NATURAL SELECTION 217 



from the study of structure will be unconvincing (cf. also p. 300). 

 For this reason evolutionary speculation may be said to be 

 halting on the very threshold of its field of inquiry. Never- 

 theless, the following statements seem justified : (1) that 

 much variation in animals is seen definitely to be of the 

 fluctuational order, and to be of no evolutionary importance ; 

 (2) that some mutations arise with no apparent cause in the 

 environment ; (3) that a limited number are known to be 

 related to extrinsic factors ; and (4) that factorial combina- 

 tion is responsible for a good deal of variation. 



It is necessary to return for a moment to the question we 

 have posed on pp. 28-9. We drew attention there to the highly 

 suggestive nature of some of the recent work on induced heri- 

 table variation, and we restated the doubt originally expressed 

 by Robson (1928, p. 254), whether ' germinal ' change is likely 

 to be a purely spontaneous phenomenon and entirely inde- 

 pendent of external stimulus. We freely admit that certain 

 gene-mutations appear to arise without any specific external 

 stimulus and in the present state of our knowledge must be 

 treated as ' spontaneous.' The recent work on the induction 

 of mutation by raising the temperature of cultures or exposing 

 them to radiation cannot be said as yet to explain the bulk of 

 ordinary mutation, and we regard the ultimate causes of gene- 

 mutations as highly problematical. With this uncertainty in 

 the background, it cannot be said that evolutionary inquiry is 

 ready to answer in a very authoritative fashion the questions 

 which it raises. 



Of course it may be argued that, even if gene-mutations are 

 ultimately due to external stimuli, we have still to account for 

 their spread and multiplication. It is, indeed, theoretically 

 possible that a local population may be transformed en masse 

 by the action of the environment. There is some slight evidence 

 in favour of this, but it is not enough to convince us that this 

 is a very important factor in evolution. Moreover, the appeal 

 to a general environmental modification of a population 

 involves us in a number of difficult questions (Robson, I.e. 

 p. 174). 



Even if we begin by admitting the possibility of some 

 induced variation being hereditary, and thereby acknowledge 

 that the general situation is obscured by doubt on a very 

 crucial issue, it is still possible to discuss a part of this question 



