﻿fowle] 



SOLAR RADIATION MEASURES 



403 



This process as applied to three different days may be seen in 

 figure 62. The deviations of the observation from the straight line 

 used in the extrapolation are given in Table 1 and serve as an indi- 

 cation both of the accuracy of the observations and the suitability 

 of the straight line for representing the points. 



T \t.le I 



I .ogarithmic deviations. 



Jan. 27, 1904. 

 +.002 

 — .001 



+.003 



Mar. 26, 190 

 0.000 

 0.000 

 —0.003 

 -j-O.OOI 

 — 0.002 

 +0.0007 

 — 0.002 

 — 0.00 1 



— .00/ 

 +.010 

 +.007 

 — .002 

 — .001 



(Jan. 27; 3 discordant 

 observations due to 

 passing clouds or 

 smoke are omitted.) 



average logarithmic deviation .002 .004 



average deviations 0.5 percent 0.9 percent 



+.014 

 —.006 

 +.007 

 —.003 

 — .011 

 +.001 

 — .007 

 —.013 

 +.001 

 +.007 

 +.001 

 — .012 

 +.001 

 .006 

 1.4 percent 



These extrapolated determinations may now be compared with the 

 values of the solar radiation computed by the spectrobolographic 

 method. The results are summarized in Table 11. 



Table II 



Log. 



Oct. 9, 



Feb. 19, 

 Mar. 25, 

 Mar. 26, 

 Apr. 17, 

 Apr. 29, 

 July 7, 

 Oct. 14, 

 Dec. 7, 

 Dec. 23, 

 Jan. 27, 

 Feb. 11, 

 May 28, 

 Oct. 5, 

 Nov. 16, 



1902 

 1903 



1903 



i9°3 

 1903 

 1903 

 1903 

 1903 

 1903 

 1903 

 1904 

 1904 

 1904 

 1904 

 1904 



Grade. 



.3 > I 



if) ^ 



Grade. 



S -z <J o 3 

 - 'J jj a ~ 



7-2.9 

 I-I.9 



5-2-7 

 0-3-8 



1.99 

 1.97 



2.02 1 

 1. 71 



Very good 



Very good 



Excellent 



Excellent 



Very good 



Had 



Short, good 



Very good 



Excellent 



Doubtful 



Excellent 



Very good 



Very good 



Very good 



Very good 



2.19 



2.27 



2.23 



2.09 



2.18 



I.96 



2.14 



I.96 



I.94 



I.99 



2.02 



2.26 



2.09 



2.32 1 



I.98 



Fair 

 Very good 

 Excellent 



Excellent 



Passable 



Very good 



Poor 



Very good 



Passable 



Passable 



Fair 



Fair 



Poor 



Excellent 



Passable 



15% 



15 



14 



15 



15 



12 



3 



5 



15 

 14 

 14 

 14 



6 



15 

 16 



1 Some doubt exists whether these values are not 

 a possible change in the constant of the actinometer 



both too high, owing 10 

 due to an accident. 



