124 THE NATURE AND ORIGIN OF LIFE 



virulent to the sheep, that is, fit to survive in the sheep, will 

 develop in the interior medium of that animal. On the con- 

 trary, those which, after variation, chance not to be virulent 

 to the sheep will die in the interior medium of the animal 

 because, as the definition of non-virulence implies, they are 

 not fitted to multiply in that interior medium. The virulent 

 bacteridia develop alone and kill the sheep ; and if afterwards 

 we search in the dead sheep's blood, we shall find only viru- 

 lent bacteridia. The sheep, once more, plays the part of a 

 sieve which lets pass only the virulent individuals fitted to 

 prosper within it. 



Here, indeed, is adaptation after the fact. Darwin gave 

 the name of natural selection to this sieve function which 

 keeps back all unfitted individuals and destroys them, and lets 

 pass only the fit. Herbert Spencer, to the same phenomenon, 

 gave the name of survival of the fittest an equivalent of the 

 other. It is clear that the example chosen is a case of pure 

 Darwinism : it verifies, in fact, natural selection by a sum 

 total of conditions differing from that in which the varia- 

 tions to be selected were produced. The variations among 

 which selection is made are, therefore, really chance varia- 

 tions with relation to the sieve that chooses the individuals 

 to survive. 



But it is also a factitious example. Generally speaking, 

 there are not in nature experimenters who take it on them- 

 selves to sift in a living sheep's body the bacteridia which 

 have varied by chance in a culture bouillon or other dead 

 media. We have been studying a case of artificial selection 

 after the fact ; but the case is none the less interesting since 

 it clearly lays down the Darwinian doctrine at a point where 

 the most convinced Lamarckian would be unable to deny 

 that the variations were fortuitous in relation to the selection- 

 apparatus used. 



