262 REPORT OF NATIONAL MUSEUM, 1802. 



ities or arbitrary tokens of the formations, in which they are found. 

 With the aid of comparative anatomy and systematic taxonomy these 

 fossils, notwithstanding their imperfection and faunal incompleteness, 

 become not only indispensable in geological investigations, but real rep- 

 resentatives of the grand succession of animal life that lias existed upon 

 the earth. 



Jn the ease of fossil Vertebrata it is the skeleton and teeth almost 

 alone that are used in classification, but so distinctly were the classifi- 

 catory characteristics of those animals impressed upon the hard parts 

 of their bodies that the nearly or quite complete structure of the whole 

 animal may legitimately be inferred from them. Furthermore, these 

 characteristics are so distinctly impressed upon the teeth and upon cer- 

 tain essential parts of the skeleton that legitimate classification of 

 extinct forms can often be established upon no more than a few scat- 

 tered teeth or fragmentary portions of the skeleton. 



The extinct Mollnsca, as has already been shown, are represented by 

 their fossil shells, and these are classified precisely as are the shells of 

 living members of that class.* Moreover, the classification of the fossil 

 molluscan forms accords more completely with that of the living than is 

 the case with the Vertebrata because living mollusks are largely classi- 

 fied by their shells alone, even when the soft parts of the animal are 



available. 



As in the case of the Mollnsca the fossil Echinodermata are classified 

 in the same manner and by the same means as are the living forms, 

 because the classification of the latter is based mainly upon those which 

 in the extinct forms become fossilized. 



The method of classification of all the fossil Annulosa is essentially 

 the same as that which is used for the living forms, the difference, if 

 any, being mainly due to the usual incompleteness of the fossil speci- 

 mens. In such cases, as well as in those of the greater part of other 

 fossil remains, more attention is given to certain characteristics of the 

 hard parts than is found necessary when the whole animal is available 

 for study. 



While the classification of those living Cflelenterata which have no 

 hard parts is necessarily based wholly upon the structure of their soft 

 bodies, that of those living forms which possess skeletal or protective 

 parts is largely based upon them. In this latter respect the extinct 

 Ccelenterata are classified in the same manner as are their living rep- 

 resentatives, namely, by means of their coralline hard parts. 



Because the soft bodies of the living Protozoa are so nearly struc- 

 tureless that they can not furnish a satisfactory basis for classification 

 it is necessarily based upon the structure and character of the hard 

 parts in the case of those Avhich secrete them. Therefore the classifica 



* The elaborate classification of the fossil cephalopoda by means of their septal 

 flexures may be taken as an exception to this statement. 



