560 REPORT OF COMMISSIONER OF FISH AND FISHERIES. [106] 



the latter have subsided, tends to compel them to maintain their irregu- 

 lar forms and thus afford the starting points for still further irregulari- 

 ties of cleavage. 



After a while also the series of successive segmentations are no 

 longer synchronous— that is, after the third or fourth series of cleavage 

 furrows have developed, it will be found that if sections of germinal 

 disks are examined there will be evidence of division in some cells and 

 not in others. In some cells caryokinetic or nuclear firgues are present ; 

 in others they are wauting. Not only is this true, but where the disk 

 contains upwards of 1,000 cells the nuclear figures of different cells are 

 also found in various stages of metamorphosis, while a much larger 

 number are quiescent or resting. This tendency to heterochronous 

 division of the nuclei of- the germ is another very striking illustration 

 of the law of acceleration and retardation of development. 



The acceleration and retardation of the metamorphosis of the nuclei 

 again probably depends upon the nutritive processes and metabolism 

 occurring reciprocally between the component cells of the disk, and by 

 way of intercellular spaces as well as the segmentation cavity. It is cer- 

 tain that respiratory processes go on during segmentation, and it may not 

 be impossible for the segmentation cavitj^ with its thin roof to be partially 

 respiratory in the ovum of osseous fishes. It is also quite certain that 

 Newport and Ransom were justified in regarding the water space around 

 the vitellus and germ or the cavity between the ovum proper and the 

 egg-membrane as resi)iratory in function. One of these iuithorities, I 

 do not recall which, first named this cavity the breathing chamber. It is 

 developed, as already described, at the time of fertilization. There can 

 also be little doubt that respiration goes on in young fishes, which are 

 without a circulation at thetime of hatching. If, as we have seen, there is 

 evidence of the existence of respiratory processes in embryos long before 

 any spontaneous movement of the body is visible, it is fair to infer that 

 such a process can and probably does influence the rate of segmentation. 



The final proof, however, that respiration occurs in fish embryos is 

 that it is positively necessary in the construction of hatching ai)paratus 

 to have it so arranged as to constantly change the water upon the eggs. 

 This is done simply to supply the developing embryos with fresh, oxy- 

 genated Avater. Still another proof that respiration must occur in fish 

 embryos before a circulation is developed is the fact that young shad 

 move by contorting the tail some time before they leave the egg-mem- 

 brane. The muscular contractions of the lateral muscular plates so 

 manifested must be accompanied by the evolution of carbonic acid, 

 which must be carried off" and replaced by oxygen. This can mani- 

 festly not be accomplished more directly than by way of the water in 

 the so-called breathing chambt'r surrounding the embryo fish. 



It will be noticed upon comparing the outline of the disk repre- 

 sented by Fig. 10 with Fig. 33, that the latter is more nearly circular 

 than the former. The latter has passed farther into the restiug stage, 



