MAGNETIC DISTURBANCES AT KEW. 299 



differences between the seasonal results for example, the difference between the 

 winter and other seasonal results under a t in H are to be regarded with some 

 suspicion. But there are several reasons for not expecting a very close agreement. 

 If we compare the phase angles in Table XIII., which are derived from all the 

 disturbed days, with those derived from the smaller number of more highly disturbed 

 days, or with those derived from sunspot maximum years, we find a difference which 

 is generally in the direction suggesting that the departure from the value of the 

 phase angle characteristic of quiet days is greater the more disturbed the conditions. 

 For instance, c^ in V is at all seasons much larger for disturbed than for quiet days, 

 and in Table XIII. the value of a x derived from the whole body of disturbed days is 

 invariably the least. Thus, even if there were no true seasonal variation in the 

 difference between the phase angles characteristic of quiet days and of days possessing 

 a standard amount of disturbance, we should expect some seasonal variation to appear 

 in Table XV. through seasonal fluctuation in the amount of disturbance. Finally, 

 quiet day phase angles do show more or less seasonal variation, and there is no 

 obvious reason why such seasonal variation should be identical for days of standard 

 disturbance. 



Comment was passed in (C) on the great increase in a in D through disturbance, 

 representing on the average an advance of over two hours in time. The corresponding 

 change in V is, however, an advance of over four hours. The corresponding change 

 in I is retrograde and roughly equal to that in D. The winter results for j in H had 

 better be regarded as unreliable. At the other seasons j in H shows an advance of 

 time, though considerably less than in D. The influence of disturbance on a 2 , 3 , and 

 4 , when regarded from the point of view of the equivalent in time, is in general 

 relatively small, and is decidedly less in I, H, and V than in U. The differences 

 between the disturbed and quiet day values of a 3 and a 4 in the three former elements 

 are in fact generally less than one would have anticipated even on the hypothesis that 

 the phase angles are really unaffected by disturbance. - 



Force Vector of Diurnal Inequality. 



23. The departure of the magnetic elements from their mean value for the day 

 may be regarded as due to a disturbing force of constantly varying intensity and 

 direction. Table XVI. gives particulars of the direction and intensity of this force at 

 each hour of the day for the year as a whole. Analogous results could, of course, be 

 derived from the diurnal inequalities for individual months or seasons of the year. 



<j) denotes the angle which the vertical plane containing the resultant force makes 

 with the vertical plane through the geographical meridian, measured from north to 

 east. 6 denotes the inclination of the resultant force or vector to the vertical, 

 measured from the nadir. The force on the north pole of a magnet is thus directed 

 below or above the horizontal plane according as 6 is less or greater than 90. 

 II denotes the intensity of the resultant force or vector, p its component in the 



2 Q 2 





