274 RIRKELAND. THE NORWEGIAN AURORA POLARIS EXPEDITION, 19021903. 



the positive equatorial perturbations in these districts, and we then suggested, that it would probably be 

 due to polar precipitation in the north of North America (cf. pp. 67 & 128). Here, however, the abnormal 

 condition is far more marked than in these two earlier storms. 



Upon the conclusion of this equatorial perturbation, we enter upon 



the second section, from i6 h to about 22 h . 



The perturbing forces appearing here are generally small; but from about i7 h 30"" to about 18'' 20 

 they are comparatively large, especially in southern latitudes. 



The conditions at I7 h 40 are shown on Chart II. If we look at the curves for the Norwegian 

 stations during this period, we find, as regards the horizontal intensity, that there is a perturbing force 

 at Axeleen directed southwards, and at Dyrafjord and Kaafjord there are perturbing forces directed 

 northwards all the time. The declination-curve oscillates at all the stations above and below the mean 

 line. We have unfortunately no registerings of H from Matotchkin Schar for this perturbation; but from 

 the other three stations there is sufficient material to enable us to conclude that the field during this 

 period is the typical one for a post-meridian storm. There are distinct effects of a positive polar storm 

 at Dyrafjord and Kaafjord, and at Axeleen the effect of a negative storm, which, after what has been 

 said, we are inclined to suppose extends eastwards on the night-side of the globe. This comes out 

 clearly on Chart II. In Europe and Asia there is a distinct area of convergence; and in America and 

 the districts east of it, there seems undoubtedly to be an area of divergence. These conditions agree 

 well with the results we have already arrived at, regarding the appearance and formation of the systems 

 at various times of day. As the forces, however, for the later part of this period are small, we have 

 contented ourselves with this one chart as representative of the period. 



The third section from about 22 h on the 23rd November, to 7 h on the 24th. 



At about 22 h , the conditions begin to alter considerably. The Norwegian stations have now entered 

 the night-side of the earth, and accordingly the deflections in H for Kaafjord and Dyrafjord swing round 

 so that we now get the westward-directed current-arrows that are characteristic of the night-storms. The 

 change in direction does not take place, however, until about 2i h 30 at Dyrafjord, and an hour later 

 at about 22 h 30 at Kaafjord. This may seem to be at variance with what we have previously 

 found to be the case, as for instance in the perturbations of the 3ist October and ist November, 1902, 

 when we found that the cause of the change appeared to move westwards with the sun. Here, however, 

 we find the opposite, as the change takes place earlier at the more westerly-situated Dyrafjord than at 

 Kaafjord. 



There are, however, several things to notice in this connection that may aid in a comprehension 

 of these conditions. 



In the first place, on the 315! October, we were considering the stations Matotchkin Schar and 

 Kaafjord, both of which are situated to the south of the auroral zone; whereas here we have one sta- 

 tion Dyrafjord to the north, and one Kaafjord to the south of the zone. It is by no means 

 improbable that this circumstance is of some importance. It would be natural, indeed, to imagine that 

 owing to the more northerly situation of Dyrafjord in relation to the magnetic axis, it would be easier 

 for the system acting at Axeleen to have an influence here than at Kaafjord, which in this respect has 

 a more southerly situation; and that on this account the positive storm of the preceding section would 

 be able to act longer at Kaafjord than at Dyrafjord. 



In the next place it should be observed that the times considered in the two cases differ very 

 considerably from one another, a fact which is undoubtedly very important; for if we assume that the 

 position of the sun in relation to the magnetic axis of the earth is of great importance in deciding the 

 position of the systems of precipitation, we must also assume that the relative motion of the earth and 

 the sun will govern the displacement of the systems from time to time. 



