78 CLIMATIC CYCLES AND TREE GROWTH 



secured the records for 1906 to 1934 in daily values. For convenience these 

 were compressed into 3-day averages. 1 The data could then be compressed 

 again if needed and longer cycles investigated. In the first examination by 

 cy olograph several striking effects appeared at once: the intervals between 

 magnetic maxima easily show several definite periods between 26 and 32 days ; 

 a period close to 27.0 days is much more persistent than any other; this is 

 occasionally reduced to 26 days and sometimes is mixed with longer periods of 

 about 30 days. One notes that a synodic period of 30 days in solar rota- 

 tion occurs near latitude 40° which is at the outer limit of sunspot area. It 

 is difficult to explain how streams of charged particles, leaving the sun in 

 lines nearly perpendicular to its surface, can reach the earth from that lati- 

 tude ; yet one finds possibilities in the curves of coronal streamers often pho- 

 tographed during the total solar eclipse. Periods near 27 days are double- 

 crested ; that is, a lesser maximum appears to bisect them as would happen 

 if two magnetic source areas were at opposite longitudes of the sun. The 

 second maximum is often fainter than the other. Near sunspot maximum 

 additional periods make the variations more complex. 



The results of cyclogram analysis confirm strongly the persistence or 

 localizing of magnetic source areas in solar longitudes. For example, there 

 seems to have been very little change in longitude of a 27.0-day maximum 

 from the beginning of 1930 to our latest data in the middle of 1935. A 27.0- 

 day period is the one commonly shown by sunspots at 5° to 10° from the equa- 

 tor. Hence we presume that near sunspot minima the source areas persist- 

 ently active in agitating terrestrial magnetism are sometimes localized for 

 long intervals at that latitude on opposite longitudes of the sun. 



Sources of the 6-Months Period — Since the 6-months maxima occur in spring 

 and autumn, the question has been raised whether they originate in the sun 

 or in the earth and the relation of the magnetic maxima to each has been 

 studied. For example, the sun's pole is inclined 7° to the ecliptic in such a 

 direction that in early March the sun's southern hemisphere is most inclined 

 toward us, and in early September, its northern hemisphere. But late March 

 and September are the times of the terrestrial equinoxes and the maxima could 

 be related to them through the changing of the earth's hemispheres regarding 

 exposure to the sun's radiation. An extensive statistical study of the 6- 

 months maxima by Bartels (1932) has demonstrated that the earth is one 

 source of this variation, that is, the magnetic forces from the sun are suffi- 

 ciently terrestrialized to show an earthly character. 



Cyclogram analysis applied in 1935 presented immediately a complex 

 feature. In the vicinity of the sunspot minima and especially in the last one, 

 1932-1933, the 6-months maxima alternate in opposite longitudes of the sun. 

 This is shown in our diagrams, Plate 19B and C and figure 33. The figure 

 summates the magnetic effects that come while opposite halves of the sun 

 (longitudinally) are presented towards us. If this effect of alternating 



1 The preparation of these curves was done by Mrs. G. Dewey and Mr. Arthur N. 

 Cowperthwait. 



