FOEM OF BODY OF MOLLUSCA. 323 



their common origin in this disposition of parts. The latter are 

 principally seen as differentiations of the edge of the mantle, and 

 are connected with the function of the branchial cavity. Part of the 

 edge of the mantle is produced into a groove, which serves to bring 

 in water, and which may be converted into a tube by the folding 

 over of its two edges. We meet with a siphon of this kind, though 

 in all stages of gradual differentiation, in a large number of aquatic 

 Gastropoda (Buccinum, Dolium, Harpa, Tritonium, Murex, etc.). A 

 second siphon, formed in the same way, but smaller in size, is 

 generally found at the opposite end of the branchial cavity ; it serves 

 to carry the water out from it. Various other kinds of processes, or 

 tentacular appendages, lead to fresh complications in its structure 

 (e.g. in Strombus, Pterocera). 



When the shell undergoes atrophy the mantle generally does so 

 too. This is mostly the case in the division of the Opisthobranchiata, 

 some of which have a more or less rudimentary shell ; while in others, 

 when adult, there is no shell at all. As all these forms had a shelled 

 larval stage, the atrophy of the shell must have been brought about 

 during their ontogenetic development ; and it follows that those 

 Opisthobranchiata, which are naked in their later stages, were derived 

 from forms that had shells. The larval shell and its accompanying 

 mantle-fold, even though feebly developed, are therefore rudimentary 

 organs, which prove that the naked Opisthobranchiata had the same 

 origin as the rest of the Gastropoda. Where these rudimentary shells 

 are retained by the adult animal they must even then be regarded 

 as degenerate parts, and not as developing shells ; for here again a 

 comparison with the larval forms shows that the shell had a much 

 greater significance than have the rudimentary structures found in 

 the adult stage of these organisms. It is of great importance also 

 as explaining the position of the anus and of the genital orifice, 

 which can be due to nothing but a former greater development 

 of the shell. 



Within smaller divisions also we meet with series of degene- 

 rating parts, as for example in the Heteropoda, where Atlanta has 

 a well-developed shell and mantle, while in Oarinaria they are both 

 rudimentary, and in Pterotrachea completely lost. A similar series 

 is observable in the Nephropneusta. 



§ 251. 



The varying extent to which the Foot is developed is of im- 

 portance as affecting the form of the body. In the larvee of the 

 Pteropoda and Gastropoda it has always very much the form of a 

 short, conical, somewhat-flattened process, placed below the mouth 

 (Fig. 170, A p). On the hinder, or dorsal surface, a shelly secretion 

 is formed, which serves as an operculum for the orifice of the shell. 

 Owing to its increase in size, especially in the aboral direction, the 

 foot of' most GastrojDocla comes to have a broad lower face, which is 



y 2 



