732 DR. E. B. WILSON ON THE DEVELOPMENT OF RENILLA. 
sometimes observed to consist of four large spheres, capped by smaller ones, exactly as 
in the early stages of many epibolic gastrulas. In the sixteen-sphere stage there are 
often three or four larger spheres which are always placed at one pole of the egg, and 
are not separated by smaller spheres. 
The first cleavage into eight spheres may be incomplete and irregular as well as 
unequal. Thus the egg shown in figs. 45 to 48 divided at first incompletely into eight 
(fig. 45), and then passed into a somewhat marked quiescent stage (fig. 46) of fifteen 
minutes’ duration, one of the spheres retaining its prominence, as shown in the figure. 
It then divided into sixteen nearly equal spheres (figs. 47, 48), and its subsequent 
development was regular. A somewhat similar case is illustrated by figs. 49 to 58. 
In this individual the first resting stage (figs. 51, 52) was very marked, the outlines 
of the spheres became quite invisible, and the embryo could only be distinguished 
from the unsegmented egg by its slightly irregular outline. In this individual it is 
shown, further, that the spheres do not necessarily divide simultaneously, though 
this is usually the case. The sphere marked a. did not divide at the third general 
cleavage (fig. 55), but delayed until the next, or fourth, cleavage, when it divided into 
two spheres, aa. 
In these cases of slightly unequal cleavage it was in several instances observed that 
the smaller spheres sometimes increased considerably in size, after the cleavage was 
apparently complete, so as to reduce the inequality considerably. This is rendered 
possible, perhaps, by the circumstance to be afterwards described, that the earlier 
cleavages do not extend to the middle of the ovum, and the spheres are continuous 
at first with a central solid unsegmented mass. Or it may possibly be due to a 
re-arrangement of the material of the spheres, such as a change from a vertical to a 
horizontal elongation. 
3. In the third form of segmentation to be described, of which a single case only 
was observed by Dr. Wixson, the egg divided at the first visible cleavage into thirty- 
two (++) spheres, passing over both the eight-sphere and the sixteen-sphere stages. The 
segmentation was somewhat unequal and became more so in later stages owing to the 
more rapid multiplication of the spheres at one pole. The egg developed perfectly, 
however, and produced a larva which appeared to be quite normal. 
In the three forms of segmentation so far described, a certain number of individuals 
were observed to undergo considerable changes of form fifteen to twenty-five minutes 
before actual cleavage took place. The eggs became slightly irregular, with wavy 
outlines, as if about to segment; but within a few minutes they became again 
perfectly spherical, and remained so until the actual segmentation began. This was 
observed only once preliminary to the eight-sphere cleavage, and occurred in the single 
example of the thirty-two-sphere division. About one-fourth of those which divided 
at once into sixteen underwent the preliminary change. 
There can be little doubt that these preliminary changes of form are attendant upon 
