XII. SPECIAL COLLECTIONS. 339 



pole of a Holtz' electric machine (a machine of Carre (Paris) was employed 

 with great advantage), of which the positive pole is in metallic connexion 

 with the earth. As soon as the machine is put in movement, the sphere 

 being charged, becomes negative-electric, and at the same time there goes 

 through all the tubes a current of reddish-lilac light, so that they altogether 

 form a shining bow-shaped belt. With an ordinary machine this pheno- 

 menon may still be observed when the lower ends of the tubes are at a 

 distance of two meters from the sphere. This proves evidently that the 

 electricity flowing out from (or into) the sphere not only traverses the 

 layer of air that is between, but goes also with such power through the 

 tubes that the gas therein becomes glowing by the heat that the electric 

 current produces, as is well known. In order that the electricity might 

 more easily flow out in the air from the sphere, this latter is furnished 

 with points. These points, as well as the metallic union between the 

 upper end of the tubes and the earth, are of no absolute necessity, for the 

 phenomenon may be produced without them, only that then the distance 

 between the sphere and the tube must be considerably reduced. 



The described light-phenomenon produced by the apparatus proves clearly 

 that a current of electricity may go through a layer of air of ordinary pressure 

 760 mm without producing the light-phenomenon, but if it meets in its way a 

 space of rarefied air of low pressure (from to 30 nm to 4o mm ) there arises 

 immediately a light-phenomenon, caused by the fact that the current makes 

 the molecules of gas glow. 



On the Theory of Polar-Light. 



The knowledge we have acquired of the electric state of the earth proves 

 that it is a conducting body, charged with a small quantity of negative elec- 

 tricity, and surrounded by the atmosphere, in general charged with positive 

 electricity. Though this latter might be produced by an influence from the 

 earth, it is still very probable that it proceeds from the process of evaporation, 

 either directly by this phenomenon itself, or by the friction of vapour against 

 particles of air. The atmospheric air possesses a very small conducting 

 power for electricity when dry and of ordinary pressure, but the conducting 

 power increases considerably as soon as the air becomes moist and rarefied. 

 It has been proved by experiments that the conducting power is highest at 

 a pressure between 5 mm and 10 mm , and goes then 10,000 times beyond 

 the conducting power at a pressure of 760 mm . If the rarefaction of the 

 air is carried further than 5 min , the conducting power diminishes again, 

 but very slowly. It is known that in proportion to its elevation over the 

 surface of the earth the air becomes more and more rarefied according 

 to an irrefragable law, which finds its expression in the formula given 

 by Laplace, and that consequently, at a certain elevation the earth is 

 surrounded by a layer of air that has a pressure of only 5 mm ; the con- 

 ducting power for electricity in this layer is sufficiently great to allow of its 

 being regarded as a conductor in comparison to the air in lower regions, 

 and even in the highest. The negative electric earth is thus surrounded 

 by a conductor for electricity concentric with it. All the positive elec- 

 tricity that attains the space of rarefied air of about S"" 11 , or, as it might be 

 called, this conductor of air, submits almost to the same laws as if it were 

 in a real conductor, and must thus set in a restricted manner according to the 

 influence of the electro-negative earth. Part of the electricity, conducted by 

 the vapours, remains on the clouds in the atmosphere and discharges in form 

 of lightning and thunder ; another part attains'the space of rarefied air or 

 conductor of air, by the fact that the vapour itself, submitting to well-known 

 physical laws, rises to this elevation, and also because electricity, according 

 to its nature, endeavours always to set on something. 



Y 2 



