This paper aims to review the development of central place studies in Japan. The situation has somewhat changed considerably in the ten years since the publication of review articles by Watanabe (1975), Morikawa (1977) and others. As reported in the above-mentioned papers, studies of central places in the early 1950 s were influenced not only directly by Christaller's theory but also by other schools of thought such as rural sociology in the United States. Trade area, traffic area, life sphere, city region, etc, were also the focus of attention in urban geography on that area. Since that time, the study of city regions has become a major field but the empirical study concerning the hierarchical structure of the central places started to be undertaken, especially by Watanabe (1954 c, 1955). And the central place studies in Japan have grown gradually both in terms of quality and quantity little later than the studies by Anglo-Saxon geographers which reached at one of the golden age around in 1960, when the IGU Symposium was held at Stockholm. Two central research directions can be recognized in central place studies. One direction aims to elucidate structural characteristics of the Japanese central place system. Wa tanabe (1975) claims that the research concerns of Japanese urban geographers remain on the level of the morphology of structural patterns and that studies of the causal consequences of structure are lacking. However, his view seems to reflect the fact that most earlier research was in this direction, including his own. Early research endeavored to clarify the structure of central place systems. Second direction of study has made efforts to introduce quantitative approaches in Anglo-Saxson geography; it has very rapidly developed since the study of Berry and Garrison (1958). With the exception of the studies of Ishimizu (1957 c) and M, Nishimura (1956, 1969), it was not until the early 1970s that Japanese central place studies using the quantitative approach were published. Both directions of central place studies have developed in parallel, though over time the quantitative approach has become greater emphasized. Table 1 shows the number of papers in central place studies published between 1951 and 1986 in units of five years. There is a decreasing trend in the number of papers pub lished after the peak years of 1971-75, when studies of quantitative approach were first published. In recent years the number of papers on theoretical and methodological approaches has increased significantly compared to the number of empirical studies; within the category of empirical studies, there is an increasing number of papers from related fields such as regional-planning application studies. The following four points seem to be most important in a discussion of central place sutdies today. 1) In recent years the quantitative approach using factor analysis has increased greatly. Although new results have been obtained using such analysis, the field of study becomes too narrow when one depends on this type of analysis alone. A pluralism of approaches is required. 2) Despite long-term development of central place studies since the early 1950 s, method of investigation easily applicable to actual analyses has not yet been established. In mu-nicipalities, for example, statistical approaches are not always appropriate, especially when more than one central place are included within the area of the municipality. In regional planning, the centrality of each settlement should not be measured simply by retail function but also in terms of other central functions such as administration, education, health service, communication etc., from more wholistic perspective. 3) The central place system on a national scale has been considered using only simple criterion. However, such research has not yet been taken on in the scale of, for example, the “Kluczka-Karte” in West Germany.
Recently, many investigators have studied the heat island phenomenon and consequently the horizontal temperature distribution of the heat island has been clarified. However, the vertical distribution of the heat island has not been clarified still. A cross-over phenomenon was certified there of in large cities, but the accurate existence of a cross-over phenomenon in medium sized cities has been uncertified up to date as the result In this paper, in order to reveal the vertical and horizontal constitution of the heat island in Tsuchiura City (36°06'N, 140°13'E), the observations carried out on October 29/30, 1983 is reported. The results are summaised as follows. 1) Vertical temperature distribution was observed by using Kytoon at Stations 1, 2, 3, 4 and 5. The heat island phenomenon was clearly certified in the central area of the city. The height of the heat island reached up to about 150 m in the sky at 5:00 a. m. on October 30, 1983, under little wind and the clear sky condition (Fig. 3). 2) To express the three-dimensional heat island intensity, the vertical temperature effect to the city area, the shaded area in Fig. 4, was measured. And the relation between vertical urban temperature effect and horizontal heat island intensity (ΔTu-r) was examined. It was certified as the result that the vertical urban temperature effect was proportional to the horizontal heat island intensity (ΔTu-r), (Fig. 5). 3) Vertical humidity distributions were measured by use of my original instruments. In order to compare urban humidity with different rural humidity, change of the urban humidity effect (Δdvu-r) was estimated. The result was that absolute humidity was sometimes higher in the central area of the city than in the rural area at height between 30 m and 70 m levels in the sky (Fig. 6).