196—42 INDIAN TERTIARY AND POST-TERTIARY VERTEBRATA. 
half of the 1st vertebral, and tlie antero-internal angles of the 1st costal. The 
carapace was evidently much depressed, and indicates an animal about the size of 
large individuals of Batagur hasJca. 
The nuchal plate forms an isosceles triangle, and tapers to a point on the anterior 
margin of the carapace, expanding again very widely on the ventral surface of the 
latter : its posterior margin is indented by a blunt projection from the 1st vertebral 
plate. The latter was apparently subhexagonal, and almost certainly longer than 
broad: its remaining borders are curiously sinuated, and the anterior border projects 
considerably in advance of the costo-marginal suture. There is no trace of a keel 
on the remaining half of the 1st vertebral plate ; and the anterior border of the 
specimen is deeply concave. The extreme length of the nuchal plate is 2*3, and the 
width of the 1st vertebral between the costo-marginal sutures 3 '25 inches. 
Distinctness and affinities. — The concavity of the anterior margin of the carapace 
distinguishes the present specimen from all the recent and fossil species of Batagur. 
The specimen indicates an animal considerably larger than any existing Indian 
species of Glemmys^ Terrapene^ or Cgclemgs ; wliile by the elongated form of the 
1st vertebrals it is distinguished from Testudo. Compared with Geoemyda grandis^ 
there are very considerable points of resemblance : thus in both forms the anterior 
border of the carapace is concave, the nuchal plate is triangular with an excavated 
posterior border, and the borders of the 1st vertebral plate are sinuous. The fossil 
presents no trace of the flat vertebral ridge of G. grandis^ but the writer is unaware 
how mucli this would show if the horny plates were removed ; and it is absent in 
the Siamese G. impressa, Giinther.^ That the fossil is specifically distinct from the 
former species (which attains a length a little short of 18 inches) is pretty evident, 
but the writer inclines to the opinion that the two are allied, and perhaj)s generically 
identical, since the same marked concavity of the anterior border of the nuchal 
scute obtains in G. impressa. More perfect specimens of the fossil form are, however, 
needed before anything definite can be determined ; and the present is one of those 
numerous instances so provoking to the vertebrate palaeontologist where a specimen, 
through its imperfections, throws only a faint glimmer of light on the affinity of the 
animal to which it belonged. 
Family III. TBIONYCHID^. 
Shell much depressed, covered with a soft continuous skin. 
Genus I. EMYDA, Gray.^ 
Shell partly ossified, with a rim of free marginal scutes, and covered with a 
shagreen-like sculpture ; the plastron has five rugose callosities. 
The genus is confined to India, Burma, etc.;® but is represented in Africa by 
1 In the maj ority of species the anterior border of the carapace is not concave, although this is occasionally so ( 0. mgricans) . 
2 Including Cuora, Gray. 3 Yide Gunther, “Eeptiles of British India,” pi. I. 
-i ‘ Proc. Zool. Soc.’ 1882, p. 344. 5 “ Catalogue of Tortoises in British Museum,” p. 46 (1844). 
I 3 The so-called Awyifa sfwcyatosis belongs to Cgclunuslcus, “ Supplement to Catalogue of Shield-Keptiles,” pt. I.p. 111. 
