70 BRITISH FOSSIL REPTILES. 



here descril)cd, and is by no means the only fresh-water tortoise which the clay of 

 Sheppy has yielded ; and since the characters of the present species have not hitherto 

 been defined nor its affinities to the land tortoises been pointed out, the interests of 

 science appeared to me to be best consulted by giving a distinct name to the present 

 species. 



The fossil here described is from the Eocene clay of Sheppy Island, and forms part 

 of the collection of J. S. Bowerbank, Esq., F.R.S. 



Emys l^vis. Bell. Plate 3. 



The only specimen I have seen of this species, I obtained from Sheppy a few 

 months since, and it is now in my collection. It has some remarkable peculiarities 

 which distinguish it, at first sight, from every other species of Emydian, either recent 

 or fossil. 



The specimen is imperfect at each extremity ; the carapace wanting anteriorly the 

 nuchal plate, and posteriorly from the eighth neural plate inclusive. The contour of 

 the carapace is remarkably even, free from all inequalities of surface, and forming, 

 from side to side, nearly a perfect segment of a circle, uninterrupted by cither 

 carina or depression of any kind. The whole surface of the bone also is remarkably 

 smooth. 



The first neural plate (fig. I, s\) is narrow, being not more than two fifths as broad 

 as it is long ; the sides parallel for the first two thirds of its length, then slightly 

 narrowed ; its sides are not interrupted by the costal sutures, as the posterior margin 

 of the first costal plate (j/j/i) joins the anterior part of the second neural. The second, 

 third, and fourth neural plates (.S2 — s\) are of an elongated hexagonal form, and nearly 

 resemble each other ; the fifth, sixth, and seventh (55-^*7) are also hexagonal, but each 

 shorter than the preceding one ; the sixth is narrowed somewhat abruptly, and the 

 seventh still more so, the latter being also shorter than it is broad. 



Although the posterior part of the carapace is considerably broken, there appears 

 evidently to be an interval between the seventh and eighth neural plates ; at which 

 part the posterior portion of the seventh costal plate and the anterior portion of tlie 

 eighth approximate to the corresponding plates of the opposite side, on the median 

 line, without the intervention of the neural plates ; a peculiarity which I do not 

 remember to have seen in any other of the Emydida. 



The first costal plate occupies in its breadth the whole length of the first neural, 

 and the anterior fifth only of the second ; but in consequence of the gradual shortening 

 of the neural plates in the portion of each, posterior to the angle at which the costal 

 sutures join them, the seventh neural receives the costal suture at about the middle 

 of its length. 



