ORIGIN OF ORGANIC FORMS 313 



and its cause. We then call the variation accidental, 

 though science cannot admit accidental phenomena in 

 the literal sense of the word. We call it accidental 

 in so far as its primary cause is concealed from us. 

 The difficulties we meet in trying to unravel the con- 

 nexion between a variation and its cause depend mainly 

 upon two circumstances : in the first place it is already 

 rather late to look for a cause when the variation has 

 already manifested itself: secondly, external influences 

 very seldom produce a lasting effect upon a fully 

 developed organism ; this probably happens more 

 frequently in the case of organisms in the embryonic 

 stage of their development, for it is obvious that the 

 earlier the influence is exerted the more far-reaching its 

 consequences. The influence of any disturbing factor is 

 generally very strong in the earliest stages of develop- 

 ment. This can be proved, for instance, by the fact that 

 it is impossible to propagate some sports otherwise than 

 by asexual reproduction, because the influence of the 

 second parent in the process of sexual reproduction is 

 strong enough to affect the whole organism to such an 

 extent as to interfere with the transmission of the 

 desired character. One of the secondary causes of 

 variation must be looked for in the action of a law called 

 the correlation of growth, which consists for instance in the 

 excessive development of one part being accompanied by 

 the under-development of another. Having at a given 

 time only a certain quantity of nutrient substances to 

 dispose of, Nature, as Goethe says, ' in order to be lavish 

 in one direction is forced to economise in another.' 

 Lastly, once formed, the organ is apparently able to 

 go on developing precisely in so far as it is being 

 used. 1 



Variations when occurring under the influence of 

 physical conditions may equally well be useful or harm- 



1 It has not yet been explained what variations obtained as the result 

 of exercise are inherited, and which are not. 



