The purpose of this article is to classify the characteristics of the distribution of Japanese towns before the industrial revolution. To qualify such towns, the author uses “Kyomhseihyo” (1880), the national statistics on population and products, and limits as town the settlements with over 2, 000 persons, but some fishing villages may be exceptionally contained among them.
The author thinks that the distribution of those towns is analogous to that “alten Kulturländer” called by G. Schwarz, which roughly speaking is related to the density of rural population (Fig. 1.) There was a dense net of towns with much urban populatin in coastal and basin regions (densely populated), where there were also large towns like Tokyo (725, 000), Osaka (363, 000), Kyoto (136, 000), Nagoya (117, 000), Kanazawa (108, 000) etc. (Fig. 2 and 3). Of course, the agglomeration of towns and urban population in those days were not in so large scale as in the present day.
But the distribution of towns in those days can not be explained by population density only. The ratio of urban population per Kuni, regional division in those days (Fig. 4) and the hierarchical structure of towns were related more closely to scale of regional centres than to economic richness of the areas. For example, the large regional centre Toitori (36, 000) and some small towns (2-3, 000) lay in Inaba-Kuni, so that the ratio of urban population per Kuni was raised (exclusively) by the urban population of Tottori.
Most of such regional contres were castle towns in feudal age, and their scale was in proportion to that of territories of “Daimyoes”, feudal lords. The origin of small towns was mosily market towns, coaching towns, and port-towns, which had grown in proportion to regional economic development.
Therefore, the distribution of towns in early Meiji-era was related to hitstorical conditions in feudal age everywhere.
