736 The American Naturalist. [August, 
stance the egg shell itself, but is owing to the reddish brown color 
of the albumen which invests the ovum proper. 
The diameter of the entire ovum is about one-fifth of an inch. This 
measurement includes the secondary egg-envelopes, i.e., the calcareous 
shell and the albumen. The ovum proper is quite small, measuring 
only two-thirds of an millimeter or one thirty-seventh of an inch in 
The structure and physiological relations of the shell, albumen, air- 
vesicle and ovum are complex ; quite as much so, in fact, as in the egg 
of the common fowl, from which Ampullaria is of course exceedingly 
remote. There is even a striking resemblance between the bird's &gg 
and that of Ampullaria in a number of respects. These are found in 
the common feature of an air space, a peripheral more liquid and a 
central more viscid mass of albumen, in which the ovum is embedded 
in both cases. I have been unable, however, to make out distinct 
chalazae in the eggs of Ampullaria. 
The size of the youngest segmented ova was two-thirds of a milli- 
meter, which is unusual for the egg of a gastropod and is not greatly ex- 
ceeded by the ova of most forms ; the more usual dimensions being far 
below this, though in some it is probably much greater, as in Bulimus, 
for example, judging from the size of the shell. 
The great size of the egg makes it certain that development pro- 
ceeds by epiboly, the germinal or animal pole of the egg being probably 
marked by a blastodisk at an early stage. The early stages, of course, 
were not observed by me, having been passed over long before the ova 
came into my possession. A very large yolk was present in the young- 
est stages and the yolk substance was homogeneous and not granular. 
The relations of the yolk and the development of the walls of the mid- 
gut are of considerable interest as revealed in sections of entire 
The account given by Semper of the development oi A. polita 
Derhayes, is very incomplete as respects the early stages. The entire 
egg of ^. /^//Va is much smaller than that of A. depressa, measuring 
only three millimeters instead of five as in the latter. But there evi- 
dently remains a large yolk mass, as shown by one of Semper's figures, 
in which there is also represented a hemispherical cap, the blastoderm 
probably, composed of vesicular cells. One of his figures, showing 
four blastomeres, gives one the impression at first that cleavage is total, 
but this view is irreconcilable with the next figure, and if I have under- 
stood his text properly he has recorded nothing which is in con- 
flict with the conclusion that the segmentation was partial ; since 
