PART I. ON MAGNETIC STORMS. CHAP. III. 255 



The conditions in North America at this point of time are peculiar. At the three stations in the 

 east of that continent, the direction of the current-arrow is east, and at Sitka south-west, or on the 

 whole rather different from that which might be expected from its resemblance to the above-mentioned 

 polar elementary storms. This lasts, however, only for about 10 minutes during the first part of the 

 perturbation, whereupon P h decreases, and for a moment is about zero ; and in the two succeeding charts 

 the directions of the arrows are the same as, for instance, on the i5th December, 1902, and the 22nd 

 February, 1903. 



The resemblance to these storms is still further increased by the circumstance that in Europe there 

 is a corresponding positive turning of the perturbing force. 



The third intermediate storm, from about 23** io m to o h 25. 



As regards the arctic regions, this polar storm is powerful at Axeleen, rather less so at Matotchkin 

 Schar, and at Kaafjord, strange to say, it is almost imperceptible in H and D, while in the vertical intensity 

 it is quite distinct. 



At the same time there is a distinct difference in the perturbation-conditions in southern latitudes, 

 these being particularly powerful and distinct in Europe, and noticeable also in the East and in the 

 United States, while at Sitka the perturbation is almost imperceptible. The oscillations are on the whole 

 uniform in direction, indicating that the forces remain in one direction all the time. We have therefore 

 considered it sufficient to show the distribution of force at one moment during the time when the 

 perturbation is at its height. This is represented on 



Chart XV; time 23* 45. 



This storm, on the whole, has a great resemblance to the previously-described elementary night- 

 storms, e. g. to that of the 23rd March, 1903. They commence at about the same time of day, i. e. 

 a little before midnight. In both of them, the distribution of force remains constant throughout the 

 perturbation, and is in the main similar. 



The perturbing forces of southern latitudes, as the chart shows, seem to indicate that we have a 

 point of convergence situated, in this case, very near Kaafjord, the effect of this system being there 

 almost exclusively in a vertical direction. The horizontal arrow drawn for Kaafjord would appear, to 

 judge from the curve, to be due mainly to the effect of the principal storm, which is still in activity. 

 At one place in the north of Canada, perhaps near Hudson's Bay, there is a point of divergence of the 

 horizontal component of the perturbing force. 



Notwithstanding the long duration of the perturbation, and its somewhat varied character, we believe 

 that we have succeeded, by means of the foregoing analysis, in elucidating the main features of the 

 perturbation-conditions, and taking out the elementary phenomena that together form the present storm 

 in all its diversity. In the course of the period of time considered, the following principal phenomena 

 have been shown : 



A positive equatorial perturbation from about n h i8 h , and the six following polar storms: 



(1) The polar storm from I2 h 25 to 13'' 15, 



(2) The polar storm from about i7 h 20 to i8 h 30, 



(3) The main polar storm from about i8 h 30 to o h 30, 



(4) The first intermediate polar storm, maximum at i8 h 34, 



(5) The second intermediate polar storm from 2o h 45"" to 21 h 2o m , 



(6) The third intermediate polar storm from 23 h io m to o h 25. 



