DARWIN AND THE ORIGIN OF SPECIES 223 



original race (near the surface of our sphere), then it will in 

 time spontaneously lose the tendency to relapse, and acquire a 

 tendency to vary outside the sphere. What is to produce this 

 change ? Time simply, apparently. The race is to be kept 

 constant, to all appearance, for a very long while, but some 

 subtle change due to time is to take place ; so that, of two indi- 

 viduals just alike in every feature, but one born a few thousand 

 years after the other, the first shall tend to produce relapsing 

 offspring, the second shall not. This seems rather like the idea 

 that keeping a bar of iron hot or cold for a very long time would 

 leave it permanently hot or cold at the end of the period when 

 the heating or cooling agent was withdrawn. This strikes us 

 as absurd now, but Bacon believed it possibly true. So many 

 things may happen in a very long time, that time comes to be 

 looked on as an agent capable of doing great and unknown 

 things. Natural selection, as we contend, could hardly select 

 an individual because it bred true. Man does. He chooses for 

 sires those horses which he sees not only run fast themselves, 

 but produce fine foals. He never gets rid of the tendency to 

 revert. Darwin says species of pigeons have bred true for 

 centuries. Does he believe that it would not be easier by selec- 

 tion to diminish the peculiarities of the pouter pigeon than to 

 increase them ? and what does this mean, but that the tendency 

 to revert exists ? It is possible that by man's selection this 

 tendency may be diminished as any other quality may be some- 

 what increased or diminished, but, like all other qualities, this 

 seems rapidly to approach a limit which there is no obvious 

 reason to suppose ' time ' will alter. 



But not only do we require for Darwin's theory that time 

 shall first permanently fix the variety near the outside of the 

 assumed sphere of variation, we require that it shall give the 

 power of varying beyond that sphere. It may be urged that 

 man's rapid selection does away with this power ; that if each 

 little improvement were allowed to take root during a few 

 hundred generations, there would be no symptom of a decrease 

 of the rate of variation, no symptom that a limit was ap- 

 proached. If this be so, breeders of race-horses and prize 

 flowers had better change their tactics ; instead of selecting the 

 fastest colts and finest flowers to start with, they ought to begin 



