The appearance of a new star in the constellation of the Swan in the autumn of 1876 promises to throw even more light than was expected on some of the most interesting problems with which modern astronomy has to deal. It was justly regarded as a circumstance of extreme interest that so soon after the outburst of the star which formed a new gem in the Northern Crown in May, 1866, another should have shone forth under seemingly similar conditions. And when, as time went on, it appeared that in several respects the new star in the Swan differed from the new star in the Crown, astronomers found fresh interest in studying, as closely as possible, the changes presented by the former as it gradually faded from view. But they were not prepared to expect what has actually taken place, or to recognize so great a difference of character between these two new stars, that whereas one seemed throughout its visibility to ordinary eyesight, and even until the present time, to be justly called a star, the other should so change as to render it extremely doubtful whether at any time it deserved to be regarded as a star or sun.

Few astronomical phenomena, even of those observed during this century (so fruitful in great astronomical discoveries), seem better worthy of thorough investigation and study than those presented by the two stars which appeared in the Crown and in the Swan, in 1866 and 1876 respectively.107 A new era seems indeed to be beginning for those departments of astronomy which deal with stars and star-cloudlets on the one hand, and with the evolution of solar systems and stellar systems on the other.

Let us briefly consider the history of the star of 1866 in the first place, and then turn our thoughts to the more surprising and probably more instructive history of the star which shone out in November, 1876.

In the first place, however, I would desire to make a few remarks on the objections which have been expressed by an observer to whom astronomy is indebted for very useful work, against the endeavour to interpret the facts ascertained respecting these so-called new stars. M. Cornu, who made some among the earliest spectroscopic observations of the star in Cygnus, after describing his results, proceeded as follows:—“Grand and seductive though the task may be of endeavouring to draw from observed facts inductions respecting the physical state of this new star, respecting its temperature, and the chemical reactions of which it may be the scene, I shall abstain from all commentary and all hypothesis on this subject. I think that we do not yet possess the data necessary for arriving at useful conclusions, or at least at conclusions capable of being tested: however attractive hypotheses may be, we must not forget that they are outside the bounds of science, and that, far from serving science, they seriously endanger its progress.” This, as I ventured to point out at the time, is utterly inconsistent with all experience. M. Cornu’s objection to theorizing when he did not see his way to theorizing justly, is sound enough; but his general objection to theorizing is, with all deference be it said, sheerly absurd. It will be noticed that I say theorizing, not hypothesis-framing; for though he speaks of hypotheses, he in reality is describing theories. The word hypothesis is too frequently used in this incorrect sense—perhaps so frequently that we may almost prefer sanctioning the use to substituting the correct word. But the fact really is, that many, even among scientific writers,108 when they hear the word hypothesis, think immediately of Newton’s famous “hypotheses non fingo,” a dictum relating to real hypotheses, not to theories. It would, in fact, be absurd to suppose that Newton, who had advanced, advocated, and eventually established, the noblest scientific theory the world has known, would ever have expressed an objection to theorizing, as he is commonly understood to have done by those who interpret his “hypotheses non fingo” in the sense which finds favour with M. Cornu. But apart from this, Newton definitely indicates what he means by hypotheses. “I frame no hypotheses,” he says, “for whatever is not deduced from phenomena is to be called an hypothesis.” M. Cornu, it will be seen, rejects the idea of deducing from phenomena what he calls an hypothesis, but what would not be an hypothesis according to Newton’s definition: “Malgré tout ce qu’il y aurait de séduisant et de grandiose à tirer de ce fait des inductions, etc., je m’abstiendrai de tout commentaire et de toute hypothèse à ce sujet.” It is not thus that observed scientific facts are to be made fruitful, nor thus that the points to which closer attention must be given are to be ascertained.

Since the preceding paragraph was written, my attention has been attracted to the words of another observer more experienced than M. Cornu, who has not only expressed the same opinion which I entertain respecting M. Cornu’s ill-advised remark, but has illustrated in a very practical way, and in this very case, how science gains from commentary and theory upon observed facts. Herr V?gel considers “that the fear that an hypothesis” (he, also, means a theory here) “might do harm to science is only justifiable in very rare cases: in most cases it will further science. In the first place, it draws the attention of the observer to things which but for the hypothesis might have been neglected. Of course if the observer is so strongly influenced that in favour of an hypothesis he sees things which do not exist—and this may happen sometimes—science may for a while be arrested in its progress, but in that case the observer is109 far more to blame than the author of the hypothesis. On the other hand, it is very possible that an observer may, involuntarily, arrest the progress of science, even without originating an hypothesis, by pronouncing and publishing sentences which have a tendency to diminish the general interest in a question, and which do not place its high significance in the proper light.” (This is very neatly put.) He is “almost inclined to think that such an effect might follow from the reading of M. Cornu’s remark, and that nowhere better than in the present case, where in short periods colossal changes showed themselves occurring upon a heavenly body, might the necessary data be obtained for drawing useful conclusions, and tests be applied to those hypotheses which have been ventured with regard to the condition of heavenly bodies.” It was, as we shall presently see, in thus collecting data and applying tests, that V?gel practically illustrated the justice of his views.

The star which shone out in the Northern Crown in May, 1866, would seem to have grown to its full brightness very quickly. It is not necessary that I should here consider the history of the star’s discovery; but I think all who have examined that history agree in considering that whereas on the evening of May 12, 1866, a new star was shining in the Northern Crown with second-magnitude brightness, none had been visible in the same spot with brightness above that of a fifth-magnitude star twenty-four hours earlier. On ascertaining, however, the place of the new star, astronomers found that there had been recorded in Argelander’s charts and catalogue a star of between the ninth and tenth magnitude in this spot. The star declined very rapidly in brightness. On May 13th it appeared of the third magnitude; on May 16th it had sunk to the fourth magnitude; on the 17th to the fifth; on the 19th to the seventh; and by the end of the month it shone only as a telescopic star of the ninth magnitude. It is now certainly not above the tenth magnitude.

Examined with the spectroscope, this star was found to110 be in an abnormal condition. It gave the rainbow-tinted streak crossed by dark lines, which is usually given by stars (with minor variations, which enable astronomers to classify the stars into several distinct orders). But superposed upon this spectrum, or perhaps we should rather say shining through this spectrum, were seen four brilliant lines, two of which certainly belonged to glowing hydrogen. These lines were so bright as to show that the greater part of the light of the star at the time came from the glowing gas or gases giving these lines. It appeared, however, that the rainbow-tinted spectrum on which these lines were seen was considerably brighter than it would otherwise have been, in consequence of the accession of heat indicated by and probably derived from the glowing hydrogen.

Unfortunately, we have not accordant accounts of the changes which the spectrum of this star underwent as the star faded out of view. Wolf and Rayet, of the Paris Observatory, assert that when there remained scarcely any trace of the continuous spectrum, the four bright lines were still quite brilliant. But Huggins affirms that this was not the case in his observations; he was “able to see the continuous spectrum when the bright lines could be scarcely distinguished.” As the bright lines certainly faded out of view eventually, we may reasonably assume that the French observers were prevented by the brightness of the lines from recognizing the continuous spectrum at that particular stage of the diminution of the star’s light when the continuous spectrum had faded considerably but the hydrogen lines little. Later, the continuous spectrum ceased to diminish in brightness, while the hydrogen lines rapidly faded. Thereafter the continuous spectrum could be discerned, and with greater and greater distinctness as the hydrogen lines faded out.

Now, in considering the meaning of the observed changes in the so-called “new star,” we have two general theories to consider.

One of these theories is that to which Dr. Huggins would seem to have inclined, though he did not definitely111 adopt it—the theory, namely, that in consequence of some internal convulsion enormous quantities of hydrogen and other gases were evolved, which in combining with some other elements ignited on the surface of the star, and thus enveloped the whole body suddenly in a sheet of flame.

“The ignited hydrogen gas in burning produced the light corresponding to the two bright bands in the red and green; the remaining bright lines were not, however, coincident with those of oxygen, as might have been expected. According to this theory, the burning hydrogen must have greatly increased the heat of the solid matter of the photosphere and brought it into a state of more intense incandescence and luminosity, which may explain how the formerly faint star could so suddenly assume such remarkable brilliance; the liberated hydrogen became exhausted, the flame gradually abated, and with the consequent cooling the photosphere became less vivid, and the star returned to its original condition.”

According to the other theory, advanced by Meyer and Klein, the blazing forth of this new star may have been occasioned by the violent precipitation of some great mass, perhaps a planet, upon a fixed star, “by which the momentum of the falling mass would be changed into molecular motion,” and result in the emission of light and heat.

“It might even be supposed that the new star, through its rapid motion, may have come in contact with one of the nebul? which traverse in great numbers the realms of space in every direction, and which from their gaseous condition must possess a high temperature; such a collision would necessarily set the star in a blaze, and occasion the most vehement ignition of its hydrogen.”

If we regard these two theories in their more general aspect, considering one as the theory that the origin of disturbance was within the star, and the other as the theory that the origin of disturbance was outside the star, they seem to include all possible interpretations of the observed112 phenomena. But, as actually advanced, neither seems satisfactory. The sudden pouring forth of hydrogen from the interior, in quantities sufficient to explain the outburst, seems altogether improbable. On the other hand, as I have pointed out elsewhere, there are reasons for rejecting the theory that the cause of the heat which suddenly affected this star was either the downfall of a planet on the star or the collision of the star with a star-cloudlet or nebula, traversing space in one direction, while the star rushed onwards in another.

A planet could not very well come into final conflict with its sun at one fell swoop. It would gradually draw nearer and nearer, not by the narrowing of its path, but by the change of the path’s shape. The path would, in fact, become more and more eccentric; until at length, at its point of nearest approach, the planet would graze its primary, exciting an intense heat where it struck, but escaping actual destruction that time. The planet would make another circuit, and again graze the sun, at or near the same part of the planet’s path. For several circuits this would continue, the grazes not becoming more and more effective each time, but rather less. The interval between them, however, would grow continually less and less; at last the time would come when the planet’s path would be reduced to the circular form, its globe touching the sun’s all the way round, and then the planet would very quickly be reduced to vapour and partly burned up, its substance being absorbed by its sun. But all successive grazes would be indicated to us by accessions of lustre, the period between each seeming outburst being only a few months at first, and gradually becoming less and less (during a long course of years, perhaps even of centuries) until the planet was finally destroyed. Nothing of this sort has happened in the case of any so-called new star. As for the rush of a star through a nebulous mass, that is a theory which would scarcely be entertained by any one acquainted with the enormous distances separating the gaseous star-clouds113 properly called nebul?. There may be small clouds of the same sort scattered much more freely through space; but we have not a particle of evidence that this is actually the case. All we certainly know about star-cloudlets suggests that the distances separating them from each other are comparable with those which separate star from star, in which case the idea of a star coming into collision with a star-cloudlet, and still more the idea of this occurring several times in a century, is wild in the extreme.

But while thus advancing objections, which seem to me irrefragable, against the theory that either a planet or a nebula (still less another small star) had come into collision with the orb in Corona which shone out so splendidly for a while, I advanced another view which seemed to me then and seems now to correspond well with phenomena, and to render the theory of action from without on the whole preferable to the theory of outburst from within. I suggested that, far more probably, an enormous flight of large meteoric masses travelling around the star had come into partial collision with it in the same way that the flight of November meteors comes into collision with our earth thrice in each century, and that other meteoric flights may occasionally come into collision with our sun, producing the disturbances which occasion the sun-spots. As I pointed out, in conceiving this we are imagining nothing new. A meteoric flight capable of producing the suggested effects would differ only in kind from meteoric flights which are known to circle around our own sun. The meteors which produce the November displays of falling stars follow in the track of a comet barely visible to the naked eye.

“May we not reasonably assume that those glorious comets which have not only been visible but conspicuous, shining even in the day-time, and brandishing around tails, which like that of the ‘wonder in heaven, the great dragon,’ seemed to ‘draw the third part of the stars of heaven,’ are followed by much denser flights of much more massive meteors? Some of these giant comets have paths which114 carry them very close to our sun. Newton’s comet, with its tail a hundred millions of miles in length, all but grazed the sun’s globe. The comet of 1843, whose tail, says Sir John Herschel, ‘stretched half-way across the sky,’ must actually have grazed the sun, though but lightly, for its nucleus was within 80,000 miles of his surface, and its head was more than 160,000 miles in diameter. And these are only two among the few comets whose paths are known. At any time we might be visited by a comet mightier than either, travelling in an orbit intersecting the sun’s surface, followed by flights of meteoric masses enormous in size and many in number, which, falling on the sun’s globe with enormous velocity corresponding to their vast orbital range and their near approach to the sun—a velocity of some 360 miles per second—would, beyond all doubt, excite his whole frame, and especially his surface regions, to a degree of heat far exceeding what he now emits.”

This theory corresponds far better also with observed facts than the theory of Meyer and Klein, in other respects than simply in antecedent probability. It can easily be shown that if a planet fell upon a sun in such sort as to become part of his mass, or if a nebula in a state of intense heat excited the whole frame of a star to a similar degree of heat, the effects would be of longer duration than the observed accession of heat and light in the case of all the so-called “new stars.” It has been calculated by Mr. Croll (the well-known mathematician to whom we owe the most complete investigations yet made into the effect of the varying eccentricity of the earth’s orbit on the climate of the earth) that if two suns, each equal in mass to one-half of our sun, came into collision with a velocity of 476 miles per second, light and heat would be produced which would cover the present rate of the sun’s radiation for fifty million years. Now although it certainly does not follow from this that such a collision would result in the steady emission of so much light and heat as our sun gives out, for a period of fifty million years, but is, on the contrary, certain that there115 would be a far greater emission at first and a far smaller emission afterwards, yet it manifestly must be admitted that such a collision could not possibly produce so short-lived an effect as we see in the case of every one of the so-called new stars. The diminution in the emission of light and heat from the maximum to one-half the maximum would not occupy fifty millions of years, or perhaps even five million or five hundred thousand years; but it would certainly require thousands of years; whereas we have seen that the new stars in the Crown and in the Swan have lost not one-half but ninety-nine hundredths of their maximum lustre in a few months.

This has been urged as an objection even to the term star as applied to these suddenly appearing orbs. But the objection is not valid; because there is no reason whatever for supposing that even our own sun might not be excited by the downfall of meteoric or cometic matter upon it to a sudden and short-lasting intensity of splendour and of heat. Mr. Lockyer remarks that, if any star, properly so called, were to become a “a world on fire,” or “burst into flames,” or, in less poetical language, were to be driven either into a condition of incandescence absolutely, or to have its incandescence increased, there can be little doubt that thousands or millions of years would be necessary for the reduction of its light to its original intensity. This must, however, have been written in forgetfulness of some facts which have been ascertained respecting our sun, and which indicate pretty clearly that the sun’s surface might be roused to a temporary intensity of splendour and heat without any corresponding increase in the internal heat, or in the activity of the causes, whatever they may be, to which the sun’s steady emissions of light and heat are due.

For instance, most of my readers are doubtless familiar with the account (an oft-told tale, at any rate) of the sudden increase in the splendour of a small portion of the sun’s surface on September 1, 1859, observed by two astronomers independently. The appearances described corresponded116 exactly with what we should expect if two large meteoric masses travelling side by side had rushed, with a velocity originally amounting to two or three hundred miles per second, through the portions of the solar atmosphere lying just above, at, and just below the visible photosphere. The actual rate of motion was measured at 120 miles per second as the minimum, but may, if the direction of motion was considerably inclined to the line of sight, have amounted to more than 200 miles per second. The effect was such, that the parts of the sun thus suddenly excited to an increased emission of light and heat appeared like bright stars upon the background of the glowing photosphere itself. One of the observers, Carrington, supposed for a moment that the dark glass screen used to protect the eye had broken. The increase of splendour was exceedingly limited in area, and lasted only for a few minutes—fortunately for the inhabitants of earth. As it was, the whole frame of the earth sympathized with the sun. Vivid auroras were seen, not only in both hemispheres, but in latitudes where auroras are seldom seen. They were accompanied by unusually great electro-magnetic disturbances.

“In many places,” says Sir J. Herschel, “the telegraph wires struck work. At Washington and Philadelphia, the electric signalmen received severe electric shocks. At a station in Norway, the telegraphic apparatus was set fire to, and at Boston, in North America, a flame of fire followed the pen of Bain’s electric telegraph, which writes down the message upon chemically prepared paper.”

We see, then, that most certainly the sun can be locally excited to increased emission of light and heat, which nevertheless may last but for a very short time; and we have good reason for believing that the actual cause of the sudden change in his condition was the downfall of meteoric matter upon a portion of his surface. We may well believe that, whatever the cause may have been, it was one which might in the case of other suns, or even in our sun’s own case, affect a much larger portion of the photosphere. If this happened117 there would be just such an accession of splendour as we recognize in the case of the new stars. And as the small local accession of brilliancy lasted only a few minutes, we can well believe that an increase of surface brilliancy affecting a much larger portion of the photosphere, or even the entire photosphere, might last but for a few days or weeks.

All that can be said in the way of negative evidence, so far as our own sun is concerned, is that we have no reason for believing that our sun has, at any time within many thousands of years, been excited to emit even for a few hours a much greater amount of light and heat than usual; so that it has afforded no direct evidence in favour of the belief that other suns may be roused to many times their normal splendour, and yet very quickly resume that usual lustre. But we know that our sun, whether because of his situation in space, or of his position in time (that is, the stage of solar development to which he has at present attained), belongs to the class of stars which shine with steady lustre. He does not vary like Betelgeux, for example, which is not only a sun like him as to general character, but notably a larger and more massive orb. Still less is he like Mira, the Wonderful Star; or like that more wonderful variable star, Eta Arg?s, which at one time shines with a lustre nearly equalling that of the bright Sirius, and anon fades away almost into utter invisibility. He is a variable sun, for we cannot suppose that the waxing and waning of the sun-spot period leaves his lustre, as a whole, altogether unaffected. But his variation is so slight that, with all ordinary methods of photometric measurement by observers stationed on worlds which circle around other suns, it must be absolutely undiscernible. We do not, however, reject Betelgeux, or Mira, or even Eta Arg?s, from among stars because they vary in lustre. We recognize the fact that, as in glory, so in condition and in changes of condition, one star differeth from another.

Doubtless there are excellent reasons for rejecting the theory that a massive body like a planet, or a nebulous mass like those which are found among the star-depths (the least118 of which would exceed many times in volume a sphere filling the entire space of the orbit of Neptune), fell on some remote sun in the Northern Crown. But there are no sufficient reasons for rejecting or even doubting the theory that a comet, bearing in its train a flight of many millions of meteoric masses, falling directly upon such a sun, might cause it to shine with many times its ordinary lustre, but only for a short time, a few months or weeks, or a few days, or even hours. In the article entitled “Suns in Flames,” in my “Myths and Marvels of Astronomy,” before the startling evidence recently obtained from the star in Cygnus had been thought of, I thus indicated the probable effects of such an event:—“When the earth has passed through the richer portions (not the actual nuclei be it remembered) of meteor systems, the meteors visible from even a single station have been counted by tens of thousands, and it has been computed that millions must have fallen upon the whole earth. These were meteors following in the trains of very small comets. If a very large comet followed by no denser a flight of meteors, but each meteoric mass much larger, fell directly upon the sun, it would not be the outskirts but the nucleus of the meteoric train which would impinge upon him. They would number thousands of millions. The velocity of downfall of each mass would be more than 360 miles per second. And they would continue to pour in upon him for several days in succession, millions falling every hour. It seems not improbable that under this tremendous and long-continued meteoric hail, his whole surface would be caused to glow as intensely as that small part whose brilliancy was so surprising in the observation made by Carrington and Hodgson. In that case our sun, seen from some remote star whence ordinarily he is invisible, would shine out as a new sun for a few days, while all things living, on our earth and whatever other members of the solar system are the abodes of life, would inevitably be destroyed.”

There are, indeed, reasons for believing, not only, as I have already indicated, that the outburst in the sun was caused by the downfall of meteoric masses, but that those119 masses were following in the train of a known comet, precisely as the November meteors follow in the train of Tempel’s comet (II., 1866). For we know that November meteoric displays have been witnessed for five or six years after the passage of Tempel’s comet, in its thirty-three year orbit, while the August meteoric displays have been witnessed fully one hundred and twenty years after the passage of their comet (II., 1862).15 Now only sixteen years before the solar outburst witnessed by Carrington and Hodgson, a magnificent comet had passed even closer to the sun than either Tempel’s comet or the second comet of 1862 approached the earth’s orbit. That was the famous comet of the year 1843. Many of us remember that wonderful object. I was but a child myself when it appeared, but I can well remember its amazing tail, which in March, 1843, stretched half-way across the sky.

“Of all the comets on record,” says Sir J. Herschel, “that approached nearest the sun; indeed, it was at first supposed that it had actually grazed the sun’s surface, but it proved to have just missed by an interval of not more than 80,000 miles—about a third of the distance of the moon from the earth, which (in such a matter) is a very close shave indeed to get clear off.”

We can well believe that the two meteors which produced the remarkable outburst of 1859 may have been stragglers from the main body following after that glorious comet. I do not insist upon the connection. In fact, I rather incline to the belief that the disturbance in 1859, occurring as it did about the time of maximum sun-spot frequency, was caused by meteors following in the train of some as yet undiscovered comet, circuiting the sun in about eleven years, the spots120 themselves being, I believe, due in the main to meteoric downfalls. There is greater reason for believing that the great sun-spot which appeared in June, 1843, was caused by the comet which three months before had grazed the sun’s surface. As Professor Kirkwood, of Bloomington, Indiana, justly remarks, had this comet approached a little nearer, the resistance of the solar atmosphere would probably have brought the comet’s entire mass to the solar surface. Even at its actual distance, it must have produced considerable atmospheric disturbance. But the recent discovery that a number of comets are associated with meteoric matter travelling in nearly the same orbits, suggests the inquiry whether an enormous meteorite following in the comet’s train, and having a somewhat less perihelion distance, may not have been precipitated upon the sun, thus producing the great disturbance observed so shortly after the comet’s perihelion passage.

Let us consider now the evidence obtained from the star in Cygnus, noting especially in what points it resembles, and in what points it differs from, the evidence afforded by the star in the Crown.

The new star was first seen by Professor Schmidt at a quarter to six on the evening of November 24. It was then shining as a star of the third magnitude, in the constellation of the Swan, not very far from the famous but faint star 61 Cygni—which first of all the stars in the northern heavens had its distance determined by astronomers. The three previous nights had unfortunately been dark; but Schmidt is certain that on November 20 the star was not visible. At midnight, November 24, its light was very yellow, and it was somewhat brighter than the well-known star Eta Pegasi, which marks the forearm of the Flying Horse. Schmidt sent news of the discovery to Leverrier, at Paris; but neither he nor Leverrier telegraphed the news, as they should have done, to Greenwich, Berlin, or the United States. Many precious opportunities for observing the spectrum of the new-comer at the time of its greatest brilliancy were thus lost.

121 The observers at Paris did their best to observe the spectrum of the star and the all-important changes in the spectrum. But they had unfavourable weather. It was not till December 2 that the star was observed at Paris, by which time the colour, which had been very yellow on November 24, had become “greenish, almost blue.” The star had also then sunk from the third to far below the fourth magnitude. It is seldom that science has to regret a more important loss of opportunity than this. What we want specially to know is the nature of the spectrum given by this star when its light was yellow; and this we can now never know. Nor are the outbursts of new stars so common that we may quickly expect another similar opportunity, even if any number of other new stars should present the same series of phenomena as the star in Cygnus.

On December 2, the spectrum, as observed by M. Cornu, consisted almost entirely of bright lines. On December 5, he determined the position of these lines, though clouds still greatly interfered with his labours. He found three bright lines of hydrogen, the strong double sodium line in the orange-yellow, the triple magnesium line in the yellow-green, and two other lines—one of which seemed to agree exactly in position with a bright line belonging to the solar corona. All these lines were shining upon the rainbow-tinted background of the spectrum, which was relatively faint. He drew the conclusion that in chemical constitution the atmosphere of the new star was constituted exactly like the solar sierra.

Herr V?gel’s observations commenced on December 5, and were continued at intervals until March 10, when the star had sunk to below the eighth magnitude.

V?gel’s earlier observations agreed well with Cornu’s. He remarks, however, that Cornu’s opinion as to the exact resemblance of the chemical constitution of the star’s atmosphere with that of the sierra is not just, for both Cornu and himself noticed one line which did not correspond with any line belonging to the solar sierra; and this line122 eventually became the brightest line of the whole spectrum. Comparing his own observations with those of Cornu, V?gel points out that they agree perfectly with regard to the presence of the three hydrogen lines, and that of the brightest line of the air spectrum (belonging to nitrogen),—which is the principal line of the spectrum of nebul?. This is the line which has no analogue in the spectrum of the sierra.

We have also observations by F. Secchi, at Rome, Mr. Copeland, at Dunecht, and Mr. Backhouse, of Sunderland, all agreeing in the main with the observations made by V?gel and Cornu. In particular, Mr. Backhouse observed, as V?gel had done, that whereas in December the greenish-blue line of hydrogen, F, was brighter than the nitrogen line (also in the green-blue, but nearer the red end than F), on January 6 the nitrogen line was the brightest of all the lines in the spectrum of the new star.

V?gel, commenting on the results of his observations up to March 10, makes the following interesting remarks (I quote, with slight verbal alterations, from a paraphrase in a weekly scientific journal):—“A stellar spectrum with bright lines is always a highly interesting phenomenon for any one acquainted with stellar spectrum analysis, and well worthy of deep consideration. Although in the chromosphere (sierra) of our sun, near the limb, we see numerous bright lines, yet only dark lines appear in the spectrum whenever we produce a small star-like image of the sun, and examine it through the spectroscope. It is generally believed that the bright lines in some few star-spectra result from gases which break forth from the interior of the luminous body, the temperature of which is higher than that of the surface of the body—that is, the phenomenon is the same sometimes observed in the spectra of solar spots, where incandescent hydrogen rushing out of the hot interior becomes visible above the cooler spots through the hydrogen lines turning bright. But this is not the only possible explanation. We may also suppose that the atmosphere123 of a star, consisting of incandescent gases, as is the case with our own sun, is on the whole cooler than the nucleus, but with regard to the latter is extremely large. I cannot well imagine how the phenomenon can last for any long period of time if the former hypothesis be correct. The gas breaking forth from the hot interior of the body will impart a portion of its heat to the surface of the body, and thus raise the temperature of the latter; consequently, the difference of temperature between the incandescent gas and the surface of the body will soon be insufficient to produce bright lines; and these will disappear from the spectrum. This view applies perfectly to stars which suddenly appear and soon disappear again, or at least increase considerably in intensity—that is, it applies perfectly to so-called new stars in the spectra of which bright lines are apparent, if the hypothesis presently to be mentioned is admitted for their explanation. For a more stable state of things the second hypothesis seems to be far better adapted. Stars like Beta Lyr?, Gamma Cassiopei?, and others, which show the hydrogen lines and the sierra D line bright on a continuous spectrum, with only slight changes of intensity, possess, according to this theory, atmospheres very large relatively to their own volume—the atmospheres consisting of hydrogen and that unknown element which produces the D line.16 With regard to the new star, Z?llner, long before the progress lately made in stellar physics by means of spectrum analysis, deduced from Tycho’s observations of the star called after him, that on the surface of a star, through the constant emission of heat, the products of cooling, which in the case of our sun we call sun-spots, accumulate: so that finally the whole surface of the body is covered with a colder stratum, which gives124 much less light or none at all. Through a sudden and violent tearing up of this stratum, the interior incandescent materials which it encloses must naturally break forth, and must in consequence, according to the extent of their eruption, cause larger or smaller patches of the dark envelope of the body to become luminous again. To a distant observer such an eruption from the hot and still incandescent interior of a heavenly body must appear as the sudden flashing-up of a new star. That this evolution of light may under certain conditions be an extremely powerful one, could be explained by the circumstance that all the chemical compounds which, under the influence of a lower temperature, had already formed upon the surface, are again decomposed through the sudden eruption of these hot materials; and that this decomposition, as in the case of terrestrial substances, takes place under evolution of light and heat. Thus the bright flashing-up is not only ascribed to the parts of the surface which through the eruption of the incandescent matter have again become luminous, but also to a simultaneous process of combustion, which is initiated through the colder compounds coming into contact with the incandescent matter.”

V?gel considers that Z?llner’s hypothesis has been confirmed in its essential points by the application of spectrum analysis to the stars. We can recognize from the spectrum different stages in the process of cooling, and in some of the fainter stars we perceive indeed that chemical compounds have already formed, and still exist. As to new stars, again, says V?gel, Z?llner’s theory seems in nowise contradicted “by the spectral observations made on the two new stars of 1866 and 1876. The bright continuous spectrum, and the bright lines only slightly exceeding it at first” (a description, however, applying correctly only to the star of 1876), “could not be well explained if we only suppose a violent eruption from the interior, which again rendered the surface wholly or partially luminous; but are easily explained if we suppose that the quantity of light is considerably augmented through125 a simultaneous process of combustion. If this process is of short duration, then the continuous spectrum, as was the case with the new star of 1876, will very quickly decrease in intensity down to a certain limit, while the bright lines in the spectrum, which result from the incandescent gases that have emanated in enormous quantities from the interior, will continue for some time.”

It thus appears that Herr V?gel regarded the observations which had been made on this remarkable star up to March 10 as indicating that first there had been an outburst of glowing gaseous matter from the interior, producing the part of the light which gave the bright lines indicative of gaseity, and that then there had followed, as a consequence, the combustion of a portion of the solid and relatively cool crust, causing the continuous part of the spectrum. We may compare what had taken place, on this hypothesis, with the outburst of intensely hot gases from the interior of a volcanic crater, and the incandescence of the lips of the crater in consequence of the intense heat of the out-rushing gases. Any one viewing such a crater from a distance, with a spectroscope, would see the bright lines belonging to the out-rushing gases superposed upon the continuous spectrum due to the crater’s burning lips. V?gel further supposes that the burning parts of the star soon cooled, the majority of the remaining light (or at any rate the part of the remaining light spectroscopically most effective) being that which came from the glowing gases which had emanated in vast quantities from the star’s interior.

“The observations of the spectrum show, beyond doubt,” he says, “that the decrease in the light of the star corresponds with the cooling of its surface. The violet and blue parts decreased more rapidly in intensity than the other parts; and the absorption-bands which crossed the spectrum have gradually become darker and darker.”

The reasoning, however, if not altogether unsatisfactory, is by no means so conclusive as Herr V?gel appears to think. It is not clear how the incandescent portion of the surface126 could possibly cool in any great degree while enormous quantities of gas more intensely heated (by the hypothesis) remained around the star. The more rapid decrease in the violet and blue parts of the spectrum than in the red and orange is explicable as an effect of absorption, at least as readily as by the hypothesis that burning solid or liquid matter had cooled. V?gel himself could only regard the other bands which crossed the spectrum as absorption-bands. And the absorption of light from the continuous spectrum in these parts (that is, not where the bright lines belonging to the gaseous matter lay) could not possibly result from absorption produced by those gases. If other gases were in question, gases which, by cooling with the cooling surface, had become capable of thus absorbing light from special parts of the spectrum, how is it that before, when these gases were presumably intensely heated, they did not indicate their presence by bright bands? Bright bands, indeed, were seen, which eventually faded out of view, but these bright bands did not occupy the position where, later on, absorption-bands appeared.

The natural explanation of what had thus far been observed is different from that advanced by V?gel, though we must not assume that because it is the natural, it is necessarily the true explanation. It is this—that the source of that part of the star’s light which gave the bright-line spectrum, or the spectrum indicative of gaseity, belongs to the normal condition of the star, and not to gases poured forth, in consequence of some abnormal state of things, from the sun’s interior. We should infer naturally, though again I say not therefore correctly, that if a star spectroscope had been directed upon the place occupied by the new star before it began to shine with unusual splendour, the bright-line spectrum would have been observed. Some exceptional cause would then seem to have aroused the entire surface of the star to shine with a more intense brightness, the matter thus (presumably) more intensely heated being such as would give out the combined continuous and bright-line127 spectrum, including the bright lines which, instead of fading out, shone with at least relatively superior brightness as the star faded from view. The theory that, on the contrary, the matter giving these more persistent lines was that whose emission caused the star’s increase of lustre, seems at least not proven, and I would go so far as to say that it accords ill with the evidence.

The question, be it noted, is simply whether we should regard the kind of light which lasts longest in this star as it fades out of view as more probably belonging to the star’s abnormal brightness or to its normal luminosity. It seems to me there can be little doubt that the persistence of this part of the star’s light points to the latter rather than to the former view.

Let it also be noticed that the changes which had been observed thus far were altogether unlike those which had been observed in the case of the star in the Northern Crown, and therefore cannot justly be regarded as pointing to the same explanation. As the star in the Crown faded from view, the bright lines indicative of glowing hydrogen died out, and only the ordinary stellar spectrum remained. In the case of the star in the Swan, the part of the spectrum corresponding to stellar light faded gradually from view, and bright lines only were left, at least as conspicuous parts of the star’s spectrum. So that whereas one orb seemed to have faded into a faint star, the other seemed fading out into a nebula—not merely passing into such a condition as to shine with light indicative of gaseity, but actually so changing as to shine with light of the very tints (or, more strictly, of the very wave-lengths) observed in all the gaseous nebul?.

The strange eventful history of the new star in Cygnus did not end here, however. We may even say, indeed, that it has not ended yet. But another chapter can already be written.

V?gel ceased from observing the star in March, precisely when observation seemed to promise the most interesting results. At most other observatories, also, no observations128 were made for about half a year. At the Dunecht Observatory17 pressure of work relating to Mars interfered with the prosecution of those observations which had been commenced early in the year. But on September 3, Lord Lindsay’s 15-inch reflector was directed upon the star. A star was still shining where the new star’s yellow lustre had been displayed in November, 1876; but now the star shone with a faint blue colour. Under spectroscopic examination, however, the light from this seeming blue star was found not to be starlight, properly speaking, at all. It formed no rainbow-tinted spectrum, but gave light of only a single colour. The single line now seen was that which at the time of V?gel’s latest observation had become the strongest of the bright lines of the originally complex spectrum of the so-called new star. It is the brightest of the lines given by the gaseous nebul?. In fact, if nothing had been known about this body before the spectroscopic observation of September 3 was made, the inference from the spectrum given by the blue star would undoubtedly have been that the object is in reality a small nebula of the planetary sort, very similar to that one close by the pole of the ecliptic, which gave Huggins the first evidence of the gaseity of nebul?, but very much smaller. I would specially direct the reader’s attention, in fact, to Huggins’s account of his observation of that planetary nebula in the Dragon. “On August 19, 1864,” he says, “I directed the telescope armed with the spectrum apparatus to this nebula. At first I suspected some derangement of the instrument had taken place, for no spectrum was seen, but only” a single line of light. “I then found that the light of this nebula, unlike any other extra-terrestrial light which had yet been subjected by me to prismatic analysis, was not composed of light of different refrangibilities, and therefore could not form a spectrum. A great part of the light from this nebula129 is monochromatic, and after passing through the prisms remains concentrated in a bright line.” A more careful examination showed that not far from the bright line was a much fainter line; and beyond this, again, a third exceedingly faint line was seen. The brightest of the three lines was a line of nitrogen corresponding in position with the brightest of the lines in the spectrum of our own air. The faintest corresponded in position with a line of hydrogen. The other has not yet been associated with a known line of any element. Besides the faint lines, Dr. Huggins perceived an exceedingly faint continuous spectrum on both sides of the group of bright lines; he suspected, however, that this faint spectrum was not continuous, but crossed by dark spaces. Later observations on other nebul? induced him to regard this faint continuous spectrum as due to the solid or liquid matter of the nucleus, and as quite distinct from the bright lines into which nearly the whole of the light from the nebula is concentrated. The fainter parts of the spectrum of the gaseous nebul?, in fact, correspond to those parts of the spectrum of the “new star” in Cygnus which last remained visible, before the light assumed its present monochromatic colour.

Now let us consider the significance of the evidence afforded by this discovery—not perhaps hoping at once to perceive the full meaning of the discovery, but endeavouring to advance as far as we safely can in the direction in which it seems to point.

We have, then, these broad facts: where no star had been known, an object has for a while shone with stellar lustre, in this sense, that its light gave a rainbow-tinted spectrum not unlike that which is given by a certain order of stars; this object has gradually parted with its new lustre, and in so doing the character of its spectrum has slowly altered, the continuous portion becoming fainter, and the chief lustre of the bright-line portion shifting from the hydrogen lines to a line which, there is every reason to believe, is absolutely identical with the nebula nitrogen line:130 and lastly, the object has ceased to give any perceptible light, other than that belonging to this nitrogen line.

Now it cannot, I think, be doubted that, accompanying the loss of lustre in this orb, there has been a corresponding loss of heat. The theory that all the solid and liquid materials of the orb have been vaporized by intense heat, and that this vaporization has caused the loss of the star’s light (as a lime-light might die out with the consumption of the lime, though the flame remained as hot as ever), is opposed by many considerations. It seems sufficient to mention this, that if a mass of solid matter, like a dead sun or planet, were exposed to an intense heat, first raising it to incandescence, and eventually altogether vaporizing its materials, although quite possibly the time of its intensest lustre might precede the completion of the vaporization, yet certainly so soon as the vaporization was complete, the spectrum of the newly vaporized mass would show multitudinous bright lines corresponding to the variety of material existing in the body. No known fact of spectroscopic analysis lends countenance to the belief that a solid or liquid mass, vaporized by intense heat, would shine thenceforth with monochromatic light.

Again, I think we are definitely compelled to abandon V?gel’s explanation of the phenomena by Z?llner’s theory. The reasons which I have urged above are not only strengthened severally by the change which has taken place in the spectrum of the new star since V?gel observed it, but an additional argument of overwhelming force has been introduced. If any one of the suns died out, a crust forming over its surface and this crust being either absolutely dark or only shining with very feeble lustre, the sun would still in one respect resemble all the suns which are spread over the heavens—it would show no visible disc, however great the telescopic power used in observing it. If the nearest of all the stars were as large, or even a hundred times as large, as Sirius, and were observed with a telescope of ten times greater magnifying power than any yet131 directed to the heavens, it would appear only as a point of light. If it lost the best part of its lustre, it would appear only as a dull point of light. Now the planetary nebul? show discs, sometimes of considerable breadth. Sir J. Herschel, to whom and to Sir W. Herschel we owe the discovery and observation of nearly all these objects, remarks that “the planetary nebul? have, as their name imports, a near, in some instances a perfect, resemblance to planets, presenting discs round, or slightly oval, in some quite sharply terminated, in others a little hazy or softened at the borders....” Among the most remarkable may be specified one near the Cross, whose light is about equal to that of a star just visible to the naked eye, “its diameter about twelve seconds, its disc circular or very slightly elliptic, and with a clear, sharp, well-defined outline, having exactly the appearance of a planet, with the exception of its colour, which is a fine and full blue, verging somewhat upon green.” But the largest of these planetary nebul?, not far from the southernmost of the two stars called the Pointers, has a diameter of 2? minutes of arc, “which, supposing it placed at a distance from us not greater than that of the nearest known star of our northern heavens, would imply a linear diameter seven times greater than that of the orbit of Neptune.” The actual volume of this object, on this assumption, would exceed our sun’s ten million million times. No one supposes that this planetary nebula, shining with a light indicative of gaseity, has a mass exceeding our sun’s in this enormous degree. It probably has so small a mean density as not greatly to exceed, or perhaps barely to equal, our sun in mass. Now though the “new star” in Cygnus presented no measurable disc, and still shines as a mere blue point in the largest telescope, yet inasmuch as its spectrum associated it with the planetary and gaseous nebul?, which we know to be much larger bodies than the stars, it must be regarded, in its present condition, as a planetary nebula, though a small one; and since we cannot for a moment imagine that the monstrous planetary132 nebul? just described are bodies which once were suns, but whose crust has now become non-luminous, while around the crust masses of gas shine with a faint luminosity, so are we precluded from believing that this smaller member of the same family is in that condition.

It is conceivable (and the possibility must be taken into account in any attempt to interpret the phenomena of the new star) that when shining as a star, the new orb, so far as this unusual lustre was concerned, was of sunlike dimensions. For we cannot tell whether the surface which gave the strong light was less or greater than, or equal to, that which is now shining with monochromatic light. Very likely, if we had been placed where we could have seen the full dimensions of the planetary nebula as it at present exists, we should have found only its nuclear part glowing suddenly with increased lustre, which, after very rapidly attaining its maximum, gradually died out again, leaving the nebula as it had been before. But that the mass now shining with monochromatic light is, I will not say enormously large, but of exceedingly small mean density, so that it is enormously large compared with the dimensions it would have if its entire substance were compressed till it had the same mean density as our own sun, must be regarded as, to all intents and purposes, certain.

We certainly have not here, then, the case of a sun which has grown old and dead and dark save at the surface, but within whose interior fire has still remained, only waiting some disturbing cause to enable it for a while to rush forth. If we could suppose that in such a case there could be such changes as the spectroscope has indicated—that the bright lines of the gaseous outbursting matter would, during the earlier period of the outburst, show on a bright continuous background, due to the glowing lips of the opening through which the matter had rushed, but later would shine alone, becoming also fewer in number, till at last only one was left,—we should find ourselves confronted with the stupendous difficulty that that single remaining line is the bright line of133 the planetary and other gaseous nebul?. Any hypothesis accounting for its existence in the spectrum of the faint blue starlike object into which the star in Cygnus has faded ought to be competent to explain its existence in the spectrum of those nebul?. But this hypothesis certainly does not so explain its existence in the nebular spectrum. The nebul? cannot be suns which have died out save for the light of gaseous matter surrounding them, for they are millions, or rather millions of millions, of times too large. If, for instance, a nebula, like the one above described as lying near the southernmost Pointer, were a mass of this kind, having the same mean density as the sun, and lying only at the distance of the nearest of the stars from us, then not only would it have the utterly monstrous dimensions stated by Sir J. Herschel, but it would in the most effective way perturb the whole solar system. With a diameter exceeding seven times that of the orbit of Neptune, it would have a volume, and therefore a mass, exceeding our sun’s volume and mass more than eleven millions of millions of times. But its distance on this assumption would be only about two hundred thousand times the sun’s, and its attraction reduced, as compared with his, on this account only forty thousand millions of times. So that its attraction on the sun and on the earth would be greater than his attraction on the earth, in the same degree that eleven millions are greater than forty thousand—or two hundred and seventy-five times. The sun, despite his enormous distance from such a mass, would be compelled to fall very quickly into it, unless he circuited (with all his family) around it in about one-sixteenth of a year, which most certainly he does not do. Nor would increasing the distance at which we assume the star to lie have any effect to save the sun from being thus perturbed, but the reverse. If we double for instance our estimate of the nebula’s distance, we increase eightfold our estimate of its mass, while we only diminish its attraction on our sun fourfold on account of increased distance; so that now its attraction on our sun would be one-fourth its former134 attraction multiplied by eight, or twice our former estimate. We cannot suppose the nebula to be much nearer than the nearest star. Again, we cannot suppose that the light of these gaseous nebul? comes from some bright orb within them of only starlike apparent dimensions, for in that case we should constantly recognize such starlike nucleus, which is not the case. Moreover, the bright-line spectrum from one of these nebul? comes from the whole nebula, as is proved by the fact that if the slit of the spectroscope be opened it becomes possible to see three spectroscopic images of the nebula itself, not merely the three bright lines.

So that, if we assume the so-called star in Cygnus to be now like other objects giving the same monochromatic spectrum—and this seems the only legitimate assumption—we are compelled to believe that the light now reaching us comes from a nebulous mass, not from the faintly luminous envelope of a dead sun. Yet, remembering that when at its brightest this orb gave a spectrum resembling in general characteristics that of other stars or suns, and closely resembling even in details that of stars like Gamma Cassiopei?, we are compelled by parity of reasoning to infer that when the so-called new star was so shining, the greater part of its light came from a sunlike mass. Thus, then, we are led to the conclusion that in the case of this body we have a nucleus or central mass, and that around this central mass there is a quantity of gaseous matter, resembling in constitution that which forms the bulk of the other gaseous nebul?. The denser nucleus ordinarily shines with so faint a lustre that the continuous spectrum from its light is too faint to be discerned with the same spectroscopic means by which the bright lines of the gaseous portion are shown; and the gaseous portion ordinarily shines with so faint a lustre that its bright lines would not be discernible on the continuous background of a stellar spectrum. Through some cause unknown—possibly (as suggested in an article on the earlier history of this same star in my “Myths and Marvels of Astronomy”) the rush of a rich135 and dense flight of meteors upon the central mass—the nucleus was roused to a degree of heat far surpassing its ordinary temperature. Thus for a time it glowed as a sun. At the same time the denser central portions of the nebulous matter were also aroused to intenser heat, and the bright lines which ordinarily (and certainly at present) would not stand out bright against the rainbow-tinted background of a stellar spectrum, showed brightly upon the continuous spectrum of the new star. Then as the rush of meteors upon the nucleus and on the surrounding nebulous matter ceased—if that be the true explanation of the orb’s accession of lustre—or as the cause of the increase of brightness, whatever that cause may have been, ceased to act, the central orb slowly returned to its usual temperature, the nebulous matter also cooling, the continuous spectrum slowly fading out, the denser parts of the nebulous matter exercising also a selective absorption (explaining the bands seen in the spectrum at this stage) which gradually became a continuous absorption—that is, affected the entire spectrum. Those component gases, also, of the nebulous portion which had for a while been excited to sufficient heat to show their bright lines, cooled until their lines disappeared, and none remained visible except for a while the three usual nebular lines, and latterly (owing to still further cooling) only the single line corresponding to the monochromatic light of the fainter gaseous nebul?.