February, 1870-but only recently published, the author has shewn that the point of no disturbance of the water on the surface of a sphere passing through it is at an angle of 49° 59', nearly, with the direction of motion-a result agreeing so nearly with the angle 49° 30′ observed by Mr. Berthon, in his Cylindrical Tube, as to induce great confidence in the practical application of the author's Theory of Stream-lines.

JAS. R. NAPIER.

VII.-On the Results of the Recent Solar Eclipse. By DR. Tнorpe.

Read before the Society, 6th March, 1872.

ON the ne sutor ultra crepidam principle I feel, in some degree, that I ought to tender an apology for my appearance here to-night. As one of the laity, and with Professor Grant in the chair, I deem that it is in no small degree presumptuous in me to attempt to address you on a subject apparently so far removed from my true vocation. But we chemists are terrible Ishmaelites: our motto is ubique. We are continually invading somebody else's preserves; and the last piece of territory we have filched, or rather, let us say, annexedthat's the proper word-is a good slice out of the province of the astronomers. For, I respectfully submit to our President, that he who occupies himself in searching for the vapours of iron and sodium in the atmosphere of the sun is rather more of a chemist than he is of an astronomer. I confess, gentlemen, that the subject of solar eclipse has an especial interest for me; and my coming before you to-night arises from a desire I felt to put together the broad results obtained during the last eclipse, and to compare them with the state of our previous knowledge, in order that we might determine how far a real advance had been made. This I take to be a subject very proper for the consideration of this Society; for the questions which it opens up are of immense importance to us, as bearing directly upon the constitution of our cosmical system.

I doubt not that it will be within the memory of everybody here that on the 12th of last December there occurred a total eclipse of the sun, visible in the southern part of India, and over a large section of Australasia. The great importance of this event, in its relation to

our knowledge of the constitution of the sun, has led to a very careful observance of it; and I purpose to-night to bring under your notice some of the chief points established by the various expeditions which have been fortunate in witnessing the phenomenon. In the outset, I may say that this eclipse has been viewed under exceptionally favourable conditions; but I regret that the time which has subsequently elapsed has not been sufficient for us to come to a definite conclusion respecting many of the vexed questions which it was hoped that the present eclipse would settle. It is true that we are in possession of many capital facts; but the greater number of the reports have yet to be sent in. Still, I think I need offer no apology for bringing the subject of this eclipse under the notice of the Society at this somewhat premature stage. The remarks which I offer may at least serve to direct the attention of some here to the more complete publication of the results which will doubtless occur in our recess.

Let me, in the first place, state the general nature of the phenomena seen during a total eclipse of the sun, in order that we may gain some insight into the character of the problems which it presents. Most of you will be aware that when the sun is eclipsed the dark disc is, among other things, seen to be surrounded by a luminous halo, which we term the corona. Now, it is touching this corona that I wish principally to speak this evening: it is the centre of interest with us at present, and has been the main object of attack during the eclipses of the last few years. You must therefore pardon me if I appear to dwell to an inordinate length upon this particular feature.

Of the chromosphere I shall say little or nothing. That is a portion of the subject which is peculiarly our President's; for, I need not remind the Society that we owe to the critical labours of Professor Grant the discovery of that portion of the luminous envelope which we to-day term the chromosphere.

The corona was noticed so far back as the year 96 by Philostratus ; but Kepler has the credit of having first drawn the special attention of astronomers to its existence, and he threw out the idea that it was due to a lunar atmosphere. Kepler conjectured that the solar rays in passing to the earth were refracted in penetrating this hypothetical atmosphere, occasioning the appearance of the luminous ring. This idea was shared by many of the earlier astronomers— for instance, by Halley, one of the most famous of eclipse observers. But I need scarcely tell you that the supposition of a lunar atmosphere of appreciable density is rendered doubtful by a multitude of observaVOL. VIII.-No. 1.

M

tions. I say of appreciable density, for the mathematical reasoning of Clausius has rendered it very probable that our own atmosphere extends even to our satellite, but exists there in a state of extreme tenuity-less, indeed, than 1-700th of the density which it possesses at the surface of the earth. Halley, however, although he seemed to favour it, never definitely committed himself to the explanation of Kepler: it seemed to him incompatible with the fact which he and others before him had detected, that the luminous ring was not concentric with the disc during the progress of the eclipse. This idea of something at the moon,-atmosphere, emanation, or call it what you like-occasioning this appearance seems to have retained its hold on the mind of observers for a long time, although the results of each successive eclipse render the probabilities of its existence more and more remote. From the apparent extension of the luminous ring, as seen during the total eclipse of 1806, it has been calculated that the lunar atmosphere, supposing that it existed and produced this effect, would be even more dense than our own. This reductio ad absurdum would seem at once to clinch the difficulty; but, unfortunately, as we shall shew, the premises, although they lead to a result consistent with our present information, were not such as to form the basis of a definite calculation. Driven to regard the hypothesis of a lunar atmosphere as untenable, other observers have sought to connect the moon with the phenomenon upon another principle. Delisle, for example, considered that the corona might be due to the diffraction of the solar rays in passing near the moon's edge, and he based his supposition on experimental evidence of no mean weight. This idea was more fully worked out by the late Professor Powell; but although the circumstances of these experiments, considered in the abstract, bear some resemblance to those which occasion the manifestation of the luminous ring, the theory (as Sir David Brewster has proved) is incapable of explaining many of the main features of the appearance of the corona. On no hypothesis, therefore, can the moon be rightly considered as the direct cause of the luminous appearance round the eclipsed sun: the evidence we possess to-day seems perfectly decisive of this point. Having thus disposed of the moon, I hope to your complete satisfaction, and whilst I am upon the question of the early notions of astronomers respecting the physical cause of the luminous ring, I may mention that it has been referred by some to the zodiacal light; and these persons have fancied that they have traced in the configuration of the corona an elongation in the direction of the ecliptic exactly corresponding to the appearance of the zodiacal light. But a more intimate knowledge of the changes of

this configuration will serve to dispel this idea completely. But I would whisper, lest it disturb the ghosts of those who formerly believed in their relation that the spectroscope has seemingly furnished a grand argument for the connection of corona and zodiacal light, in the fact that the famous green line in the spectrum of the corona (about which more anon) is said by some to be identical with that which Angström has detected in the zodiacal light. I may also state in this connection (you taking it for what it is worth) that Professor Adams and his party saw a very marked display of the zodiacal light in Sicily on the day before the eclipse of December, 1870.

People are not wanting who imagine that the phenomenon of the luminous ring is of purely electrical origin; and Professor Osborne Reynolds exhibited at the Edinburgh meeting of the British Association an experiment designed to illustrate this supposition. He finds that when a ball is electrified in a rarified atmosphere, under a certain relation of electrical tension and atmospheric rarifaction, the ball is surrounded by a luminosity which recals in a very striking manner the general appearance of the corona; but it remains to be seen how far this idea can be brought to harmonize with the revelations of the spectroscope.

But it is not necessary to wait for an absolutely total eclipse to see this corona: this fact, formerly not generally credited, although the observation had been made by perfectly trustworthy observers, has now been incontestibly established. Thus, in the famous eclipse of 1842, Petit, at Montpellier, noticed the ring about six seconds before the sun had totally disappeared behind the dark body of the moon; and his observation was confirmed by Mr. Baily in Italy: nay, the luminous ring has been seen from places outside the limits of the shadow, and where consequently the eclipse was throughout its entire course only partial. Mrs. Airy, who observed the same eclipse, has given us a circumstantial account of the fact. "A bright line," she says, "seemed to form round the right side of the moon before the disappearance, but not quite round, so that the ring was not complete; but at the moment of total disappearance the ends seemed suddenly to join and form the complete ring." To come down to more recent times, M. Rumker, in 1860, observed the corona one and a-half minutes before totality; and Mr. Brett and Professor Watson, in Sicily, in 1870, saw it when a large portion of the sun still remained unobscured. Mr. Lockyer also made the same observation during the last eclipse: he says that for nearly three minutes after totality did the coronal structure impress itself on his retina,

until at last it faded away in the rapidly increasing sunlight. These observations are of great significance in the argument for solar origin; they lend probability to the conjecture that the corona may yet be traced, perhaps, by some process analogous to that by which Lockyer and Jannsen have picked up the chromosphere, by diluting the glare which masks it.

But now, to come to particulars about this corona. Let us inquire more minutely into it. What does it look like? What is its shape? What is its colour? Is it the same thing throughout? And lastly, What is it made up of?

First, then, as to its appearance. This is such that scarcely two observers, although placed side by side, can agree about it. The corona is the very Proteus of celestial phenomena. This fact came out very strikingly in the famous eclipse of 1851, visible in Sweden. Here are some representations of the corona made at that time. First, we have a drawing by Mr. Carrington, in which you see the corona figured as a modest little circle of light, with nothing particularly striking in its appearance; here is a drawing made during the same eclipse by Mr. Airy, presenting a totally different effect from the other: you will notice the " compass-card" ornament surrounding the narrow circle. Here, too, are representations made by the Americans during the eclipse of 1869: you will see what a gorgeous sight Mr. Gilman must have enjoyed: a pyrotechnical display exceeding any 4th of July demonstration. Others of the party were not so fortunate: the corona appeared to Professor Eastman as a far tamer affair. Here is another drawing made also by Professor Eastman in Sicily, in 1870: compare it with this made at the same time and place by Captain Tupman.

I would direct your attention to these outlying portions of the corona or "streamers," as they are called. These are only seen after totality; and they change from side to side, and dance about almost as energetically as the streamers of the Northern Light. It has long been noticed that the corona is of greatest size and brilliancy on the side where the sun has just been covered; but as the eclipse progresses, the brightness and size of this part gradually diminish, and a corresponding increase is seen on the opposite side, where the sun is about to emerge. Maraldi, at the eclipse of 1724, stated that at the commencement of the total immersion the ring appeared broader on the east than on the west; but that at the end of the total immersion it appeared broader on the west side. The same fact was also noticed independently by Rumker and Plantamour in 1860; and in three drawings, made in Spain in 1860, by Mr.