402 NOTES ON THE CANIDE OF THE WHITE RIVER OLIGOCENE. 
from the Old World, belonging to the series which leads from the Oligocene Cephalogale 
to the Phocene Simocyon. The dogs of the Loup Fork, with the exception of the aber- 
rant d/urodon, are very imperfectly known and the remains of them which have been 
found are not, according to present knowledge, generically separable from Canis, though it 
hardly seems probable that the modern genus had actually been differentiated so early as 
the upper Miocene, and we may regard it as extremely likely that these supposed repre- 
sentatives of Canis will eventually prove to belong to more primitive genera. None of 
the forms which haye hitherto been found in the Loup Fork beds can be referred to the 
Cynodictis line. 
The mutual relationships between the two canine series, which are already so well 
distinguished in the Uinta, are quite obscure and puzzling, although there is nothing to 
forbid the ‘assumption that both series converge to a common ancestor in the Bridger, per- 
haps the genus JMacis. The Cynodictis series, when we first meet with it, is decidedly 
more adyanced than the other phylum, as is shown in the deyelopment of the skull, the 
reduction of the dentition, the character of the limbs and feet and the digitigrade gait. 
Continuing through the White River age and, so far as North America is concerned, at- 
taining its maximum of deyelopment in the abundance and yariety of its species in the 
John Day, the line apparently disappears and can be traced no farther. Whether the 
series actually died out at the end of the John Day, or whether it continued farther and 
possesses representatives even at the present time, are questions which cannot yet be defi- 
nitively answered. Schlosser (88, p. 247) has suggested that some of the species of Cyno- 
dictis may, perhaps, be of phylogenetic significance in the canine stem, but if so, they 
can hardly be placed in the thooid series, which apparently has no place for them. M. 
Boule (89, p. 321), in an article upon the Pliocene Canis megamastoides Pomel, comes to 
the conclusion that the modern Canidw are diphyletic, and have arisen by a process of 
convergence, the thooids and the bears being divergent groups derived from Amphicyon, 
while the alopecoids and viverrines are descended from Cynodictis. In discussing the 
affinities of the Pliocene form Boule says : 
“Ta description précédente nous montre que le fossile de Perrier se rattache de plus 
pres aux Renards qu’ aux autres représentants actuels de la famille des Canidés. Par 
son crane, le Canis megamastoides ressemble beaucoup le Renard de nos pays. Par la 
forme de sa mandibule, il se place au contraire pres des Renards américains (Canis 
cancrivorus, C. azarae, C. cinereoargentatus) et prés de ? Otocyon megalotis de V Afrique 
australe. Ces espoces, notamment la derniére, sont regardées par tous les auteurs com- 
me des formes primitives. 
“Tout en ratifiant ce premier rapprochement, la dentition presente des caractéres 
particuliers que nous retrouyons en grande partie dans les Cynodictis et Cephalogale du 
Miocéne (p. 527). 
