ATMOSPHERIC ATTENUATION OF ULTRA-VIOLET LIGHT. 



By E. R. Schaeffer. 



Received March 29, 1922. Presented by Theodore Lyman. 



Synopsis. 



Absorption of ultra-violet light by the lower atmosphere. — Photographs 

 of the spectrum of a Cd spark, in the interval X3500-2500 were taken 

 from several stations where the air path ranged from 160 to 8000 

 meters. The apparent absorption for the shorter wave-lengths was 

 very marked as the distance between the source and the spectrograph 

 increased. A method of photographic photometry, similar to that 

 devised by Stetson to determine stellar magnitudes, was used to 

 measure the changes in relative intensities of the lines as the air path 

 was increased. The values of the relative intensities of the spectral 

 lines were plotted with the curve for molecular scattering computed 

 from the Rayleigh formula. The results agree with Strutt's work in 

 showing that the ozone concentration in the lower region of the 

 atmosphere is negligible but molecular scattering will not explain the 

 magnitude of the effect as has been previously supposed. The 

 absorption of light in this region of the spectrum by long columns of 

 oxygen and loss by ionization of the air may account for the observed 

 attenuation. 



Introduction. — It has long been known that the ultra-violet solar 

 spectrum ends quite abruptly near X2900. Cornu ^ found that the 

 limit depended upon the altitude of the sun and concluded that the 

 earth's atmosphere was the cause of the apparent absorption. On the 

 assumption that the concentration of the absorbing material did not 

 change with altitude, Cornu calculated the amount the spectrum 

 should be extended with increase in altitude. He then made observa- 

 tions at several stations from sea-level to a height of 2560 meters and 

 obtained experimental results in accord with this amount. 



About the same time Hartley ^ was studying a number of gases to 

 get information on absorption spectra. He photographed ultra-violet 

 spectra after the light had been transmitted through measured quanti- 

 ties of gas at atmospheric pressure. The oxides of nitrogen, CO2, 



1 Cornu, Journ. de Physique, 10, 1881. 



2 Hartley, Journ. Chem. Soc, ^9, 1881. Nature, p. 475, 1889. 



