THE REGULATION OF NUMBERS 293 



bringing about an approximation to the desirable number is not 

 easily answered. As regards those primitive races which have 

 come under European observation there is no definite evidence 

 because, before exact observation could be made, the conditions 

 of life had wholly changed. But, putting aside the fact that 

 the evidence, such as it is, does suggest an approximation, it 

 may be observed that conditions wef^ such as to render an 

 approximation easy. Skill increased so slowly that for long 

 periods of time the desirable number remained about the same ; 

 the factors of elimination, such as war and disease, were not 

 erratic in their action ; social organization was not complex in 

 the economic sense and therefore the danger of break-dow^ns 

 followed by changes in the desirable number was absent. Thus 

 we may conclude that, in all probability, in primitive society an 

 approximation was normally attained. 



In the third period we meet with wholly different conditions. 

 Social organization becomes elaborate in the economic sense, war 

 and disease become erratic in their operation, and, relatively to 

 what was the case before, skill increases rapidly. As a result the 

 desirable number frequently changes, becoming on the whole 

 larger, particularly in those periods with which we are best 

 acquainted. There has in consequence grown up an idea that 

 throughout human history population has been increasing, 

 whereas in fact it is far more correct to regard population as 

 normally having been stable. It may not improbably turn out 

 that the third period is in this repect peculiar and we may be 

 approaching a period when population will again normally be 

 stable. However that may be, the changes mentioned tend to 

 make adjustment more difficult ; on the other hand, there have 

 been at the same time at w^ork changes tending to facilitate 

 adjustment among which growing freedom from convention and 

 increasing sensitiveness to the economic situation may be 

 particularly noticed. As regards the methods of adjustment, 

 evidence has been adduced to show that, up to the opening of 

 what we have called the mediaeval period, one or other of the 

 same methods were in use that were in vogue before, that all 

 these methods ceased to be practised in the mediaeval period and 

 were replaced by postponement of marriage, and that finally 

 in the modern period dehberate restriction has replaced postpone- 

 ment of marriage. 



