172—18 INDIAN TERTIAEY AND POST-TERTIARY VERTEBRATA. 
collected by Mr. Theobald in the Siwaliks of the Punjab, exhibits the nearly complete 
shell of a small (? female) individual, but unfortunately with the exception of the 
anterior marginals, and 1st and 5th vertebrals, the boundaries of the plates are 
invisible. The shell agrees exactly in form with the t3^pe specimen, exhibiting very 
clearly the characteristic flattened and keelless vertebral region. The width is 4*8 
inches, and the height of the carapace P9 inches. The 5th vertebral is of moderate 
width ; and the lateral edges of the carajDace are nearly straight. 
Distinctness and affinities . — The three specimens show clearly that the characters 
of the type are constant ones. Compared with the large series of shells of C. 
crassicollis in the British Museum, the fossil form differs in the characters mentioned 
by Mr. Theobald, and also by the wider posterior border of the 1st vertebral plate. 
The anterior portion of the plastron (fig. lb) shows more marked differences, the 
gular plates being relatively longer, their posterior extremity forming a more acute 
angle, and the suture between the postgulars being only J the length of the inter- 
gular suture, instead of about -j, as in C. crassicollis. These differences, especially 
the absence of a nuchal plate, and the extremely flattened vertebral region, fully 
confirm Mr. Theobald’s conclusions as to the distinctness of this form from 0. 
crassicollis.! and as it is totally different from any other species known to the writer it 
may rank as a distinct species. That it is intimately allied to C. crassicollis is quite 
evident, but the absence of the nuchal plate may render it doubtful whether it can 
be considered as the direct ancestor of that species. C. crassicollis occurs in India, 
Tenasserim, the Malay Peninsula, and probably in Sumatra and Java. 
Species 2. Clemmys hydaspica, n. sp. noUs. 
Shell . — The evidence for the existence of this species of Clemmys is afforded by 
a nearly complete shell collected by Mr. Theobald in the Siwaliks of the Jhelam 
district of the Punjab, which is figured of one-half the natural size in plate XX. 
figs. 4. 4a. The specimen has lost the anterior and some of the posterior marginal 
scutes, and the hinder half of the plastron, but is otherwise very perfect, and totally 
unaffected by crush. The boundaries of all the costal and vertebral plates, with the 
exception of the 4th vertebra? are distinctly defined ; and some of those of the 
underlying bony scutes are also visible : the plastron is so damaged that only the 
form of the gular and postgular plates can be traced. As the plastron is united by 
bone with the carapace, while the pygal plates are separate, and the general 
resemblance of the vertebral plates to those of the last species is very close, the 
specimen may be referred to the present genus ; and its aflinities may be indicated 
by comparing it with C. sivalensis. 
From that species the present form is at once distinguished by the more vaulted 
and regularly oval carapace, of which the lateral surfaces are highly arcuated, and 
the sides of the anterior portion instead of shelving rapidly are full and convex ; 
the vertebral region is also highly convex, having no trace of the flatness so 
1 This is restored in the figure. 
