EARLY DEVELOPMENT OF CRIBRELLA OCULATA. 397 



the anterior ends of the posterior elements in defining their exact separation from the 

 anterior, and at their posterior ends we have still more difficulty in distinguishing 

 between right and left. These impediments have been recognised recently by Goto * in 

 both Asterina and Asterias, and his conclusions on this point are, I think, largely in 

 agreement with mine, though not exactly in his way of stating them. It is only fair to 

 say that I have not examined Asterina nor Asterias pallida, and that MacBride, who 

 has worked exhaustively upon the former type, casts doubts upon Goto's results. Goto 

 ascribes MacBride's conflicting results to the use of osmic acid as a fixing reagent, 

 after which there is always a difficulty in staining. MacBride replies by indicating the 

 uselessness of corrosive-acetic and glycerine, and the necessity of celloidin imbedding 

 for pelagic larvae, and so on. I confess that all this nicety of technique is somewhat 

 beyond me ; and not having worked on either Asterina or a complete series of Asterias, 

 I cannot presume to judge between these two workers. I can only indicate that with 

 regard to the origin and fate of the body-cavities and the central coelom, Goto's work 

 agrees more closely with the facts as in Cribrella. As regards questions of symmetry 

 and orientation, my facts agree with MacBride's account of Asterina. 



In Asterias pallida Goto finds a pair of ' enterocceles,' right and left, which early 

 fuse anteriorly, producing a stage of a large coelom with an unpaired pre-oral part and 

 a pair of lateral wings closely similar to that in Asterina. The left wing gives rise to 

 the hydrocoele and a posterior portion which he calls the left posterior enteroccele ; the 

 right gives off a dorsal portion, the epigastric coelom, and a residue which fuses with 

 the left posterior ccelom to form the hypogastric coelom of the adult. A comparison 

 of Goto's figures with fig. 53 will show that my right lateral coelom lies over the right 

 posterior coelom in the same position as Goto's epigastric coelom. Hence the total 

 result of the coelomic changes in Asterias pallida is that the left side produces a 

 hydrocele and a residue, the right produces an epigastric coelom and a residue, and 

 the two residues fuse to form a hypogastric codom. This result is disguised by Goto, 

 because he terms the whole of the right wing the right posterior ccelom, whereas from 

 my point of view it contains the lateral element as well as the posterior. 



If Goto's results be looked at in this light, they agree absolutely with mine as to 

 the fate of the body- cavities ; Goto found the epigastric coelom arising from 'the 

 dorsal posterior corner ' of the right wing (right posterior ccelom of Goto), a position 

 so closely agreeing with that in Cribrella, clearly brought about by the growth of the 

 right posterior ccelom forwards and ventrally to the epigastric ccelom. Only ^ this 

 difficulty could have prevented Goto from recognising the homology of the hydrocoele 

 and the epigastric coelom as left and right elements. In a subsequent examination of 

 Asterina gibbosaj Goto found the same arrangement and fate of the body-cavities. 



* Goto, S., Journal of the College of Science, Imp. Univ. Tokyo, Japan, vol. x. part iii., 1898 ; vol, xii. part iii., 

 1898. 



i Goto, S., ibid., vol. xii. part iii., 1898. We may note that he very clearly indicates the epigastric ccelom to be 

 formed of one-half of the right posterior enteroccele, 



TRANS. ROY. SOC. EDIN., VOL. XL. PART II. (NO. 19). 3 o 



