130 Permanence and Evolution. 



supposed to rest on positive evidence ; yet, on 

 the other hand, when they attempt to account for 

 the origin of the differences which divide closely 

 allied species, while, no doubt, the thing con- 

 jectured is less startling to the imagination, and 

 is nearer to what is commonly supposed to be a 

 known fact, the origin of domestic breeds from 

 one stock, yet, at the same time, it is found 

 simply impossible to give any even plausible 

 theory of the mode of development on the doc- 

 trine of natural selection. 



Take, for instance, the Fringillidae or the 

 Columbae. What we see is a number of 

 forms, closely similar in general organisa- 

 tion and type, differing in a great number of 

 minor points, apparently unconnected with 

 each other, or with any teleological purpose 

 these forms existing on the same area, with 

 much the same habits, food, etc. I think it is 

 impossible to overrate the enormous difficulties 

 this state of things offers in the way of pure 



