476 MELLOM ON THE NOCTURNAL COOLING OP BODIES. 



nocturnal cooling of snow, have been obtained by a method 

 which we believe to be inaccurate, in consequence of the unequal 

 circumstances in which the thermometers used to measure the 

 temperatures of the air and of the radiating bodies were placed. 

 If properly measured, the cooUng of the snow and of the swans- 

 down would certainly be less; but it matters little what the 

 absolute value is here, since we are only considering the con- 

 stancy of the effect under variations of atmospheric temperature. 



The results which we have announced prove therefore the truth 

 of the proposition above asserted, namely, that a body exposed 

 during the night to the influence of a sky of equal clearness and 

 calmness is always cooled to the same extent, whatever may be 

 the temjjerature of the air. 



This fact alone, thoroughly established by experiment, will 

 lead us to a clear and complete explanation of all the pheno- 

 mena which precede and accompany the formation of dew. 



