414 JBigelow — Meteorological Elements of the United States. 



be taken up at a later time after the facts are known. In 1889* 

 I concluded that the variation of the solar radiation is respon- 

 sible for the annual variations in the terrestrial magnetic field, 

 because the magnetic disturbing vectors follow the sun in its 

 changes in declination. In 1894f evidence of the effect of the 

 solar variation upon the magnetic field of Europe, and upon the 

 temperatures of the United States, together with the action of 

 the same upon the position of the storm tracks and the velocity 

 of the movement of cyclones in the United States, was added. 

 In 1904^: the comparison was extended to the temperatures 

 and pressures of the entire earth's atmosphere, with the gen- 

 eral result that the temperatures respond directly to the solar 

 variation in the tropics, but inversely in the temperate zones; 

 while the pressures correspond directly in the eastern hemi- 

 sphere, but inversely in the western hemisphere. It thus 

 became apparent that the circulation of the atmosphere has to do 

 with the prevailing annual temperatures quite as much as the 

 direct radiation of the sun. To separate these two terms is a 

 task of great difficulty, and it is the purpose of this paper to 

 record one step in that direction. 



The variations in the annual values of the temperature, 

 vapor tension and barometric pressure are in fact comparatively 

 small quantities, and the question whether they are to be 

 ascribed to the influence of the solar variations can only be 

 determined by using data of several kinds reduced to strictly 

 homogeneous series. There are several causes for the fact 

 that the meteorological observations of the United States, as 

 originally made, are not sufficiently free from accidental errors 

 to permit of this refined use, and advantage was taken of the 

 necessity of providing for the Weather Bureau some new nor- 

 mals of temperature, vapor pressure and barometric pressure, 

 to make the series from 1873 to the present time as homo- 

 geneous as possible. In the report of the Chief of the Weather 

 Bureau, 1900-1901, the results for the barometric pressure 

 were published, the succeeding annual volumes containing the 

 supplementary data. In 1908, the temperatures and vapor 

 pressures reduced to homogeneous systems are to be published, 

 and this completes the re-reductions for the United States for 

 one-third of a century. Evidently these data are very valuable 

 for the study of the solar-terrestrial problems. There are 

 about 100 stations available in the United States for these 



*The Causes of the Variation of the Magnetic Needle, this Journal, Sept. 

 1891. 



f Inversions of temperatures in the 26 '68 day solar magnetic period, ibid., 

 Dec. 1894. 



| Synchronism of the variation of the solar prominences with the terrestrial 

 barometric pressures and the temperatures, Monthly Weather Eeview, Nov. 

 1903. 



