The pachytene stage is characterized by a great contraction of the chromosomes, which leads to their being gathered up in a charac- teristic " bouquet " at one side of the nucleus of course that side adjacent to the centrosome. Then the nuclear wall is dissolved and the chromosomes separate as if repelled from one another: this process is known as " diakinesis." During this process the line dividing the sister chromosomes which paired may reappear, and according to Agar in Lepidosiren these chromosomes may entirely separate from one another, only to pair again at a slightly later stage in mitosis. In most cases, however, this complete separation does not take place, but only a partial separation, which leads to the compound chromosome assuming the form of a ring or loop. Each half of the ring corresponds to one of the two original chromo- somes, and these halves are dragged apart in the ensuing mitotic division. Most frequently each half shows a constriction in the middle of its length, which was formerly interpreted as a precocious appearance of the longitudinal division of the chromosome which is consummated in the second maturation division. For this reason the name " tetrad " has been bestowed on the shortened, thickened and partially split chromosome. Agar 1 has shown, however, that in the vast majority of cases this constriction has no relation what- ever to the splitting of the chromosome in the second maturation division.
Modern theories of heredity assume that a chromosome consists of a linear series of rudiments, each of which has its particular part to play in the up-building of the embryo. A side-by-side pairing enables us to see how corresponding rudiments belonging to maternal and paternal chromosomes are brought together: an end-to-end pairing would of course render such a process impossible.
But the side-by-side union of homologous chromosomes does not always take place in a straight line. In the germ cells of the newt Balrachoseps Janssens 2 has shown that one filament becomes spirally wrapped round the other. He believes that he has demon- strated that each filament likewise becomes split longitudinally and that when the two chromosomes separate the now separated chro- mosomes are no longer the same as those which became united with one another but each has appropriated one strand of the other. Janssens' theory in its extreme form is not accepted by other cytolo- gists; but he certainly has demonstrated cross connexions between the pairing chromosomes, and if the chromosomes are the actual bearers of the hereditary qualities, as seems to be proved from the fact that they alone constitute the head of the spermatozoon, then there are a good many facts (see GENETICS) which seem to require for their explanation an interchange of substance between the two paired chromosomes. This is termed by Morgan the " cross-over."
In the ripening of the egg very peculiar phenomena occur which have only recently received an explanation. The unripe female germ cells or oogonia show nothing peculiar in their mitosis, but during the prophases of the first ripening division an enormous increase in size of the egg cell takes place. The leptotene threads are at first clearly visible and can be seen to pass into the pachytene stage, but then they fade from view. The nucleus becomes very large and gorged with nuclear sap, from which circumstance is derived the name, " germinal vesicle," which the' older authors bestowed upon it. The nucleolus becomes large and conspicuous. At the close of the growth period the nucleolus has been completely dissolved ; the nuclear wall disappears and the nuclear sap mingles with the cyto- plasm: then the chromosomes can again be detected as minute tetrads which begin to arrange themselves on the mitotic spindle of the first ripening division. Now by the examination of specially favourable cases it has been shown that what happens during this episode of growth is that the chromosomes swell up, become pressed against the nuclear wall and almost lose their capacity for absorbing stain. It seems to be clear that a chromosome consists of at least two substances a framework which does not stain and an embedded material which stains intensely, and to which alone the name chromatin is, properly speaking, applicable, and that during the growth of the egg cell the framework swells up enormously. De Baehr 3 has shown that in the male germ cells of the annelid Sacco- cirrus, a similar growth period exists, though it is of very much shorter duration than the corresponding period in the life of the female germ cell, but, short though it is, it is long enough to cause the chromosomes to swell up and temporarily fade from view.
The nucleolus or "germinal-spot" of the older authors, which is so conspicuous a feature in the unripe egg, has formed the subject of some most interesting researches. Hogben has shown that in the cockroach Periplaneta the nucleolus becomes vacuolated and that portions of it are extruded and that these can be recognized by their peculiar staining properties, scattered in the nuclear sap and even in the act of passing through the nuclear membrane. They can also be detected in the cytoplasm outside the nucleus. There seems to be no reasonable doubt that it is by this process of vacuolation and emission of pieces of itself that the nucleolus ultimately disappears.
1 W. Agar, " Transverse Segmentation and Internal Differentia- tion of Chromosomes," Quart. Jour. Micr. Sc., vol. Ixxxviii. (1912).
2 F. A. Janssens, " La Theorie de la Chiasmotyos," La Cellule, vol. xxv. (1909).
'V. B. de Baehr, " La SpermatogenSse et 1'Ovogenese chez le Saccocirrus Major," La Cellule, vol. xxx. (1920).
Gatenby 4 and Hogben 6 have shown that in certain cases these emitted fragments may assume the appearance of nuclei, since they seem to secrete round themselves both nuclear sap and a nuclear wall. Ultimately they all disappear. Though it has not been pos- sible to connect them directly with the formation of yolk spheres, yet it is an interesting fact that the beginning of this emission from the nucleolus coincides with the first appearance of yolk spheres, and as we know by experimental evidence that the nucleus presides over assimilation, it is a reasonable hypothesis that the absorption of these pieces of nucleolus in the cytoplasm leads directly to the synthesis of yolk. It has recently been asserted by Carleton 6 that in ordinary tissue cells where the same disappearance of the nucleolus occurs as the cell grows, this is not complete that a small kernel remains which can be stained in certain silver salts and that this kernel takes its place on the mitotic spindle at the next division of the cell. This " nucleolinus," as Carleton terms it, becomes equally divided into two and the halves pass into the two daughter cells. If these observations should be confirmed we should have in the nucleolinus a part of the nucleus as permanent as are any of the chromosomes, the function of which was to form a centre for the synthesis of a mass of chromatin which constitutes the nucleolus and is destined to be emitted into the cytoplasm, where it no doubt profoundly affects the metabolism and determines the formation of cytoplasmic structures.
Parthenogenesis. We have seen that the normal history of the egg cell is to undergo two ripening divisions, at the first of which the chromosomes are reduced in number by one half. When the egg is fertilized by the spermatozoon not only is the full number of chro- mosomes restored by the addition of those brought in by the sper- matozoon but the division of the egg is initiated by the centrosome which is carried into the cytoplasm of the egg along with the head of the spermatozoon. An interesting question now comes as to what happens in the case of those eggs which develop without fertiliza- tion or, as it is termed, "parthenogenetically."
Now parthenogenesis may be either artificially induced or it may be a natural event in the history of the species. If we take the case of " naturally " parthenogenetic eggs first, we find that a great deal of light has been thrown on the subject by the investigations of de Baehr. 7 He took for his subject the plant-louse Aphis palmae, the eggs of which develop without the aid of the male throughout the summer. He shows that in these eggs the preparations for the reducing division occur. Out of the apparently irregular chromatin network leptotene threads differentiate themselves. These pair so as to form thicker pachytene threads but then at diakinesis these pairs become completely dissociated from one another, and the full number of chromosomes is thus established. Then the period of growth supervenes and the chromosomes become indistinct, but when they reappear in the metaphase they are in the full number and only one maturation division takes place at which all the chromosomes are longitudinally cleft. From these facts de Baehr draws the conclusion that the reducing division is suppressed and only the second maturation division takes place.
A somewhat different case is presented by the egg of the bee. The egg if fertilized gives rise to a female but if unfertilized grows into a male. In the latter case of course the resulting animal has in all its nuclei only the reduced number of chromosomes. When the male produces germ cells, the reducing division is suppressed. The nucleus of the spermatocyte enters on the prophases of mitosis and the cell divides, but one of the daughter cells is devoid of a nucleus and dies. The nucleus in the other cell goes back into the resting stage; and then like the egg in Aphis, it enters on a single maturation division in which the chromosomes are divided longitudinally, and the spermatozoon has therefore the same number of chromosomes as that possessed by the nuclei of the tissue cells of the male, which is the reduced number as compared to the number in the nuclei of the cells of the fertilized female.
In still other cases, as in the eggs of the small crustacean Artemia, the two ripening divisions may occur, but the first one can give rise to a nucleus which is not extruded as a polar body but remains in the egg and, reuniting with its sister nucleus, restores the full number of chromosomes.
Parthenogenesis can, however, be brought about in eggs which normally require fertilization by the application of external stimuli. 8 This stimulus in the case of the frog's egg may take the form of a prick with a needle. Under these circumstances an immense de- velopment of astral fibres takes place, centring on a particle lying
- J. B. Gatenby, "The Cytoplasmic Inclusion of Germ Cells:
VI. On the Origin and Probable Constitution of the Germ Cell Determinant, etc.," Quart. Jour. Micr. Sc., vol. Ixiv. (1920).
6 L. T. Hogben, "Studies on Synapsis: III. The Nuclear Or- ganization of the Germ Cells in Libellula Depressa," Proc. Roy. Soc., Series B, vol. xcii. (1921).
6 H. M. Carleton, " Observations on an Intronucleolar Body in Columnar (Male) Epithelial Cells of the Intestine," Quart. Jour. Micr. Sc., vol. Ixiv. (1920).
7 V. B. de Baehr, " Recherches sur la Maturation des CEufs parthenogenetiques dans 1'Aphis," La Cellule, vol. Ixxx. (1920).
8 For a full account of recent work see A. Brachet, " L'CEuf et le Facteurs de 1'Ontogenese," Encyclopedic Scientifique (Paris 1916).
