BULLETIN OF THE UNITED STATES FISH COMMISSION. 
466 
10. The yolk is a waning structure aud can scarcely be taken into consideration 
in accounting for the growth of early stages. 
17. During the whole of gestation respiration is carried on by the osmotic action 
between the general surface aud the closely applied ovarian structures. When the 
alimentary tract is opened a current is kept flowing through it and aeration is, in all 
probability, effected by the alimentary tract. In later stages the fins become highly 
vascular aud doubtless serve both for purposes of aeration aud food absorption. 
18. There is present in the eutodermic pole of the developing egg a body the like 
of which has not beeu observed in any other egg. It consists of a mass of protoplasm 
imbedded in the yolk. It is dissolved near the time of the closing of the blastopore. 
Mr. J. W. Hubbard, one of my students, has connected its history with that of the 
yolk nucleus, which is a conspicuous structure in the ovaries of adult fishes in eggs 
from 20 /< up to maturity. It is a general extrusion from the nucleus of the young 
ovum, and probably represents the histogeuetic or somatic portion of the nucleus, and 
this in part at least corresponds to the macronucleus of ciliate infusoria. 
19. Before segmentation begins the whole of the germ is separated from the deuto- 
plasm. The first cleavage plane extends entirely through the germ to the yolk before 
the second cleavage begins. 
20. A segmentation cavity is not formed during segmentation, but appears later 
by a separation of the ectoderm and entoderm. 
21. The third cleavage plane is not parallel with the first, as is usual in fishes, but 
is semiequatorial. This has nothing to do with the horizontal cleavage claimed to 
have been seen by Hoffmann and by Brook. It is taken to be a pseudoreversion to 
primitive methods of segmentation, with the reservation that this condition is not 
perfectly homologous with the third segmentation of the frog or Branchiostoma and 
would not be, had the yolk entirely disappeared. 
22. The periblast is formed from a few of the marginal cells. Like the yolk it is a 
waning structure. Only about 12 cells are ever formed. They take no part whatever 
in the formation of the embryo. All of them persist as long as a trace of the yolk is 
left. It, with the final part of the yolk, is absorbed by the blood of the sinus venosus. 
The liver has nothing to do with its final absorption, as Wilson has claimed, but simply 
mechanically incloses the nuclei above and behind. 
23. During an early stage of segmentation some of the marginal cells of the blas- 
toderm creep over the yolk till they nearly, if not entirely, cover it. 
24. Before gastrulation the yolk sinks into the mass of the blastoderm, the cells of 
which rearrange themselves about it and nearly inclose it. 
25. The gastrula is finally formed by a process of delamination of entoderm from 
ectoderm and is completely diplastic and symmetrical, the blastopore closing at the 
entodermic pole of the e^g. 
26. Before any other organs become evident the sex cells become conspicuous. 
Their fate I have discussed elsewhere. 
27. The earliest stages of the formation of the embryo have not been clearly 
made out with the material at hand. It is, however, certain that in one of the figures 
published by me in the “Journal of Morphology,” I mistook the tail for the head. 
The conditions are extremely similar to those found in the mammalian embryos, except 
that the central cavity is filled with yolk instead of fluid. 
