POLAR-LIGHT APPARA TUS. 161 



as on their form. The earth might, without a remarkable difference, 

 be considered as a sphere, and likewise the conductor of air, but in 

 their reciprocal position to each other it appears that the space of 

 rarefied air of 5 mm approaches much nearer to the earth at the poles 

 than at the equator, principally in consequence of the inequality of 

 the temperature of the air in the two places. If we assume the mean 

 temperature of the air round the equator to be 25, at the poles 12, 

 and everywhere on the conductor of air 60, and we suppose at the 

 same time the air everywhere half saturated with moisture, and that 

 the temperature is reduced in proportion to the elevation, we find, 

 if the above-mentioned formula of Laplace* is applied, that the 

 conductor of air, at the equator, must be at an elevation of 37*47 

 kilometres, and at the poles but 34*25. 



In consequence of this relative position, and if the two conductors 

 are regarded as conducting surfaces, the electric density on them both 

 becomes about 9 per cent, greater at the poles than at the equator, and 

 the power, by which the two electricities endeavour to join again, at 

 least 20 per cent., but probably 30 or 40 per cent, greater, if all the 

 circumstances are considered, at the poles than at the equator. It is 

 in these facts we have to seek the principal cause of the accumulation 

 of electricity at the poles of the earth and of the phenomenon that 

 occurs there, and is called polar light or aurora borealis. 



It is a remarkable fact that thunderstorms diminish as well in 

 number as in intensity in proportion as we remove from the equator, 

 and that at the 7oth degree of latitude they cease completely, after 

 having shown once more in the highest north vestiges of their primitive 

 intensity. In Finnish Lapland, for instance, thunderstorms are very 

 uncommon, but when they occur they are extremely severe, and are 

 almost always accompanied by thunderbolts. This peculiarity has 

 probably its cause in the fact that the region of thunderclouds lowers 

 towards the earth in accordance with the same rule as the before- 

 mentioned conductor of air. The reduced number of thunderstorms 



* X= 18-393 metres (1+0-002837 Cos. 2 <f>) (i+'o<>4T+*) /H where x s5gn;fies the eleva . 



2 A 



tion, <j) the latitude, T the temperature at the surface of the earth, t at the upper point H, 

 and h the stand of the barometer for the same points, but duly reduced. 



M 



