On Riipid Periodic Variations of Terrestrial Magnetism. 



21 



a very cliaracteristic train is repeated two or three times with 

 nearly equal intervals of 15™ 40™ etc., as if the same train of 

 waves were recrn'ring to the point of observation with a definite 

 period (Fig. 5, B). 



/) It is frequently observed tliat while a fairly regular train 

 of waves is traced in both the X- and Y-components at the same 

 time, the periods of waves are quite different for the two com- 

 ponents (for examples, see Fig. 7). Among (:)S conspicous cases 

 chosen, 42 were those in which the periods were longer in X- 

 than in Y-component. The distribution of such cases in different 

 hours of the day may be seen in the following table: 



Table I. 



A tendency is suspected, though not very apparent, that the 

 cases where the periods of X-waves are greater than those of Y, 

 are most frequent near midnight and noon, while they are com- 

 paratively rare in the morning and evening hours. 



g) Though the most conspicuous regular trains are chiefly 

 observed witli periods ranging from 30^ to 300' still longer waves 

 are not at all rare, if we take those trains together in account, 

 with only a small number of maxima and minima. The longest 

 wave traced was of a period nearly amounting to one hour, which 

 may of course be detected in the ordinary magnetographic record 

 of low sensibility, when the amplitude is sufficiently large. The 

 intermediate periods are rather evenly represented in our list of 



