88 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1919. 



of temperature change to solar change at the time of greatest solar activity 

 was found from the averages of several years to be 1.4 C. for each change of 

 1 per cent in solar radiation. Since the extreme solar values range about 6 per 

 cent on either side of the mean, there might result departures from the normal 

 at Buenos Aires from this cause of about 8.5 C. The extreme departure from 

 the normal observed at Buenos Aires during the last 13 years has been 11.5 C. 

 The results of these researches have led me to believe that the existing abnor- 

 mal changes which we call weather have their origin chiefly, if not entirely, 

 in the variation of solar radiation. 



Naturally, these results, which are supported by an enormous 

 amount of careful and conscientious computation on the part of the 

 forecasting division of the Argentine meteorological service, are of 

 extreme interest. They point to the great desirability of equipping 

 in different cloudless regions of the world several observatories de- 

 signed for the measurement of the solar constant of radiation. The 

 chief of the Argentine weather service, Mr- Wiggin, desires very 

 much to take over the South American station of the Sinithsonian 

 Institution, to be maintained by the Argentine meteorological serv- 

 ice. Tentative arrangements were entered into between Dr. Abbot 

 and Mr. Wiggin for this purpose, which, however, require the further 

 approval of the Argentine Government to become effective. If suit- 

 able arrangements for the transfer can be made, it is hoped to employ 

 the funds thereby set loose for the establishment by the Smithsonian 

 Institution of a solar station in Egypt. 



From Argentina, Messrs. Moore and Abbot returned immediately 

 to Calama. 



Measurements of the Solar Constant of Radiation at Calama^ 

 Chile. — When Dr. Abbot reached Calama he found that Messrs. 

 Moore and Abbot had prepared data giving the pyrheliometry, the 

 transparency of the atmosphere for nearly 40 wave lengths, the func- 

 tion p/psc, and pyranometer values representing the intensity of the 

 radiation of the sky in a zone 15° wide surrounding the sun. All 

 these values were tabulated with solar constant values for 60 days 

 of observation and for each day at periods when the air masses were 

 '2 and 3, respectively. 



We have felt very keenly the desirability of devising some method 

 of determining the solar constant of radiation which would be inde- 

 pendent of changes in the transparency of the atmosphere during 

 the period of observation. It had been hoped that this might be 

 done in some simple way by the aid of the pyranometer, that instru- 

 ment which we devised several years ago for the purpose of measur- 

 ing the brightness of the sky. It is well known that when the sky 

 becomes more hazy the direct beam of the sun is reduced in intensity, 

 but the scattered light of the sky at the same time is increased. Ac- 

 cordingly, it would seem that a pyranometer measurement of the 

 brightness of a limited area of the skv near the sun would furnish 



