262 Prof. C. B. Adams on (he 



classes two are wholly testaceous, and may be distinguished 

 not by the shells alone, but by their position on the soft parts ; 

 and three are partly testaceous, of which two may be dis- 

 tinguished by their shells, so far as they are testaceous. 



In the orders, we find their value not much increased. The 

 two orders of Brachiopoda are distinguished by a single cha- 

 racter of the shells, namely, the presence or absence of arti- 

 culation between the valves. This is not the basis of their 

 ordinal classification, which would be artificial if founded on 

 one character, but it is an index of difi^erences which charac- 

 terize the entire animals. 



The two Lamarckian orders of Conchifera are separated 

 on account of a difference in their soft parts, namely, the 

 number (whether one or two) of the adductor muscles, which 

 character is clearly indicated in the shells of all the families 

 except in the Tridacnidse, where the approximation of the 

 two muscles presents the appearance of a single impression. 

 In the Mytilidse, which have several adductor muscles, some 

 of which are inconspicuous, the defect in the ordinal cha- 

 racter of the soft parts is itself indicated in the shells. In 

 Mr Gray's list of the Genera of Recent Mollusca five orders 

 of Conchifera are enumerated. I have not seen them cha- 

 racterized, but it is obvious from their genera that ordinal 

 characters cannot be found in the shells. The two orders of 

 Pteropoda in Mr Gray's list, as in several other authors, are 

 indicated by the presence or absence of a shell, although the 

 classification of Mollusca in that elaborate and excellent 

 paper was designed to include all the modifications that were 

 suggested by an examination of the soft parts of 5000 species. 



The two Lamarckian orders of Gastropoda, Gastropoda 

 proper and Trachelipods, each contain both naked and testa- 

 ceous mollusca ; but, with some important exceptions, the 

 shells in each order may be distinguished. Mr Gray's five 

 orders of this class cannot be distinguished by ordinal cha- 

 racters of the shells. 



In the three orders of Cephalopoda one is indicated by 

 the shells of Nautilus. The other two orders, including 

 Argonautidse, contain each one testaceous genus, with many 

 naked genera, but these testaceous genera are widely different. 



