SMITH : MOLLCSCA OF LAKE TANGANTIKA. 83 



therefore, at all remarkable that these Tanganyikan thalassoid species 

 should in their anatomy exhibit more or less close similarity to marine 

 families. Tiphohia, for example, is said to resemble Xenophora, Stromlus, 

 and Capuhis as regards the radula, whilst, in respect of the nervoiis 

 system, it approaches both Melania (amarula) and Cerithium. "The 

 whole anatomy of Chytra is singularly like that of Capulus.'^ 

 Paramelania and Bythoceras are "regarded as a group of rather 

 primitive Cerithoids,'.' and " Spekia would in many ways appear to 

 be very like a primitive JRissoa." 



This now brings me to the conclusion of the introductory" discussion 

 of these most remarkable shells, which, I am well aware, is all too 

 brief and superficial, but still it is the expression of opinion of a pro- 

 fessional conchologist which can be compared with the results criticised. 

 I will now give a short resume of the species which compose the fauna 

 of the lake, but before doing so I would call attention to a remarkable 

 classification of the thalassoid forms published by H. Nicolas in 1899. 

 From the fact that some of the genera had been placed in different 

 families by various writers he resorted to the plan of placing the 

 whole of the twenty-four genera which had been previously described 

 in the single family Tanganyikidae, which had been proposed by 

 M. Nourry in 1897. The genera, he points out, have relationship with 

 eight marine families, which he enumerates, and finally distributes 

 them into the following ten gi'oups or series, namely — Buccinopsidae 

 with Bourguignatia as the typical form, Nassopsidse with Paramelania, 

 Muricidopsidse with Tiphobia, Trochodopsidae with Limnotrochus, Neri- 

 topsidae with Spekia, Eissopsidae with Horea, Cancellopsidae with 

 Lavigeria, Naticidopsidae with Rumella, Littorrinidopsidae with Stanley a, 

 and PyramidellopsidEe with Syrnolopsis. 



The absurdity of this classification at once becomes apparent upon 

 a very slight investigation. Here we find Bourguignatia, which is 

 generically the same asFaramelania, placed in one section, Buccinopsidae, 

 whilst Paramelania is located in Nassopsidae. Moreover, it has been 

 shown by Mr. Moore that this genus bears relationship to the Cerithia, 

 so that these group names suggested by M. Nicolas are misleading 

 with respect to the affinity of the form in question. The same may 

 be said of all the other sectional names suggested by that author for 

 genera which have been studied anatomically. Lavigeria { = Nassopsis), 

 for instance, placed in Cancellopsidae, has no relationship with Cancel- 

 laria, if that genus is referred to, Tipliohia has nothing to do with 

 Murex, nor Spekia with Nerita or Neritopsis, Limnotrochus is not 

 a Trochus, Syrnolopsis is not a Pyramidellid, and Bathanalia is not 

 allied to the Turbinidae. Of course, these sectional names were 

 suggested to their author by the general form of the shells, but I must 

 confess I do not see the resemblance in several cases ; for instance, 

 Spekia does not seem to recall Nerita, Lavigeria a Cancellaria, or 

 Paramelania a Nassa. Here I may also refer to the classification 

 proposed by Bourguignat (i), which was based on a collection brought 

 from Tanganyika by M. Giraud. Here two new families are proposed, 

 Hauttecoeuridae and Giraudidae, the former to include the genera 

 Tanganikia and Hauttecoeuria, and the latter to include Giraudia and 



