COMPARISON OF ARCTIC AND ANTARCTIC DISTURBANCES. 



247 



After having read Prof. BIRKELAND'S volume, I carno, somewhat reluctantly I must confess, to the 

 conclusion that those responsible for the production of the present volume would be open to criticism if 

 the opportunity afforded for contrasting Arctic and Antarctic records were not utilised. This "appendix 

 has accordingly been written with the object of supplying information as to what was happening in the 

 Antarctic during the disturbed times selected by Prof. BIRKELAND. Its primary object is to inform, not 

 to criticise, and if it contains anything that savours of criticism, this is mainly in order to explain why 

 Prof. BIRKELAND'S exact procedure has not been followed. 



S3. The method adopted by BIRKELAND in dealing with magnetic disturbances is fundamentally 

 different from that adopted in Chapters IX and X. Practically following SABINE, he defines disturbance 

 at any given hour as the difference from the normal for that hour ; and what his tables give, and his charts 

 illustrate, are the values of disturbances so defined at different stages of each disturbed period. He 

 believes that the cause of the disturbance is an electrical discharge, or a series of electrical discharges, 

 in the Earth's atmosphere, such as he has succeeded in producing in the vicinity of his terrella, and his 

 ultimate object apparently is to calculate the position and intensity of these discharges. 



One great difficulty, as I remarked in 90, Chapter IX, is the fixing of a normal value. In doing this 

 BIRKELAND seems to have been materially helped by the fact that at some observatories, including those 

 provided with Kew-pattern magnetographs, it is usual to have two days' curves on the same photographic 

 sheet. If the one day's curve happens to be undisturbed, its form greatly assists the eye in deciding 

 as to the nature of the disturbance in the other. This is an advantage which I have often had occasion 

 to appreciate myself. The Arctic stations, however, and some of the others had magnetographs of the 

 Eschenhagen pattern, like those of the " Discovery," and there must have been considerable difficulty 

 at times, as BIRKELAND himself allows, in deciding what the departure from the normal really was. This 

 difficulty had probably a considerable indirect influence on BIRKELAND'S choice of disturbance periods. 

 Those he has selected are largely represented at Kew and other non-polar stations by "bays" of 

 comparatively short duration. 



What is meant by a " bay " will be readily grasped by reference to the accompanying figure. The 



continuous line ABCDEFGH represents an imaginary magnetic curve having two bays, one, BCD, 

 occurring at a time when the regular diurnal variation is slow, the other, EFG, occurring at a time when 

 it is rapid. The ordinates CM, FN drawn on to the base line represent the excess above the constant 

 base-line value of the values of the magnetic element answering to the times M and N. The broken lines 

 BC'D, EF'G are intended to represent the imaginary undisturbed curve, and the intercepts CC', FF' on 

 the ordinates represent from BIRKELAND'S point of view the disturbances. 



In such a case as that represented by the figure the method appears simple, especially when the 

 disturbance occurs at a time when the diurnal change is slow. In practice, however, there is usually 

 a difficulty in deciding where the " bay " begins or ends, and the relative position of its two extremities 

 is usually not quite what one would anticipate from the trend of the curve prior to its commencement. 

 This latter difficulty naturally increases the longer the duration of the bay, and the more rapid the regular 

 diurnal inequality changes at the hour. 



As a general rule, during really active magnetic disturbances, whilst bays of a kind are not infrequent, 

 the curve adjacent to them is itself disturbed and sinuous, and affords very imperfect guidance as to where 

 the normal curve would come. On days of real disturbance, one is usually obliged to have recourse to the 



