BLASTOGENIC VARIATIONS. 151 



average correlation between parent and offspring was 

 thus .335, instead of the theoretical .3. Pearson thinks 

 this high value may be due to assortive mating.* The 

 number of data available for calculating these constants 

 was not very great, so that they cannot be accepted 

 as final, but there seems little doubt of the existence of 

 a small degree of male prepotency. It should be no- 

 ticed, also, that the intensity of the heredity is stronger 

 in the son than in the daughter, and this not only for 

 stature in the English race, but also for cephalic index 

 in the North American Indians. A similar relation 

 was found in respect of eye-colour, hence Pearson con- 

 siders that in man it may be a general rule for the male 

 to inherit more than the female. 



A comparison of other coefficients of correlation 

 seemed to show that the hereditary resemblance be- 

 tween brother and brother, or sister and sister, is greater 

 than that between brother and sister. This was true for 

 stature, head index, and eye-colour in man, and also for 

 coat-colour in thoroughbred-race horses. It would 

 therefore follow that inheritance in a line through one 

 sex is prepotent over inheritance with a change of sex, 

 or that, for instance, a man would resemble his paternal 

 more closely than his maternal grandfather. 



It is not to be imagined that prepotency of the male 

 over the female is in any way a general law. Thus in 

 thoroughbred horses, sire and dam are equipotent in the 

 transmission of coat-colour. In Basset hounds, on the 

 other hand, Galtonf found that the female was pre- 

 potent over the male in transmitting colour in about the 



*L. c.,p. 457. 



fProc. Roy. Soc., Ixi. p. 404, 1897. 



