402 BIRKELAND. THE NORWEGIAN AURORA POLARIS EXPEDITION, igO2 1903. 



At Ssagastyr there is a short, comparatively small, but well-defined negative storm at about 17''; 

 and at Uglaamie there is also a negative storm. A maximum is found here a little before 19''. 



At Cape Thordsen we find negative deflections that very much resemble the positive deflections 

 at Bossekop and Jan Mayen, for at all three stations we find deflections of fairly constant strength for a 

 period of some length. The similarity is immediately seen on looking at the curves. 



Lastly we also find considerable perturbing forces at Kingua Fjord; but, as already remarked, 

 they will be more easily studied by looking at the charts. The most powerful storms, however, are 

 between 2o h and 22 h . At almost all the polar stations here mentioned, there are negative deflections, 

 as a rule very well defined. Only at Sodankyla do we find a considerable positive deflection, this being 

 at about 2o h 3o m . 



There is a certain amount of time-displacement here. The great negative deflections, for instance, 

 begin a little earlier at Ssagastyr than at the European polar stations; but as we unfortunately have no 

 observations from Little Karmakul, this circumstance cannot be closely studied. Moreover there are 

 other phenomena which encroach upon it: for in all probability there will be positive storms occurring 

 simultaneously in districts from which we have no observations. Now and then too, we find positive 

 deflections, which may be interpreted as effects of such a system, e. g. the one just mentioned at 

 Sodankyla, a small positive deflection in Jan Mayen at about 22 h 30"", and two or three distinct positive 

 deflections at Fort Rae, in the interval between 22 h and the close of the period of observation. 



On looking at the declination-curve for Kingua Fjord, we are at once aware of a peculiar circum- 

 stance. This is the jagged, disturbed character of the curve before 2o h , and the wide, but regular 

 deflection after that hour. We have seen that as a rule the curves in the polar regions during equatorial 

 storms are of an exceedingly jagged, disturbed character, whereas the curve during well-defined polar 

 stoms may frequently exhibit a fairly quiet course, even if the deflections are large. It may well be, 

 therefore, that this transition to a more quiet course is an indication that the equatorial system is dis- 

 appearing. 



At the southern stations we find the most powerful forces in declination, and the deflection here 

 begins at the time that the more powerful negative forces appear in the northern regions. 



We now pass on to consider the last eight charts, on which the perturbation-conditions for the 

 last part of the period are shown. 



On Chart V, for 77'' 20"' we see evident traces of the positive polar storm, its district of pre- 

 cipitation extending from Godthaab across Jan Mayen to Bossekop. At all the other polar stations there 

 are distinct effects of negative storms; while at the southern stations the equatorial system is still 

 evidently at work. 



On Chart VI, for 2o h 20, the effects of the equatorial storm have disappeared, and those of the 

 positive polar storm are found only at Bossekop and Sodankyla. Everywhere else in the polar regions, 

 we find more or less pronounced effects of negative precipitation, these being especially marked in the 

 district Uglaamie to Ssagastyr. At Fort Conger there is also a more or less westerly-directed current- 

 arrow, which in strength considerably surpasses those at Kingua Fjord, Godthaab and Cape Thordsen, 

 This should probably be regarded as a continuation of the system of which traces were found in Jan Mayen. 



The divergence of the current-arrows for Christiania, Gottingen and Pawlowsk, especially the 

 westerly direction that they have at the first two stations, seems clearly to indicate the existence of a 

 system of positive precipitation in the regions westward from Bossekop along the auroral zone; for 

 these two arrows appear to be enclosed in an area of divergence corresponding to such a system. The 

 positive vertical arrow for Gottingen is also in accordance with this. 



The arrow for Pawlowsk, on the other hand, seems to be in the eastern area of convergence, 

 answering to the negative storm in the north of Asia, but may also be considered as belonging to the 



