466 THE GRAMMAR OF SCIENCE 



environment, then only intensive selection can keep the 

 community stable (see p. 447). If natural selection be 

 suspended, there results a progressive change ; the most 

 fertile tend to multiply, and multiply at an increasing rate. 

 In our modern societies natural selection has been to 

 some extent suspended ; ^ what test have we of the identity 

 of the most fertile and the most fit ? It wants very few 

 generations of genetic selection to carry the mode, the 

 type, from the fit to the unfit. Are the aristocracy of the 

 intellectual and of the artizan classes equally fertile with 

 the mediocrity of those classes ? I doubt it. I have not 

 yet obtained statistics for England, but from Danish 

 statistics I have shown ^ that : — 



{a) The absolute fertility of the working classes is 

 greater than that of the intellectual or professional 

 classes. 



{b) The net fertility per marriage of the latter is, 

 however, greater than that of the former, owing to a 

 selective death-rate, but 



{c) The marriage rate of the working classes is so much 

 higher than that of the intellectual classes, that their total 

 net fertility is relatively higher. 



Hence if the professional and intellectual classes are 

 to be maintained in due proportion they must be recruited 

 from below. 



Now this is a much more serious result than might 

 appear at first sight. The upper middle class is the back- 

 bone of a nation, it depends upon it for its thinkers, leaders, 

 and organisers. This class is not a mushroom growth, 

 but the result of a long process of selecting the in- 

 tellectually abler and fitter members of society ; roughly 

 speaking, its members marry within the caste, and they 

 form opinion and think for a nation. We want every 

 possible ladder for attracting to that class able members 

 of the hand-working classes ; but with very considerable 



^ By no means so much as is sometimes Supposed, it is only autogeneiic 

 selection, not inorganic or heterogeneric selection (see p. 378) which has been 

 reduced to a minimum. 



- The Chances of Death, and other Studies in Evolution, vol. i. p. 63, 

 " Reproductive Selection," 



