Dr. J. E. Gray on the Mud-Tortoises of India. 327 



The sternum being furnished with Haps on the sides (which 

 cover the legs when they are contracted), or being narrow at the 

 sides and leaving the legs bare (as in most freshwater and marine 

 tortoises), furnishes a most natural and easily observed character 

 for the division of the group, and as such has been used by most 

 authors. But it has been shown that each of these groups con- 

 tains animals with very different skulls ; and it is a matter of 

 serious consideration whether the form of the skull, on which 

 such important peculiarities in the animal economy depend, is 

 not of more importance than the covering or exposure of the 

 feet when they are withdrawn. When first the covering of the 

 feet was observed, it was connected with a bony margin to the 

 dorsal disk ; but it is now well ascertained that many species 

 with covered feet have the margin flexible and without bones, 

 like the other mud-tortoises. It is to be remarked that all the 

 tortoises that have flaps to cover their feet have callosities on 

 the two anterior bones of the sternum, which have never yet 

 been observed in those which have naked feet. This character is 

 common to those that have thin skulls and jaws and narrow 

 alveolar edge, and those which have thicker skulls and wider 

 alveolar surface. 



Cuvier and Wagler described and figured the skulls of 

 two or three species of this group ; but all the skulls which 

 they had the opportunity of studying belonged to a single 

 type of form, of a thick and solid consistency. In my ' Cata- 

 logue of Shield Reptiles in the British Museum ' I figured 

 a few skulls of the species which we then possessed, pointing- 

 out that they belonged to two different groups — one solid, and 

 the other light and thin ; and in the ' Supplement to the Cata- 

 logue of Shield Reptiles' I figured and described the skulls 

 of many more species. I used this character to separate the 

 soft-disk mud-tortoises into two families, Trionychidse and 

 Chitradte — one having a solid, and the otlier a thin and light 

 skull ; and I divided the genera of each family according to 

 the form of the skull, especially the form of the alveolar edge 

 of the jaws. I consider this one of the most important steps 

 towards the proper division of the species and defining them, 

 as it affords us the power of dividing them into natural 

 groups : for example, Chitra indica, Trionyx gangeticuSj and 

 Tyrse nilotica have been considered specimens of the same 

 species, but they belong to two very different families ; Chitra 

 indica and Pelochelys Cantori have been regarded as the 

 same species, the one having a very long ovate, and the other 

 a short square skull. In the same manner Fordia africana and 

 Tyrse nilotica (the one having a broad, flat alveolar surface, 

 and the other a sharp thin one) have been regarded as the 



