OZONE IN THE UPPER ATMOSPHERE 453 



assumed that if all the light proceeding from this region were 

 polarised, its origin would be solely due to the diffraction of 

 molecules and small particles. The measurement of polarisation 

 gives accordingly a method of ascertaining definitely the part 

 played by selective scattering. The result of such measurements 

 is to show that the light reflected at this angle of 90° is by no 

 means completely polarised, and that the proportion of polarised 

 to total light varies very largely from time to time. 



Some recent measurements made by Boutaric 1 in Switzerland 

 have shown that the degree of polarisation of light scattered at 

 an angle of 90 varied between o - 4 and 07 of the total light. 

 Measurements were also made on the constant of solar radiation 

 which is discussed below. The degree of polarisation was found 

 to vary concomitantly with these radiation values, or in other 

 words, inversely as the absorption of the atmosphere. The 

 variation in these values of the polarisation, and deductions which 

 have been made from measurements on the relative luminous 

 intensities of sunlight and skylight, have shown that it is necessary 

 to assume that a large amount of light is reflected from the sky 

 under conditions which do not conform to the theory of selective 

 scattering. This is probably due to the reflection of light from 

 particles which are large compared to the waves of light, and 

 also to some extent from direct illumination by light reflected 

 from the earth. The admission of these sources of light opens 

 the possibility of the operation of such factors as the colour of 

 the air itself due to elements which exert a selective absorption. 



Experiments have been carried out on the degree of polarisa- 

 tion of the light of different wave-lengths proceeding from the 

 sky. It was found that if blue rays are removed from this light 

 by causing it to pass through a medium of complementary colour 

 (red), so arranged that the sky appeared white when viewed 

 through this liquid, then the light thus filtered showed exactly 

 the same degree of polarisation as when measured after proceed- 

 ing directly from the sky. It would appear from this that the 

 blue light from the sky is not polarised, but that the polarised 

 light is white, and that the blue colour results from the 

 absorption by the air of reflected non-polarised light, or else 

 possibly it is produced by fluorescent phenomena in the atmo- 

 sphere. 



Evidence of Solar Radiation on Composition of Upper Atmo- 



1 Le Radium (1914), 11, 15. 



