244 ON THE NATURAL CLASSIFICATION OF THE GASTEROPODA. 



Suborder 2. Anaclodonta. Cusps recurved from the fore part of the basal 

 plates. Dentition typically septiserial, but in some instances reduced to 

 5 or 3 rows by suppression. 



VeUitinidce. 

 Naticidce. 



Tritonidcs 



Strombidce. 



BanellidcB. 

 DoliidcB. 

 Cass id ides. 



Order II. Rostrifera. Muzzle simple or proboscis rudimenta ry. 



Suborder 1. Orthodonta. Cusps direct. 



Pelagic. Heteropoda. 

 Marine. Phoridce. 



Suborder 2. Anaclodonta. Cusps recurved. 



( CyprmdcB. 



VermetidcB. 



Cah/ptrcsidce. 

 Marine and littoral... •{ Planacidce. 



Littorinidce. 



Bissoida. 



Truncatellidce. 

 I* MelaniadcB. 



Aquatic \ Paludinidoe. 



[ ValvatidcB. 



f CyclostomidcB and 



\ their allies. 



Terrestrial 



CeHthiidce. 

 Cerithidea. 

 Potamidince. 



Cyclophoridce. 

 DiplommatinidcB. 



All the families in the first column have otoliths in their ear- 

 sacs ; the few on the right have otoconia. This may be signi- 

 ficant ; but the subject requires further study. 



Just as we have found terrestrial, aquatic, and marine Nerites, 

 there is good promise that corresponding groups may be disco- 

 vered in relation to other types of Anaclodontous Tfcostrifera with 

 septiserial ribbons, the grouping of which is at present very im- 

 perfect. In this research, however, the shell-characters must be 

 subsidiary to the most critical record of the anatomy of well- 

 determined species, so as to afford legitimate grounds for their 

 adoption or rejection as the case may require. 



Indeed, from my own experience, I am quite sure that without 

 this test the assumption of the alliance of even one so-called 

 species with another founded on superficial resemblances can only 

 be guesswork, allowable certainly for convenience and provisional 

 a "angement, but it must always be amenable to the dictum of 

 mor'5 precise anatomical knowledge. 





