The purpose of this paper is to bring out, largely by means of observational material, outstanding features concerning the formation of hurricanes in the south Pacific. Because these data were few, fairly marked hurricanes had to be chosen for study, so that a sufficient number of observations might be obtained within one and the same hurricane. While the conclusions reached may have been influenced by this particular selection of cases, it is believed that the results are generally applicable. Now take up the storminess of the south pacific, longitude 160°E to 140°W, by island groups. It will be seen that the hurricane season(1)extends from December to April, inclusive. During this five month period about 90-95 per cent of the recorded storms have occurred. After making conservative allowance for the incompleteness of the record, and counting all gale-producing storms, not only true hurricanes, it appears that on the average fully a dozen tropical cyclones occur annually in the south Pacific. Tropical storms form most frequently in the south Pacific when the thermal equator is farthest south. The season when they are most frequent occurs during the winter and the early spring of the northern Hemisphere. The outbreaks of the cold air from the morth polar region move southward and surge against the Equatorial Front in the southern Hemisphere. But it is the opinion of the author that the hurricane in the south Pacific do not show any warm sector at the surface to the south of the hurricane. On account of the distribution of land and sea over the surface of the earth, both northern and southern air masses generally arrive at the Equatorial Front after a long journey over water. Both masses are there fore warm and humid in the surface layer. The occurrence of the outbreak of cold waves in the north Pacific from China to Kamtchatka inclusive is considered in Table 1. It is based on the data obtained from the weather maps analysed by the Central Meteorological Observatory, Tokyo, Japan. Table 1. Annual Frequency of Cold Waves in the North Pacific 1933-1939. The cold wave season extends from December to April, inclusive. During this five-month period about 20 cold waves have occured. After making allowance for the imcompleteness of the record of hurricanes in the south Pacific and the completeness of the record of cold waves in the north Pacific, it appears probable that the formation of hurrjcanes in the south Pacific is somewhat related to the outbreak of cold waves in Siberia. Utilizing this method for the hurricane in the south Pacific of the 8th to the 12th February, 1932, in which(2) there are appropriate data, a marked outbreak of cold air from the north polar region which moved southward and surged against the Equatorial Front in the southern Hemisphere is proved. The results are generally applicable for the hurricane of the 27th December, 1932, to 5th January, 1933, (1)and the hurricane of the 28th January to the 3rd February, 1936.(2) The work of S. Li(3)shows that the formation of typhoons in the North Pacific is somewhat related to the outbreak of cold waves in Australia. The conclusions stated above agree with that Li had to say regarding the formation of tropical cyclones in 1936.
(1) Im August wird die Tagesschwankung der Bodentemperatur in der Tiefe von 36cm 0.1°C, die Jahresschwankung wird im 9.9m Tiefe unmerklich. (2) Die Eintrittszeit des Maximums des täglichen Ganges verspätet sich um 4 Stunden pro Dezimeter und die des Jahresmaximums verzögert sich 30 Tage pro Meter. (3) Als die Temperaturleitfähigkeit haben wir aus der Amplitudenabnahme und aus der Phasenverschiebung des täglichen Ganges die Werte 0.0020-0.0040 c. g. s. abgeleitet. (4) Durch das Ausdruck wurde der Tagesverlauf der Wärmemenge, die durch die Bodenoberfläche ein-und ausgeht, bestimmt. Also könnte man nach Newton' schem Abkühlungsgesetze, eine empirische Formel annehmen und h, a wurden durch die Methode des kleinsten Quadrates berechnet. (5) Der Wärmegehalt des Bodens hat im März dentiefsten, im September den höchsten Stand, und die Amplitude des Wärmegehaltes beträgt 127.6 cal/cm2.
In this paper are shown some differencess between the climatic factors of the central part of Tokyo and those of its western suburbs. The materials used for this research was collected from following seven years, 1932-38. 1. Maximum Air Temperature Little, if any, difference is seen between the maximum temperature in the district from Kitizyozi to Tatikawa and that of the central part of the metropolis, but if we once cross the river Tama to Hino, it suddenly rises about 0.8°C higher, while at Asakawa the temperature is 0.8°C lower than that of the central part. Generally speaking, the maximum air temperature in the western suburbs is rather higher on clear days in summer time. 2. Minimum Air Temperature The minimum temperature gradually falls lower as we go through Kitizyozi to Tatikawa, but at Hino we see an abrupt fall. In summer it falls slowly, but in winter rapidly, -the minimum temperature in the western suburbs in winter is far lower than that of the central part, and it is not seldom that the difference is 4°C or 5°C, especially early in the morning on clear days. At Asakawa the minimum temperature is high as compared with that of Tatikawa and Hino. 3. Range of Air Temperature The range of air temperature in the western suburbs gradually widens as we proceed farther from the centre through Kitizyozi to Tatikawa, but once we cross the river Tama, it takes, a sudden and discontinuous turn, the range being the greatest at Hino. Asakaw a is in the remotest situation and also in the highest elevation, but the range is not so wide. This is probably because Takawo and Kobotoke Hills to the northwest of the village protect it from winds in the winter season, and moreover, the dense woods on the hills temper the change of temperature. 4. Amount of Precipitation The amount of precipitation in winter time is nearly equal in each place, but in July, August and September, it grows gradually greater as we go farther away from the centre to the west.
It has been theoretically proved that the diffusivity of small particles depends on the size of particles. And it has been derived that the diffusivity constant becomes a constant which coinsides with the kinematic viscosity when the size is very small; on the other hand when the size is large, diffusivity constant becomes very small compared with the kinematic viscosity.
It has long been noticed that the effect of wind velocity acts to diminish the amount of rainfall by a rain-gauge. Thus the rain-gauge installed at a higher elevation is considered to generally measure the smaller amount of rainfall compared with that at the lower elevation, since the wind velocity increases rapidly with height near the earth's surface. In the present paper this windeffect is examined statistically by inspecting the observational materials for the year 19.3-1929 at the Zinsen Meteorological Observatory. In this observatory one rain-gauge is installed 0.3m high above the ground surface and the other automatic recording rain-gauge 3m high. The author has made a carefull study on the difference of precipitation-amount between the above two gauges. The result thus obtained may approximately be expressed by the formula: y=1.2-1.6x (see Fig. 1), in which y denotes the difference of amount of precipitation between the two gauges in mm per day and x the mean wind velocity in m/s measured by the Robinson's cup anemomenter at 14m elevation. The same research has been extended to the Saisyu and the Genzan observatory and the analogous results have been obtained.
In the present paper the author deduced a formula to estimate the quantity of rainfall due to the discontinuous surface. Results of the calculation by the formula and the actual quantity of rainfall was compared in several cases. The agreement was generally good. It is necessary to consider the special mechanism of rainfall to account for the remarkable variation of the strength of rainfall in the heavy rainfall.