70 A REVISION OF THE COTYLOSAURIA OF NORTH AMERICA 



can be determined. These figures are in part estimates, as the upper jaws are 

 imperfect in some specimens. 



mm 



No. 4350 Am. Mus., D. molaris '4 



No. 4349A1H. Mus.,D. phaseolmus H 



No. 4839 Am. Mus., D. phaseolinus '4 



No. 1078 Univ. of Chicago Diadectes sp 141015 



In all members of the genus the posterior tooth is smaller and more simple 

 than the other cheek teeth. The next seven or eight have the form characteristic 

 of the family and are described in detail under the various species. The anterior 

 teeth of the maxillary series become smaller and the root rounded in section, lead- 

 ing Cope to the suggestion that some specimens had a distinct, enlarged canine. 

 The posterior premaxillary teeth were of the same form, cones slightly flattened 

 transversely. The three anterior incisors were larger and elliptical in section, but 

 attrition speedily produced flat, concave faces on the outer side of the upper teeth and 

 corresponding surfaces on the inner side of the lower. These teeth were as emi- 

 nently fitted for cropping herbage as the posterior teeth were for crushing it. It is 

 to be noted that only in old specimens is there much evidence of wear on the cheek 

 teeth; the tubercular and cuspate tops, which did not fit accurately upon those of the 

 opposite jaw, are singularly well preserved. This indicates the absence of any 

 grinding action of the jaws and coincidentally the lack of use of hard or tough 

 vegetation as food. Perhaps the animals fed upon succulent water plants or upon 

 soft, easily crushed shellfish. The wear of the teeth produced flat faces on the inner 

 cusps of the lower teeth and in advanced stages a deep triangular groove on the 

 inner side of the middle cusp. The same effect was produced on the outer half of 

 the upper teeth. 



All the teeth were placed in well-defined sockets. The root is fairly long and 

 shows a distinct longitudinal corrugation of the surface (plate 6, fig. 2). Several 

 specimens show successional teeth rising to displace functional ones. 



The lachrymal was elongate, reaching from the orbit to the nares. It sent a 

 process upward from its posterior end, partly separating the prefrontal and the 

 nasals. Just within the edge of the anterior part of the orbit is a good-sized lachry- 

 mal foramen. 



The prefrontal formed more than half of the upper rim of the orbit and did not 

 extend very far anterior to it. 



Lower surface of the skull: The basioccipital is separated from the basisphenoid 

 by an indistinct suture. The basisphenoid ends abruptly in front, just anterior to 

 the basipterygoid processes. There is no presphenoid rostrum apparent on the 

 lower surface of the skull; a short and very stout process, excavated longitudinally 

 on its lower surface by a deep pit, is attached to the anterior end of the basisphe- 

 noid. In other specimens the anterior end of the basisphenoid is perfectly smooth. 

 The lower surface of the basisphenoid is penetrated in the median line by a single 

 foramen; in the Pelycosauria and Rhyncocephalia, generally, there are two foramina 

 in this place for the passage of the internal carotid arteries. In other specimens 

 of Diadectes I have been unable to detect any foramina at all in this position, but 

 its apparent absence was probably due to the condition of the specimen. Poste- 

 riorly the basisphenoid widens and the posterior edge is continued backward as a 

 sort of flange overhanging the basioccipital. The flange is excavated by a notch 

 in the median line, and the notch is divided by a narrow partition which is continued 

 backward as a narrow ridge on the surface of the basisphenoid and then divides, 

 probably following the basioccipital-basisphenoid suture. On either side of the 



