With the numerical calculation of the theoretical formula of primary and secondary Rayleigh scattering in a plane-stratified atmosphere of uniform composition, an ifluence of variation of total scattering coefficient of the atmosphere on the angular distribution of the intensity of ultraviolet sky radiation in the sun's vertical plane was discussed, when the sun's elevation angle was 30° and 60° respectively. Taking the intensity at the zenith as a unit, the relative intensities from varying zenith angle were computed and were computed and were compared with the observed values of C. Dorno and F. Lindholm or O. Eckel which were obtained at the position of varying elevation from sea level. The theory shows that the relative intensities at a comparatively large zenith angle especially at 60° always increase with decreasing optical thickness of the atmosphere and these were confirmed by the observed results at both of the sun's elevation angles.
In this paper, the author deals theoretically with the vertical turbulent diffusion of minute particles emitted continuously form a line source in the lower layer of the atmosphere, taking into consideration the non-homogeneity of turbulence in that layer. Assuming the scale of the largest turbulent element at any point to be in proportion to the height from the ground surface, and also assuming Taylor-Karman's correlation coefficient of vertical turbulent wind velocities to be of a reasonable functional form, the shape of diffused smoke is determined theoretically, following the same procedure as that used in previous papers of the same author. By means of results obtained here, it is to some extent made clear how far the turbulent diffusion in the homogeneous atmosphere differs from that in the lower layer of the real non-homogeneous atmosphere.
A comparative and critical study of the angular-momentum-transfer and vorticity-transfer methods in the investigation of zonal circulation of the atmosphere is given. The essential difference between the two methods is that the fluctuating variations of atmospheric pressure in the west-east direction, which certainly exist in a turbulent field of flow, are ignored in the former method, while in the latter the variations are taken into account. According to the former method, the displaced particle moving towards the pole gains the west component of wind speed and the displaced particle moving towards the equator loses the west component of wind speed. While, according to the latter method, the displaced particle moving north or southward usually gains the west component of wind speed. Very high velocities will result when a mass is moved horizontally from a given latitude towards the pole or equator under conservation of its angular momentum, while quite reasonable velocities will be given when a mass is moved horizontally under conservation of its absolute vorticity. Such an essential difference can be readily seen from Tables l•a and l•b.
If we wind the coil so as to be uniformly distributed along the core, it will be able to reduce the error of angle transmission to about ±3°, and it is clear that D. C. self synchronous-motor can be utilized as the wind direction meter. And if we use as the core an alloy of constant permeability, for instance, dust core, the error will be decreased to a half of the former.