434 



NA TURE 



{^March lo, 1887 



system and all the various parts composing a fully deve- 

 loped auroral display : the corona forming in the mag- 

 netic zenith, the streamers flitting parallel to the line 

 of the dip, the arch erecting its summit in the magnetic 

 meridian. When Faraday succeeded in obtaining lumin- 

 ous effects through magnetic action, and Rudolf Wolf 

 demonstrated the subjection of aurora: to an identical 

 periodicity with magnetic variations, the case might have 

 appeared complete. 



Yet the pure and simple magnetic theory of the aurora 

 borealis, when attempted to be realised, either eluded the 

 grasp of thought, or was found to involve admissions not 

 very easy to make. Dalton, whose propensity towards 

 forming distinct conceptions was a primary quality of his 

 mind, had the courage to give it definite shaps in 1793 

 (in his "Meteorological Observations and Essays"). 

 Compelled, as he supposed, by the exigencies of observa- 

 tion, he conceived an envelope of an elastic fluid partak- 

 ing of the nature of steel to replace atmospheric air at 

 a height of about a hundred miles, and to supply the 

 material alike of arches and streamers, shown, by their 

 disposition in space, to be of a ferruginous character. It 

 can scarcely be wondered at that the idea, in spite of 

 Biot's adoption and development of it, failed to strike 

 root. More vitality was in Canton's interrogative sugges- 

 tion forty years earlier : " Is not the aurora borealis the 

 flashing of electrical fire from positive towards negative 

 clouds at a great distance, through the upper part of the 

 atmosphere, where the resistance is least?" (Phil. Trans., 

 vol. xlviii. p. 357.) 



It was not, however, until the electrical illumination of 

 rarefied gases came to be studied with detailed attention 

 that the full effect of the visual identity of the two kinds 

 of phenomenon became sensible. The analogy was 

 defined and completed by some experiments made by 

 De la Rive in 1853. They showed that a luminous dis- 

 charge in an exhausted vessel, when influenced by a 

 powerful electro-magnet, condenses into a ring of light 

 encircling its pole, endowed, moreover, with a rotatory 

 movement such as ha; frequently been observed to affect 

 a system of auroral streamers. Here, for the first time, 

 their true part in such displays was assigned to the forces 

 emanating from the earth's magnetic poles. It is a 

 directive, not a formative one. The structure, not the 

 occurrence, of aurorae is conditioned by them. 



A marked change in scientific opinion resulted, accord- 

 ingly, from these investigations. The " magnetic efflu- 

 vium" theory of the aurora borealis finally disappeared, 

 and what we may call the "illuminated vacuum" theory 

 took its place. The change was acco.iipanied by a shift- 

 ing of the ground of inquiry. What was urgently needed 

 in order to render intelligible the mode of action pro- 

 ducing the gorgeous flame-vesture of northern skies, w'as 

 perceived to be, not so much improved knowledge (how- 

 ever desirable) of the laws of terrestrial magnetism, as a 

 sound doctrine of atmospheric electricity. Here De la 

 Rive broke down. His hypothesis of a polar accumula- 

 tion, through the agency of the winds, of the positive 

 charge of the air, was obviously untenable. 



Nothing better was, however, proposed until 1S7S, 

 when Prof Edlund, of Stockholm, applied the principle 

 of '■ unipolar induction," discovered by Weber in 1S41, to 

 explain the phenomena of atmospheric electricity {Phil. 



Mag., vol. vi. p. 360). The effects thus designated are 

 really derived, as a particular case (in M. Edlund's 

 probable view); from the well-established laws of magnetic 

 action upon electric currents. 



Each element of the system of currents formed by the 

 electrified particles of the rotating terrestrial crust and 

 atmosphere is thus urged, by the powerful magnet which 

 the earth may legitimately be regarded as inclosing within 

 it, along a path at right angles to the line drawn from 

 each of its poles to the current-element. The direction 

 and relative strength of the impelling forces are indicated 

 in the accompanying figure, copied from the work before 



us (p. 166), where the particle m' is solicited towards X 

 by the south pole (pointing north) of the magnet Sn, and 

 towards _§■ by its south pole. The combined result is to 

 drive the particle upward and poleward along a line every- 

 where perpendicular to the swing of the dipping-needle. 

 The vertical component, accordingly (represented by the 

 arrows standing erect on the circle ;;/ »i in"'), attains a 

 maximum at the magnetic equator, where the dip vanishes ; 

 the tangential component is there = o, and attains its 

 highest value in middle latitudes. 



Several remarkable effects ensue : first, that atmo- 

 spheric electricity gains potential with elevation — an 

 observed fact ; next, that it is constantly travelling away 

 from the equator towards either pole. The circulatory 

 process, however, thus set on foot, must be carried 

 further ; and in its continuance and completion M. 

 Edlund finds the key to the auroral mystery. 



At and near the equator, recomposition of the positive 

 electricity of the air with the negative electricity of the earth 

 is opposed by the whole strength of magnetic inductive 

 repulsion, there acting vertically. When effected at all, 

 it can only then be by sudden, violent, disruptive dis- 

 charges, apparent to our senses as lightning. But the 

 increasing inclination of the magnetic needle in higher 

 latitudes renders the line of no resistance marked by it 

 continually more practicable as an avenue of descent for 

 the accumulating positive fluid. Hence, when it hai 

 attained a certain potential, gradual discharges take 

 place over two polar zones, along the line of the dipping- 



