448 



LORANDE LOSS WOODRUFF. 



This leads to the third point of interest which is the death of 

 the cultures after the cessation of endomictic phenomena. This 

 is true in each instance: At, AEt, and Ot (cf. Fig. 2). The data 

 from these experiments perhaps are not sufficiently extensive to 

 make sure that this is not a case of post hoc sed non propter hoc, 

 but they make it highly probable that endomixis is necessary 

 for the continued life of the race. 



C. 



The data which have been presented in the study of the 

 periodicity of rhythms and endomixis in different races of Para- 

 mecium aurelia when bred under varied and under constant cul- 

 ture conditions, may now be analyzed from the point of view 

 of the periodicity of these phenomena in the same races under 

 varied and under constant culture conditions, since experiments 

 A and B of this series were conducted simultaneously. 



This analysis is readily made by a study of Figs. 3, 4, 5 and 6, 

 which consist of a combination of the graphs already presented 

 from the other point of view. The graphs are paired, one above 

 the other, so that identical five-day periods coincide. For ex- 

 ample, in Fig. 3, period 10 of A is directly above the same period 

 of At. Since the endomictic periods of all these cultures have 

 just been considered, it is only necessary to tabulate them for 

 reference : 



From this table it is apparent that the synchronism of endo- 

 mixis is practically perfect in the same races when bred under 

 different environmental conditions. The first two experiments 

 showed that it was the same for different races when bred 

 under the same environmental conditions whether varied or con- 

 stant. Therefore this experiment corroborates and broadens 

 the conclusions derived from the former ones and shows clearly 



