5o6 



NATURE 



[August 28, 1919 



were of an unusual character. In declination swings 

 of id' to west, 20' to east, and again 62' to west 

 followed in immediate succession. In horizontal force 

 there was, as usual at the start, a rapid rise, 

 amounting to 75 y ; but in less than a minute the 

 movement was reversed, and a fall exceeding 4507 in 

 less than twelve minutes took the trace off the sheet. 

 Horizontal force remained depressed fof nearly 

 i^- hours, but then for a few minutes it was above 

 the normal value. Another large fall then ensued, 

 which carried the trace off the sheet from gh. 15m. 

 to gh. 25m. Between I4h. and i7h. of August 11 

 horizontal force was usually above the normal. The 

 maximum, which appeared synchronously with that 

 in vertical force at about i6h. 2m., exceeded the value 

 prior to the sudden commencement by 4607. The 

 disturbance in vertical force, though exceptionally 

 large, was of the usual type. During the afternoon 

 of August II, from I4h. to i8h., the curve was of a 

 pyramidal shape, the value of the element being much 

 enhanced. By 23h. the curve had resumed its normal 

 level, and a depression then set in, the minimum 

 being reached just after 2h. on August 12. 



The ranges recorded during this storm have seldom 

 been approached at Kew Observatory. In fact, it is 

 unhkely that so large a range has ever been recorded 

 there before in vertical force. But for the great 

 reduction in sensitiveness made of late years to meet 

 the conditions caused by electric trains and trams, the 

 maximum would have been far beyond the limits of 

 registration. C. Chree. 



Kew Observatory, Richmond, Surrey, 

 August 13. 



The magnetic storm which began on the morning 

 of August II was one of the largest recorded in recent 

 years, and was probably of world-wide distribution. 

 It . attracted public attention chiefly through the 

 notable — though not unusual — extent to which it 

 interfered with telegraphic work. On account of its 

 somewhat exceptional features, the following state- 

 ment of results of observations at Eskdalemuir 

 Observatory may be of some interest, and is com- 

 municated by permission of the Director, Meteoro- 

 logical Office. 



The times given below are Greenwich mean times. 

 The unit of i 7 is o-ooooi C.G.S. It should also be 

 mentioned that the principal magnetographs at Esk- 

 dalemuir are so arranged as to give directly the 

 vertical (V), north (N), and west (W) components of 

 terrestrial force ; a declination magnetograph is also 

 in operation. 



The conditions prior to the advent of the storm 

 were those of a magnetically quiet day. Very slight 

 disturbance was recorded between 2oh. and 2ih. on 

 August 10, and pulsations of about three minutes' 

 period were observed on N about an hour after mid- 

 night. The beginning of the storm as observed at 

 Eskdalemuir may be taken as having occurred at 

 6h. 58m. on August 11. But this beginning differed 

 very considerably in its character from the usual type 

 of what is known as a "sudden commencement." 

 Ordinarily, this phenomenon exhibits a rise in the 

 value of the horizontal force; a rise also, though 

 usually smaller, in declination ; and in some instances 

 a fall in the value of the downward directed vertical 

 force. So far as is known, these abrupt changes 

 take place simultaneously at any one place, and (in 

 spite of attempts to prove the contrary) there is no 

 trustworthy evidence to show that they are not syn- 

 chronous at all observatories. In the case of the 

 storm now considered, however, a minor disturbance 



NO. 2600, VOL. 103] 



of somewhat unusual type began on the north com- 

 ponent thirty-two minutes before the other com-. 

 ponents experienced the sudden commencement of the 

 storm. There is nothing to show that this minor 

 disturbance had any relation to subsequent events 

 or was other than "accidental," but it is men- 

 tioned for what it may be worth, and as being the 

 cause of a doubtful estimate as to the time of the 

 sudden commencement on the north component. At 

 all events, the disturbance began at 6h. 58m. so 

 suddenly as to send the light spot completely off the 

 reoording_ sheet, and did so with such rapidity that it 

 is impossible to state whether the change was one of 

 increase or decrease in force. But while there is no 

 photographic trace immediately after 6h. 58m. on the 

 + _ side of the undisturbed value, there is distinct 

 evidence of its being helo-w that value within a minute 

 after that time. In another respect, the beginning 

 of the storm was altogether exceptional in that the 

 sudden commencement on the vertical-force magneto- 

 graph showed but the faintest trace of any decrease 

 in value, and in reality was followed by a large 

 increase. On the west component record there is 

 shown a sudden rise and fall, the difference between 

 the extremes being 172 7. 



After the rapid changes associated with the sudden 

 commencement of the storm, the first minimum 

 value of N occurred at some time between 7h. and 

 7h. 30m., the trace being off the sheet in that interval. 

 The first maximum value of W after the sudden 

 commencement was at yh. lom., when it reached 

 1887 above the undisturbed value. The declination 

 at this time was 1° 18' to westward of its amount 

 before the storm began. The vertical force rose to a 

 maximum at 7h. 13m., it being then 447 above its 

 undisturbed value. Then followed a fall, on which 

 were superposed numerous pulsations, to a minimum 

 at 7h. 39m., and a recovery to a maximum 667 above 

 the undisturbed value at yh. 58m. Such changes in 

 V during the early part of the storm are entirely 

 unusual, both in character and amount. 



During a magnetic storm the value of the vertical 

 force usually rises to a maximum about lyh., the 

 rise occupying about four hours, and being gradual 

 though irregular. In the present instance, after the 

 first few hours of the storm had passed, during which 

 time the oscillations in V were unusually rapid, the 

 value rose suddenly at i4h. 28m., the trace leaving 

 the sheet at i4h. 55m., having risen 2507 in this 

 interval of twentv-seven minutes. 



Other unusual features of the storm may be referred 

 to, and one of these is the early hour at which the 

 fall, after the maximum, in V took place. Usually 

 this occurs about midnight, and includes two sudden 

 drops in value. In the present case both occurred at 

 unusually early hours, the first beginning at i9h.22m., 

 the second at 23h. 9m. The gradual! recovery of the 

 vertical force to its normal value is occasionally 

 accompanied by pulsations. These were prominent 

 on the) morning of August 12. For example, during 

 half an hour after 5h. thirteen oscillations were 

 recorded with a mean amplitude of 47. Another note- 

 worthv feature of the storm was the intense agitation 

 (" internal activity ") in the horizontal components, 

 especially after 2h. on August 12. As a rule, this is 

 more prominent during the daylight hours of a 

 storm; here it occurred during the^ night hours. 

 Lastly, the disturbance was peculiar in the sudden- 

 ness with which it ended about igh. on August 12, 

 and in llie magnetically quiet conditions which suc- 

 ceeded it. 



A. Crichton Mitchell. 

 Eskdalemuir Observatory, August 14. 



