1893 ON SELF- ADAPTATION 359 



they grow precisely in the same way. It is only 

 the weakest (from badly nourished seeds) which get 

 crowded out of existence. Here, then, is definite 

 variation without the aid of natural selection. Ex uno 

 disce omnes. 



2. Delimitation of varieties and species by the 

 non-reproduction of intermediate forms. 



It is generally said that if ' good species ' are 

 isolated, the intermediate forms have been killed off 

 by natural selection. I maintain that they were 

 never reproduced. Thus if a has passed by succes- 

 sive generations, a', a", a"', &c, to a" ; a and A n being 

 now only in existence, then a', a", &c, represented a 

 single generation apiece, each offspring being one 

 degree nearer to A n , but could never be reproduced, 

 as the environment was continually acting upon the 

 whole series, urging each generation forwards till it 

 became stable in A a . 



This is precisely what takes place in cultivating a 

 wild plant like the parsnip. Each year the grower 

 selects a slightly improved form, till the required 

 type is fixed. The ' Student ' is now A n , a more or 

 less permanently fixed form, each of the intermediate 

 forms, lasting one year, having ceased to be reproduced. 



3. The geographical distribution of varieties and 

 species by self-adaptation. 



That is, if a number of plants migrate to a new 

 locality with new environmental conditions, half of 

 them may die ; because they cannot adapt them- 

 selves ; the other half may live — change, and become 

 fixed forms, by their power of adaptation. The final 

 conclusion of the whole is that plants require nothing 



