EVOLUTION 47 5 



understand how any selection could have changed a type 

 unless heredity had an a priori existence. Shortly, those 

 who, like Mr. Sedgwick,'^ argue for an early period, not only 

 of great variation, but of little or no heredity, not only 

 destroy the basis of the Darwinian theory, but can never 

 have examined the perfectly definite quantitative determina- 

 tions of variability and heredity we have already reached. 

 Heredity and racial variability are quantitatively ex- 

 plicable on the basis of a law, which shows the individual 

 producing undifferentiated like organs with a certain 

 degree of resemblance only. What is the bathmic source 

 of this limited variation within the individual ? That is 

 the great mystery on which inheritance and genetic varia- 

 tion depends. As for bi-sexual reproduction, its physio- 

 logical function may also be a mystery," but that mystery 

 will not be solved by asserting that sex is the fundamental 

 source of variability. 



^ I I. — On the Law of Ancestral Heredity 



We have dealt at length with the problem of bi-parental 

 inheritance. This will have suggested to the reader the 

 methods which must be employed when we deal with 

 the whole ancestry of an individual. In precisely the 

 same way as we formed the mid-parent from the organs 

 of father and mother, we can form a mid-grandparent ; 

 we reduce the female organs, i.e. those of the two grand- 

 mothers, to their male equivalents by multiplying them 

 by the ratio of male to female variability in that genera- 

 tion, and then take the mean of the organs of the four 

 grandparents. Similarly from the eight great-grandparents 

 we form a mid-great-grandparent, and so on. Thus a 

 man's 1024 great-great-great-great-great-great-great-great- 

 grandparents give a mid-ancestor of the tenth order. For 

 brevity we will term it his tenth mid-parent. Now it is 

 easy to see that, unless there has been much in-and-in 



1 See locus cited, p. 473. 



^ It has been attributed by some to the advantaije due to a division of 

 labour, but the share of the labour is not very apparent in the case of the 

 males of a considerable number of species. 



