259 
NOTES ON THE CANID® OF THE WHITE RIVER OLIGOCENE.. oo 
The asymmetry of this phalanx is quite marked: its tibial side is straight, while the 
fibular border is quite coneaye, and the dorsal surface is hollowed, or cut away, near the 
distal end, allowing a retraction of the claws, to a limited extent, as may be readily seen 
when the second and third phalanges are put together. This asymmetry of the second 
phalanx is much less conspicuous than in Dinictis, not to mention the modern felines, 
but it is, nevertheless, unmistakable and is certainly one of the most surprising features 
in the whole structure of Daphenus. 
That an animal with the skull and dentition of a primitive dog should prove to pos- 
sess even imperfectly retractile claws is not what our previous knowledge of the early 
carnivores would have led us to expect. So unlooked for was this character, that at first 
I was strongly inclined to believe that the association of the hind foot shown in Pl. XX, 
Fig. 21, with the skull of D. hartshornianus was an accidental one, and that the pes 
must belong to some genus of felines or Machairodonts as yet unknown. Fortunately, how- 
ever, the collection contains a number of other individuals with more or less well-pre- 
served hind feet, and the agreement among them all is complete. Curiously enough, the 
characteristic second phalanges are preserved only in connection with the specimen 
figured, but other specimens haye parts of the tarsus, metatarsus, proximal and ungual 
phalanges, and a comparison of them shows that the reference of this particular hind 
foot is not open to question. The fact that the pes and the skull were found enclosed 
in the same block of matrix corroborates this inference, though, of course, such a fact is 
not of itself entirely conclusive. 
The ungual phalanx is hardly less peculiar than the second, being short, very much 
compressed laterally, and bluntly pointed ; it is very little decurved and has a plainly 
marked groove on the plantar face near the distal end. The narrowness, compression 
and straightness of this claw are in very decided contrast to the heavy and strongly 
decurved ungual phalanges of the modern Canidae, though among the latter there is con- 
siderable variation in these respects. The articular surface for the second phalanx is 
much more strongly concave than in Canis, permitting a greater freedom of motion in 
this joint, as was necessary in order to provide for the retraction of the claw. The sub- 
ungual process is not so large as in the modern genus and does not project so promi- 
nently upon the plantar face of the bone, but it is produced much farther proximally, 
extending beneath the distal end of the second phalanx, when the two are in their nat- 
ural position. The long hood which envelopes the base of the claw is of about the same 
size and shape as in Canis, though the space between this hood and the body of the 
ungual phalanx is narrower. The ungual phalanx of Dinictis is shorter, more compressed, 
but deeper in’ the dorso-plantar diameter than in Daphenus, and has a decidedly larger 
subungual process, in correlation with the more complete retractility of the claws. The 
A. P. S.—VOL. XIX. 27. 
