88 ON THE HOMOLOGIES AND ORIGIN OF THE 



Bathmodon ; the small third trochanter of the femur is also much like that in 

 Bathmvdon. The osseous horn-cores may be compared with those on the front of 

 Loxolophodon. The knee was probably free from the integument of the abdomen, 

 as in Proboscidians. In all other respects there is no approximation to this order. 



Proboscidia. 



The approximations to other orders in the structure of the feet are only to be 

 seen in the Eocene genera Bathmodon and Eohasileus. The latter, or its ally TJin- 

 tatherium, presents, according to Marsh, but four toes on the hind foot; the ante- 

 rior has five. In the former point we have a resemblance to Hippopotamus; but 

 one of little significance, in view of the radical difterences between the two in the 

 form of the astragalus, calcaneum, and cuboid bones. The former is essentially 

 Proboscidian in all respects, with the addition of a cuboid facet alongside of and 

 behind the navicular, as in Symhorodon ; thus constituting a Perissodactyle char- 

 acter, but leaning to the forms of that order which betray probably the closest, 

 though slight, approach to the omnivorous division of the Artiodactyla. Thus 

 while the EohasiliidcB present the Proboscidian type of feet and molar dentition, if 

 they present any ordinal characters resembling those of the Artiodactyla, they are 

 equally shared by certain extinct Perissodactyla. 



From the hints above furnished, we may regard the succession of modifications 

 of foot structure to be nearly as follows : — 



Ruminantia. Khinocerus. Equus. Elephas. 



I 1 . „.i 



Omnivora. Palseotherium. Hipposyus. 



Symborodon. Palteosyops. Eohasileus. 



Bathmodon. 



V. The Ancestral type of Mammalia Educabilia. 



I trust that I have made it sufficiently obvious that the primitive genera of this 

 division of mammals must have been Bunodonts with pentadactyle plantigrade 

 feet. It therefore follows that Elephas was not the descendant of Eohasileus nor 

 Bathmodon in a direct line, but from some common ancestor with tubercular teeth, 

 through Mastodon. We may anticipate the discovery of such a genus, and believe 

 that it will not be widely removed from the Eocene Hyopsodus, or perhaps Ach(S- 

 ncdon. This will then be the primitive ungulate. 



