22 



SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL, ']'] 



and compared them with the solar-constant values secured by Smith- 

 sonian observers at Calama, Chile. We were soon perfectly agreed 

 that we perceived the following relation : When a sun spot, or group 

 of sun spots, crosses the central diameter of the solar disk, in course 

 of the solar rotation, the next following day almost invariably shows 

 a minimum value of the solar constant. We perceived this to hold 

 in so very large a proportion of cases that all doubt of it was 

 dispelled. 



Jan Mar May July Sep Nov Jan Mat- May July Sep Nov Jan Mar May July Sep Nov 

 1921 1922] ,| 



Fig. 14. — Comparison of solar variation with variation of visible phenomena 

 on the sun from results of the observatory at Ebro. 



A conspicuous case occurred in March, 1920, as shown in figure 15. 

 More recently, Mr. Clayton has made a quantitative examination of 

 this relation extended over several years of observation. His result 

 entirely confirms ours. Still more recently, my colleague, Mr. Fowle, 

 has taken the quantitative data of the Observatory of Ebro in Spain, 

 where they give sun-spot areas within 15° of the sun's center. He 

 finds a plain correlation of the same sort. Some examples of it are 

 shown in figure 16. Large spotted areas near the sun's center are 

 nearly always associated with lower solar constants. 



