500 GEOLOGICAL SURVEY OF THE TEERITOEIES. 



are very much elongate and resemble those of the fore limb. There is, 

 on the external, a proximal excavation for the rudiment of the fifth 

 metatarsus, like that which contains the fifth metacarpus. Probably 

 there is a similar second, but the indications are lost. The distal end 

 of the fibula is not co-ossified with the tibia. 



The above analysis determined some interesting relations of this 

 genus. The cervical vertebrae indicate affinity to the Camelidce, and 

 there is nothing in the remainder of the structure to contradict such 

 relation. The separation of the os trapezoMcs is found in the camels, 

 and very few others only among Euminantia ; but, in the presence of 

 the trapezium Fo'ebrotlierium shows relationships to more ancient 

 types, as Anoplotlieriidce, &c. The reduction of the digits to two and 

 the separation of the metacarpals point in the same direction ; indeed, 

 the number of carpals and metacarpals is prcisely as in Xiphodon. But 

 the mutual relations of these bones are quite different from what exists 

 in that genus, and is rather that of the Camelidce and other ruminants, 

 or what Kowalevsky has called the " adaptive type." This author has 

 seen, in the genus GeJocus, Aym., from the Lowest Miocene, or Upper 

 Eocene of France, the oldest ruminant and the probable ancestor 

 of a number of the types of the order ; but among these he does not 

 include the Camelidce. The present genus is a more generalized type 

 than GeloGus ; in its separate trapezoid and distinct metacarpals, it rep- 

 resents an early stage in the developmental history of that genus. It also 

 presents affinity to an earlier type than the Tragulidce, which sometimes 

 have the divided metacarpals, but the trapezoides and magnum co-ossi- 

 fied. In fact, Foehrotheriitm, as direct ancestor of the camels, indicates 

 that the existing Buminantki were derived from three lines represented 

 by the genera Gelocus for the typical forms ; Po'ebrotlierkmi for the cam- 

 les ; and Hycmiosclius for the Tragulidw. The first of these genera can- 

 not have been derived from the second, on account of the cameloid 

 cervical vertebrae of the latter, and all three must be traced to the 

 source whence were derived, also, the A7io])lo]teriidce, and perhaps the 

 little-known Dichodontidce. 



The two distinct metacarpals, separate trapezium and trapezoides, 

 cameloid cervical vertebrse, and dentition characterize this type as a 

 peculiar family, which may be called FoehrotJieriidce. The genus from 

 which it takes its name was originally referred by Leidy to the Camelidce. 

 I have been unable to detect any characters by which Protomeryx 

 liallii, Leidy, can be placed in a distinct genus from the present one. It 

 rests on a portion of a lower jaw of an individual somewhat larger than 

 the usual size of P. vilsonii. 



PoEBROTHERiUM viLSONii, Leidy, loc. cit.,]).14:l ', Copc, Bull. U.S. Geol. 

 Surv., No. 1, 24. 



Several ^individuals procured. The size was about that of a sheep, 

 but the limbs and neck were much longer. The latter resemble, in their 

 slender proportions, those of the XipJiodon gracile of the Paris Eocene, 

 and exceed those of any of their contemporaries in this respect. 



Measurements. 



M. 



Length of coutiiiuous sis molars 0.058 



Depth of mandible at M. 2 022 



Length of atlas, (on centrum) -■ 035 



Length of third cervical vertebra 056 



Width of centrum behind 020 



Depth of centrum behind , 015 



