which A B C D represents the portion of the Moon's shadow in the atmosphere, E the place of an observer towards whom this shadow is swiftly advancing. Then, until the shadow actually reaches E (at which moment totality begins) a line drawn from E to D will pass to the left of the line A D m-as in the direction E Df; so that, since the line A D m is necessarily directed towards the Moon's limb, the directly illuminated air (bounded, of course, by a n) extends between the observer and the Moon. Hence, the Moon's disc seems lighted up by this atmospheric glare until the

FIG. 84.

H

712

D

H'

Illustrating the condition of the Earth's atmosphere a few minutes before totality in a solar eclipse.

very moment when totality begins; and as the curtain of glare is drawn suddenly away towards the edge where the last sickle of the Sun's direct light is vanishing, all the circumstances tend to give an appearance of agile motion in that direction.

As respects the rotatory motion which seemed to accompany this leap forward on the Moon's part, we can very well understand it as referring to the completion of the corona, which must needs take place by a sweeping round of the bounding rays to close in upon each other opposite the point where the last part of the

Sun's disc disappears. It will be well to notice, however, as we proceed, whether we have convincing evidence of an apparent motion of the coronal beams after totality has begun. The evidence on this point cannot but have an important bearing on the views we are to form respecting the corona.

Dr. Wyberd saw a corona of very limited extentindeed only half a digit wide-corresponding to a height not half so great as that of many prominences which have been observed during recent eclipses.

Our next observation refers, however, to a much more favourable view of the corona.

In May 1706 Plantade and Capiés saw a bright ring of white light surrounding the eclipsed Sun, and extending to a distance equal to about one-tenth part of the Moon's apparent diameter. Outside this bright ring a fainter light could be recognised, which extended no less than four degrees from the eclipsed Sun, fading off insensibly, until its light was lost in the obscure background of the sky.

In 1724 Maraldi noted a circumstance of some importance. At the beginning of the total eclipse which was observed in France in that year he perceived that the corona was wider on the side towards which the Moon was advancing than on the opposite side. At the close of totality the widest part of the corona was on the opposite side. As this would exactly correspond to what would be observed if the corona lies beyond the Moon, and so is traversed by the Moon precisely as the Sun himself is, it will be well for us to

notice as we proceed whether Maraldi's observation has been confirmed or disproved, or how the evidence stands with respect to it.

In 1733 a total eclipse occurred in Sweden, which was observed in a manner reflecting great credit on the astronomers of that country. The Royal Society of Sweden invited all who could spare the time to assist, as far as their ability permitted, in recording the phenomena presented during total obscuration. Accordingly there are few eclipses - perhaps there is not one-to the phenomena of which so many independent witnesses give testimony. At Catherinesholm the pastor of Forshem noticed that a ring of light which appeared round the disc of the Moon was of a reddish colour-an observation confirmed by Vallerius, another pastor, who noticed, however, that at a considerable distance from the Sun the ring appeared of a greenish hue. The pastor of Smoland states that "during the total obscuration the edge of the Moon's disc resembled gilded brass, and that the faint ring around it emitted rays in an upward as well as in a downward direction, similar to those seen beneath the Sun when a shower of rain is impending." The mathematical lecturer in the Academy of Charlesstadt, M. Edstrom, observed these rays with special attention, and remarks, respecting them, that "they plainly maintained the same position, until they vanished along with the ring upon the reappearance of the Sun." On the other hand, the ring, as seen at Lincopia, seemed to have no rays. Professor Grant remarks that "from the descriptions

given by several observers, it would seem that at the commencement of the total obscuration the ring appeared brighter and broader at the part of the Moon's limb where the Sun had disappeared, but that towards the close of the obscuration it was more conspicuous in both these respects at the part where the Sun was about to emerge.

It is observable, therefore, of this well-watched eclipse that it confirms Maraldi's observation, while the variety of appearance presented by the corona at different stations would point to the conclusion that if the corona is not a phenomenon of our own atmosphere, its light must for the most part be of a very delicate nature, insomuch that seemingly unimportant differences in the circumstances under which it is viewed suffice to modify its aspect to a very noteworthy extent. We shall see presently that during recent eclipses similar evidence has been afforded of the extreme faintness of a large portion of the coronal light.

During the eclipse of 1766 the corona exhibited four remarkable expansions, separated from each other by nearly equal intervals.

In the account given by Don Antonio d'Ulloa of the appearance presented by the total eclipse of 1778, we again find a reference to the appearance of rotatory motion in the corona. Five or six seconds after the commencement of the total obscuration,' he writes, a brilliant luminous circle was seen surrounding the Moon, which became vivid as the centre of

* From an article in the Cornhill for October 1870 (by the present writer).

that body continued to approach the centre of the Sun. About the middle of the eclipse its breadth was equal to one-sixth of the Moon's diameter. There appeared issuing from it a great number of rays of unequal length, which could be discerned to a distance equal to the lunar diameter. It seemed to be endued with a rapid rotatory motion, which caused it to resemble a firework turning around its centre. The colour of the light was not uniform throughout the whole breadth of the ring. Towards the margin of the lunar disc it appeared of a reddish hue; then it changed to a pale yellow, and from the middle to the outer border the yellow gradually became fainter, until at length it seemed almost quite white.'

The next eclipse during which new features of importance were noticed, was that of 1842. We have seen how carefully the prominences were observed during that eclipse, and how many eminent astronomers were engaged in noting the phenomena presented during the total obscuration of the Sun. The corona was also carefully, studied. Indeed, one is inclined almost to regard it as a misfortune that the same astronomers who tell us about the appearance of the prominences, are those from whom we derive our information respecting the corona. One cannot but feel that the accurate observation of both phenomena was more than could be expected even from the most skilful astronomers, and that a division of labour might have been advisable.

The apparent motion of the corona was noticed by