128 BIRKEI.AND. THE NORWEGIAN AURORA POLARIS EXPEDITION, IQO2 1903. 



equatorial perturbations. At Kaafjord also, the direction of the arrow is in accordance with the rest of 

 Europe, but the force is somewhat greater. At Dyrafjord and Axeleen we find a peculiar circumstance, 

 namely, that in the course of a few minutes the force oscillates very violently. A number of arrows 

 are placed upon the chart, answering to various hours, the scale being the same for the southern stations 

 as for the northern. At Dyrafjord, the current-arrow makes a negative turn of about 180 from SW 

 to NE. At i3 h 4'", the direction of the force is uniform with that of the arrows in the south of Europe. 

 At the same time, the arrows on Axeleen turn from S in a positive direction, until at 13'' 7 they 

 point NE as at the other European stations. 



We will not here attempt to give an explanation of this peculiar circumstance, but will only say 

 that this turning in different directions at two places so near to one another, must necessarily lead to 

 the conclusion that in the north at any rate, the perturbation is to some extent of a local character. 



We thus see that while there is a current that acts powerfully and almost symmetrically on both 

 sides of the equator, there will be exactly simultaneous perturbations of a local character in the north. 

 These currents in the north, which are very slight, cannot, on account of the extent of the perturbation, 

 be the cause of the perturbation as a whole; for, as we see, the force diminishes from the poles south- 

 wards as far as Tiflis and San Fernando, whereupon it increases, and even at Christchurch is great. 

 In the vertical intensity this first oscillation is noticed, in southern latitudes, only at Tiflis, where it 

 indicates a force directed upwards. The reason why it is not felt at Zi-ka-wei can only be that the 

 sensibility there is so small; but on the other hand, it seems stranger that nothing is noticed at Pawlowsk 

 and Pola, where the sensibility is fairly great. 



After the first deflection, the equatorial perturbation continues with a small deflection in H, answer- 

 ing to a perturbing force directed northwards along the magnetic meridian. Judging from the character- 

 istically serrated appearance of the //-curve in low latitudes, the perturbation seems to last until the 

 polar storm is over, or from about ia u 57"" until midnight. 



The distribution of force, as it is on the whole maintained on account of this equatorial perturba- 

 tion, is shown on 



Charts II and III, for //* o m and /p* /o m . 



The current-arrows in somewhat more southern latitudes lie, as we see, almost parallel with the 

 magnetic parallels, and the force there is comparatively great. We notice that the force at the Central 

 European stations varies greatly in magnitude. We must not, however, immediately draw conclusions 

 from this circumstance; for it may be accounted for partly by the difficulty there is in determining the 

 normal line for so long an interval, and partly by the fact that, owing to the rapid changes in the 

 deflections, a mistake in the time will easily occasion a mistake in the determination of the perturbing 

 force, of which the percentage becomes all the larger, when the perturbing force is small. 



At the arctic stations the force is comparatively great, and we see that the current-arrow bends 

 northwards, and indicates a circle round the magnetic pole, showing that it is not the axis of the earth, 

 but the magnetic axis, that determines the phenomenon. At Sitka too, the current-arrows are some- 

 what abnormal, as we also found them to be in previous equatorial perturbations. This must be due to 

 the polar precipitation that is always present during these storms. If we look at fig. 37, we see that 

 the light parts in the terrella's auroral zone, come more or less in the region answering to the north 

 of N. America. It is possible that this drawing-in of rays may also to some extent be the cause of the 

 abnormal smallness of the perturbing force at Baldwin on Chart I. We shall find this confirmed in the 

 conditions during the equatorial storm of the isth December, 1882, described in Chapter III. 



In the vertical intensity, the perturbation is almost imperceptible, being only slight at Tiflis, where 

 it is directed upwards at the moment of observation. At Pawlowsk it is not noticeable, and at Dyra- 



