754 
DR. E. B. WILSON ON THE DEVELOPMENT OF RENILLA. 
suggested in the case of Planorbis that they play a part in the absorptive activity of 
the cells. I have not been able to discover any such function in the nuclei. 
§ 6. Comparison with other forms. 
Upon comparing the formation of the digestive cavity in Renilla with that of other 
Anthozoa, we find, in some cases, a close agreement, but in other cases the phenomena 
are entirely different. All of the Alcyonarian forms, so far as known, excepting 
Monoxenia, agree in their general features with Renilla, developing as solid delaminate 
planulas, in which the gastric cavity is hollowed out by the disintegration and absorp¬ 
tion of a central mass of yolk-cells, and the latter are at first indistinguishable from 
the true or permanent entoderm cells. Gorgonia, according to Kowalevsky, is an 
exception to the rule ; for the embryo contains a central cavity surrounded by a layer 
of ciliated rounded cells, which are in turn enclosed in a layer of columnar true 
entoderm cells. The ciliated cells are believed by Kowalevsky to be absorbed, and are 
considered as homologous with the yolk-cells of other forms. It is noteworthy that 
Leptogorgia, though far more nearly allied to Gorgonia than to Renilla, agrees entirely 
in development with the latter, and does not have a permanent segmentation cavity. 
Among the Zoantharia, the greater number of forms agree with the Alcyonaria in 
developing as solid delaminate planulte in which the gastric cavity is formed by the 
absorption of a central yolk-mass. A few forms, on the other hand, viz.: Cerianthus 
(Kowalevsky, Jourdan), Actinia equina L. (Jourdan), and perhaps an allied Actinia, 
and probably Caryophyllium (Kowalevsky), develop as invaginate gastruke. Balfour 
states, on the authority of Kleinenrerg, that in some of the apparently delaminate 
types the segmentation is unequal, which “ probably indicates an epibolic gastrula.” 
While the occurrence of epibolic gastrulse among these forms is by no means improbable, 
it cannot be accepted on this evidence alone ; for the segmentation of Renilla shows that 
such an inference may be entirely false. 
It is a curious fact that in two at least of the invaginate forms, viz. : Actina equina 
(Jourdan), and Cerianthus (Kowalevsky), a yolk-mass is formed in the gastric 
cavity some time after the invagination has occurred, though no traces of it exist in 
earlier stages. Thus, of the former species Jourdan states: “ L’espace entre les 
cloisons est toujours occupe par une masse probablement vitelline, et qu : on croirait 
exsudee des tissus de la larve; cette masse nutritive est formee par de grosses 
vesicules semblables a des cellules adipeuses et par des noyaux fortement colons par 
les reactifs.”^ Kowalevsky regards the yolk-mass of Cerianthus as a secretion of 
the deeper layers of the entoderm, and considers its elements as fat globules. In 
both cases the yolk-mass is eventually resorbed. If the origin of the yolk-mass 
is correctly described by these eminent observers, it is clearly not homologous with 
that of the Alcyonarian forms. 
* Ann. d. Sci. Nat., 6 me serie, tome x., p. 120. 
