THE TANGANYIKA PROBLEM. 
277 
a labial commissure, which, as Bouvier has shown, is a 
special feature of the Rhipidoglossa and their allies. But, 
not only is Nassopsis an archi-taenioglossan in the character 
of its nerves, we find, as I have shown above, that this 
type has a most unique and primitive gastric apparatus ; 
which, like Chytra , unites at the same time the presence of 
a style sac, and a style with the spiral stomachic caecum 
common to the Rhipidoglossa. Nassopsis is thus, on 
anatomical examination, shown clearly to be a new type of 
archi-taenioglossa, and, consequently, a genus which will be 
of great value in the future study of the relationships of the 
different sections of the Prosobranchiata. As a matter of 
fact, Nassopsis, in the Rhipidoglossate character of its 
stomach, together with the possession of a crystalline style, 
is a much more striking archi-tsenioglossate form, than 
even Vivipara itself.* 
Having followed these matters relating to the affinities of 
Nassopsis, we have completed the consideration of the 
structure of the halolimnic gastropods in Tanganyika ; and 
so far as our knowledge of the vast hordes of Prosobranchs 
which live in the sea and the fresh-water of the earth at the 
present time make such an attempt possible, we have esti- 
mated the affinities of the different individual members of 
the halolimnic group. None of these molluscs, as we have 
seen, appear capable of being regarded as in any way 
related to the fresh-water molluscs which exist elsewhere in 
the world to-day ; while the whole group presents us with 
an array of characters which at once carries them back 
to the position of ancestors of the numerous forms, chiefly 
marine, which their structure appears distinctly to fore- 
* For the reasons stated in the last chapter, I have not included the genera Neothaumus 
and Stanleya in the halolimnic group, while for Syniolopsis we have as yet not sufficient 
material for a complete investigation. 
