The preceding description indicates the general features of the appearance of the A. B. ; but several auroras have been described which presented striking peculiarities. Sometimes the phenomenon assumed the form of one or more curtains of light, depend ing from dingy clouds, whose folds were agitated to and fro, as if by the wind. Some times this curtain seemed to consist of separate ribbons of light, arranged side by side in groups of different lengths, and attaining their greatest brilliancy at the lower edges. In this country the A. B. seldom occurs with the distinctness and brilliancy which attend it in northern latitudes, but the description just given portrays the type to which such appearance of the meteor more or less approximates.
The height of the aurora has been variously estimated. The first .observers were inclined to place the seat of it beyond the atmosphere; but this hypothesis is untenable, as the aurora does not seem to be affected by the rotation of the earth, but appears to be in every respect a terrestrial phenomenon. By taking observations of the altitude of the highest point of the arch of the same aurora at different stations, heights varying from 5 to 500 in. have been calculated. The cause of these widely differing results may be found in the probability that exists of each observer seeing a different arch of the aurora for himself, and he is, in consequence, furnished with no comparable or reliable data for his calculations. It is now, however, generally admitted, on what are con sidered to be sufficient grounds, that the A. B. occurs at various heights, and that it is seldom found beyond 90 m. above the surface of the earth. The distance of the stations at which the same aurora has been visible, indicates the enormous geographical extent, and likewise the great altitude which the phenomenon frequently attains. One aurora, for instance—that which occurred on the 3d of Sept., 1839—was seen in the Isle of Skye by M. de Saussure; at Paris, by the astronomers of the observatory; at Asti, in northern Italy, by M. Quetelet; at New Haven, in Connecticut, by Mr. Merrick; and at New Orleans, by credible observers. On the other hand, some observers of eminence assert that the aurora sometimes descends to the region of the clouds, and appears almost as a local phenomenon. A brilliant aurora was seen by Mr. Farquharson, the minister of Alford, in Aberdeenshire, on the 20th of Dec., 1829, from 8 to 11:30 in the evening, above a thick bank of clouds, which covered the tops of the hills to the n. of where he lived, and which never attained an altitude of more than 20°. The same aurora was
seen in the zenith, at 9:15, by Mr. Paul, another minister, at Tullynessle, which is about 2 in. n. of Alford, so that the height of it could not have been quite 4000 ft.
The noise that is alleged to accompany the A. B. in high latitudes would indicate for it a comparatively moderate height. Some of those who have heard it, compare it to.the nose that is produced by the rolling of one piece of silk upon another; and others, to the sound of the wind blowing against the flame of a candle. In Siberia, it has been related that this noise sometimes resembles that attending the discharge of fireworks; and that the dogs of the hunters, when overtaken by such an aurora, lay themselves with terror on the ground.* The intimate connection between the A. B. and the magnetism of the earth is shown by various facts. During the occurrence of the phenomenon, the magnetic needle appears very much disturbed, sometimes deviating several degrees from its normal position, and appearing to be most affected when the aurora is brightest; and this oscilla tion is frequently perceived far beyond the district where the aurora is seen. The vertex, likewise, of the luminous arch is almost always found to be in or very near the Magnetic and the boreal crown has its scat in the prolongation of the freely suspended needle. There seems, moreover, to be a connection between the magnetic poles of the earth in regard to the aurora for, so far as has been ascertained, the mete oroccurs simul taneously at both. The A. B. appears to be an electric discharge connected with magnetic disturbance. If one of Gassiot's vacuous tubes be brought near an electric machine, or between the poles of an induction coil, flashes of light pass between the ends, which bear a striking resemblance to the A. B. A comparison of the spectra of the two goes far to establish identity. The auroral spectral line, according to Angstrom, is a yellow line near the sodium line, and is the same as the air line seen in the solar light when the sun is near the horizon. Other lines, however, have been seen, which cannot as yet be produced by the physicist from any known substance. • A line drawn through the s. of Spain to the n. of the Sandwich islands, and through Cuba, marks the southern limit of the A. B. (in the northern hemisphere); though occasional displays have been noticed even further s. To the n. of a line passing through Edinburgh, the frequency of the A. B. rapidly increases, until a maximum is reached in a line through the n. of Spitzbergen, after which the frequency diminishes as the north pole is approached.