THE AUROEA BOREALIS, 745 



The results of tlie observations entirely agree with M. Ang- 

 strom's explanation of the origin of the rays. There are only the 

 relatively bright bands 528*9 and 535*9, which cannot be ac- 

 counted for. The brightest gi-oup of iron lines is very near, but 

 does not correspond with them. 



22. Observations on the Electricity of the Air made in the 

 Swedish Arctic Expedition in 1872-73, by M. Auguste 

 WijKANDER. — (Archives des Sciences Physiques et Natu- 

 relles, vol. li., p. 31.) 



Many attempts have been made to discover the nature of the 

 electricity of the air in high latitudes, because of its great impor- 

 tance, especially with reference to the Aurora Borealis. But the 

 result of all these researches has been that not a trace of elec- 

 tricity has been found until the observations of MM. Bravais and 

 Lottin, in the winter of 1838-39, at Bossekop in Northern Nor- 

 way. On their way to the north, in experiments made on board 

 during the summer, positive electricity was collected by shooting 

 arrows into the air ; but in Northern Norway, and in the month of 

 October, at Bossekop, the experiments were repeated without any 

 result. 



On February 25 five, and on March 1 eight, experiments with 

 kites were more successful, giving weak charges of positive 

 electricity. 



The observers concluded that the air is charged with positive 

 electricity in those regions as well as in lower latitudes, but they 

 had also found that the air is a much better conductor of electricity 

 in those regions than elsewhere. 



In the first Swedish Expedition in 1 868, M. Lemstrom made 

 observations with an electrometer by M. Lament as well as with 

 a straw electroscope (Snow Harris ?), but these instruments were 

 not sufficiently delicate to detect the slight traces of electricity 

 which are found in these regions, and no traces were obtained. 



In the second Swedish Expedition, 1872-73, a modification of 

 Thomson's electrometer was made by M. Holmgren, who also 

 gav^e excellent directions for making the observations. 



The Leyden jar of Thomson's electrometer was replaced by an 

 alcohol battery, the poles of which are attached to the quadrants, 

 the opposite ones being connected together. The needle is in- 

 sulated from the quadrants and is in connexion with the collect- 

 ing ball. The whole was sealed almost hermetically by a brass 

 cover, and the air dried by chloride of calcium. In employing an 

 alcohol battery in place of a Leyden jar, the readings made at 

 different times were more nearly on a uniform scale ; but, for the 

 better comparif=on of the different readings with the strength of 

 charge, an alcohol battery of 25 pairs was always connected up to 

 the electrometer, and its effect measured before each observation. 



For the winter observations a special room or hut was built by 

 means of sacks filled with northern moss (?) at a distance of a 



