Euler justly observes, that the theory of M. de Mairan is not only exposed to the objection of resting upon hypothesis, rather than upon observed facts, but that it is inconsistent with the direction in which the aurora is constantly observed to move ; which is not from the equator towards the poles, but conversely from the poles towards the equator. He himself as cribes the aurora, not to the zodiacal light, but to the luminous particles of our own atmosphere, driven be yond its limits by the light of the sun, and sometimes ascending to the height of several thousand miles. This, it must be owned, is no very brilliant specimen of the philosophical acumen of this celebrated ma thematician. See Alen?. Acad. Berlin, 1746, p. 117.
As soon as the phenomena of electricity, and the laws by which they are governed, were tolerably un derstood, philosophers very naturally had recourse to this agent, as affording a satisfactory explanation of the aurora borealis. Tlie brilliancy of its light, the rapidity of its motions, and the instantaneous chan res of feral which it underwent, all seemed plainly to poi:It to this powerful element as the cause. of these striking phenomena. Mr Hawkcsbee, too, had very early shewn, that the electrical fluid assumes an appearance resembling the aurora borealis, when it passes through a vacuum or highly rarefied atmo sphere. If a glass tube, resembling a Florence flask in size and shape, be exhausted of air by means of a stop-cock and syringe fixed to its mouth, and be then excited by friction, it will appear filled with a pale light, resembling the aurora borealis, which will go and come at intervals, sending forth brilliant flashes, exactly as this meteor does in the heavens. If either end of the flask be presented to the conductor of an electrical machine, the other being held in the hand, a constant stream of pale light will be transmitted through it, proceeding from the conductor. Mr Canton, also, contrived to exhibit an imitation of the aurora borealis, by means of electricity transmitted through the Torricellian vacuum, formed in a glass tube about three feet long, and hermetically sealed. When one end of the tube is held in the hand, and the other applied to the conductor of an electrical machine, the whole tube is illuminated from end to end, and will continue luminous for a considerable time after it has been removed from the conductor. If, after this, it be drawn through the hand either way, the light will be uncommonly intense, extending without the least interruption, from one hand to the other, even throughout its whole length. By this
operation, however, a great part of the electricity is discharged ; nevertheless the tube will flash at inter vals, if held at one extremity and kept quite still ; but if it be grasped by the other hand at the same time in a different place, strong flashes of light will hardly ever fail to dart from one end to the other, which will continue twenty-four hours and longer, without any fresh excitation. An arched double ba rometer, of a considerable height, exhibits these phe nomena in a still more striking .manner.
Thus we find that a small quantity of electricity, excited in a highly rarefied atmosphere, or in a me dium approaching to a perfect vacuum, will exhibit luminous appearances entirely resembling the aurora borealis, for a very considerable space of time. With respect to the variations of colour which we find in the aurora borealis, these seem fairly ascribable to the different of rarefaction of the air ; for the same electricity which appears white in a very rare medium, becomes blue, purple, or red, in a medium of increased density ; as is fully evinced by the fol lowing experiment. Let an electrical machine and an air pump be so disposed, that while the machine is worked a succession of strong sparks shall be com municated from the prime conductor to a metallic knob attached to the top of the receiver of the air pump. Let now the exhaustion of the receiver pro ceed, and we shall soon perceive the electricity for cing itself through the air within it, in a visible. stream. At first this stream is of a deep purple co lour ; but, as the exhaustion advances, it changes to blue ; and at length to an intense white,. with which the whole receiver becomes completely filled.
This experiment would appear to establish the iden tity of the aurora borealis with. electric light ; and it may be mentioned as collateral proofs of this iden tity, that the atmcspliere is found, by the clectrome ter, to abound with electricity when the aurora shines forth ; that the aurora, when strong, is accompanied with the whizzing or crackling sound of electricity ;_ and that the magnetic needle is evidently disturbed by the aurora, as well as by the action of an electrical machine, or by the natural electricity of a thunder storm.