THE THEORY OF NATURAL SELECTION. 299 



extends back to the beginning of the human period ; 

 but, so far as it goes, it would rather support the view 

 that variation is limited and definite both in amount and 

 direction. The 'artificial selection,' for example, which 

 man has exercised in the case of his domestic animals 

 for some thousands of years, has not, so far, resulted in 

 the production of a single new 'species.' New 'varieties' 

 have been produced, but that is all; and we know that 

 these may appear suddenly (as in the instance of the 

 Black-shouldered Peacock), without the direct or indirect 

 action of man at all. Besides, if variation be indefinite, 

 it is difficult to account for the constantly-recurring pheno- 

 menon of the extinction of species a phenomenon which 

 is, on any hypothesis, very difficult to satisfactorily explain. 

 So far as wild animals are concerned there is no direct 

 evidence to show that a single 'species' has come into 

 existence since the beginning of the historical period ; nor 

 is there any evidence to show that during the same period 

 a single wild species has become extinct, except only 

 where its extinction has been the result of the interposition 

 of man. 



The points above enumerated are sufficient to show 

 that there are great difficulties in the way of accepting 

 ' Natural Selection ' as the sole agent in the production of 

 species. That it is one agent, and an important one, 

 is a matter that does not admit of doubt. Under any 

 circumstances, however highly we may rate 'natural 

 selection' as an agent in the production of species, it 

 remains certain that we are still almost entirely ignorant 

 of the causes of the two fundamental laws which have 



