i68 EXPERIMENTAL EVOLUTION LECT. 



man himself out of wild existing species which in 

 some cases are yet living in the feral state, the mere 

 comparison between the different types which have 

 been evolved out of the original stock, at different 

 times not always very remote is enough to show 

 the influence of domestication. I may add that the 

 divergence might have been even more marked. 

 For we must remember that man has always had in 

 view his own profit, excluding all other considerations ; 

 the selection which he has consciously or uncon- 

 sciously performed has always been directed along 

 the same lines ; and all variations which were of no 

 direct and positive use to him have been swamped. 

 This is well displayed by the cases in which man has 

 been less subservient to utility ; in pigeons, for in- 

 stance, where the production of new types or varieties 

 has been the business of amateurs and dilettanti who 

 pursued no particular useful object, but merely wished 

 to obtain the greatest number of variations, there arc 

 more different types than among other species of 

 which only a small number of useful characters 

 useful to man, not to the animal itself arc required, 

 all other characters being considered useless and un- 

 interesting, and therefore neglected, the result being 

 that the variations are swamped by intercrossing. 

 As Geoffroy Saint Hilaire 1 observed, there is a 

 1 Histoire Natnrelle generate des Regnes Organiques,\o\. iii., p. 133 



