Feb. 19, [885 



NA TURE 



37: 



yet they have QOt been finally worked out, gave the following 

 results : — 



1 1 1 When two galvanometers, as nearly equal as possible, were 

 introduced into the two circuits lying east-west near Sodankyla, 

 and which, as we have said, were moved towards each other so 

 that the old circuit was 2 '5 km. further east, the variations of 

 the two galvanometers were nearly identical. This appears 

 clearly from the graphic account of the deviations as they were 

 observed on Oct. 16 from 5I1. 25m. to 5I1. 55m. p.m. (see Fig. 2). 

 In tlie abscissa each centimetre represents two minutes, and in 

 the ordinate each millimetre represents a deviation of 20 divisions 

 of the scale, equal to an arc of 20'. The plates of the circuits in 

 question having been -unk to a depth of i - 3 m., it is clear that 

 the variations observed arose from changes in the electromotive 

 force of the terrestrial current, and could not have their origin 

 in the changes in the electromotive force arising front the con- 

 tact of the plates with the earth, for had the latter been the 

 case, the variations could scarcely have shown such extraordinary 

 coincidence. Other similar experiments show, however, that 

 small inaccuracies may occur. 



The two curves do not correspond exactly in the intensity 

 of the variations, which arises from the fact that the resistance 

 of the old circuit was greater than that of the new. 



While the variations which were greater and more numerous 

 in an east-west direction occurred nearly always at Sodankyla, so 

 that the needles of the galvanometer at that place were rarely 

 at rest, the contrary was the case at Kultala, that is to say, the 

 on which the needles of the galvanometer were in 

 motion were very rare. 



were already observed by me in 1871 and 1882, 



Fig. 2. — Curve I. shows the deviations of the galvanometer in the old con- 

 ductor. Curve II. shows the deviations of the galvanometer in the new- 

 conductor. Each centimetre of the abscissa indicates 2 minutes. The 

 observations were made every half minute. Each millimetre of the 

 abscissa indicates 20" of the scale = 20' of an arc. 



it seems fair to assume that the North Pole of the earth is sur- 

 rounded by a belt in which the terrestrial currents are strongei 

 and more variable than they are north and south of this belt. 



The northern limit of this belt seems to be about 68° N. The 

 position of this belt of terrestrial currents depends probably 

 upon the belt of the aurora borealis. 



(3) The magnetic variations and the changes which govern 

 the terrestrial currents follow each other closely. We know that 

 the former depend very much upon the aurora borealis, and that 

 this dependence also influences the latter. However, the laws 

 of this dependence cannot be determined until the materials of 

 the observations have been fully analysed. 



The Eltctric Currents of the Atmosphere studied by means of 

 the Discharging-Apparalus. — Since Franklin and Dalibard 

 proved— about the middle of the last century — by practical 

 experiments that lightning is an electric phenomenon, many 

 attempts have been made to measure the electricity which is 

 always present in the atmosphere. These experiments have 

 become more general since the discovery by Lemonnier, a 

 Frenchman, that electricity was present in the air even without 

 thunderstorms. Numerous methods have been invented and 

 employed for examining this electricity, while all had for their 

 object the measurement of the electricity present in a given spot 

 at a given moment. In this manner the atmospheric electricity 

 was carefully observed and registered, and by means of these 

 records it was hoped to arrive at some definite conclusions. 

 Sometimes researches were made to determine the variations of 

 tensions in different directions, particularly the vertical direction. 

 As a general result, but not without exception, these experiments 



showed that the electrical tension (or potential) increased with the 

 distance from the earth's surface. 



The knowledge of the electric charge, or the quantity of elec- 

 tricity present in a given atmospheric space, does not yet convey an 

 exact idea of the electric phenomena which take place therein, but 

 the knowdedge of the variations accompanying it in different direc- 

 tions enable us to ascertain the movements of the electricity, or, 

 in other words, the electric currents of the atmosphere. When we 

 know by experience that the generality of effects, and the most 

 important, which produce electricity arise from electric currents, 

 it will easily be understood that the examination of the atmo- 

 spheric electricity should have for its principal object the visible 

 demonstration of these currents, and an explanation of the laws 

 which regulate them. 



The reason why the question has not as yet been studied from 

 this point of view is probably that the air has been regarded as 

 an insulating medium in which only momentary electric discharges 

 occur, and not electric currents. 



In the aurora borealis we have a " brilliant " proof of the exist- 

 ence of these currents, but up to the present the cause has 

 always been sought elsewhere. 



It is of course understood that a great number of savants have 

 long been of the opinion that the aurora borealis was of electric 

 origin. Having obtained, while with the Swedish Polar Expe- 

 dition of 1S68, some experience of electric phenomena in Arctic 

 regions, I made some attempts during the expedition of 1871, 

 near the church of Enare, to see if it was possible with the few 

 means at my disposal to examine this supposed electric current 

 (Nature, vol. xxvii. pp. 322 and 389). I then succeeded, by 

 means of a small discharging-apparatus, in demonstrating the 

 presence of this current and in producing luminous phenomena, 

 but, owing to certain external difficulties which I could not 

 overcome at that time, these results are uncertain. During 

 the year 1882-83 the Expedition at Sodankyla had occa- 

 sion to make some similar but more extensive experiments, 

 which were crowned with success, as I have previously stated 

 in this journal (vol. xxvii. pp. 322 and 389). An electric 

 current from the air towards the earth was proved to exist. 

 Close to the village of Sodankyla we produced, by means of a 

 large "discharging apparatus" or network of pointed con- 

 ductors erected upon the summit of Orantunturi (1000 feet in 

 height) a diffuse yellowish light, which, in the spectroscope, 

 iwed the ordinary auroral spectrum ; and later on a veritable 

 ray of the aurora borealis was produced on the Pietarintunturi 

 Mountain, close to Kultala. On both occasions the electric 

 current was measured. 



Important as were the results of these experiments, they were, 

 however, only of a provisional character, because, in carrying 

 them out, difficulties of every description had to be overcome. 

 In all these experiments the apparatus was connected with the 

 earth by a wire leading to a zinc plate immersed in a well. 

 Owing to the contact of the zinc with the water, an electro- 

 motive force w as produced, and it was therefore probable that 

 the current observed by the galvanometer might have its prin- 

 cipal, or perhaps sole, origin in this electromotive force. 



The expedition of 1883-84 was supplied with instruments for 

 overcoming these difficulties as well as others, and has exa- 

 mined as closely as possible the laws which this current obeys. 



After the arrival of the expedition at Sodankyla about the 

 middle of September, a provisional apparatus was constructed 

 on the mountain Komattivaara, lying 6 km. from the station, 

 and about 437-5 feet = 1297m. high. 



A conducting wdre, supported by Mascart insulators, was 

 placed from the apparatus on the mountain to the station, where 

 it was joined to a galvanometer, which was connected with the 

 earth by a plate of zinc (amalgam) placed in the neighbouring 

 river. After some preliminary trials with this apparatus, 

 which showed that, in spite of the lowness of the mountain, the 

 atmospheric current could be examined, a " discharging ap- 

 paratus," or network of pointed conductors, was erected upon 

 a solid wooden structure, and was ready by October 19. The 

 apparatus consisted of iron wires with welded points 0"5 m. 

 apart. The wire was arranged in a series of squares I '5 m. 

 from each other, resting upon sulphuric acid insulators attached 

 to poles which were nailed to a wooden frame. The 

 wire with the points covered a surface of 364 square metres. 

 With this apparatus extended experiments were made, chiefly 

 relative to the different kinds of conducting plates to the earth, 

 but space does not permit me to discuss these experiments here. 

 The galvanometer showed a current from the earth towards 



