1010.] Crcodonts not Related to Sparassodonts. 303 



(tympanic process of the alisplienoicl) in Crcodonts, but on the contrary only 

 a true tympanic bulla, which in Dromocyon is inflated and has a tubular 

 meatus. The basi-occipital in the carnivorous Marsupials is short and wide, 

 whereas in placental carnivores it is progressively long. (It is, however, 

 relatively short in certain specimens of Mesonyx.) As to the inflected 

 angle of the jaw, Wortman admits (1901, p. 33(3) that this is probably second- 

 ary in the Mesonychidne, since according to Matthew (1901, p. 34) the angle 

 is less inflected in the Basal Eocene Dissacii.s- and becomes progressively 

 more inflected as we j)ass from this to the Upper Eocene Mesonyx uinte^isis 

 "where it is almost like that of a Marsupial." The accessory posterior 

 palatine foramina have the appearance of nutritive foramina, such as occur 

 in Hyocna, and do not appear to be homologous with the posterior palatal 

 vacuities of Marsupials. The thickened round posterior border of the palate 

 differs from the high transverse ridge with projecting angles which is so 

 characteristic of carnivorous Marsupials and many Insectivores. However, 

 ThyJacynns approximates in this respect the Creodont type. In the Insecti- 

 vores this character seems to be secondary (p. 286). The Crcodonts appear 

 to lack the foramen on the postero-external angles of the palatal ridge which 

 is very characteristic in Marsupials and many Insectivores, but which again 

 is reduced or wanting in Thylacynus. 



The remaining Crcodonts and Carnassidents, according to Wortman's 

 descriptions, are much more advanced toward the normal Placental Car- 

 nivore type and do not show so many resemblances to the Marsupials. 



Supposed relations of the Crcodonts to the " Sparassodonta" {Borhyeeninoe). 



If the Crcodonts and especially the Mesonychidfe were derived from 

 Cretaceous carnivorous Marsupials they might be expected to show some 

 special palseotelic resemblances, to the carnivorous Marsupials (BorhyEeninse) 

 of the Patagonian Miocene, described above (p. 207). But a careful com- 

 parison of the two groups, as figured by Wortman [op. cit.) and Sinclair 

 (1901) respectively, fails to reveal any resemblances which may not be 

 interpreted either as due to similar life-habits or to the retention of very 

 ancient Marsupio-Placental characters. 



On the contrary, there are many differences which may not lightly be set 

 aside as due to differences in hfe-habits and which appear to point to very 

 ancient separation between the "Sparassodonts" and the Crcodonts. A 

 number of these characters have been recorded above (p. 207) but two more 

 may now be added: first, in the manus of Crcodonts and other primitive 

 Placental orders the lunar is large and the magnum is typically small, whereas 

 in all the carnivorous Marsupials, including even Thylacynus and the cur- 

 sorial Santa Cruz forms the reverse is the case. Secondly, leaving out of 



