A PHYSIOLOGICAL CHARACTER. 115 



themselves to explanation as due to differences in relative vigor. For 

 example, during the period from October 1915 to May 1916, inclusive, 

 the plus strains of Lines 795 and 796 were both greatly superior in 

 reproductive vigor to the minus strains of the same lines (figures 

 12, a and b, and 13, a and b), yet during this time the plus strain of 

 Line 795 was much less reactive than its minus strain, while the plus 

 strain of Line 796 was more reactive than its minus strain. 



The curves for Line 740 (figures 14b and 15) are also instructive 

 on this point. In general, the minus strain averaged slightly less 

 reactive than the plus, yet it was on the whole consistently the more 

 vigorous. 



It is obvious that reaction-time and reproductive vigor are on the 

 whole, very slightly, if at all, related in S. exspinosus, and that 

 differences in vigor do not explain the divergence in mean reaction- 

 time so pronounced in Line 757. 



Further, there is no basis for an assumption that the reaction- 

 time differences which arose in Line 757 are due to differences in 

 swimming ability. If the difference in reaction-time in Line 757 

 arose from differences in swimming ability, the results are fully as 

 interesting from a genetic standpoint as though they were due to 

 differences in reactiveness to the light per se. But differences in 

 swimming ability were not the source of the differences in reaction- 

 time between the two strains of Line 757. As compared with the 

 plus strains, individuals of the minus strain less often showed any 

 tendency to react to light, i.e., many more were immobile throughout 

 the time of the test, and individuals of the minus strain which did 

 reach an end of the tank in general did so after more and longer inter- 

 ruptions of swimming than occurred for individuals of the plus 

 strain. It is not believed that anyone who observed the behavior 

 of individuals of these two strains in the experimental tank and in 

 the culture-bottles would countenance any suggestion that the dif- 

 ferences between the two strains were due to differences either in 

 general activity or in swimming ability. 



Special Features of the Reaction-Time Curves. 



The approximation of the curves (figure 18a) for mean reaction- 

 times in Line 757, during 1914 and 1915, at first sight appears more 

 significant than it really is. For 1 two-month period, June-July 

 1914, the minus strain was more reactive than the plus strain, and 

 for the following two-month period the difference was only —2 

 seconds, while for the June-July 1915 period the plus strain averaged 

 only 20 seconds lower in reaction-time than the minus strain; but in 

 the intervening February-March period, for which the curves appear 

 so nearly to approach, there is a difference of 64 seconds, or 16 per 

 cent. That such fluctuations are to be expected occasionally would 

 seem to be indicated by the fact that during the last longer (nine- 



