SELECTION WITHIN PURE LINES OF PESTALOZZIA 



163 



lines parallel each other in a remarkable manner but the differences 

 between them are not great. In one case (generation 11) the two lines are 

 so nearly coincident that parent cultures certainly different for appendage 

 length, could not be secured from the plus and minus lines. The range 

 of variation in appendage lengths did not consistently increase from gen- 

 eration to generation as they should if selection were effective but under- 

 went successive increases and decreases as the two lines drew nearer to- 

 gether or deviated more widely. The differences between the plus-selected 

 parent and the minus-selected parent vary from generation to generation, 

 but are large enough to be of significance except in the one case mentioned. 

 Figure 7 shows the means of the generations for the whole experiment 

 and from this figure it is at once apparent that the two lines of descent did 



TABLE 5 



Mean lengths of spore appendages of cultures of strain 29, grown prior to beginning experiment 2. 



Measurements in fi. 



not become more widely divergent as the number of selections increased. 

 Instead the same upward and downward swings appear as were seen in 

 experiment 1, but in this experiment the two lines parallel each other in 

 these fluctuations in a surprising way. 



The means of the generations of each line and the mean of means of 

 each line for the whole experiment are presented in table 6, while table 13 

 (in the appendix) gives the data for all the cultures grown during the course 

 of the experiment. After examining figure 7 one is not surprised to find 

 that the differences between the three different experimental lines are of no 

 significance. In one case the difference is less than its probable error. Each 

 case where the plus line has grown longer and the minus line shorter has 



GENETICS 7: Mr 1922 



