Many old documents show that the Ainu in the Edo period (1603-1867), who lived on fishing, hunting, and collecting, were migratory people. And it is widely accepted that the Ainu moved seasonally from their fixed bases. This probably indicates that the residents of the bases were relatively stable. But a detailed analysis of other documents shows that many households moved their bases into other settlements in the southeastern part of Hokkaido during the years 1856-1858.
The purpose of this paper is to show that during 1856-1877 the households in the Monbetsu district of Hokkaido were open, flexible and variable in composition through inter-household movements, of its members though most of the households had fixed bases within the same settlements.
The analysis of inter-household movements was made by tracing the name, age, and kinship of each of the household members. For example, if the name of a member of household Al in 1856 was found in the list of members of household B2 in 1858, he or she was recognized as having moved from Al to B2. The kinship relationships of each member show which members of a household moved together and whether they moved for marriage or not. The documents used in the analysis are the lists of the inhabitants in 1856 1862, 1868, 1872, 1876, and 1877, which were compiled by the Japanese. The details of the findings are as follows.
In the Monbetsu district during 1856-1877, the bases of most of the households were fixed within the same settlements (Table 2), and the residents of the settlements were relatively stable (Fig. 2). But many resident members moved between households (Table 3), and about 70% of the movements were intra-settlement. The number of persons per 100 inhabitants during the 10 years who moved between households was very large in the Monbetsu and Takashima districts, but it was small in the Shizunai district and the southwestern part of Sakhalin (Table 5).
Of the 556 persons in the Monbetsu district whose names appeared in more than one list, 283 (50.9%) moved to other households once or more than once (Table 6). Such movements caused the membership of household A 31, for example, to change greatly during the period 1856-1877 (Fig. 3). The household members of A 31 were as follows:
(a 31-1, a 31-2, a 31-3, a 31-4, a 31-5), 1856
(a 31-1, a 31-5, a 14-3, a 14-4, a 26-2), 1862
(a 31-1, a 31-5, a 14-3, a 26-2, d 4-4), 1868
(a 31-1, a 31-5, a 14-3, d 4-4), 1872
(a 31-1, a 14-3, a 14-4), 1876 and 1877
_??_a 31-1, a 31-2, etc., represent individual household members._??_
During the period 1856-1877, the Monbetsu inhabitants who moved between households moved individually in most cases, rarely together (Table 6). When they moved together, they were in most cases parents and their children, especially mothers and their children (Table 8). When they moved individually, 59.6% (146/245) of them moved for marriage. 36.6% (30/82) of those cases where they moved together also involved persons who moved for marriage.
The number of persons who moved and/or stayed together all the time from 1856 to 1877 as if they were one unit was one in most cases and two to four in others (Table 7). Therefore, household membership was not stable in the Monbetsu district. The stability of the household members was much lower in the Monbetsu and Takashima districts than in the Shizunai district and the southwestern part of Sakhalin (Figs. 4 and 5).
Thus, two types of mobility were found in the Ainu society: the inter-settlement movement of households and the inter-household movement of the inhabitants.

View full abstract

The purpose of this paper is to clarify the regional characteristics of three main neighborhood movements in the drainage basin of Lake Kasumigaura, in particular, the environmental conservation movement. Two voluntary groups are mainly investigated. One is the Nature Conservation Society of Tsuchiura, and the other is the Liaison Conference for the Conservation of Lake Kasumigaura. Eutrophication of Lake Kasumigaura affects all the residents of the drainage basin, but only those in the urbanized area took action and agitated for environmental conservation. Eutrophication is essentially an urban problem.
Development of Lake Kasumigaura has progressed since a dam was constructed in 1963 in order to turn the lake into a freshwater lake. Its water resources have supported industrialization and population growth in the area. But water resource development, industrialization, urbanization, carp culture and stock raising have combined to result in the lake's steady eutrophication, and nowadays every summer the lake is host to dense growths of plankton which damage the carp culture seriously and give a foul smell to tap water in the surrounding area.
The area around the lake can be divided into four main zones (Fig. 4). The zone along the JR Joban-line is rapidly urbanizing because of the expansion of the Tokyo Metropolitan Area and the development of Tsukuba Academic City. The rate of population increase in this zone is the highest of the four (Fig. 5). A big steel plant is located in the Kashima zone. In the northern lakeside zone, carp culture and hog raising are important. The fourth zone includes the lowland along the Tone River and the upper and middle reaches of the Sakura River.
After the water resource development of the lake, three main neighborhood movements sprang up over the use of the lake. These movements centered around the first three of the above-mentioned zones (Fig. 6). In the JR Joban-line zone, environmental conservation is the object. In the Kashima zone, the residents are opposed to an Industrial Development Policy that prescribes drawing water from the lake for use in industrial areas. In the northern lakeside zone, the fishermen are protesting the Land Reclamation Project of Takahama-iri Inlet, begun in 1967.
In the JR Joban-line zone, two groups have been organized for environmental conservation. The Nature Conservation Society of Tsuchiura, organized in 1972, has played an important role in the movement. The Liaison Conference for the Conservation of Lake Kasumigaura was organized in 1981. Their chief interest is the solution of the eutrophication problem. Most of the members of the two groups live in Tsuchiura and Tsukuba Academic City. Eutrophication affects all residents who live in the drainage basin, but most of those who are active in the groups live in the urbanized area (Fig. 7). In the area around the lake, only the residents of the urbanized area have to drink water from Lake Kasumigaura. The lake poses two problems for them. One is that the lake is a pool of sewage, and the other is that it is the source of drinking water. Water pollution inevitably causes doubts about the safety of drinkingwater from the lake. This fear motivated some of the residents to act. The movement did not begin only in response to water pollution. The JR Joban-line zone has rapidly urbanized, and environmental deterioration is most serious in the area around the lake. Rapid environmental changes have made the residents nervous about environmental deterioration, including water pollution. Many residents of the urbanized area are interested in the environment. The water in Tsuchiura has a special meaning for persons who remember the area's past unspoiled beauty. Some of the residents who moved to this area from Tokyo are also keenly interested in the environment, and some of them have taken part in the movement.

View full abstract

Since we cannot directly capture the cognitive map as an internal representation, researchers are obliged to explore its nature indirectly through spatial products (eg. sketch maps) (Liben, 1981) derived from subjects in various ways. Geographers, who attempt to elucidate the nature of cognitive maps by comparing the spatial products with the actual map, must acquire cognitive maps represented in two-dimensional space that are comparable with the actual one.
The author has previously presented a quantitative method to capture the spatial properties of cognitive maps, and attempted to apply the method to high school students in the Tokyo metropolitan area (Wakabayashi, 1989a). On the basis of the previous study, the aim of the present paper is to examine several remaining problems: (1) a theoretical explanation of distortions in cognitive maps based on the results of experimental research in cognitive psychology, (2) improvement of analytical techniques, (3) investigation of non-Euclidean properties of cognitive maps, and (4) generalization of the patterns of distortions in cognitive maps.
Concerning the first problem, the author presented a framework of research derived from Lloyd and Heivly (1987) and Lloyd (1982, 1989) (Fig. 1). In this framework, the cognitive mapping process consists of 1) acquisition of information from the environment, 2) encoding, 3) memory storage, 4) decoding, and 5) reproduction of external representation. The distortions appearing in the external representation can be operationally divided into absolute and relative ones, employing Euclidean regression (Lloyd, 1989). While absolute distortion is mainly due to the encoding and decoding process, relative distortion is difficult to explain with general theories, because it might reflect the characteristics of the physical environment. This study is focused on the influence of the physical environment upon the relative distortions in cognitive maps.
The study area is the central part of Sapporo citys streets are laid out in a grid pattern and a regular address is assigned to each grid according to rectangular coordinates (Fig. 2). Such a simple and regular form is an exception among major Japanese cities.
The sample points for the research are nine places shown in Fig. 2, which are wellknown to the students at Hokkaido University. The data were collected from 170 students through ratio estimation of interpoint distances and sketch maps for the nine sample points.
In the same way as was done by Wakabayashi (1989a), cognitive configurations of sample points were obtained, applying a nonmetric mutidimensional scaling (MDS) to each estimated distance matrix; the configurations of sketch maps were directly measured. Euclidean regression (Tobler, 1965) was then employed to fit the actual configuration to the cognitive ones. Finally, all fitted configurations are aggregated by means of standard deviational ellipses for each point. The results obtained are summarized as follows:
(1) A relatively better fit of cognitive maps to the actual one is detected. Particularly for sketch maps, the average value of bi-dimensional correlation coeffcients reached 0.953, while the configurations of MDS maps are better congruent with the route distance space (Fig. 7) recovered by MDS than with the actual one
(Tables 1 and 2). (2)The displacement of mean centers of ellipses and their area, which measure distortion or accuracy and fuzziness (Gale, 1982), is smaller in sketch maps than in MDS maps (Figs. 5 and 6; Table 3). Such a tendency was also reported in previous studies (eg. Buttenfield, 1986). Especially in Sapporo, rectangular reference lines may enable subjects to easily encode and decode locational information in their memories as well as to produce relatively accurate sketch maps; the MDS maps are distorted according to route distance in the street pattern with a rectangular grid.

View full abstract