DEVELOPMENT OF CYPRINODONTS. 907 
lies is brought into direct communication with the general ovarian space, which, sin- 
gularly enough appears to be occluded from without by a temporary closure or plug- 
ging up of the oviduct or canal from the posterior end of the ovarian sac, a state of 
affairs, which, if it can be confirmed, approximates, or, to some extent, resembles, the 
condition found to obtain in a pregnant mammal, where the uterine os or mouth is tem- 
porarily occluded during gestation. 
‘‘We found ourselves unable to determine the species of the form, the structure of 
which is described above; none of those described in Jordan’s Manual appear to agree 
with our species (Gambusia patruelis, B. & G—D.S.J.). It may be, as some of us have 
surmised, that the isolation of the form on the eastern peninsula of Virginia for a great 
length of time may have served to develop specific characters, and that it ig undescribed. 
We leave the determination of the species to the systematic ichthyologists. 
“Thus far our account has dealt only with the structure of the adults and the pecu- 
liar contrivances by means of which reproduction is «ffected ; we will now take up the 
- discussion of the egg and embryo. 
‘‘The globular vitellus measures about a line in diameter including the embryonic or 
germinal portion. The germinal protoplasm probably occupies a peripheral position 
covering the nutritive or vitelline portion of the egg as a continuous envelope with 
strands of germinal matter running from if through and among the corpuscles of the 
vitellus. This peripheral germinal layer, when the*egg is ready to be fertilized, mi- 
grates toward one pole and assumes a biscuit shape, This is essentially the history of 
the formation of the germinal disk of the Teleostean egg as worked out independently by 
Professor Kupffer and the writer. Little of a trustwortby character is known of the 
history of the germative vesicle and epot, which bear the same relation to the egg as 
the nucleus and nucleolus do to the substance of the cell of the ordinary type. When 
cleavage of the germinal disk has begin, it is the first positive evidence that impregnation 
has been successful. The disk then begins to spread over the vitellus or yelk and soon 
acquires the form of a watch glass with its concave side lying next the surface of the 
yelk. Coincident with the lateral expansion of the germinal disk, a thickening appears 
at one point in its margin which is the first sign of the appearance of the embryo fish. 
With its still further expansion, the embryo is developed more from the margin of the 
disk toward its centre ; in this way it happens that the axis of the embryo lies in one of 
the radii of the disk; its head toward the centre, its tail at the margin. 
‘¢ But before the embryo is fairly formed, a space appears under the disk limited by 
the thickened rim of the latter, and the embryo at one side. This space, the segmenta- 
tion cavity, is filled with fluid and grows with the growth of the germinal disk, as the latter 
becomes converted into the blastoderm, «nd does not disappear until sometime after the embryo 
has left the egg as a young fish ; and then it often remains as a space around the yelk sac 
for as long as a vestige of the latter remains, as may be seen in the young of Cybium, 
Parephippus, Gadus, Elecate, and Syngnathus. In regard to this point, I hold views en- 
_ tirely different from any other observers, but inasmuch as the writer has had opportu- 
nities for the study of the development of a greater number of species representing a 
greater number of families than any previous investigator, and because the observa- 
tions are based on material studied without the use of hardening re-agents which either 
deform or obliterate the segmentation cavity, and also because it was found to be present 
in all of the forms which were sufficiently we!l studied, it is believed that it will be found 
in the developing ova of most or all Teleostean fishes. Should this prove to be the fact, the 
Teleostean egg willbe asdistinectly defined in respect to the sum of the developmental char- 
