The annexed chart from Prof. Loornis s paper (Sill. Jour., April 1873) shows, in a very striking manner, the correspondence of auroræ;, magnetic variation, and sunspot area since 1776. It is not improbable that there may also be changes of longer period which our observations are yet insufficient to determine.
Diagram showing Correspondence of Auroræ, Magnetic Variation, and Sunspots.
Annual dis-
tribution.It has frequently been stated that the aurora returned periodically on certain days in the same manner as meteors. On the 3d of February brilliant auroræ occurred in 1750 and 1869, and on the 4th in 1869, 1870, 1871, 1872, 1873, and 1874; on the 13th February in 1575, 1821, 1822, 1865, and 1867; on the 6th March in 1716, 1777, 1843, 1867, and 1868; on the 9th September in 1776, 1827, 1835, 1866, 1868, 1872, and on the 29th in 1828, 1840, 1851, 1852, 1870, and 1872. This conclusion, however, is not supported by systematic investigation. A considerable catalogue of auroræ was divided into decennial periods, and it was found that the maxima of one period rarely coincided with those of others, and that the larger the number of years taken into account the less prominent the maxima appeared,—evident proof that they were only accidental. It may be s however, that if only prominent auroræ had been considered, more periodicity might have been found, or that the periodicity is constant for very short periods only.
Although no daily periodicity can be affirmed, there are two well-marked annual maxima in March and October, of which the latter is the greater, and two minima—the greater in June and the less in January. In this respect the aurora differs from the sporadic meteors, which have a maximum in autumn and a minimum in spring. It also differs from meteors in the hours of its appearance, the former being most frequent before and the latter after midnight.
Meteoric
hypothesis.Although the electric hypothesis is the one generally accepted by scientific men, it is only fair to allude to one that has been recently proposed independently by Dr Zehfuss (Physikalische Theorie, Adelman, Frankfort) and by H. J. H. Groneman of Gröningen (Astronomische Nachrichten, No. 2010-2012). According to this view, the light of the aurora is caused by clouds of ferruginous meteoric dust, which is ignited by friction with the atmosphere. Groneman has shown that these might be arranged along the magnetic curves by action of the earth’s magnetic force during their descent, and that their influence might produce the observed magnetic disturbances. The arches may be accounted for by the effects of perspective on columns suddenly terminated at a uniform height by increase of atmospheric density, while the correspondences with iron lines in its spectrum are sufficiently close to favour the idea. Ferruginous particles have been found in the dust of the Polar regions (E. A. Nordenskiold, Ast. Nach., 1874, § 154), but whether they are derived from stellar space or from volcanic eruption is uncertain. The yearly and eleven-yearly periodicity of auroræ tends to support the theory, but it is a formidable difficulty that, while shooting stars are more frequent in the morning, or on the face of the earth which is directed forwards in its orbit, the reverse is the case with auroræ. Groneman meets this difficulty by supposing that in the first case the velocity may be too great to allow of arrangement by the earth s magnetic force, and that, consequently, only diffused light can be produced. He accounts for its unfrequency in equatorial regions by the weakness of the earth’s magnetic force, and the fact that, when it does occur, the columns must be parallel to the earth’s surface. Without pronouncing in favour of this hypothesis, it must be admitted that it furnishes a plausible explanation of the phenomenon, although we have no evidence that meteoric dust, even if it exists, would produce the observed spectrum, and, as has been already remarked, the iron coincidences are of little weight.
Although we must confess that the causes of the aurora are very imperfectly explained, we may hope that the rapid progress which the last few years have witnessed in bringing terrestrial magnetism under the domain of cosmical laws may soon be extended to the aurora, and that we shall see in it fresh evidence that the same forces which cause hurricanes in the solar atmosphere thrill sympathetically to the furthest planets of our system in waves, not only of light and heat, but of magnetism and electricity.
The following is a list of the most important papers, treatises, and Biblio-
graphy.works on this subject: Berlin Mem. 1710, i. 131 ; Halley, Phil. Trans. 1716, 1719, xxix. 406 xxx. 584; Hearne, Phil. Trans., xxx.
