356 



NA rURE 



[May 19, 1910 



practically over, but the force shows a marked depression 

 compared to its value at the same time on the previous 

 day. ■ This is a very common after-effect of magnetic 

 storms; the greater part of the depression usually dis- 

 appears in two or three days. Fig. i is a good example 

 of an ordinary disturbance in which the magnetic changes, 

 though considerable, were seldom rapid. It differs con- 

 spicuously in this respect from the recent great storm of 

 September 25, 1909. Many of the movements on this 

 occasion were too rapid to be shown clearly in the photo- 

 graphic traces. 



Dr. Schmidt, the leading German authority on our 

 subject, assigns to this recent storm the first place of all 

 recorded since the Potsdam Observatory came into exist- 

 ence some twenty years ago. Table IX. gives his estimate, 

 on an arbitrary scale, of the intensity of the seven largest 

 storms recorded at Potsdam. 



Table IX. 



Hr. Ad. Schmidt's Estimate of Intensity of Magnetic 



Storms. 



Date of Storm 



September 25, 1909 

 October 31, 1903 

 February 14, 1892 

 July 20, 1894 



Disturbance 

 at Potsdam 



3SC0 



28fo 



over i8co 

 1580 



Dale of Storm 



Disturbance 

 at Potsdam 



Stptember 11. 190 

 August 20, 1894 

 February 9, 1907 



1520 

 J410 

 '34° 



An old question which has received a good deal of recent 

 attention is whether there is a cyclic period approaching 



7- 



which the declination range conspicuously overtops the 

 average is considerable. During these days there is usually 

 a distinct fall, in the horizontal force,, a circumstance also 

 indicative of magnetic disturbance. The following days 

 were considerably disturbed : — .August 29, 30, September 21, 

 25, 30, and October 18, .19, 23, 24; while a variety of 

 other days, e.g. August 31 and October 2, 8, and 9, were 

 decidedly more disturbed than the average. If we 

 associate August 30 and September 25 we get a twenty- 

 six-day period ; if we associate August 29 and Sep 

 tember 25, or September 21 and October 18, we get 

 twenty -seven-day period; if we associate August 31 an 

 September 30 we get a thirty-day period ; and we haw 

 any number of other possible combinations left. Disturbet 

 conditions are seldom limited to a few hours of a particular 

 da)', and often extend over two or more daj-s. Thus 

 there is usually a good deal that is arbitrary in the value 

 deduced by observation for the interval between two 

 specified storms. 



The disturbances of .September 21, 25, and 30 led to a 

 fall in the horizontal force, from which it is doubtful 

 whether the element had entirely recovered even by the 

 middle of November. 



Mr. . Maunder and Dr. Schmidt both associate thei 

 periods with that of the revolution of the sun relative to 

 a point, on the earth. This period exceeds the true period 

 of the sun's. rotation — which varies considerably with solai 

 latitude — because the earth is . travelling round the sun i 

 the direction in which the sun rotates. 



The view most in favour at the present time is that 

 magnetic storms are due to some solar discharge, probably 

 '^lom sun-spot areas, and of an electrical nature. W( 



Kjaat fhm/J.^^'yu^ti^ /(rrc£^ hnx^ryU^A. u-^i, IffXf., ^jOCy. = o-ooi Cf.^. 



Fig. I. 



a month in the occurrence of magnetic storms. J. A. 

 Broun, an early pioneer of magnetic work, believed his 

 observations to indicate a period of about twenty-six days. 

 From an elaborate study of many years' storms at Green- 

 wich, Mr. E. W. Maunder deduced a period of 27-275 

 days, and Mr. Arthur Harvey independently, from a study 

 of storms at Toronto, deduced the remarkably similar 

 period of 27-246 days. The latest result of this kind is 

 due to the eminent German magnetician already men- 

 tioned, Dr. Schmidt, who believes in a period of 29-97 

 days. Schmidt found evidence of this period in a number 

 of recent storms, and he declares that it exists in the case 

 of very large storms even when separated by many years. 

 He found that the dates of occurrence of five out of the 

 seven largest storms recorded at Potsdam (see Table IX.) 

 could be deduced to a high degree of accuracy from the 

 expression 24io,ooo-H303i-o + mX29-97, which counts time 

 in days from the commencement of the Julian era. 



Fig. 2, which serves as a chronicle of magnetic history 

 at Kew from August 20 to November 16, 1909, will illus- 

 trate some of the difficulties in the way when one attempts 

 either to prove or disprove the existence of a period in- 

 magnetic storms. 



The upper curve shows the value each day of the 

 absolute declination range at Kew, the lower the value 

 at each midnight of the horizontal force. We see incessant 

 variations from day to day, and the number of days in 



NO. 2 II 6, VOL. 83] 



may suppose a solar discharge to traverse space like a je 

 of water ; when it overtakes the earth a magnetic storm 

 begins, which continues until the full width of the jet 

 has passed over. If the solar discharge continues long 

 enough, it may sweep over the earth during ' several 

 successive revolutions of the sun, and so give rise to 

 series of magnetic storms at nearly equal intervals. 



Theories accepting a solar origin for magnetic storms 

 differ as to the nature of the solar discharge. 



Nordmann has suggested Rontgen rays, Birkeland 

 kathode rays, and Arrhenius negatively charged particles. 

 On Nordmann 's hypothesis the terrestrial phenomena 

 should follow the solar in a few minutes, on Birkeland 's 

 hypothesis in a few hours, while according to Arrhenius 

 the interval might be two days or more. 



The most elaborate investigation hitherto made into the 

 supposed solar origin of magnetic storms is due to Prof. 

 Kr. Birkeland, of Christiania, who belie%-es kathode or 

 analogous rays to be the vehicle by which the solar disturb- 

 ance is propagated to the earth. He has made numerous 

 experiments with kathode rays in a vacuum tube which 

 contains a miniature earth or " terrella." By means of 

 electric currents in wires wound on the terrella, a magnetic 

 field is produced similar in type to the earth's field. It 

 was apparently his experiments that suggested his explana- 

 tion of a certain type of magnetic storm which he termsi 

 the "equatorial." These "equatorial" disturbances are. 



