TRIONYCHID^. 529 



This species resembles in some respects A. francisccv. However, the carapace of the latter 

 is somewhat excavated at the miclline in front, the rear is more truncated, and the ridges of 

 the sculpture have broad and rounded summits. The anterior border of the hyoplastron is 

 very concave; that of A. mira is gently sigmoid. Likewise, the mesial border of the hypo- 

 plastron of A . franclscce has a greater number of digitations. 



There are points of resemblance between A. mira and Platypeltis heteroglypta. In the 

 latter, as in others of the genus, the eighth costals are wanting or only feebly developt. The 

 form of P. heteroglypta is different from that of A. mira, the front of the carapace is more 

 rounded and the rear more truncated. The nuchal is relatively longer and there is an 

 unsculptured band along its front. It is observed, too, that the ribs on the underside of the 

 costals are broader in P. heteroglypta than in the subject of the present description. The width 

 of those of the second, third, and fourth costals of P. heteroglypta, taken near the distal end, is 

 24 mm. each. In A. mira, the rib of the second is 22 mm. wide; the third, 20 mm.; the fourth, 

 18 mm. P. heteroglypta was probably a flatter species, but the specimens are much crusht. 



Amyda? tritor Hay. 



Tcit-figs. 687-689. 



AspiJonectes tritor, HAY, Science (2), XIX, 1904, p. 254. 

 Amyda? tritor, HAY, Amer. Geologist, XXXV, 1905, p. 336. 



This species is based on a large and well-preserved skull which was collected by the 

 writer in the Bridger deposits, about 3 miles above the mouth of Cottonwood Creek, east of 

 Fort Bridger, Wyoming. The matrix in which the specimen was found is a calcareous and 

 argillaceous sandstone belonging to the base of the middle third of the Bridger beds. The 

 number of the skull, in the American Museum catalog, is 5913. 



The total length of the skull (figs. 687-689) is 162 mm.; and it has therefore belonged to a 

 large animal. In the American Museum of Natural History there is a specimen of Platy- 

 peltis ferox, the carapace of which has a length of 265 mm. and whose skull has a length of 116 

 mm. If the skull and carapace of A . tritor had the same ratio that obtains in these parts in 

 P. ferox, the carapace of the former would have had a length of about 370 mm. This is some- 

 what less than the length of the carapace of Leidy's Trionyx uintaensis, and considerably less 

 than that of the carapace of Cope's T. scutumantiquurn. The skull under consideration may 

 therefore have belonged to one or the other of these or to some other described species. For 

 the present it is necessary to give a distinct name to this skull and await further discoveries. 



The following are the principal measurements of the skull: 



Millimeters. 



Length from snout to end of supraoccipital spine 162 



Length from snout to end of occipital condyle 123 



Width from outside to outside of quadrats 91 



Width across pterygoids 47 



Width of interorbital space 20 



In general form the skull resembles that of Platypeltis ferox; but the width is somewhat 

 greater, the snout is shorter and less pointed, the interorbital space is broader, and the choanae 

 are more constricted. The facial angle is about the same in the fossil as in the living species, 

 the roof of the mouth being bent down about 25 below the plane of the pterygoids. The 

 sutures between the various bones are more finely dentated than in P. ferox. The skull 

 appears not to have suffered distortion during fossihzation. 



The snout is broad and blunt. The narial opening is 22 mm. wide and only 12 mm. high. 

 The co-ossified premaxillae are small, measuring from side to side only 9 mm. The external 

 surface of the maxilla is perpendicular, and the outline, seen from above, is slightly concave. 

 Seen in profile, the skull is nearly flat behind the orbits, whilst more anteriorly there is a gradual 

 curve to the end of the snout. In P. ferox there is an abrupt change of direction between the 

 hinder borders of the orbits. 



The paroccipitals have about the same relative size and disposition as in P. ferox. The 

 portion of the quadrate exposed between the prootic and the squamosal appears to be some- 

 what narrower than in the living species mentioned. 

 34 



