766 
DR. E. B. WILSON ON THE DEVELOPMENT OF RENILLA. 
rudiment of the peduncular septum (p.s.) appears as a rounded mass of entoderm 
cells, at the base of which is a delicate supporting lamella running inwards from the 
lamella of the body-wall, and becoming insensibly lost among the cells. The anterior 
part of the stomach still contains a considerable quantity of unabsorbed yolk, and 
the stomodreum is just beginning to be formed. From this point the septum grows 
rapidly forwards, ending by a free edge in front. As the septum extends forwards its 
lateral portions grow more rapidly than the middle, so that the free edge becomes 
deeply concave in front. I have no figures of this in early stages where it is most 
pronounced ; it is shown in fig. 136 at e, where it has extended very far forwards. 
By reason of this structure of the septum, the posterior part of the body is com¬ 
pletely divided into a dorsal and ventral chamber (fig. 154) ; while farther forwards, 
in front of the edge of the septum, the gastric cavity is undivided, and a section 
shows only the lateral forward extensions of the septum (fig. 154, a ., a.). These have 
exactly the appearance of two independent septa, situated on opposite sides of the 
body. If they be traced forwards they are found to be continuous with the dorsal 
septa at their point of union with the dorso-lateral pair, as explained at p. 765. 
The free edge gradually extends forwards until it reaches the point at which the 
lateral portions join the radial septa, and then remains stationary for a long period. 
Its subsequent development is described at p. 795. 
A transverse section through the peduncular septum behind its free edge (fig. 151) 
shows that it is composed mainly of two thick layers of clear, rather ill-defined 
entoderm cells, separated by a peculiar membrane (ax.). At the sides this membrane 
appears like the ordinary lamella of the septa, and joins the lamella of the body-wall. 
Towards the middle, however, the membrane splits into two layers enclosing a narrow 
space in which appear numbers of conspicuous nuclei similar to those of the entoderm 
cells. Cell-outlines can only faintly be distinguished, but there can be no doubt that 
the nuclei belong to cells which are enclosed in the lamella and may conveniently be 
termed the axial cells. These cells are confined to the central portion of the septum 
behind the free edge. The forward extension of the lateral parts of the septum show 
no trace of anything like the axial cells in their lamella. As development proceeds, 
the axial cells become more and more flattened between the enclosing layers of lamella 
and at length nearly or quite disappear (fig. 173, p.s.). The lamella of the peduncular 
septum has then the same appearance as that of the radial septa in which no axial 
cells were ever observed. 
In order to ascertain the origin of the axial cells it is necessary to study sections 
of still earlier stages of development. Fig. 153 represents a longitudinal section of a 
somewhat younger larva (forty-eight hours). The entoderm cells forming the main mass 
of the septum are here very distinct and of a high columnar form. The axial cells are 
larger and their outlines are more distinct. Their appearance is more clearly shown m 
fig. 136, from another specimen. The lamella is simple behind but splits further 
forwards into two delicate membranes between which lie the axial cells. The latter 
