CORRELATED VARIATIONS. 87 



organism, and indeed is probably a good deal more so. 

 Supposing that the quality of fertility is correlated 

 with some particular character or characters more than 

 it is with other characters, then it follows that more in- 

 dividuals bearing the character in question will be born 

 and propagate their kind, and so, in course of time, the 

 whole race will be modified in this direction. This prin- 

 ciple has been termed by its discoverer, Professor Pear- 

 son, " Reproductive ' or " Genetic Selection." Its 

 existence as a real factor in evolution depends on the 

 validity of the assumption that the characteristic of 

 fertility is inherited. That this is so, Professor Pear- 

 son, in conjunction with Miss Alice Lee and Mr. L. 

 Bramley-Moore,* has recently proved by a most labori- 

 ous research on inheritance in man and in the thor- 

 oughbred race-horse. Their results show that fertility 

 is undoubtedly inherited from mother to daughter, and 

 also from father to son. It was also found that a 

 woman's fertility is as highly correlated with that of 

 her paternal as with that of her maternal grandmother. 

 In other words, the latent character fertility in the 

 woman is transmitted through the male line, and with 

 an intensity which approximates to that required by the 

 law of ancestral heredity. Again, it was deduced that 

 fecundity in the brood-mare is inherited from dam to 

 mare, and also from grand-dam to mare through the 

 dam. Also the latent quality of fecundity in the brood- 

 mare is inherited through the sire, and by the stallion 

 from his sire. In these latter two cases, the degree of 

 inheritance approaches fairly closely to that required by 

 Galton's law of ancestral heredity, but, in the two for- 



*Proc. Roy. Soc.,lxiv. p. 163, 1899. 



