28 THE VARIATION OF ANIMALS IN NATURE 



directions, while other treatments which have been supposed 

 to produce ' one way ' mutation have given only a very low 

 percentage of mutants. It is possible, therefore, that all these 

 agencies merely alter a mutation-rate which, even without 

 special treatment, would slowly lead to the production of 

 mutations which the treatment makes more numerous. 



Before considering the experimental evidence for alteration 

 of the mutation-rate, there is one other point that must be 

 considered. Those who do not believe in the possibility of the 

 inheritance of acquired characters sometimes write as if the 

 experiments carried out in this connection were designed to 

 investigate the factors controlling the mutation-rate. Thus 

 Sonneborn (1931), commenting on Macdougall's experiments 

 on rats (see p. 40), writes as if Macdougall had produced 

 a series of adaptive mutations (i.e. assuming Macdougall's 

 claim to be technically sound). In our view this is a confusion 

 of the point at issue. The question is rather whether there 

 is not a special process, in addition to mutation, by which 

 characters gradually become inherited under prolonged environ- 

 mental influences. We have to distinguish between (a) induced 

 mutations which are hereditarily stable from the start and do 

 not revert back to type except by a jump as sudden as that 

 by which they arose, and (b) induced modifications which 

 gradually become more intensified and more stable as the 

 stimulus lasts longer and are often slowly lost when the stimulus 

 is removed. Variation of this second category is considered 

 in our next section. At the moment we shall consider only 

 examples of what is clearly induced mutation. 



It was long thought that gene-mutations were spontaneous 

 because they are so rare, so erratic in occurrence, and appar- 

 ently so unrelated to any known factor in the environment. 

 It has been held that mutations observed in animals kept 

 under standard cultural conditions cannot be related to an 

 environmental cause, and the mode of origin of the Drosophila- 

 and Gammas-mutations has been regarded as evidence of 

 first-class importance. It may be noted that a great deal of 

 the evidence relates to mutations in eye-colour and develop- 

 ment (20 per cent, in Drosophila, 100 per cent, in Gammarus) 

 and nearly all the mutants are more or less of the nature of 

 defects. This cannot but arouse suspicion that some dis- 

 turbing external agency may be involved. 



