THE CANID^E, OE DOGS. 273 



what is most remarkable, the same loss is met with 

 in two other species (C. crassirostris, leptorhynchiis), 

 the same modification of the carnassial tooth. It 

 is not known what change in the mode of life 

 caused these same changes in the teeth in several 

 different species. We are content with knowing in 

 what manner so-called new species and genera 

 appear on our earth ; in fact, not suddenly, but 

 by imperceptible shades of difference, which in- 

 crease in the course of thousands of generations, 

 until, finally, what seemed at first an exception to 

 the rule becomes the prevailing state of things. 

 The objection so frequently brought forward 

 that these ' accidental ' deviations would always 

 again be neutralised by crossings with unchanged 

 members of the species if geographical isolation 

 did not come to assist them have no founda- 

 tion whatever; for our discoveries in palaeonto- 

 logy prove the contrary. It must be remembered 

 that the expression * accident ' applies only in so 

 far as it conceals our ignorance of causes and 

 occasions. In many cases for instance, in the 

 transformation of Mastodons into Elephants 

 we can with some certainty determine the altered 

 conditions of food to which the teeth had to adapt 



themselves. 

 25 



