PART I. ON MAGNETIC STORMS. CHAP. III. 237 



If we take 2 as an average value of /.i, we obtain 



In this way we should expect to find values of P v at the magnetic poles about double the value 

 of PI, observed near the equator. For greater values of , the proportion P, : P h will increase, and 

 vice versa. 



From about i6 h to i8 h we really find conditions that seem to favour our assumptions, when 

 we compare the values of P, at Axeleen with the value of P h at Dehra Dun and Batavia. Later on, 

 however, we find that P increases greatly, while PI,, at the equatorial stations, is slowly diminishing 

 and that before this period P, is much less and even sometimes directed to the opposite side. 



We cannot, of course, draw any further conclusions from this, as it is impossible to determine how 

 great a part of P, at Axeleen is due to polar precipitations. There is all the greater need of caution 

 in drawing conclusions, from the fact that the conditions at Christchurch which is in a comparatively 

 high southern latitude show that at that place there is only a very slight perturbation in the vertical 

 intensity, and from about I3 h 30 onwards, the corresponding P, is directed downwards, not upwards 

 as we should expect when only the equatorial perturbation is acting. We there find, moreover, com- 

 paratively poverful perturbing forces in the horizontal components, and it would thus appear that there 

 were precipitations of a more polar character in the southern hemisphere also. 



If, with the assumed value of /<, we make the force PI, at the equator equal to 75 y, we find that 



~ . ~ 3 TCI 



/r + F ' = 4^ = 75 ' I ' 



and / must then be equal to about 2 . io 6 amperes, a value of the same order as that which we shall 

 find in the calculation of the current-strength in the polar perturbations (see Chap. IV). 



The first intermediate storm, with maximum about 13'' 42 occurs during the same time and with 

 great violence, at Sitka and at Axeleen. Its local character at these places shows that the current-systems 

 are comparatively near to both stations. 



It is plain from the simultaneous appearance of the intermediate storms at Sitka and at Axeleen, 

 that these two storms must be closely connected with one another; but whether they are the effect of a 

 single system, or of separate and more limited systems of precipitation in the vicinity of the two stations, 

 it is impossible to decide with any certainty. 



We have seen in Art. 52 (cf. fig. 68) how well the assumption of separate fields of simultaneous 

 precipitation agrees with our theory; and circumstances are actually found here that seem to favour such 

 a view. The maximum occurs, indeed, at about the same time, namely at I3 h 42, but the storm begins 

 at Sitka about a quarter of an hour before that at Axeleen, and perhaps does not end until a quarter 

 of an hour after the latter has ceased. If we look at the declination at Baldwin, where the intermediate 

 storm is well defined, it appears that the storm there begins at 13'' 8 m , and concludes at 14!' 34. 



If we look at the //-curve for Kew or Wilhelmshaven, we notice that during this perturbation the 

 course of the curves is as follows: first at i3 h 12, there is a deflection answering to a diminution of H; 

 at I3 h 24, H has an intermediate minimum, then increases until I3 h 42, then decreases until I4 h 5 m , 

 when it again increases, and at I4 h 30 the effect of the impulse has ceased. The Z)-curve has a similar 

 course. It may perhaps therefore be natural to interpret the conditions in Europe in the following manner. 



Between I3 h I2 m and 14'' 30 there is a perturbation of uniform direction, occurring simultaneously 

 with the perturbation in America. P h and P d are directed respectively south and west, answering to a 

 current-arrow pointing north-west or west-north-west. This is interrupted by another perturbation, which 

 lasts from i3 h 24"" to I4 b 5, and acts in almost exactly the opposite direction; and at the moment 

 when this latter storm reaches its maximum at Kew, it causes the effect of the former perturbation to 



