174 



CARL DOWNEY LA RUE 



Only the cultures of the fifth and sixth generations were measured. 

 In the fifth generation the culture of the plus line had a greater mean spore 

 length than that of the minus line, but in the next generation the result 

 was reversed, and the culture of the minus line had the higher mean. The 

 summation of the two generations shows that the two lines are not dis- 

 tinct, the difference between them being barely larger than its probable 

 error. The result of this experiment is the same as that for the three 

 preceding experiments, namely, that selection has no appreciable result. 



22 



20 



18 



16 



14 



M 



123456 



FIGURE 10. Graphs showing the lengths of the individual spores selected as parents of 

 the plus and minus lines of experiment 4. P indicates the plus selections; M, the minus selec- 

 tions. Lengths in n are given as ordinates; the successive generations as abscissae. 



The evidence gained from two experiments, in which visible characters 

 were selected, is in entire agreement with that secured from the more 

 extensive experiments in selection according to progeny, and all the experi- 

 ments, whatever the method and strain used, consistently show that 

 selection within pure lines of Pestalozzia is entirely ineffective, but that 

 rarely mutations occur which are significantly different from the parent 

 line. 



DISCUSSION 



JENNINGS (1916, 1920) considers that if the experiments which have 

 been supposed to demonstrate the negative result of selection had dealt with 

 characters which were less likely to be influenced by degree of maturity 

 and environmental influences, had been more carefully conducted, and 

 had involved a larger number of selections, they would likely have shown a 



