742 
DR. E. B. WILSON ON THE DEVELOPMENT OF RENILLA. 
egg may be able to do this at the first division of the segmentation nucleus into two, 
or may be unable to effect it until six successive divisions of the nuclei have taken 
place, and the egg therefore divides into thirty-two spheres at the first cleavage. 
In searching for the cause of these variations in the activity of the vitellus, the idea 
at once suggests itself that it lies in the variations of the amount and distribution of 
the deutoplasm. It has been pretty clearly established by the researches of late years 
that the protoplasmic and deutoplasmic constituents of the vitellus are, in a certain sense, 
antagonistic to each other in their influence upon the rate of development. The 
protoplasm is the active part, while the deutoplasm, as such, is inert, and, until absorbed 
and converted into protoplasm, exercises a retarding influence upon development. 
The egg of Renilla is heavily laden with deutoplasm spheres, which, as we shall 
see, long remain inert, and are not converted into protoplasm until a late stage of 
development. If we suppose—and the assumption appears fully justifiable—that the 
amount and distribution of the deutoplasm in the vitellus are subject to slight 
variation, most of the variation receives a simple explanation. It is, of course, possible, 
or even probable, that the activity of the protoplasm may vary also ; but since the 
two constituents of the vitellus are, as it were, counterbalanced against each other, a 
variation in the amount or activity of the protoplasm must have the same effect as 
the opposite variation in the deutoplasm, and hence we may for the sake of simplicity 
consider the amount of deutoplasm alone. 
The researches of Flemming, Strasburger, and others have within a few years 
clearly shown that the division of the nucleus of a cell produces, or is at any rate 
closely associated with, a tendency to division in the body of the cell. If then 
the deutoplasm of an egg be scanty, this tendency may be strong enough at the first 
division of the nucleus to overcome the inertia of the mass of the vitellus and the egg 
divides into two cleavage spheres at the start. This condition is permanently retained 
in the eggs of many animals, but in Renilla occurs only as a rare variation. With an 
increasing amount of deutoplasm, equally distributed, the cleavage of the vitellus 
is longer and longer delayed, though the ineffectual efforts of the vitellus may be 
expressed in slight changes in the form of the ovum. 
Bearing these considerations in mind it is exceedingly interesting to compare the 
various modes of development of Renilla with those of other animals, and especially 
of certain forms existing among the Arthropods. 
In Lucifer, as described by Brooks (Phil. Trans., 1882), the egg is transparent and 
nearly destitute of deutoplasm. The segmentation is regular and total, the nuclei 
and bodies of the spheres divide regularly and simultaneously into two, four, eight, 
&c., as far as the segmentation can be followed, and the spheres remain perfectly 
distinct from one another. In Palcemon, described by Bobretsky (whose Russian 
paper I know only from German abstracts), the deutoplasm is abundant, but the 
segmentation is regular and total at first. Late in the development, however, the 
inner ends of the high columnar cells (“ yolk pyramids ”) fuse together to form a 
