cumulate again. It was thought that years of dry weather were years of maxima of auroras, and it seemed natural to suppose that moisture would hinder exhalations. Extensive efforts were made, without success, by studying the properties of the recently discovered phosphorescent substances, to determine the nature of the stuff that thus shone in space. Previous to this, an explanation of the phenomenon had been suggested by supposing a fermentation of gross exhalations from the earth's surface which were driven toward the pole and there took fire.
Quite different from this was Mairan's theory; and the reading of his book, "Traité physique et historique de l'aurore boréale" ("A Physical and Historical Treatise on the Aurora Borealis"), which appeared in 1733, is still indispensable, after a hundred and fifty years, to any person wishing to study the meteor to-day. Rejecting the ideas outlined above, and another curious hypothesis, that the rays of the sun were reflected from the polar ice, and sent back to the observer from the concave surface of the upper strata of the atmosphere, he had recourse to the zodiacal light which had been observed by Cassini some fifty years before. While some explained this phenomenon by supposing a ring of light concentric with the sun, and surrounding it without touching it, others, and Mairan among the number, considered it a prolongation of the solar atmosphere, accumulated chiefly in the plane of the ecliptic or of the solar equator, and extending beyond the orbit of Venus. Emanations from the sun, or rather the corona that surrounds it, according to Mairan, strike our atmosphere and illuminate our globe. Then, must we suppose that the zodiacal light shines of itself? That is not necessary, says Mairan. A chemical combination, an essentially luminous precipitate, results from the mixture that takes place in the upper regions of the atmosphere. This supposition is hazardous, and Mairan seems to be a little too fast. It is, however, indisputable that then, as now, auroras were more frequent in March and September, or the months when the zodiacal light is brightest. It is also worthy of remark that Angström, in 1867, and Respighi, in 1872, found in the spectrum of the zodiacal light a green ray identical with a line of the same color characteristic of the aurora borealis.
Mairan found a redoubtable antagonist in the celebrated mathematician Euler, who did not admit the hypothesis of an immense solar atmosphere, and believed only in the existence of a ring. He invented, in explanation of the meteor, a somewhat obscure theory, according to which the subtile and rarefied portions of the air were driven away from the surface of the globe, and the particles, having become lumi-
- Mairan observes that, the centrifugal force being less toward the poles than at the equator, the parts of the globe at the tropics will repel the foreign matter, and it will accumulate toward the high latitudes. Hence there will be few auroras except in the frigid and temperate zones; and this is the case.