106 INTRODUCTION. 



they be without a known prototype, we might form 

 a system as philosophically admissible as our present 

 knowledge ^vill suffer, or as those already established. 

 We might view the dogs residing nearest the arctic 

 circle, or considered as descended from that quarter, 

 and resembling wolves, as their offspring, more or 

 less modified by the conditions of their being, during 

 a long process of ages, and their fertile intermixture 

 with other more southern species. We might take 

 the more aberrant forms, whose original types we 

 want, from that point in their zones where the most 

 vigorous race of them is known to exist ; and next 

 connect, in a similar manner, the wild and domestic 

 species of the tropics in the Old World, and those of 

 South America, with the group of their congeners 

 in that part of the New. We would in this manner 

 distinguish the large wolf-headed and long-haired 

 species of the north from the smaller jackal-headed 

 tribes of the south ; and find the species belonging 

 to the mastiff race, known to degenerate alike in 

 cold and hot countries, strong, numerous, and t3rpi- 

 cal in central Asia ; spreading eastward into Chinese 

 Tartary and westward to Great Britain, — always in 

 temperate regions ; while the long-nosed greyhound 

 group, in the highest state of procreative vigour and 

 serviceable activity, occupies another belt of the old 

 continent centering in the Persian Taurus, west of 

 Hinducoh, and spreading from China through north- 

 em India, Persia, and Arabia, including both sides 

 of the Mediterranean to Morocco, and to the extre- 

 mity of Europe. 



