10 CLAVID^E. 



must confess to some doubt as to its validity. The prin- 

 cipal character that separates it from C. multicornis is the 

 diffusion of the reproductive bodies. But this occurs 

 occasionally in other species, and can hardly be accounted 

 a character of much significance. Muller* represents 

 such a condition in his figure of the Hydra squamata ; and 

 Agassizf, in his account of Clava leptostyla, says that the 

 bunches of reproductive buds frequently extend doAvnward, 

 either continuously or in detached masses, over one-third 

 of the distance below the base of the tentacles. Possibly 

 we may have in C. diffasa a species of which this is the 

 normal and constant characteristic ; but at present I am 

 inclined to regard it as a mere variety. 



It is right to add that Prof. Allman found the clusters 

 of gonophores scattered in all the zooids of the colony which 

 he examined. The polypites, too, were attenuated, and of 

 a delicate rose-colour. 



Hab. In rock-pools at low water of spring tides. Out- 

 skerries, Shetland Isles (G. J. A.) . 



Genus TUBICLAVA, Allman. 



Der. Tubus, a tube, and Clava, the name of a Hydroicl genus. 



GENERIC CHARACTER. Stems erect, simple or branched, 

 given off at intervals from a creeping stolon; the whole 

 ccenosarc invested by a polypary ; polypites claviform, with 

 scattered filiform tentacula : reproduction by means of fixed 

 sporosacs, borne in clusters on the body of the polypite } 

 immediately behind the posterior tentacula, or on the sum- 

 mit of very short stems developed on the creeping base. 



THE genus Tubiclava is most nearly allied to Clava. 



* Zoologia Danica, pi. iv. 



t N. H. United States, vol. iv. p. "2'22, fig. &.'. 



