PART I. ON MAGNETIC STORMS. CHAP. III. l6g 



THE PERTURBATION OF THE 28th DECEMBER, 1902. 



(PI. XIII.) 



56. This perturbation is not one of those that it was originally intended to describe, and the time 

 is therefore not given in my circular dated June 1903. There are thus only a few more or less chance 

 observations besides those from the Norwegian stations. What has determined us nevertheless to describe 

 it is the peculiarity we find on comparing the curves for Dyrafjord with those for the American stations. 

 The perturbation occurrs chiefly between 4 h 40 and 6 1 ', that is to say about midnight, local time, at 

 the three easterly North American stations. 



The well defined deflection in the curves for Dyrafjord indicates that the storm could be a polar 

 elementary one, of which the district of precipitation perhaps is in the vicinity of that station. The time 

 of the perturbation, however, differs from that generally found in the best examples of polar elementary 

 storms at the Norwegian stations. The conditions at Kaafjord and Matotchkin Schar also show with 

 sufficient distinctness that there is no field of precipitation at those stations, the perturbing forces there 

 being quite inconsiderable. At Axeleen, on the other hand, there are more powerful perturbing forces, 

 and the perturbation there is of somewhat longer duration than at Dyrafjord, as it begins earlier and 

 concludes at about the same time. The character of the curve too, is so different that it is difficult to 

 decide whether the perturbing forces at these two stations arise from two separate systems or not; but 

 this question is of no great actuality in our study of this storm. The main thing is to prove the con- 

 nection between the perturbations at Dyrafjord and the American stations. The form of the curves has 

 a very great resemblance to those found in Europe during the polar elementary storms occurring at 

 about midnight on, for instance, the I5th December. We should therefore imagine that in this instance, 

 the field on the midnight side was similar to that previously found at the Norwegian stations; and a 

 closer investigation seems to verify that so is the case. 



On Chart I, for 4 h 45 and j h , there appears to be an area of convergence in the east of North 

 America, and adjoining part of the Atlantic, and in the west of Europe. This should indicate that in 

 the neighbourhood of Dyrafjord, possibly a little to the west of it, there should be a stormcentre with 

 current-arrows directed westwards. It is impossible to determine the size and position of the field of 

 precipitation more precisely with the comparatively few data that we have to go upon ; but the conditions 

 at Sitka indicate that it must extend comparatively far westwards in North America. Judging from the 

 curves for Sitka, we may suppose that the same system is at work there as at Toronto and Cheltenham. 

 The similarity between the curves at these places is great enough to allow such an assumption. The 

 centre of gravity, so to speak, of the field of precipitation may be assumed to be about the south of 

 Greenland. Sitka should be situated almost on the main axis. 



The rest of the course of the perturbation may now be very simply explained by a westward 

 movement of this storm-centre. On surveying the curves closer, we see that at Toronto PI, turns from posi- 

 tive to negative a little earlier than P d from E to W. In consequence of this the arrows will turn with 

 the hands of a clock, the current-arrow from S by W to N. Their size at this time, about 5 h 20, is 

 very small. In Cheltenham P k and P d change the sign at nearly exactly the same time, so that here 

 one does not get a rotation, but more a sudden change of direction from S to N of the current-arrow. 

 Thus in Toronto the conditions are such, as if the point of convergence passes just a trifle south of the 

 place, while the conditions in Cheltenham indicate that the point just passes the same. At any rate we 

 may conclude from this that the point of convergence will pass near these stations. But to determine its 

 course more exactly is difficult, as precision in the fixing of time here plays an important part. 



At Sitka the directions of the arrows are at first rather constant, but then turn with a counter- 

 clockwise movement, showing that as the system moves westwards, the place comes into the area of 



Birkeland. The Norwegian Aurora Polaris Expedition, 19021903. 



