THE ORIGIN OF VARIATION 37 



x. Kammerer Ciona ; truncation Repeated by Fox 

 (1923) of siphons (1924), who did not 



obtain the same 

 result. 



xi. Tower (1906). Colour changes induced in Leptinotarsa 

 by alterations in temperature and humidity. 



This very extensive series of experiments brings to light the 

 fact that, if the stimuli were applied to the eggs or larvae, 

 little or no change was effected. If they were applied to the 

 pupae, changes were induced which were not inherited. But 

 if the adults were exposed to the stimuli during the period of 

 maturation, the offspring alone were modified and the effects 

 were inherited. Tower's results have been very adversely 

 criticised, unfortunately on the score of the actual accuracy of 

 the results claimed. It is difficult to judge whether the criti- 

 cisms are finally destructive or inspired by prejudice. The 

 work has not been repeated, so that in all fairness it cannot be 

 used as evidence. 



More recent work on the effects induced by temperature 

 and humidity in insects suggests that Tower's results must be 

 at least very exceptional, though sublethal temperatures may 

 induce mutation (p. 29). 



xii. Diirken (1923) and Harrison (1928a). Colour changes 

 in PzVm-pupae. 



Diirken studied P. brassicae ; Harrison, P. napi. In the 

 former species under normal conditions about 4 per cent, of 

 the pupae are green, in the latter about 2 1 per cent. If the 

 pupae are exposed only to orange light a much higher per- 

 centage becomes green — in P. brassicae, 69 per cent, in the first 

 generation, 95 per cent, in the second ; in P. napi, 93 per cent, 

 and 95 per cent, respectively. In Diirken's experiment 

 offspring of the first generation reared in normal light gave 

 41 per cent, green. Harrison's broods of the second genera- 

 tion gave 100 per cent, green in the third generation and 

 58 per cent, in the fourth. In both experiments the initial 

 stock may have been somewhat mixed and there was con- 

 siderable mortality, which may have involved some selection. 

 Further, in both experiments only green pupae were bred from 

 to obtain the pupae which were returned to normal conditions. 

 In both experiments, and especially in Harrison's, the in- 

 herited modification occurred in far more of the offspring 



