When the Ritsuryo (_??__??_) state (the statutory state ruled by the Ritsuryo Code in ancient Japan) with its nuclear region in Kinai (the central part of ancient Japan, i. e., Yamato, Yamashiro, Kawachi, Settsu, and Izumi provinces which altogether constitute the chief part of the Kinai district of Japan today) advanced to northeastern Japan, it came to contact with the system and culture of Yezo (_??__??_), a strong peripheral people in northeastern Japan. The author calls this region where both systems or cultural powers met the transitional zone of the Ritsuryo state and Yezo. It is necessary to analyse ar-chaeological materials and historical records of both the Ritsuryo state and Yezo, but we cannot find any historical materials clearly recorded by Yezo. Accordingly, for the study of the regional structure of the transitional zone, the author thinks that we should inquire the gradually dissolving process through which Yezo people or the Yezo territory became other than what they had once been and came to be integrated into the Ritsuryo state system. Such an analysis involves, the author thinks, an important problem and significance in historical geography. This paper is one of the author's serial historical studies of the transitional zone.
In this paper, however, the author does not intend to study the historical changes, but to inquire the physical geographical conditions of the transitional zone which might support his . further work. The study is not a mere restoration of the ancient natural landscapes, but an investigation of the natural conditions and geographical factors needed for the understanding of the regional structure of the ancient transitional zone.
The author focuses his attention on the period in which josaku (_??__??_) (the forts built in northeastern Japan in the ancient period) were constructed; i. e., from about the middle to the end of the Nara period. This was the period in which the Tai-i (_??__??_) Policy (the policy taken to apply the Ritsuryo state system over Yezo) was actively enforced in northeastern Japan. The Ritsuryo government constructed five forts (socalled the Tempyo no go-saku _??__??_) and two forts of Momoo (_??__??_) and Ichi (_??__??_) in the Tempyo time. Thus, the Sempoku area (the northern half of the Sendai plains) was the stage of the significant Tai-i activities. And yet, in the Sempoku area, settlements were only gradually incorporated into the local administrative system. Before the Ritsuryo Era, the yokoana-kofun (_??__??__??__??_) culture of the Kofun (_??__??_) Era (the first half of the ancient age) had advanced to the basin of the River Hazama running in the northern fringe of the Sempoku area. Therefore, it is clear that the advance of Kinai culture and of the Ritsuryo state system into northeastern Japan stopped in the Sempoku area. Questioning why the advance was stagnating in that area is a significant problem in historical geography.
On the factors of the stagnation in the transitional area, the author speculates as follows. Analysis of the northern culture (the culture of the prehistoric Hokkaido earthen vessel) shows that the southern limit of distribution of the Hokkaido-type earthenwares is generally found at the north fringe of the Sempoku area. Moreover, the dense distribution of the Ainu placenames has its southern limit roughly in the basin of the River Hazama, namely, at the northern fringe of the Sempoku area.
To the north of the Sempoku area, namely, in the middle course basin of the River Kitakami, a powerful Yezo people settled. For this reason, when the Ritsuryo state system forged ahead into the middle course basin of the Kitakaani,

View full abstract

As located in a mountainous region, the Shimanto River has little or no plains alongside both its upper and lower banks. In case of torrential rains caused by typhoons in July-September periods, extremely large quantities of water flow down to the river and the flow of 17, 000 m3/sec., which was the greatest in this country, was recorded by the heavy rain on August 29, 1935. As this water passed through narrow valley plains, the water surface was extremely raised upward. This means that its surface was raised up by 18.5 m above the average surface of 7.5 m. The figure of 18.5m may be regarded as the biggest overflow in the past 50 years or so in this country.
Kawanobori is a tiny agricultural village located 22 km upstream the Shimanto River. This village is on a natural levee formed by the Shimanto River and is seem-ingly free from the danger of overflow. In actuality, however, the natural levee was flooded 4.5 m deep on August 29, 1935, nevertheless it is located 7.5m higher than its backmarsh. From these reasons, this natural levee is apt to be misunderstood as either a terrace or a sand dune, However, we understand that this natural levee is neither a terrace, since the deposit is formed by sand mixed with fresh clay of more than 11 m in thickness, nor a sand dune from the analysis of the shape of grains of the deposit. We also understand that the natural levee with a height of 7.5 m above the backmarsh, the biggest figure to be seen in this country, implies that the level of flood along the Shimanto is exceptionally high and neither mud nor sand is supplied from the mountain sides by landslides. Although there were 45 houses in Kawanobori located on the natural levee, all of them were totally washed away by the flood of 1935.
The extraordinarily high floods of the Simanto River strongly influences the floods by the Nakasuji River located a little further downstream. As the Nakasuji River is exceptionally almost level, large quantities of water flowed into it from the main stream of the Shimanto, not from upper stream, in the case of the flood of 1935. As a result, a depth of 7-10m of water overflowed the low terrain of 2-5m high.
The fact that the level of the natural levee along the Nakasuji River goes down from down to upper stream is explained by the reverse overflow of the main stream into the tributary. The deposit of the natural levee is almost similar to that of the backmarsh and the contents are mainly silt mixed with clay. However, in 1964, the confluence with the main stream was changed about 7km downstream from the former point. In this way it was successful in lowering the water level by about 5m at the time of flood.

View full abstract

It is well known that the advection fog invades from the sea to the Pacific coast of the eastern part of Hokkaido in summer season. The summer climate of this region is characterized by the fog, which brings lower temperature, the shorter sunshine dura-tion, and the higher humidity. In this study, the present writer tries to clarify the seasonal and diurnal variation of fog frequency, the speeds of fog invasion and dis-sipation, and other climatological factors during the invasion or dissipation of fog.
The results are as follows:
1. In Kushiro District, fog frequency becomes higher from the 32nd half decade to the 47th. Especially the 33rd to 36th and the 41st to 43rd are the most frequent. In these periods the pressure pattern of the Baiu type and the summer type prevail. This variation is shown in Pig. 2.
2. Fog frequency has great diurnal changes. Generally, it is the most frequent during 5:00 to 6:00 in the morning, and reaches over 40% at its peak. The diurnl change of fog frequency diminishes on the sea shore, and increases in the inland, where the fog invades at night and dissipates early in the morning, as shown in Fig. 3. The fog in the inland is characterized by the radiation type, but mixed with the advection type.
3. The time of fog dissipation is mostly between 7:00 to 8:00 in the morning, and has a little seasonal fluctuation. The speed of fog front retreat is about 2.5km/h, but the higher speed is observed on the seashore in contrast to the lower speed in the inland.
On the other hand, fog invasion occurs mostly between 19:00 to 22:00 with the average speed of 2.0km/h. It is lower on the seashore and higher in the inland.
4. In the daytime (9:00-15:00), the invasion of shallow advection fog makes the air temperature lower by 1. 5°C to 2. 3°C on an average and the relative humidity higherr by 5 to 12% for 30 to 70 minutes.
5. The temperature distribution patterns during the time of fog invasion are shown in Figs. a-, -10. The difference of temperatures between on the seashore and in the inland (6.0km from the coastal line) amount to 5 or 6°C at maximum, but it is only about 2°C after the fog dissipation.
6. The fog front marches from Shinto Cape, southern part of Kushiro Port, or from the eastern coast of Kushiro District to the inland. It invades faster in the eastern and western parts of the New Kushiro River, but slowly in the highly urbanized area, the leeward area of upland. The air temperature of this area is higher even during the time of fog invasion.
7. In summer, the winds of this district blow mostly from the sea to the inland. This is also true at the time of the fog invasion. As given in Figs. 8_??_10, the wind during the time of the fog invasion blows at right angles to the isothermal line. This means that the cold wet air mass advects to this area from sea surface. Fig. 11 shows wind roses during the time of the fog invasion in the day time. The average wind speed is 4.5m/sec on the coast and 3.0m/sec in the inland.
8. In order to clarify the wind conditions in detail, the wind shaped trees as an indicator were observed in this district. The distribution of wind conditions as revealed by the wind shaped trees shown in Fig. 12 resembles to that of the results obtained by the instrumental observation shown in Fig. 11.

View full abstract

Kurara-Seto is a narrow passage about 7 kilometers in width located between the Chikuzen O-Shima (Island) and the northern coast of Genkai-Machi, Pukuoka Prefecture.
There exist several islands and reefs in the strait and strong tidal currents is passing through them. Two striking ridges formed by linearly arranging islands and reefs cross the strait. These ridges are composed of hard rock called the Kanmon Group of Mesozoic Age. The basin between the two ridges has a smooth bottom with a depth of 10 to 20 meters. The basin is composed of soft rock belonging to the Munakata Group of Tertiary Age. Such features of the ridges and basin are originated from selective erosion of tidal currents acted on the Kanmon Group and the Munakata Group. Two caldrons develop on the outer sides of the ridges along the axis of the strongest tidal current. The caldron on the south side of the north ridge was formed by the erosion of the tidal currents acted on the Munakata Group. The caldron related to the south ridge is not a result of abrasion by tidal current but an old submerged valley which have been free from sedimentation by biolent action of tidal currents.

View full abstract