Besides the slow variations of its radius and its thickness, besides the oscillations which displace its center movements, the laws of which are worth studying, the luminous arc rises, falls, and fades away for intervals of some hours. Its light, generally uniform, is heightened by "knots of light" that play from one end to the other. Sometimes a second arc is formed parallel to the first; according to M. Nordenskjöld, this is nearly always concentric with the usual arc and situated in the same plane with it, but farther from the surface. Sometimes, also, the two arcs amalgamate, and a vertically flattened aurora results. Not rarely, supplementary arcs intervene, and frequently luminous rays play between the two arcs and into the undefined exterior space. If, now, we imagine the phenomena growing more complicated and becoming irregular, with the arcs rising above the horizon and the rays multiplying, shooting through the curves in such a way as to illuminate the vacant space, and extending themselves out toward the magnetic south in somewhat oblique directions, we have the common aurora borealis passably explained. Within the projection of the corona, toward the magnetic pole, is a zone where we may observe the auroras in a southerly direction, and, still nearer to that pole, the meteor only rarely illuminates the horizon. A few travelers, Dr. Hayes, for example, noticed this fact some time ago. The zone of no auroras embraces a circle having a radius of about eight degrees.
The labors of M. Lenström, in Lapland, are of particular interest, because they constitute a direct and definite proof of the electrical nature of the aurora borealis. They go further than those of M. de La Rive, for the Swedish observer, instead of operating in his laboratory, has succeeded in reproducing the meteor itself in the open air, and has compelled it to manifest itself, as Franklin forced the lightning to come down from the sky, so that he could examine it scientifically. We must not forget, furthermore, that it is a very meritorious thing to work in a cold of twenty degrees below zero, with a strong wind blowing and the frost all the time clogging the apparatus, having to be constantly on the watch, and enjoying no better shelter than a charcoal-burner's hut.
Not satisfied with provoking artificial auroras, the Finnish expedition, of which M. Lenström was a part, has collected a number of important data relative to the free manifestation of the phenomenon. The observations were made at Sodankylä (lat. 67° N., long. 27° E.), and Kultala (lat. 78° 30' N., long. 27° E.), Lapland, in November and December, 1882. In the former place "the polar aurora appeared frequently of a very great intensity, but did not exhibit much variation. It generally began with a faint arc in the north, which shortly developed into an arc with rays and sometimes into draperies extending from the east to the west, most frequently a little toward the north. But little change of color took place; nearly always a pale-yellow
