90 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 1 28 



The upper molars of L. (H.) edwardsi show a somewhat less 

 sharply flexed outer wall with less-prominent styles and ribs than 

 observed in L. marshi. In the lower molars the lingual median cuspule 

 is well developed and the cingulum appears to encroach somewhat 

 farther lingually on the principal inner cusps, but this would not 

 appear to be significant. 



As Stock has indicated in his description of Hesperomeryx, the 

 differences in the premolars between Leptoreodon (H.) edwardsi and 

 Leptoreodon marshi are rather significant but probably not of full 

 generic importance. 



Genus POABROMYLUS Peterson, 1931 



Type. — Poabromylus kayi Peterson. 



Discussion. — Poabromylus was described by Peterson as a camelid 

 and regarded as such by Scott (1945). I am convinced, however, 

 regardless of its large size, that it is a leptotragulid. In the structure 

 of the lower molars it more closely resembles the leptomerycids in 

 general than it does either the oromerycids or Poebrotheriinae. P4 

 has a metaconid developed similarly but relatively not so large as in 

 Leptoreodon, and the anterior crest in both P3 and P4 extends forward 

 and inward as in Leptoreodon without the separate parastylid cusp 

 noticed in the camelids. The heel of P4 has a basin somewhat re- 

 sembling that in Protylopus, but this portion of the tooth is much 

 broader, as it is in Leptoreodon. The talonid portion of P3 is slightly 

 damaged posterointernally, but the posterior half of this tooth is not 

 unlike the camelids ; neither is it unlike Leptotragulus or even the 

 Sespe subgenus Hesperomeryx. 



Of more fundamental significance would appear to be the structure 

 of the molars. In the lower series the inner wall is continuous as in 

 both the leptotragulids and poebrotherines, not deeply interrupted as 

 it is in the oromerycids. On the other hand, the highly crescentic 

 outer cusps have the leptotragulid pattern, not the oromerycid ar- 

 rangement where the entoconid remains relatively isolated except in 

 advanced wear, or the poebrotherine arrangement where the separa- 

 tion of the two outer crests from each other is very distinct and a 

 union is made primarily with the opposite inner cusp to form an 

 irregular ellipse in both the trigonid and talonid portion. Moreover, 

 the heel or hypoconulid portion of M3 is quite like that characterizing 

 Leptoreodon and Leptotragulus rather than the camelids. 



The Poebrotherium-like slenderness of the lower jaw attributed to 

 Poabromylus cannot be verified, as the lower margin of the jaw of 



