1 9© I.] Matthew, Additional Observations on the Creodonta. J 



Didymictis of Viverravus. The name Miacidae therefore cannot 

 be used, although the family Viverravidae there proposed by us 

 may perhaps be extended so as to take in Vulpavus and Uintacyon, 

 should these prove to have the Creodont carpus. 



The analysis of the Creodonta on the primary basis of the 

 specialized carnassial will stand thus : 



I. Creodonta Primitiva. No specialized carnassial ; tritubercular upper 

 and lower molars, shear rudimentary or absent. Claws unknown. 



Oxycla:)iidc€. Includes some genera with Lemuroid affinities in the dentition. 



II. Creodonta Adaptiva. Carnassial when present on p^ and m^. 

 Claws, where known, of modern type, and probably carried more or less free 

 of the ground. Scapholunar-centrale early uniting (podials tending towards 

 true Carnivore type). 



1. Post-carnassial teeth disappearing PalcBonictida. 



2. Post-carnassial teeth becoming tubercular Viverravidce. 



3. No carnassials, molars becoming flat-crowned, premolars disap- 

 pearing A 7-ctocyonidcB. 



III. Creodonta Inadaptiva. Carnassial when present not on p^ and 

 my. Claws, where known, blunt, hoof-like, resting on the ground. No ten- 

 dency to union of the carpals (podials tending towards Ungulate type). 



1. Carnassials xa\. Shearing teeth OxvicnidcB. 



2. Carnassials mf " " Hycenodontidce. 



3. No carnassials, teeth with high, round, blunted cusps, upper molars tri- 

 tubercular, lower molars premolariform MesonychidcE. 



There is probably a definite correlation between the hoof-like 

 claws and the lack of union among the carpals in Oxycena, 

 Mesonyx, and Hycsnodon ; but that the division in foot-characters 

 coincides with that based on carnassial specialization is not yet 

 proven. 



The Creodonta Adaptiva early gave rise to the true Carnivora 

 (by union of the carpals), and are not known to have existed 

 later than the Wasatch, although the Middle Eocene of Vulpavus 

 and Uintacyon may, when the carpus is known, prove to be 

 Creodonts. The Creodonta Inadaptiva persisted side by side with 

 true Carnivora until the close of the Oligocene, but none of 

 them can be admitted as ancestral to any true Carnivore, unless 

 the supposition that has successively been raised about each of 

 the three families, connecting them with the Pinnipedia, should 

 be proven by satisfactory evidence. 



Of the modern Carnivore families we may derive the Viver- 

 ridae, Canidse and Procyonidse, and perhaps the Mustelidas, from 



