46 R. T. YOUNG. 



other, show a periodicity differing from both the earlier and later 

 period in the parent line II; while lines IV and VI, derived from 

 II on 4/13 and 4/15 respectively, show periodicities differing from 

 each other and from those of all three preceding lines (II, I la and 

 III). In line XII marked variations in periodicity are shown. 



Since these experiments were not designed to test the peri- 

 odicity of endomixis it is difficult in most of the above cases to 

 determine the cause of the variations noted, whether extrinsic 

 (environmental) or intrinsic. In the following experiments how- 

 ever we have very clear evidence of the influence of environment, 

 in this case temperature, in altering the endomictic period. Com- 

 paring the length of period in lines XI la and b with that in the 

 parent line XII for example it is seen that increasing the tem- 

 perature reduced not only the generation period, but also the 

 time period, from an average of twenty-five days in XII to one of 

 seven in XI la and of twelve in XI Ib. Even more striking 

 evidence of the same character is furnished by experiments 14, 

 15 and 16. 



Notwithstanding Woodruff's emphasis upon the constancy 

 of the time period of endomixis, in contrast to the gener- 

 ation period, which latter was markedly affected by changed 

 environment; it is reasonable to assume that the latter is the 

 more important, indicating as it does the amount of growth and 

 hence the metabolic activity of any period. He himself says 

 (/. c., p. 458), "This is a most surprising result, because such 

 profound reorganization phenomena as are involved in endomixis 

 must have a more or less definite relation to the physiological 

 activity of the protoplasm, the best criterion of which is generally 

 considered to be growth and reproduction as indicated by the 

 division rate. More experiments obviously are needed to resolve 

 this 'time factor' into its significant elements." In the absence 

 of further experiments and in the light of the contradictory results 

 of my own experiments and of many of those of Woodruff himself 

 as analyzed above, I would suggest that the time synchronism, 

 when present, is a result of generation synchronism ; and that in 

 those cases in which the latter, but not the former was modified 

 by environment, the constancy of the former in those cases in 

 which it was constant or approximately so (twelve out of seven- 



