224 THE PROBLEMS OF EVOLUTION 



zation, transmit agglutinating ability to their own 

 offspring," and successive generations are able to 

 build up increasing degrees of immunity. But the 

 possibility that these results are due to "a placen- 

 tal rather than a truly hereditary transmission" is 

 a serious difficulty in a study of evolution. I feel 

 that the experiments have evolutionary signifi- 

 cance in either case, for placental transmission is 

 a normal factor in mammalian life, but for the 

 establishment of individual responses as a factor 

 in evolution generally, the possibility of prenatal 

 influence must be eliminated. Mammals are there- 

 fore barred as experimental subjects, and it is very 

 desirable to confine experiments to species in which 

 no association between parents and offspring exists 

 beyond the production of the fertilized ovum or 

 other reproductive unit. 



It is further necessary that the individual re- 

 sponse to a fixed stimulus over an extended period 

 of time, and to a gradually increasing stimulus, be 

 determined. In succeeding generations it is desir- 

 able to determine how the young of treated parents 

 respond to a continuation of the initial stimulus 

 and to an increase of it, with adequate provision 

 against the possibility of natural selection. Are 

 the young able to adjust themselves successfully 

 to a more extreme condition than their parents 

 could meet.^ If so, the influence of individual re- 

 sponse upon the reproductive cells is definitely 



