190 &. P. Langley—Selective Absorption of Solar Energy. 
EXAMPLE OF THE MODE OF OBSERVATION. 
As an example of the first class of measures, let us consider 
the observations made with the Hilger Prism on June 22, 1882. 
The high sun observation was made at 0" 15”. The suns 
zenith distance at this time was 17° 10’; the air mass* was 
me as at noon, and the air mass by the same formula was 
5:18 times that overhead, or 7:39 X 5°18 = 38-27 decimeters, 
so that the mass of air traversed in the second observation 
exceeded that in the first by an amount capable of supporting 
30°58 decimeters of mercury. 
The galvanometer deflection obtained in the part of the 
spectrum whose deviation is 44° 30’ (a part which is near the 
extreme lower limit of the present cbservations, far below the 
visible red) was at noon 17, and in the afternoon 11. In the 
violet, where the deviation is 50° 00’, the corresponding deflec 
tions were 45 and 0:39. Let us take these two feeble extreme 
rays as types with which to illustrate our process. Considering 
first the infra-red ray we have, deflection at noon = d, = 17, 
deflection in afternoon = d, = 11, difference in mass of alr 
traversed = M, 8, — M,?, = 30°58 decimeters, which, by 1 
absorption, has produced the difference in the deflections. 
¢ representing the amount of energy transmitted by a layer ° 
air equivalent to 1 decimeter of mercury, we find from the 
formula 
t= (M,6,—M,6,)4/2 
¢= ‘986; that is, a mass of air capable of supporting 1 deci- 
meter of mercury in the barometer, transmits 98°6 per cent of 
the energy of this particular kind of ray. This quantity ¢ we 
call the coefficient of transmission of the ray. 
nowing now the amount of energy transmitted by one 
such layer of air, we can find the amount transmitted by the 
7-74 layers which intervened between the observer and the 
sun at noon, namely 986" = 895. Only 89°5 per cent, there- 
fore, of the original unknown heat of the ray, which we will 
0°0174 Tabular Refraction 
cos, app. altitude. ae 
+ In general it is not advisable to make observations at so great a zenith dis- 
tance as this. 
* Computed from the formula M = 
