430 Prof. E. Edlund on Atmospheric Electricity 



It is evident, therefore, that aurora boreales must become 

 more numerous in proportion as we come nearer to this ring 

 from the south, and that their greatest frequency will be under 

 this ring itself, while they commence to decrease again in 

 number and in brightness at still higher latitudes. Southward 

 of the ring, the observer sees the aurora in the north ; if he is 

 beneath it, the aurora occupies, when seen under favourable 

 circumstances, the greater part of the sky ; and, lastly, if he 

 is northward of the ring, the aurora appears on the south. 

 It can hardly be admitted that this ring occupies an absolutely 

 fixed position in the atmosphere ; rather is it probable that on 

 one occasion it is situated more to the south or more to the 

 north than on another — a circumstance which may depend on 

 the modifications of the electric conductivity of the terrestrial 

 atmosphere. If, then, the place of the observer is at a point 

 of the earth's surface over which the ring is usually situated, 

 he may see the aurora borealis sometimes to the north, some- 

 times to the south. If this ring formed a true circle with the 

 magnetic pole for its centre, if the intensity of the descending 

 current were the same at every point, and consequently pro- 

 duced everywhere the same intensity of light, an observer on 

 the earth to the south of the ring would necessarily see the 

 crown of the boreal arc in the plane passing through the place 

 of observation, the centre of the earth, and the magnetic pole. 

 Now, if the declination-needle placed itself entirely in this 

 plane, one would, in consequence, always perceive the summit 

 of the boreal arc in the plane of terrestrial magnetic declina- 

 tion. But the ring in question does not form a perfect circle, 

 nor can we assume that the descending currents possess every- 

 where the same luminous intensity. Besides, the plane in 



At a certain distance from the knob some Geissler tubes were fixed to an 

 insulated stage permitting them to be brought near to or moved away 

 from the knob. The tubes were, as usual, furnished at their extremities 

 with thin platinum wires. The posterior extremities were connected with 

 the earth hy a conducting wire, while the anterior extremities, or those 

 turned towards the knob, were insulated in the air. Although there was 

 no metallic communication between the tubes and the knob, they never- 

 theless commenced to be luminous as soon as the machine was put in mo- 

 tion, and that even when the distance between the tubes and the knob 

 rose to 2 metres. The current which produced the luminous appearance 

 must therefore have traversed a length of 2 metres through a stratum of 

 air of ordinary density. It was natural that no luminosity was produced 

 in this layer of dense air. These experiments appear to me to have much 

 analogy with the phenomena produced on a grand scale at the formation 

 of the aurora borealis in the terrestrial atmosphere : the electric currents 

 coming from the upper strata descend into the earth without producing 

 luminous phenomena in the lowest strata of the atmosphere. 



