SOLAR RADIATIOISr ABBOT 183 



THE MEAN VALUE OF THE SOLAR CONSTANT OF RADIATION 



The solar-constant methods just described were used exclusively 

 at Washington from 1902 to 1905; at Bassour, Algeria, in 1911 and 

 1912, and at Mount Wilson, Calif., from 1905 to 1920. Only one, or 

 at most two, determinations per day resulted from them, and they 

 were at the mercy of changes in atmospheric transparency during 

 the several hours per day required. If the sky clears as the sun 

 mounts higher, the solar-constant value is too high, and vice versa. 

 But in the average of many days of highest apparent excellence, 

 atmospheric changes were largely eliminated. We believe that our 

 mean value determined during this period, 1.94 calories per square 

 centimeter per minute, will never be greatly corrected. 



THE SHORT METHOD OF SOLAR-CONSTANT DETERMINATION 



In 1919 we devised a new and briefer method, less affected by 

 atmospheric changes though dependent on the long method for 

 indispensible data of average atmospheric transmission. If the sky 

 is hazy and therefore bright, the atmospheric transparency is low. 

 We found it iDOSsible to express empirically the observed atmospheric 

 transmission coefficients at our 40 wave lengths as functions of the 

 brightness of the sky near the sun. This brightness we measure 

 with an instrument called the pyranometer, which we invented. We 

 have so far perfected this short method of solar-constant measure- 

 ment that five independent values of the solar constant can be 

 observed and computed by two observers in five hours of work. We 

 still use the fundamental long method of Langley occasionally, but 

 only as a check to assure us that the empirical short method is still 

 sound. 



Soon after the Mount Wilson Observatory was founded by that 

 great scientific organizer. Dr. George E. Hale, he invited Doctor 

 Langley to conduct solar-constant work there. We carried on solar- 

 constant measurements at Mount Wilson each summer and autumn 

 from 1905 to 1920. By simultaneous measurements at Washington 

 and Mount Wilson and again at Mount Wilson and Mount Whitney, 

 we proved that the fundamental method of Langley yields equal 

 results whether carried on at sea level, at 1,700 meters, or 4,400 meters 

 of elevation. 



THE BALLOON PYRHELIOMETER 



About 1914 we constructed a form of automatic pyrhcliometer 

 which could be attached to a group of sounding balloons and used to 

 record the total intensity of the solar beam at very much higher 

 elevations. In July, 1914, my colleague, L. B. Aldrich, assisted by 

 by the United States Weather Bureau, sent up one of these balloon 

 pyrheliometers from Omaha. It was recovered uninjured in Iowa, 



