OPIIIACTIDAE 



26s 



shape. Adoral shields joining rather widely within. Two broad scale-like outer mouth 

 papillae (exceptionally only one). The infradental papilla is distinctly pointed. Ventral 

 plates a little longer than broad, with convex outer edge ; they are rather broadly con- 

 tiguous in the proximal part of the arms. Dorsal arm plates broadly contiguous in the 

 proximal part of the arms, with convex distal edge ; they are somewhat broader than 

 long. Four, exceptionally five, arm spines, which are rather robust ; the middle ones are 

 more or less distinctly truncate, flattened, with the edges finely spinulose, the end 

 being somewhat axe-shaped. One large, oval tentacle scale. Colour light greyish or 

 white, the younger specimens with two to three well-defined brownish bands on the 



arms. 



The unusual condition of the ventral interradii being naked in their proximal part 

 forms a marked character of this species ; it is, however, not easily ascertained in the 



a b 



Fig. 9. Ophiactis semhtuda, n.sp. Part of oral side (a) and dorsal side (h) X24. 



young specimens. The species does not seem very closely related to any other known 

 species. 



Two of the regenerating specimens are remarkable in having only two arms in 

 regeneration, so that here would seem actually to be a case where a five armed adult 

 would be derived from a self-dividing, possibly originally six-armed specimen. But this 

 is the only case I have ever seen among the very numerous self-dividing specimens of 

 Ophiactis which I have examined. Another specimen has only one full-grown arm, one 

 somewhat smaller, and three young, regenerating arms. Yet another specimen has two 

 full-grown arms, one opposite the other, and two young regenerating arms on either 

 side. On the whole there is evidently a good deal of irregularity in the propagation by 

 autotomy in this species. At any rate, this single case of regenerating five-armed speci- 

 mens cannot prove it to be the rule that six-armed self-dividing young specimens end 

 as five-armed adults, as appears to be the opinion of H. L. Clark. 



9-2 



