LECTURES IN BIOLOGICAL SCIENCES 



exhibit not only some flexibility, which allows tor the produc- 

 tion of adaptive modifications, but also some tendencies towards 

 inflexibility, which insure that development frequently attains 

 the normal end result even if the circumstances in which the 

 individual develops are somewhat unusual. If a population is 

 subjected for many generations to natural selection for its ca- 

 pacity to undergo a certain developmental modification in an 

 unusual environment, it may acquire such a strong tendency in 

 this direction that, if it is transferred back into the original 

 environment from which it came, the developmental modifica- 

 tion is still produced, even though the precipitating environ- 

 mental stress is now absent. The character which was originally 

 acquired may in this way, after many generations of selection, 

 become independent of the environmental cause. It becomes 

 "assimilated by the genotype." 



It is important to distinguish this mechanism, by which ac- 

 quired characters may be forced by selection to become in- 

 herited, from the earlier theories which supposed that acquired 

 characters may be inherited without selection having played 

 any part in the process. The distinction can be seen very 

 clearly in some of the experiments which have been performed 

 to demonstrate genetic assimilation. For instance, if Drosophila 

 pupae are submitted to an environmental stress in the form 

 of a high temperature for a few hours, many of them exhibit 

 developmental modifications which take the form of changes 

 in the venation of the wings. One modification which has been 

 studied in detail is the partial or complete disappearance of 

 the posterior cross-vein. This acquired character is not in- 

 herited. Even if the environmental stress is applied to a popu- 

 lation for many generations, there is no change in the frequency 

 with which the response occurs, unless selection is also operating. 

 This can be demonstrated most clearly if one makes the experi- 

 ment with an inbred strain. In such a strain there is no genetic 

 variability available, and even if selection occurs it cannot be 

 effective. On the other hand, if selection is applied to a hetero- 

 genous population, which is submitted to the environmental 

 stress in every generation, the frequency of the response can 

 be rapidly increased. It can also be decreased, in spite of the 

 repetition of the environmental stress, if selection is exerted 



32 



