38 THE VARIATION OF ANIMALS IN NATURE 



than would be expected if the result was entirely due to 

 selection, considering the small amount of elimination in- 

 volved. Further, in another experiment of Durken's (see his 

 fig. 8) selection of non-green pupae did not eliminate the 

 individuals with power to become green, so that there is no 

 reason why reverse selection should have given a pure line of 

 green. We believe a prima facie case has been made out for 

 the inheritance of this modification. 



xiia. Wladimirsky (1928). Colour of pupa of Plutella 

 maculipennis. 



In this moth the amount of black pigment in the pupa 

 case appears to depend jointly on temperature, light and on 

 hereditary constitution. In view of this complicated relation- 

 ship it is rather difficult to come to certain conclusions. 

 Wladimirsky's experiments, which were carried on over twelve 

 generations, gave results not unlike those of Diirken and 

 Harrison, though the author himself does not regard them as 

 evidence for the inheritance of induced modifications, selection 

 being at least partly responsible. The question how far selection 

 was exercised in this case is a difficult one to decide, owing to 

 the heterogeneous nature of the material. 



xiii. Kammerer (1913). Induced colour-change in Sala- 

 mandra. 



These experiments were carried out and the results are 

 presented in such a way as to make it impossible to draw any 

 conclusions as to the inheritance of induced modifications. 

 They were initiated with wild material, which may well have 

 been genotypically diverse. No exact numerical data are 

 given, so that it is impossible to discover whether any form of 

 selection may have been practised. The number of individuals 

 in which the induced changes were supposed to have been 

 inherited is not explicitly stated. 



xiv. Metalnikov (1924). Immunity of Galleria larvae to 

 the Cholera Vibrio. 



The account of these experiments is not sufficiently detailed 

 to enable one to draw any certain conclusions. There is no 

 description of the stock used, no detailed lineages are set out, 

 and the system of mating adopted is not stated. As far as 

 can be gathered, larvae were immunised against the Vibrio 

 and the survivors in each generation were bred from. There 

 was thus a stringent selection in favour of immunity and it is 



