48 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 128 



and lower Pi, and the general, though possibly not invariable, reten- 

 tion of the upper incisors. Moreover, the outer walls of the upper 

 molars of Protoreodon may be slightly more ribbed at the paracone and 

 metacone. This is somewhat more noticeable in the Uinta B than in 

 the Uinta C material. Also, the mesostyle has a more pinched or 

 constricted appearance ventrally than is customary in Agriochoerus. 



Resemblances to Agriochoerus are seen in the basic structure of 

 both the upper and lower molars, structures that would appear to be 

 more conservative and to show generic relationships. In the upper 

 molars it is particularly significant that the posterior crest of the 

 protocone in Protoreodon is directed at about right angles to the an- 

 terior crest of the metaconule -^ and terminates abruptly just short 

 of a union, quite as in Agriochoerus. In Merycoidodon the posterior 

 crest of the protocone, together with the anterior crest of the meta- 

 conule, extend parallel and deep into the valley between the paracone 

 and metacone, somewhat as in Leptoreodon. In the lower molars a 

 crest arrangement somewhat analogous to that in the upper molars 

 persists through Protoreodon and Agriochoerus. Here the anterior 

 crest of the hypoconid is directed toward the posterior crest of the 

 protoconid, leaving a conspicuous pocket anteroexternal to the ento- 

 conid. In Merycoidodon this pocket is not developed inasmuch as 

 the anterior crest of the hypoconid as well as the posterior crest of the 

 protoconid are directed lingually together into the valley between the 

 metaconid and entoconid, again rather as in Leptoreodon. 



The lachrymal in Protoreodon is like that in Agriochoerus. It has 

 nowhere near the forward expansion, nor the characteristic depression 

 seen in Merycoidodon. The orbits are widely open posteriorly, and 

 the posterior narial aperture extends well forward between the molars. 

 Moreover, in the Protoreodon foot the distal phalanges have a some- 

 what compressed or more-acute dorsal surface than in Merycoido- 

 don, resembling in this way, also, the agriochoerid almost clawlike 

 structures. 



I find difficulty in visualizing derivation of Merycoidodon from 

 any of the better-known species of Protoreodon. There seems to be 

 no certain point of origin within the Protoreodon complex, but I am 

 inclined to believe that the line represented by P. minor and leading 

 to the new Myton species, Protoreodon petersoni, is less unlikely than 

 any part of the sequence of larger protoreodonts represented by P. 

 parvus and producing the more distinctly Agriochoerus-like P. pumilus 

 annectens. 



2iWortman (1898) has shown that this is not the hypocone. 



