162 D"R. ,T. D. MAGDONALD ON THE 



If we now turn to the analogy of shell-characters in a general 

 way we find : — 



1st. That some Grasteropodous families are altogether naked, 

 having no representatives furnished with a shell. 



2nd. That some have both naked and shell-bearing species. 



3rd. That others are all furnished with a shell, having no naked 

 representatives. 



Under the first head would come the Nudibranchiata of authors, 

 and the second would include the Tectibranchiata, the bisexual 

 Pulmonifera, and the Heteropoda ; while the third would comprise 

 all the remaining families. But for the position here given to 

 the Heteropoda, this arrangement would seem to answer very 

 well. If, however, we take a formula derived from the above 

 heads and apply it to the second division, some interesting results 

 will be obtained. Referring to the Heteropoda, first, by way of 

 illustration, we find six typical genera ; two of these are entirely 

 naked, namely Firola and Cerophora ; two others have a shell of 

 only sufficient size to protect the viscera, viz. Cardiapoda and 

 Carinaria ; while the two remaining genera have an operculate 

 shell, completely including the retracted animal, Oxygyrus and 

 Atlanta. On attempting to arrange the shells we see that those 

 of Cardiapoda and Oxygyrus are cartilaginous with an involute 

 nucleus, while those of Carinaria and Atlanta are hard and 

 vitreous-looking with a spiral nucleus. Two distinct groups are 

 thus clearly indicated. 



First Group. 

 absent Cerophora. 



{protecting the viscera Cardiapoda. 

 operculate, including the 1 q 

 retracted animal J v™ 



Second Group. 

 absent Firola. 



(protecting the viscera Carinaria. 

 operculate, including the j .,-, , 



Shell 



Shell 



spiral) { retracted animal 



It might be asked, why Firola should be placed with Carinaria 

 and Cerophora with Cardiapoda. But the only reason that need 

 be given is, that the general appearance of the animals and the 

 comparison of their lingual dentition would appear to indicate 

 the position assigned to them. 



Many persons think that the Heteropoda should be separated 

 from the Grasteropoda as a distinct class, but this would scarcely 

 appear to be necessary. There is a much wider difference between 



