138 BULLETIN OF THE UNITED STATES FISH COMMISSION. 



the highest degree probable that it is secreted from the celhilar walls of 

 the sack or follicle in which the egg grows and is matured. 



The youngest ova of the mackerel do not appear to be inclosed in 

 independent follicles; these seem to be developed only after the egg 

 has attained some dimensions. Very young ova are found to contain 

 a relatively large nucleus or germinative vesicle inclosed in a thin layer 

 of transparent homogeneous protoplasm, and for a considerable time 

 this condition seems to be maintained, but as they increase in size it is 

 found that, while the germinative vesicle increases in dimensions, the 

 protoplasmic envelope also grows in thickness, and that there is a tend- 

 ency to multiply the number of nucleoli or germinative spots included 

 in the germinative vesicle. At a still hiter period the nucleus becomes 

 apparently granular, and finally, when the egg is mature and ready to 

 rupture the follicle in which it grew, the germinative vesicle, as well 

 as the spots, seem to have disappeared ; at any rate it is now generally 

 held that when the egg has attained maturity the germinative vesicle 

 undergoes disintegration, andiierhaps a reorganization, by which a por- 

 tion of it becomes what is known in recent years as the female pro- 

 nucleus, which conjugates or becomes fused with a similar body called 

 the male pronucleus, which results from the metamorphosis of the head 

 of the male element, or spermatozoan, after its entry into the germinal 

 matter of the egg. It is this body, made up, as it is, in part of male 

 and in part of female or ovarian protoplasm, which constitutes the 

 nucleus of the first segmentation furrow across the germinal disk in 

 which it is embedded, and which must be regarded as the initial or 

 starting j)oint in the development of the young fish. It is also doubt- 

 less a fact that during the iirocess of division of the germinal disk we 

 would find the nuclei elongated with granular rays extending from 

 their ends through the surrounding protoplasm, as well as bands or 

 fibrils of denser protoplasm running from one end to the other of the 

 nuclear figure. When the cleavage is completed the rays and bands 

 appear to be withdrawn, and each pole of the formerly elongated 

 nucleus becomes rounded independent of its fellow, and so two nuclei 

 result, the dynamic or forcfe centers of two cells, the j^roducts of this 

 process of cleavage of the original cell. This process has been named 

 l-aryoJcmesis by Prof. W. Flemming, of Kiel, in allusion to the ajiparent 

 exhibition of the modes of motion of the matter of the life centers of 

 cells. Nuclear figures of great complexity, but always bipolar during 

 cleavage, have been described by this author and others in both animal 

 and vegetable tissues. That they exist and will be found well devel- 

 oped in the very early stages of the cleavage of the germinal disk of the 

 embryo fish, I have not the slightest doubt ; it will depend upon the 

 method of demonstration as to whether or not they will be made visible. 

 My reason for this statement is the fact that the early stages of the 

 segmentation of the germinal disk of all the species of fish ova which I 

 have observed closely enough are essentially rhythmical ; that is to say^ 



