COMPARISON OF ARCTIC AND ANTARCTIC DISTURBANCES. 



271 



perturbation ... is in reality . . . composed of two principal phenomena, an equatorial perturbation 

 and a short, well-defined, comparatively powerful elementary polar storm." Of the " equatorial " 

 perturbation he writes: "This . . . begins quite suddenly, at 12h. 58m., with an oscillation that is 

 noticed simultaneously all over the world. In the equatorial regions, this sharp deflection . . . appears 

 principally in H. About the auroral zone the curve oscillates, and the perturbation is noticeable both in 

 D and H." The time at which this commencing disturbance reached a maximum is given as 13h. 4m. 

 BIRKELAND concludes, apparently, that the small enhancement of H caused by the commencing 

 distiirbanee, or at least a fraction of it, persists in low latitudes right through the subsequent polar storm 

 until midnight. At Dyrafjord and Axeloen, however, he notes "a peculiar circumstance," p. 128, viz., 

 t hat the commencing movement is distinctly oscillatory. The H curve, in fact, at the two stations shows 

 a fall, a rise and then a second fall, and possibly a second, but much smaller, rise. 



According to BIRKELAND'S table, p. 130, the time of commencement of the "polar" storm varied from 

 20h. 30m. to 21h. 30m., the time at non-polar European stations being about 21h. 10m. At most stations 

 the termination is put at about 23h. 45m. At most non-polar European stations the maximum occurred 

 about 22h. 10m. ; in North America it was about half an hour later. The maximum value of the horizontal 

 disturbing force is given as varying from about 370y at Axeloen to 12 '4y at Batavia, the value at 

 Christchurch being too small to measure satisfactorily. The value at Kew, 44y, is about a mean for 

 non-polar Europe. 



My examination of the original Kew curves makes the sudden commencement 4 or 5 minutes earlier 

 than the hour, 12h. 58m., given by BlRKELAND. But the movement appears distinctly oscillatory in both 

 H and D, a small fall, lasting 2 or 3 minutes, preceding the rise. In the case of H the fall seems about ly, 

 the rise about 13y ; while in D the fall is I'-O, the rise 2' -4. There are a number of small movements at 

 Kew in both H and D throughout the remainder of the 22nd, but much the most conspicuous phenomenon 

 is a bay on the D curve from about 21h. 6m. to 23h. 50m., the element being depressed in value. The 

 largest value of the depression is about 8' '8 and it occurred about 22h. 5m. 



Looking at the curves in Plate XX, it is easy to recognise in most, if not all, a second movement which 

 somewhat resembles the commencing movement both in size and character, and which occurred about an 

 hour and fifty minutes later. At Kew this second movement occupied some 1 4 minutes. During it H 

 fell 4y and rose lOy, while D fell 0' - 8 and rose l'-0, the turning-point in each case coming at about 

 14h. 46m. The reason for mentioning this is because on looking at the Antarctic curves one's eye is 

 caught by two oscillatory D movements larger than their neighbours, occurring during an otherwise 

 somewhat unusually quiet time. So far as I can judge, these movements correspond in time to the 

 commencing movement and the second movement just referred to. The following are the results of the 

 measurement made of the two groups of movement ( + denotes a rise, - a fall) : 



In each case it is a little doubtful whether the final rise in D forms part of the same system as the 

 preceding rise and fall. The H movements do not stand out very conspicuously, and it is mainly their 

 approximate coincidence in time with the others that has led to their enumeration. The V movements, 

 though small, do stand out. 



Unless we assume an accidental coincidence of a truly remarkable kind, the conclusion seems inevitable 

 that the above two groups of movements are due to the same cause as the movements seen at Kew and 

 elsewhere at the same times. If this be so, then the fact that the Antarctic movements represent 



