276 THE TANGANYIKA PROBLEM. 



ing of the halolimnic gastropods, Nassopsis, we find our- 

 selves confronted with a type of organisation which, from a 

 morphologist's point of view, is at once important and 

 unique. The radula of this form is somewhat like that of 

 the genus Vivipara, but it is much more closely similar 

 to that of a number of Littorinas ; in like manner, the 

 nerves are arranged upon a plan which at once recalls the 

 more primitive types of the Prosobranchiata and even the 

 Rhipidoglossa themselves. To make this quite clear, I 

 would, in the first place, point out that if we consider the 

 disposition of the anterior ganglionic masses in the different 

 groups of the Prosobranchiata, we find that in all the forms 

 of the Rhipidoglossa, both diatocardiate, and monatocar- 

 diate, and their derivatives, the pleural ganglia are below 

 the oesophagus and closely applied to the pedal ganglia. 

 And I have, consequently, spoken of this condition of 

 the nervous system as being hypoathroid. In Nassopsis, 

 Vivipara, and some other forms of archi-taenioglossa, 

 the pleural ganglia are more or less half-way between 

 the cerebrals above and the pedals below the oesophagus. 

 The ganglia are scattered, and I have, consequently, 

 spoken of this condition as being dystenoid. In all the 

 higher Tsenioglossa and their derivatives, such as the Pteno- 

 glossa, the Toxoglossa, and the Rachiglossa, the pleural 

 ganglia are intimately related with the cerebral ganglia, 

 both the cerebrals and the pleurals forming an almost indis- 

 tinguishable ganglionic mass. And, in consequence, we 

 may speak of this condition of the nervous system as 

 being epiathroid when compared with the other two. 



Nassopsis, like Vivipara, appears thus to stand half-way 

 between the primitive Rhipidoglossate types and the 

 higher Tsenioglossa and their derivatives ; and this conclu- 

 sion is further driven home by the presence in Nassopsis of 



