1. The author has experimentally proved that the hygroscopic nature of volcanic ashes mainly depends upon their water soluble sublimates. The chemical compositions of the water soluble sublimates are as follows. 2. There is a distinct group distribution on the size of ashes, for example Mt. Asama of which the larger one consists the main part of the ash. When we observe the fresh samples, we can see that the smaller ashes adhere on the larger particles, as if they are small magnets and iron powder. The force between these particles is perhaps the adhesion force, and not the electrical or magnetic force.
In this paper were investigated experimentally the relations between the rate of evaporation from several sorts of soils and the quantity of water hold in the soils. In order, to be free from the complex meteorological influences, we adopted the ratio of the evaporation from soils to that from free water as the representation of the rate of evaporation. Soils used were divided into 4 classes according as the magnitude of their grains; (1) 2.0_??_0.55mm., (2) 0.55_??_0.145mm., (3) 0.145_??_0.075mm., (4)<0.075mm. Results of experiments, are as follows:- (1) When every kind of soils holds equal quantity of water near its surface, the rate of evaporation is as the first approximation independent of the magnitude of the soil grains. (2) But by more precise measurements it was found that: (a) Soils of large grains (soil (1)) are permeated by water very quickly but cannot suck it up till high place. Therefore the rate of evaporation from them is in the beginning pretty fast, but the surfaces of them dry up soon and evaporation becomes faint before long. (b) The rate of evaporation from grains of moderate magnitude (soil (2) & (3)) is largest. (c) With regard to soils of small grains (soil (4)) water penetrates most slowly, but can go up highest by their strong capillary action, so that the rate of evaporation is moderate but does not decrease till the quantity of water in them falls to scanty. (3) After the sruface of soil dries up the rate of evaporation decreases very quickly and it is nearly inversely proportional to the depth of the dried portion of the soil. (4) Except the effect of solar radiation most meteorological influences were same on the evaporation from soils and that from free water. As to the effect of solar radiation it was found that the above mentioned ratio is about 20% larger when the sun is shining (during day-time) than when there is no sun-shine (during night). But the effect of the sunshine is not so serious if we consider the daily mean value.
This paper presents theoretical basis from the thermodynamical standpoint in favou of C. G. Rossby's priuciple for the atmospheric turbulence vez., that lateral mixing is greatest where vertical mixing is the least. That is, the more stable is a layer of atmosphere the more rapidly the lateral mixing will take place.
It is important not only from the climatological but also from the agricultural and industrial point of view to ascertain the variability of summer precipitation (May to September) in Japan. Although many works have been undertaken, there is no fundamental guidance exactly to determine the coefficient of variability. Present work was intended to determine the variability of summer precipitation in Japan. The results obtained are presented as follows: (1) The standard deviation of frequency distribution of the summer precipitation for 53 meteorological stations hitherto since 40 years was calculated; it was found that the correlation between the standard deviation of frequency distribution above mentioned and the mean amount of summer precipitation was highly intimated, and the coefficient of correlation was as follows: +0.938±0.011. From this, we see that as the variability coefficient it could be cboosen the standard deviation already described. As numerical relation between the variability coefficient x and the mean amount of summer precipitation y (cm) we get x=0.0118y+0.1392 (2) The geographical distribution of the variability as represented on a map (Fig. 3) gives interesting results. The Pacific coast from southern Kyûsyû to Izu peninsula has greater variability than that of the other portion of Japan, and the variability decreases with increasing latitude. Hokkaido has lowest variability. (3) The variability after Maurer's scheme was also calculated. It was found that the summer precipitation fluctuates within a wider range from year to year than that of annuals.
The 27-day recurrences of magnetic activities were treated by many authorities, but J. Bartels said that no trace was found of the existence of any disturbance-interval which is an exact submultiple of 27 days. In this short report, I have examined whether the 14-day recurrences exist or not. After K. Takahasi and K. Husimi's method, which is applicable to the determination of the period and the damping factor of a vibrating system exposed to irregular forces, the character figures for calcium flocculi, the sun-spot numbers, the international character numbers and the numerical character numbers at Kakioka Magnetic Observatory, were treated. The materials used were taken from April 1 to Oct. 24, 1930, for all the four elements. In the recurrence-diagram we find that not only 27-day recurrences but also 13-14 days recurrences are apparent, and moreover, the maxima of these four curves occur almost simultaneously. These facts show that the new appearance of eruptive phenomena on the solar disk, are the main causes of the terrestrial magnetic activities, and moreover, that the phenomena occur simultaneously on nearly opposite sides of the Sun. But it cannot be concluded with confidence that such two regions are antipodal or symmetrical. According to the result of calculation of the variability made by Dr. H. Hatakeyama, using the observated values at Toyohara Magnetic Observatory during the period between VIII 1932-VIII 1935, it seems that the 27-day recurrence lasts about 4 or 5 months, and it constitutes one system. And it is also interesting that the phase difference between the two consecutive systems are often nearly 13 or 14 days.
The author found that at the great part of the meteorological stations in Japan Proper the seculer variation of the monthly mean air temperature of February and that of December correlated as follows; i) the positive correlation for 1905_??_1918 (or 1919), ii) the negative correlation for 1918 (or 1919)_??_1933 (or 1934), that is, the correlation turned into negative from positive in the year 1918 or 1919. The distribution of this turning year resembles the winter path of “Kuro-sio” along the Proper coast and the year 1917 is the most maximum year of Wolf's number of solar spot in a good long period. Thus the turn of this correlation is likely to have a relation with Kuro-sio along the Proper coast. Indeed it has been reported by Mr. Uda that the abnormality of Kuro-sio at the southern Offing of “Kisyü” had caused for 1905_??_1906, 1916_??_1918, and 1934_??_. Since about the year 1934 each seculer variation is at random one another, but the positive correlation will begin about that maximum year of solar spot number.