268 THE TANGANYIKA PROBLEM. 



while others are inhabitants of the quiet dark water which 

 covers the deep floor of the lake. In the second place — and 

 it is this fact which will carry most weight — the type of 

 alimentary canal which the halolimnic gastropods present 

 is not confined to them, but is present in part or in its 

 entirety in very widely divergent oceanic forms. We have 

 only to compare the alimentary canal of Typhobia with that 

 of Strombus to see that both are built entirely on the same 

 plan. In each there is a straight oesophagus, a curiously 

 bent intestine, a stomach with two chambers and a crys- 

 talline style. Still further, we find that the possession of an 

 anterior stomachic chamber and a crystalline style is not 

 confined to the prosobranchiate mollusca ; it occurs also, 

 and presents the same relationships in the Lamelli- 

 branchiata, and, when we realise this fact, there can, I think, 

 be little doubt, as the investigations of Collier, Huxley, 

 Woodward and myself have tended to show, that the 

 type of alimentary apparatus which is found in the Lamel- 

 libranchiata, in Strombus, Pteroceras, Bitkynia, and which 

 now turns up again in every member of the halolimnic 

 group, is indicative of a retention in them all of a primitive 

 type of organisation which once belonged generally to the 

 older molluscan forms.* 



As a group, then, and, taken as a whole, one of the most 

 remarkable characters which the gastropods of the halolimnic 

 series possess is, that they all present a similar and primitive 

 character in the arrangement of their alimentary apparatus, 

 and consequently appear to be a series of forms which 

 retain the characteristics of a time when most gastropods 



* This view has been strikingly confirmed by a. comparison recently made by one of 

 my students and myself between the gastric apparatus of Spirula, Natitilus and Nassopsis, 

 whereby it was shown that the style sac of the Gastropods is unquestionably present in 

 the ancient Cephalopods as well. Pro. Roy. Soc, vol. 70, p. 231. 1892. 



