790 DR. E. B. WILSON ON THE DEVELOPMENT OF RENILLA. 
compartments between the mouth and the calyx-teeth. They are at first simple, but 
soon become pinnate, the pinnz developing somewhat irregularly in pairs at the bases 
of the tentacles. 
c. Development of the mesenterial filaments. 
The mesenterial filaments are formed in essentially the same manner as in the 
primary polyp, appearing as thickenings on the edges of the septa. They differ 
however in one striking feature from the filaments of the primary polyp, viz.: in the 
order of their appearance. The dorsal filaments in the latter are last to appear aud 
slowest in development, whereas in the buds they are in many cases first to appear, 
and in all cases develop at first more rapidly than the other six. This agrees 
entirely with KoriiKer’s observations on the development of the secondary polyps 
in Halisceptrum (‘ Pennatuliden,’ p. 161) in which the dorsal filaments are well formed 
before a trace of the other six can be made out. 
This remarkable contrast between the development of the filaments of the primary 
and secondary polyps shows clearly that in searching for the relations between the 
various groups of polyps, as indicated by their embryology, we are not justified in 
comparing the egg-development of one form with the bud-development of another, or 
in taking the structure of the bud as any necessary indication of the succession of 
the parts in the egg-embryo. ‘That the importance of this principle has been 
unconsciously disregarded will, I think, be clear from the following citations, 
MosELEY writes in his admirable paper on Heliopora and Sarcophyton (Phil. Trans. 
Vol. 166, 1876, p. 121): ‘It seems extremely difficult to reconcile the extraordinary 
succession of the mesenteries in the development of the Zoantharians, discovered by 
Lacaze-Duraiers, with the facts presented by Alcyonarians. Did the development 
of the eight mesenteries of Alcyonaria correspond with that of the first eight 
mesenteries formed in Actiniade, the first mesenteries formed would be either the 
lateral dorsal or lateral ventral; but these are those which are most rudimentary in 
the zooids of Sarcophyton. Moreover the mesenterial filaments of the two lateral 
pairs of septa are in the development of Actiniadz the first to appear, and not the 
dorsal, which are longest in the Alcyonarian polyps and most persistent in the 
zooids. Apparently, however, development in Alcyonarians follows a different course.” 
These words seem clearly to imply that the greater length of the dorsal filaments 
and their persistence in the zooids indicates their earlier development in the embryo. 
K6LLIKER states explicitly (‘ Pennatuliden,’ p. 427): “Die Septa und Septula sind 
Falten des Entoderm und entstehen wahrscheinlich alle zugleich (ich), dagegen bilden 
sich in erster Linie nur an zweien derselben Verdickungen des Entoderma (Mesenterial- 
filamente) und spiiter erst treten solche auch an den andern 6 Septa gleichzeitig auf.” 
At p. 434, he even extends this statement so as to apply to the entire group of 
Aleyonaria: “Bei den Alcyonarien treten nun allerdings auch zwei Mesenterial- 
