I i 



polar-regions a belt (see thc dark ring fig. 1) where the 

 number of polarlights is in maximum, it also exists in the 

 same regions a belt, in which the earth-currents attain a 

 maximum. The earth-current belt seems to be situated a 

 little more to the south than the aurora-maximum belt. 



The general conclusion is drawn in the close of the 

 next article. 



5. 



Electrical current from the atmosphere. 



An electrical current in the atmosphere seems not to 

 be in harmony with our conception of the air as an insu- 

 lator. The existence of such an electric air-current is ho- 

 wever proved by a number of facts which are incon- 

 testable. 



Though many different explanations of the nature of 

 the polarlight have been given, the general opinion seems 

 now to agree about the only possible conception: that the 

 polarlight or aurora is caused by electrical currents in the 

 atmosphere. 



Since \ve had, by studies and numerous observations 

 in the polar-regions, convinced ourselves of the correctness 

 of this explanation our aim has been to produce the deci- 

 cive proofs of it. 



It is very singular that studies of the atmospheric elec- 

 tricity, made in polar-regions with electrometer, have never 

 given definite results showing changes in this electricity by 

 times of auroras, a fact which has been used as a proof 

 against the electrical nature of the phenomenon. 



Such a change does not however need to happen with 

 necessity. By the electrometer we determine the difference 



