Ch. V.] IX CHRONOLOGICAL ORDER. 49 



species of several other classes. We even find the skeletons of 

 extinct quadrupeds in deposits wherein all the land and fresh- 

 water shells are of recent species*. 



Necessity of accurately determining species. The reader 

 will already perceive that the systematic arrangement of strata, 

 so far as it rests on organic remains, must depend essentially 

 on the accurate determination of species, and the geologist 

 must therefore have recourse to the ablest naturalists, who have 

 devoted their lives to the study of certain departments of 

 organic nature. It is scarcely possible that they who are con- 

 tinually employed in laborious investigations in the field, and 

 in ascertaining the relative position and characters of mineral 

 masses, should have leisure to acquire a profound knowledge of 

 fossil osteology, conchology, and other branches; but it is 

 desirable that, in the latter science at least, they should become 

 acquainted with the principles on which the specific characters 

 are determined, and on which the habits of species are inferred 

 from their peculiar forms. When the specintens are in an im- 

 perfect state of preservation,, or the shells happen to belong to 

 genera in which it is difficult to decide on the species, ex- 

 cept when the inhabitant itself is present, or when any other 

 grounds of ambiguity arise, we must reject, or lay small stress 

 upon, the evidence, lest we vitiate our general results by 

 false identifications and analogies. We cannot do better than 

 consider the steps by which the science of botanical geography 

 has reached its present stage of advancement, and endeavour to 

 introduce the same severe comparison of the specific characters, 

 in drawing all our geological inferences. 



Tables of shells by M. Deshayes. In the Appendix the reader 

 will find a tabular view of the results obtained by the compa- 

 rison of more than three thousand tertiary shells, with nearly five 

 thousand living species, all of which, with few exceptions, are 

 contained in the rich collection of M. Deshayes. Having enjoyed 

 an opportunity of examining, again and again, the specimens 

 on which this eminent conchologist has founded his identifica- 



* See vol. i. chap. vi. 

 VOL. III. E 



