ATMOSPHERIC ATTENUATION OF ULTRA-VIOLET LIGHT. 373 



ganic matter would be certain to prevent an appreciable concentration 

 of that gas in the lower atmosphere. 



Results. — The result of the present paper shows, first, that the con- 

 centration of ozone near the surface of the earth is negligible — as may 

 be expected from the properties of ozone given above; second, the 

 attenuation of ultra-violet light by the lower atmosphere cannot be 

 explained by atmospheric scattering alone. There are several other 

 agents which, no doubt, ha^'e some effect. The absorption by a long 

 path of oxygen, and the ionization of the air by the light of wave- 

 length under consideration deserve attention. Sulphur dioxide and 

 methane absorb ultra-violet light, but no data are at hand which show 

 the effect to be important for concentrations of these gases too small 

 to be detected easily by chemical means. 



Little is known about the absorption of long columns of oxygen in 

 this region of the spectrum. The effect of oxygen on shorter wave- 

 lengths is well known; Kreusler ^^ found that a column of air 20 cm. 

 long absorbed about 9 per cent of the light at X1860, while he detected 

 no absorption at X1930. By compressing the gas in a tube, Liveing 

 and Dewar ^° extended the absorption band of oxygen into the longer 

 wave-lengths. In one experiment the spectra of the carbon arc and 

 of the iron spark were photographed through a tube 18 meters long 

 containing oxygen under pressure of 85 atmospheres. Under these 

 conditions the gas was found to be quite transparent for violet and 

 ultra-violet rays to about X2745. From that point, the light gradually 

 diminished, and beyond X2664 appeared to be wholly absorbed. 



The mass of oxygen used in this experiment is approximately the 

 same as that traversed in a column 8000 meters long at N.T.P., but 

 the absorption is no doubt somewhat more at the higher pressure than 

 for an equal mass at a lower pressure. But the shoulder of the great 

 absorption band in the Schumann region may extend far enough into 

 the longer wave-lengths to affect the intensity of the light appreciably. 



The formation of ions by the action of ultra-violet light on air may 

 be important. Bloch ^^ found that strong ultra-violet light of wave- 

 length greater than 1800 produced a certain ionization on inifiltered 

 air. Lenard and Ramsauer ^^ verified this, but found there was no 

 volume effect on the pure gases of the atmosphere and that the effect 

 on air containing dust depended upon the presence of moisture. No 



19 Kreusler, Ann. d. Phvs., 6, p. 412, 1901. 



20 Liveing and Dewar, Phil. Mag., 26, p. 286, 1888. 



21 Bloch, Comp. Rend., 146, p. 892, 1908. 



22 Lenard and Ramsauer, Sitz. Ber. d. Heidelberger Akad. 1910-1911. 



