356 De la Rive on the Aurora Borealis. 
M. de Tessan cites an observation made in 1818, by M. Baral, 
another F'rench naval officer, on the same coasts of New Holland, 
who found that he had been making a wrong course from follow- 
ing his needle; there had been no storm, and the compass had 
not been touched. But on the evening of the same day, there 
was a brilliant aurora, and to this he attributes the deviation—a 
conclusion which could not have been dictated by theory, since 
at the time (in 1818) the relations between electricity and mag- 
netism were not known. 
The intimate connection between the aurora and terrestrial 
magnetism, has led Humboldt to designate as a magnetic storm 
a succession of disturbances of equilibrium in the magnetic for- 
ces of the earth. The presence of this storm is indicated by the 
propose to bring some direct experiments, as well as the results 0 
2. Proposed Theory. 
The atmosphere in its normal state is constantly charged with 
a considerable quantity of positive electricity, which increases 4S 
we ascend, starting at the earth’s surface where it is zero. ; 
I will not inquire into the origin of this electricity: what 1s 
certain is that its production is connected with the action of the 
sun, since its intensity is subject to diurnal variations. It may 
be a question whether the sun acts directly, either through 1ts 
light or its heat, on the constituents of our atmosphere, and so 
produces the electricity ; or whether it is an indirect effect of the 
