454 SCIENCE PROGRESS 



sphere. — Direct measurements made on the surface of the earth 

 of the solar constant of radiation — which is defined as the 

 radiant energy falling on unit area of the earth's surface— are of 

 course affected by any absorption which takes place in the 

 atmosphere. The determinations made of this constant, which 

 is of very important astronomical significance, have shown 

 considerable variations. Much careful research has been carried 

 out in recent years to determine the numerical value of this 

 constant, but on account of the varying results obtained, there 

 is still a deal of uncertainty even about the approximate value. 



Careful determinations made by Abbot and Fowle ! in 

 America, at an altitude of 14,500 feet, give a mean value of 

 1*922 calories per minute per square cm. of the earth's surface. 

 However, marked fluctuations, covering a range of 8 per cent., 

 in the radiation received were observed. In all cases the values 

 obtained agreed very closely with those made simultaneously 

 at an altitude of 5,800 feet. It was concluded that the absorp- 

 tion of extreme ultra-violet rays by the atmosphere did not cause 

 an error greater that 1 per cent, in the total radiation received. 

 The observed fluctuations were attributed to changes in the 

 actual emissivity of the sun. 



In conversation, it was suggested by Mr. R. Rossi to the 

 writer that the fluctuations are caused by ozone. While the 

 absorption of visible light rays by oxygen and nitrogen is 

 negligibly small, water vapour, on the other hand, has a 

 considerable influence. Ozone, if present in only small amounts, 

 would act similarly in causing a marked absorption of the sun's 

 rays. As mentioned above, the presence of this gas in the 

 upper atmosphere has been assumed as an explanation of the 

 fact that the solar spectrum ceases abruptly in the ultra- 

 violet at a point where ozone is known to have a deep 

 absorption band. The question of solar radiation had been 

 carefully studied by F. W. Very. 2 



Attention is drawn to the fact that where Nichols found 

 an apparent emission band in the violet region of the spectrum 

 of sky light, which attained a maximum about noon each day, 

 other observers have found here an absorption band. This 

 discrepancy is explained by Very on the supposition that the 

 absorption band originates from the illumination of the lower 



1 Astro-physical Journal (191 1), 33, 191. 

 « 7^.(1910,34,371; (I9i3),37,3i. 



