In order to clarify climatic regionality in relation to synoptic meteorological phenomena, the daily duration of sunshine and the daily mean temperatures at about 140 sites in Hokkaido are analysed with time variation (day-to-day variation). Both principal compo-nent analysis and cluster analysis using factor loadings are applied. The values of the factor loadings correspond to the correlation coefficients between the principal component and the climatic element of each point, and also indicate the time coefficients between them.
The daily data for 1978_??_1980 were used in this study. Each datum was converted into a deviation from the areal mean for each day. The data for 1978 were mainly used for the analysis, although the results were compared with those for other years to examine the propriety of the results.
The first three dominant patterns of the daily variability of sunshine duration account for more than 50% of the total variance. The first eigenvector (24.8% of the total vari-ance) depicts the area of positive (negative) departure to the northwestern side of Ho-kkaido with the area of negative (positive) departure to the southeastern side (Fig. 2-a). The positive pattern (i, e., the total duration of sunshine over the area with a positive sign is relatively greater than that of the area with a negative sign) prevails in summer. This is due to the reduction of sunshine in the southeastern side as is influenced by cyclon-es passing through the southeast of Hokkaido or by sea fogs which are caused when southeasterly warm monsoons blows over the cool Chishima Current. In winter the nega-tive pattern prevails; it reflects the diminution of sunshine in the northwestern region as is influenced by snow clouds accompanying winter monsoons.
The second eigenvector (17.2% of the total variance) indicates the contrast between the northeastern side (Okhotsk side) and the southwestern side (Donan side) of Hokkaido (Fig. 2-c). Each of these positive or negative patterns is formed with the influence of both the front and the passage of cyclones in the warm season.
The pattern of the third eigenvector (9.3 % of the total variance) describes three areas, i. e., one positive area which extends from the northeast to southwest with two negative areas on either side (Fig. 2-e). This positive pattern appears clearly in summer. It is formed with the influence of both the sea fog mentioned above and cyclones passing over the north of Hokkaido. Furthermore, this pattern tends to appear at the same time as the first component of the sunshine duration has a dominant positive value. On the other hand, the negative pattern, which is caused with the influence of fronts passing over the center of Hokkaido, does not turn out clearly.
From these results, four regions with different time variation types of sunshine duration are pointed out. These are the Pacific side type, the Japan Sea side type, the Sea of Okhotsk side type, and the Donan side type (Fig. 8). Especially, the southeastern side of Hokkaido which is belonging to the Pacific side type has fairly specific regional properties of the sunshine duration in comparison to other regions.
The first three eigenvectors of daily mean temperature variations account for 43.1%, 14.4%, and 10.6% of the total variance, respectively (i. e., cumulative for 68.1% of the total variance). The pattern of the first eigenvector which distinguishes the inland area from the coastal area is caused by the land-sea heating contrast (Fig. 4-a).
The distribution pattern of the second eigenvector closely resembles that of the first eigenvector of the sunshine duration (Fig. 4-c). Both positive and negative patterns prevail in early summer and winter, whose causes are explained by the cooled or warmed wind over the sea in each season.

View full abstract

In this paper, the writer dealt with lotus root growing at the Tamura and Okijuku districts in Tsuchiura city which are located in the western shore of Lake Kasumigaura. He made clear the relationship between land conditions and lotus root farm management in these districts by his research, which was carried out before the rice acreage reduction policy and there after. The results are summarized as follows;
The alluvial plain along, the northern shore of Tsuchiura inlet, the western part of Lake Kasumigaura, has a gentle slope, and it can be divided into upper alluvial surface and lower surface. The former is a coastal plain, which was formed by upheaval of the land that had been below the sea-level at the high water level period. And the land is semi-ill-drained field. The latter is a lake shore lowland that continues to be formed now and it might be subject to flood if it were not for the dikes. Therefore, the land is vast ill. drained field.
At the northern shore of Tsuchiura inlet, the ratio of lotus fields to all paddy fields is more than 80 per cent, indicating one of the lotus root producing centers. But the writer found that the different introduction periods of lotus root and land conditions resulted in different agricultural management groups in these districts. First of all, the transition from rice growing to lotus root growing at the Tamura district began at an early stage because most areas of the district had vast ill-drained paddy fields with silt and clay. Thus, the farmers had already become producers skilled with high lotus root productivity before the rice acreage reduction policy was enforced. On the other hand, in the Okijuku district, rice has been a major crop because of large area of semi-ill-drained fields with silt and sand, and the transition to lotus root growing occured after the enforcement of the rice acreage reduction policy.
Yields of lotus root per ten ares at the Tamura district is high at every class of the farm scales by utilizing suitable land conditions of ill-drained paddy fields and the production experience, and family income per ten ares exceeds 250, 000. This figure is three and one-half times as large as the average income of the rice growing farm. This is the economic basis of the formation of farmer groups specialized in lotus root growing in this district. On the other hand, yields of lotus root per ten ares the Okijuku district is lower than that of the Tamura district because the land is silty sand that is poor in keeping the fertilizer, and family income per ten ares ranges fro¥110, 000 to ¥170, 000 at each class of farms. It shows high profitableness, however, compared with that of the average rice growing farms. And the subsidy for changing rice to other crops has made up for the gap of profit against the Tamura district. Here the writer found that the lotus root growing at the Okijuku district depends upon the subsidy for changing rice to other crops under the rice acreage reduction policy.

View full abstract

The Pleistocene glacial landforms in the Northern Japanese Alps have been studied by many researchers, and the existence of several glacial stades, ranging from 60, 000 to 10, 000 years B. P., has been proposed by Kobayashi (1958), Iozawa (1962, 1972), Koaze et al. (1974), Fukai (1975) and Ono (1980). The glacial extent and the valley filling before 60, 000 years B. P. have been little recognized. This paper describes the glacial landforms and the accumulation terraces before 60, 000 years B. P., located along the River Karasu, at the eastern part of Mt. Chogatake in the Northern Japanese Alps.
At Chogatake Cirque, a rather dissected old cirque at the eastern side of Mt. Chogatake (2, 664 m), the cirque bottom, situated about 2, 000 m in altitude, is partly eroded away by the dissection of the River Chozawa. In this area, the present writer found out two sets of terminal moraines: the Honzawa and the Mameuchi Moraines (Fig. 5). The former, located 1, 600 m high, indicates the Pleistocene maximum glacial extension along the Hon-zawa Valley, and the latter, located at a height of 1, 900 m on the frontal edge of the Chogatake Cirque, shows a recessional feature.
Along the downstream of the River Karasu, five accumulation terraces develop which are classified into Terraces I to V in this paper (Fig. 8). Terrace II which extends to the Honzawa Moraine on the terrace profile (Fig. 9), consists of fluvioglacial deposits judging from the sedimentary facies. The pumice fall deposits of Tateyama Volcano (DPm) are intercalated in the periglacial deposits covering the gravel of Terrace II. The pumice fall deposits of the EPm or OPm are intercalated in the periglacial deposits cover-ing Terrace III (Fig. 7).
In view of the presumed age of the DPm fall, a stage of glacial extension and accumu-lation terracing (Tarrace II) in this area should be one of the cold periods before 100, 000 years B. P.. This glacial stage is named here the Chogatake Glacial, which can be divided into two stades, judging from the existence and location of the two terminal moraines. After the Chogatake Glacial, a periglacial environment seems to have governed this area during the period between 60, 000 and 35, 000 years B. P., on the basis of the presumed age of the EPm and OPm. Almost no glaciation occurred around the Chogatake Cirque, although the existence of a glacial stade of this period is widely recognized in other parts of the Northern Japanese Alps by many researchers (Fig. 10).

View full abstract