78 REPORT OF THE. SECRETARY. 



A part of the measurements have heen made in Washington, and therefore 

 practically at sea level, and a part at Mount Wilson, in California, at about 

 1,800 meters, or nearly 6,000 feet elevation. The radiation of the sun has been 

 studied, not only in the total, but also as dispersed into its spectrum, and not 

 only in the part visible to the eye, but also in those portions whose wave lengths 

 are too long or too short to affect the eye. For all these different rays the 

 earth's atmosphere produces different degrees of absorption, or of diffuse reflec- 

 tion, and in the course of the work the transparency of the earth's atmosphere for 

 many different rays has been extensively investigated. The reflecting powers 

 of the clouds and the air have been measured, and also the quality of the 

 sky light as regards the relative intensity of its rays of different colors. 



We use as our unit of measurement that intensity of radiation which, when 

 fully absorbed for one minute over a square centimeter of area, placed at right 

 angles to the ray, would produce heat enough to raise the temperature of a gram 

 of water 1° centigrade. This unit is termed 1 calorie per square centimeter per 

 minute.^ 



The mean result of 130 measurements conducted on Mount Wilson in the 

 summer and autumn months of 1905 and 1906 fixes the intensity of solar radia- 

 tion outside the atmosphere at mean solar distance as 2.023 calories per square 

 centimeter per minute. 



The mean result of 41 measurements at Washington from 1902 to 1907 is 

 2.061 calories. 



It is probable that the mean result of such measurements, if conducted for a 

 long term of years, would be higher, and the probable mean value of the solar 

 constant may be estimated in round numbers at 2.1 calories per square centi- 

 meter per minute. 



Expressed in another way, the solar radiation is capable of melting an ice- 

 shell, 35 meters (114 feet) thick, annually over the whole surface of the earth. 



The results of Langley, while seemingly in contradiction of these, in reality 

 support them. For, as he states on page 211 of the Report of the Mount 

 Whitney expedition, his value (3 calories) for the "solar constant" depends 

 upon an allowance which he made for an apparent " systematic error in high 

 and low sun observations at one station," of such a nature as becomes manifest 

 "by calculating at the lower station, from our high and low sun observations 

 there, the heat which should be found at a certain height in the atmosphere, 

 then actually ascending to this height, and finding the observed heat there con- 

 spicuously and systematically greater than the calculated one." As shown in 

 Chapter VII, Part I, of the present volume, this seeming discrepancy arose from 

 a misapprehension of the requirements of the calculations. In fact, there is 

 no such systematic error, no correction for it should have been applied by 

 Langley, and the best mean value of his experimental determination of the 

 " solar constant " at Mount Whitney and Lone Pine is 2.14 calories per square 

 centimeter per minute. 



Substantial agreement as to the magnitude of the "solar constant" is there- 

 fore reached by observations at sea level, at 1,800 meters, and at 3,500 meters 

 elevation. 



The solar radiation is far from being constant in its intensity. The values 

 determined on Mount Wilson range from 1.93 calories to 2.14 calories, and 

 those in Washington from 1.89 calories to 2.22 calories. A change of the in- 

 tensity of solar radiation of 3| per cent, due to the decrease in solar distance, 

 occurs from August to October, and this is readily discernible in the work done 



°As above stated, it is possible that the numerical results to be given in Vol. 

 II of the Annals may be 1.5 per cent higher than they should be in these units. 



