DR. E. B. WILSON ON THE DEVELOPMENT OF RENILLA. 745 
development in different parts of the egg, will not, however, account for some of the 
forms of unequal segmentation in Renilla. When, for instance, the egg divides into 
four larger and four smaller spheres, the former do not contain a greater number of 
nuclei than the latter, since at the following cleavage all are divided alike into two 
parts, and further, we have seen that the inequality existing at first may be 
considerably reduced without the occurrence of any visible cleavage. It is improbable 
that the cause is simply a lack of precision in the action of the vitellus, since the 
arrangement of the spheres is constant, so far as observed, the larger spheres being at 
one pole of the egg and the smaller spheres at the other. The resemblance of the egg 
at this stage to an epibolic gastrula has already been noted, and the idea naturally 
suggests itself that this resemblance may be due, not to accident, but to actual 
reversion of the gastrula form, though the essential features of the development are 
entirely different from those of the gastrula. There are a number of facts which 
indicate the derivation of the delaminate planula from an epibolic gastrula like that of 
Euawes; and if the planula has had such an origin, it is not improbable that it might 
occasionally revert to the original unequal form of segmentation. 
§ 4. Changes of external form and further histological differentiation. 
At the close of segmentation the embryo is roughly spherical in form, varying 
considerably in outline. As development proceeds the body elongates slightly so that 
a longer axis (antero-posterior) can usually be made out, but the larvee both of Renilla 
and Leptogorgia assume the most irregular and strange forms (figs. 100%, 100’, 100°, 
107). Occasionally a larva develops very regularly, preserving a nearly even oval out- 
line until the cilia make their appearance. But in far the greater number large irre- 
gular prominences and depressions make their appearance over the whole surface of the 
embryo, and the form becomes so strangely modified that it is difficult to believe the 
shrunken and distorted larvee capable of further development. In fact I unhesitatingly 
considered them at first as abnormal or dying specimens. No two of them have the 
same form, and they sometimes appear almost like huge Amabe with short rounded 
pseudopodia extended in various directions. Nevertheless the larvz are perfectly 
normal, as I repeatedly proved by isolating them in small vessels and following then 
development. A regular oval form is once more gradually assumed (fig. 101), and 
most of the larve of twenty-four hours show no trace of the strange changes of form 
through which they have passed. The various prominences and processes are not 
capable of active movement, and the change of form is exceedingly slow. I am 
entirely unable to say what the significance of this curious change of form may be, 
and can hardly find a parallel to it in the development of other animals. 
The rate of development varies exceedingly in differeut individuals, being sometimes 
twice as rapid as in other cases. In nearly all instances, however, the embryo acquires 
a dense and uniform covering of cilia when about twenty-four hours old, the body 
