336 Wortman — Studies of Eocene Mammalia in the 



acters which we may assume are specialized or secondarily 

 acquired, we shall have left an assemblage of primitive features 

 which must have certainly belonged to the ancestors of the 

 Carnivores. These are especially seen in (1) the narrow more 

 or less elongated type of skull, much constricted behind the 

 orbits, (2) the stout, heavy zygomata, (3) the large lachrymal, 

 spreading out upon the face, (4) the prominent sagittal crest 

 terminating in a rather high, overhanging occiput, (5) the 

 relatively large, downwardly projecting paroccipitals, (6) the 

 double condyloid foramen, (7) the peculiar thickening of the 

 posterior border of the palate, (8) the large hatchet-shaped 

 neural spine of the axis, (9) the large size of the lumbar verte- 

 brae as compared with the dorsals, and their tendency in some 

 forms (Opossums) to develop the double tongue and groove 

 articulations, (10) the large deltoid crest and characteristically 

 broad distal end of the humerus, (11) the fusion of the 

 scaphoid and centrale (Opossums, Myrmecobius, and Dasyures), 

 (12) the subequal size of ulna and radius, (13) the large size of 

 the lesser trochanter of the femur, (14) the large size of the 

 fibula and its extensive articulation with the proximal surface of 

 the astragalus instead of upon its outside, and (15) the very 

 primitive form of the astragalus. To these may also be added, 

 (lf>) the small size of the brain, (17) the dorso-lumbar vertebrae 

 formula of 19, and (18) the posterior spreading of the nasals so 

 as to exclude contact between frontals and maxillary in front. 

 With respect to the forms of the teeth, they are with few excep- 

 tions primitive. 



Now nearly all the foregoing characters are actually possessed 

 in varying degrees by some members of the Creodonta, as far, at 

 least, as we know them ; and if the replacement of the dentition 

 were complete instead of partial, and the so-called first molar of 

 the carnivorous Marsupials, which is undoubtedly homolo- 

 gous with a persistent last milk molar of the diphyodont 

 dentition, were replaced by a permanent simpler successor, as is 

 invariably the case where its succession is accomplished, the 

 analogy would then be complete and there would be no diffi- 

 culties whatever in deriving the Creodonts from this source. 

 In like manner the origin of the Insectivora would be trace- 

 able to a form similar to Myrmecobius. That the inflection of 

 the angle of the jaw and the partial repression of the second 

 set of teeth were secondarily acquired, is rendered probable 

 from the recent discovery of Matthew* in which it would 

 appear likely that both these characters have been acquired by 

 the Mesonychidae, among the Creodonts. Just what the 

 Cretaceous Marsupials, when more fully known, will show 

 with respect to these characters cannot now be predicted ; 



*Bull. Amer. Mus. Nat. Hist., January, 1901. 



