242 The Sporting Dog 



more theorems and present more deductions in 

 an hour than can be proved in fifty years of 

 experiment. 



For it must be remembered that breeding is not 

 mathematics, but merely experiment and empiri- 

 cism ; that, except within certain broad Hmits, 

 nobody can tell where a calculation will land. 

 John H. Wallace, the trotting horse authority, 

 once said that in breeding two and two sometimes 

 make four, but often only three. That phrase 

 condenses the story as far as it has gone. To 

 every breeding formula the answer is : It may be 

 so ; sometimes it is and sometimes it isn't. 



There are two broad rules which may be counted 

 upon. One is that a breed or variety, in propor- 

 tion to the length and thoroughness of its estab- 

 lishment, will reproduce its general characteristics. 

 The other is that nothing can change within the 

 purview of a human generation the essential char- 

 acteristics of a genus. Each characteristic can 

 only be increased or diminished. None will dis- 

 appear and there will not be new ones. For 

 example, every animal of the dog tribe, from a 

 coyote to St. Bernard, has an acute nose and 

 depends much on the olfactory sense for its 

 knowledge of objects. Every one of the tribe 

 also " points " more or less in approaching hidden 

 game, and every one retrieves or carries things 

 about in its mouth. These characteristics are 



