410 PROP. J. A. THOMSON AND MR. W. D. HENDERSON ON [Apr. 10, 



Order II. ALCYONAOEA, Verrill {j^ro 2)arte). 

 Family X E N 1 1 D ^. 



Xenia umhellata Savigny. 



,, quinqueserta May. 



,, memhranacea Sclienk. 



,, ccei'idea Ehrenberg. 



,, ternatana Scherik, var. elongata, nov. 



,, rigida, sp. n. 

 Heteroxenia eliscd>ethce Kolliker. 

 Gesjntidaria coeridea May. 



JVote on the Sjiecies (j/'Xenia. 



In the genus Xenia, as in the genus Clavidaria, the question of 

 species is a difficult one. All the species are within a relatively 

 narrow I'ange, and the differentiating characters are, when taken 

 separately, somewhat tri^dal. Even in the same colony there ai-e 

 sometimes noteworthy difierences in the adjacent polyps — differ- 

 ences which are sometimes as marked as any one of the sepaiute 

 items which are used to distinguish species. It seems likely that 

 many of the diffei'ences ai'e purely modificationcd, and referable to 

 differences in nutrition and the like. 



The number of rows of pinnules is a character which has been 

 much relied upon, but it is apt to lead one astray unless the 

 tentacles observed ai'e equally extended. Moreovei', there may be 

 3 rows at the proximal end and 4 about halfway up, or 2 at the 

 proximal end and 3 about halfway up the tentacle. There can be 

 no confusion between a species with oiie row of pinnules on each 

 side and a species with four rows on each side, but to distinguish 

 two species because one has three rows and the other four 

 appears to us quite misleading unless this detail is supplemented 

 by many others. In X. umhellata we found from 2-4 rows, in 

 X. qioinqueserta 3-5 rows. 



Some workers have attached importance to the presence of wart- 

 like pinnules along with othei'S of the usual elongated type ; but 

 the presence of a few wart-like pinnules at the proximal end 

 ajDpears to us to be very common, and may be naturally expected 

 when a polyp is not fully grown. At the same time, the minute 

 wai-ts of, foi' instance, X rigida are diagnostic in conti'ast to the 

 long pinnules of X. umhellata. 



Another diagnostic featui-e is the presence or absence of a bare 

 strip on the surface or surfaces of the tentacle, but this is apt to 

 be obscured or exaggerated by the degree of contraction. The 

 bare streak may be present at the proximal end and absent higher 

 up, or quite distinct along the distal half and quite obscured by 

 contraction lower down. 



Xenia umbellata Savigny. 



The length of the staUi is 24*5 mm., with a maximum basal 



