476 CONCLUSION 



have to begin among the upper classes. Though, therefore, 

 differential fertility by producing unfavourable germinal changes 

 is to be to that degree deplored, yet we have to remember that, 

 so far as quantity is concerned, failure to meet economic require- 

 ments might be a much greater misfortune. 



3. As regards quantitative problems we saw that from the 

 first period of history onwards — from the time, that is to say, 

 that it began to be possible for man to reap the benefits of 

 co-operation — it was of the utmost importance for every group 

 to approximate to the optimum number. This is the number 

 which — taking into consideration the nature of the environment, 

 the degree of skill employed, the habits and customs of the 

 people concerned, and all other relevant facts — gives the highest 

 average return per head. This number is not fixed once and for 

 all. On the contrary it is constantly varying as the conditions 

 referred to vary, and, as skill has tended to increase throughout 

 history, so has the number economically desirable tended to 

 increase. The errors underlying the wholly different exposition 

 given by Malthus have been indicated ; for him there was no 

 such thing as over-population. In his view population had at 

 any one time increased up to the possible limit and was in process 

 of being checked. In the modern view numbers may approximate 

 to the desirable level, may not reach it, or they may exceed it, 

 and if either of two latter positions arise, the return per head 

 will not be as high as it might be. 



The quantitative problem presents itself to all races at all 

 times. There is no escaping it. The common notion that it 

 only presents itself at certain times and in certain places is based 

 upon a failure to grasp the strength of fecundity. Almost without 

 exception those factors, which incidentally restrict increase and 

 produce elimination, are insufficient so to reduce fertility as to 

 keep numbers down to the optimum level. There thus arises 

 the need for factors which directly restrict fertility and cause 

 elimination ; among primitive races these factors take the form 

 of abortion, infanticide, and prolonged abstention from inter- 

 course. There is no correlation between these factors and the 

 economic stage reached^ and therefore we have no grounds for 

 assuming any one factor to have been prevalent at any one stage 

 in prehistory, though we must assume that one or more of these 

 factors was always at work. This assumption is confirmed by 



