i'iv dii:l Mai/nctic " 



As pointed out by Dr. Chree,* there is a distinct relation between the " magnetic 

 character" at (.'ape Evans and the brilliance of tin- aiirurn- observed at th" same hour 

 Irom thai station. The following figures for unrnnc of Classes I to IV, arranged in 

 descending order of brightness, are given below : 



Cape Evans, 1911. Class I. Mean magnetic character number 1-28. 



,, U . ,, ,, ,, ,, 1 Uu. 



III. 0-93. 



IV. 0-66. 

 Mean character number lor May-August, 0-84. 



Cape Evans, 1912. Class I. Mean magnetic character number 0-93. 



II. ,, 0-(>3. 



,,III. 0-71. 



IV. 0-25. 



Mean character number for May-August, 0-72. 



Though the variations in mean character number are not intrinsically large, we 

 must recollect that these figures refer to only a portion of the whole time, periods of 

 bad weather and daylight being excluded. In these circumstances, it is not surprising 

 that the mean for the whole of the winter monthsf is higher than the mean for very 

 faint aurora). It is, in fact, fairly clear that the correlation between aurora) and 

 magnetic character number is not close, except possibly for aurora) of exceptional 

 brilliance. 



We had a definite impression while in the Antarctic that the correspondence 

 between aurora 1 and magnetic disturbances was closest in the case of coloured aurora) 

 and those which showed much movement. That this impression is justified is shown 

 when we take from Dr. Chree's figures the character numbers corresponding to the 

 few cases when mention was made of much movement, or of colour, in the aurora 

 log. 



The figures for these two classes at Cape Evans in 1911 are : 



Coloured aurora; . . . . Mean character number =1-82. 

 Aurora; with much movement ,, ,, =1-44. 



As mentioned before, it is interesting to note that the greater number of both 

 these classes were recorded in the afternoon, a result which equally applied to the 

 aurora of Class I. 



* " British Antarctic Expedition, M'lo r.iKJ." 'Terrestrial Ma^ui-iism,' Chap. XIV. The m.iLrn. n, 

 character number of a (lav or an hour is a iiimilii'i (between O and 2) wliirli ijives a mrasinv <if the 

 .l.-'jiee nl magnetic (list urlianee diirlnj; tliat period. Maximum dist urbanee i-s marked _>. anil minimum 

 disturbance, 0. 



I The mean for tin- same |>erii>d, lull evludin'j I!H' ilayliulit hours, \\nuld lie lower. 



33 c 



