It is generally believed that river piracy is a rare phenomenon in our mountainous country. In Jaapn, however, there are many upheaved peneplains where river piracy often occurs, and we have many examples of crustal movements which are the direct causes of river piracy. On account of these facts, various examples of river piracy are found in Japan, and these phenomena took place by way of surface streams and subterranean capture of surface streams, and faulting tilted block movement, tectonic line and selective erosion and so forth. In Japan, spill-overr is due mainly to volca-nic activity and river load. As regards regions, Chugoku and Abukuma districts attract our attention with their abundance of examples of river piracy and spill-over.
The author did research on the number of passengers at each station, the distances covered by the passengers and railway traffic density in Hokkaido during 1949. The research was based on railway passengerr statistics of the same year. Passenger-traffic density between Sapporo and Otaru is more than 66, 000 persons per day (single trips), while in local areas, less than 250 persons (Fig. 1). It is observed that there is negative correlation bettieeu the numbers of passengers and the distances covered (Fig. 2). The amount of passenger-traffie between the two principal cities is the function of the populations and the distances between the two. For enovenierce sake the following marks are used: n=number of passenger in one year, A=population of a city (in 10, 000). B population of the other city (in 10, 000). Then a formula m=_??_ is made. In this case “m” is called the traffic rate. It is noticed that the following relationship exist between “m” and “D”km. (distance between the two cities): log. m=a-b log.D The absolute value of. “b” decreases in the case of a larger city, while it increases in the case of a smaller town (Fig. 4). In Hokkaido the frequency distribution of each station based on the number of its passengers has logarithmic normal distribution (Fig. 5). Especially at those stations where the numbers of passengers are from 1, 000 to 100, 000 (in a month) the distribution is in a straight line on the probability paper (Fig. 6). Therefore the author classified those stations, in strata, and indicated their distribution on the map (Fig. 7). Research was made of the distances covered by passengers at 25 stations. The average distance covered is 61km. in Hokkaido. At large stations more than 75% of all passengers make trips shorter than the average distance, and at small stations more than 90% make shorter trips. The percentage of passengers who make trips shorter than 30km. is very large. The average distance covered by the passengers in Hokkaido is the longest throughout Japan. Yet even in this area the passengers making trips shorter than 60km. constitute a great majority. This feature may be said. to be one of the characteristics of Japanese railway passenger traffic.
The author has studied the problem of how many people can live at the present state of water supply dependent mainly upon rain water in the volcanic island of Oshima. In this island, all the precipitated water upon roofs of rural houses is collected into the Avater tanks of each house. Inve-stigations of amounts of water necessary for rural houses were done at Moto-mura, Nomashi and Habu-no-minato village The results are as follows; 1) The relation between the number of people and the amount of water needed by families at each village is expressed in the following experimental formula. Y=AXb Y signifies amount of water needs, X means numberr of people in a family, and A and b are constant values. At several villages, he obtained the following values. Moto-mura Y=69X0.5 Nomashi Y=52X0.5 Habu-no-minat o Y=39X0.5 2) The increasing coefficient of water needs of rural families was equal to each other or all villages, but basic amounts of water needs (A) showedd dillerent values. The value of A was the biggest at the most humid village, and tine smallest at the most arid one in the island. 3) Using the former formula, he calculated total amounts of water needs at all rural families in Oshinia island underr the present conditions, and he estimated the total amount of water needs by cattle, ships, factories, hotels, etc.. Adding these two values, mentioned above he found the total amount of water needs in Oshima. This value is only 0.016% of rain water on Oshima island.
The author made a local climatological observation on humidity distribution at Uchihara, Iberagi Pref.. There is a flat diluvial plateau surface dotted with forests, paddy fields and upland fields. He studied effects of such surface covers on the distributions of humidity in daytime in late summer. 1) In mobile observation, unavoidable errors are caused by microfluctuations of atmospheric conditions and by assuming that the time change is the same at every place in the whole area, but these two kinds of errors do not seem to exceed ±2 %. 2) In oiler to lessen the latter errors, he made a new attempt of repeated ob ervations at the same place at different times. He got a revised time series of the reduction terms. 3) Humidity differences between various sorts of surface covers reached 5-10%. The most humid areas were forests and paddy fields. Upland fields were less humid, and barelands and grasslands were still drier.