EMYDID^. 



289 



set of peripherals. No fontanels occur anywhere, except apparently a small one on each side 

 of the suprapygal. 



On account of missing parts, the outline of the front of the carapace can not be determined. 

 The shell was broadest across the inguinal region. The hinder border shows a broad shallow 

 excavation. An attempt has been made to map out the areas occupied by the neurals, costals, 

 and peripherals. The position of most of these has been determined pretty satisfactorily. The 

 extent of the peripherals is somewhat conjectural, as shown by the interrupted lines in the 

 drawing. Most of the sutures had to be traced on the inside of the carapace. As there seen, 

 the costals approach to within 45 mm. of the free border of the carapace. 



So far as can be determined, the inguinal buttresses rest against the fifth and sixth costals. 

 The buttresses did not rise to a great height. The ribs were quite prominent on the under side 

 of the costal plates. The rib-heads were strongly developt. The tenth rib formed an abutment 

 for the upper end of the ilium. 



The first neural is missing. The second has the front border wanting, but it was about 

 48 mm. long and 27 mm. wide. The third is hexagonal, 46 mm. long and 48 mm. wide. The 



357- 



358. 



FlGS. 357 AND 358. Gyremys spectabilis. Carapace and plastron of type. Xj. 



357. Portion of carapace. 



358. Plastron. 



limits of the fourth are uncertain; it may have been wider than represented. The plotting of 

 the areas behind the eighth neurals and costals is to some extent conjectural. The thickness 

 of the costals is from 7 mm. to 10 mm.; but the ribs added something to this in the middle of 

 the breadth. 



The scutal areas are even more difficult to determine than are the bones. Enough, how- 

 ever, is determinable to prove that the vertebral scutes were very broad, the second and the 

 third having had a breadth of about 170 mm. The boundaries of the fourth are conjectural, 

 but probably not greatly different from the drawing. Toward the borders of the carapace no 

 sulci are to be detected and no attempt has been made to represent them. 



The plastron (plate 44, fig. 2; text-fig. 358) is well preserved, altho the external layer of 

 bone on its upper surface is eroded away. The lower surface is in better condition, but the 

 bone is much fractured. The total length of the plastron is 425 mm. The anterior lobe is 

 broad and rounded in front; the hinder is somewhat wedge-shaped, with the extremity trun- 



