WEATHER GOVERNED BY THE SUIST — ^ABBOT 97 



monies at the same time. The harmonics are, indeed, the sources 

 of the distinguishing quality of the violin sound. The sun's radia- 

 tion seems to behave similarly. But it is not clear why the sun, 

 a great ball of compressed hot gas, should vibrate in a fundamental 

 and numerous harmonics, as a violin string or a bell does. Though 

 the cause is obscure, it is believed that the fact is demonstrated. 

 So confident of it was I, that in 1930, and again in 1932, though not 

 then in possession of analyses of the solar variation as satisfactory 

 as figure 2, 1 ventured to forecast publicly for 2 years in advance the 

 solar variation. Figure 3 shows these forecasts and the event. It 

 will be seen that maxima and minima fall about at the times ex- 

 pected, but that the fit is by no means so close as in figure 2. This 

 was due to the lack, at the time of those predictions, of the more 

 recent discovery of the terms of 34, 39^^, 92', and 276 months. 



Everyone is aware that the weather is in the main controlled by 

 the earth's relation to the sun. Our alternate exposure to the sun 

 and to dark space by the daily rotation of the earth produces day 

 and night, with their warming and cooling. Revolution about the 

 sun in a plane 23i/^° out of the plane of the Equator causes the 

 sun to appear far south in January and far north in July, with 

 attendant cooling and warming of the Northern Hemisphere. The 

 ellipticity of the earth's orbit removes us to 3,000,000 miles farther 

 from the sun in July than in January, and causes the sun's heat 

 upon the earth to be 6 percent more intense in January than in 

 July. This tends to make winters in the Northern Hemisphere more 

 mild, and in the Southern Hemisphere more severe than otherwise 

 they would be. All these effects are well known. We may now 

 inquire whether the recently discovered variation of the sun's emis- 

 sion is also of suflficient importance to affect the weather. 



We have studied this question in many ways. In the Smith- 

 sonian publication called " World Weather Records " are given 

 monthly mean temperatures and precipitations for several hundred 

 stations in many countries. Some of these stations, as, for instance, 

 Helsingfors, Berlin, Copenhagen, Greenwich, Capetown, and Ade- 

 laide, present observations covering nearly, or quite, a century The 

 records of all the cities just named were studied. To avoid per- 

 plexing details, the original monthly mean observations were 

 smoothed by 5-month traveling means. For instance, for March use 



Jan. + Feb. + Mar. + Apr. + May 

 6 



and for April, 



Feb. + Mar. + Apr. + May + June 



etc. 



