1881.] 



the Atmosphere which absorbs Radiant Heat. 



219 



radiant heat. We can inquire "which constituent of the atmosphere is 

 subject to a diurnal variation most resembling that of the absorption 

 coefficient, and also which constituent is distributed vertically in a 

 manner most closely approaching to that of the absorptive substance. 



If the coefficient K were truly constant for each day, as has been 

 assumed, the positive and negative differences between the observed 

 and computed values of r should be irregularly distributed ; but we 

 find that, almost without exception, the computed values are too low 

 in the forenoon and too high in the afternoon at Mussooree, while at 

 Dehra there is no such regular order in the differences. The excess of 

 the morning over the afternoon radiation at Mussooree and their 

 approximate equality at Dehra are very clearly seen in the averages of 

 the three sets of observations, which are the following : — 



Hour 



8 



9 



10 



11 



12 



13 



14 



15 



16 







Mussooree . . 



790 



898 



946 



964 



967 



949 



927 



875 



763 





659 



779 



845 



880 



885 



880 



849 



779 



600 





The departure from complete symmetry at Dehra is due chiefly to 

 one observation at 4 p.m. on the 4th November, 1869, when probably 

 there was a slight haze partially obscuring the sun. 



The absorption of heat by the atmosphere on a clear, calm day 

 cannot be, in any appreciable degree, a mere apparent effect due to 

 the scattering of the rays brought about by disturbances set up in the 

 atmosphere through heating from below ; for any variations in the 

 a-mount of such scattering must be even more apparent at the lower 

 station that at the upper one, because, to get to the lower station, the 

 rays must pass through more of the disturbed atmosphere. Moreover, 

 such disturbances, and the apparent absorption caused by them, would 

 probably be at least as great during the hours of rising temperature 

 (up to about 1 p.m.) as when the temperature is falling. The increased 

 absorption of the afternoons at Mussooree must therefore be due to 

 the fact that more of the absorbent substance — whatever it be — lies 

 above the level of the station in the afternoons than in the mornings, 

 while above the lower station the total quantity of this substance is 

 practically constant throughout the day ; that is to say, the absorbent 

 substance is carried upwards during the day and probably sinks down- 

 wards again at night. 



Such an upward movement of the total atmosphere occurs under 

 the influence of diurnal heating, but what proportion of the air that 

 is lifted above Mussooree by expansion from below remains there 

 cannot be exactly determined. The barometer falls from 10 a.m. to 



