380 FOSSIL TURTLES OF NORTH AMERICA. 



to 132 mm. There is a slight notch in the midline, and the lateral angles are rounded off. The 

 notch in the hinder lobe of the plastron is 55 mm. deep. The humero-pectoral sulcus tails 

 about 18 mm. behind the entoplastron. The pectoral scutes are 55 mm. wide at the midline. 



It appears to the present writer that the specimen on which Professor Cope based his H. 

 octonarius belongs really to H. corsoni. Cope states that the shell "is of an elongated form, 

 strongly contracted at the bridge, but expanded and arched above the limbs. " This narrowness 

 of the fossil is deceptive, having, largely at least, been produced by distortion during tossiliza- 

 tion. The right side is crusht inward from 50 mm. to 100 mm. and the hyoplastron and hypo- 

 plastron of one side have been pushed so as to overlap those of the other side. From the mesial 

 border of the left hyoplastron at its hinder edge to the left outline of the shell is about 260 mm., 

 which would make the width of the shell about 520 mm., instead of 437 mm., as given by Cope. 

 This estimate makes the shell but little narrower than that of the specimen first described above, 

 and this can hardly be regarded as H . octonartus. Cope has also stated that the entoplastron 

 is heart-shaped; but this is an error, caused by not carefully distinguishing fractures from 

 sutures. Its form is that seen in H. corsoni. The upper surface of the lip is concave, whereas 

 in other specimens it is commonly convex or flat. 



It is evident that there are great variations in the form of the lip and likewise in the size 

 and form of the notch in the hinder lobe. Furthermore there appears to be no unvarying 

 association of a large lip with a large hinder notch. It is therefore necessary to consider the 

 various forms as belonging to a single species. 



Professor Cope identified some fragments of turtles found in the Wasatch deposits of New 

 Mexico, on Gallinas River, as belonging to H . corsoni. It is more probable that these materials 

 belonged to the species which the present writer has named HaJrianus majusculus. 



Portions of the internal skeleton of this species are rarely found. No skull has yet been 

 secured. No. 1068 of the American Museum of Natural History, a half-grown individual col- 

 lected by Cope, furnishes the greater part of both scapula?, both coracoids, both ischia, 

 both pubes, the acetabular end of one ilium, 2 phalanges, and 2 damaged cervical vertebrae. 



Fig. 476 represents the scapula and coracoid of the left side, as viewed from below; it also 

 shows the left scapula as seen from in front. It will be observed that the scapula is a stout bone, 

 with more than a right angle between the body of the bone and the procoracoid process. Both 

 the body and the process are comprest, except at the ends distal from the glenoid process. The 

 coracoids broaden toward the mesial border, as in the genus Testudo. 



Fig. 477 presents a view of the pelvis from below. The pelvis resembles more that of 

 Testudo, especially that of T . radiata of Madagascar, than that of any of the Emvdidae at the 

 writer's command. The lateral process of the pubis is placed mostly in front of the ischio- 

 pubic foramina. The section of the process is nearly circular in section. From the hinder border 

 of each ischium there projects a pointed process. The extremities of these processes are distant 

 from each other more than half the distance between the lower borders of the acetabula. 

 In fig. 478 the pelvis is seen from the left side, a portion of the ilium being missing. Fig. 479 

 shows the right ilium of No. 6026 A. M. N. H. From its upper end there projects backward 

 a sharp process, which received the outer ends of the two hinder sacral ribs. 



The radius is a stout bone. All the bones of the shoulder-girdle and pelvis indicate an 

 animal that was accustomed to bear its heavy body about on land. 



One of the vertebrae present is the eighth cervical. It resembles closely that of T . radiata. 



Hadrianus tumidus sp. nov. 

 Fig. 480. 



The specimen on which this species reposes was obtained in the upper Uinta deposits of 

 Utah, by the American Museum's expedition of 1895. Its number is 6076. It consists of 

 nearly the whole of the left half of the plastron and the peripherals above the bridge. From 

 the plastron are missing the epiplastron and the entoplastron. The parts present are in their 

 natural relations, so that some idea may be obtained of the appearance of the shell. 



The length of the portion of the plastron present (fig. 480), from the anterior end of the 

 epiplastron to the extremity of the xiphiplastron, is 533 mm. The whole length of the carapace 

 could then have been hardly less than 800 mm. The breadth of the half plastron, from the 



