REDISTEIBUTION OF MANKIND DICKSON. 563 



required is greatly diminished. One remembers when a 50-shilling 

 Swiss watch, although then still by tradition regarded as sufficiently 

 valuable to deserve inclosure in a case constructed of a precious metal, 

 was considered a marvel of cheapness. American machine-made 

 watches, produced by the ton, are now encased in the baser metals 

 and sold at some 5 shillings each, and the watch-making industry 

 has ceased to be specially suited to mountainous districts. 



In considering the differences which seem likely to arise in wdiat 

 we may call the regional pressures of one kind and another, pressures 

 which are relieved or adjusted by and along certain lines of trans- 

 port, I have made a primary distinction between " east-and-west " 

 and " north-and-south " types, because both in matters of food supply 

 and in the modes of life w^iich control the nature of the demand for 

 manufactured articles climate is eventuallj^ the dominant factor ; and, 

 as I ha\'e said, climate varies primarily with latitude. This is tiiie 

 speciallj^ of atmospheric temperature; but temperature varies also 

 with altitude, or height above the level of the sea. To a less extent 

 rainfall, the other great element of climate, varies with altitude^ but 

 the variation is much more irregular. More important in this case 

 is the influence of the distribution of land and sea, and more especi- 

 ally the configuration of the land surface, the tendency here being 

 sometimes to strengthen the latitude effect where a continuous ridge 

 is interposed, as in Asia, practically cutting off " north-and-south " 

 communication altogether along a certain line, emphasizing the par- 

 allel-strip arrangement running east and west to the north of the 

 line, and inducing the quite special conditions of the monsoon region 

 to the south of it. We may contrast this with the effect of a " north- 

 and-south " structure, which (in temperate latitudes especially) 

 tends to swing what we may call the regional lines round till they 

 cross the parallels of latitude obliquely. This is tyjjically illustrated 

 in North America, where the angle is locally sometimes nearly a right 

 angle. It follows, therefore, that the contrast of " east-and-west '" 

 and " north-and-south " lines, which I have here used for purposes 

 of illustration, is necessarily extremely crude, and one of the most 

 pressing duties of geogi-aphers at the present moment is to elaborate 

 a more satisfactory method of classification. I am very glad that we 

 are to have a discussion on " Natural regions " at one of our sede- 

 runts. Perhaps I may be permitted to express the hope that we shall 

 concern ourselves with the types of region we want, their structure 

 or " grain," and their relative positions, rather than with the precise 

 delimitation of their boundaries, to which I think we have sometimes 

 been inclined, for educational purposes, to give a little too much at- 

 tention. 



Before leaving this I should like to add, speaking still in terms of 

 ■• east-and-west " and " north-and-south," one word moi-e nbout the 



