6 4 



IIIROSHI OI1SH1MA : 



of the Cucumariida: have hitherto been principally the number, size 

 and arrangement of tentacles, the distribution of pedicels and papilla?, 

 and the structure of calcareous deposits in the perisome as well as of 

 the calcareous ring. The presence or absence of the posterior 

 prolongations in radial calcareous segments seems to be scarcely of 

 more than specific import, though their presence in association 

 with the prolonged " main canals " seems to be very frequent in the 

 sporadipodous forms, such as Thyonc and Phyllophorus. I should 

 think the construction plan of calcareous deposits in perisome may be 

 of use for the distinction of genera or subgenera ; however, we 

 have as yet, as ÖSTERGREN has said [37, p. 212], no sufficient data 

 in that regard to avail of. That the anterior notches in the radial 

 calcareous segments are in close relation with the number, size and 

 arrangement of tentacles, was first pointed out to be of importance in 

 classification by ÖSTERGREN [36, p. 5]. 



As a result of my studies I have come to see that, at least 

 within the subfamily Phyllophorinae, the ordinal rank of the tentacles 

 present— not a mere statement of their number — taken in conjunction 

 with the anterior notches in the radial calcareous segments, is of 

 greater importance than the distribution of pedicels and papillae in 

 distinguishing genera. 



The distribution of pedicels and papilla? often varies very much 

 according to individuals or to age in one and the same species. 

 Both the genera Semper ia Lam pert and Oc nus Forbes present in 

 that respect no distinct demarcation against the genus Cucumaria, 

 and therefore must be included in the latter [Ludwig, 26, p. 345]- 

 HÉROUARD [16, p. 9] has shown that young Cucumaria mendax 

 ThÉEL has five rows of pedicels, as is said to be characteristic of 

 Ocnus, but older ^individuals of that species resemble Semperia in 

 that they possess additional pedicels scattered in a few number on 

 interambulacra. A similar fact is also seen in C. vegœ ThÉEL [46, 

 p. 114]. According to Lampert [19, p. 61], Orcula cucumiformis 



