 76, 101, 136, 168, 185, 231,

247, 257, 263, 269, 271, 280, 281,

287, 298, 311, 313, 314
Winiwarter, 133, 135
Wolff, 22, 355
Woodruff, 211

X-chromosomes, 307; in maturation,
diagram, 309; in relation to sex,
diagram, 310; variations in, 311,
312

Yatsu, 183, 207, 287

Y-chromosomes, 311, 312

yolk, arrangement in ova, 93; lobe,
of Dentalium, 271; nucleus, 89;
pyramids, 242 ; relation to develop-
mental period, 108; sac, 350;
stalk, 351

Zeleny, 287

Ziegler, 81

Zoja, 280

zona radiata, 95

zoospores, 4

Zur Strassen, 253, 255, 282

zygonema, 135

zygote, 11

S •;•-,$•
•f: "-•>„

—

University of Toronto

Library

£

DO NOT /f

§

REMOVE /

W fe

THE If

§1

CARD

!?*

'FROM ^

O fl)

o fl

THIS V

M
•H

POCKET X"

Acme Library Card Pocket
LOWE-MARTIN CO. LIMITED