768 DR. KE. B. WILSON ON THE DEVELOPMENT OF RENILLA. 
entirely disproved this view. The entodermic origin of the axial cells is placed 
beyond all doubt, by the fact that typical entodermic spicules are sometimes developed 
in them. These are unmistakable in form and optical characters, and are never 
developed in ectoderm cells. 
In its earlier stages the peduncular septum forms a complete partition, extending 
from side to side, and reaching the posterior end of the body. At a later period it 
becomes perforated along its sides, and at its posterior extremity by rounded openings, 
which place the chambers of the peduncle in communication, The posterior opening 
(fig. 155, p.) becomes very large, and the lateral openings (0.) also increase in size, 
until the septum has the appearance of being suspended by narrow threads from the 
lateral walls of the peduncle (see figs. 181, 182). The lateral openings subsequently 
become much reduced in size, or even close entirely (figs. 206, 207), but the posterior 
opening remains permanently in the adult, and has been described and figured by 
KOLLIKER. 
In Leptogorgia the eight radial septa are visible when the larva ceases to swim, and 
attaches itself to the bottom (fig. 113). So far as could be determined, they develop 
simultaneously, and extend throughout the entire length of the body, without joining 
one another, or otherwise departing from a strictly radial disposition. They have, 
however, the same bilateral arrangement with respect to the cesophagus as in [enilla. 
The mouth and cesophagial cavity are distinctly elongated in a definite plane, which 
may by analogy be regarded as the dorso-ventral. Nothing like the peduncular 
septum in its fully formed condition was observed, but there is an accumulation of 
entoderm cells at the aboral end of the larva (fig. 116), developed in connexion with 
the axis, which is very similar to the peduncular septum in its earliest stages. 
Review. 
The radial septa and the peduncular septum are structures widely different from 
one another in structure and origin. The former have a simple cuticular supporting 
lamella, consist of two strata of entoderm cells, arise at the anterior extremity of the 
body and grow backwards; the latter, on the other hand, has a double supporting 
lamella, consists of three layers of entoderm cells, arises at the posterior end of the 
body and grows forwards. 
The eight radial septa are of universal occurrence among the Alcyonaria, have 
in all cases the same grouping about the cesophagus, possess an entirely similar 
musculature, and for these reasons are clearly homologous throughout the group. 
I had strong hopes that a careful study of the early development of the radial septa 
might give some indication of the relation in which they stand to the septa of other 
groups of polyps. The result is, however, a purely negative one, and affords absolutely 
no new basis for speculation upon the systematic affinities of the Alcyonaria. Their 
