xlviii INTRODUCTION 



spontaneous, while the variations acquired by the action of 

 the environment during hfe — strictly called modifications — are 

 due to the definite influence of surrounding circumstances. 

 But it is of course merely a, f agon de parler to call any variation 

 spontaneous. A variation must have some cause, and that 

 cause must be of physico-chemical nature : this is one of 

 the most fundamental axioms of science. Darwin, when 

 he spoke of variations as spontaneous, did not mean that 

 they rose up like a phoenix out of nothing : he meant that 

 they occurred in a purely fortuitous manner, without any 

 reference whatever to the preservation either of the species 

 or of the individual which developed them. He meant that 

 they had no purpose or design behind them, and in fact 

 used the word in opposition to finalism or teleology in the 

 old sense. Nevertheless, congenital variations must have 

 some physico-chemical cause, and it yet remains to be 

 proved that that physico-chemical cause is of different 

 character from the cause which produces acquired modifica- 

 tions. It may indeed be pointed out that modifications 

 are purposive, or adapted to meet the individual's special 

 needs, whereas variations are fortuitous. But the observa- 

 tion is of no relevance : for natural selection would speedily 

 extinguish all races where the modifications acquired were 

 not adapted to the individual's special needs ; moreover, 

 the distinction between purpose and chance is at best a 

 shallow and subjective one. 



It may further be pointed out that the factor causing 

 an acquired modification acts on a particular organ or tissue, 

 whereas in a congenital variation it acts upon the undif- 

 ferentiated protoplasm of the germ. For inheritance to 

 take place in the former case, the somatic modification 

 would have to be conveyed to the germ-plasm, whereas 

 in the latter case it is impressed there from the first. But 

 this is merely a repetition of Weismann's argument that it 

 is very difficult to see how a somatic modification can affect 

 the germ-plasm in a corresponding specific sense : it is 

 difficult to see, no doubt ; but nature's methods are not 



