C oS C athe - — eee but rather that the scatter coefficient is a complicated function of the ratio of the particle diameter to the wave length of the radiation, As a matter of fact the scattering area coefficient, Kg, versus & curve shows at least two maxima and two minima and according to Houghton probably such maxima and minima continue to oscillate about the value of Kg = 2 with decreasing amplitudes. However, such va: ations of ransmission with wave length will be evident only for small dand for monodisperse aerosols, In natural aerosols such as fog and clouds, « is so large and the drop size distribution is so broad that no variation of trans- mission with A is expected to be -vident. verified by the frst that the s. viewed through fog or thin cloua:. This conclusion is s disk appears white when Stratton (6) was able to produce artificial fog in the laboratory using steam and natural nucleating agents found in the air, This fog was composed of water droplets which were considerably smaller than in natural fog. Stratton found a definite variation of transmission with “s wave length. At o& # 11,2 and using A * 0.49 micron there was a maximum transmission, Hence the radius of the fog particles was calculated to be 0.875. This is a remarkable confirmation of Mie's theory of scatter and an experimental verification of the first minimum in the Kg versus & curve, Since Equation & applies only to a civen wave length, to obtain the correct value for solar radiation the value of I should be integrated over the range of wave lengths in sunlight. In order to simplify the computational problem, it will be assumed that solar wave length is equal to 0.57 micron, And because it is known that the atomic cloud is not a monodisperse aerosol, probably the Jobst asymptotic curve of Ks versus«would produce more realistic results than the complex Kg versuscurves. If the particle size distribution in the atomic cloud were known with some accuracy, it may be worthwhile to determine I by the summation process indicated in Equation 8 and then to integrate I over the range of q values. -As it is, there is no merit in such a procedure until the particle size distribution in the atomic cloud is better known, It will be assumed that m # 1.55, and it will also be assumed that the particles are transparent since the absorbtivity of the dust particles is not known, Under these assumptions Kg may have a maximum walue of approximately 4 when ¢ has a value between 3 and 4, and it will be assumed that Kg reduces asymptotically from its maximum value to a value of 2 at« = 50, Ce alculations of Layer Aloft the Reduction of Insolation due to the Dust It will be assumed that the numerical median particle . distribution in the atomic cloud 1s approxinately 0.6 microns in diameter for large yield atomic bombs or thermonuclear weapons (10 megatons) exploded on the surface or underground. However, since the particle size distribution of the atomic bomb cloud is admittedly not known with any high degree of accuracy, calculations 9008832 6°: C2~38529