Urban System in Geography

Abstract

Geographical ‘urban systems’ order and classify cities according to criteria. There are two central concepts: (a) the rank-size rule, from Zipf, whereby there is a fixed relationship between the rank of a city and its size; this appears to work well in self-contained regions, but less so in regions with many outside contacts; (b) central place theory, from Christaller, wherein there is a constant relationship between cities at different levels in a hierarchy, depending on the kinds of services they provide to their surrounding areas. The commonest, based on the ‘market principle,’ gives k=3; each level of the hierarchy has three times the number of cities, and an average city size one-third, the one above. This is consistent with the rank-size rule. General urban-location theory, from Lösch, extends central place theory into a general system of cities, arrayed in ‘city-rich’ and ‘city-poor’ sectors around a first-order metropolis. City systems change over time. Pred showed how interdependencies between cities evolved with improvements in transport technology in the nineteenth-century United States. It appears that, recently, the lowest levels of the urban hierarchy have been demoted to villages, while ‘global’ cities have appeared at the top.