The purpose of this article is to clarify the followings: (1) ecological conditions conducive to the genesis of urban communities in Southern England during the Iron Age, (2) spread patterns of such urban settlements in their chrysalis stages, and (3) territories over which such urban settlements controlled. The study results are summarized as follows.
The genesis of an urban community can be generally accounted for by the formation of a habitat of non-agricultural groups, being supported by the over-surplus of agricultural products. The enlargement of agricultural produce in England in the above period can be verified by means of pollen analysis administered in the following areas; Hockham Mere (Norfolk), Old Buckenham Mere (Norfolk), Decoy Pool wood (Somersetshire), and Thrang Moss (Lancashire). A supposition could be made out of these findings that forest clearances had been quite vigorously pursued in order to obtain arable land.
The fact that defensive settlements called hillforts took urban appearances, can be known from the following archeological findings: in such places as Maiden Castle (Dorsetshire), South Cadbury (Somersetshire) and Heathrow (Middlesex) either a shrine or a temple had been excavated, and using these discoveries we can surmise that these places had been religious centres. It can also be assumed that Hod Hill (Dorsetshire), Chalburry (Dorset shire), and Credenhill Camp (Herefordshire) had a large population.
The regularity shown in the distribution of a hillfort whose area covers over 15 acres, leads to the presumption that these hillforts presented themselves as central places in the hexagonal network with a distance of 13 kilometers, when measured from the centre of one cell to the centre of the next one nearest to it.
An additional clarification is the fact that Credenhill Camp had a higher ranking than the other central places with its sphere of influences amounting to a circle with a radius of 30 kilometers. Judging from tribal territory and distribution of pottery works, a cultural sphere of influence can be discerned as a circle with a radius of some 30 to 40 kilometers.
The territorial scale as recognized in the above-mentioned findings has been maintained in the regional size of iron-aged oppidum (Belgic oppidum) as the central place. Never-theless, the most powerful state Catuvellauni seems to have enjoyed a radius of some 60 kilometers as her sphere of influence. As such, we can surmise that the regional system prevailing then was somewhat akin to the Kolb's Central Place Model.
The circular regional system discussed above, had been so recapitulated, even in the civitas under the Roman rule, that we can surmise an existence of some ecological balance that had been sustaining such a regularity.
In conclusion, the author should like to thank Drs. K. Fujioka and I. Suizu, Professors at Kyoto University, for their continual assistance. His thanks also go to Dr. A. M. Lambert, of the London School of Economics and Political Science for her generous assistance and cooperation in securing the necessary materials contained in this paper.