HATCHER : OLIGOCEISrE CANIDiE 79 



Measurements. 



Length of premolar-molar series from alveolar border of canine 83 mm. 



Length of molar series 32 " 



Antero-posterlor diameter of sectorial 17 " 



" " " "canine 12 " 



Transverse diameter of canine 8.5 " 



Summary op Crantal and Dental Characters. 

 From the above description and the accompanying figures it will readily appear 

 that the character of the skull, mandible and dentition of Daphosnus is essentially 

 canine, and although possessing a number of creodont characters, more especi- 

 ally in the skull, on the whole it is not very different from that which we 

 might expect among representatives of the Oligocene canidse. I fail to recog- 

 nize those resemblances between the skull of Daphosnus and that of the early 

 Machairodonts referred to by Scott. On the other hand all the creodont characters 

 noted by Scott are confirmed in our material, and in addition I may mention the 

 following : 



1. The unusual development of the sagittal and occipital crests, which is even 

 more marked than in Hysenodon. Owing to the imperfect nature of the material at 

 his command Scott erroneously described the sagittal crest in Daphoenus as being 

 low, it is in fact extremely high and sharp, as shown in Fig. 1. 



2. The widely separated premaxillaries and frontals. 



3. The incurved pterygoids, which show a tendency to arch over the posterior 

 narial tract inferiorly as is completely done by these bones in Hysenodon. 



4. The imperfect development of the auditory bulla. 



5. The nearly horizontal zygomata, which are but slightly arched vertically. 

 The orbit however is more widely open posteriorly than in most modern canidse, 

 and the prominently developed and deflected postorbital process of the frontals seen 

 in Hyxnodon are only very moderately developed in Daphoenus. 



The angular process and general conformation of the mandible as well as the 

 limited area of the symphysial surface by which the rami are imperfectly united are 

 essentially canine rather than feline or creodont. 



While the dentition is as a whole undoubtedly more primitive than that of any 

 single living species of dog, I do not recall any single character which could be 

 regarded as primitive that is not paralleled or surpassed in some of the recent dogs. 

 The character of the sectorials is very like that of those teeth in Canis urostictus, as 

 has already been remarked, and is less primitive than obtains in 0. parvidens. The 

 dental formula, though normal for heterodont mammals, is surpassed by Otocyon, 



