474 THE GRAMMAR OF SCIENCE 



(t) Whatever be the physiological function of sex in 

 evolution, it is not the production of greater variability. 



The variability of the mid-parent is less than that of 

 the single parent, and the position of the mid -parent 

 appears to be closely allied to that of the parthenogenetic 

 mother. 



{d) Without heredity at all and without a bathmic 

 bias (p. 375), the parent could not be supposed to re- 

 produce individuals all like itself, nor, again, all like the 

 race type, but would produce an array of the same type 

 as the race and of equal variability, i.e. there would be 

 a reproduction of the race on a small scale. 



The effect of heredity is to draw this " race on a small 

 scale " from the racial towards the parental type, and at 

 the same time to diminish its variability. This is the 

 verbal interpretation of the formulae for regression and 

 variability (pp. 397 and 401), 



(f) Whatever amount of selection has taken place 

 there seems no possibility of reducing variability beyond 

 some 10 to II per cent (see p. 472). It is, therefore, 

 erroneous to suppose a greatly reduced variabilit}' could 

 have appeared in the progress of evolution. 



{/) Heredity is the law which accounts for the change 

 of type between parent and offspring, i.e. the progression 

 from the racial towards the parental type. It is perfectly 

 consistent with the most diverse degrees of racial varia- 

 bility, and it is completely wrong to suppose that when 

 race variability is large then heredity is small, or vice versa. 

 For a given race variability, the variability of the offspring 

 of one pair of parents does vary with the intensity of 

 heredity, but it cannot be increased above the racial 

 variability. 



{g) So far as we are able to judge from a considerable 

 number of organs in a considerable number of forms of 

 life, there is an approach, although not very close, to the 

 law, that all organs and characters in all forms of life are 

 inherited at the same rate. 



If this law were absolutely true, then heredity certainly 

 must have preceded evolution. Anyhow it is difficult to 



