100 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1916. 



that whose secant is 1.2) were found to be almost wholly governed 

 by three things — the atmospheric humidity, the distance of the sun, 

 and the variations of the sun's emission. Hence from measurements 

 of the humidity by the psychrometer it was possible to compute from 

 the observed radiation the probable intensity of the solar radiation 

 outside the atmosphere for each day. These empirical solar-constant 

 values from Arequipa observations confirm the variations of the sun 

 observed at Mount Wilson by the complete spectrobolometric process. 

 Indeed, it appears that if eight or ten well-separated stations at high 

 altitudes should be equipped wdth the pyrheliometer and psychrome- 

 ter their combined results might well be expected to determine closely 

 enough the sun's variations. A most interesting feature of Arequipa 

 observations is that there is nothing anomalous about the observations 

 of 1912 to suggest that the volcanic eruption of Mount Katmai (of 

 June 6, 1912), which produced a great deal of dust all over the 

 northern hemisphere, produced any turbidity of the atmosphere 

 whatever south of the Equator. 



Results of Mount Wilson solar-constant observations have been 

 furnished in advance of publication to Dr. Bauer of the Carnegie 

 Institution for comparison with magnetic data. He finds a close 

 correlation betw^een certain fluctuations of the earth's magnetic field 

 and the variations of solar radiation. 



The tower-telescope observations of the distribution of radiation 

 along the diameter of the sun's disk, made at Mount Wilson in 1913 

 and 1914, having been fully reduced, a preliminary publication of 

 them has been made by the Smithsonian Institution.^ These results 

 show distinctly that the average distribution of solar radiation over 

 the solar disk varies from year to year. Greater contrast of bright- 

 ness between the center and limb of the sun prevailed in 1907 and 

 1914 than in 1913. The change is greater for short wave lengths 

 than for longer ones. Changes also occur from day to day. Both 

 of these kinds of changes are found correlated with changes of the 

 solar constant of radiation, but in opposite senses. High values of 

 the solar radiation attend periods of greater solar activity and are 

 associated with increased contrast of brightness between the center 

 and edge of the solar disk. For short-period fluctuations of solar 

 radiation, however, low values of solar radiation are associated with 

 increased contrast. It seems reasonable to suppose that the first kind 

 of phenomena is caused by increased convection in the sun, bringing 

 fresh radiating surfaces forward more rapidly, thus increasing the 

 effective solar temperature. The second kind of phenomena may be 

 caused by temporary increases of the turbidity of the outer solar 

 envelopes, restricting the solar emission especially at the limb. 



1 On the distribution of radiation over the f?un's disk and new evidence of the solar 

 variabUity, Smithsonian Misc. Coll., Vol. 66, No. 5, May, 1916. 



