190 $. V. PRINCIPLES OF 



4. When two or more parts are fixed on as diag- 

 nostics of a permanent species, they must be such 

 as are found immediately connected in the recent 

 subject. 



5. When permanent species depend on more parts 

 than one, similar parts in a detached state are not 

 to be considered as the foundation of permanent 

 species When the parts in question thus occur, 

 they must be arranged as temporary species, ex- 

 cept they be ascertained to belong to spceies al- 

 ready established, to which they must be referred 

 as imperfect specimens, (v. specimen.) 



According to the above principles, the parts on 

 which permanent species may be established are as 

 follows f . 



f The parts we have fixed on for the construction of permanent 

 species, are those which experience has pointed out, as best calcu- 

 lated for the purpose Other parls might answer, however, in a 

 certain degree the end we have in view ; and perhaps, to many, 

 may appear even preferable to those adopted It may not be im- 

 proper, therefore, to state more at large the reasons, that have led 

 to each particular selection especially in those genera where 

 various parts might have been taken, agreeably to the principles 

 above assumed 1st. In respect to the genus Mammodolithus We 

 must observe, there is no doubt that the united vertebrae, and in- 

 deed otiier parts of the skeleton, are as capable of affording cha- 

 racteristic marks of the orders and other natural tribes m Mammi- 

 ferous animals, as the bones of the head, if the general knowledge 

 of comparative anatomy were equal to their selection But it is the 

 tetth and their situation in the head, on which Linue has chiefly 



