48 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 65 



value of 0.439 caL f° r ver y large values of p, as one would expect 

 from the expression that has been adopted. On the other hand, the 

 observations bring us very near the zero value of humidity and the 

 question arises, whether we may not be entitled to attempt an extra- 

 polation down to zero without causing too large an error in the limit- 

 ing value. We wish to answer the question : how does the atmosphere 

 radiate, if there is no water vapor in it? As I have pointed out 

 previously, the possibility of an extrapolation to zero is doubtful, 

 because in the non-homogeneous radiation of the water vapor there 

 are certainly terms corresponding to wave lengths, where even very 

 thin layers radiate almost to their full value. Consequently these 

 have scarcely any influence upon the variations of the radiation from 

 thicker layers. Will the curve that gives the relation between the 

 radiation and the radiating mass of water vapor for values of the 

 humidity lower than 0.4 show a rapid decline of which no indication 

 is apparent in the investigated interval 0.4—12 mm.? For compari- 

 son I may refer to a curve drawn from a calculation by N. Ekholm 1 

 of the transmission of water vapor according to Langley and 

 Rubens and Aschkinass. The curve represents the radiation from 

 a black body at 15 ° temperature as transmitted through layers of 

 water vapor of variable thickness. The same curve evidently also 

 gives the radiation from the identical vapor layers, provided that 

 the law of Kirchhoff holds, and that the water vapor itself is at 15 . 

 As far as the result may be depended upon, it apparently shows 

 that laboratory measurements give no evidence whatever of a sudden 

 drop in the radiation curve for very thin radiating layers. It would 

 be rather interesting to investigate the radiation of the atmosphere 

 compared with the radiation of the water vapor and of the carbon 

 dioxide and possibly also that of the ozone contained in the upper 

 layers, with proper regard to the temperature conditions and to care- 

 ful laboratory measurements on the absorption and radiation of these 

 gases. A first attempt in this direction is made by Ekholm. How- 

 ever, it appears to me that he does not give due attention to the fact 

 that the magnitude of the effective radiation to space depends upon 

 the capacity of the atmosphere to radiate back to the earth, and 

 only indirectly upon the absorption capacity of the atmosphere. 

 Quantitative calculations of the radiation processes within the atmos- 

 phere must necessarily take into consideration the temperature con- 

 ditions in various atmospheric layers. The laboratory measurements 

 upon which such a computation should be based are as yet very in- 



1 Met. Zt., 1902, pp. 489-505. 



