nous because it is colder. Again, M. Faye holds that a spot is due to an uprush the Kew observers, that it is due to a downrush. At the outset there were many arguments against M. Faye's hypothesis. The law of exchanges was utterly against his idea of the darkness of a spot, for if it were the interior of the sun which we saw, and its radiation were feeble, then its absorption would have been equally feeble and the sun would be spotless; for where the photosphere was torn away on the side nearest us, we should be able to see, through the sun, the lower surface of the photosphere on the opposite side. Again, the arguments in favour of an uprush, in the case both of spots and faculæ, are not very clear, nor have we a satisfactory explanation of the falling behind of the faculæ. But we had not long to wait for facts which, as far as we can see, have entirely settled the question. First, as to the downrush into a spot. In 1865 two observersone in France, the other in Englandcarefully observed the fine spots from time to time visible on the sun's disc in that year; and the observations of both tend to show the absolute certainty that if spots are not caused by downrushes, they are, at all events, fed by them. Let us hear the French observer first:1 "La rapidité des changements est telle, que l'on peut suivre dans une même "journée des courants des matières pho"tosphériques se précipitant dans le The spot had a tongue of facula stretching half-way over it. When the observation commenced at 11:30 on April 2, this tongue of facula was extremely brilliant; by 1 o'clock it had become less brilliant than any portion of the penumbra: at the same time the faculous mass seemed to be giving out its end, veiling the umbra gradually with a kind of stratus cloud evolved out of it, which after a time again condensed into masses resembling the willow-leaves in the penumbra, only less distinct. The argument for the downrush is to be found in the fact of the diminution of brightness; accepting as proved, first, that the faculæ are higher than the general surface, and, secondly, that a spot is a cavity. But it does not wholly depend upon this, for the masses or granulations on the general surface of the sun appear to lengthen out when they reach the penumbral region, as if they were acted upon by a current, and this may also explain the constantly observed difference in the shape of the F "Willow-leaves" detaching themselves from the penumbra. A very faint one at F. cloud masses on the general surface and in the penumbra. In this connexion it is worthy of remark, that when a solitary willow-leaf is seen over the centre of a spot, it is often observed to be nearly circular, as if its longer axis were tipped down. It is fair to add, however, that observations of the requisite delicacy can be very rarely made, owing to the many coincident conditions necessary. The fact that a spot is due to absorption has next to be considered. On M. Faye's theory, as it will doubtless have already suggested itself to the reader, could a sun-spot be observed by means of a spectroscope; as, by hypothesis, we have radiation from a gas in a state of dissociation, the resulting spectrum would be a gaseous one- -that is, it would consist of bright lines. We, in fact, should get from a spot a spectrum absolutely different from that which belongs to the light emitted from the general surface, the latter being a band of rich colour going from red through yellow, green, blue, indigo, to the intensest lavender, crossed by innumerable black lines of different intensity, the former consisting only of three or four thin bands of light, located in the green portion of the spectrum. On the absorption-hypothesis there would be none of these bright lines; we should get a spectrum in the particular region of the spot similar to the average solar one, but showing evidence of greater absorption. This was put to the test in 1866.1 The method adopted was to apply a direct-vision spectroscope to a 64-inch equatorial, so that it was possible to observe at one time the spectra of the umbra of a spot and of the adjoining photosphere or penumbra. On turning the telescope and spectrum-apparatus, driven by clock-work, on to the sun, the solar spectrum was observed in the field of view of the spectroscope with its central portion (corresponding to the diameter of the umbra falling on the slit) greatly enfeebled in brilliancy. All the absorption-bands visible in the spectrum of the photosphere, above and below, were visible in the spectrum of the spot; but they appeared thicker where they crossed the spot spectrum. There was not the slightest indication of any bright bands. The dispersive power of the spectroscope employed was not sufficient to enable it to be determined whether the decreased brilliancy of the spot-spectrum was due in any measure to a greater number of bands of absorption. 1 "Spectroscopic Observations of the Sun," by J. Norman Lockyer (Proceedings of the Royal Society, vol. xv. p. 256). The Royal Society at once recognised the importance of this discovery, although it was put forward with much hesitation, as the instrument employed was not of sufficient dispersive power, and the spot itself was not a very favourable one for the experiment. A larger instrument has now been constructed, and detailed observations are now about to be commenced under the auspices of that body. In the meantime, however, this settlement of the long-debated question has recently been entirely endorsed by Mr. Huggins, whose discovery of the physical constitution of nebulæ, and spectroscopic observations of the fixed stars, make his opinion of the greatest possible weight. We have thus, as briefly as possible, traced up our knowledge of the sun's surface from the times of Galileo to our own. That surface, we have learnt, is of a cloudy nature, the light and heat being derived from the solid incandescent particles of which the clouds are composed. Further, there are exchanges perpetually going on between the cooler exterior and the interior. The descending current is accompanied by a spot, the ascending one by a facula; and finally, the dark appearance of a spot, like the darkening of the limb, is due to the absorptive properties of the sun's atmosphere. Let us, for one moment, compare the sun's envelope with our own, and observe the action of the latter when the sun is withdrawn. The general surface of the ground is a good radiator. On the other hand, the atmosphere is at once a feeble absorbent and a feeble radiator. When the sun's influence is withdrawn from the earth's surface, and the sky is clear, the general surface of the ground and the leaves of plants give off their heat, which is radiated into space unimpeded by the very feeble absorbing power of the air; on the other hand, the air, being a feeble radiator, gives back little or nothing in return. As far as radiation is concerned, there fore, the ground and leaves get rapidly cooler, nor is this loss of heat made up by any other process. Little or no heat can reach the cooled surface by conduction, for ground, leaves, and air are bad conductors. Further, convection does not operate, for the particles of air next the cooled surface becoming cooler themselves become also heavier, and remain where they are. There is, therefore, no hindrance to the cooling of the earth's surface, which in its turn cools the air in contact with it until the air has reached so low a temperature that it cannot longer retain all its vapour. Part of the vapour is, therefore, deposited as moisture (or hoar frost if the temperature be below freezing-point), on the surface of the ground and the leaves of plants; and this is the explanation of dew and hoar frost, which we get when there is free exposure to the open sky. If there be cloud, a glass frame, matting, or any obstacle in the shape of a good radiator, interposed between the body and the sky, there will be no deposition of dew, because a quantity of heat will be derived from the radiator which has been interposed. Heat will therefore be lost very slowly, and moisture will not be deposited. It must be borne in mind that the presence of cloud makes an essential difference. We may suppose something equivalent to the deposition of dew, or, at all events, great radiation, to be taking place on the upper surface of the cloud, not under the cloud. The heat of the bodies is retained in the latter region, the radiation being diminished, or rather compensated, by counterradiation. It may be instructive to place ourselves in imagination above the surface of such a cloud, the sun being withdrawn, and consider for a moment what probably takes place. The small deposited particles, being great radiators, will rapidly get colder than the surrounding air; they will, at the same time, cool the air around them; and the air, being cooled, and thereby rendered heavier, will descend. There will thus be descending currents of air. But descending convection currents are naturally accompanied with ascending ones. There will therefore be ascending currents, conveying upwards some of the comparatively warm air from below. It is not impossible that such currents may assume in nature somewhat large dimensions, and that the cloud may therefore present to a beholder regarding it from a great distance above, an irregular, pitted, notched shape; in fact, exactly such an appearance as we see on the sun, the envelope of which, parvis componere magna, may resemble in its mechanism that of a planet like our own with its sun withdrawn. So far we have only referred to the phenomenon ordinarily visible to us. Another part of the sun's physical constitution is rendered visible during total eclipses. We allude to the nature of its atmosphere. Eclipse-teachings, therefore, are of high value; but they certainly are not of such high value as ordinary observations of its surface, although they are in their nature much more sensational, for a total eclipse of the sun is at once one of the grandest and most awe-inspiring sights it is possible for man to witness. All nature conspires to make it strange and unearthly. The sky grows of a dusky livid, or purple, or yellowish crimson colour, which gradually darkens, and the colour appears to run over large portions of the sky, irrespective of the clouds; the sea turns lurid red; the moon's shadow sweeps across the surface of the earth, and is even seen in the air; all sense of distance is lost; the faces of men assume a livid hue; fowls hasten to roost; flowers close; cocks crow; nor does the animal world escape the general excitement. Soon the stars burst out; and surrounding the dark moon on all sides is seen a glorious halo, generally of a silverwhite light this is called the corona. It is slightly radiated in structure, and extends sometimes beyond the moon to a distance equal to her diameter. Besides this, rays of light, called aigrettes, diverge from the moon's edge, and appear to be shining through the light of the corona. In some eclipses parts of the corona have reached to a much greater distance from the moon's edge than in others. It is supposed that the corona is the sun's atmosphere, which is not seen when the sun itself is visible, owing to the overpowering light of the latter. When the totality has commenced, apparently close to the edge of the moon, and therefore within the corona, are observed fantastically-shaped masses, full lake-red, fading into rose-pink, variously called red-flames and red-prominences. Two of the most remarkable of these hitherto noticed were observed in the eclipse of 1851. It has been definitely established by the exquisite eclipse photographs of De la Rue and Secchi, that these prominences belong to the sun, as those at first visible on the eastern side are gradually obscured by the moon, while those on the western are becoming more visible, owing to the moon's motion from west to east over the sun. The height of some of them above the sun's surface is upwards of 70,000 miles. It is not yet known what these strange red prominences are; but while we write astronomers are trooping to India to settle the question at the coming total eclipse. England, France, Prussia, and other European states will be represented, while a happy evidence of the sooner or later prevalence of truth the successor of Galileo's persecutor, will be represented by by one of the most accomplished astronomers of modern times-Father Secchi, a Jesuit, who, we trust, will be among the foremost to crown the edifice of which Galileo laid the foundation-stone; for, in fact, a knowledge of the nature of the red prominences seems now to be the only thing wanting to complete a sketch of the visible solar phenomena apart from their causes. Of course there is much detailed drawing to be added afterwards. But, even at present, we are in a position to imagine what the real nature of the prominences may be. In the first place, a diligent spectroscope sweeping round the edge of the sun has not revealed any bright lines. This is strong negative evidence that they are not masses of incandescent vapour or gas; for as the light from such vapour or gas is almost monochromatic, it should be as easy to detect as that of the immeasurably distant nebulæ. Secondly, we know that the atmosphere of the sun is colder than the photosphere, and that in the latter we have incandescent particles of solid matter. As the prominences are possibly not due to incandescent vapour, the question remains whether they may be attributable to sub-incandescent particles of solid matter at a red glowing heat only, suspended in the atmosphere. In fact, whether the particles in the photosphere itself may not be likened to a white-hot poker, and those in the atmosphere to merely a red-hot one. It has In the previous part of this article attention has been directed solely to the immediate cause of a sun-spot; and an attempt has been made to show that a downrush of comparatively cold atmosphere from above, accompanied with an uprush of warm atmosphere from below, is the only sufficient explanation of the phenomena observed. also been shown, as the result of a careful scrutiny of the whole surface of the sun, that there are probably convection currents in constant operation all over the disc-a condition of things which we might expect from the intensely hot state of the sun's surface combined with the enormous gravity of matter there placed. A sun-spot may thus not improbably be regarded as an enormous development under exceptional circumstances of what is constantly occurring all over the sun's surface. This remark brings us a step further in our inquiry by suggesting the question, What are the exceptional circumstances that cause the ordinary convection currents of the sun's surface to develop themselves occasionally into sun-spots? This inquiry may be rendered more general by dismissing from the mind all idea of the nature of sun-spots: it is not essential to know what they are, whether convection currents or something else. The question now is not what is their nature, but what is their cause, or rather, in the present state of our ignorance, are they connected with any other phenomena that may serve to throw light upon their cause? This inquiry divides itself into the four following heads : 1. Does the amount of spotted surface of the sun vary from time to time? 2. Is the region of outbreak of a spot confined to any particular part of the sun's disc? 3. When a spot is formed, does it obey any laws with regard to increase and diminution? 4. And finally, are spots connected with any other phenomena on the earth's surface or elsewhere? The remainder of the article will consist of an attempt to answer these four questions. Now, in the first place, as has been already noticed, the amount of spotted surface has a ten-yearly period. This has been discovered through the labours of the veteran astronomer Hofrath Schwabe, of Dessau, who has now for about forty years been engaged without intermission in registering the number of spots which appear on the sun's surface. Herr Schwabe has found as the result of his labours, that in the year 1828 there were 225 groups, against 161 groups in 1827, and 199 in 1829; the year 1828 was therefore a year of maximum. After this the number of groups gradually decreased until in 1833 there were only 33 new groups observed. After this year they began again to increase, and in 1837 they attained another maximum. The next year of maximum was 1848, and the next after it 1859. We may therefore expect another in the course of a few years; indeed at the present moment the number of spots is increasing. We are still ignorant of the ultimate cause of this periodicity, but independent observations by the Kew Observers,1 and by Hofrath Schwabe, lead to the impression that in years of minimum there is a less amount of cold-absorbing atmosphere above the photosphere and consequently a smaller tendency to the downrush of cold matter in large quantity. The observations above referred to seem to indicate for years of minimum a more uniform brightness of the sun's surface,-that is to say, a less amount of absorption or falling off towards the limb, a phenomenon which, it has been already shown, depends upon the amount of cold-absorbing atmosphere above the region of light. We pass on to the second question, as to the region of outbreak of a spot. This question has been answered in an admirable manner by Carrington, who showed in a complete discussion of all the spots extending from 1854 to 1860, that, generally speaking, the region of outbreak of spots is the equatorial zone of the sun. At certain periods, however, he has shown that the zone is very closely confined to the equator, though at other periods it opens out. Such an opening out began about September 1856, at which epoch the generality of spots were for the most part found at a latitude of 30° either north or south of the solar equator. After this they gradually narrowed in towards the equator. The date of the next widening out cannot be given until the Kew records are reduced, but it is believed that at the present moment, or very recently, there has been a similar phenomenon. Thus while, generally speaking, spots attach themselves to the equatorial region of the sun, they are nevertheless inconstant in their attachment; and just as we have a small ripple proceeding on the back of a large wave, so we have minor periods of opening out proceeding on the back of the large period described by Carrington. The Kew Observers have very |