266 Prof. C. B. Adams on the 



it relates to the peculiarities of colour. But the details of 

 form are more perfectly represented by the skins of Ver- 

 tebrata than by the shells of Mollusca. Yet most of the 

 details of form in the soft parts, which are not moulded in 

 the shells, are indicated by constantly associated characters. 



In their functional relations, shells are skeletons. They 

 serve for the attachment of muscles, and the support of the 

 soft parts. On this groimd, therefore, we may claim for 

 them a value similar to that of the skeletons of vertebrata, 

 but admit the amount of the value to be less, on account of 

 the greater simplicity of shells. 



It is, therefore, true that a collection of the shells of Mol- 

 lusca corresponds to a collection of both the stuffed skins and 

 the skeletons of vertebrata, but is of inferior value, both for 

 the reasons just mentioned, and because their presence is not 

 universal in their department. 



It is perhaps unnecessary to add that while we believe 

 that the shells of Mollusca are sufficient for distinguishing 

 species, we would not regard them as only the basis of species. 

 This would be an artificial method, although in the results it 

 would coincide with a natural method of classification. But 

 the shells are partly the basis of, and partly the index to, 

 specific characters. Professor Agassiz remarked that, as 

 valuable as such investigations are at the present time, he 

 would like to suggest the importance of a series of investi- 

 gations into the relative value of the different parts of the 

 animal, but not a comparison, for this has long ago been done. 



It is the great advantage of embryology in this respect, that 

 we are enabled to learn the real value of characters, for the 

 purposes of classification. There is so much with common 

 classification that is at the discretion of the naturalist, and 

 which, therefore, has no real foundation, that it may be justly 

 called artificial, while nature has afforded sufficient data in 

 growth, to interpret her correctly. If we take Cuvier's clas- 

 sification of the Mollusca, from which naturalists have not 

 much deviated, we shall find that upon embryological data it 

 cannot be carried out. This is illustrated in the cephalopods 

 and gastropods, and Nature's own expression should be re- 

 garded above the opinions of all men. I have examined 



