EARLY DEVELOPMENT OF C-RIBRELLA OCULATA. 411 



The second method of reconciliation is to suppose that the anterior and posterior 

 coeloms are homologous in the two forms, and that the asterids and the crinoids are 

 divergent from the very fixation of the bilateral stage. Although this view, as we 

 noticed above, has been advocated on other grounds, I do not think we should accept 

 such a drastic resource until all else has failed. 



Hence in Antedon (Bury and Seeliger) the first division is into 1 + 2 + mesenteron 

 and 2'+ 3, according to our interpretation. 



The anterior then divides into 1+2 and the mesenteron, whilst the posterior divides 

 into 2' and 3, 2' to its own dextral side, to become aboral coelom, and 3 to the left 

 posterior end of the larva to form the oral ccelom. 



/1+2 



< anterior coelom < , 



\mesenteron 

 i /.t, o\/ 2 (aboral coelom) 

 posterior coelom (2 +3)^ ' 



\3 (oral ccelom) 



If Bury's or Seeliger's memoirs be read over carefully with this interpretation kept 

 in view, several strange features in the early development seem thereby to be explained. 

 Thus Bury first pointed out that the two elements of the posterior ccelom become 

 partially separated into a dumb-bell shape, the connecting stalk passing completely 

 through the annular mesenteron. Such a remarkable feature might well be caused by a 

 persistent connection between posterior ccelom (3) and right lateral ccelom (2') pushing 

 itself forwards into the mesenteron. Again, we find that the fully developed bilateral 

 larva of Antedon, as figured by Bury (e.g., fig. 17), shows the right lateral ccelom (' right 

 body-cavity of Bury ') lying on the right side exactly opposite the hydroccele, whilst the 

 posterior ccelom (left body-cavity) is exactly posterior to the mesenteron (cf. also fig. 

 20). His earlier stages (figs. 7, 8 and 9) # and also those of Seeliger appear to favour 

 this view. On such an interpretation the aboral ccelom of Crinoids = the epigastric 

 ccelom of asterids = the right lateral ccelom of the bilateral ancestor, from which equation 

 there follow many evident interesting deductions which cannot be pursued here. 



Holothuroidea. 

 In Holothurians we have Synapta described by SEMONt and confirmed by Clark J. 

 Here we have the first division into mesenteron and anterior coelom, which must have 

 in itself the elements 1 + 2 + 2'+ 3. A division into two leaves the vestigial pre-oral 

 coelom (anterior ccelom of Bury) and the hydroccele together (1+2) on the one hand, 

 and the other elements in one sac, namely, 2' + 3. This division therefore agrees with 

 that of Antedon, for we have now (l) mesenteron, (2) 1 + 2, (3) 2' + 3. The latter sac, 

 corresponding with the ' posterior ' coelom of Antedon, then divides into 2' and 3, 2' 

 forming the right body-cavity and 3 the left. 



* These figures seem to have their true axes of symmetry at 45° to those assumed by Bury. 



t Semon, R., " Die Entwicklung der Synapta digitata und die Stammes geschichte der Echinodermen," Jen. Zeitsch. 

 f. Naturvriss, vol. xxii., 1884. 



% Clark, H. L., Johns Hopkins Univ. Memoirs, iv. 



