THE LAMARCKIAN PRINCIPLE 67 



take into consideration the possibility of such transference and 

 that even in the case of disease,' direct evidence is difficult 

 to find : that neither exercise nor external conditions can change 

 the nature of any organism, but can only develop characters that 

 are already present. " Every acquired character is," as Weis- 

 mann has put it, " simply the reaction of the organism upon a 

 certain stimulus." No organ can be originated by exercise, 

 though an existing organ may be developed to its maximum. 

 The Lamarckian view of exercise is incorrect : growth does not 

 depend on exercise except under certain circumstances where 

 the association of the two is not difficult to explain. 



However close the correspondence between an animal or plant 

 and its environment, it is easy in many cases to show that the 

 adaptation that has arisen cannot possibly be due to the direct 

 action of the environment. Lamarckism fails to account for the 

 antlers of the stag, for the skill of the worker bee, for the 

 cocoon-spinning of the caterpillar or the remarkable instinct of 

 the Yucca moth. In some cases the Lamarckian principle cannot 

 possibly be applied and we have even in one and the same 

 animal to imagine it working side by side with a principle that, 

 if not actually antagonistic, is yet incongruous. 



The reproduction of lost parts by embryos shows that the 

 organism, even at that early stage, follows its own course and is 

 not the creature of circumstances. The argument that varia- 

 tions may and must be caused by diet is unsound. The organ- 

 ism does not passively submit to the influence of any diet that 

 may be served up to it, but selects in accordance with its own 

 inherited character, and this character has been inherited equally 

 by the germ-plasm. 



If there is any force in what I have urged, Lamarckism fails 

 to make good either of its contentions. First, it cannot account 

 for the origin of variations ; secondly, granting for the sake of 

 argument, that there are such things as acquired characteristics, 

 the evidence of their transmission is unreliable ; the machinery 

 in the organism by means of which they can be transmitted has 

 yet to be discovered. 



