1910.] Additional Notes on Peralestes, Kurtodon, Panrodon, etc. 195 



Avidth also recalls the conditions in that reptile, and strengthens the inference 

 (p. 193) that in the ancestors of the Trituberculata the upper molars were 

 very wide transversely. The molars also are analogous to those of Microqale 

 among Insectivores. From the characters of the molars and of the high 

 piercing canines it seems likely that there may have been a transverse 

 grinding action of the posterior part of the mandible as in Sesamodon (p. 119) 

 the ]Marsupials and the Zalambdodont Insectivores. 



Dicrocynodon. Mr. Gidley has drawn attention to the fact (190(), p. 100) 

 that the outer cusps and cingulum of the Dicrocynodon molar are comparable 

 with those of Dryohstes and indirectly with those of Triconodon, but he 

 regards the large internal cusp as a secondary heel, "differing widely in 

 character from that of Dryohstes." 



The fact that it is supported by two rudimentary fangs would not disprove 

 its homology with the single cusp in Dryolestes. The transverse ridge seen 

 in Diademodon hrachytiara, which is present in Kurtodon (p. 194) and reduced 

 in Dryolestes, is highly developed in Dicrocynodon. The interspace back 

 of this ridge shows that the protoconid was large while the talonid was also 

 very large. In short, so far as the evidence goes, the Dicrocynodon molar 

 might be derived as readily from the hypothetical type figured on page 185 

 as from the Triconodont type. 



Panrodon. The type specimen (see Osborn, 1907, }). 29, Fig. 2G) should 

 be reexamined in order to make sure that the tooth interpreted by Marsh as 

 a canine is not a broken p^. If it is a canine the reduction in the number of 

 post-canine teeth to 7, taken in connection with the number of incisors, 

 canines and premolars in related genera, would show that the typical dental 

 formula of the Placentals, ;|4xij: appeared in some instances as far back at 

 least as the Upper Jurassic or Ivower Cretaceous. 



Summary of the possible relations of the Trituberculata to the 



Higher Mammalia. 



Amphitherium so far as indicated by the lower jaw and teeth seems to 

 have been the most generalized member of the order. Its dental formula 

 (p. 179) might give rise by reduction to that of either the Polyprotodont 

 Marsu])ials or the Placentals. The lower incisors and canines (if resembling 

 those of related genera), the simple premolars and the lower molars realize 

 the recpiired ancestral conditions for the Polyprotodonts, Creodonts and 

 Insectivores. The probable characters of the upper molars, in so far as 

 they can be inferred from a careful study of the lower molars and from the 

 upper molars of related genera, point in the same direction. The angle of 

 the lower jaw was partly inflected in Amphitherium prevostii but not in 



