502 EMBRYOLOGY OF THE TURTLE. Part III. 
eggs from the embrace of the ovary to the last third of its own channel, there to 
be endowed with an albuminous and calcareous covering, and withal to assume the 
shape peculiar to each species. In reference to the shape of the eggs of vari- 
ous genera, it is important to mention that they vary greatly in form, and _ that 
their outline does not answer to the prevalent theory that their passage through 
the narrow channel of the oviduct gives them their form, since we have those 
which are perfectly spherical, and yet sustain as great a lateral pressure from the 
embracing walls of the shell-forming conduit as those which are more or less 
oval. We need therefore adduce nothing more against this mechanical theory 
beyond the statement of such an obviously conflicting fact as the one just men- 
tioned. We would, however, refer to the plastic power which gives to the 
embryo its typical form while it floats in the midst of a uniformly pressing 
fluid, in order to answer the question as to what renders some eggs almost 
cylindrical, others oval, and those of certain species more or less curved, approach- 
ing even to a kidney shape, whilst others are broadly oval, and finally, some 
perfectly spherical. 
Since no eggs were found in the oviduct before the shell membrane had 
already been deposited, at least partially, it might be presumed as a matter of 
course that the albumen also had already taken its place around the yolk. 
This supposition is negatived, however, by the occurrence of eggs, observed espe- 
cially in one well marked instance in Glyptemys insculpta, which formed a series 
of five in one oviduct, situated at the extreme posterior end of that organ, and 
close to one another, presenting just as many different grades of albumen and 
shell-lining deposit. The albumen was thicker and the shell lming more opaque 
for each successively more posterior egg, showing at a glance that not only the 
albumen, but the shell lining, was depositing at one and the same poimt of the 
oviduct; and moreover that the albumen, in order to reach its destined position, 
must filtrate through the meshes of the fibrous shell limmg. There is no disput- 
ing this fact, which readily proves the normality of another single case which 
we have noticed, showing a still greater disparity between the amount of albumen, 
and the shell lining by which it was covered. The egg in question was found 
in company with another, in the right oviduct of the same animal; it was covered 
by a shell lining as thick and opaque as the most posterior of the left oviduct, 
but the albumen was not half deposited (Pl. 9b, fig. 4b); the more tenacious and 
denser portion, (a) which clings so closely to the yolk sac (y) when a young egg 
is broken open and the more external and nearly fluid portion drops away, was 
all that presented itself Now under such circumstances, in order to allow the 
albumen to attain its destined bulk, the very elastic shell lmmg must stretch to 
a great extent; moreover the former probably solidifies as rapidly as it infil- 
