20 GENEEAL CHARACTEEISTICS. 



nothing. Indeed, experience shows ns more and more clearly 

 every year, that do reliance can be placed upon the test depending 

 upon fertile intercommunion, which, especially in birds, is shown 

 to be liable to various exceptions. Still it has been supported by 

 respectable authorities, and for this reason we have given insertion 

 to the above extract. 



GENERAL CHARACTERISTICS. 



In every variety the dog is more or less endowed with a keen 

 sight, strong powers of smell, sagacity almost amounting to rea- 

 son, and considerable speed, so that he is admirably adapted for all 

 purposes connected with the pursuit of game. He is also furnished 

 with strong teeth, and courage enough to use them in defence of 

 his master, and with muscular power sufficient to enable him to 

 draw moderate weights, as we see in Kamtschatka and Newfound- 

 land. Hence, among the old writers, dogs were divided into Pug- 

 naces, Sagaces, and Geleres; but this arrangement is now super- 

 seded, various other systems having been adopted in modern times, 

 though none perhaps much more satisfactory. Belonging \o the 

 division Vertebrata, class Mammalia^ order Ferce^ family FelidcB^ 

 and sub-family Ganina, the species is known as Ganis Jamiliarii^ 

 the sub-family being distinguished by having two tubercular teeth 

 behind the canines on the upper jaw, with non-retractile claws, 

 while the dog itself differs from the fox with which he is grouped, 

 in having a round pupil in the eye instead of a perpendicular slit, 

 as is seen in that animal. 



The attempt made by Linnaeus to distinguish the dog as having 

 4 tai\ curved to the left, is evidently without any reliable founda- 

 tion, as though there are far more with the tail on that side than 

 on the right, yet many exceptions are to be met with, and among 

 the pugs almost all the bitches wear their tails curled to the left. 

 The definition, therefore, of Gani% familiaris caudd {sinistrorsum) 

 recurvatd^ will not serve to separate the species from the others of 

 the genus Ganis, as proposed by the Swedish naturalist. 



