REPORT OF THE SECRETARY 77 



at one station may be justifiably rejected, because of this sort of inter- 

 nal evidence, Avhen they differ widely from reasonaljle results of other 

 stations. The application of these criteria at the three stations will 

 lead to a new increase in the accuracy with which we may measure 

 the solar variability, and to a much closer agreement between the 

 results of the three stations. 



This long-desired improvement in accuracy is now shown to be 

 indispensable to a proposed application of studies of solar variability 

 to weather forecasting. 



Dr. Abbot published in May and August of 1936 two papers en- 

 titled, respectively, "The Dependence of Terrestrial Temperatures 

 on the Variations of the Sun's Radiation" and "Further Evidence on 

 the Dependence of Terrestrial Temperatures on the Variations of 

 Solar Radiation." These papers appear to prove that the short- 

 interval changes of solar radiation, such as run their courses in a few 

 days, are of major influence on the weather for the ensuing 2 weeks or 

 more. Figure 2 of the first-mentioned paper shows such effects for 

 the temperature of Washington. The curves shown each represent 

 the average results for some 10 to 20 cases occurring in each of the 

 months of the year of the years 192-^-1935. 



It is observed that opposite solar changes produce opposite tem- 

 perature effects ; that these effects differ from month to month ; that 

 other investigations reported in the papers cited show that they also 

 differ from locality to locality ; that the effects are of large magnitude, 

 sometimes reaching 15° F., sufficient indeed to account for most varia- 

 tions from the normal temperature; that these effects follow solar 

 changes seldom as great as 1.5 percent and averaging only about 0.7 

 percent. 



After discussion of these results with the Chief of the United States 

 Weather Bureau and with several gentlemen of the Weather Bureau 

 Advisory Committee, these gentlemen all agreed that the investiga- 

 tions offered reasonable promise of a method of forecasting some 

 features of the weather for at least 2 weeks in advance. They 

 unitedly signed a memorandum recommending increased support to 

 the Astrophysical Observatory to enable it to set up seven additional 

 solar stations in the best localities of the world, and also to investi- 

 gate the possibility of automatic determinations of solar variability 

 from sounding balloons. This measure was approved by the Presi- 

 dent and the Bureau of the Budget, passed the Senate as an amend- 

 ment to the Urgent Deficiency Act, but was rejected in conference 

 with the House. 



The Institution has since made a small grant from private funds 

 to promote preliminary studies for the balloon investigation referred 

 to, but the application of solar variation observations to weather fore- 



