446 



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



probably merely an enij^hasis of conditions partly foreshadowed in the 

 Permian Theriodesmus. 



From this primitive type the carpus of Clcenodon (p. 304) may be derived 

 by the fusion of the centrale and scaphoid, the close appression of the lunar 

 to the scaphocentrale and the broadening of the limar-unciform contact; that 

 of Dromocjjon may be derived by the suj^pression of digit I, the increase in 

 size in digit II, the broadening of the huiar-uncifonn contact. It should 



not be forgotten that the known 

 Creodont manus pertained to 

 animals that were already sev- 

 eral times larger than the small 

 insectivorous forms from which 

 they were probably descended. 

 The manus in the primitive Oxy- 

 chienidae is unfortunately not 

 known; but it appears probable 

 that the minute insectivorous an- 

 cestors of the Creodonts had a 

 rather slender hand, with flat- 

 tened carpals, a large trapezium, 

 free centrale, a small lunar-unci- 

 form contact, and a slightly di- 

 vergent hallux. 



Fissipedc Carnivores. Mat- 

 thew (1909, p. 388) has shown 

 that the ancestral type of manus 

 is realized in the ]\Iiacidie, in which the scaphoid, lunar and centrale, 

 although closely appressed, are not yet fused into a single bone. The hand 

 vyas primitively of the grasping type and the lunar-unciform contact is always 

 broad. 



Condijlarthra. All authorities now agree that the famous Phenacodus 

 •privioevus of Cope is less primitive as regards size and characters of the 

 dentition than its small Basal Eocene forerunner and presumable ancestor 

 Euprotogonia puercensis. As regards the carjnis, IMatthew (1897, pp. 308- 

 309) has shown that Euproiogonia has an alternating, not serial, carpus, in 

 which there was a distinct oblique hmar-imciform contact, the magnum had 

 "a comparatively small upper surface" and was "extended proximally 

 into a keel separating two nearly equal facets, one for the lunar, the other 

 su])porting the scaphoid, or as may be suspected, a centrale." After showing 

 that in many characters Euprotogonia was intermediate betw^een the later 

 Phenacodus and the Basal Eocene Creodonts, Matthew concluded that the 



Fig. 29. Generalized sclieme of the Basal Eocene 

 Creodont carpus; based on a study of the carpus of 

 Clcenodon, Triisodon, Sinopa, Oxyccna, etc. Many- 

 features of this type were foreshadowed in the 

 manus of Permian Therapsids (p. 434), and accord- 

 ing to the writer's view it is itself prototypal to the 

 carpus of all the Placental orders. 



