NOTES ON THE CANIDH OF THE WHITE RIVER OLIGOCENE. 401 
Still another species should be mentioned in this connection. In the American 
Museum of Natural History, New York, are the remains of a small eynoid animal from 
the Uinta beds, which may belong to Cynodictis, or if not, should be referred to some 
closely allied genus. It is important to observe that in the Uinta stage (uppermost 
Kocene or lowest Oligocene) we find that the two canine series, represented in White 
River times by Daphenus and Cynodictis, had already been established. 
Tur PHYLOGENY OF THE CANIDA. 
It seems probable that the fossil genera of this family already known are sufficient 
to indicate to us the main outlines of its phylogenetic history. The problem of recon- 
structing the series is, however, obscured by two circumstances ; first, the variety and 
multiplicity of nearly allied genera, the mutual relationships of which are very complex 
and difficult to disentangle ; and in the second place, by the fact that only rarely do we 
obtain satisfactory material of any of the genera. Most of the forms are known only 
from the skull and teeth, and the skeleton has, so far, been found in but few of the 
species. Cynodictis, Daphenus, Temnocyon and dlurodon are now known from more 
or less complete skeletons, but we shall need to learn far more than we know at present 
concerning the structure of the other genera before we can reach a solution of the many 
problems of canine phylogeny. 
Before taking up the discussion of these phylogenetic problems, it will be conveni- 
ent to establish the order of geological succession in which the various genera make their 
appearance. We haye seen that in the Uinta there appear to be two distinctly sepa- 
rated canine series, one of which is represented by ? Iacis and the other by a genus 
which is very closely allied to, if not identical with Cynodictis. The former series would 
seem to be continued into the White River by Daphenus and the latter, of course, by 
Cynodictis. The latter genus may well prove to be of Old World origin, for in the 
European Oligocene it attains such a variety and fullness of development as it never 
reached in America, although, on the other hand, the American creodont genus MJiacis, 
from which Cynodictis probably took its origin, has not yet been found in Europe. In 
the John Day stage the canine phylum underwent an extraordinary expansion. Daphe- 
nus persisted, but is represented only by a single small species, D. cuspigerus, while the 
series branched out into several distinct and more or less specialized genera, such as 
Temnocyon, Hypotemnodon, Cynodesmus, Enhydrocyon, and perhaps even the little known 
Hycnocyon. No new genera of the Cynodictis series haye yet been detected, but that 
genus itself became differentiated into many more species than occur in the White River, 
and some of these may, on better knowledge, prove to be generically distinct. On the 
other hand, Oligobunis probably represents, as Schlosser has suggested, an immigrant 
