1889.] NATURAL SCIENCES OF PHILADELPHIA. 237 



14. There may or may not be a pollex ; but in the large majority 



of species there is one. 



15. There may or may not be a hallux ; but in the large majority 



of species there is one. 



16. The ungual phalanges may or may not be strongly arched, hut 



there is not so wide a lamina of bone to shelter the base of 

 the claw as in the Felidon.. They are hardly ever (except 

 perhaps in Poiana and Prionodori) completely retractile, and- 

 often not at all so. 



17. The humerus may be without epicondylar foramen. 



18. The dorsal vertebra? never number more than fourteen. 



19. The bone of the penis is small (except in Cryptoproda). 



20. LI is generally and ^ constantly developed. 



21. There is always an ^~; and generally an *L1. 



22. M il is always present and generally large. 



23. The antero-external cusp of Li is generally very small. 



24. ~^\ has almost always a considerable talon. 



2.3. The limbs are short in proportion to the body. 

 To these Ave may add : — 



26. Except in Cryptoproda, the lower sectorial has a well developed 



postero-internal cusp. 



27. The second lower incisor is pushed backwards out of the line 



of the others. 



28. The mandibular rami are tapering. 



29. The metacarpals are less complexly interlocked than in the 



cats. 

 :H>. There are no clavicles. 



On comparing the description already given of Dinictis with the 

 characteristic features of the cats and viverrines here set forth, it 

 appears that, so far as the base of tfie cranium is concerned, these 

 two groups agree together much more nearly than Dinidis does with 

 either of them, the latter corresponding rather to the cynoids and 

 very closely approximating the condition stated above as character- 

 istic of the primitive stock of the Carnivora. In other respects the 

 skull of Dinidis may fairly be said to be intermediate between the 

 cats and viverrines in structure, but the shape of the mandible is 

 peculiarly feline. The dentition exhibits the same intermediate 

 characters, and though it approximates that of the cats in several 

 important respects, it is distinctly less feline than is the dentition of 

 Cryptoproda. One very characteristic feature of the Felidce is how- 



