1880.] DENTAL CHARACTERS OF THE CANID^. 287 



Thooids, side by side, f'roir. South Africa, through Central Asia, to 

 India and North and South America ; (2) the limitation of Otocyon 

 to South Africa ; (3) the limitation of Aguarine Thooids to South 

 America and North-east Asia, including Japan ; (4) the limitation 

 of the most specialized Thooids, namely the Wolves and the ordi- 

 nary Foxes, to the Northern hemisphere; (5) the exclusion of Foxes 

 from South America ; (G) the distribution of CSjon, which curiously 

 resembles that of the Tiger. 



If provinces of distribution were marked out by the Canidse, they 

 would by no means correspond with those generally recognized. 

 There is nothing peculiar about the Australian dog, while the 

 American continent contains within itself all the chief types of Canine 

 animals, except Otocyon. The presence of this form, with its ancient 

 type of dentition, in South Africa is not improbably due to the fact 

 that this region contains the remains of a very old Mammalian fauna. 



VII. The morphological relations of the living Canidae are such as 

 to suggest that they result from the gradual accumulation of small 

 variations in the general direction of increase of size and of differentia- 

 tion of the teeth, superinduced upon a primitive stock which presented 

 the full microdont dentition of Otocyon. 



VIII. Though the paleeontological history of the Canidse is incom- 

 plete, the facts which are ascertained tend in the same direction. In 

 skull and dentition, the older Tertiary Canidoe either, as in the case 

 of Cynodictis, resemble the less-differentiated Canidse, or, as in Am- 

 phicyon, present a third upper molar, such as occasionally exists in 

 Canis cancnvorus. But if, as I suppose, Cynodictis and Amphicyon 

 were preceded by forms having four molars above and below, they 

 have yet to be discovered, as no Eocene mammals with four molars, 

 except Opossums, have as yet been brought to light. 



IX. The primitive stock of the dogs, for which we thus have to 

 seek in older Eocene or earlier deposits, is theoretically required to 

 have been a pentadactyle plantigrade animal provided with clavicles 

 and possibly with bony epipubes. Such an animal, if it existed now, 

 would probably be regarded as an lasectivore with more or less 

 marked didelphous affinities. 



In conclusion I desire to express my thanks to the President, to 

 Dr. Giinther, and to Dr. Rolleston for the ready access afforded me 

 to the abundant materials for the study of the Canidse in the Museum 

 of the Royal College of Surgeons, the British Museum, and the 

 Oxford .Museum, to Sir Joseph Fayrer and to Mr. Wood-Mason, of 

 the Indian Museum at Calcutta, for the great trouble they have been 

 good enough to take in supplying me with specimens of Indian 

 species, and to Professor Peters, of Berlin, for the loan of a skull of 

 V. corsac. 



[P.S. I ought to mention that large additions have been made to 

 this paper since it was read before the Society ; but I have deferred 

 the consideration of the origin and relations of the domestic dogs 

 until the evidence which I am at present collecting is more com- 

 plete. July 4th, 1880.] 



Proc. Zool. Soc— 1880, No. XIX. 19 



