762 DR. E. B. WILSON ON THE DEVELOPMENT OF RENILLA. 
embryological history, the cesophagus is apparently a true stomodzum, comparable 
to that of the higher Metazoa, The general occurrence of this structure—which, 
so far as the evidence at command shows, is homologous throughout all the groups 
in which it is found—is a very striking fact which probably has an important 
phylogenetic significance. Its universal occurrence among the Anthozoa and complete 
absence from the Hydrozoa is a strong argument in favour of the more primitive 
nature of the latter group. From the fact that the Anthozoa are the most primitive 
group in which the stomodzum appears, it might be concluded that this group 
represents the stock from which the higher Metazoa have descended. It seems, 
however, much more probable that the line of descent has been through some primitive 
Turbellarian form which, in common with the polyps, derived the stomodzeum from a 
still earlier group. What this origin of the stomodzum was is still an unsolved 
problem. The hypothesis that the stomodzeum is to be regarded as the introverted 
manubrium of a Hydrozoan, though a plausible one, has no embryological facts in its 
favour, and can hardly be accepted without additional evidence. 
§ 9. Development of the septa. 
The septa make their appearance at about the same time with the stomodzum, and 
are well developed within a few hours. As we shall see below, the eight radial septa 
of the anterior part of the body, which are characteristic of all Aleyonaria, differ 
entirely in structure and mode of origin from the peduncular septum, a structure 
which is found in the Pennatulacea alone, Hence it will be convenient to describe 
separately the development of the two forms of septa. 
a. Formation of the radial septa. 
Although the peduncular septum makes its appearance some time before the radial 
septa, it is preferable to describe the development of the latter first, They make their 
appearance simultaneously at the oral extremity of the larva at the time when the 
stomodzeal invagination takes place, and gradually extend thence backwards about to 
the middle of the body. Although I have made many sections through the septa at 
the time of their first appearance, and have given special attention to the matter, I 
have not been able to discover any difference in the time of theirappearance. In later 
stages, as described further on, they are of different lengths, and the differences are 
perfectly constant. This is, however, the only indication of a regular succession in . 
the development of the septa, and in the earlier stages no difference can be observed. 
The septa appear upon longitudinal section (fig. 136, fifty-two hours) as thick plates 
of entoderm cells (s.s.) extending downwards from the oral end and ending by free edges 
below. Inwardiy they are continuons with the entoderm covering the stomodzum ; 
outwardly they join the entoderm of the body-wall. In transverse section (fig. 142) 
they are seen to radiate at nearly equal intervals from the stomodzeum. The centre 
