OZONE IN THE UPPER ATMOSPHERE 451 



Spectro-photometric measurements have been made by a 

 number of workers under different conditions, such as varying 

 altitude and latitude, with clear and overcast skies, and at 

 different hours of the day and times of the year. Recent 

 measurements have clearly shown that in some cases the 

 curve of distribution of intensity follows the general direction 

 demanded by Rayleigh's formula, but that as a rule the intensity 

 in the violet and blue part of the spectrum exceeds the theo- 

 retical value in a very variable manner, and in some cases 

 amounts to about double the normal value. Measurements 

 made by E. L. Nichols 1 in Switzerland and the Tyrol showed 

 that in the early morning curves agreeing closely with the 

 theoretical values were given, but later in the day a com- 

 paratively large upward peak developed in that part of the 

 curve representing the blue region of the spectrum. This 

 peak reached a maximum in the early afternoon, and then 

 diminished ; the effect was particularly marked at high altitudes. 

 The nature of this peak in the curve has the appearance of 

 an emission band superimposed on the normal spectrum of 

 the sky, and lends support to a view which has been put forward 

 that ozone, which is known to be a fluorescent gas, plays a part 

 in the illumination of the sky in virtue of this property of 

 fluorescence. An alternative view is that this comparative 

 increase in the intensity of the violet is due to the selective 

 absorption of the atmosphere for ultra-violet on one side and 

 green on the other, an effect which also harmonises with the 

 presence of ozone. Measurements which have been made to 

 show the relation between wave-length and intensity of sky 

 light, as reflected at an angle of 90 to the incident light, have 

 shown that the exponent of 4, which is demanded by Rayleigh's 

 formula, does not apply at all generally. This factor shows 

 very large variations. A continuous increase has been traced 

 in proportion to the zenithal distance of the sun, and it has 

 been found to vary with the humidity of the atmosphere. 

 These discrepancies and variations show that factors other than 

 scattering operate in the illumination of the sky. 



Colour of Setting Sun. — In considering selective absorption 

 by the atmosphere, it is obvious that the phenomenon of scatter- 

 ing which causes reflected light to be blue leaves the transmitted 

 light red. In consequence of this, light coming from the sky is 



1 Physical Review (1909), 28, 122. 



