30 INDIVIDUAL DIFFERENCES AND FAMILY RESEMBLANCES 



same task, namely, 76.00 P.E. 8.3 seconds and 2.9 error per trial. 

 In the multiple choice test the males took 25.76 P.E. 2.5 seconds 

 and 2.8 error per trial, while the females made an average of 52.08 

 P.E. 4.9 seconds and 3.9 error per trial. The final average of the 

 males for all four tests amounted to 33.72 P.E. 4.9 seconds per 

 trial, while that of the females was 68.45 seconds. This shows that 

 the females in the total record took slightly more than twice as long 

 to learn the tests as did the males. 12 



(&) COMPARATIVE DISTRIBUTIONS OF THE MALES AND FEMALES 



In Fig. 8 are given the comparative distributions for the 34 males 

 and 37 females that have been tested in all four of the tasks for 

 which time averages are given in Table IX. The distributions for 

 the males and females, in the initial learning period of the maze test, 

 show that the curve for the males is skewed, for, although there is an 

 equal number of individuals, namely, 12, between and 20 seconds, 

 and 20 and 40 seconds, still from that point on the curve drops 

 quickly, while no individuals are to be found beyond 160 seconds. 

 The distribution curve for the females shows that the mode for that 

 sex is about the same as that found for the males, also that several 

 females are distributed toward the right end of the figure, which 

 indicates that they were the animals that made poor time records. 

 Eight females are to be found between and 20 seconds, and while 

 the curve for this sex drops steadily until 80 seconds is reached, still 

 it rises steadily from that point until it reaches a maximum in the 

 region between 140 and 160 seconds, where five females are to be 

 found. This almost gives the curve a bimodal appearance, but there 

 is no evidence at hand which leads the writer to believe that there 

 were two distinct classes of females among the mice that were tested. 

 The important point to be noted in Fig. 8 is that there were six 

 females that exceeded any of the records made by the males; there 



12 It might be expected that this sex difference would have its effect on the 

 validity of the correlations that have already been given, since they show that the 

 total population was made up of two groups, one, the males, which were relatively 

 quick learners, and the other the females, relatively slow learners, but the accom- 

 panying correlations that were made for the separate sexes show that they fol- 

 low rather closely the combined correlations that have been given in Table VIII. 

 Correlating the last 15 trials of the maze test with the first two trials of the 

 interference test gave the following correlations : ^s = + -60, $s = + .54. The 

 correlations for the same 15 trials of the maze test with the last 10 trials of the 

 interference test gave : ^s = + .35, $s = -f .49. Again the same maze trials when 

 correlated with the retention test gave the following correlations: ^== + .18, 

 $s = + .27. The last sex correlation was made between the last 15 trials of the 

 maze test and the last 23 trials of the multiple choice test, and here the following 

 correlations were found: Js = .26, $s + .16. 



