Book - Vertebrate Embryology (1913) 3

From Embryology
Embryology - 21 Jan 2019    Facebook link Pinterest link Twitter link  Expand to Translate  
Google Translate - select your language from the list shown below (this will open a new external page)

العربية | català | 中文 | 中國傳統的 | français | Deutsche | עִברִית | हिंदी | bahasa Indonesia | italiano | 日本語 | 한국어 | မြန်မာ | Pilipino | Polskie | português | ਪੰਜਾਬੀ ਦੇ | Română | русский | Español | Swahili | Svensk | ไทย | Türkçe | اردو | ייִדיש | Tiếng Việt    These external translations are automated and may not be accurate. (More? About Translations)

Jenkinson JW. Vertebrate Embryology. (1913) Oxford University Press, London.

Vertebrate Embryology 1913: 1 Introduction | 2 Growth | 3 The Germ-Cells, their Origin and Structure | 4 The Germ- Cells, their Maturation and Fertilization | 5 Segmentation | 6 The Germinal Layers | 7 The Early Stages in the Development of the Embryo | 8 The Foetal Membranes of the Mammalia | 9 The Placenta | Figures
Historic Disclaimer - information about historic embryology pages 
Mark Hill.jpg
Pages where the terms "Historic Textbook" and "Historic Embryology" appear on this site, and sections within pages where this disclaimer appears, indicate that the content and scientific understanding are specific to the time of publication. This means that while some scientific descriptions are still accurate, the terminology and interpretation of the developmental mechanisms reflect the understanding at the time of original publication and those of the preceding periods, these terms and interpretations may not reflect our current scientific understanding.     (More? Embryology History | Historic Embryology Papers)

Chapter III The Germ-Cells

The male and female germ-cells - the spermatozoon and ovum - are highly specialized structures and as different from one another - except in their nuclei - as any two cells could well be, the former being a small, active body, the latter large and inert. But though so unlike in their completed form, they are derived from cells which are apparently identical in the two sexes, the primordial germ-cells.

During the development and growth of the body of the parent which encloses them, these primordial germ-cells pass through a series of changes, the final result of which is the formation of the ova and spermatozoa. The history of these changes is very similar in the male and in the female. In each case it may be divided into three periods, a period of multiplication, a period of rest and growth, and a period of maturation (Fig. 6).

In the male sex the primordial germ-cells divide to form small cells, the spermogonia, which in their turn divide a large number of times. In all these divisions the nucleus divides by karyokinesis, and the number of chromosomes formed at each division is the same as that observed in all the tissue-cells of the body. This somatic number is constant for any given species of animal (or plant), and is (except in certain insects and some other forms) an even number. We shall speak of it as 2 w.

After a time, however, these spermogonia cease dividing and enter upon the second period of rest, during which they grow. The growth is not very great, but quite well marked. During this time the nucleus undergoes intricate changes which are the prophases of the first maturation division. The male germ-cells are now known as primary spermocytes.

At the end of the resting period the primary spermocytes prepare once more to divide. Each is halved to form two secondary spermocytes, the nuclear division being of an altogether peculiar character, and the number of chromosomes reduced to one-half (n) of the normal number. Each secondary spermocyte then divides again to give rise to two equal cells, the spermatids, the number of chromosomes being again one-half that observed in the spermogonia. Each spermatid becomes directly metamorphosed into a spermatozoon, there being no further division.

Fig. 6. - Diagram to illustrate the history of the germ-cells in the male (on the left) and the female (on the right).

I. Period of multiplication (many more divisions occur than are here represented).

II. Period of rest and growth,

III. Period of maturation. Sp.g., spermogonia ; sp.c.l, primary, sp.c.2, secondary spermocytes ; sp., spermatids ; sp.', spermatozoa ; o.g., oogonia ; o.c.l, primary, o.c.2, secondary oocyte; o., ovum ; ]).b.l, first polar body; p.b.2, second polar body; p.b.1.2, halves of first polar body. (After Wilson, after Boveri.)

In the female the primordial germ-cells divide to produce oogonia, and these in their turn divide, the nucleus breaking up into the full number of chromosomes (2»). When the period of multiplication has come to an entl each oogonium rests while the nucleus passes through the prophases of the first maturation division. The whole cell then grows into a primary oocyte. This growth is much greater than in the male sex, since it is during this time that the yolk is deposited in the cytoplasm to the accompaniment of other and very complex nuclear changes. In the third or maturation period the ovum, like the spermatozoon, undergoes two divisions, and two only ; but whereas in the male these two divisions are equal, giving four spermatids, eventually four spermatozoa of the same size, in the female they are markedly unequal. The primary oocyte divides unequally into a large cell - the secondary oocyte - and a small cell, the first polar body, the number of chromosomes being reduced to n. A second unequal division results in the production of one large cell - the matme ovum - and a small cell, the second polar body ; meanwhile the first polar body has divided (usually) into two small cells of the same size. The number of chromosomes is again one-half the normal number.

The parallel between these processes in the two sexes is evident, since each primary spermocyte or oocj^e by two divisions produces four cells, each one of which possesses only one-half the number of chromosomes seen in the spermogonial and oogonial mitoses. While, however, in the male the four cells are all of the same size, in the female one, the ovum, is large, while the remainder are small.

The germ-cells are supported, invested, and nourished in the testis and ovary by certam elements known as follicle-cells. These, like the primordial germ-cells, appear at an early stage in the development of the parental body, and our first duty will be to inquire into the origin of both. We shall then be at liberty to discuss the structure and chemical composition of the mature sexual elements, the disposition of the protecting follicles, the nature of the membranes by which the ovum is enclosed, the intracellular and nuclear processes accompanying the deposition of the yolk, the nuclear phenomena involved in the reduction of the number of the chromosomes during the two maturation divisions, the metamorphosis of the spermatid into the spermatozoon, and finally the union of the two germ-cells in the act of fertilization.

I. The Origin of thk Germ-Cells and of the Follicle-Cells

The gonads - testes and ovaries - first appear, at an early date in the development of the embryo, in the form of what are known as the genital ridges. The genital ridges are a pair of longitudinal bands of tissue in the abdominal region, each placed between the root of the mesentery of the gut on the inside and the Wolffian body or mesonephros on the outside, and each projecting downwards into the peritoneal cavity.

Each genital ridge is covered by the peritoneum or coelomic epithelium, which is here usually columnar. Under this epithelium are a number of small cells which are probably derived by proliferation from the epitheUum itseK, together with others which come from the retro -peritoneal tissue behind. In addition there are conspicuous certain large cells - ^usually with large nuclei unlike those of the surrounding cells and in many cases with yolk-granules in the cytoplasm, derived from the yolk of the egg from which the embryo itself has arisen. These large cells are placed some in, some below the epithelium. They are the primordial germ-cells, and their position in the columnar epithelium covering the genital ridge has very nationally given rise to the belief that they are formed by modification in situ of the cells of that epithelium, which has hence been termed the germinal epithelium, The researches of recent years have, however, brought forward very strong evidence to show that the first germ-cells are not formed in or from the germinal epitheUum, but elsewhere in the body, and that they only reach what is to be their final, resting-place by migrating there, the source from which they spring being in general the endoderm or splanchnopleure (mesoderm) of the gut or yolk-sac. Later on, however, it is generally admitted that germ-cells may arise from the germinal epithelium. This is also probably the source of the future follicle-cells, since these are derived from the small cells which are proliferated from that epithelium. The other small cells, retro-peritoneal, give rise to the thecae of the follicles and to the vascular connective tissue (stroma) olf the ovary or testis. Let us consider a few cases.

Fig. 7. - Primordial germ-cells in the dogfish {Sci/llium).

A, Germ-cells in the mesoderm surrounding the gut.

B, Germ-cells (g.c.) creeping up the mesentery (m.).

c. Germ-cells in the germinal epithelium (g.ep.) of the genital ridge. The yolk-granules are beginning to disappear.

Fio. 8.- Primordial germ-cells (r/.c.) in the tadpole of the common frog {Rana kmpomria). A, In the mesentery (hi.) ; b, In the genital ridge. g.ep., germinal epithelium ; f.c, follicle-cell.

In the Elasmobraiich fishes the gerni-celLs are first found in the extra-embryonic blastoderm, either between the yolk and the mesoderm, or under the ectoderm. Thence they migrate into the body of the embryo by way of the yolk-stalk (Fig. 7) ; passing up by the splanchnopleure surrounding the gut, sometimes in the gut epitheKum itself, they reach the mesentery, and thence to right and left into the two genital ridges, where they make their way into the epithehum. The cells are large, the cjrtoplasm crowded with yolk -granules, which, however, are presently digested and disappear, the nucleus large, provided with a large nucleolus, its chromatin in the form of small granules. The accompanying table (from Woods) gives the number found in the unsegmented mesoderm or ventral to the mesentery, in the mesentery, and in the genital ridges, in successively older

Number in genital rid"cs.

34 69 193 710

the germinal

In the lamprey {Petromyzon) the primordial germ-cells, similar m character to those of the Elasmobranchs, first appear in the lateral plate mesoderm, whence they migrate to their definitive position.

So also in the trout and salmon : in these there is also a later formation of germ-cells from the epitheUum of the ridges.

In another bony fish {Cynmtogaster) the sex-cells can be distmguished even in segmentation stages; later they are found at the posterior end of the body (where all the germ-layers are fused together), whence they move forwards into the mesoderm of the genital ridges.

Amongst the Amphibia the germ-cells of the frog appear at

these cells.

Length of embryo

Number in

Total number of germ-cells.

unsegmented mesoderm or

Number in

in mm.

ventral to




mesentery. 98

















More germ-cells are later on formed from epithelium.

an early stage, in the ncwly-hatchcd tadjiolc ; they are derived from the large yolk-cells of the gut. Being separated off on the dorsal side they move up the mesentery (Fig. 8), and so, passing to the right and left, reach the genital ridges. The cells are at first crowded with yolk, but this soon disappears ; the nuclei are not peculiar.

Fig. 9. - Section of a 12-clay rabbit embryo, showing the migration of the primordial germ-cells (indicated by black dots). Most are in the yolkstalk mesoderm, some in the mesentery (m), some in the genital ridge {g.r.) internal to the mesonephros (M)., opening of the yolk-stalk into the yolk-sac.

It is remarkable that, in the female at any rate, many of these young germ-cells are expelled from their follicles and disintegrate in- the peritoneal cavity. There is an extensive formation of fresh ones by modification of cells of the germinal epithelium, which is also the source from which the folUcle-cells are derived. The theca comes from retro-peritoneal tissue. (An account of the follicle and the theca will be found in a subsequent section.)

When we pass to the Reptiles we find the same migration taking place. In the tortoises {Glmjsemys) the sex-cells arise in the endoderm of the yolk-sac posteriorly. Becoming amoeboid they migrate towards the middle line into the embryonic region. Passing out of the endoderm into the splanchnopleure (the mesoderm covering the gut), they travel up the mesentery into the ridges. Many, however, fail to reach their destination, and remain for some time in the epithelium of the gut. It is stated that in this animal no germ-cells are ever formed from any other source.

Fig. 10.- Primordial gorm-cells {g.c.) in the rabbit.

A. Early stage in the formation of the genital ndge, covered by the germinal epithelium {g.cp.). Below this are some germ-cells and connective tissue and blood-vessels {b.v.). Germ-cells arc also seen in the mesentery {m.).

B, A germ-cell in the epithelium of the gut.

c, One from the yolk-stalk.

D, An epithelial cell from the yolk-sac.

In the Birds, again, the germ-cells appear early, in a chick on the third day of incubation. They seem- to originate from the splanchnopleure of the yolk-sac, and pass, in the way already described, to the genital ridges, and there into the germinal epithelium. By the fifth day the migration is complete and the cells begin once more to multiply.

In all these cases the identification of the primordial germcells is considerably facilitated by their retention of yolk-granules at a time when these bodies have disappeared from the surrounding cells.

In the (placental) Mammals, however, where there is practically no yolk, the distinction of these cells from the surrounding elements is a matter of some difficulty, and it has been, and is still contended, that the germinal epithelium is their only place of origin. Nevertheless, there is good reason for believing that the Mammalia are no exception to the general rule.

In a rabbit embryo of eleven or twelve days (Figs. 9, 10) there are to be found in the splanchjiopleure of the yolk-stalk large numbers of rounded cells, distinguishable from the surrounding cells by their cytoplasm- which includes large oxyphile granules - and their nuclei, which are round and large, with a fine achromatic reticulum bearing small granules of chromatin. The nuclei consequently look pale. There is one, sometimes more, large nucleolus. In all these respects the cells bear a close resemblance to the large cells in the endodermal yolk-sac epithelium (Fig. 10 d). Precisely similar cells may be found in the body of the embryoround the sides of the gut, and sometimes in the gut epithelium in the mesentery, and finally below and in the columnar epithelmm (germinal epithelium) internal to the mesoncphros which IS the beginning of the genital ridge. In earlier stages the same ceils are found in increasingly smaller numbers in the genital ridges and mesentery, in increasingly larger numbers in the yolkstalk and endoderm. It can hardly be doubted, therefore, that the same migration of these cells from the yolk-sac to the genital ridges is occurring here as we have already observed in other forms. There is equally little doubt that these cells, arrived at the genital ridges, become germ-cells.

In the rabbit embryo of twelve days the genital ridge (Fig. 10 a) is very slight, consisting of a band of columnar epithelium, below which are a few cells derived probably by immigration from that epithelium. These cells differ from the germ-cells amongst which they lie in their nuclei - which are oval in shape, have more than one nucleolus, and a more open reticulum with coarser granules of chromatin, and are of smaller size - and in their cytoplasm - which includes no granules. Those which lie below the surface are destined to give rise to the follicle-cells, which will eventually be disposed in layers round the germ-cells. Deeper still are connective tissue-cells and blood-capillaries derived from the retro-peritoneal tissue ; from these will come the thecae and the vascular stroma. Let us follow the development of this genital ridge into the sexual organ, and first into the ovary.

Fig. 11. - Ovary of rabbit fiom embryos of a, 18 days, and B, 21 days, showing formation of cortex (G) and medulla, b.v., blood-vessels in stroma {str.) of medulla ; md.c, medullary cords.

The whole genital ridge is enlarged and made to project into the body-cavity by the increase of the connective tissue elements and blood-vessels, or stroma, which thus forms a central core or medulla to the whole organ. At the same time, by continued proliferation of the germinal epithelium at the surface, an external layer or cortex is formed. The germ-cells lie mainly in this cortex, but a few - those presumably which in migrating to the genital ridge have never reached the surface - lie in the stroma, where they are grouped in rows known as medullary cords (Figs. 11 A, 12 a). They seem to degenerate.

The cortex increases in thickness and becomes divided up into columns or blocks by the ingrowth of vascular connective tissue from the stroma. These columns - ^which were at one time believed to be produced by hollow invaginations of the germinal epithelium and known as the epithelial tubes of Pfliiger - are the sex-cords (Figs. 11 b, 12 b). They consist of folHcle-cells derived in all probability from the germinal epithelium, and of germ-cells which have migrated into then- present position from their source, the yolk-sac." The germ-cells have been increasing in numbers : in the resting condition their nuclei present the same characters as before. The cytoplasm, however, loses the oxyphile granules. At about the twenty-second day the germ-cells cease to divide and enter on the period of rest : they are in fact primary oocytes, and their nuclei begin to undergo the changes characteristic of the prophases of the first maturation division.

This mode of origin of the germ-cells does not of course preclude the formation of others from the cells of the sex-cords, that is, from the germinal epithelium, and it is Indeed quite possible that this occurs.

In the mouse and other Mammals (guinea-pig, mole, cat) the germ-cells appear to come from a similar source. In the mouse they are large cells with rounded bodies, dense, rather deeplystaining cytoplasm, and large nuclei, with a close reticulum, small, scattered chromatin granules and one or two large nucleoli (Fig. 13 A). They divide by mitosis for a time, but pass into the resting condition at a comparatively early date, about the fifteenth day (Fig. 1 3 B, c). They He intermingled with a number of cells, the future follicle-cells, which may be regarded as of epithehal origin. The medulla of the ovary is formed late (Fig. 13 e), when the nuclei of the germ-cells have already reached the pachytene ^ condition (see below), by ingrowth of connective tissue from the base ; prior to this the whole thickness of the organ is composed of follicle-cells and germ-cells, with but a few capillary blood-vessels. The sex-cords (' tubes ') of the cortex arise, therefore, not so much by downgrowths of epithehum as by rearrangement of the cortical, that is, foUicle and germinal, and medullary, that is, stroma cells (Fig. 13*).

â– p-rp 13* _A B, Formation of medulla in ovary of mouse. sir stroma of medulla; md.c, medullary eords; c, cortex of gerra-cells and

^Sestisof embryo mouse, ep., eoelomie epithelium ; L, seminiferous tubules.

A few germ-cells are found in the medulla, but as in the rabbit, these probably never become mature. When the gerincell nuclei have reached the diplotenic or dictyate stage (J^ig. 13 F) (in the new-born animal) the formation of folUcles begnis, by the grouping of the follicle-cells round the oocj^es, to form a single flat layer. At the same time the oocytes enlarge. These two processes always occur first at the deep end of the sex-cords. Later (Fig. 14) the cells of the follicle become cubical, and then increase in number till several layers are formed. Hence the cortex of the young ovary comes to comprise several layers of small oocjrtes, each surrounded by a single layer of flat cells under the surface, and larger oocytes, surrounded by cubical foUicle-cells, disposed in the more advanced deeper down in many layers. It is not, therefore, that the folHcles and oocytes enlarge as they pass in from their (supposed) origin at the surface but that those which are inmost are the first to enlarge The oocytes of the outermost layer often lie practically in the epithehum at this stage. Their nuclei, however, are not in the condition seen in newly-formed germ-cells (oogonia), but in the dictyate stage characteristic of oocytes

  • A full explanation of this tern, will be found in the section of the next chapter deahng with maturation.

Fig. 14.- Part of cortex of ovary of young mouse (8 davs) In the deeper parts the foUieles (/. 3) consist of two- or three-cell layeS" and Se oocytes are arge. In the middle layers the follicles (/. 2) areTntlayered but the cells are cubical, the oocytes smaUer. Under the s^SfaTe the oocytes are smaUer still and the follicle-cell flat (/. 1). ep surf^e enithe lium (germinal epithelium) ; h.v., blood-vessel ; strl\tfo^T,Z tW

In the male (Figs. 13 Bo^, 13* c) the sex-cords become early shut ofi from the suiiace epithehum (peritoneum) by the formation of a sheet of comxective tissue [iunim albuginm), and from one another in the same way. The sex-cords are the rudiments of the seminiferous tubules of the testis. Each consists of an outer layer of foUicle-cells, and an inner mass of germ-cells, presenting the characters already described. These are spermogonia and divide mitoticaUy many times. Intermmgled wi h Lm are a few of the folUcle-cells. In this sex there seems to be no doubt that many of the first-formed g-m-cells d^^^^^^^^ and that in the adult fresh spermogoma are difierentiated from

"Lfy"on of the germ from the body or somatic " ee^thet first appearance in a part of the body remote from, ^iCr gradualmigrationto their ^f^^f^^^ eesses which find a parallel in many, if not m aU, groups ot tne aSmal ldngdom. Thus in the Hydroid Coelenterates, the germ on xo become medusae or gono

tLrir^s Sgr^LX often pass foxward. and backphores. In tins migi<^t j round-worm

waxds from one 87;7„V ;t:eUs is distinguishable Ascaris the parent cell ol tne g ^ t^e

segmentmg oTum. Similaily ™ ^^Me Oephalopod Moltaca and in S<=°:^P ^/^^^ i^cts during the formation of the ^e™-' ^appearance they may be -P--"f ^j'^^Cre^Ue difierLiation of the blastoderm-or rather later, ^^e

the germ-layers or ater ^J^.^y ^o^Hged to migrate mesoderm, and m aU these oases

forwards into their ^f^Z::::Z ^ n Z from the germinal saos ; lastly, they may be different epitheUum of the ^j'-^^ ^ere we y,,,,^,,^, either

the double origm which we have .^^ subsequent

at an early date, independently °' ^ aate, from

migration into the f f no hard-and-fast the germinal epithehum itself. It is clear rule can be laid down. All the germ-cells may be precociously separated from the somatic cells and elsewhere than in the generative organ, or some may have such an origin, while others arise in the generative organ itself, or lastly, all may be developed by the second method, as appears to be the case in most Annelid worms and in Ascidians.

Nor need the conversion of what look like tissue-cells into cells endued with the capacity of reproducing all the characters of the species cause any particular astonishment when the widespread capacity for regenerating lost parts possessed by the adult tissues, and the remarkable facts of bud-reproduction, are borne in mind. In these cases germ-plasm or reproductive substance must be present in the regenerating or budding tissues, and yet there is no obvious continuity between this germ-plasm and that of the germ-cell from which the regenerating or budding individual sprang ; as little should we expect to find a demonstrable continuity in the case of sexual reproduction.

II. The Structure of the Mature Germ-cells A. The Ovum

The egg-cell is large and inert : it is quite incapable of locomotion ; only occasionally does it exhibit peristaltic contractile movements, as in the formation of the polar rings at the time of fertilization in Annelids, or slow changes of shape as in the protrusion of the animal ends of the blastomeres in Petromyzon, or the flattening at the animal pole of the frog's egg prior to segmentation.

In shape it is nearly always spherical : exceptionally, as in Myxinoids and Amia, ellipsoid or ovoid.

Size of the ovum. The ovum is always a large cell compared to other cells of the body, even where, as in Placental Mammals, it is actually very small, and it may be very large indeed, as in the large-yolked ova of most fishes, and of birds and reptiles. The size of the ovum is due to the contained reserve food material or yolk, the amount of which varies very greatly in the different groups.

A small-yolked (microlecithal) egg is found in the lamprey (Peiromyzon) and in the Anurous and Urodelous Amphibia. In the frog the diameter of the ovum is about 1-6 mm. In the Gymnophiona, and most 'Ganoid' fishes {Acipenser, Amia, Lepidosteus), there is more yolk in the egg, while in the Myxinoid Cyclostomes, Elasmobranch and Teleostean Fishes, Reptiles, Birds, and Monotrematous Mammals, the egg is large-yolked (megalecithal). Finally, in the Placental Mammals- which are descended fiom large-yolked forms- the yolk has been reduced to a very small amount.

The following table brings out the contrast between the size of the eggs in the large-yolked Monotremes and the smallyolked other forms. It will be seen that amongst the latter the Marsupials have the largest ovum. In this respect, as m others, they are intermediate between the Monotremata and the Placentalia.


Marsupialia :

Placentalia :

Echidna Ornithorhynchus

Dasyurus Didelphys









3-4 mm. 2-5 mm.

0-28 mm. 0-13 mm.

0-18 mm. 017 mm. 015 mm. 0-15 mm. 0 09 mm. 0-08 mm. 0 06 mm. 0 06 mm.

The yolk. The yolk is frequently termed deutoplasm m distinction from the living substance or protoplasm m which it hes. ir s Losited in the cytopl-m of the ovum during the period o ^oXh t the form of smaU bodies spoken of a. granules, g ofls or platelets (Fig. 15). The size, shape, and structure of these vary. In the lamprey, frogs and toads, newts and salamanders, the granules are oval or ellipsoid bodies, sometimes vacuolated. In the Elasmobranch fishes they are oval plates, sometimes spherical and vacuolated. In the Teleostean fishes the separate yolk-globules run together at an early stage to form one continuous yolk-mass. In the Birds there is white yolk and yellow yolk, the former consisting of small globules enclosing still smaller ones of varying size, while the latter is made up of

Fig. 15.- Yolk-gvanules. a, Dogfish, b, Axolotl, the smaller from the anmial, the larger from the vegetative hemisphere, c, White volk. D, yellow yolk, from the Hen's egg.

larger spheres, each including a multitude of minute droplets. In both kinds of yolk the smaller bodies are often set free by the rupture of the larger enclosing envelopes. In the Placental Mammals the yolk-granules are usually globular (Fig. 18, b). All the granules which have been mentioned are protein in nature, but in addition to these fat globules are not uncommon. Fat is present in the hen's egg, in some Mammalia (guinea-pig) (Fig. 18, c), while in the Teleostei a single large oil-drop is characteristically present (Fig. 72).

The chemical composition of the yolk of a hen's egg is as follows. The yolk- that is, the ovum- weighs from 12 to 18


grammes include

47-2 % of this is water ; the remaining solids

Protein . Salts . Fats Lecithin Cholesterin

15-63 %

0- 964 % 22-84 % 10-7 %

1- 75 %

51-884 %

The proteins include ovo-vitellin (for the greater part) and some albumin.

The former is not a globulin but a nucleo-proteid ; on digestion with pepsin it yields an iron-containing body, a pseudo-nuclein known as haematogen, since it is supposed that it is the source of the haemoglobin of the embryonic blood corpuscles. With the ovo-viteUin the lecithin of the egg is closely associated. The fats are oleates, palmitates. and stearates. With them must be included certain phosphatides.

The salts are chlorides of sodium, potassium, magnesium, and calcium. ,

The reaction of the yolk is alkaline. The colouring is due to lutein, a lipochrome. ,

Other ova have not been so fully investigated, but it is known that the ichthulin of certain fish eggs (carp, cod) is a nucleoproteid, and lecithin (6 %) and nucleo-proteid (94 %) can be demonstrated in the yolk of the frog's egg. The significance of the presence of nucleo-proteids will be more evident when we consider later on the part played by the nucleus during the deposition of the yolk.

The yolk of the Monotreme egg is of a yellow colour, in the lamprey it is a faint yellow, in the dogfish greenish, in the Ganoid fish Amia brown. In Placental Mammals the yolk is colourless. ,

The yolk is not scattered Irregularly through the cytoplasm, but arranged in a very definite fashion, known as the telolecithal ; that is to say, while the cytoplasm (or protoplasm) is concentrated on one side of the egg, the yolk (or deutoplasm) is conceiitrated on the opposite side. This does not imply, of course, that all the yolk is on one side, all the protoplasm on the other side, but that most of the cytoplasm is on the one, with fewer and smaller yolk-granules, while on the other the yolk-granules are more abimdant and larger, with less cytoplasm in between them. The transition from one extreme to the other in a small-yolked egg such as that of an Amphibian is quite gradual (Fig. 16) : there is a graded diminution in the concentration of cytoplasm, an increase in the concentration of the yolk in passing from one side to the other.

Fig. 16. - Diagram of a meridional section through a full-grown oocyte of the frog. The yolk-granules are represented by stippUng, the pigment by the thin black line. The arrow marks the egg-axis, its head the animal pole.

As the yolk increases the distinction between protoplasmic and deutoplasmic portions becomes more and more marked, until the limit is reached in the megalecithal type. Here the amount of yolk is so enormous that the cytoplasm is reduced to a small cap or disc- the blastodisc- at one side, the bulk of the ovum being occupied by the yolk (Fig. 17). Yet eVen here small yolkgranules are found in the blastodisc, and the transition from blastodisc to yolk is not absolutely abrupt.

In the Placental Mammals the telolecithal arrangement of the yolk can still be seen, in spite of the small amount, at least when the nucleus, with some cytoplasm, goes to the surface just before maturation (Fig. 18).

In the ovum of the Marsupial Dasyurus (Fig. 18, a) the yolkglobules run together at this time to form a single rounded mass - the yolk-body - placed on the opposite side to the nucleus.

In the ova of Birds (Fig. 17) the white yolk is disposed in the form of a central plug - the latebra - under the blastodisc. This is surrounded by successive layers of yellow and white yolk, alternately. The same feature is observable in the ova of Reptiles, Gymnophiona, Amphibia, and Elasmobranch fishes, where sheets of coarse and fine granules alternate.

The telolecithal disposition of the yolk confers upon the Vertebrate ovum a very definite structure and symmetry. In most cases the ovum is a sphere, and it is evident that a line may be drawn passing through the centre of the protoplasmic portion, at the surface, the centre of the egg, and the centre of the deutoplasmic portion at the opposite surface. This line is the egg-axis, and it is clear that its two ends, or poles, are unlike. The former, the protoplasmic, is known as the animalpole, the latter as the vegetative pole. These terms took their origin in the observation that in such an egg as that of the hen the chick or animal is developed from the blastodisc, at the side opposite to the inert or vegetative yolk.

From what has akeady been said it further foHows that the yolk and protoplasm are distributed about this axis in such a way that the egg would be divided into precisely similar halves by any section which included the axis, but by none other. Hence the egg is said to possess a polarity and a radial symmetry about the axis. In any one plane at right angles to the egg-axis all radii are alike. The plane at right angles to the axis and including the centre of the egg is equatorial.

In cases where the egg is ovoid or elUpsoid (Myxinoids, Aima) the egg-axis is the major axis.

Yolk is heavier than protoplasm. Hence the Amphibian egg which, after fertilization, is free to rotate inside its jeUy membranes, always turns over till its axis is vertical with the white, vegetative pole below. The fuU-gromi ovarian egg-whether alive or dead-behaves in the same way when floated m a fluid of the same specific gravity as itself. Similarly the ovum (yolk) of the hen's egg always turns over inside the shell till the blastodise is uppermost. So in Elasmobranchs.

I^ct 17 - Hen's longitudinally biRected. (After Balfour, modified.) The section includes" the axis of the ovum, the animal pole bemg to the upper side of the figure, sli., shell, underneath it the external shellmembrane ; i.m., internal shell-membrane ; a.di., au' chamber ; c7i., chaiaza ; hi blastodisc ; I., latebra of white yolk ; v.m., vitelline membrane.

Fig. 18. - Mammalian ova.

A, Dasijurus (a Marsupial). 1, The ovarian egg (oocyte) ; the nucleus is near the surface at the animal pole ; the cytoplasm contains spherules of jo\k. 2, The second maturation division. The first polar body has been extruded, and the second polar spindle is seen. The yolk-spherules have run together to form the yolk- body [y.h.) placed at the vegetative pole. (After Hill.)

B, A ha,t [Vesferiilio). Both polar bodies have been extruded and fertilization is taking place. The two pronuclei are seen. In the cytoplasm are numerous globules of yolk (protein). (After Van Beneden.)

c, The guinea-pig (Cavia). Full-grown oocyte. In the cytoplasm are mitochondria (cln'oraatic bodies) and fat globules (the former are black, the latter clear in the figure). (After Lams and Doorne. ) z. , zona pellucida.

Pigment. The polarity and radial symmetry thus conferred upon the egg by distribution of the yolk may be further emphasized by the disposition of the pigment where that is present apart from the colouring matter of the yolk itself. In many Amphibia (Anura and Urodela), in Ceratodus and Acipenser, pigment is present in the egg. The dark brown, almost black pigment of the frog's egg will be familiar. Chemically it is a melanin. In other cases {Siredon, for example, and the edible frog) it is of a much lighter colour.

The pigment lies (Fig. 16), in the form of minute droplets, in a dense superficial layer in the animal hemisphere of the egg, extending a greater or less distance into the vegetative hemisphere. There is left round the vegetative pole as a centre a circular unpigmented area. The symmetry of the egg, as determined by the position of the yolk, coincides with that due to the distribution of the pigment. There is also a less dense mass internally in the animal hemisphere.

The Nucleus. The nucleus - germinal vesicle - of the fullgrown oocyte is characterized by the presence of one (Placental Mammals) or more nucleoli, usually chromatic. The history of these nucleoli and of other parts of the nucleus wiU be dealt with later. What interests us at the moment is the position of the nucleus. This is always in the axis of the egg, but excentric (Fig. 16), and always nearer the animal than the vegetative pole. In a microlecithal egg the nucleus lies in the protoplasmic portion, in. a megalecithal egg in the blastodise (Fig. 17). It is placed, therefore, in what is termed, in Oskar Hertwig's first rule, the centre of its field of activity. The importance of this will be appreciated when we come to the study of the phenomena of segmentation.

Structure and symmetry of the ovum. It will be obvious from the foregoing that the egg is no homogeneous body, but heterogeneous with a definite polar structure - radially symmetrical about an axis determined conjointly by the disposition of the yolk, the distribution of the pigment, when that is present, and the position of the nucleus. The first two characters are purely cytoplasmic. The significance of this initial structure of the

Fig. 19.-0vary of the tadpole, showing development of the ovarian cavity (o.c.) and numerous germ-cells in diSerent stag^ 1 ^Wntene with the nuclei in different conditions. 1, Earliest stage ; 2, bynaptene Tpachvtene 4 Diplotene ; 5, Formation of nucleoli. The largest oocytes L sSnded by f ofiicles. ep.', coelomic epithelium (germmal epithehum).

egg cytoplasm in development cannot be over-estimated, for it is related in a perfectly definite way to the structure of the embryo which will come from it. Thus, to take one example.

Fig. 20. - Small ovarian egg of the frog surrounded by Its follicle (/.) and theca {ill.), which is continued into the pedicle (^.). h.v., a blood-vessel between follicle and theca ; v.m., vitelline membrane ; dr., chromatin filaments, now aclu-omatic ; n., chromatic nucleoli, ejected from the nucleus in' .\ and becoming achromatic (54".).

Fig. 21. - a, young, and b. older oocytes from the pigeon's ovary. r'^T cytoplasm of the oocyte; v.m., vitelline membrane;

/., follicle ; Ih., theca.

the anterior end of the embryo is developed always near the animal pole of the frog's egg, the egg-axis making a certain constant angle with the longitudinal axis of the embryo ; or, in other words, the anterior and posterior regions of the embryo are predetermined in the structure of the egg.

That the relation is a necessary and causal one is shown by those experiments- performed on the eggs of various animals - in which, some one part of the cyioplasm being removed, some definite organ of the embryo or larva is lacking. The different portions of the egg cytoplasm are therefore so many organforming substances, and since the organs are part of the sum total of the inheritable characters of the species, the cytoplasmic substances, on which their development depends, are factors determinant of inheritance.

The egg-follicle. In the ovary the egg-cell is invested by one or more layers of follicle-cells, the function of which is not only to protect, but also to nourish, the growing oocyte. These are derived, as we have seen, from the germinal epithehum of the genital ridge. The follicle in its turn is surrounded by a theca of flattened connective tissue-cells.

In the Amphibia there is but one layer of cells in the follicle ; they are flat. The ovary is hollow (Pig. 19), and the theca cells are continued into the stalk by which each ovum is suspended to the wall of the ovarian cavity (Fig. 20). Between theca and foUicle there are blood-vessels.

In the Elasmobranchs and Birds (Fig. 21) there is but one cell-layer in the young follicle, but the number is subsequently increased to two or more. The cells are cubical or polyhedral in shape.

Li the Monotreraata the number of cell-layers is only one or two, but in all other Mammals (Fig. 22) it is greatly increased, and a cavity filled with an albuminous fluid - the liquor foUicuh - is developed in between the cells, thus leading to the development of the characteristic hollow Graafian follicle. The cavity appears first on one side of the ovum as a narrow crescentic sUt ; soon this enlarges and extends round the ovum, which is then attached to the wall of the cavity only by a short stalk- the so-called discus prohgerus. On its free side a few layers of follicle-cells remain adherent to it, the cumulus proligorus. Finally, by the further extension of the cavity, the stalk is ruptured and the ovum, with its corona of cells, floats freely in the folUcular cavity. The ripe foUicle, which has now returned from the deep parts to the surface of the ovary, bursts, and the ovum, with its corona, is expelled and passes into the mouth of the oviduct (Fig. 43) to be fertilized.

The expulsion of the ovum is known as ovulation. In multiparous Mammals several are, of course, expelled at the same time and from both ovaries.

Fto 22 - Part of the cortex of an adult mouse ovary. </./., cavity of JSaSan follicle ;/.,fomck^

proligerus ; ej)., surface (coelomic) epithelium ; th., theca , b.v., Dlooa vessel.

After ovulation the foUicle collapses, but it does not immediately degenerate. It becomes altered mto a corpus luteum. The foUicle-cells divide for some little time, and then, ceasmg to do so, hypertrophy (Fig. 23). They secrete fat and lutem (to which the corpus luteum owes its yeUow colour). Amongst these enlarged folUcle-cells grow vascular strands from the mnermost layer of the theca. The theca cells, which mcrea.e m numbers by division, are fusiform, and. lymg obliquely, or tangentially, or radially, iii the follicle, divide up the luteal tissue into ii-regular blocks. The larger strands contain blood vessels. There is a central cavity filled with stellate cells and extravasated blood corpuscles.

It has been shown that the corpus luteum secretes a substance which passes into the blood, and by that channel reaches the wall of the uterus, Avhere it appears to be necessary for the proper attachment of the embryo by means of the placenta.

Fig. 23.- Marginal portion of a section through the corpus luteum of a mouse 14 days after parturition (i.e. after ovulation). /., hypertrophied follicle-cells ; s., septa of connective-tissue cells ; th., theca ; b.v., blood vessels.

The membranes of the ovum. These may be of three kinds, primary, secondary, and tertiary. A primary membrane is one secreted by the cytoplasm of the egg itself; a secondary, one secreted by the follicle-cells and often termed ' chorion ' ; while tertiary membranes (albumen, shell) are secreted by the epithehum of the oviduct as the egg passes to the exterior.

The ovum of Vertebrates is always immediately surrounded by a vitelline membrane, frequently termed a zona pellucida (Figs. 16, 17, 18, 20, 21). This membrane may be traversed by fine radial pores, by nieans of which nutrient material passes from the folhcle -cells to the ovum : it is then spoken of as a zona radiata.

It is a matter of great difficulty in most cases to determine whether the vitelline membrane is primary or secondary, but it is stated that there is a membrane secreted by the ovum itself, inside another secreted by the folhcle -cells, in most forms (Elasmobranch fishes. Amphibia, Reptiles, Birds). The radial striations of the uuier primary membrane disappear before the ovum is full grown. The viteUine membrane of the ripe egg is possibly the result of the fusion of both the primary and secondary membranes of an early stage.

The so-called ' chorion ' of Teleostei and the Ganoid Lepidosteus, a very thick membrane, is apparently primary. In it the radial striations are persistent. The Myxinoids, however, possess a true chorion which is provided, at the animal pole, with a number of hooks, by which the egg is attached. In Petromyzon, Teleostei, and Lepidosteus, the vitelhne membrane is perforated by a passage at the animal pole through which the spermatozoon enters. This is the micropyle.

In Mammals (Marsupiaha and PlacentaUa) there is much uncertainty as to the origm of the vitelline membrane. It varies a good deal in thickness, and is not generaUy radiate unless thick. It is a zona radiata in the rabbit (Fig. 60), mole, pig, and sheep. ChemicaUy the vitelline membrane (of Birds) is an albummoid

alhed to keratin.

The tertiary membranes are secreted by the oviduct. The innermost of these is the albumen, white of egg, or jelly. This is found in Elasmobranch fishes, Amphibia, Tortoises, and Crocodiles but not Snakes and Lizards, Ends (Fig. 17), Monotremata, Marsupials (Fig. 68), and sometimes in Placental Mammals


The white of the hen's egg is wound round the ovum m layers, spiraUy arranged. The layers are separated from one another by a thin but tough membrane, the albumen in between successive membranes being fluid. Owing to the rotation of he egg as it passes down the oviduct these layers are spiraUy tw^ted up into cords (the chalazae) on two opposite sides. The chalazae

23*.- Section tliroTigh the shell of the egg of the ostrich Cafter VAaldeyer, after Konigsborn). c, cuticle ; sp., spongy layer of stratified snbstance pierced by canals, which open internally between the bases of the conical processes of the mammillarv layer {m.) sJi.m., shell-membrane. are always placed iii the equator of the ovum so that the blastodisc is midway between them, and they lie in the long axis of the egg-shell.

The white of the egg, which has an alkaUne reaction, contauis

85-88 % water

10-13 % protein

0-7 % salts

0-5 % . . . • • dextrose

and traces of fats, soaps, lecithin, cholesterin, and lutein (to which the faint yellow colour is due).

The proteins are ovo-globulin (6-7 %), ovo-albumin (a mixture of at least two proteins), and ovo-mucoid.

The salts are sodium and potassium chloride, phosphates, and salts of calcium, magnesium, and iron.

In some birds (Insessores) the egg-white does not become opaque on boiling, but gives a transparent jelly, similar to alkali albuminate.

In Fishes and Amphibia the egg-white is a jelly, composed of mucin.

A shell is present, outside the egg-white, in Elasmobranch jfishes. Birds (Fig. 17) and Reptiles, Monotremata, and some Marsupials (Fig. 68).

In the fishes referred to the shell is horny and attached by tendril-like strings to some foreign body. It is composed of keratin.

In Birds and most Reptiles the outer layer of the shell is calcified, the iimer layer being then known as the sheU-membrane. Calcification, however, does not occiu: in some cases (Lacerta vivipara). The shell-membrane is made up of a network of fibriUae of keratin.

The calcareous layer consists - in a Bird's egg - of three sheets (Fig. 23*) : an outer dehcate porous cuticle, a middle spongy sheet, and an inner mamnullary sheet of columns whose conical ends impinge upon the sheU-membrane.

The shell contains 3-7 % of organic matter (keratin), 90 % of calcium carbonate, and small quantities of magnesium carbonate and earthy phosphates.

The colour of the shell is due to bile-pigments. In the hen's egg the shell-membrane is separable into two sheets : between these two air collects at the blunt end of the shell after the beginning of incubation, so forming the air-chamber. This air is for the chick to breathe just before hatching (Pigs. 17, 121).

The Monotremes possess a shell which in Ornithorhynchus is calcified.

Amongst Marsupials a horny shell is present in Dasyurus and Phascolarctos. In Placentaha the shell is invariably absent.

B. The Spermatozoon

In striking contrast to the large inert egg-cell, the spermatozoon is a smaU, actively-moymg body, capable of swimming towards and enteruig the ovum in fertiUzation.

While there is great variety in the form of the animal spermatozoon, two principal types may be recognized, the flageUate or tailed, and the tailless.

The Vertebrate spermatozoon is flagellate. It consists typically of two parts, a head and a tail (Fig. 24).

In the head there is at the anterior end the acrosome or perforatorium, used in perforating the surface of the ovum, and behmd this the nucleus. The nucleus is always dense and homogeneous, and highly chromatic.

The tail consists of an axial filament and a cytoplasmic envelope. Centrosomes are always present in it. Three portions may be recognized : an anterior part including the centrosomes ; this is the pars conjunctionis ; a middle part, pars principalis, as far as the end of the cytoplasmic envelope of the taU ; and a pars' terminalis, in which the axial filament

is naked. -v *

The axial filament (probably the seat of the contractihty of the tail) runs throughout the length of the tail. Anteriorly it termmates m the most anterior centrosome- referred to sometimes as the end-knob-placed immediately behmd, or even embedded m, the nucleus. Behind this are one or more other centrosomes.

The cytoplasmic envelope of the tail extends from the front end of the first to the hmd end of the second region. In the third region only the axial filament is present.

The small part interposed between the head and the tail, that is, between the hind end of the nucleus and the front end of the axial filament, and . containing only the anterior centre some, is sometimes spoken of as the neck, or middle piece. This usage cannot be justified in all cases, as the axial filament may pass right through the anterior centrosome to the nucleus (as in the Amphibian Discoglossus) . The term should therefore be dropped, or apphed to the anterior region of the tail, including aU the centrosomes Though always of the flagellate type, the form of the Vertebrate spermatozoon is variable. Thus, to take a few illustrations (Figs. 25, 26), the acrosome may be large and flattened (spoon-shaped) as in the guinea-pig, or, as is more usual, narrow and pointed (some Amphibia, Reptiles, and Elasmobranch fishes), or much reduced {Phalangista), or apparently absent (Teleostean fishes, possibly Bu-ds). Whether it is really absent or not can, however, only be stated when the origin of the spermatozoon from the spermatid has been studied in these forms. In Birds there is often a remarkable spirally-coiled membrane round the head.

The nucleus may be short and rounded (Teleostei), or short and flat (guinea-pig), or cuneiform (Phalangista), or oval

Fig. 24.- Diagram of a typical vertebrate spermatozoon. H., head; a., acrosome ; n., nucleus ; T, tail; T.l, pars conjunctionis ; T.2, pars principalis ; T.3, pars terminalis; a.c, anterior centrosome ; p.c, posterior centrosome ; /, axial filament ; c, cytoplasm ; e., envelope.

{Tropidonotus), or pointed and elongated, sometimes excessively (Urodela). The anterior end-knob, single (Fig. 25, 2, 3, 25, 1), or much enlarged, as in

centrosome may be a small 4, Fig. 26) or multiple (Fig. Urodeles especially. In Bom

17ra 2^5 -Various spermatozoa. 1. Guinea-pig (Cama) (after Meves). 4. FringUla (the chaffinch). (3 and 4 after Ballowitz.)

binator (a toad) its position near the anterior end of the nucleus is remarkable. The tail filament xs mser ted, there fore, near the front end of the head m this form Thexe ma; be (Pnalangista) one or more intermedxate es. The posterior one, at the end of the first portion of the^d, and therefore some way back, is frequently rmg- ox disc shaped. Iii Urodeles it lies very far back indeed. The cytoplasm of the anterior region often presents transverse or spiral

Fig. 26. - ^Various spermatozoa. 1. Bufo (the toad) (after King). 2. Bomhinator (a toad) (after Broman). 3. Siredon (the Axolotl). 4. Perca (perch). 5. Eaia (skate). (4 and 6 after Ballowitz.)

markings. In the anterior and middle regions (or in the middle region only) the cytoplasm is frequently in the form of a fin, which may have a thickened undulating border, or be spirally coiled round the axial filament. Spermatozoa also vary very greatly in length, as the following table will show. The lengths are given in thousandths of a millimetre.

Crocodilus 20-27

Esox 43

Homo 52-62

Boa 66

Bufo 62-91

Erinacells ..... 85

Cavia . . . . . 93

IIus 107

Eaia 215

Siredon 360-^30

Discoglossus .... 2250

The gigantic spermatozoa of the Amphibian which comes last in the Hst are not, it is hardly necessary to say, proportionately broad. It may be added that even this length is exceeded by the spermatozoa of an Ostracod Crustacean, Pontocypris monstrosa, which are 5-7 millimetres long.

The Chemistry of the Spermatozoon. The most accurate determinations of the chemical composition of the spermatozoon are those carried out on fish sperms.

In the salmon the head (nucleus) of the sperm consists of fat and nuclein and other substances. The nuclem is itself a compound of nucleic acid (C.^ H^g N^^ 0,J with a protamine known as salmm (C30 Hg^ N,^ 0^), the proportions bemg roughly 60% and 35 % respectively of the head, after removal of the fat. The remaining 5 % consists of inorganic matter (Ca3(P04)2, CaSO^) 2-5%, and an iron-containing organic material (the remainder).

In the herring the protamine known as clupein is apparently the same ; scombrin (mackerel) and sturin (sturgeon) are other protamines obtained from the sperm-heads of fishes.

The tail, in the salmon, contains heat coagulable proteids to the extent of 42 %, and fatty substances, 58 %. The latter include lecithin (50 %), fat (30 %), and cholesterin (20 %).

The metamorphosis of the spermatid into the spermatozoon. As has already been stated, four small cells, the spermatids, are produced from each primary spermocyto by the two maturation divisions. Each spermatid is then directly metamorphosed into a spermatozoon.

Fig. 27.- Metamorphosis of the spermatid into the spermatozoon in the salamander (after Meves) 1-6, the whole cell ; 7-9, the anterior end; 10-ld, the posterior end of the head. (For explanation see text. )

The investigation of thjs process in many forms, including several Vertebrates, has shown that there is a remarkable constancy in the changes that take place. One or two examples will suffice.

As a first, let us take the salamander (Fig. 27). The spermatid, emergmg from the second maturation division, is a rounded cell, in which the chromosomes are clumped together while the centrosome, lying in the middle of the centrosphere or sphere of attraction (Idiozom), has divided into two, placed tangentially with regard to the surface of the cell (Fig. 27, 1). While the nuclear membrane is being formed round the chromosomes, the centroeomes detach themselves from the sphere (Fig. 27, 2) and adopt a radial position. When the chromosomes break up into granules, a fine filament- the axial filament of the tail- grows out from the centrosome nearest the surface (Fig. 27, 3). This is the posterior centrosome, and it soon becomes first discoidal, then ring-shaped, the axial filament passing through the ring to attach itself to the other or anterior centrosome (Fig. 27, 4). Meanwhfie the sphere- in which a spherical vacuole has been developed - moves away from the centrosomes to the opposite side of the nucleus, which is of course the anterior end (Fig. 27, 4-6). Here it becomes gradually changed into the acrosome or perforatorium. It becomes oval and an axial rod is formed in it. It is protruded from the cell, becomes pointed, and finaUy much elongated and barbed at its extremity, while the vacuole disappears (Fig. 27, 7-9). At the other, the posterior end of the cell, further changes are taking place. The anterior centrosome first attaches itself to the hinder end of the nucleus (Fig. 27, 5)- now elongated and finely granular- then enlarges and embeds itself in the nucleus (Fig. 27, 6), finally lengthening to form a long eUipsoid body. The axial filament has remained inserted into Tt (Fig. 27, 10). In the meantime, an outgrowth of cytoplasm has occurred on one side (dorsal) of the tail filament to form the fin (Fig. 27, 10). When the fin is weU developed, the posterior ring-shaped centrosome breaks into two halves. One half travels down the other (ventral) side of the tail, carrying some cytoplasm with it, and eventuaUy reaches a point near the end of the middle region (pars principaHs). The other haK remains behind and is fused with the anterior centrosome (Fig. 27, 11-13). The nucleus continues to elongate to form the sperm-head, finally becomes homogeneous, and is divested of its cytoplasmic covering.

The history of the sperm in other types, the gumea-pig for instance, is almost the same (Fig. 28).

In the spermatid can be seen the sphere- including some dark granules- the chromatoid accessory body, and the two centrosomes (Fig. 28, 1). These are dumb-boll-shaped, the outer or posterior is placed radially and bears the axial filament, the inner tangentially.

Fig. 28. - Metamorphosis of the spermatid into the spermatozoon in the guinea-pig (after Meves). 1-4 show the whole cell; 5-9 the head and the front part of the tail ; 9 is seen in profile. In 1 the accessory chromatoid body is rendered in black. In this and the following figures the sphere is shaded or (acrosome) stippled. In 6 and 7 the granules of von Ebner are shown in black. (A full explanation will be found in the text.)

The chromatoid body disappears. The sphere moves round to what will be the anterior end ; in it two portions are distinguishable. A spherical body with a dense central spherule, this is derived from the dark granules of the previous stage, and an irregular body applied to the first, derived from the outer portion of the original sphere. This irregular body presently moves back to the hind end and disappears, but the spherical part becomes transformed into the acrosome (Fig. 28, 2-4). It is applied to the front end of the nucleus, and becomes lenticular (concavo-convex) (Fig. 28, 5). The central dense body then vanishes, the whole projects from the front end of the cell, being attached to the front and sides of the nucleus. Finally it becomes thin and curved (spoon-shaped) (Fig. 28, 9). The nucleus meanwhile having become homogeneous is also flattened and curved, its curvature being opposite to that of the acrosome.

The centrosomes have aU this time been passing through comphcated changes. The anterior one becomes flattened against the nucleus, the posterior hook-shaped, one hmb of the hookdirected outwards- bears the tail filament, while the other, or anterior Kmb, is at right angles to it (Fig. 28, 2).

The hinder limb of the posterior centrosome now becomes divided into a ring behind and a knob in front (Fig. 28, 4). The tail filament passes through the ring, on to the Imob, and then on to the middle of the anterior Hmb. The anterior centrosome and the anterior Hmb of the posterior centrosome then become divided, each into three knobs (Fig. 28, 5). The arrangement is therefore as foUows. A row of three knobs united by filaments next the nucleus ; each of these knobs being similarly united to one of the three knobs of the next row, also united together. The middle knob of the second row is miited to another knob, and into this is inserted the axial filament of the taU which passes through the ring. Later the ring passes backwards some Httle way; it marks the end of the first region of the tail (Fig. 28, The tail filament thickens, the posterior knob being fused with it.

A curious, quite transitory, structure is the tail-sleeve ^Fie 28 4 5) This is a felt-work of fibriUae developed round the froni end of the filament to form a sort of tube. Its existence is short. Most of the cytoplasm-which has by this time passed away from the nucleus to the middle -piece-is peeled off (Fig. 28, 6, 7), with the remains of the sphere and a number of stainable bodies - the granules of von Ebner - which always appear at this time. The remains of the cytoplasm form a sheath round the middle-piece (pars conjunctionis), and apparently a thin investment for the principal part of the tail. In the middle-piece the characteristic transverse (? annular) striations appear (Fig. 28, 8, 9).

Fig. 29. - Sperm-cells of Amphibia in tlieir cysts or follicles.

A, Section of a single seminiferous tubule from the immature part of the testis of a newt (in winter). The spermogonia [^sp.g.) are enclosed ni follicles (/".c.) ; the theca surrounding the tubule, -ex

B, "Bundle of ripe spermatozoa inside a cyst (c), from the testis of tbe Azo'lotl. B.C., Sertoli-cell in which the acrosomes of the spermatozoa are embedded.

These examples are tj^ical of spermogenesis in general. The acrosome is formed from the sphere, the tail filament grows out from the centrosome ; and even where (as in many Crustacea) the tail is absent, the two centrosomes are still present, and the posterior one becomes transformed into a ring.

The close relation between the locomotory organ of the cell and the centrosome is not peculiar to spermatozoa. In certain Protozoa there is a central corpuscle which not only -acts as an organ of cell-division, but also serves as a base of insertion for the flagella or for the axial filaments of the pseudopodia {Dimorpha, Acanihocystis).

The centrosome of the spermatozoon and the sphere - which becomes the acrosome - are both parts of the original division apparatus. They have, however, distinct functions to perform in fertihzation, for while the latter is the perforatorium, employed for ensuring the penetration of the sperm below the surface of the egg, the latter is the centre round which the primary spermsphere is formed. We shall see, nevertheless, that these distinct processes probably depend upon a property which is common to both bodies, and may be due to their community of origin.

The follick-cells of the testis. Like the ova, the male cells are associated with certain nutrient cells in the testis, known also as foUicle-cells, though they do not always form a covering for the germ-cells. Their origin from the germinal epitheUum has already been referred to.

In the lower Vertebrates the germ-cells commonly occur in bundles, each of which is enclosed in a wrapping of folKcle -cells, a cyst, or foUicIe. The cysts are arranged round the walls of the seminiferous tubules. In the immature tubules of the testis the cysts will be found to be small, each containing only one or two spermogonia (Fig. 29 a). But the number of the latter is soon moreased by division, and quantities of germ-cells are subsequently found in each single cyst. In each cyst all the germ-cells are usually in the same stage- whether spermogonia, spermocytes, spermatids, or spermatozoa, complete or incomplete- but in the different cysts in the same tubule different stages are found, though they are not usually very widely different in adjacent cysts (Fig. 30).

Fig 30 - Four cysts or follicles from the same seminiferous tubule of the testis of the Axolotl. 1, contains spermatids ; 2, 3 and 4, three successive stages of the metamorphosis of the spermatid into the spermatozoon. /., follicle ; b.v., blood-vessel outside the tubule.

The mature spermatozoa become arranged in bmidles, and inserted, in each bundle, by their acrosomes into a smgle basal cell of the cyst (Fig. 29 b), facing the wall of the tubule. The cyst which has been much distended, then gives way, and the tails of the sperms project freely into the lumen of the tubule.

The basal cells in which the heads of the spermatozoa are embedded are apparently nutrient as well as supportmg. They are known as the cells of Sertoh, a term first apphed to the corresponding cells of the Mammahan testis.

In the Mammaha the germ-cells are also grouped m bundles but are not enclosed in cysts ; further, they are disposed in several layers, and different stages in development are found in the several layers at one and the same point in a seminiferous tubule (Fig. 31).

Fig 31 - Testis of mouse. A small part of a section tlu-ongh a seminiferouV tubule in six different conditions {a-f) 1-20 stages in spermogeneiis (for further explanation see text). 8., Sertoli-cell ; th., theca.

The basal layer contains the supporting cells or cells of Sertoh, which (in the mouse) are recognizable by the presence in the nuclei of one large nucleolus and two large spherules of chromatin. In addition to these there are the spermogonia (indifferent cells), which are derived , either from the central cells of the young testis (see Fig. 13) or from the surrounding follicle-cells, or from both.

Internal to this basal layer are about three others, in each of which the germ-cells are in a different stage. For the sake of illustration the whole spermogenesis may be divided into twenty stages, as follows :

  1. Indifferent cell or spermogonium.
  2. Transition to spermocyte.
  3. Primary spermocyte : leptotene stage.
  4. Transition to synaptene stage.
  5. Advanced synaptene.
  6. Pachytene,
  7. Pachytene to diplotene.
  8. Later diplotene.
  9. Commencement of ring-formation.
  10. Ring-shaped (heterotypic) chromosomes formed.
  11. First matviration division.
  12. Secondary spermocytes.
  13. Second maturation division.
  14. Spermatids.
  15. Later spermatids.
  16. Commencing metamorphosis, with short tail filament.
  17. Later stage.
  18. Appearance of von Ebner's granules.
  19. Peeling off of cytoplasm ; spermatozoon complete.

Beginning with, for example, a stage in which the spermogonia of the basal layer are in stage 1, the cells of the second layer m the fourth,, those of the third layer iii the tenth, and those of the fourth layer in the sixteenth stage, the progress of development in each layer may bo readily watched. As layer ii passes into stage 5, layer in shows first maturation spindles, while the spermatids of the fourth layer begin to be metamorphosed into spermatozoa, and so on. By the time that the spermatozoa of the fourth (inmost) layer are ripe and ready to drop into the tubule, the cells of the third layer have reached the spermatid stage ; those of the second are in a late prophase of the first maturation division, whUe the somewhat flattened spermogonia of the basal layer are becoming cubical and preparing to grow into spermocyiies. By the time the ripe sperms have been thrown off, the young spermocytes have detached themselves from the basal layer and he in a distinct second layer below the third and fourth layers, which are respectively in the ninth and fifteenth stages, and so the starting-point is reached once more.

The germ-cells, therefore, originating in the basal layer, are brought nearer and nearer the lumen of the tubule as the ripe sperms of the inner layer are cast ofE and fresh layers formed from below. As the spermatids undergo their metamorphosis, they become grouped into bundles ; and, in each bundle, the spermatozoa are inserted by their acrosomes into the extremity of an elongated Sertoh cell, which is retracted once more when the spermatozoa have been set free. The tails, therefore, float out into the seminiferous tubulev

The ovum and the spermatozoon are obviously different from one another in almost every respect. The former is large, inert, and, when fully ripe, as we are shortly to see, without a centrosome. It is rich in cytoplasm, contams reserve food material, arid has a structure, related in a very definite way to the structure of the embryo which is to be developed from it.

The spermatozoon, on the other hand, is motile and small, has httle cytoplasm (except hi the tail, which is of no importance in fertiUzation, smce it may be left outside the egg) but is provided with one or more centrosomes, as well as with an apparatus for entering the ovum.

We have still to examine the structure of the nuclei of the germ-cells, a structure which is the result of the pecuhar nuclear changes mvolved m maturation. This exammation will show us that in their nuclei the germ-cells are alike.


B. M. Allen. The origin of the sex-cells of Chrysemys. Anal. Am. xxix, 1906.

E. Ballowitz. Untersuchungen uber die Struktur der Spermatozoen. L Arch. mikr. Ami. xxxii, 1888. III. Arch. mikr. Anal, xxxvi, 1890.

E. Ballowitz. Die merkwiirdigen, 2\ Millimeter langen Spermien des Batrachiers Discoglossus pictus. Arch. mikr. Anal. Ixiii, 1904.

J. Beard. The germ-cells. Journ. Anal, and Phys. xxxviii, 1904.

E. VAN Beneden et C. Julin. Observations sur la maturation de I'cellf chez les Chiropteres. Arch, de Biol, i, 1880.

U. Bbm. Beitrage zur Entwickelungsgeschiehte der Leibeshohle und der Genitalanlage bei den Salmoniden. Morph. Jahrb. xxxii, 1904.

I. Beoman. Ueber Bau und Entwickelung der Spermien von Bomhinator igneus. Anal. Anz. xvii, 1900.

R. BuRLiN. Chemie der Spermatozoen. Ergebn. Physiol, v, 1906.

C. A. EiGENMANN. On the precocious segregation of the sex-cells in Micrometry aggregatus. Journ. Morph. v, 1891.

0. Hammaesten. Text-book of Physiological Chemistry, trans, by J. A. Mandel. New York, 1911.

H. D. King. The egg of Bufo lentiginosus. Journ. Morph. xvii.

K. VON KoRFF. Zur Histogenese der Spermien von Phalangisla vulpina. Arch. mikr. Anal. Ix, 1902.

E. KoKSCHBLT u. K. Heider. Vergleichende Entwickelungsgeschiehte der wkbellosen Thiere. Allg. Th., Lief. L ii, Jena, 1902.

H. Lams et J. Doorme. Nouvelles recherches sur la maturation et la f6condation de I'cellf des Mammiferes. Arch, de Biol, xxiii, 1908.

F. McClendon. On the nucleo-albumin in the yolk-platelets of the frog's egg. Amer. Journ. Phys. xxv, 1909.

F. Meves. Ueber Struktur und Histogenese der Samenfaden von Salamandra. Arch. mikr. Anal. 1, 1897.

F. Meves. Ueber Struktur und Histogenese der Samenfaden des Meerschweinchens. Arch. mikr. Anal, liv, 1899.

W. Rubaschkin. Ueber das erste Auftreten und Migration der KeimzeUen bei Vogelembryonen. Anal. Hefte, l^e Abt., xxxv, 1908.

W. RuBASCHKiN. Ueber die UrgeschlechtzeUen ' bei Saugetieren Anal. Hefte, I'e Abt., xxxix, 1909.

J. SOBOTTA Ueber die Entstehung des Corpus luteum der Saugethiere. Aiiat. Hefte, 2'^ Abt., viii, 1899.

O VAN DER Stricht. La structure de I'cellf des Mammiferes. Arch, de Btol. XXI, 1905.

W. Waldeyer. Die Geschlechtzellen, in 0. Hertwig, Handhuch der JLnlwicklungslehre der Wirheltiere. Jena, 1906.

1902 nt and inheritance. New York.

'ovogen^se et I'organogendse de 1 ovaire des Mammiferes. Arch, de Biol, yiyn, \mi

III. The Maturation of the Germ-cells A. In the male

The Urodelous Amphibia have always been a favourite object for the study of these changes, and may conveniently be taken by us as a type.

It AviU be recalled that duriag the spermogonial divisions the full somatic number of chromosomes is seen. The mitosis is of the ordmary character (Fig. 32). The granules of chromatm increase, run together m the form of beaded rows, which become the V-shaped chromosomes. The nuclear membrane has m the meantime broken down, the centrosome has divided, and around each daughter centrosome an aster is appearmg. The chromosomes then undergo longitudmal fission and, so spht, are placed on the equator of the spmdle now developed between the two centrosomes. The daughter chromosones are then puUed apart by the spmdle-fibres attached to them to the opposite spmdle poles, and there passmg through the same series of changes in the reverse order become the daughter nuclei. Meanwhile a celldivision has occurred m the equatorial plane of the spmdle, m which process the mtermediate bodies-thickenings of the spmdle fibres- play an important part.

AU the features of an ordmary mitosis are here : the chromatm is the only part of the nucleus to be divided ; for that purpose it is thrown mto the form of chromosomes, which ^ht lengthways mdependently of any external agency , a divis on a^tatuLasters and sphidle-is ^onst^f^^/^^^ f^e centrosomes and probably by them, the function of -^^^f^ '^ puU apart the halves of the aheady divided chromosomes and to LL?the division of the cell. When the W--/^^ceased dividing they enter upon a

time the nucleus passes through complex changes, whlch are in reality the prophases of the fost of the two maturation divisions. This first division is of a very different character to an ordinary mitosis. There ensues the second division. This, with one important exception, resembles the mitoses of the spermogonia.

Fig. 32. - Stages in the karyokinetic division of the spermogonia of the newt.

The first maturation division (Figs. 33, 34). In the nucleus of a spermogonium the chromatin is in the form of fairly coarse lumps uniformly distributed over a wide achromatic reticulum. As the growth of the cell and its nucleus begin the chromatin becomes subdivided mto finer granules, which soon arrange themselves m rows or filaments ; in each row the granules are connected by threads of the achromatic reticulum, while similar threads pass from one filament to another. This is the narrow thread or leptotene'^ stage. As the nucleus enlarges still more

Fig. 33.- Prophases of the heterotype division in the male Axolotl. 1, Nucleus of spermogonium or young spermocyte ; 2, Early leptotene ; 3, Transition to synaptene; 4, Synaptene with the double filaments converging towards the centrosome ; 5, Contraction figure ; 6, 7, Pachytene ; 8, Early, 9, Later diplotene ; 10, The heterotypic double chromosomes ; the nuclear membrane is disappearing.

it is seen that on one side some of the filaments are arranged in pairs, and converge towards one point, the point where the centrosome m its centrosphere is placed. On the other side of the nucleus the filaments pass into the general network. This is the paired thread or synaptene stage.

1 These and the followmg terms were first proposed by von Winiwarter in his classical work on the oogenesis of the rabbit.

By coalescence of the component granules the filaments become shorter and thicker : at the same time in each pair the filaments approach one another so closely that only a narrow slit is left between them. On one side the pairs of filaments still converge towards the centrosome, but on the other are inextricably coiled and tangled together into a bunch which is withdrawn some little way from the nuclear membrane. The pairing of the filaments can, however, be seen in the tangle. The several pairs are still imited by achromatic threads, the filaments being toothed at each point of insertion of such a thread. A few threads stretch across the empty space between the tangle and the membrane. This is the contraction figure.

Fig. 34. - ^First maturation division in the male. 2, Salamander, the remainder Axolotl. 1, 2, The heterotypic chromosomes on the spindle (metaphase) ; 3, Anaphase ; 4, 5, Telophase ; 6, Resting nuclei ; 4-6, Celldivision into two secondary spermocytes.

The members of each pair of filaments now unite throughout their length, so that the longitudinal sHt disappears. The thick filaments still converge towards the centrosome side, where apparently they end against the nuclear membrane. There is, therefore, not one continuous filament or spireme, but several.

The other ends of the filaments pass into the tangle, which is still retracted from the nuclear membrane, but becoming looser as the nucleus enlarges. The coil is soon still more unravelled and occupies the whole of its side of the nucleus. This is the pachytene stage.

The several filaments now separate from one another, so that the polar convergence is lost, and coil in various directions through the nucleus. At the same time the longitudinal sUt reappears in each, and the filaments are once more paired, so reaching the diplotene condition. Their surfaces are still toothed where the connecting achromatic threads are inserted. Soon, however, these cross threads disappear and the filaments become smooth. At the same time the members of the several pairs begin to separate a little from one another, in places if not throughout their length.

The nuclear membrane now breaks down and disappears, the pairs of filaments shorten and thicken, and assume the most various shapes and sizes. A pair may be in the form of two straight parallel rods, or two curved parallel rods, either V-shaped, or C -shaped, or two rods parallel at one, divergent at the other extremity, and so -||--shaped ; or the sht between them may be expanded in two or more places, and then the two may be tAvisted over one another mto a figure of g or or by expansion of the whole sht, while the rods are united at the ends, may be ring-shaped, while finally the ring may be pushed in in four places and assume the form of a cross, =[}:. These bizarre double bodies are the chromosomes of the first maturation division. It seems clear that they are derived from the separate paired filaments of the diplotene stage, these from the thick filaments of the pachytene stage, and these again from the paired filaments of the synaptene nucleus. The origm of these we shall have to discuss later on.

The number of the double chromosomes, and therefore of the several double filaments in the earlier diplotene, pachytene, and synaptene stages, is one-haK that seen in the spermogoma. The reduction from the somatic number {2n) to the germ number (n) has already taken place. It seems that this half number must be estabhshed in the synaptene nucleus.

The actual division now occurs (Fig. 34). A spindle is formed in the ordinary way, and the double chromosomes are thrown upon its equator in such a way that the two ends of each member of a pair he in the equatorial plane. This is easily seen where the pair retains the original form of two closely-parallel rods separated by a longitudinal slit, and can often be made out in the ring- and cross-shaped and other chromosomes.

The members of the pairs now come apart and travel to opposite spindle poles, where they coalesce and pass into the condition of resting nuclei. The cell, meanwhile, has divided and the two secondary spermocytes have been formed. The nucleus of each of these, it is clear, contains only one -half of the ordinary number of chromosomes.

The division which we have just witnessed is unlike an ordinary mitosis in at least two respects. First, the number of chromosomes is reduced from the somatic to the germ number, and second, the chromosomes are double and frequently of extraordinary shape. For these reasons the division is spoken of as heterotypic, or xmlike the usual type. The term meiotic or reducing, also applied to it, refers to the numerical lessening of the chromosomes.

We have now to inquire whether this division is or is not like an ordinary mitosis in another respect, the manner ia which the chromosomes are divided. Ordinarily, as we know, the chromosomes are longitudinally divided ; but on this occasion it is held by many observers that the division, albeit in appearance longitudinal, is in reahty transverse.

The interpretation of the nuclear changes is a matter of considerable difficulty, and very diverse opinions are entertained (1) as to the origin of the double filaments seen in the synaptene and later stages of the prophase, and (2) as to the mode of formation of the ring-shaped chromosomes seen in the actual mitosis ; different combination of these diverse opinions has led to the formulation of three principal views.

I. It is held that (I) the double filaments of the synaptene stage arise by longitudinal fission of the filament, that the longitudinal split disappears, but reappears (2) to form the cavity of the rings. Hence the actual division is longitudinal (Meves). This is illustrated in the accompanying diagram (Fig. 35, I).

For the sake of simplicity we will suppose that the full number of chi'omosomes is four, the reduced number two. We will further suppose ttot these four chromosomes are really different

Fig. 35.- Diagram to illustrate three interpretations of the first matoasee text.)

Tne another though apparently identical. Let us call them 7 A' B, and B: In the prophase of the mitosis two mstead of' four filaments appear. We may suppose that each of th«a consists of two ordinary chromosomes muted end to end say aTo 4' and B to B'. Each filament becomes then split lengthways 1). the sHt widens out untU each filament assumes a ling shape (I, 2), and the rings are then so placed on the equator of the spindle that the ends of the chromosomes lie in the equator (I, 3). Hence, since each half ring consists of an A and an A', or of a 5 and a jB', when the halves are separated and travel towards the spindle poles, each daughter nucleus of a secondary spermocyte will receive a chromosome of each kind, A, A', B, and B'.

II. On the second view (von Winiwarter, Schreiner, Agar), (1) while the paired filaments of the sjmaptene stage are believed to arise, not by longitudinal fission of the leptotene, but by apposition of distinct chromatin filaments (that is, chromosomes), the formation (2) of the rings from these double ' filaments is in accordance with the first view.

The diagram (Fig. 35, II, 1) shows the four chromosomes united in pairs by their entire length, though presenting every appearance of longitudinally spht rods : A is paired with A', and B with B'. The chromosomes of each pair then separate to form rings, remaining united only by their ends, and then are placed on the spindle ua such a way that these ends lie in the equator. It follows that A and B face towards one. A' and B' towards the opposite pole, and hence that each nucleus of a secondary spermocyte receives not all tour chromosomes, but only two, say A and B, ov A' and B'.

The division, therefore, is not really but only apparently longitudinal : the result is the same as though A and A' (and B and B') had been united end to end, and then separated by a transverse division of the double chromosome so formed.

III. On the third view (Farmer, Montgomery), (1) the double thread of the synaptene and pachytene is formed by the longitudinal splitting of the chromatin filament ; but (2) the rings do not arise by the opening out of the spUt. The longitudinal division disappears, and the filament is first gathered up into half as many loops as there are chromosomes in the spermogonia, and these loops then separate as the n ring-shaped chi omosomes. The rings are therefore open at one end only, and the cavity of the ring arises, not by the opening out of the longitudinal split (for that has disappeared), but by the bending of the two halves, united end to end, of each double chromosome upon one another (Fig. 36, III). That is, the filament, consisting of Aj A', B, and B', is first gathered up into two loops, A being bent on A', and B on B', and then the loops separate. In the mitosis (III, 3) the rings are so placed on the spindle that A becomes separated from A' and B from B', so that one secondary spermocyte receives A and B, the other A' and B' (or, of course, A and B', A' and B).

Fig. 36.- Second maturation division in the male (Axolotl). 1, Prophase (split spireme); 2, The homoeotypic spht chromosomes on the spindle; 3, Polar view of the same ; 4, Anaphase ; 5, Telophase ; 6, Resting nuclei and completion of cell-division ; in each spermatid the centrosome has divided, and the sphere has become detached.

The result is therefore the same as on the second hypothesis.

Considering the diversity of opinion, it would be rash to dogmatize, but it may be pointed out that the evidence on the whole is against the mode of formation of the rings adopted by the third view. It does seem as though the rings were made by the opening out of the double filaments. We are left, therefore, with the choice between the first and second hypotheses. We can only say that the way in which the members of the pairs of filaments diverge into the general network in the fourth stage (Fig. 33, 4) suggests apposition rather than fission, and this involves ultimately a transverse division of double chromosomes, and that the phenomena of maturation observed in a number of Invertebrate forms corroborate this view.

Before discussing the theoretical significance of this mode of division, we shall describe the second maturation division (Fig. 36).

The nucleus of the secondary spermocyte soon emerges from the resting condition, and a chromatic filament appears. This filament becomes longitudinally split and then divided into a number of V-shaped chromosomes, themselves therefore split lengthways. The number of chromosomes is the half somatic, n. A spindle is developed, the spht chromosomes are placed on its equator, and division takes its ordinary course, resulting in two spermatids, the nucleus of each of which therefore possesses n chromosomes. La the V-shape of the chromosomes, as well as in their longitudinal division, this second mitosis is of the ordinary type. Hence it is called homoeotypic. Each spermatid becomes metamorphosed into a spermatozoon in the fashion already described.

The phenomena of maturation in the male are, as far as is known, similar in other forms {Myxine, Elasmobranchs, Mammaha). Each ripe male cell, therefore, is provided with only haK the number of chromosomes seen in the spermogonia and in the tissue cells of the body. Whether the n chromosomes in all the spermatozoa are or are not alike depends upon the interpretation placed on the first maturation division, as well as upon our views of the nature of the chromosomes.

B. In the female

While m the male the first or heterotypic division follows immediately upon the prophases, in the female the two episodes - prophase and division- are separated by an interval, sometimes of great length, a year or more- during which the yolk is deposited in the cytoplasm to the accompaniment of complex nuclear changes.

Prophases of the heterotypic division. The oogonial divisions come to an end at a fairly early period, and growth of the oocyte begins almost at once. The prophases of the heterotype are therefore usually found only in very young animals- in the tadpole of the frog, or the new-born or embryonic Mammal.

These two afford good examples. The nuclear changes which are readily seen in the tadpole's ovary (Fig. 37) are obviously closely j)arallel to what we have observed in the other sex.

A stage in which the chromatin is in the form of scattered granules is followed by one in which the granules run together to form the leptotene filament. Then comes the synaptene, with parallel filaments, followed by the contraction figure. The paired filaments emerge from the tangle to converge to one pole, the tangle itself beiiag withdra-WTi from the other side of the nucleus. The pachjiiene and diplotene follow in due course. A remarkable change now occurs in the straining capacity of the chromatin filaments. Up to the diplotene stage they behave in the usual way, showing great affinity for chromatin stains (carmine, haematoxylin, and basic aniline dyes) ; but from now onwards they lose this faculty and stain only with the acid plasma dyes. Meanwhile, the number of nucleoli (these also stain in acid dyes) is increasing, and presently it is seen that granules of chromatin (that is, granules which are coloured by the ordinary chromatin dyes) begin to settle upon (? be precipitated round) the nucleoli. By what appears to be a continuation of this process the nucleoh become converted into highly chromatic bodies.

The filaments (chromosomes) persist for a while, but will eventually disappear.

Precisely similar phenomena are seen in the young Mammahan ovary (Fig. 38), and only one or two points require to be mentioned. There is a very obvious centrosphere with included centrosome on one side of the nucleus (this usually goes by the name of the yolk-body of Balbiani), towards which the filaments of the synaptene and pachytene converge. In the early stage of contraction the paired filaments are seen to emerge from the rather open tangle on this side, while on the other a few filaments, also paked, stretch out to the nuclear membrane. In the later contraction figure the latter are retracted and the tangle, much closer, hes wholly on the side of the centrosphere.

After the diplotene stage the ring-shaped figures of eight and other forms of double chromosomes are seen, but then the chromosomes break up into their constituent granules and range themselves along the achromatic threads which make a network through the nucleus. This is the diciyale condition, and in this the nucleus remains through the growth period until the moment of maturation arrives.

Fig. 37. - Prophases of the heterotypic division in the female (ovary of tadpole). 1, Nucleus of oogonium ; 2, Leptotene ; 3, Synaptene ; 4, 5, Contraction figures ; 6, Pachytene ; 7, Later pachytene, multiplication of nucleoli ; 8, 9, Diplotene : the chromatin filaments are becoming achromatic ; granules of chromatin are being deposited on the nucleoli.

Fig. 38. - Prophases of the heterotypic division in the female (Mammals). 1-6, Kitten three days old ; 7, Mouse embryo shortly before birth ; 8, Mouse eight days old.

1, Nucleus of oogonium or young oocyte ; 2, Leptotene ; 3, Synaptene ; 4, Contraction figure ; 5, Pachytene ; 6, Diplotene ; 7, Heterotypic clnomosomes ; 8, Dictyate.

In 2-5 the centrosphere and centrosome (volk-body of Balbiani) are shown with the chromatic filaments of the nucleus converging towards them.

j'j-G. 38*.- Small ovarian egg of the frog surrounded by its follicle (/.) and theca (th.), which is continued into the pedicle {p.). b.v., a blood vessel between follicle and theca ; v.rn., vitelline membrane ; ch., chromatin filaments, now achromatic ; n., chromatic nucleoli, ejected from the nucleus (n'.) and becoming achromatic (w".).

The period of growth and deposition of yolk. The nuclear changes accompanying the deposition of the yolk in the oocyte have only been studied in the Amphibia, to which we must now accordingly return (Fig. 38*).

We have seen that after the prophases the chromatin filaments become achi-omatic, while the nucleoH increase in number and become chi-omatic. The filaments gradually break up into a number of small granules, which disappear, or at least become indistinguishable from the general ground substance - or magma - of the nucleus.

The nucleoli become more numerous, larger, and more chromatic. They pass into the cytoplasm in one of two ways : either they are bodily ejected from the nucleus, lose their staining capacity and break up into small fragments, or else they disintegrate inside the nucleus, the products of their disintegration then passing out - either in the form of small particles or in solution - through the nuclear membrane into the cytoplasm. The nucleoli consist of nucleo -protein, and the result of their transference to the cytoplasm is that the latter first acquires an affinity for the chromatin stains, and then begins to secrete yolk-granules. There is thus a direct connexion between the nucleo-protein of the nucleoli and that which, as we have seen, is demonstrable in the yolk.

It appears that this cycle of changes is repeated many times during the growth of the oocyte, fresh nucleoli being formed, moving to the centre of the nucleus, and there disintegrating.

This passage of material from the nucleus to the cytoplasm of the egg-cell during the time of growth and yolk-formation is of constant occurrence in animals. The material may be solid and bodily ejected or liquid and diffusible, it may be chromatic or achromatic, but it is always given off and is always concerned m yolk-secretion. The chemical changes are, unfortunately, not fully understood.

The material is known generally as ' yolk-nucleus '. It has sometimes been confounded with another quite distinct structure, the sphere and centrosome. In Mammalian ova the sphere has indeed long been known as the yolk-body of Balbiani (Pig. 39). In the Mammals the sphere usually divides into two or more bodies, which persist for some time, but disappear (in the bat) when there are two or three cell-layers in the follicle.

In some Mammals (Cavia, Vespertilio) chromatoid bodies are found in the cytoplasm. These may be of nuclear origin and correspond to the yolk-nucleus of Amphibia.

The yolk-nucleus is a very important contribution made by the nucleus to the structure of the cytoplasm : a second contribution has still to be made.

With the growth of the oocyte the nucleus has been enlarging pari passu, and by the time growth is completed is of considerable size. It lies in the axis, but excentrically, near the surface in the animal half of the egg. The oocyte is now ready for the first maturation division.

Maturation. The nuclear membrane breaks down and disappears. From a very small part of the achromatic substance of the nucleus a spindle - the first polar spindle - is formed (Fig. 41), and on this are placed the heterotypic chromosomes, of which we shall speak in a moment. The whole of the rest of the contents of the nucleus - chromatic nucleoli and achromatic granular ' magma ' - are cast into the cytoplasm. This is the second contribution made by the nucleus to the cytoplasmic structure, and it is of considerable importance, since on it in part depends the diflEerence between animal and vegetative hemispheres.

As we have already had occasion to observe, there is a definite relation between the polar structure and symmetry of the egg and the structure of the embryo which is to come out of it, inasmuch as the anterior end is always developed near the animal, the posterior end near the vegetative pole. The structm-e of the embryo is at this moment bemg predetermmed in the egg, by the dispersal of the contents of the nucleus.

This is a fact of universal occurrence. When the germinal vesicle breaks down, only a smaU part of it is utihzed m the formation of the chromosomes which take part in the maturation mitosis. The remainder is given to the cytoplasm, of which it forms henceforward a definite and integral part. Experiment has sho^vn that that part is causally related to the development of certain organs, is therefore a vehicle of inheritance. It will be noticed that this process is without parallel in the male sex.

We return to the first polar spindle and its chromosomes.

The chromosomes appear first, as beaded filaments of heterotypic form - wrings, crosses, figures of eight, curved rods, and so on (Fig. 40). Their number is the half -somatic or germ-number n. It has been disputed whether these chromosomes are identical with those which were formed at the end of the prophases, in the young oocyte.

It must be remembered, in discussing this question, that the hypothesis of the individuahty of the chromatin A does not necessarily involve 'p^-'\X^ that of the individuality of

the chromosomes. We have so™1-from^?SSi? th?o»y?; seen elsewhere that there is of the Axolotl (Siredon) just before reason for beheving that the membrane breaks down.

chromatia of the nucleus comprises a number of quahtatively unlike bodies- not merely that the chromosomes are different, but that they are composed of individually different granules. It is also probable, to say the least, that chromosome formation is a nratter of precipitation from solution, for there is certainly much more chromatin in a dividing than in a resting nucleus, and the chromatin often disappears from view in the latter condition. But a body endowed with certain properties wiH retain those properties in solution and emerge from solution with the same, and in a mixture of unlike bodies each wiU retain its own properties in solution and exhibit them afresh when reprecipitated. The chromatin granules are such bodies, and wo may weU suppose that they do retain their properties in spite of their disappearance. It does not foUow, however, that the granules are associated always in the same order to form chromosomes, though that may be so. Hence the chromatin granules may well retam their individuaUty while the chromosomes do not.

The chromatin, therefore, of those heterotypic cliroinosomes that now apj)ear may, on this view, bo regarded as identical with the chromatin of the prophases.

Fig. 42.- The maturation divisions in the female (Axolotl). 1, First polar spindle with heterotypic chromosomes ; 2, Extrusion of first polar body ; 3, Appearance of second polar spindle ; longitudinal division of chromosomes in egg and in first polar body ; 4, Second polar spindle radial ; homoeotypic chromosomes on equator (metaphase) ; 5, Polar view of the same ; 6, Anapliase ; 7, Extrusion of second polar body ; 8, Second polar body with resting nucleus ; 9, Female pronucleus in resting condition, closely surrounded by yolk-granules.

When the spindle is formed the chromosomes are placed on it and shorten and thicken (Fig. 41). The spindle then moves to the surface at the animal pole, Avhere it takes up a radial position, closely surrounded by yolk-granules. The actual maturation divisions now occur.

Fig. 41. - Germinal vesicle of the oocyte of the frog just before maturation (after Carney). The nuclear membrane has disappeared. The first polar spindle, bearing the heterotypic cliromosomes, is seen in the middle of the nucleus {p.s.). n., nucleoli ; v.m., vitelline membrane ; /., follicle ; th., theca.

Fig. 43. - Oocyte of mouse with heterotypic spindle from the Fallopian tuDe. ihe oocyte is still surrounded by the cumulus of follicle-cells.

The first maturaiion division (Fig. 42, 1, 2). The heterotypio chromosomes are placed upon the spmdio in the same way as in the male - that is, Avitli the extremities of the half -rings in the equators. The half-rmgs break away from one another and pass to the spindle poles. Cell-division now occurs. This is extremely unequal. The outer group of chromosomes, with a small quantity of cytoplasm, is cut off as the first polar body from the egg ; it lies in a depression at the surface. The inner group of chromosomes remain in a clear area in the egg, now the secondary oocyte.

The first polar spindle is found (in Siredon, and generally in Amphibia, also in Bnds) in the egg as it passes mto the oviduct. Li Mammalia - ^where it is also known to be heterotypic (Fig. 43) - ^it may be formed while the egg is in the ovary, or after it has passed into the Fallopian tube.

The first maturation division in the female evidently involves similar changes to those seen in the male : the prophases, the number and form of the chromosomes are all exactly the same. The interpretation of the manner of division of the chromosomes - ^whether longitudinal or transverse - which is adopted for the one, may therefore be applied to the other.

The second maturation division (Fig. 42, 3-9). Without passing into a resting condition the V-shaped chromosomes in the egg undergo longitudinal fission, as also do those in the first polar body. A number of parallel fibres, tangentially placed, now appear - the second polar spindle. The spindle is soon rotated into a radial position and the V-shaped chromosomes, already split, are thrown upon its equator with their apices towards the spindle axis, as in the male. Their number is, of course, n. The halves of the chromosomes then separate and pass to the spindle poles. Another unequal cell-division now occurs. The outer group of chromosomes, together with a httle cytoplasm and one or two yolk-granules, is extruded as the second polar body, while the inner group remain in the now mature ovum as the female nucleus, or rather pronucleus, to employ the more usual term.

In both the second polar body and the ovum the chromosomes break up, a membrane is formed round them, and the nucleus passes into the resting condition.

Since the chromosomes are V-shaped, are longitudinally divided, and are present in half the normal number, this division is evidently homoeotypic, as in the male.

The second polar spindle is formed as the egg passes down the glandular region of the oviduct (in Siredon and most other Amphibia). In the uterus the polar spindles are in metaphase (with the chromosomes in the equator). The division is not completed until after the egg has been fertilized (which is just after the egg is laid).

Where fertihzation is internal (Elasmobranchs, Birds, Reptiles, Mammals) the second polar body is extended while the egg is in the oviduct.

Although the chromosomes of the first polar body have divided, cell-division (in Siredon) does not usually follow. In other cases the first polar body does divide.

A centrosphere - if not an actual centrosome - is present at the poles of both the first and second spindles. In the mature ovum there is, however, no trace of it. The female pronucleus is immediately surrounded by yolk-granules (Fig. 42, 9).

Nature of the reducing division. We have already assumed for the purposes of illustration that the several chromosomes of a nucleus are genuinely different from one another. We may now add that there is experimental evidence (which we cannot discuss here) in support of this ; it is further probable that the granules of which each chromosome is composed are again of different values. Secondly, there are cases where the chromosomes are of different sizes (certain Insects), and in these cases they are found in pairs (in tissue- and in young germ-cells), the two members of a pair being of the same size. In the heterotypic division of maturation the members of the paks get separated from one another, so that each secondary spermocyte (and consequently each spermatid after the second homoeotypic division) receives a similar set of different-sized chromosomes.

Attention has akeady been called to the difference in size of the ring-shaped chromosomes in Siredon.

Now when a row of granules (or chromosome) is divided lengthways each half contains its due portion of each granule, and hence each daughter nucleus receiving half of each chromosome receives ipso facto a specimen of each different granule. The two daughter nuclei are therefore alike and a longitudinal division of the chromosomes is merely quantitative.

If, on the other hand, the row of granules (or chromosome) is transversely divided, or, what is the same thing, if two different chromosomes are separated from one another, each daughter nucleus will not receive a specimen of each different granule or chromosome, but only one-half, the remainder passing to the other nucleus, and the division is quaKtative.

The first condition may be represented by some such formula as this (where a-h are the quahtatively different granules in a chromosome, A,A',B, B', &c., whole chromosomes) : abcdefgh A A' B B' G C D D' abcdefgji' A A' B B' G C D D

the line being the division, while the second condition will be represented by

abed A B G D or

efgh A' B' G' D' Ordinary somatic mitoses are therefore quantitative, and so is the second homoeotypic maturation division. If, however, we adopt the view that in the heterotypic mitosis a transverse division of the chromosomes is involved, then we must further beheve that the division is quahtative, and consequently that the secondary spermocytes, and eventuaUy the spermatozoa, receive chromosomes of different kinds. Of every four spermatozoa produced from a single primary spermocyte, therefore, two wiU be aUke of one kind (containing, say, A, B, Sec), while two will be alike of another kind (containing A', B', &c.).

But it is evident from the foregoing that identical nuclear changes occur during maturation in the two sexes. The prophases of the first division- with the leptotene, synaptene, pachytene, and diplotene stages- are the same, and whatever view is taken of these phenomena must hold good for both sexes. In the female the growth period intervenes between the prophases and the actual division, but when this division occurs it is of the same form as in the male, heterotypic. The second division is homoeotypic in both sexes.

While, however, the cell-divisions are equal in the male- resulting in four spermatozoa - in the female they are unequal - giving one large ovum which receives practically the whole of the cytoplasm and the yolk, and three small polar bodies. The similarity of the nuclei shows that in spite of their small size these polar bodies are in reality potential ova, and there are cases where they are large - as large as the ovum - and can be fertilized and develope.

Like the spermatozoa, the ovum (and polar bodies) receives only one-half the somatic number of chromosomes. As we shall see more fully in the next section, these chromosomes form a complete set, as do those of the male. If- as is probably the case- there are varietal differences between individual spermatozoa in respect of these chromosomes, the same will be true of the ova.^

But what the further significance of these differences is, if they exist, we do not know. The chromosomes of the spermatozoon and ovum are certainly vehicles of inheritance - ^that is, concerned in the transmission of at least some of the inheritable characters of the species from one generation to the next. But since every spermatozoon or ovum can perform this function as well as every other, we are driven to conclude that each one possesses a complete set of the necessary specific chromosomes ; but that in different spermatozoa or ova the chromosomes may be of different varieties- that is, be concerned in the transmission of different varieties of the same inheritable character. This may be expressed by the following scheme. A B, G, D, Sec, are the n different specific chromosomes. In the tissue-cells and young germ-cells there are 2n, each kind bemg represented by two slightly different varieties, namely, A and A', B and B', G and G', &c. In the prophases of the heterotype division A and A' unite,^ and so B and B', G and G'.

In the actual heterotype division A and A', B and B', G and G' are separated from one another, so that each secondary spermocyte or oocyte has A or A', B or B', and so on.

1 Provided of course that priraary oocytes differ inter se in the arrangement and distribution of the heterotypic chromosomes.

2 If the union is by parallel apposition it is further possible to suppose that the individual granules of"^ which 4 and^' are composed pair ofi each with each, namely a with a', b with b', and so on.

The homoeotypic division is quantitative, hence each spermatozoon or ovum obtains A or A', B or B' , and so on ; that is, a complete set of the various kinds of chromosomes.

In only one respect are there chromosomal differences between the two sexes. In certain forms (Insecta), and possibly in others also, there is an accessory chromosome or heterochromosome (often paired), which not only differs in size and behaviour from the ordinary chromosomes, but is not the same in spermatozoon and ovum. The variations in the behaviour of this body or bodies are too complex to be discussed here, but those who have investigated it beheve it to be concerned in the determination of sex. Apart from the heterochromosomes and the varietal differences of the ordinary chromosomes, the germ-nuclei are exactly alike.

We have now to see how the two nuclei - each containing one-half the somatic number of chromosomes - are brought together when the germ-cella unite in the act of fertilization.

IV. Fertilization

The Axolotl- 8iredon- will serve as a type (Fig, 44). The spermatozoon- which is of the same form as that of the newt and salamander - after passing through the mucin jelly surrounding the egg, reaches the surface of the latter. It approaches the egg with its anterior end- acrosome - and always in the pigmented animal hemisphere, sometimes near the equator, but more usually near the animal pole.

The acrosome pierces the surface-layer of the egg-cytoplasm, and immediately the egg reacts in a remarkable manner. From all sides there begins to flow towards the acrosome what appears to be a watery albuminous fluid : it is hyaline, but coagulable. This becomes concentrated round the acrosome in the form of a conical plug, the base of which projects at the surface, the apex towards the interior of the ovum (Fig. 44, a). This plug is the entrance-funnel, its base bemg known as the entrancecone (' cone of attraction ' is an erroneous expression, as it is not formed prior to the contact of the sperm with the egg). The entrance-funnel enlarges and extends more and more . into the interior of the ovum, being directed usually towards the axis : it carries in with it a number of the superficial pigment granules and the spermatozoon. The latter, therefore, after moving actively up to the surface of the ovum and penetrating it with its acrosome, is passively carried in by the inflow of the entrance-funnel ; this movement is apparently due to a difference in surface tension between the entrance-funnel and the surrounding cytoplasm. The acrosome presently gets caught in the side of the entrance-funnel, but the substance of the latter, still moving on, carries the head and tail of the sperm with it. The result is that the anterior end of the head now faces outwards, while the posterior end Kes at the bottom of the funnel, where the head is bent on the tail, and the whole sperm-head has been rotated through 180°. Between the head and the tail - and therefore now at the inner end of the funnel - is the large anterior centrosome (Fig. 44, c).

Fig. 44. - Fertilization in the Axolotl.

A and B. Meridional sections of the whole egg. a, Formation of entrancefunnel (first part of sperm-path), b, Formation of sperm-sphere and aster ; o3 male pronucleus ; ? female pronucleus ; p.b the t^^^ polar bodies.

c Formation of the sperm-sphere round the middle piece (anterioi centrosome) ; narts only of the head (black) and tail are sho^vn.

X. Formation of the sperm-aster. The centrosome has disappeared ; the head besinning to be vacuolated, is separated from the tail.

T'FuSr shortening and vacuolation of the sperm-nucleus. There is still no centrosome.

F, Appearance of the definitive centrosome. g, h. Division of thn centrosome.

(In c-H the arrow marks the direction of entrance of tlie spermatozoon.) I, Approach of the two pronuclei. Formation of spindle-fibres J, i?ormation of asters, elongation of spindle, further enlargement of pronuclei, and appearance of clu'omosomes.

K, Further elongation of spindle, and formation of a ccntrosnhere

irfhe'Toindr^r""" ^he pronudear men.branes are breakSg 5ow,x ana trie spindle-hbres passing in.

L, The fully.formed fertilization spindle. In the equator are the chromoomes, now longitudinally split, and attached to large spiiidle fi'bre Tu each centrosome the centriole has divided.

The entrance-funnel soon disappears, but the pigment carried in by it remains for some time as a streak, usually known as the first part of the sperm-path (Fig. 44, b).

A clear, yolk-free area now appears round the centrosome ; this is the sperm-sphere (Fig. 44, c). Very soon radial fibres or processes of some kind begin to pass out from the sphere amongst the yolk-granules ; this is the sperm-aster (Fig. 44, d). Meanwhile the head or sperm-nucleus has become detached from the tail, and the centrosome which was between them has totally disappeared. It seems that the formation of the sperm-sphere and aster- like that of the entrance-funnel- is due to the extraction of water from the cytoplasm, in the case of the entrancefimnel by the acrosome, in the present case by the centrosome ; and that the centrosome is completely used up, in fact dissolved, in the process.

The tail of the spermatozoon will not concern us : it degenerates and vanishes. The head of course remains to become the sperm-nucleus or male pronucleus. It shortens and thickens : as it does so it becomes vacuolated. By further shortening and vacuolation it becomes transformed into an ordinary nucleus (Fig. 44, E). It Hes on the outside of the sperm-aster.

It is at this moment that the definitive centrosome makes its appearance (Fig. 44, v). On the side towards the sperm-aster the nuclear membrane breaks down, and through the aperture something comes out of the nucleus which appears, when outside, as a rounded granular body. This is the definitive centrosome. It is not preformed in the sperm-nucleus and then ejected, but, probably, is due to a precipitation of the albumins of the cytoplasm by the nucleic acid of the sperm-nucleus. But, whatever interpretation be put upon the process, the centrosome is of male origm.

The male pronucleus, preceded by its centrosome and aster, now advances to meet the female pronucleus which has aheady left its position at the animal pole and is retm-ning towards the centre of the egg. The line in which the male pronucleus is now moving is knomi as the second part of the sperm-path. This does not necessarily lie in the same straight line, nor even in the same meridional plane as the first or entrance part of the path. This depends in part on the position of the female pronucleus (Fig. 46).

The first or entrance part of the path is usually directed towards some point in the egg axis, that is, it Hes in a meridional plane of the egg. If, as also is usual, the female pronucleus hes in the axis, it is evident that the second part of the sperm-path or line of union of the two pronuclei will he in the same plane. In that case it may be in the same straight line with the first part, or, more usually, make an angle with it, smce the pomt in the axis at which the pronuclei meet is at a fairly constant distance from the animal pole, while the point of entrance o the spermatozoon in the animal hemisphere is variable. If, however while the first part of the path is in a meridional plane the female pronucleus is not in the axis, then the sperm -nucleus must turn out of its meridional plane to meet the female pronucleus at some point which is not in the axis. The converse of this is seen when the entrance-path is not m a meridional plane while the female pronucleus is in the axis ; m this case also the sperm must turn aside. Thirdly, both sperm-entrance path and female pronucleus may be out of their normal direction

Afterwords, the meridional plane which includes or is parallel to the entrance-path does not necessarily coincide with the meridional plane which includes or is parallel to the line of union of the pronuclei.

During the advance of the sperm-nucleus the centrosome divides (Fig. 44, g, h) at right angles to the direction in which the sperm-nucleus is travelKng, that is, to the second part of the sperm-path, and also to the meridional plane in which the path lies. The daughter centrosomes therefore lie in a plane parallel to the equator of the egg. Hence, when the pronuclei have met, they lie together between the daughter centrosomes, which lie in a plane parallel to the equator of the egg.

The two pronuclei are now closely apposed, but not fused, inside the sperm-sphere and aster. Next, the centrosomes send out fine fibres in all directions (Fig. 44, i, j). On the one hand these impinge upon the pronuclear membranes - ^these are the begimiing of the fertilization spindle ; on the other hand they radiate out until they pass into the radiations of the original aster inside which they He.

The pronuclei enlarge, and presently in each granules of chromatin appear and run together in rows to form chromosomes (Fig. 44, j). The number of these in each pronocleus is the same as that which entered into it at the close of maturation, namely n, the germ-number. Meanwhile the asters round each centrosome have been growing larger, the spindle-fibres longer, and the latter now break through the pronuclear membranes to meet their fellows from the opposite pole (Fig. 44, k). The membranes, achromatic network, and nuclei are now all dispersed, and the two sets of chromosomes, paternal and maternal, are placed side by side on the equator of the fertilization spindle, where they undergo longitudinal fission as in ordinary mitosis (Fig. 44, l). Hence, when the daughter chromosomes pass to the spindle poles, each daughter nucleus will receive a complete set of paternal, and a complete set of maternal chromosomes. The full somatic number, 2 n, is now restored, and with each repetition of nuclear and cell-division each cell in the body comes to possess 2 w chromosomes, one-half of which are derived from the father, one-half from the mother. With the apposition of the two sets of chromosomes in the equator of the division apparatus - asters and spindle - the act of fertilization may be said to be complete.

The whole falls into two periods. In the first the spermatozoon is carried into the egg by means of the entrance-funnel, which in turn is due to a stimulus of some kind imparted to the egg cytoplasm by the acrosome ; the acrosome is the modified centrosphere. In the second the definitive centrosome is formed from the male pronucleus and the division apparatus made between its two halves while the pronuclei meet. The mechanisms involved in both periods are therefore centrosomal.

The details of fertilization have been studied in many animals, including several Vertebrates. In Vertebrates it is a rule for the sperm to enter during the second maturation division of the ovum, as in the Axolotl {Petromyzon, Salmo, Triton, Mus), but in other cases it may enter at an earlier or later period. The tail may be left outside {Mus), but is more often taken in : it always degenerates.

The pronuclei may fuse to form a segmentation nucleus, from which 2 n chromosomes arise {Pristiurus, Salmo, Petromyzon) ; but the newt and the mouse resemble the Axolotl in the separate formation of the chromosomes in each pronucleus.

It is certain that in all cases the female centrosome disappears. Whether the definitive cleavage centrosome is identical with the centrosome seen in the spermatozoon, that. is, in the spermatid, or is, as in the Axolotl, a new formation from the sperm-nucleus, is not certainly known, but there is little doubt that it is invariably a male centrosome.

As a rule only one spermatozoon enters the egg, and the presence of more than one leads to serious derangements of development (pathological polyspermy).! in what is known as physiological polyspermy, however, two or more, sometimes a great number, normally get in, as in some Amphibia (including the Axolotl), Reptiles, Birds, and Elasmobranch fishes, in which last they are very numerous and known as 'merocytes' (Riickert). In these cases only one of the sperm-nuclei fuses with the egg-nucleus. The remainder lie about in the yolk, each develops its own centrosome and aster, and may divide (with n chromosomes) many times. Ultimately the accessory sperm-nuclei degenerate without contributing to any embryonic structure.

It remains for us to discuss the significance of fertilization.

1 As in the sea-urchin, where the several nuclei fuse and their chromosomes become irregularly distributed. Where, however as m the frog Srseveral nuclei remain apart the polyspermy need not cause abnormal ^eveToprenHM. Herlant, Arch, de Biol. xxvi. 1911) although the superfluous sperm-nuclei do take part in the edification of the embryo.

It has commonly been supposed that its essence is to be found in the union of the pronuclei of the germ-cells, both nuclei being held to be necessary for the development of a normal individual. This view is based partly on the phenomena of conjugation in certain Infusoria, but also very largely on the assumption that the nuclei of the germ-cells are the sole vehicles for the transmission of inheritable characters ; this again rests upon the fact that it is only in their nuclei that the germ-cells are alike, while in every other respect they differ, and upon the supposition that the paternal and maternal contributions to the total inheritance are equal.

Now, whatever view we may take of the parts played by nucleus and cytoplasm respectively in the handing on of the characters of the species, it is most assuredly certain that for the production of a normal individual both pronuclei are not a necessity. In the first place, there is the phenomenon of parthenogenesis, natural and artificial. In the former the ovum develops without fertilization by the sperm and without artificial assistance (as in Aphidae and some other Insects, and in certam Crustacea). In the latter the stimulus usually given by the sperm is replaced experimentally by some physical or chemical agent. Thus the ovum of a sea-urchin or Mollusc may be stimulated by treatment with hypertonic sea-water, or butyric acid or other substance, or by mechanical shock, or a lowering of the temperature ; in the case of the frog it is sufficient to pierce the egg with a fine needle. In all these instances some physical or chemical alteration (or both) IS produced in the egg, as a result of which it begins to segment and develop. The process, if care is taken, may be perfectly normal, and the individual reach the adult condition A sexually mature (male) sea-urchin has been reared in this way In all cases of parthenogenesis only the female pronucleus is

The converse is seen in what is called merogony, where the egg (of a sea-urchin, Worm, or Mollusc) is divided into two halves, only one of wlucli contains the nucleus. Both halves can be fertilized, the nucleate and the enucleate, and will develop into normal larvae. In the latter case only the male pronucleus is present.

On the other hand, a nucleus must of course be present, and actual experiment has shown that what is really necessary for normal development is the presence in the ovum, and ultimately in every cell of the body developed from it, of a complete set of the n unlike chromosomes characteristic of the species.

Hence, both male and female pronuclei are not necessary, and we must look elsewhere for the significance of fertilization.

As we know already, the germ-cells of both sexes pass through two maturation divisions, and two only, after which their capacity for reproducing themselves is lost. The first effect, or almost the first effect, of their union is that their product, the fertilized ovum, begins to segment and continues to do so. In other words, the power of reproduction by cell-division which was previously lost is in fertilization restored. It is mutually restored.

That the ovum regains the power of nuclear and cell-division is obvious : we see the maternal chromosomes undergo longitudinal fission, as they lie on the spindle, and subsequently we see the egg cytoplasm divide. In the case of the male Ave see the male chromosomes divide in ordinary fertilization as they lie alongside the female ; in the fertilization of enucleate eggfragments the stimulus imparted by the female cytoplasm to the male chromosomes is still more evident.

A study of fertilization reveals the mechanism by which this stimulation is effected. For ordinary nuclear and cell-division an apparatus is necessary, the spindle with its asters ; this apparatus is made by the centrosomes in the cytoplasm, the two centrosomes proceeding from the division of one, and its function is first to pull apart the halves of the divided chromosomes, and second, to ensure cell-division by the cell-plate or intermediate bodies developed in the equator.

The mature ovum possesses no centrosome : the mature spermatozoon possesses little cjrtoplasm, and that only in the tail. In fertilization the centrosome is either introduced by the male cell or made by it after entering the egg : the necessary cytoplasm in which this centrosome can divide and make the asters and spindle is provided by the female. The wholly different structures of the two germ-cells are therefore mutually complementary in the stimulation by which the lost power of cell-division is restored, and this is the significance of fertilization.

The experiments on artificial parthenogenesis suggest that a physico-chemical expression may be found for this stimulus.

This is not, however, its only effect. A very common, if not universal, result of the approach of the sperm is the exudation by the ovum of a peri vitelline fluid. In some cases (for instance, the sea-urchin) a membrane which prevents the entry of more spermatozoa is secreted at the same time and pushed out by the peri vitelline fluid. In the frog it remains as a thin fluid layer between the ovum and the jelly ; it is the exudation of this fluid which enables the egg previously adherent to the mucin jelly to turn over till its axis is vertical and the white pole below : this occurs shortly after insemination.

Of greater importance than this is the change in the cytoplasmic structure of the egg brought about at this time.

A few hours after insemination there appears in the frog's egg a crescentic grey patch on one side along the border of the pigmented area (Fig. 45). The grey crescent is due to the immigration of pigment from the surface into the interior, and this in turn is caused by the entrance of the spermatozoon. The grey crescent always appears on the side of the egg opposite to that on which the sperm has entered. We know that a watery fluid flows towards the sperm from the cytoplasm (the entrancefunnel flrst, and later the sperm-sphere, are due to this), and we may suppose that this streaming movement drags the pigment granules away from the surface on the opposite side, whence the grey crescent.

The grey crescent is actually opposite to- that is, in the same meridional plane as- the first or entrance part of the sperm-path (Fig. 46). Hence it does not necessarily lie in the same meridional plane as that which includes the line of union of the pronuclei.

We shall see in the next chapter that the meridional plane of the first division always includes the line of union of the pronuclei, and hence does not always coincide with the meridional plane of the grey crescent.

It is clear that, whereas the unfertilized egg was radially symmetrical about its axis, it can now be divided into similar halves by only one plane, that which includes the axis and the middle point of the grey crescent. About this plane it is bi-laterally symmetrical. The greatest interest attaches to this alteration of symmetry, since the side of the grey crescent will become the dorsal side of the embryo, the side on which the sperm

Fig. 45. - Formation of the grey crescent in the frog's egg (R. ternporaria). a, b from the side ; c, d from the vegetative pole. In A, c there is no crescent, in b, d a part of the border of the pigmented area has become grey.

entered its ventral side. Since the animal and vegetative poles mark respectively the future anterior and posterior ends (approximately), it follows that the plane of symmetry of the fertiUzed but unsegmented egg coincides with the median longitudmal or sagittal plane of the future embryo. The whole bilateral symmetry of the embryo is now predetermined in the cytoplasmic structure of the egg.

That the blastodisc has a bilateral structure in Birds and Elasmobranch fishes also seems to foUow from the fact that the cells in both these cases are larger at one end of the blastoderm than at the other. Further, this structure is definitely related to that of the embryo since the large-celled end becomes anterior.

Whether the change from the original radial to the definitive bilateral symmetry is in these cases also brought about by the spermatozoon, future researches must show.

In the Teleostei the concentration of the superficial cytoplasm (periblast) to form the blastodisc is an effect of fertihzation.

l^' ^^r" Diagrams to show the relation between the first and second parts of the sperm-paths. The paths are projected on a plane perpendicular to the axis. In a the two parts are in the same meridional plane, in B m different meridiona.1 planes. 1, First part of the sperm-path ; 2. Second part; o^, male pronucleus ; ?, female pronucleus ; ^.c, grey crescent : on the opposite side (side of entrance of the sperm) the superficial pigment Z rS; ' «,<^^ has divided in a plane perpendicular to thf a^

at right angles to the second part of the path. '

In conclusion we may attempt to estimate the parts played by the cytoplasm and the nucleus of the germ-cells in inheritance.

That some at least of all the inheritable characters of the species- and not only specific but varietal and individual characters as well- can be inherited from the father as readily as from the mother is obvious. Since the nucleus, beside the centrosome which is merely an organ of cell-division, and the acrosome which merely provides for the entrance, is the only part of the male cell which is always incorporated in the fertilized ovum for the tail may be left outside, we are obliged to regard the nucleus, that is, the chromosomes, as the vehicles by which tliese characters are transmitted.

The chromosomes of the nuclei of the germ-cells - which, as we have already pointed out, are different from one another - are in some sense the determinants of inheritance in the offspring : on their presence depends the ultimate appearance in the offspring of certain characters, and, in respect of their capacity for transmitting these characters, the two germ-cells are similar : each possesses a full set of the necessary chromosomes. In ordmary sexual reproduction the offspring receives two such sets, but one will suffice, as in parthenogenesis and merogony.

It does not, however, follow that the determinants for the whole of the inheritance are located in the nucleus.

As we have just seen, the material for the different parts of the body of the embryo is present in the cytoplasm of the fertilized but unsegmented egg ; to that structure the spermatozoon has . contributed nothing, beyond the rearrangement of material, the substitution of a bilateral for a radial symmetry. Experiment teaches us that the various parts of this structure are so many organ-forming substances, causally related to the development of certain organs, and therefore determinants of a part of the whole inheritance ; and recent researches on heterogeneous hybridization show clearly what this part is. The ovum of a sea-urchin, if the proper precautions are taken, may be fertilized by the sperm of a starfish, a feather-star (both of which of course are, like the urchin, Echinoderms), or even of a Mollusc or Worm. The result is always the same. A typical sea-urchin larva is developed. Even an enucleate egg-fragment will develop a little way when so fertilized, and exhibits the maternal characters alone. The spermatozoon employed does nothing but convey to the egg a stimulus, which sets the process in action ; its chromosomes sometimes persist, sometimes do not.

Hence the characters, the determinants of which reside in the cytoplasm, are the large characters which put the animal in its proper phylum, class and order, which make it an Echinoderm and not a Mollusc, a Sea-urchin and not a Starfish ; and these large characters are transmitted through the cytoplasm and therefore through the female alone. The smaller characters - generic, specific, varietal, individual - are equally transmissible by both germ-cells, and the determinants of these are in the chromosomes of their nuclei.

And yet the cytoplasm of the egg-cell is indebted very largely for its structure to the activity of the nucleus. As we have seen, the nucleus makes two contributions to the cytoplasm, first, the so-called ' yolk-nucleus ', the substances concerned in the deposition of the yolk, and second, the contents of the germinal vesicles dispersed when the latter breaks down at matm'ation. These processes are perhaps independent of the chromosomes. Further, they find no parallel in the male sex.

Even if, therefore, the cytoplasmic determinants are ultimately to be assigned to the nucleus, the share taken by the female in the transmission of the whole heritage is greater than the part played by the male.


W. E. Agab. The spermatogenesis of Lepidosiren paradoxa. Quart. Journ. Micr. Set. Ivii, 1911.

G. Belikens. Die Reifung und Befruchtung des Forelleneies. Aiiat. Hefte, x; 1898.

J. B. Caenoy et H. Lebeun. La v6sicule germinative et les globules polaires chez les Batraciens, La Cellule, xii, xiv, 1897, 1898.

J. B. Farmer and J. E. S. Moore. On the meiotic phase in animals and plants. Quart. Journ. Micr. Sci. xlviii, 1905.

K. Herfort. Die Reifung und Befruchtung des Eies von Pelromyzon fluviatilis. Arch. miJcr. Anat. Ivii, 1901.

J. W. Jenkinson. Observations on the maturation and fertilization of the egg of the Axolotl. Quart. Journ. Micr. Sci. xlviii, 1904.

E. KoRSCHELT u. K. Heider. Vergleichende Entwicklungsgeschichte der wirbellosen Tiere. Allg. Th., Lief. 2, Jena, 1903.

F. Meves. Ueber die Entwicklung der mannlichen Geschlechtszellen von Salamandra mactdosa. Arch. mikr. Anat. xlviii, 1896.

F. Meves. Es gibt keine parallele Conjugation der Chromosomen ! Arch. Zellforsch. i, 1908.

T. A. Montgomery. The heterotypic maturation mitosis in Amphibia and its general significance. Biol. Bull, iv, 1903.

A. Oppel. Die Befruchtung des Reptihencies. Arch. mikr. Altai, xxxix,

J. RijOKEET. Zur Bofruchtung des Selacliioroies. Ami. Am. vi, 1891

A. u. K. E, SoHEEiNER. Uio Rcif ung der maiinlichen Geschlechtszelleu von Salamandra maculosa, Spinax niger und Myxine gluHnosa. Arch, de Biol, xxii, 1906.

J. SoBOTTA. Dio Befruchtung und Furohung des Eies der Maus. Arch, mikr. Anat. xlv, 1895.

E. B. Wilson. The cell in development and inheritance. New York, 1902.

H. VON WiNrwABTER. Rechcrches sur I'ovogentee et I'organogcneso de I'ovaire des Mammif^rea. Arch, de Biol, xvii, 1901.