NO. 12 TEMPERATURES AND SUN S RADIATION ABBOT I3 



variation of the sun appears to be greater for the shorter wave lengths. 

 It may be even lo times as great at wave length 3200 angstroms as 

 for the total radiation of all wave lengths combined. Suppose, then, 

 that a silvered hollow ball of Corex glass should be carried by a sound- 

 ing balloon to 30,000 meters elevation. Every day would be fine there, 

 and a flux of ultraviolet rays for which silver is transparent would 

 flow through the ball, practically as unimpeded by the atmosphere as 

 if actually in free space. If, therefore, by a photoelectric cell, or a 

 photographic record, the intensity of these selected ultraviolet rays 

 could be measured to an accuracy of i percent, and quick intelligence 

 of these measurements could reach the home station on the ground, the 

 problem might be solved perhaps better than by multiplying solar- 

 constant observing stations. Yet this new plan has several uncertain 

 features. It would be well if both plans could be tried. 



Should an accurate daily series of determinations of solar variation 

 become available, it may well prove that several radically different tem- 

 perature effects will be discovered for each given station differing with 

 the magnitudes of the solar changes concerned. For it seems reasonable 

 to suppose that until a solar change reaches a certain magnitude local 

 obstacles would impede its effect on temperature. Indeed it must not 

 be expected that the solar variation is a complete guide to weather, 

 but rather a factor in a highly complex problem, the recognition of 

 which may yield valuable progress in weather forecasting. 



The reader will note in figures 2, 5, and 6 that the curves dift'er from 

 month to month and from station to station. As stated by Clayton and 

 others, almost immediate effects of solar changes are found at certain 

 places on the globe which he calls " centers of action." From these 

 the effects spread by atmospheric waves to distant localities. The 

 paths followed by these atmospheric waves doubtless differ with secu- 

 larly changing meteorological conditions governing atmospheric states 

 in different hemispheres. Hence, doubtless, arise the differences above 

 noted. Note for instance in the month of January that the broadest 

 separation of the pair of curves occurs first at Helena, later at St. 

 Louis, and last at Washington. 



It was suggested above that the temperature eft'ects discovered seem 

 at first sight disproportionally large compared to the solar changes 

 which induce them. However, temperatures in the temperate zones de- 

 pend largely on the prevailing direction of the wind. Clayton has 

 shown that solar changes are accompanied by wanderings of the at- 

 mospheric " centers of action." Thence come shiftings of the centers 

 of cyclonic motions of the atmosphere, which, in their turn govern the 



