NO. 2187. A NEW PALEOCENE MAMMAL— GIDLEY. 433 



foramen are in about the same relative positions respectively as in 

 the OUgocene Titanotheres. 



The principal differences observed are: (1) The relatively low and 

 less transversely expanded talonid (in Paleosyops the talonid is also 

 depressed, shghtly in the molars, very pronouncedly in the premolars) ; 

 (2) the somewhat greater angulation of the principal cusps of the 

 trigonid; (3) the rudimentary condition of the talonid of p^; and (4) 

 the absence of a third, or heel lobe of the last molar. These morpho- 

 logical differences and especially the last two cited, are undeniably 

 very considerable, yet it seems to me they may all be attributed to the 

 more primitive stage in which we find the teeth of Titanoides. 



In a general way the teeth of Titanoides also resemble those of the 

 Ambliypoda. Especially is this noticable in the form of the premolar, 

 which is in the same stage of development as the corresponding one of 

 CorijpTiodon. Comparing the teeth of Titanoides with those of Pan- 

 tolamhda, Corypliodon, and UintatTierium, however, they show the fol- 

 lowing important differences, and be it noted these same differences 

 are observable in comparing the Titanotheres with the Ambliypoda. 

 In the Ambliypoda the basal cingulum, when present, is confined to 

 the anterior and posterior borders of the tooth crown, continuous in 

 the Fort Union species; trigonid short, its anteroposterior diameter 

 being much less than that of the talonid, trigonid and talonid about 

 equal in length in Titanoides; paraconid small and low in Pantol- 

 amhda, much depressed and vestigial in Corypliodon, and Uintatheriunn 

 relatively high, prominent and well separated from the metaconid in 

 Titanoides} The Ambhypoda are further distmguished (1) by the 

 progressive tendency to depression of the anterior lophs of the trigonid 

 and talonid respectively in Pantolamlda and finally to their ahnost entire 

 disappearance in Corypliodon and the Uintatheres where the teeth 

 have developed two subequal cross lophs with an open valley be- 

 tween; and (2) by the development on the talonid of a low median 

 shelf, or spur, which is incipiant in Pantolamlda, well masked in 

 Corijpliodon and the Uintatheres. 



From the above it will be seen that, whether related to the Titan- 

 otheres or not, the teeth of T. primaevus have no amblypod character- 

 istics, and the line of development marked out is evidently funda- 

 mentally different from that observed in the Amblypoda. 



It is unfortunate that the upper dentition of T. primaevus is not 

 known, since this series is usually more characteristic than the lower, 

 and therefore more reliable for purposes of group determination. 

 However, because of certain definite mechanical relationships between 

 cusps of corresponding upper and lower cheek-teeth in mammals, the 



1 In the Chalicotheres, which also resemble the Titanotheres in some respects, the paraconid is depressed 

 while the entoconid is high and prominent. 



65008°— Proc.N.M.vol.52— 17 28 



