370 THE AURORA BOREALiS. 



rnagnetisin. This vapor must consist of volatilized iron or other mag- 

 iielic metals, ejected from polar volcanos and forced to great heights 

 into the atmosphere, vvhere, forming strata, it would perform the othce 

 of electric conductor. If the metalic particles were sufficiently near 

 each other or the cloud sufficiently dense, the electricity would flash 

 along without producing light- but if the cloud were very rare, the elec- 

 tric light would be seen between them and so produce the appearance 

 of luminous lines, and the particles themselves would become luminous. 

 The electricity he also supposed originated from the polar volcanos. 

 But to this theory it may be objected that we know of no such polar 

 volcanos as are adequate to produce the effects ascribed to them, and 

 the volcanic vapors as far as known consist principally of non metalic 

 gases, and comminuted earthy matters. 



S. The most plausible theory yet suggested is that in which the light 

 is referred to electricity, and the aurora is regarded as an electrical dis- 

 play. This is, indeed, in part the theory of Biot, which has, in some 

 respects, been deemed insufficient. But the electrical theory, in. vari- 

 ously modified forms, has been advocated by the most eminent electri- 

 cians of the past and present centuries : such as liawksbee, Canton, 

 Beccaria, Franklin, Faraday, and others. 



The first two showed that the principal appearances of the aurora can 

 be exhibited by means of conimon electricity — an experiment which al- 

 most every lecturer on that branch of science now performs as a class 

 illustration. If, for example, a tube of any convenient length and diam- 

 eter be made air-tight, and exhausted by means of an air-pump, it will 

 exhibit flashes of light diffiised through the space within resembling the 

 auroral streamers and waves, if either end be held in the hand of the op- 

 erator, and the other be presented to the prime conductor of an electrical 

 machine.' As each successive spark passes upon the cap of the tube a 

 flash passes through the latter to the other end. The color of the light 

 will be influenced by the extent to which the exhaustion has been car- 

 ried ; if this be nearly perfect, the light will be white : but if only par- 

 tial, it will be of some shade of blue, purple or red. Perhaps, however, 

 the color still more depends upon the state of condensation of the elec- 

 tric light; in the ordinary atmospheric flashes, the electric matter, in 

 order to overcome the resistance, must pass in a condensed stream and 

 consequently in large quantities, from point to point, and is then white; 

 .so in the exhausted tube the quantity passing in a given time, and there- 

 fore its density, may be greater than when the exhaustion is only partial ; 

 and this may explain the greater whiteness of the aurora near the hori- 



