﻿400 SMITHSONIAN MISCELLANEOUS COLLECTIONS [VOL. 47 



In a paper published by Air. Langley in the Astrophysical Jour- 

 nal, 1 a possible change in the value of the solar radiation was sur- 

 mised. This possibility is indicated by several kinds of circum- 

 stantial evidence : 



1. By the spectro-bolographic 2 determinations of the solar radia- 

 tion by the study of homogeneous rays, thus avoiding the erroneous 

 assumption of uniform atmospheric transmission coefficients for all 

 wavelengths. 



2. 3 In several cases by apparent changes in the value of the trans- 

 missibility of the solar atmosphere, indicating some change in solar 

 condition coincident with the observed changes in the radiation 

 values determined as just mentioned. 



And 3. By observed changes in terrestrial temperature as noted 

 over the greater portion of the northern hemisphere of our earth. 



If such changes in the solar radiation actually do take place, then 

 their detection and measurement by some easy and rapid method is 

 of great importance, since the solar radiation is one of the most im- 

 portant factors in determining meteorological phenomena. The 

 holographic method just mentioned involves not only the use of 

 costly and extremely delicate apparatus requiring two skilled ob- 

 servers, but the subsequent reductions are so laborious that a single 

 reduction requires at least a week's time. The second method, 

 though leading to results with much less time-consuming reductions, 

 yet involves still more costly and quite as delicate apparatus. Hence 

 if some trustworthy method could be determined for computing the 

 so-called solar constant from ordinary actinometric measures, the 

 desired values of the solar radiation could then be obtained not only 

 quickly but cheaply. A secondary actinometer costing probably 

 only $10 or $15 would be good enough for such measures, and 

 indeed as good as any, since all actinometers at present in use must 

 be considered as secondarv instruments. 



1 Astrophysical Journal, xix, p. 305, 1904. 



2 A full description of the spectro-bolometric determinations may be found 

 in the Smithsonian Miscellaneous Collections, Quarterly Issue, Volume 

 I, page 74, 1903. Briefly, solar energy curves are taken at frequent interva.s 

 during an afternoon. These serve for the determination of Bouguer's " a "s 

 for each wavelength. After an energy curve has been corrected for all in- 

 strumental absorptions, it is then, by means of these atmospheric transmis- 

 sion coefficients, corrected to its value outside the atmosphere. The ratio of 

 this final area to that before this last correction, multiplied by the simultaneous 

 actinometer reading, gives the so-called spectro-bolographic determination of 

 the extra-atmospheric solar radiation. This is then reduced to the mean solar 

 distance. 



3 Nature, vol. 70. p. 19S. 1904. 



