The problem of atmospheric oscillations is one of the most important problems in dynamical meteorology. Many authors treated the problem from various points of view, Their methods are to solve the differential equations of atmospheric oscillations, but the equations are, in general, much complicated and are solved only for the cases with constant lapse rate of temperature. In the present note we treat the problem with the perturbation theory, but numerical investigations are left for further studies.
The distribution of wind velocity near the earth's surface was calculated, assuming that the atmosphere is built up of two layers which have different eddy viscosities. And the following relation between surface and gradient winds was derived: where v1 and v2 are kinematic eddy viscosities of two layers, h thickness of lower layer, λ and η constants dependent on latitude. Next, the angle between wind directions and the ratio _??_ were measured from the synoptic chart and their mean values were calculated. It was found that the angle and ratio vary with the seasons and it was explained by the seasonal variation of the stability of lower layer.
One of the authors, Kato, has constructed an apparatus for measuring the total amount of radiation from the whole sky after the same method as Ångström's pyrheliometer, in company with Mr. K. Sinozuka, a member of the trial manufacture department of delicate machinery attached to the Physical Institute of the Imperial University of Tokyo. The chief part of the apparatus is composed of three thermopiles. The elements which form these thermopiles are extremely thin foils of manganin and constantan. Each thermopile is made of 14 elements and has electric resistance of 12.11±0.015Ω, and it needs 9 sec till the full electro-motive force caused by the constant radiation is measured. It is laid on a circular disk of brass having a diameter of 22mm. Each element is 0.8mm broad, 12mm long and 10 microns thick. These three thermopiles are denoted as A, B and C for convenience' sake. Each of A and B is furnished with a thin foil of manganin; this foil is 19mm long, 2mm broad and 10 microns thick, and is blackened with soot of camphor and jointed to the high temperature part of the thermopile. A slit of 19mm length and 8mm breadth is mounted up on the thermopile and manganin foil. C is jointed by an extremely thin foil of manganin to the high temperature part, but this foil is not blackened and is covered with thin plates of copper. Other parts of the thermopile are of the same construction as those of A and C. The principle of measurement is on the whole the same as that of K. Ångström's electric compensation pyrheliometer, namely: A and C are exposed to the radiation, while the manganin plate of B is heated by electric current. When the heat energy which A receives from the radiation is equal to what B receives from the electric current, the thermoelectric currents generated in the thermopiles will be also equal. This two thermo-electric currents are compensated each other by the zero method making use of the pointing galvanometer. Let the length of the manganin plate be l, its breadth b, the absorption coefficient a, the electric resistance per unit length r, and the strength of the current i, then the heat which B receives from the current will be_??_per unit time, where J denotes the Joule's constant. If the total amount of radiation falling on a horizontal surface per unit area per unit time is q gram calories, the beat which the manganin plate of A receives will be qb_??_a. Equating the two values, we obtain the following formulae as the total amount of radiation fallng on a horizontal surface of the earth per unit area per minute: In this equation i will be determined by practical observation, and r and a must be determined experimentally. The result of the measurement is r=0.2446 and a=99.85%. Substituting these values, the equation becomes Q=17.499 i2 It is necessary to explain the usefulness of the thermopile C. When the pile A is heated by the radiation, its part between the manganin plate and the slit is also exposed to the radiation and the heat produced by the latter is conveyed to the active junctions by heat conduction. This effect should be by all means removed. The pile C, as its construction indicates, only produces the thermo-electric current by the above effect, because the active junctions are unable to receive the heat directly from the radiation. Accordingly, in the practical observation the pile A is exposed to the radiation, together with C, and if the thermoleectric current produced in the latter is reverse to that in the former, this combination of A and C may be compared with the thermo-electric current produced by B in which the manganin plate is heated by electric current. Next, B and C are exposed to the radiation, there also the two thermo-electric currents are reverse each other, and A is heated by the current.
The actinographs of Gorcainsky's type were used for the measurement of the total amount of radiation from the whole sky. The observed records used in this investigation are those taken in the interval from April 1937 to February 1940. The chief part of the apparatus is a thermopile. The elements which form this thermopile are extremely thin fo_??_ls of manganin and constantan. The thermopile is composed of 14 elements and has electric resistance of 12.11±0.015Ω, and it needs about 9 sec till the full electro-motive force caused by the constant radiation is measured. It is laid on a circular disk of brass having a diameter of 22mm. Each element is 0.8mm broad, 12mm long and 10 microns thick, and is blackened with soot of camphor and is exposed to the radiation all over the whole length Owing to the good heat conduction of the junctions and to the small capacity of the active junction, it reaches its equilibrium of temperature instantaneously. The recording part of the apparatus is a recording millivoltmeter of ordinary type. made by the Shimazu Mauufacturing Factory. The portion which gives motion to the needle is just the same as the usual millivoltmeter with moving coil. In front of this part there is a drum rotating around a horizonatal axis by clock-work. The recording sheet is rolled up to this drum. The needle marks the values of the thermo-electric voltage generated in the thermopile every four minutes In this investigation, we used the records of clear days only. On clera days a beautiful sinusoidal curve is recorded. In this sheet we must obtain the time corresponding to the sun's altitudes of 10°, 20° and so on. It is possible to obtain the hour angle from the fundamental formulae of spherical astronomy giving the sun's altitude and declination and the latitude at the observing place. But, for convenience'sake, we used the graphical method of nomograph. To determine the scale value of this actiongraph, it is necessary to observe the absolute value of total radiation. In our case the new apparatus of direct measurement of the absolute value of the total amount of radiation from the whole sky was used. We can easi_??_y get the reduction factor of the actinograph by comparing the reading value r with the observed absolute value of the total radiation Q. The (Q. r) curve thus obtained is nearly a straight line. This fact indicates that the Berlage's formulae are in accord with our observation, and so we will be able to obtain conversely the coefficient of atmospheric transmission by comparing the recording sheet with the claculated value from the theoretical formulae given by Berlage. Thus we obtained the averaged annual variation of the transmission comefficient. In July and August the transmission coefficient of the sky is very small. This is of course attributed to the vast amount of the atmospheric water vapour in these months. The above investigations support, contrary to Reitz's objection, the formulae given by Berlage
The coefficients of correlation between the yield of sweet potatoes and the weather factors (monthly mean temperature, monthly amount of rainfall and monthly total hours of sunshine) have been calculated by the method previously reported in this journal, for each prefecture in Japan and for each month from May to November. The correlation coefficients calculated are shown in tables 1, 2 and 3, and althouhg they do not indicate well the influence of the weather factors on the yield of sweet potatoes. the following results are found from them at least. 1. The correlation of the yield with the temperatures in early time of growht and in harvest time is positive for every prefecture. and that of the yield with the temperature from July to August is generally negative for the other districts than the northeastern. The former correlation may be explained by the fact that the sweet potato is a crop of subtropical origin, while the latter correlation is explained by the other fact that high temperature, accompanied by sunny weather, often brings dorought injuries, especially in hte southerm part of Japan. 2. The correlation of the yield with the rainfall is almost positive in the southern district and negative in the nortern. That the local difference in the correlation of the yield with the rainfall is remarkable, is considered to be due to the local difference in soil nature. 3. The correlatian of the yield with the sunshine is similar to that with the temperature. This is because the sunny weather is liable to accompany high temperature.
It is generally said in Isigakizima that when the rain throughout spring is of small amount, strongy typhoon is expected in the following summer. This common saying was confirmed from the data of 18 years' observations. and it was attributed to the state of upper air currents.