NOTES ON THE CANID.E OF THE WHITE RIVER OLIGOCENE. 347 



The distal end is large and has a well-developed trochlea, which is much more strongly 

 . convex than in Cards and of a different shape, the modern genus having here a trochlea 

 which is more like that of a phalanx than of a typical metacarpal. In Dap Ik en us, but 

 not in Canis, there is a well-defined palmar carina, and the lateral processes for ligamen- 

 tous attachment are more prominent than in the recent type. 



The second metacarpal is much longer and stouter than the first, though very short 

 with reference to the size of the animal and to the length of the other segments of the 

 fore limb. The proximal end is not much expanded transversely, but has a great dorso- 

 j^almar extension, the head projecting much farther behind the plane of the shaft than in 

 Canis. The facet for the trapezoid is less concave transversely than in the modern genus 

 and is of more uniform width, narrowing less toward the palmar side ; the ulnar border 

 rises more above the head of mc. iii and has a more extensive contact with the magnum. 

 Though larger than in the recent Canidce, this contact with the magnum is much smaller 

 than in existing felines, and is of about the same proportions as in the early sabre-tooth, 

 Hophpjhoneus. The combined facets for the magnum and for mc. iii form a broad, 

 curved band upon the ulnar side of the head, which is made slightly concave to receive 

 the adjoining metacarpal. No distinctly marked facet for the trapezium is visible upon the 

 radial side. The shaft is short, weak, of transversely oval section, and is arched toward 

 the dorsal side. The distal end is expanded, and made broad by the large, rugose pro- 

 cesses for the attachment of the lateral metacarpophalangeal ligaments, processes which 

 are much better developed than in Canis. The distal trochlea is of a quite different shape 

 from that seen in the modern genus, being narrower, higher and of more nearly spherical 

 outline, and is demarcated from the shaft by a deep depression, such as does not occur in 

 the existing members of the Canidce. The palmar carina is prominent and thins to a 

 narrow edge. 



The third metacarpal is incomplete in the only manus found in the collection 

 {D. felinus, No. 11425, PL XX, Fig. 17) as it lacks the distal end. The portion pre- 

 served is, however, as long as the whole of mc. ii and the complete bone was evidently 

 considerably longer. The shape of the proximal end is much as in Canis, except for the 

 relatively greater dorso-palmar diameter. The magnum facet is narrow, but deep, some- 

 what concave transversely and strongly convex antero-posteriorly, but less so than in 

 existing dogs. The facet on the radial side for mc. ii is larger, more oblique and more 

 prominent, and is more extensively overlapped by mc. ii than in the latter, and the 

 surface for mc. iv, while not so deeply concave, is larger. When the third and fourth 

 metacarpals are placed together in their natural positions, it is seen that the former rises 

 higher proximally than the latter and has a contact with the radial side of the unciform, 

 which, though narrow, is larger than in Canis. The shaft is somewhat more slender than 



