48 ON THE RECENT SECULAR PERIOD 



But we are not to regard the nebulous body, or bodies, from which the auroral 

 matter is derived, as floating loosely in space, but as revolving around the sun, with 

 an orbit and periodic time of its own ; subject, however, to perturbations from the 

 action of the planets, and possibly also from the "resisting medium." We can 

 think of no other way in which this nebulous matter can come into the earth's 

 atmosphere, and produce the aurora borealis, than to suppose that the earth, in its 

 revolution around the sun, comes so near to the nebulous body, or bodies, as to 

 bring within its own sphere of attraction some portion of them. The swiftness of 

 motion accompanying the auroral displays (by which we mean the motions of trans- 

 lation, and not mere flashes or undulations) may naturally result from the relative 

 motion of the earth and the nebulous matter, or from that motion combined with 

 the effect of the earth's attraction. The magnetic phenomena may be referred to 

 the earth, acting as a magnet on ferruginous, or other magnetic matter, of which the 

 auroral vapor gives clear indications of being constituted. The luminous appear- 

 ances, it seems to me most probable, are owing to the evolution of light and heat, 

 due to the sudden condensation of air, and naturally consequent upon the rapid 

 passage through it of such matter as the auroral matter. The great returns which 

 we have denominated secular periods, imply that the nebulous body from which 

 they are derived, returns to the same relative position with respect to the earth 

 after a cycle of years, and remains near to it, on account of the near ecpiality of 

 their periods, for a long time. Such a relation between the two bodies may be con- 

 ceived to produce the great periods of auroras, while the ordinary forms of the 

 phenomenon may result either from a more distant approach of the nebulous body, 

 or from the earth's encountering other portions of the same material, in a more 

 diffusive state. It will readily be perceived that a nebulous body, like that inferred 

 to exist in the neighborhood of the earth's orbit, could not remain stationary, since, 

 without a motion of revolution, it would descend directly to the sun. It must, 

 therefore, have an orbit and a period of its own. The duration of the great periods 

 of the aurora for upwards of twenty years together, and the almost nightly occur- 

 rence of the exhibitions in their humbler forms at the higher latitudes, prove that 

 some portions of this body are always hovering near the earth. 



The idea of a connection between the aurora borealis and meteoric showers, 

 occurred to me as early as April, 1835, according to a record made at that date in 

 my note-book. Many great auroras have occurred about the time of the November 

 meteors, and during a grand auroral display, an uncommon number of shooting 

 stars have frequently been observed. In the great auroral exhibition of August, 

 1827, they made a very conspicuous figure. In the Regents' Report of the State 

 of New York, for 1836, Professor Joslin remarks, that in the great aurora of 

 November 17, 1835, many shooting stars were seen, and he expresses the opinion 

 that shooting stars are intimately connected with the aurora. On this point the 

 illustrious Humboldt thus expresses himself: " The aurora borealis showed itself 

 with great intensity during the occurrence of the most magnificent display of 

 meteors j'et observed, that described by Olmsted on the 12th and loth November, 

 1833. The aurora was also seen during the periodical phenomenon in 1S38, at 

 Bremen, where, however, the fall of meteors was much less striking than at 



