258 ' ENDEAVOUR " SCIENTIFIC RESULTS. 



genus from a Selaginopsis without one ; here, however, as 

 Stechow accepts the presence of some form of operculum as a 

 primary character of the whole family, it must be assumed 

 that he excludes such species now assigned to Dictyocladium 

 and Selaginopsis as are inoperculate ; only he has then 

 provided in his system no place for these species, any more 

 than for the inoperculate species hitherto included in the 

 genera Sertularia and Sertularella. 



What has doubtless contributed largely to the reluctance 

 of systematists to adopt Levinsen's views is the difficulty 

 often found in determining the nature of the operculum in 

 species with which we are only partially acquainted, and of 

 which we may often examine many specimens without being 

 able to detect any trace of such a structure, or, where such 

 traces are visible, to ascertain the true form. 



For example I have mentioned elsewhere that in Thyros- 

 scyphus marginatus one sometimes finds empty hydrotheca? 

 with opercula sufficiently preserved to enable their character 

 to be clearly made out, while in other cases hydrothecse are 

 found containing the hydranths in good condition but 

 entirely devoid of opercula. Levinsen's contention that he 

 can discern from the form of the hydrotheca-margin what 

 that of the operculum has been, is no doubt correct in general, 

 but I have seen hydrothecse as to which I was quite unable 

 to satisfy myself from the form of the border whether they 

 belonged to Levinsen's genus Odontotheca or to Sertularia. 



In regard to these two genera, I have great doubt as to 

 the sufficiency of the distinction between them. In both 

 cases the border of the hydrotheca has two lateral teeth, 

 between which are two sinuses, the adcauline the larger in 

 Odontotheca, the abcauline in Sertularia ; in each sinus is 

 fixed an opercular membrane. In Sertularia the abcauline 

 membrane is, in most species, provided with a free distal 

 valvular portion, which is wanting in Odontotheca. As, 

 however, it is stated that some species of Sertularia are 

 without the usual free valvular portion, while some species 

 of Odontotheca possess it, it seems very doubtful whether 

 the differences are sufficient to justify the creation of a new 

 genus. The same remark applies to the genus Hydrallmania, 

 as modified by Levinsen in accordance with the opercular 

 characters ; and I may add that even according to the 

 colonial characters I see no sufficient ground for this genus : 

 the hydrotheca are as truly biserial as in the rest of the 

 family, though both series are seated so near to the front as 

 to be very nearly (but not exactly) in one line. 



