THE TANGANYIKA PROBLEM. 269 



possessed a similar primitive type of alimentary canal. Or, 

 in other words, it would appear that we have in these 

 features a clear indication that the halolimnic mollusca 

 are primitive as a whole. When, however, we try to 

 push their relationship further, we find that owing to 

 the present unsatisfactory state of our knowledge of 

 the Prosobranchiate group of gastropods, it is by no means 

 so simple an affair as one might suppose, even after 

 having carefully examined the anatomy of each of the 

 members of the halolimnic series to determine satisfactorily 

 with what recognised groups their affinities really lie. 

 Something of the nature of these difficulties will become 

 apparent if we consider the following facts. In the first 

 place, Typhobia and Nassopsis have been already said to 

 belong to the Melaniadae by the conchologists, but from 

 what we now know about the very wide morphological 

 difference which distinguishes the organisation of Typhobia 

 from that of Nassopsis, it is quite clear that both these 

 organisms cannot possibly be regarded as belonging to the 

 same family. If one of them is a Melania, the other is 

 not. And, as a matter of fact, what is a Melania ? The 

 genus Melania was founded by Lamarck upon the Mada- 

 gascan species, Melania amarula, but as Bouvier rather 

 plaintively points out, there is not one single anatomical 

 feature which can be said to distinguish M. Amarula from 

 Cerithium vulgalum, a form which, according to existing 

 classifications, is relegated to another family, the Cerithidce. 

 Melanopsis, as Bouvier found, is as widely separated from 

 the type of Melania in one direction as Nassopsis is from 

 Typhobia in another ; and these few examples will, it is 

 trusted, bring to light the fact that the so-called group, the 

 Melaniadae, is no real group at all ; as it exists in the litera- 

 ture of to-day it is an entirely heterogeneous assemblage 



