CiiAP. II. THE TESTUDININA. 357 



as a sub-family, calls them Testiulinina. Tn 1835, Dumeril and Bibron^ admit this 

 group again as a familj^ but change the name to Chersitcs. As this family stands 

 at the head of the series, it needs only to be compared with the Emydoidiv, which 

 are next below. 



As in the Emydoidiv, the body rests upon a broad, flat surfoce, but here it 

 continues broad and full much higher up. There is a general equililnimn through- 

 out the body; and corresponding parts, between a middle transverse section and the 

 two ends, pretty evenly balance one another. The whole form is distinguished by 

 the division of its outlines into three well defined regions : a middle region, includ- 

 ing the organs of digestion, respiration, and circulation, and extending from the first 

 and second pairs of ribs, or, what is the same, from the scapular arch nearly to 

 the seventh pair, and two other regions situated at the ends, including and pro- 

 tecting the extremities and some adjoining organs. The middle region is very high, 

 broad, and long, and forms much the larger part of the body ; its sides arch out- 

 ward IVom end to end, but the cords of their arcs are nearly parallel ; the top 

 is straight, or arched upward ; when straight, it is nearly jiarallel to the lower 

 surface, and when arched, its cord is so. Thus the whole region is quite sym- 

 metrical, and its ends are nearly equal, and very large. The anterior and posterior 

 regions are comparatively short and small, and the curves which close the ends of 

 the body necessarily drop al)ruptly down, and turn abruptly about them, to meet 

 the outlines of the middle region at sharp and well defined angles. 



In most genera, the top and sides of the middle region are only slightly arched 

 from end to end ; but in Psammobates, and in Cylindraspis, they are so much raised 

 as to obscure, at first sight, the distinction between the bulk of the Ijody and the 

 ends. Again, the symmetry of the middle region is somewhat disturbed by varia- 

 tions in the thickness of the shield, and by a somewhat greater elevation of the hind 

 end ; I)ut neither of these modifications rises to importance in reference to the essen- 

 tial characters of the form ; and on examination, the upper surface, divided and spe- 

 cialized as it is, is readily distinguished from the simply arched, bowl-like upper sur- 

 face of the Emydoida?. The regions at the ends very evenly balance one another 

 in bulk, but differ considerably in form ; the front one is shorter and broader 

 at the front end, the other more elongated and narrowed toward the hind end ; 

 the upper surface descends also much lower behind than in front. As in the 

 Emydoidas, the openings about the ends, for the })rotrusion of the extremities, are 

 narrow and small. The carapace is raised considerably above the plastron, a 

 part of its edges turned abru[)tly downward and inward, and joined to the corre- 

 sponding edges of the plastron, which are turned abruptly upward and outward, and 



' El-put. goiicr. vol. -ill, 1835. 



