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



smaller members of the family may have been arboreal; and there is con- 

 siderable evidence (pp. 288, 307) for regarding their arboreal adaptations 

 (e. g., divergent hallux, astragalo-cuboid contact) as i)rimitive Placental 

 characters. Other primitive features are: the entepicondylar foramen on 

 the humerus the third trochanter on the femur, the dental formula of |4tt; 

 so that in case the skull and skeleton are more fully made known some of 

 the smaller Periptychidsie may prove to stand almost as near to the general- 

 ized Placental prototypes as do any known Creodonts, Primates or Insecti- 

 vores (including Hyopsodus-). In this connection it is interesting to note 

 that Matthew (1909, p. 512) has recently noted many resemblances between 

 the Eocene Insectivore Hi/opsodu.'^ and the Miochienidtie, which latter are 

 probably of "i)rotungulate" affinities. The astragalus of the Periptychidiie 

 (cf. p. 454 Ednconus) ap])roaches the Marsupial type in several characters: 

 e. (]., the slight development of the trochlear ridges, the broad surface for 

 the internal malleolar region of the tibia, the relations of the sustentacular 

 facet. In other respects it approaches the Creodont type. 



IV. The Hyracoids. 



The small modern Hyraces present a confusing mixture of characters: 

 some cjuite unique among mammals, others primitive, others recalling such 

 diverse groups as Rodents, the jNIeniscotheres, Santa Cruz "Protypotheres." 

 Embrithopoda, Proboscidea, Horses and Rhinoceroses. Although the 

 correct phylogenetic evaluation of these characters is by no means easy, 

 the following principles may assist to some extent. 



(1) It is obvious that some differences between modern and Eocene 

 or Miocene forms may be divergent adaptations, and a small number of 

 distinctive or even primitive characters in common may be of great signi- 

 ficance in indicating interordinal affinities, which may be obscured by a 

 large number of more recently acquired specializations. For example, in 

 " Hijrax" the dental formula (which is a direct derivative of j^^), the 

 complete milk dentition, the possession of an alisphenoid canal, the inde- 

 pendent optic foramen, the retention of interparietals, of a third trochanter 

 on the femur, a free os centrale carpi, abdominal testes, uterus bicornis, 

 structure of the endoturbinals, etc., collectively prove that we are dealing 

 with a primitive Placental and more particularly with an offshoot of the 

 stock which gave rise to the Insectivores, Creodonts, Condylarths, Peri- 

 ptychids, etc.; and these few characters outweigh any number of peculiar 

 specializations which distinguish the Hyracoidea from all other Placentals. 



(2) The greater the number of homoplastic resemblances to tlifTerent 

 orders the more probable is it that the remote ancestors of the Hyracoidea 



