THE DEVELOP.MKNT OF OPIIIOTHRIX FRAGILIS. 589 



rachis wliicli eventually surrounds tlie disc and from wliich 

 the genital organs arise as buds. The ampulla of the stone- 

 canal il.ax.) is seen in fig. 52 to be still in continuity with 

 the perihasnial space to which it gave rise. This ampulla — 

 the original left anterior ccelom — into which both pore-canal 

 and stone-canal open (cf. fig. 51) is the equivalent of the 

 axial sinus of Asteroidea and Echinoidea, but has nothiuer to 

 do with the space which originally received that name in 

 Ophiuroidea. This latter space, lettered sinus h in my paper 

 on Aniphiura squamata, is not formed till the post-larval 

 period, and hence is not described in this paper. Its outer 

 wall separating it from the general hypogastric (i.e. left 

 posterior) ccelom is a thin membrane, and the space itself 

 appears to be due (as far as I can make out from a re- 

 examination of my sections) to an invagination of the coelomic 

 cavity into the tissue lying above the stone-canal. Now in 

 Asterina gibbosa the cells which give rise to the genital 

 rudiment are likewise iuvaginated, but the cavity of the 

 invagination is occluded. When we reflect also that the true 

 axial sinus of Ophiuroidea is of limited extent, the confusion 

 between the true axial sinus of Asteroidea and the wrongly 

 named axial sinus of Ophiuroidea becomes explicable. The 

 right hydrocoele is seen in fig. 51 as a closed space {r. hij.). 

 When I commenced my work on the late development of 

 Aniphiura squamata I had made no researches on the 

 early development of Echinodermata, and hence I was much 

 puzzled by the appearance of this space, whose origin I could 

 not trace. After long wavering I persuaded myself on the 

 evidence of one or two sections that it, too, was derived from 

 an evagination of the general coelora, and so it is lettered as 

 if it were part of sinus b, the wrongly denominated axial 

 sinus. Of course, this is a mistake which I endeavoured to 

 correct in my paper on Asterina gibbosa, and wliich 

 receives renewed correction here. 



As the metamorphosis nears its conclusion the ciliated epi- 

 thelium covering the long postero-lateral arms begins to de- 

 generate, and the animal sinks lower and lower in the water ; 



