of air as the separating medium. An actual proof that the northern lights are caused by electric currents in the atmosphere was attempted by Lemström in 1883. He covered the plateaus of two mountains in northern Finland with a network of copper wires raised several metres above the ground and provided it with hundreds of metal points. The whole was insulated and connected with a zinc plate buried in damp ground in the plain below. A continuous electric current from the air to the ground was noticeable, and a light which appeared hovering over the metal points showed, when examined by the spectroscope, the characteristic line of the auroral spectrum.
The theories, however, according to which the northern lights are a flowing together of terrestrial and atmospheric electricity of opposite kinds, leave unanswered the question as to the origin of these electric fluids. As no adequate cause could be found on the globe for such a tremendous evolution of electricity, attention was directed to the sun as the source of it all. Why should not Helios, the giver of all light and life of our world, be as well the creator of that inexhaustible force of nature that is revealed in the splendors of the northern lights?
As the endless supply of light and heat which is radiated into space by the sun, is accounted for by the contraction of that body, this may also be assigned as the cause of the stupendous generation of electricity. According to the theory of Kant and Laplace, the sun and other heavenly bodies are assumed to have been formed by the condensation of vapors which originally filled all space. This condensation is still going on in the sun in consequence of the enormous radiation of heat into space, and with it the consequent contraction.
Possibly also there might be suggested as a cause the cooling process which the sun is undergoing. It may be assumed, too, that vast amounts of electricity are hurled into space with the ignited masses of gas, whose eruption from the sun may be constantly observed. But it is more probable that the sun acts upon the earth by induction. Try the following experiment: Two insulated spheres are placed near one another, but without being in contact. On one of these spheres a bar of metal is placed, to which there is fastened a screen made of some good conducting material. If one of the spheres is charged with a certain kind of electricity, say, for instance, negative electricity, the opposite kind—in this case positive electricity—will, by induction, be generated on the other sphere. A corresponding amount of negative electricity will in the mean time be discharged on to the screen. An action similar to this may be assumed to be going on between the sun and the earth.
The sun's electricity, which may be assumed to be negative on
