500 THE MAMMALIA OP THE UINTA FORMATIOlSr. 



require correction before we proceed to the comparison with Protoreodon. Cope's 

 description is as follows : " The ectocuneifoi-m is distinct, and much wider than lon^. 

 The mesocuneiform is extero-posterior in position, and the transverse diameters are 

 small. It is produced distally overlapping the head of the second metatarsus. 

 Entocuneiform wanting." In reality the ecto- and mesbcuneiforras are coossified, 

 the line of junction between them being marked by a slight step, or difference of 

 level on the distal surface, which indicates the two facets for the second and third 

 metatarsals i-espeetively, and what Prof. Cope has called the mesocuneiform is really 

 the entocuneiform. If Pi-of. Cope's statements were correct, Oreodon would present 

 the remarkable anomaly of having the ectocuneiform support two digits, while the 

 mesocuneiform supports none at all, or in other words, having the second metatarsal 

 shifted outwards from its ordinary attachment. Besides this, the ectoeiineiforra per- 

 sists, with remarkable constancy in tlie ungulate series, and its absence in such an 

 unreduced pes as that of Oreodoi/, would be very extraordinary. But as we have 

 seen, these anomalies do not exist. In Protoreodon almost exactly the same conditions 

 are found as in the Miocene genus, only here the difference in height between the 

 meso- and ectocuneiforms and consequently the distal step is more pronounced. The 

 entocuneiform is not preserved in any of the specimens, but the facets on the navicu- 

 lar and second metatarsal show that it was shaped very much as in Oreodon. 



The metatarsals are four in number and entirely free ; no trace of the hallux has 

 been found, and in all probability none existed. The metatarsals are of more equal 

 development than in Oreodon, the lateral ones being somewhat largei* in propoi-tion to 

 the median, as would be inferred from the structure of the distal row of tarsals ; in 

 other respects they are closely alike. The second metatarsal has a somewhat latei'al 

 bearing on the ectocuneiform, and metatarsals iii and iv are closely interlocking. 

 The ridge on the distal end of the raetapodials is confined to the palmar surface. 



T^he, 2^cda7iges also resemble those of Oreodon, those of the first row being long 

 and depressed, those of the second i-o\v shorter and with the distal trochlea even less 

 asymmeti'ical than that of Oreodon., showing a less degree of convergence of the 

 hoofs than in that genus, and hence very much less than in any of the recent artio- 

 dactyls except the Tylopoda. The ungual phalanges are, as would be inferi'ed from 

 this, but slightly asymmetrical ; they are higher, narrower, more pointed, and alto- 

 gether more claw-like than in the Miocene members of the group, with the exception 

 perhaps of Merycliyus. 



'Wc only cleai'ly characterized species of Protoreodon is 1*. parvus S. & O., of 

 which the type specimens are the skull and lower jaws figured on PI. I, Figs. 1 and 

 2. This was a very small animal, inferior in size even to Oreodon gracilis. The 



