88 LOGGERHEAD TURTLE. 



account for the contradictory descriptions met 

 with in authors relative to species which, in every 

 respect, but the number of scales, appear to be the 

 same. An instance of this occurs in the seem- 

 ingly accurate figures of Gottwald, which agree 

 in general appearance with those of the Mydas, 

 l>ut at the same time have more numerous scales 

 on the shield, and consequently do not correspond 

 with the established character of that animal ; or 

 if we suppose them, which is more probable, to 

 represent the Caretta, they still exhibit a variety 

 with sixteen instead of fifteen scales on the 

 shield. 



The species figured in Aldrovandus, Quad. Ovip. 

 pp. 714, 715. was probably intended for the Ca- 

 retta, having fifteen dorsal scales : it is not, how- 

 ever, to be considered as a very accurate repre- 

 sentation of the animal, and is merely admitted 

 into the present publication in order to enable the 

 scientific reader to exert his own judgment on the 

 subject. 



In reality it is not without a very careful exa- 

 mination that the true specific differences of the 

 marine tortoises can be well understood ; since, 

 exclusive of the plates of the shell, they are known 

 to vary in those marks which have been some- 

 times fixed upon as specific characters, and parti- 

 cularly in the number of external or visible claws 

 on the fins, from which Linnasus attempted to 

 distinguish them ; subsequent observations hav- 

 ing proved that this mark is perhaps less to be 

 depended upon than any other; and we are ex- 



