192 BIRKELAND. THE NORWEGIAN AURORA POLARIS EXPEDITION, 1902 1903. 



In the perturbations that follow, we shall moreover have an opportunity of studying the fields that 

 form at this time of day; and we shall see that conditions similar to those that we have here pointed 

 out will be continually repeated. 



At the three hours shown on Chart V i6 h io m , T7 h , and I7 h 30 we also find on the whole 

 the same conditions as at I4 h 45, the only difference, besides a diminution in the strength of the 

 forces, being a change in the direction of the arrows in the eastern hemisphere, as if the precipitation on 

 the day-side were moving westwards with the sun. The change, however, also may be due only to the 

 diminution in the strength of this system upon the night-side. 



We have previously mentioned that the curve for the field now under discussion gives the im- 

 pression of several relatively independent systems succeeding one another. In this case therefore, it 

 would perhaps be natural to consider the one system as vanishing, and new systems being formed, in 

 such a manner that they advance towards the west. The curves for Dyrafjord seem perhaps to make 

 such an assumption of new systems doubtful, as the conditions there remain fairly constant. The move- 

 ment may also be explained by the assumption that the night-system moves westwards, and little by 

 little destroys the effect of the eastern part of the day-system. 



The conditions at Sitka and Honolulu indicate, though only faintly, an area of convergence answer- 

 ing to a precipitation on the night-side. At Baldwin, Cheltenham and Toronto, there is a very small 

 force. It appears, from investigations of the material from 1882 83, that systems on the night-side have 

 a west-ward motion. The reason why the forces in eastern America are so small in the present instance, 

 may therefore possibly be that the storm-centre has now moved too far away. This, moreover, is in 

 accordance with the fact that its effect in Europe becomes more noticeable. 



On Chart IV, for i8 h o m , the same conditions continue at Axeleen. Matotchkin Schar also seems 

 now to be mainly influenced by this precipitation. 



At Dyrafjord the force is now particulary strong, and the current-arrow is still directed towards 

 the east. It seems to be this precipitation on the day-side, which now lies farther west that especially 

 gives to the field in lower latitudes its character, as there is here an area of divergence. At Kaafjord 

 the force is smaller, but seems mainly to be determined by the precipitation at Dyrafjord. 



The third section, from i8 h to 23**. 



We have already, in the preceding section, had an opportunity of observing that the powerful 

 storm breaks in upon one of long duration. This we found to be the case both at the Norwegian sta- 

 tions and, on the whole, at stations in the eastern hemisphere. This is a well-known circumstance, and 

 we will only refer to the perturbation of the I5th February. With the same reason as on that day, 

 we can, by drawing a normal line that forms a harmonious connection between the conditions before 

 and after, obtain a more exact determination of the perturbation, in so far as it is dependent upon the 

 powerful polar storm. It will be in the main for the horizontal component as the perturbations in D at 

 most places seem to be chiefly connected with the polar system. 



(a) The conditions at the Norwegian stations. 



The violent storm is powerful at all the four Norwegian stations simultaneously, most powerful at 

 Axeleen and Matotchkin Schar. It is very varied in its details, but the oscillations retain in the main 

 an uniformity of direction. 



At Dyrafjord the powerful storm commences at i8 h 33, and is over at 22 h 17. After this time, 

 perturbations still appear for a time; but they are principally in accordance with the conditions before- 

 hand. The perturbation is at its height between I9 h 8 m and 2o h 14. At about 2o h 37 m the oscillations 



