398 
Mil. LOUIS VESSOT KING ON THE SCATTERING AND 
expressions for sky radiation liitherto olRained ; in fact the whitish colour of the 
sky' near the liorizon was considered by Kei.vin"^ to he an objection to the theory of 
scattering. 
By calculating the yalues of the intensities of solar radiation which readies the 
earth’s surface for various zenith distances of the sun and for various wave-lengths, 
and by Integrating the curves so olitained, we obtain the values of the total intensity 
of solar radiation reaching the two stations Mount Wilson and Washington. Taking 
the value of the solar constant to be 1’922 calories per square centimetre per 
minute, t the results are given in the same units and are calculated for solar 
radiation incident on a plane normal to the sun’s rays as well as for the solar radiation 
incident on a horizontal plane. The results are given in Tables VII. and IX., and 
are shown graphically in Diagrams IX. and X. 
These results enable the value of the total sky iiitensity in any direction to be 
compared with the intensity of direct solar radiation at the station in the form 
sl'i//sun for equal solid angles (semi-diameter of sun taken as IG' of arc). The results 
are given in Tables VII. and IX., both for Mount Wilson and Washington, and are 
shown graphically in Diagram XI. 
By interpolation from the diagram just mentioned the values of skyjsnn for 
various directions of sky and for various zenith distances of the sun are compared 
with the values observed at Mount Wilson.| The comparison is made in Tables XI. 
and XIII. ; the results of calculation are in fair agreement with observation 
except for regions of sky near the sun and near the horizon. The first of these 
discrejiancies is prol)ably brought about by the simplifying assumption made in 
equation (17) by writing g for //(9) and //(rr'). The existence of such a term 
depending on an angular co-ordinate complicates the integral equation beyond hope 
of solution ; it can be seen, however, that its existence gives rise to a bright region 
of sky in the neigldjourhood of the sun not represented l)y tlie approximate solution 
considered in the present paper. Tlie discrepancy which exists in the case of direc¬ 
tions of sky near the horizon is due to the failure of the approximate formula (66) 
for large values of (p. In such cases the complete fornnda (60) should be employed. 
The use of this fornnda would require the tal)ulation of functions of two and three 
variables, which might be undertaken when more numerous and more accurate 
observations on sky radiation are available. It must be remembered in making these 
comparisons that the result of sky observations on certain specified days are com¬ 
pared with values based on mean coefficients of attenuation. The (mly satisfactory 
method of making the comparison is to olRain sky radiation observations on the 
same day as observations are made for tlie coefficients of attenuation of solar 
* Kelvin, loc. rif., p. 307. 
t Abuot, C. G., and Fowle, E. E., “ 
Journal,’ XXKIII., April, 1911, p. 191. 
I ‘ Annals,’ vol. II., Table 32, j). 151. 
The Value of the Solar Constant of Ihidiation,” 
‘ Astrophysical 
