A PHYSIOLOGICAL CHARACTER. 113 



Correlations involving the iiidividual reaction-times and the re- 

 productive indices of the mothers were made. For the plus strain 

 this data included every individual of Line 757 whose reaction-time 

 was obtained in making the selections and the reproductive index 

 of whose mother was known — a total of 1,992 individuals. The 

 correlation was —0.000828 ±0.0151. For the minus strain (with 

 1,842 individuals) the corresponding correlation was 0.2154 ±0.0149. 



A negative correlation between vigor and reaction-time would 

 have obvious biological significance, while a positive correlation would 

 not, for it is difficult to see how less vigor would be causally asso- 

 ciated with greater reactiveness (lower reaction- time). Both the 

 correlations for the 757 plus strain are negative, but although sug- 

 gestive, they are much too small to be statistically significant. The 

 correlations for the 757 minus strain are both positive and the 

 second, while not large, is 14 times its probable error. This is not 

 convincing evidence, but on its face it indicates that in the minus 

 strain of Line 757 the greater the vigor the higher the reaction- time; 

 that is to say, a greater vigor is associated with less reactiveness and 

 a lower vigor with greater reactiveness. Some possible corroboration 

 of this anomalous state of affairs is seen under (4) and (5) above 

 (page 110),^ but it is difficult to understand how this relation is 

 possible. 



Further analysis of the data for Line 757 was attempted. Cor- 

 relations were made between age of the mother at the time of pro- 

 ducing her first brood and the reaction-times of her young; between 

 the number in the brood (size of brood) and their reactiveness, and 

 between the age of the mother and the size of her first brood. 



The correlations between the age of mothers and reactiveness 

 of young were -0.03865 ±0.0205 for the plus strain (1,992 young) 

 and -0.01529 ±0.0157 (1,842 young) for the minus strain. Between 

 the number in the brood and their reactiveness the correlations 

 were -0.07619 ±0.0150 for the plus strain and 0.01606 ±0.0157 for 

 the minus strain. 



Other things being equal (temperature, food, etc.), the more 

 vigorous mothers reproduce earher and the less vigorous mothers 

 later. Hence a correlation of obvious biological significance between 

 age of mothers and reactiveness of young should be positive. Both 

 the correlations obtained were negative, but they were very low, and 

 neither was of statistical significance. A biologically significant cor- 

 relation between size of brood and reaction-time should apparently 

 be negative. The correlation obtained for the data for the plus 

 strain was negative and was 5 times its probable error, but it is 

 numerically so small (-0.07619) that it at most indicates an ex- 

 tremely slight relation between size of brood and reaction-time. The 



^Likewise, this fits with the observation recorded under (7) stated on page 111. 



