106 MEMOIRS OF THE CARNEGIE MUSEUM 



especially with the very Kmited osteological material at his command with which to 

 make the necessary comparisons. From the many feline characters which have been 

 shown to occur in DajDhosnus and other Oligocene dogs and from the many skeletal 

 characters held in common by these dogs and the contemporaneous Machairodonts 

 there would seem but little doubt that both these groups as well as the true felidse 

 came from a common ancestral stock which most likely belonged to some member 

 of the Creodonta adaptiva of Matthew in the early Eocene. 



In a paper entitled " The Ancestry of Certain Members of the Canidse, the 

 Viverridse and Procyonidse," published as Article VI. of Vol. XII. of the American 

 Museum Bulletin, Wortman and Matthew have devoted a great deal of attention to 

 the phylogeny of certain genera and species of recent canidse. With characteristic 

 ingenuity and commendable patience they have developed and proposed two lines 

 of descent, one for the Dholes, and another for those South American foxes with 

 reduced premolars, Canis urostictus, parvidens, etc. Considering that these theoret- 

 ical phylogenies are, for the most part, based on tooth-structure alone, and that 

 throughout vast periods of time we have not yet discovered the intermediate forms 

 which must of necessity have existed, such phylogenetic work while extremely 

 interesting, not to say alluring, must necessarily be quite provisional and theoretic, 

 and it would therefore seem to the present writer a little premature to say that 

 "The Dhole or Red Dog of India (Cyon) can be confidently considered as the living 

 representative of the John Day genus Temnocyon " when not a single intermediate 

 form is known from the close of the Oligocene to recent times. By this I do not 

 mean to deprecate such attempts at working out the ancestry of living animals, 

 which is and should be one of the chief aims of the paleontologist, but rather to 

 emphasize how fragmentary our actual knowledge of the ancestors of the modern 

 canidse really is. Of the extinct canidse of North America we are fairly well 

 acquainted with the osteology of but four genera, Vulpavus of the Eocene, and 

 Cynodictis, Daphoenus and Temnocyon of the Oligocene. 



The plates accompanying this paper were drawn by Mr. F. von Iterson, the text 

 figures by Mr. Sidney Prentice. 



Carnegie Museum, July 15, 1902. 



