This theory accounts for nearly all the points on which there is now any uncertainty. There only remains the spectroscopic side of the subject, as intimated, to account for all the lines in the spectrum of the aurora. The yellowish-green line (the fourth in the spectrum) to which we have referred is only found in the "zodiacal light" which travels along the zodiac at certain times of our year. But while there is otherwise no identity between the two forms of light, this would indicate an intrastellar or supermundane source of both lights. So it is very probable that cosmic causes, foreign to our globe, may determine on our globe the production of the aurora. We have dwelt in other papers on the influence of the sun spots, so it is not necessary to revert to them, except to confirm our previous theory and impressions. These phenomena all appear to obey the successive phases of solar activity, and the electric theory of the aurora fully accounts for their relation to each other. If sun spots can cause a widespread action of the aurora in polar regions, why can they not induce magnetic disturbances, electric storms, and all the fury of tempests in temperate regions, acting along exactly the same lines of current, only deflected sooner to the earth than the rays of force which reach the earth nearer the poles? It is a very simple question, and the answer self-evident.
Laboratory study of electric discharges in rarefied air and gases has not yet reached a point where the spectrum of the aurora borealis can be artificially produced. A study of the zodiacal light, and of the influence of magnetism on light, will be necessary to reveal this factor of the problem. This opens an immense field of further research in which scientists must still delve.
