orate care with which the student of the noblest of all sciences is informed that an opaque body can cast a shadow, and that this shadow will have such and such characteristics. I am not here speaking of elementary treatises. It is reasonable enough, perhaps, in a first book for children to explain that "when the moon stops the sun's light its shadow falls on a part of the earth," and that the "people who live on that particular part of the earth where the shadow falls cannot see the sun because the moon is in the way." This is very pleasing and instructive for very small people; but when in treatises of a higher class the student is gravely informed of these things, as though they involved entirely new and striking conceptions, the idea is suggested that as tronomers think but lightly of the capacity of those who chance not to have made astronomy their chief subject of inquiry. On the other hand, the points about which most readers would care to hear something are commonly left untouched. Scarcely any reader of the usual explanation of eclipses fails to feel interested in the question of the laws according to which the moon comes between the sun and the earth, or the earth between the sun and the moon. The student feels that it may be very well to show him the consequences which follow when these bodies assume particular positions; but that he would also like to know a little about the causes of their becoming so placed as well as of the laws according to which the sequence of such events is determined. We are thus led to a mode of considering the subject which is very generally useful in the study of astronomy. I cannot, indeed, too earnestly recommend the student of the science to employ this method at every opportunity. It consists in imagining oneself placed at some suitable standpoint whence all the movements of such and such celestial bodies may be watched. In this case, the proper standpoint is the sun himself, and the bodies to be watched in imagination are the earth and moon. The student must picture to himself this earth on which we live, as a small globe circling around his standpoiut once in a year. He must conceive this globe as no larger in appearance than any one of the planets as seen from the earth. He would, indeed, require a good telescope to see the earth (from his place on the sun) actually as a globe. Now let him further conceive that around this small globe a much smaller orb is circling once in rather more than four weeks; but that the direction in which he looks at the circular path of the smaller orb is always such that this orb seems to travel backwards and forwards across or close past the larger one. To show exactly how long this path would look as seen from the sun, as well as to illustrate other points of interest connected with this explanation, the following process may be employed. Let the reader draw a circle ten and three-quarter inches in diameter to represent the sun or moon as we see these orbs. At the centre of this circle draw a small one, one-tenth of an inch in diameter; this will represent the earth as seen from the sun. Three inches from this small circle set another, a fortieth of an inch in diameter; this will represent the moon as seen from the sun when at her greatest range of distance from the earth. Exactly on the opposite side of the little circle representing the earth, and three inches from that circle, set another little picture of the moon; this represents the moon as seen from the sun when at her greatest range of distance from the earth on the other side. The observer in the sun would see the moon pass backwards and forwards from one position to the other in rather more than four weeks. In thus moving backwards and forwards. the moon passes always close (in appearance) to the earth, but sometimes closer than at others, and sometimes right across or right behind the earth's face. The path, in fact, opens out into an oval whose greatest width, on our scale, is slightly more than five-tenths of an inch, then closes up, then opens out to the same degree, only tilted the other way, then closes up again, and so on continually, while the earth all the time is circling round the observer's standpoint once in a year, and the moon round her path (thus varying in aspect) once in twenty-nine and a half days. Speaking roughly, we may say that once a fortnight the imagined observer in the sun would see the moon crossing the earth's place. He would always see the moon close to the * Of course the path is not a real entity, and could not therefore be seen, as supposed. It is conve We may thus nient, however, to regard it as such. compare it to the outer rim of Saturn's ring-system: and precisely as we see that ring-system closing up. and opening out systematically in the course of about twenty-nine years, so certainly an observer on the sun, watching our moon's course, would find her path opening out and closing up systematically in the course of eleven months eleven days, the seeming length of the path remaining appreciably unchanged, and about equal to three-fifths of the seeming diameter of the sun as seen from the earth. earth, since we have seen that the whole length of the moon's path, as seen from the sun, is much less than the breadth of the sun's globe as we see it; but twice in a month the moon would be very close by the earth. reach in five months and three weeks, the date December 27, which is the middle of the next "eclipse month." And so on continually. Other matters connected with the recurrence and peculiarities of these "eclipse months" belong, or should belong, to treatises on astronomy. What has been said above suffices for my present purpose, only a partial one, and closed the series. Now our observer in the sun would see that the moon's path passed from its 'greatest opening to a seeming line, and thence to its greatest opening again (but which is to explain the sequence of the with opposite tilt) in five months and late eclipses. It will be observed that about three weeks; passing back to a about eleven months and eleven days sepseeming line and to its original opening arate an eclipse month in one year from again, in all respects as at first, in the the corresponding eclipse month in the same time. Eleven months and eleven next. We thus see why the great Indian days complete the whole set of changes. eclipse of August, 1868, had its analogues, When the path seemed most open the so to speak, in the total eclipse of August moon would not at any time actually cross 29, in the preceeding year, and in the the earth's face, or pass actually behind it. American eclipse of August 7, 1899. These In other words, the moon would neither three eclipses, occuring eleven days earlier hide any part of the earth from the sun in each succeeding year, were all three nor be hidden by the earth. Hiding any total. But the series did not end with parts of the earth from the sun means the eclipse of August, 1863. On July 27, obviously eclipsing the sun as viewed 1870 (again eleven days earlier) there was from those parts of the earth; while an eclipse of the sun. It was, however, to say the moon is hidden from the sun by the earth means (no less obviously) that the moon is thrown into shadow, or eclipsed. So that when the moon's path, as seen from the sun, is most open forming then a long oval-there can be no eclipses either of the sun or moon. But when this path has in appearance closed up to a line, or nearly to a line, the moon can no longer pass by the earth (as viewed from the sun) without actually crossing the earth's disc or passing actually behind that disc. So long as this state of things lasts there must be an eclipse whenever the moon's backward and forward motion carries her past the earth. We have seen that the moon's path has this aspect, or is closed up into a straight line, as seen from the sun, at intervals of about five months and three weeks. For rather more than a month the path is sufficiently closed for eclipses to occur. I have suggested for these occasions the title of "eclipse months." To show how they succeed each other, take the following illustrative instance - Let January in any year be an eclipse month, the middle of January being the time when the moon's path appears closed up into a line as seen from the sun. Then five months and three weeks later, or about the 6th of July, the path is again closed up into a line as seen from the sun; and a period of rather more than a month, having this date for its middle - or from about June 22 to about July 23-is again an "eclipse month." Passing on from July 6, we Now the eclipse of the present month belongs to another series. It will be remembered by every one that there was an eclipse on December 22, last year; that eclipse was the first of the series to which the approaching eclipse belongs. This series, like the former, includes four eclipses. Last December the moon as seen from the sun crossed the earth's face near its northern edge. In the eclipse of Tuesday, December 12, the moon, as supposed to be seen from the sun, will pass slightly to the north of the middle point of the earth's face.* Thus the eclipse will be more important than that of last year, and the length of the actual track of the moon's shadow considerably greater. The third eclipse of the series will occur on November 30, 1872. In one respect it will be one of the most remarkable ever recorded; for it must be described as at once an annular and a total eclipse of the sun. This is readily explained, though the occurrence is altogether exceptional. It is a singular circumstance that the earth will present almost exactly the same face towards the sun at the moment of central eclipse on the 12th inst., as at the middle of the transit of Venus, on December 8, 1874. The fifteen pictures of the ro tating earth, in Plate VIII of my treatise on the sun, illustrate the approaching eclip-e as exactly as though drawn for the purpose. The first shows the earth's face as seen from the sun just before the moon's passage begins; the next thirteen show the earth's face at successive intervals of a quarter of an hour during the progress of the eclipse; and the last shows the earth's face as seen from the sun just after the moon has passed off that face. The reader is aware that the point of the | tions of the moon's shadow-track to which moon's conical shadow sometimes extends it has been judged advisable to send obbeyond and sometimes falls short of the servers. The track crosses the southern earth. In the former case an eclipse is extremity of the Indian peninsula, and total, in the latter it is annular. But in along this part of its course there will the eclipse of November 30, 1872, the apex probably be several observing parties, of the shadow falls short of the earth's the arrangements being superintended by surface at the beginning of the eclipse; it Mr. Pogson, the Government Astronomer encounters the earth as the shadow-track at Madras, and by Colonel Tenant and passed onward towards the bulging cen- Captain Herschel, both known to fame tral part of the earth's illuminated hemi- through their observations during the sphere; and presently, towards the close great eclipse of 1868. Thence the shadowof the eclipse, falls again short of the earth's track passes to the northern part of Ceysurface. So that there are two points lon, and along this part of its course the on the earth's surface where, on Novem- English eclipse expedition will be staber 30, 1872, the eclipse will be exactly tioned. It will probably be in the rememtotal, the moon just hiding the sun and no brance of most of my readers that the more, and only for a single instant. The English Government granted (several totality will nowhere last more than about months back) the sum of £2,000, as well three-quarters of a minute; and as the as transport and the means of camping, place where this will happen lies very far for an expedition to Ceylon. It was south in the Pacific Ocean, it is not likely hoped that Professor Stokes would have that any observer will witness this eclipse. been able to take charge of this expediIt is, however, the most considerable solar tion; but these hopes were disappointed. eclipse of the year 1872. The last eclipse Mr. Lockyer, however, has been able to of the series occurs on November 19, 1873, give his services, and doubtless the expeand, like the last of the former series, it is dition will be a highly effective one. The altogether unimportant. The moon, as shadow-track passes from Ceylon to Java, supposed to be seen from the sun, will just where a French party under M. Janssen graze the most southerly part of the will be stationed. Lastly, the shadowearth's disc. "The circumstances of the track passes to the northern part of the eclipse are such," says the Nautical Alma- Australian continent, and a strong observnac, "that a map has not been considered ing party has proceeded from Sydney and requisite." There will be no total solar Melbourne to the stations along this part eclipse at all in 1873. of the shadow's course. Not until April 16, 1874, will any total eclipse worth observing take place, after the eclipse of the present month. Nor are the circumstances of the eclipse of 1874 such as to encourage favourable hopes that much will be learned during its progress. On April 6, 1875, there will be, I believe, a much more important eclipse visible (as I judge from a rough calculation) in America; but I shall probably be excused from entering into an exact calculation of its circumstances, more especially as the Nautical Almanac for 1875 will, I believe, be published before this essay appears. It will be inferred that a considerable degree of interest is attached by astronomers to the eclipse of the present month, followed as it will be by two years and four months during which there will be no solar eclipses worthy of special observation. Although the eclipse of the 12th inst. is not nearly so favourable for observational purposes as the great Indian eclipse of 1868, yet there is a considerable variety as respects the choice of stations. In fact there are no less than four distinct sec The totality will last longest in North Australia, where its duration will be more than four minutes, or nearly two minutes longer than the duration of the eclipse of last year at the best stations. In Java the totality will last more than three minutes. In Ceylon the duration of totality will barely exceed by a few seconds the duration of totality last December. A somewhat curious mistake was made on this point in a scientific journal. Mr. Hind, in his first and comparatively rough estimate of the course traversed by the moon's shadow, had placed Trincomalee on the border of the track, so that the duration of totality at Trincomalee would have been very short. But after his final and more elaborate calculation. he set Trincomalee close to the centre of the shadow-track, with a duration of total obscuration amounting to two and a half minutes. Strangely enough the increase of the estimated duration was alone noticed by the writer of an article in Nature and it was reasoned that since the duration is so considerable at Trincomalee on the border of the track, it must be very much greater at places on the centre of However, as Professor Young remarks the track. I need scarcely point out that on this point (writing in 1871), "considerthis inference was unwarranted. In fact ed as a demonstration of self-luminosity the duration of totality can never under one bright line is just as conclusive as any circumstances be considerable for many." places close to the border of the shadowtrack. In southern India the eclipse will last about as long as the eclipse of last year at the best stations. It cannot be doubted that the observers this year will have a much more difficult task than those who have added so importantly to our knowledge during the eclipses of the last three years. This will appear on a brief consideration of the progress and present position of the problem with which the observers are to deal. In 1868, the observers of the great Indian eclipse discovered that the solar prominences are vast masses of glowing vapour, hydrogen being the chief constituent of these marvellous objects. But the solar corona, that glory of light which appears around and beyond the coloured prominences, did not at that time receive its interpretation. In 1869, the American observers directed their chief attention to this beautiful phenomenon; and they were singularly successful in their observations. One result of a very remarkable character was obtained by several observers. The light of the corona when analyzed in the spectroscope was found to be in large part monochromatic, the coronal spectrum showing one bright line. Now the reader is doubtless aware that in spectrum analysis the essential point is to determine where any bright or dark lines may lie along the range of that rainbow-tinted streak which we call the solar spectrum. In this instance the position of the bright line has been most satisfactorily determined by a very skilful spectroscopist, Professor Young, of America. The line agrees in position with one of the lines in the spectrum of iron, a line also seen in the spectrum of the aurora borealis. But the spectrum of iron contains upwards of 400 lines, while even the simpler spectrum of the aurora contains several lines; that of the corona, on the other hand, has not been proved to contain any other bright lines except the one just mentioned. Others have been suspected, but the degree of their brightness has not been such as to prove beyond all possibility of question that they belong to the solar It was in fact demonstrated by this observation alone that the corona, for a considerable part at least of its extension, is a self-luminous object. "Nor can there be any doubt," we may add with Professor Young, "as to the location of the selfluminous matter. It cannot be in our atmosphere, for no possible reason can be assigned why the particular molecules of the air that happen to lie near the lines which join the eye of the observer with the edge of the moon should become luminous rather than others in a different portion of the sky. Nor can it be at the moon; otherwise, of course, it would always be visible round her disc." "Accordingly," he adds, "it is now universally, I think I may say, acknowledged that one important element of the corona consists of a solar envelope of glowing gas reaching to a considerable elevation. Mr. Lockyer, who is still disposed to assign to the solar element of the corona a lower relative importance than most other astronomers, concedes a thickness of from six to ten minutes"that is from a fifth to a third of the solar diameter. This, as I have said, was written by Professor Young in 1871, but before a certain most important fact had come to his knowledge, which without at all affecting what he here puts forward, renders it possible to say much more as to the real extension of the corona. We have seen that a certain object, surrounding the sun on all sides to a distance of from 160,000 miles to 290,000 miles from his surface, is demonstrably a self-luminous envelope. It was to this envelope, or perhaps rather to its brighter portion as seen from the earth, that some proposed to assign the barbarous name "leucosphere," to distinguish it from the bright layer of prominence-matter close by the sun, which is called the sierra, or chromatosphere. But the visible extension of the corona is greater yet, and before the eclipse of 1870 doubts still existed as to the actual extent of that solar corona, which all had now begun to recognize as a real entity. That some portion of the light seen around the sun during total eclipse is in reality only due to the illumination of our own atmosphere is altogether beyond question. It is true, indeed, as was pointed out by Professors Young and Harkness, Dr. Curtis, and my The question which arises here, then, is this at what distance from the eclipsed sun has the light of the solar corona so diminished, and that of the atmospheric glare so increased, that the latter light predominates over the former. This ques self, that none of the coronal light for several degrees from the sun's place, can be solar light reflected by our atmosphere, as had been mistakenly supposed; but it is no less certain that our atmosphere is illuminated not merely in directions lying close up to the moon's edge, but evention is not only exceedingly nice, but, as towards the body of the moon herself, by the light of the coloured prominences and of the real solar corona. The observer himself sees these luminous objects during totality, and therefore the air all round him must be illuminated by them.* actually stated, it is wholly unanswerable, unless as a matter of fact the real solar corona has definite limits, recognizable perhaps by more refined methods of observation than have yet been applied. But although it is unlikely that the Now here a question of extreme delicacy utmost actual extension of the corona can arises. The true solar corona undoubtedly be determined by means of such appliances grows fainter and fainter with increased as are at present available, yet it was extension from the sun. That is, if we possible last December to demonstrate the could see the corona from some point extension of the corona to a distance far raised above the earth's atmosphere, so exceeding the six or ten minutes acknow}that no terrestrial illumination could de-edged by those who had once sought to ceive us, we should see the corona gradu- reason away the corona almost wholly. ally diminishing in lustre with distance It is clear that if any definite coronal from the sun, until at last it became too feature extending more than ten minutes faint to be discerned at all. On the con- from the place of the eclipsed sun, could trary, the illumination of our atmosphere be seen at stations far apart, then beyond during totality must necessarily increase all question that feature would be shewn with distance from the direction of the to be extra terrestrial. For instance, it eclipsed sun. This is obvious, because those molecules of the air which lie directly towards the moon's place are themselves suffering total eclipse from the sun's direct light, and are illuminated by a rather less proportion of prominence and coronal light than the observer himself, whereas those molecules which lie in directions far removed from the place of the eclipsed sun are suffering either but a partial eclipse, or else, though their eclipse be total, they are yet illuminated by more lustrous portions of the corona and prominence-matter. So that so far as atmospheric glare alone is concerned, we should have, as I wrote in March, 1870, a relatively "dark region around the eclipsed sun and a gradual increase of light with distance from him." One cannot but be surprised at the stress which was laid by some soon after the eclipse of last December, on the fact that even directly towards the moon's place, light was received which the spectroscope showed to be similar in character to that of the bright inner portion of the corona. Not only was the fact dwelt on repeatedly as a proof that the corona lies on our side of the moon, but it was commended to my own special attention as a proof that I had been mistaken in urging before the eclipse of 1870 that the corona is demonstrably a solar appendage. In the very paper in which I urged this view before the Royal Astronomical Society, on March 11, 1870, I ponted out that our air must be illuminated towards the moon's place by the light of all the visible solar appendages — as the prominences, chromatosphere, and corona as well as by reflected earth-light. My words were sufficiently distinct. They rau as follows: The light from all these sources should extend over the moon's disc, since it would illuminate the air between the observer and the moon's body." could not possibly be imagined that some peculiarity in the air over Syracuse could reproduce a feature of this sort precisely as it appeared to the observers near Xerez, owing to a peculiarity of the air over this station. Now, soon after the eclipse occurred, it was announced that the observers in Spain had recognized a peculiar gap, shaped like a letter V, in the lower portion of the corona- -on the left hand. This gap was pictured and described to me by my friend, Mr. W. H. H. Hudson, M.A., and Fellow of St. John's College, Cambridge, before any of the other accounts had come under my notice; and it was with some interest that I awaited the January meeting of the Royal Astronomical Society, before which the records of the observers in Spain were to be presented. At that meeting a picture was exhibited by Lieutenant Brown, in which this V-shaped gap was a very prominent feature. But in the discussion which ensued after Lieutenant Brown's paper had been read, Mr. Hudson remarked that the gap had seemed somewhat larger to him, on which Lieutenant Brown admitted that perhaps the size of the gap had not been quite adequately presented in his drawing. After the meeting a photograph, taken during the eclipse by Mr. Willard, of America, was shown to a few of those present. Why the picture was not exhibited and described at the meeting itself I |