300 Bulletin American Museum of Natural History. [Vol. XXVII, 



the one hand to Paniolestes (p. 305) and on the other to the ancestral Lipo- 

 typhla, with a dental formula of g^xl' simple premolars, simple trituber- 

 cular upper and tuberculosectorial lower molars and rather small canines; 

 skeleton retaining generalized arboreal characters; tymi)anic region with 

 little or no "alisphenoid bulla." 



(3) That the resemblances between Creodonts and carnivorous Marsu- 

 pials arc due: first, to the inheritance of primitive characters from a very 

 remote, i)crhaps Jurassic, Marsupio-Placental stock; secondly, and perhaps 

 chiefly, to the independent development in the two groups of similar adapta- 

 tions to predatory carnivorous habits. 



Dr. Wortman cites some nineteen characters (1902, ]>. 336) which he 

 regards as "primitive features of the Marsupial Carnivores." These he 

 thinks strengthens the hypothesis of the derivation of the Creodonts from 

 Mesozoic Implacental Metatheria. 



Many of these characters are directly correlated with Hcsh- eating, \)ve- 

 daceous habits: for examjile, the carnivorous dentition, the high sagittal 

 and lambdoidal crests, the stout zygomata, the large downwardly projecting 

 paroccipitals, and the hatchet-shaped spine of the axis. These characters 

 as a class do not necessarily indicate dose kinship between flesh-eating 

 Marsupials and flesh-eating placentals any more than the resemblances 

 between the Wombat and the Beaver indicate close kinship betw^een Diproto- 

 donts and Rodents. Nor -would the lack of such adaptive characters in 

 common in two given groups indicate a wide genetic separation, as is shown 

 by comparison of the smaller insectivorous Dasyures with the Creodont-like 

 Sa7Tophilus of the same family. Another character in the list, namely the 

 fusion of the scaphoid and centrale, seems an unsafe criterion of close genetic 

 kinship, since the same coalescence has probably occurred also in the 

 Edentates (Fig. 23, p. 393), Perissodactyls (p. 394), Artiodactyls (Fig. 26, 

 p. 405) and other groups. 



The remaining points of resemblance listed by Dr. Wortman appear to 

 the writer to be for the most part primitive Marsupio-Placental characters, 

 which are preserved to a greater or less extent in the Insectivora and other 

 orders beside the carnivorous Marsupials and Creodonts. Under this 

 category mav be included the following: 



No. 3. "The large lachrymal spreading out upon the face." In the 

 more primitive members of the Dasyuridiie and Didelphiichie this feature 

 is much less emphasized. 



Matthew- (1906, p. 210) has shown that the facial expansion of the lachry- 

 mal appears to be correlated with the position of the orbits with respect to 

 the cheek teeth. In the modern Carnivores the orbits are further forward 

 and the facial expansion of the lachrymal is reduced. 



