NO. 2 



SOLAR CHANGES ABBOT 



II 



thereby became opaque, except for a narrow range of wave lengths 

 centering about 0.316 micron, he was able to select a very narrow 

 region of ultra-violet spectrum, and compare its intensity outside 

 the atmosphere with the intensity of the solar radiation in the green. 

 The ratios of violet to green, in Pettit's monthly mean values, show 

 a range of 60 per cent at a mean wave length of 0.316 micron. 



It would naturally be expected that these large ultra-violet varia- 

 tions observed by Pettit would accompan}' exactly in point of time 

 the variations of total radiation as determined by Smithsonian ob- 

 servations. Pettit has kindly communicated some of his results 

 to me, and in figure 7 the two sets of observations are brought 



Fig. 7. — Comparison of Smithsonian monthly mean solar constant values 

 with ultra-violet solar radiation values of Pettit. 



together, with the scale of the ordinates of the Smithsonian work 

 expanded to match that of Pettit. The agreement between the two 

 series seems very satisfactor}', in view of the fact that the range of 

 total solar radiation is only about 1.5 per cent, so that one can not 

 hope that the accuracy of the Smithsonian determinations is sufificient 

 to give perfect correspondence on this very wide scale. Further- 

 more, Pettit has observed only on four days per month in the earlier 

 part of his investigation ; while in the latter part he has included 

 every possible day, and among them some of doubtful uniformity 

 of sky. In view of these circumstances it is not to be supposed that 

 his monthly results are without considerable error. 



Also, interestingly associated with solar variation are results re- 

 cently communicated to me by Dr. Austin on the variation of inten- 



