Brilliant aurora? appear, therefore, to occur most often near the time of the 

 secondary auroral maximum, at which time there is at least a suggestion of a secondary 

 maximum for the mean magnetic character number during the winter months. As may 

 be seen from Dr. Chree's figures for the winter months,* this secondary maximum for 

 magnetic character number is at 19-20 hours, the mean value at 4 hours being low and 

 the chief maximum at about 9 hours. We therefore infer that the occurrence of auroras 

 close to, or over, the station in the morning does not (on the whole) correspond with 

 more disturbed magnetic conditions at the same station, but with less disturbed 

 conditions. 



The secondary maximum in magnetic character number occurs only in the winter 

 months and seems, according to Dr. Chree, to be associated with short disturbances 

 of a " special type."f One is tempted, therefore, to infer that the secondary auroral 

 maximum is also associated with these special magnetic disturbances and with the 

 occurrence of brilliant, coloured or quickly moving forms, appearing at a time when 

 auroras are rarely seen. Such an inference is hardly justified on the scanty evidence 

 to hand, though a correspondence between secondary maximum and type of 

 disturbance seems to be indicated, both as regards magnetic and auroral characteristics, 

 while the relation between magnetic character number and auroral intensity, on the one 

 hand, and between the " special type " of magnetic disturbance and the strongly 

 coloured and quickly moving forms of aurora, on the other hand, is very slightly 

 indicated. The whole relation between magnetic and auroral disturbances, as will be 

 seen later, may be stated as follows : 



(i) A fairly strong correspondence between periods of disturbed auroral and 

 magnetic conditions, particularly in the stronger disturbances. 



(ii) A slight correspondence between disturbed conditions of both types in the 

 same hour. 



(iii) A possible secondary maximum in the mean character number for disturbance, 

 of both types, occurring about the same time in the afternoon, in both 

 cases. (The secondary maximum in mean magnetic character number 

 seems to be connected with magnetic disturbances of the " special type," 

 while the secondary auroral maximum seems to be connected with the 

 occurrence of brilliant, coloured and quickly moving aurora?.) 



Even if we admit that there is a real correspondence between magnetic and 

 auroral disturbances, no deduction can be drawn as to whether the magnetic disturbances 

 in the evening are directly caused by the particles to which the aurora is due, or 

 whether the auroral display at this hour is due to a change in the curvature or 



* Loc. oil., Table LXXX ; also p. 143. 

 | IMC. cit., i>]>. -27U and 143. 



34 



