98 THE AMERICAN BOTANIST 



disease resistance keeps pace with that increased infection 

 without the use of selection. 



When seeds from non-resistant plants are sown on dis- 

 ease infected soil, the fact that some die and others survive is 

 not evidence that the survivors had greater inherent resistance 

 than tJiose which died. Soil which is supposed to be uni- 

 formly infected is in fact spotted. Disease producing bac- 

 teria are in colonies and are not uniformly distributed. Plants 

 which fall on heavily infected spots are killed, while those on 

 lightly infected spots survive. Under these conditions, some 

 will live and some will die if they all have the same degree of 

 resistance. Even when different seeds have inherent differ- 

 ent degrees of resistance, the most resistant ones may die by 

 reason of falling on heavily infected spots, and the least re- 

 sistant ones may survive by reason of falling on lightly in- 

 fected spots. The appearance which is called "selection" 

 comes from different degrees of infection, and not from dif- 

 ferent degrees of resistance. 



It may be assumed that uniform infection gradually in- 

 tensified w'ould result in selection in the strict meaning of the 

 term, and that ordinary infection in a large field has parts 

 which are near enough to the uniform to bring about that re- 

 sult. That this is not so may be seen by referj-ing to the char- 

 acteristics of living protoplasm wherever it has been f6und. 

 One of tlie most fundamental of characteristics is that the 

 powers of protoplasm increase when exercised, and that such 

 development extends indefinitely under continually increasing 

 exercise. While there is no known limit to the development 

 of powers by exercise, the observed development is not uni- 

 form under uniform exercise. Rapid development, slow de- 

 velopment, and apparrent cessation of development are inter- 

 spersed along a period of activity. In studying mental de- 



