( 16 ) 



P. tecta, but is at once known by its strongly dentated margin, 

 and by the three hinder marginal plates on each side only occu- 

 pying the hinder margin of the fifth vertebral plate" — App., 

 Cat. S. E., p. 19 ; or in other words Jerdonella is based on a 

 specimen or variety, whose three hinder marginals are so narrow 

 as to only occupy the space usually taken up by two and a half. 

 Jerdon was undoubted nearer the truth when he hesitated to 

 consider this form as ever specifically distinct from tectuTn, and 

 doubtless the more varieties or local races are discovered, the 

 more shall we be compelled rather to reduce the number of 

 species, than to multiply genera ! Grows to 8 inches. 

 Inhabits streams in Sylhet. 



P. ventricosa, Gray. Sup., Cat. S. R., p. 60. 

 Thorax vertricose. First vertebral shield elongate, very 

 narrow. Eounded behind, second narrow, attenuated behind, 

 margin entire, third small, keeled, fourth elongate, oblong six 

 sided suddenly narrowed and produced in front. Hab. India 

 (Jerdon). 



P. Leithii, Gray. Sup., Cat. S. R., p. 60. 



This species is based on a skull from the " River Poonah," 

 which, if not belonging to P. flaviventer, is probably referrible to 

 some other genus. 



Having enjoyed the advantage of examining a large series of 

 specimens of Pangshura in the Allahabad Museum, through the 

 courtesy of J. Cockburn, Esq., the Curator of that institution, I 

 am led to believe that we have in India but three species of the 

 genus, sub-divided into fairly recognizible races or varieties. In 

 the first place, with a large series before us, we are forced to 

 admit, that the shape of the plates is a very variable character, 

 and of little use in discriminating species. Equally variable is 

 the colouration of the sternum, the sternum of" flaviventer " being 

 either uniform {ak in Gunther's type) or marked as in more 

 usual in the genus. The form of the shell is perhaps a better 

 mark and this seems to follow a geographical range. Taking P. 

 flaviventer as our type, occupying as it does the centre of the 

 area of Pangshura (so far as longitude goes), we find that the 



