tablishing this very interesting result, was led to express the opinion that the sun's atmosphere has no such extent as had been imagined, and that the corona is an appearance (only) in our own air, "an atmospheric effect merely," "due to the passage of the sun's rays through our own atmosphere."
This conclusion was, however, not very generally accepted. Several astronomers at once pointed out that the air which lies toward the place on the heavens where the corona is seen, is not illuminated at all by the sun's rays during total eclipse. I also pointed out that whatever light that particular part of the air receives during totality—not direct sunlight, but light from the prominences, and so much of the corona as might be admitted to be solar—would extend over the very place of the moon, and gradually increase thence on all sides instead of gradually diminishing, as happens with the corona. This would not be the place to exhibit the reasoning by which these results can be demonstrated; for mathematical considerations, not altogether simple, are involved in the complete discussion of the matter. Let it suffice to say, as respects the air between the observer and the moon, that, since the observer can see the colored prominences and the inner bright corona during totality, the air all around him (toward the moon as well as elsewhere) must be lit up by their light. And as respects the gradual increase of brightness on all sides of the place where the eclipsed sun is, let the reader consider that, if, at any time during totality, a bird were to fly (with enormous rapidity) from the observer's station directly toward the moon's centre, that bird would remain in the moon's shadow as he so flew; but if he flew in any other direction he would presently pass out of the shadow—that is, he would reach a place where the air is illuminated. And he would so much the more quickly reach the illuminated air, as he flew more directly from the moon's place on the sky. So that, putting the line of the observer's sight instead of the swiftly-flying bird, we see that this line will so much the sooner reach illuminated air, according as it is turned farther from the place of the moon on the heavens. Thus the air toward the place of the moon, though illuminated, is less brightly illuminated than that lying toward any other part of the sky; and the atmospheric illumination must gradually increase the farther we turn our eyes from the moon's place.
So matters stood when preparations were being made for the expeditions to view the eclipse of 1870. Evidence had, indeed, been obtained during the eclipse of 1869 in America, which seemed to show that the substance of the corona is gaseous; and singularly enough it appeared as though this substance, whatever it might be, shone with a light resembling that of the aurora borealis. But those who regarded the corona as a mere glare in our own atmosphere, rejected these results because they seemed "bizarre and perplexing in the extreme." The American astronomers, however, were not willing to have their
