EARLY DEVELOPMENT OF CRIBRELLA OCULATA. 409 



A study of a selected asterid, crinoid, ophiurid, echinid, and holothurian, not to 

 mention the comparison of Cribretta with Asterina, convinces us that it is impossible to 

 base homologies upon the method of origin of the several ccelomic elements. Hence 

 our only hope in phyletic conclusions lies in the homology of adult ccelomic cavities. Thus 

 we assume, until the contrary is proved, that the hydroccele is homologous in all cases, 

 and for similar reasons we strive to prove the homology of the epigastric with the 

 aboral ccelom and the other cavities in a similar manner. 



M acBride (loc. cit. ) has already advocated the view that asterids are to be regarded 

 as likely to show primitive characters in their development, and he relies on such 

 characters as the ectodermal nervous system, the retention of the pre-oral lobe and the 

 period of fixation, summing up with the remark, " It is, however, difficult for me to 

 see how anyone can doubt that the asterids are the least modified group of the 

 Echinoderms " (p. 398). My own belief to this effect led me originally to undertake the 

 task here published, and I believe myself that we have in stage D of Cribrella the nearest 

 approximation to the bilateral ancestor of the Echinoderms yet described. We might 

 have added to MacBride's dictum the fact that in asterids the hydroccele is developed 

 normally on the left side and remains there, whereas in other groups the hydroccele 

 moves on to the ventral surface. As Bury has incidentally shown, the proof for the 

 former existence of a true bilateral ancestor comparable to animals outside the 

 Echinodermata rests upon the assumption that asterids are the least modified of all 

 the division, a conclusion to which he was logically led by starting with his 

 pentactula-like prototype. 



If, then, we grant that asterids are the most primitive, especially in their development, 

 and that Cribrella is the most primitive, in respect of the arrangement of the ccelomic 

 cavities, that has yet been described, we can proceed to the inferences in regard to 

 the other orders. Let us speak of the pre-oral ccelom (or protoccele) as 1, # the left 

 and right lateral (or mesocceles) are 2 and 2' respectively, and the posterior ccelom (or 

 opisthoccele (paired) ) is 3. 



Ophiuroidea. 



Grave t has, as already indicated, published a paper on an ophiurid, Ophiura 

 brevispina, which has a demersal larva not unlike that of Cribrella, and also having a 

 remarkable resemblance to the larva of Antedon. We have already alluded to the 

 resemblance in its earlier stages to the former. In following out the fate of the body- 

 cavities, Grave finds, apparently beyond dispute, that the ccelom arises as a large 

 anterior ccelom and a posterior ccelom. 



* On the nomenclature usually adopted, 1 = anterior ccelom, 



2 = hydroccele, 



2' = right hydroccele, 

 3, 3' = left and right enterocceles. 

 t Grave, Caswell, Johns Hopkins Univ. Memoirs, iv. 



