98 THE PROBLEMS OF EVOLUTION 



guarded by many factors, among which seem to 

 be selective absorption and differential diffusion 

 of dissolved substances through its protecting walls 

 or membranes. Nevertheless unusual influences 

 must occasionally change it materially without 

 destroying it, and along this line experiment may 

 be directed. It may be that the germ plasm can 

 be affected through the somatoplasm, as well as 

 by direct means." ^^ Even Muller, one of the 

 ablest geneticists of the time, makes a concession 

 to the possibilities of the internal environment 

 by saying that mutations may be due to "'acci- 

 dents' occurring on a molecular scale. When the 

 molecular or atomic motions chance to take a par- 

 ticular form, to which the gene is vulnerable, then 

 the mutation occurs." ^^ And Petrunkevitch is cer- 

 tain that " the mutations themselves must be the re- 

 sult of some physico-chemical change and this in the 

 end-analysis is due to environmental factors." 22 



When we consider the implications of this point 

 of view in connection with the normal variation of 

 hereditary characters in response to fluctuating 

 environmental conditions, the question inevitably 

 arises, are the genes involved in these variable 

 responses .f^ We speak of them as somatic characters, 

 acquired characters, or individual responses. We 

 look upon them as cytoplasmic. Are we to under- 



20 Science N. S., Vol. LVI, p. 325, 1922. 



21 Am. Nat, Vol. LVI, p. 43, 1922. 

 ^ Organic Adaptation, p. 109, 1924. 



