86 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 65 



average distribution over land. This will give a deviation from the 

 assumed conditions and consequently a different absolute value to 

 the radiation, but it will probably only to a small extent change the 

 relative values and the general form of the curve. 



Melloni 1 concludes his first memoir on the cooling of bodies ex- 

 posed to the sky, published about 70 years ago, with the following 

 remarkable statement, upon which he seems to lay a certain stress : 

 "..'.'. Un corps expose pendant la nuit a Taction d'un ciel egalement 

 pur et serein se refroidit tou jours de la meme quantite quelle que soit 

 la temperature de Fair." 



One may at first be inclined to attach very little importance to this 

 statement. It seems in fact to be in contradiction with the most 

 elementary laws of radiation. If we consider the temperature of the 

 radiating surface as the only variable upon which the radiation 

 depends, we would expect the cooling of the body below the tem- 

 perature of the surroundings to be proportional to the fourth power 

 of its absolute temperature. At o° C. the cooling would for instance 

 be only about three fourths as much as at 20 C. 



Now the effect of temperature is generally a double one, as far as 

 the radiation process is concerned. With a rise in temperature there 

 generally follows an increase in the absolute humidity, which causes 

 an increase in the radiating power of the atmosphere. The increase 

 of the temperature radiation from the radiating surface is balanced 

 by a corresponding increase in the radiation of the atmosphere ; and 

 the observed effective radiation is therefore only subjected to a small 

 variation. The observations, discussed in previous .chapters, seem 

 now to indicate that the law of Melloni is approximately true with 

 the following modification : 



The cooling of a body, exposed to radiate to a clear night sky, is 

 almost independent of the temperature of the surroundings, pro- 

 vided that the relative humidity keeps a constant value. 



This conclusion, which can be drawn from the observations on the 

 influence of humidity and temperature on the effective radiation, 

 must be regarded as remarkable. It includes another consequence, 

 namely, that a high incoming radiation (sky and sun) and a there- 

 from resulting tendency to an increase of the temperature, is gen- 

 erally not counterbalanced by a corresponding increase in the 

 effective radiation from the surface of the earth to space. The vari- 

 ations of the incoming radiation are therefore, under constant tem- 

 perature conditions, almost entirely counterbalanced by variations in 

 convection, and evaporation (or other changes) of water. 



1 Melloni, loc. cit. (chapter II). 



