22 



SMITHSONIAN MISCELLANEOUS COLLECTIONS 



VOL. 



65 



During the years 1913 and 1914 the solar constant was determined 

 at Mount Wilson by spectro-bolometric observations on some of the 

 days when these formulae are applicable to Arequipa observations. 

 From 34 comparisons of Arequipa and Mount Wilson solar con- 

 stant values, the average deviation of individual days is about 2.5 

 per cent. Omitting 5 days when unusually great discrepancies oc- 

 curred, owing to poor sky at one station or the other, the average 

 deviation is only 2 per cent. 



Under the circumstances it seemed unreasonable to hope that for 

 individual days the empirically derived solar constant results from 

 Arequipa observations would be of sufficient accuracy to show the 

 short-period fluctuations of the solar constant. It might reasonably 

 be expected, however, that monthly mean values would seldom differ 

 by more than 1 per cent from the values obtained in corresponding 

 months at Mount Wilson. Thus a new confirmation of the varia- 

 bility of the sun in its longer periods may be hoped for from 

 pyrheliometry and psychrometry at Arequipa alone. This hope 

 seems to be confirmed by the following Table 3. Both Arequipa 

 values (formulae I and II) are given, but the number of days 

 relates to the first method values, which are more numerous. 



Table 3 — -Monthly Mean Solar Constant Values 



Month 



Arequipa 



No. days 



Mount Wilson 

 No. days 



1913 

 July 



Aug. 



Sept. 



Oct. 



Nov. 



1914 

 June 



July 



Aug. 



Sept. 



Oct. 



I.87 I.89 



I.89 |i.89 



17 



1.925 

 3 



1. 931 

 18 



1.90 

 1.92 



18 



1.920 



25 



1.8b 



1.89 



11 



1.874 

 24 



1.876 



5 



t .96 



1. 91 



11 



1.952 

 14 



[-95 



1.956 

 14 



1.96 



1.94 

 18 



1.964 

 22 



1.94 



1.94 



13 



1-943 

 18 



The comparisons of July and November, 1913, have little weight 

 because of the small number of days observed at Mount Wilson. 

 Apart from these months only one, August, 191 3, shows a difference 

 of more than 1 per cent between Arequipa and Mount Wilson. Both 

 stations agree in showing the interesting result that the solar constant 

 was decidedly higher in 1914 than in 191 3. 



With the word of caution that individual day's values may often 

 be in error by as much as 5 per cent, and on the average by as much 

 as 2 per cent, we have included in Table 1 two columns giving the 

 daily solar constant values determined from Arequipa pyrheliometry 

 by means of formulae I and II. Table 2 gives the mean monthly 

 solar constant values by formulae I and II. Months for which no 

 values of vapor pressures are available are supplied by taking the 



