THE ORIGIN OF TRADITION 423 



4. Fertility then is relative, but, other things being equal, 

 where fertihty is greatest there we find the greatest incentive 

 to increase in skill. From this follow certain important con- 

 sequences. So long as progress in skill keeps along any one line, 

 so long as it takes the form, for instance, of improvement in 

 methods of hunting, so long, as a general rule, will the same areas 

 remain the more fertile areas. But this does not always follow. 

 It may be that some remarkable invention, that of the bow and 

 arrow perhaps, renders areas, previously not the most fertile, the 

 most remunerative. Hitherto it may have been that in such 

 areas the hunting of game was for various reasons difficult until 

 a higher degree of skill was reached, but this higher degree having 

 been reached, these areas became more remunerative than those 

 previously the most fertile. With the progress of skill there is 

 thus a shifting of the centres of the highest fertility. 



This shifting of centres becomes more marked where progress 

 takes a different turn, where, for instance, agriculture arises. 

 For this there are two main reasons. First, it generally happens 

 that an area which is very fertile to a high degree of development 

 of one kind of skill, as the north-west coast of America is fertile 

 to hunting and fishing, is relatively infertile to another type of 

 skill, as, for instance, agriculture, in any case in its lower forms. 

 It is rare that any area is so highly endowed that at one and the 

 same time it offers a large return to the development of a particular 

 form of skill and is also relatively fertile to the first beginnings 

 of another and ultimately far more remunerative type of skill. 

 Secondly, it is not as a rule where skill is specialized that we find 

 the beginnings of a new form of skill. Tradition tends to move in 

 grooves and where attention is with success concentrated upon one 

 type of skilled process, it is not there that we should expect to 

 find the origin of quite another type of skilled process. This is 

 only an application of a principle noted in another connexion, 

 namely that evolution proceeds from the generalized and not 

 from the specialized type. 



In this manner the differences in natural endowment which 

 distinguish one part of the world from another bring about 

 a shifting of the centres of progress. There are other factors 

 still to be considered which work in the same direction ; but 

 apart from them this shifting comes about. Though this shifting 

 has been a feature of every stage of human history, it must have 



