240 Prof. S. Arrhenius on the Influence of Carbonic Acid 



of the Moon,' with the aid of which it seems not impossible 

 to determine the absorption of heat by aqueous vapour and 

 by carbonic acid in precisely the conditions which occur in 

 our atmosphere. He has measured the radiation of the full 

 moon (if the moon was not full, the necessary correction 

 relative to this point was applied) at different heights and 

 seasons of the year. This radiation was moreover dispersed 

 in a spectrum, so that in his memoir we find the figures for 

 the radiant heat from the moon for 21 different groups of 

 rays, which are defined by the angle of deviation with 

 a rocksalt prism having a refracting angle of 60 degrees. 

 The groups lie between the angles 40° and 35°, and each 

 group is separated from its neighbours by an interval of 15 

 minutes. Now the temperature of the moon is nearly the 

 same as that of the earth, and the moon-rays have, as they 

 arrive at the measuring-instruments, passed through layers of 

 carbonic acid and of aqueous vapour of different thickness 

 according to the height of the moon and the humidity of the 

 air. If, then, these observations were wholly comparable with 

 one another, three of them would suffice for calculating the 

 absorption coefficient relatively to aqueous vapour and 

 carbonic acid for any one of the 21 different groups of rays. 

 But, as an inspection of the 24 different series of observations 

 will readily show, this is not the case. The intensity of 

 radiation for any group of rays should always diminish with 

 increasing quantity of aqueous vapour or carbonic acid 

 traversed. Now the quantity of carbonic acid is proportional 

 to the path of the ray through the atmosphere, that is, to the 

 quantity called " Air-mass " in Langley's figures. As unit 

 for the carbonic acid we therefore take air-mass = l, i.e. the 

 quantity of carbonic acid that is traversed in the air by a 

 vertical ray. The quantity of aqueous vapour traversed is 

 proportional partly to the " air-mass," partly to the humidity, 

 expressed in grammes of water per cubic metre. As unit for 

 the aqueous vapour I have taken the quantity of aqueous 

 vapour that is traversed by a vertical ray, if the air contains 

 10 grammes per cubic metre at the earth's surface *.. If we 

 tabulate the 24 series of observations published by Langley 

 in the work cited with respect to the quantities of carbonic 

 acid and aqueous vapour, we immediately detect that his 

 figures run very irregularly, so that very many exceptions 

 are found to the rule that the transmitted heat should 

 continuously decrease when both these quantities increase. 



* This unit nearly corresponds to the mean huniidhty of the air (see 

 Table VI. p. 284). 



