ture is not real, but the effect of disturbance in our own atmosphere." And Mr. Hind states that though it frequently happens that "the minute shady dots" (called pores above) seem to be undergoing rapid changes, "it is most likely that this is caused by the variable conditions of the earth's atmosphere, and consequently is only apparent." Considering, indeed, that the motions in the visible regions of the sun, though so rapid when referred to terrestrial standards, take place on so great a scale that they are in fact majestically slow, when referred, that is, to the magnitude of the moving masses and of the spaces traversed by them (a circumstance which will of itself have a tendency to elude the visual perception of the motion), it may be doubted whether the arc described, or space passed over by any during the direction of the eye upon them, can ever be great enough, at the distance of the sun, to allow of the motion being actually seen. If the rapid fluctuations of the porous structure were real, they would immensely exceed in velocity all the motions and changes of the spots and other elements of the sun's surface, which, though perceptible from day to day or even from hour to hour, are insensible while under observation. Even the unique case of the sudden outburst of light simultaneously witnessed by Mr. Carrington and Mr. Hodgson (p. 32, note), which travelled 7,000 miles in a minute, was scarcely an exception to this. But the system of waves in Mr. De la Rue's photographs probably represents an actual solar phenomenon, being the momentary aspect of the undulating surface of the photosphere.

It happens, however, most remarkably, as a consequence of what we may call the accomplished vision, as well as sagacity of the veteran astronomer whose view of the photosphere has led to this discussion, that not only does his illustration derived from the subsidence of a flocculent precipitate in a transparent fluid faithfully represent the atmospheric fluctuations which in this instance seem to have been mistaken for solar changes, but it also represents with equal fidelity, though as it were on a smaller scale, the actual intestine motions of the photosphere, of which the facts we shall now state form an example.

M. Chacornac, a French observer, cited by Mr. Walker, describes in the following terms the motion of certain elements of the photosphere and penumbral region, apparently identical with, or associations of, the lenticular bodies of Nasmyth, as witnessed by him on one occasion. A sudden transformation took place of the luminous part of the photosphere into dark parts; luminous bridges were seen crossing the spots, and then gradually becoming dark: as these luminous bridges darkened, they at the same time plunged into the deeper parts, and became covered with other luminous bridges which formed above them. We have here the motions, as viewed, like the precipitate, perpendicularly from above, of a liquid containing distinct masses in their nature homogeneous with itself, heated from below, and cooling by radiation, at the same time, from its upper surface. The refrigerated portions, increasing in specific gravity, sink through the lighter and heated parts rising from below, by which they become covered. The visible action perhaps would be more precisely represented by congealed solid portions resulting from the cooling of the surface sinking down

C

through the lighter and hotter liquid, becoming liquefied by the heat below, and then rising to become solid again. Mr. Nasmyth's comparison with a shoal of fishes would also represent this circulation. To the system of currents thus produced we shall return in the sequel. A similar explanation of what is going on in the outer regions of the sun, has been given, but on purely theoretical grounds, by Professors W. Thomson and Tait.

The Rev. J. Challis, Plumian Professor of Astronomy in the University of Cambridge, and for many years Director of the Cambridge Observatory, in a paper communicated to the British Association in 1862, of which an abstract appears in the Report of that meeting, and which has been published at length in the Monthly Notices of the Royal Astronomical Society for June, 1863, has made a remarkable and important contribution to our knowledge of the sun, while discussing "the Indications by Phenomena of Atmospheres to the Sun, Moon, and Planets," employing the knowledge we have of our own atmosphere in inquiring whether the existence of other atmospheres of the same kind is indicated by phenomena. In the course of this investigation, he is led to infer a theory of the constitution and consequent refraction of the sun's atmosphere, explaining the cause of the unbroken circle which the contour of the disc presents, as already mentioned. In the present condition of our positive knowledge of the sun, as well as that of the considerations relative to its nature and the causes of its phenomena, which now occupy the minds of scientific men, whether astronomers, physicists, or chemists, it seems impossible to exaggerate the importance of this explanation, offered by its author in true philosophical simplicity.

The fact of the uniformity of the contour of the sun, is the more remarkable," Prof. Challis points out-tersely summing up the visible phenomena of the luminary, as characterised by astronomers"because the unevenness of the Sun's surface, as indicated by the mottled appearance spread over its whole extent, with the occasional occurrence of deep depressions (the spots), surrounded generally by lofty ridges (the faculæ), gives reason to expect that inequalities would be perceptible on the periphery." And in fact-reasoning from what would happen if the earth were seen from a distance, and from what actually does happen with the moon when observed in the telescope, whose inequalities are then distinctly observed on its edge or limb, breaking the regularity of its contour-it has often been reasonably objected to the conclusion that the spots are depressions or cavities in the sun, that, if they are truly such, when they arrive at the edge of the disc they ought to be seen in vertical section, if not as notches, at least as cavities, perceptible as such by differences of light and configuration; whereas the edge or limb of the sun is always-to use an expression common in descriptive natural history-entire,† a circular curve altogether free from inequalities.

Mr. Howlett has lately recorded two observations, the only two in his very considerable experience, in which portions of the penumbræ of spots extended absolutely to the sun's extreme margin, which is the phenomenon above anticipated from theory. Mention of one other instance will be found in p. 32, note.

The margin of a leaf, or of a shell, or other object, which has no serrature or teeth or other inequalities, but is undivided, is said to be entire.

How much this remarkable character of the sun involves, may in some degree be estimated from the intense interest which the apparent departure from it alluded to by Prof. Selwyn as above, and to which we shall return, excited in the astronomical world, and the pains which were taken to establish or to invalidate the evidence which it at first seemed to afford, and determine its true value. But why the visible sun is characteristically an unbroken circle, even in the telescope, and why, in fact, however great may be the inequalities of the luminous outer regions of the globe of the sun, they cannot, except in extreme and abnormal cases of rare occurrence, and, as it were, of special causation, impair the regularity of its visible contour-why, under ordinary circumstances, the depressions cannot be seen in section on the limb, we must now, after Prof. Challis, proceed to show.

The inference that the sun has an atmosphere, Prof. Challis considers to be justified by the phenomenon of a gradual diminution of brightness from the centre to the circumference plainly discernible in a photographic picture of the sun, or in an image received on a screen; the sun's atmosphere intercepting more of the rays coming from the parts near the periphery than of those coming from near the centre of the disc, because the former would pass through a greater thickness of atmosphere. Sir J. Herschel, long ago, had drawn the same inference from the same phenomenon; and had further considered decisive evidence of the existence of the solar atmosphere to be afforded by the production of the bright ring of light called the corona, concentric with the sun, seen when it is totally eclipsed; the interposition of the moon concealing the disc of the sun itself, the brilliancy of which, under ordinary circumstances, does not permit the light transmitted by and reflected from the atmosphere to be separately perceived. The astronomer Bessel had before regarded the corona as an indication of the same fact. The radiation of heat also from the central regions of the disc has been proved by Secchi, and that of the chemically active rays, recently, by Professor Roscoe, to be much greater than from its borders. The three elements of solar radiation, or distinct powers of the solar rays, light, heat, and chemical action, thus unite in the same testimony. At the present time, indeed the belief of astronomers in the reality of the sun's atmosphere is fully established; but it appeared desirable to place the evidence on which this belief is founded briefly before the reader, partly on account of its own interest, partly on that of the importance of the subject now entered upon; and also because it is noticed by Prof. Challis, as above, after finding, deductively, what would be the effect of an atmosphere on the distribution of light upon the disc of a distant luminous spherical body, which, he shows, would be identical with that actually exhibited by the sun.

We have now to consider a necessary difference of constitution between the Earth's atmosphere (which the atmospheres of the other Planets resemble) and that of the sun.

The most notable effect of the Earth's atmosphere with respect to visible phenomena is the refraction of the rays of light which pass through it, in consequence of which the apparent magnitude of the

Earth, as seen by a distant spectator, would be in some degree greater than that which it would have if it had no atmosphere, because the rays which pass from the periphery of the globe* itself to the position of the supposed spectator must have a curved course concave to the straight line joining that position and the Earth's centre. The amount of this augmentation of the disc of the Earth, supposing the spectator to be placed at the Moon, has been calculated by Prof. Challis to be, almost exactly, the very small one of 1"-one second of arc or of angular space-that is, 13200 of the apparent diameter of the Earth as seen from the Moon.

It follows from this that a mountain five miles high-an elevation which is exceeded by only four or five peaks in the Himalayan range, the highest in the world-situated at the Earth's apparent border, would subtend at the Moon an angle of 41" only, or equal to 233 of the apparent diameter of the earth as there seen, and its top would be somewhat less elevated by refraction, in proportion, than the base, on account of being higher up in the atmosphere. The atmospheric refraction, therefore, would only produce the effect of a slight apparent depression or diminution in altitude of the mountain; not augmenting the earth's diameter to a perceptible amount, nor materially altering the forms of the inequalities on the periphery when supposed to be seen by a telescope from a distant station in the solar system.

A ray of light from a star or other object seen from the earth passes through the atmosphere in a course which grazes, or is a tangent to, the interior globe itself. The same of course is true of the reverse case, of a ray from a point on the earth seen from the moon or other locality in space outside the atmosphere. It is evident that in such a case the curvature of the path of the ray is less than the curvature of the surface of the globe. "But it is conceivable," observes Professor Challis,—and in this conception of the effect of the refraction of the atmosphere of a heavenly body he is not aware that he has been anticipated by any previous investigator,"that there may be such a relation between the gradation of density of the atmospheric strata, and the curvature of the globe, that this condition cannot be fulfilled. For instance, it is reasonable to suppose that this is the case in the Sun's atmosphere, when the vast magnitude of the globe is considered," and the mathematical result is taken into account, which shows that the decrement of density of the atmosphere corresponding to a given increment of height above the globe varies conjointly as the density and the Sun's gravity.

"It seems necessary to suppose," continues Prof. Challis, "that an atmosphere has an upper boundary like that of an ocean, because, the density continually decreasing with the height, a point must at length be reached at which the upward repulsive force of an atmospheric stratum is just equal to the force of gravity, in which case there can be no downward repulsive force, and therefore no further extension of the atmosphere." In this conception of the necessary existence of an upper boundary to an atmosphere, it may * Throughout this explanation of the unbroken circularity of the edge of the sun's disc the term globe is used in reference to a planet, or other heavenly body, as distinguished from the atmosphere surrounding it, or within which it is contained.

66

be remarked, Prof. Challis agrees with the late Dr. Wollaston, and also with Sir J. Herschel, who, in adopting the inferences of the former relative to the finite extent of the Earth's atmosphere, says: Arguments are not wanting to render it, if not absolutely certain, at least in the highest degree probable, that the surface of the aërial, like that of the aqueous ocean, has a real and definite limit, beyond which there is positively no air.” *

*This conclusion is opposed to the idea which was formerly prevalent that the earth's atmosphere had in fact no termination-an idea which is still retained by certain physicists. Even those who, by comparing the calculable density of the air at certain assumed heights with that to which it would be reduced in the receiver of an air-pump, have ascribed an extent of 40 or 50 miles to the atmosphere, and others who have implicitly accepted that inference have maintained the same belief. They have supposed that the atmosphere becomes indefinitely rare beyond a certain but unknown distance from the earth, ceasing to affect the rays of light, having neither refractive nor reflective power, graduating or evanescing into nothing it may be said-into that the existence of which it would not only be impossible to recognise by any means we possess were it on the surface of the earth, but is really not conceivable by the mind-a condition of matter (!) devoid of properties, but still supposed to extend through spaceand from this attenuated condition of virtual nonentity to become gradually condensed again into the tangible and ponderable condition of the air as we know it,-around other bodies of the solar system or of the spaces beyond. Accepting, on the contrary, as proved, the doctrine that atmospheres generally have definite boundaries, at which, mathematically speaking, their densities, though small, have finite values, it will be as near a representation of the truth as the present condition of science enables us to arrive at, to say, that as the surface or upper boundary of the aqueous ocean of the earth is succeeded by the aërial ocean or atmosphere, so the upper boundary of the latter, "beyond which there is no air," is succeeded by the luminiferous ether, which extends through and fills the interplanetary spaces, surrounding all the heavenly bodies of the solar system, with their respective aërial atmospheres; and indeed, as the visibility of the stars evinces, extending, with certain modifications, throughout all known space. But while the aërial atmosphere of the earth rests upon the aqueous nearly as it does upon the land, the ether does not rest upon, but pervades the former, together indeed with the waters and the solid earth, the ether thus resting upon itself alone; independently of the ponderable matter, whether solid, liquid, or aëriform, constituting the earth and other planets, and at least the outer regions of the sun also, which it permeates.

Though the ether, when in equilibrium, thus rests upon itself alone, if we identify it with the resisting medium in the planetary spaces, which Professor Encke, the Director of the Observatory of Berlin, has shown to retard the motion of the comet bearing his name, we must ascribe to it a sort of friction with the molecules of certain forms of ponderable matter, of great tenuity, of which comets consist.

The subject of the termination of the atmosphere above, necessarily involves that of the medium which succeeds it, occupying the space beyond, and hence the consideration of the ether is inseparable from that of the nature of atmospheres in general, as known or believed to encompass the earth and other bodies of the solar system. On this account it is here noticed; but it has no further immediate bearing on the subject of this paper.

It has not yet been recorded in the history of science, though occasionally mentioned by the writer in lectures on the atmosphere, that Wollaston had been anticipated by Henry Thomas Colebrooke, F.R.S., sometime President of the Asiatic Society of Bengal, and an eminent contributor to its "Transactions,' called "the Asiatic Researches" He was one of the ornaments of our Indian Empire, whom that empire itself in great measure contributed to form, and by whom as its reward it was maintained and extended. The remarkable paper which contains his theory of the atmosphere will be found in the Quarterly Journal of Science' (vol. ix., p. 57, 58) for April, 1820. Dr. Wollaston's paper was read before the Royal Society, January 17th, 1822, and published in the Philosophical Transactions' for that year. But the two philosophers were led independently to the same important conclusion by distinct roads. Similar views had been vaguely entertained before; but, until the modern limitation of the theory of the infinite divisibility of ponderable matter by the discovery that its elements would unite only in fixed and definite proportions, and the conception of an imponderable ether at once pervading and extending beyond the aërial atmosphere consisting of such elements had been expressed, the finite extent and termination of the latter, it may fairly be said, even if they could be definitely conceived of, could not be intelligibly described, nor securely maintained. In the use of the last phrase the objections urged by Dr. Whewell against the conclusions of Dr. Wollaston are not ignored; even he admits the height of the atmosphere to be finite, though he denies that this involves the consideration of atoms, supposed to be inseparable from the truth that bodies combine only in definite proportions.