ASTRONOMY is usually thought of as the study of the bodies visible in the sky. And such it largely is when the present state of the universe alone is considered. But when we attempt to peer into its past and to foresee its future, we find ourselves facing a new side of the heavens—the contemplation of the invisible there. For in the evolution of worlds not simply must the processes be followed by the mind’s eye, so short the span of human life, but they begin and end in what we cannot see. What the solar system sprang from, and what it will eventually become, is alike matter devoid of light. Out of darkness into darkness again: such are the bourns of cosmic action.

The stars are suns; past, present, or potential. Each of those diamond points we mark studding the heavens on a winter’s night are globes comparable with, and in many cases greatly excelling, our own ruler of the day. The telescope discloses myriads more. Yet these self-confessed denizens of space form but a fraction of its occupants. Quite as near, and perhaps much nearer, are orbs of which most of us have no suspicion. Unimpressing our senses and therefore ignored by our minds, bodies people it which, except for rare occurrences, remain forever invisible. For dark stars in countless numbers course hither and thither throughout the universe at speeds as stupendous as the lucent ones themselves.

Had we no other knowledge of them, reasoning would suffice to demonstrate their existence. It is the logic of unlimited subtraction. Every self-shining star is continually giving out light and heat. Now such an expenditure cannot go on forever, as the source of its replenishing by contraction, accretion, or disintegration is finite. Long to our measures of time as the process may last, it must eventually have an end and the star finally become a cold dark body, pursuing as before its course, but in itself inert and dead; an orb grown orbÉd, in the old French sense. So it must remain unless some cosmic catastrophe rekindle it to life. The chance of such occurrence in a given time compared with the duration of the star’s light-emitting career will determine the number of dark stars relative to the lucent ones. The chance is undoubtedly small, and the number of dark bodies in space proportionally large. Reasoning, then, informs us first that such bodies must exist all about us, and second that their multitude must be great.

Valid as this reasoning is, however, we are not left to inference for our knowledge of them. There is a certain star amid the polar constellations known as Algol,—el Ghoul, the Arabs called it, or The DÆmon. The name shows they noticed how it winked its eye and recognized something sarcastically sinister in its intent. For once in two days and twenty hours its light fades to one-third of its usual amount, remains thus for about twenty minutes, and then slowly regains its brightness. Seemingly unmoved itself, its steady blinking from the time man first observed it took on an uncanniness he felt. To untelescoped man it certainly seemed demoniacal, this punctual recurrent wink. Spectroscoped man has learnt its cause.

Goodricke in 1795 divined it, and research since has confirmed his keen intuition. Its loss of light is occasioned by the passing in front of it of a dark companion almost of its own size revolving about it in a close elliptic orbit. That this is the explanation of its strange behavior, the shift of its spectral lines makes certain, by showing that the bright star is receding from us at twenty-seven miles a second seventeen hours before the eclipse and coming towards us at about the same rate seventeen hours after it; its dark companion, therefore, doing the reverse.

Algol is no solitary specimen of a mind-seen invisible star. Many eclipsing binaries of the same class are now known; and considering that the phenomenon could not be disclosed unless the orbital plane of the pair traversed the observer’s eye, an unlikely chance in a fortuitous distribution, we perceive how many such in truth there must be which escape recognition for their tilt.

But if dark stars exist in connection with lucent ones, there must be many more that travel alone. Our own Sun is an instance in embryo. If he live long enough, he will become such a solitary shrouded tramp in his old age. For he has no companion to betray him. The only way in which we could become cognizant of these wanderers would be by their chance collision with some other star, dark or lucent as the case might be. The impact of the catastrophe would generate so much light and heat that the previously dark body would be converted into a blazing sun and a new star make its advent in the sky.

Star births of the sort have actually been noted. Every now and then a new star suddenly appears in the firmament—a nova as it is technically called. These apparitions date from the dawn of astronomic history. The earliest chronicled is found in the Chinese Annals of 134 b.c. It shone out in Scorpio and was probably the new star which Pliny tells us incited Hipparchus, “The Father of Astronomy,” to make his celebrated catalogue of stars. From this time down we have recorded instances of like character.

One of the most famous was the “Pilgrim Star” of Tycho Brahe. That astronomer has left us a full account of it. “While I was living,” he tells us, “with my uncle in the monastery of Hearitzwadt, on quitting my chemical laboratory one evening, I raised my eyes to the well-known vault of heaven and observed, with indescribable astonishment, near the zenith, in Cassiopeia, a radiant fixed star of a magnitude never before seen. In my amazement I doubted the evidence of my senses. However, to convince myself that it was no illusion, and to have the testimony of others, I summoned my assistants from the laboratory and inquired of them, and of all the country people that passed by, if they also observed the star that had thus suddenly burst forth. I subsequently heard that in Germany wagoners and other common people first called the attention of astronomers to this great phenomenon in the heavens,—a circumstance which, as in the case of non-predicted comets, furnished fresh occasion for the usual raillery at the expense of the learned.”

The new star, he informs us, was just like all other fixed stars, but as bright as Venus at her brightest. Those gifted with keen sight could discern it in the daytime and even at noon. It soon began to wane. In December, 1572, it resembled Jupiter, and a year and three months later had sunk beyond recognition to the naked eye. It changed color as it did so, passing from white through yellow to red. In May, 1573, it returned to yellow (“the hue of Saturn,” he expressly states), and so remained till it disappeared from sight, scintillating strongly in proportion to its faintness.

Thirty-two years later another stranger appeared and was seen by Kepler, who wrote a paper about it entitled “The New Star in the Foot of the Serpent.” It shone out in the same sudden manner and faded in the same leisurely way.

Since 1860 there have been several such apparitions, and since 1876 it has been possible to study them with the spectroscope, which has immensely increased our knowledge of their constitution. Indeed, this instrument of research has really opened our eyes to what they are. Nova Cygni, in 1876, Nova AurigÆ, in 1892, and Nova Persei, in 1901, besides several others found by Mrs. Fleming on the Arequipa plates, were excellent examples, and all agreed in their main features, showing that novÆ constitute a type of stars by themselves, whose appearing in the first place and whose behavior afterwards prove them to have started from like cause and to have pursued parallel lines of development.

As a typical case we may review the history of Nova AurigÆ. On February 1, 1892, an anonymous post-card was received by Dr. Copeland of the Royal Observatory, Edinburgh, that read as follows: “Nova in AurigÆ. In Milky Way, about 2° south of ? AurigÆ, preceding 26 AurigÆ. Fifth magnitude slightly brighter than ?.” The observatory staff at once looked for the nova and easily found it with an opera glass. They then examined it through a prism placed before their 24-inch reflector and found its spectrum. It proved to be that of a “blaze star.”

Dr. Thomas D. Anderson turned out to be the writer of the anonymous post-card—his name modestly self-obliterated by the nova’s light. He had detected the star on January 24, but had only verified it as a new one on the 31st. Harvard College Observatory then looked up its archived plates. The plates showed that it had appeared sometime between December 1 and 10. Its maximum had been attained on December 20, after which it declined, to record apparently another maximum on February 3 of the 3.5 magnitude. From this time its light steadily waned till on April 1 it was only of the 16th magnitude or ¹/100000 of what it had been. In August it brightened again and then waned once more.

Meanwhile its spectrum underwent equally strange fluctuations. At first it exhibited the bright lines characteristic of the flaming red solar prominences, the calcium, hydrogen, and helium lines flanked by their dark correlatives upon a continuous background, showing that both glowing and cooler gases were here concerned. The sodium lines, too, appeared, like those that come out in comets as they approach the furnace of the Sun. An outburst such as occurs in miniature in the solar chromosphere or outermost gaseous layer of the Sun was here going on upon a gigantic scale. A veritable spectral chaos next supervened, staying until the star had practically faded away. Then, on its reappearance, in August, Holden, Schaeberle, and Campbell discovered to their surprise not what had been at all, but something utterly new: the soberly bright lines only of a nebula. Finally, ten years later, January, 1902, Campbell found its spectrum had become continuous, the body having reverted to the condition of a star.

Now how are we to interpret these grandiose vicissitudes, visually and spectrally revealed? That we witnessed some great catastrophe is clear. The sudden increase of light of many thousand fold from invisibility to prominence shows that a tremendous cataclysm occurred. The bright lines in the spectrum confirm it and imply that vast upheavals like those that shake the Sun were there in progress, but on so stupendous a scale that, if for no other reason, we must dismiss the idea that explosions alone can possibly be concerned. The dark correlatives of the bright lines have been interpreted as indicating that two bodies were concerned, each travelling at velocities of hundreds of miles a second. But in Nova AurigÆ shiftings of the spectral lines implying six bodies at least were recorded, if such be attributed to motion in the line of sight, and Vogel was minded to throw in a few planets as well—as Miss Clerke pithily puts it. There is not room for so many on the stage of the cosmic drama. Other causes, as we now know, may also displace the spectral lines. Great pressure has been shown to do it, thanks to the labors of Humphreys and Mohler at Baltimore. “Anomalous refraction” may do it, as Professor Julius of Utrecht has found out. Finally, changes of density may produce it, as Michelson has discovered. To these causes we may confidently ascribe most of the shiftings in the stellar spectrum, for just such forces must be there at work.

Mr. Monck suggested the idea that new stars are the result of old dark stars rushing through gaseous fields in space and rendered luminous by the encounter. Seeliger revived and developed this idea, which in certain cases is undoubtedly the truth. Probably this occurred to the new star of 1885 which suddenly blazed out almost in the centre of the great nebula in Andromeda. It behaved like a typical nova and in due course faded to indistinguishability. Something like it happened, too, in the nova of 1860, which suddenly flared up in the star cluster 80 Messier, outdoing in lustre the cluster itself, and then, too, faded away.

But just as psychology teaches us that not only do we cry because we are sorrowful, but that we are sorrowful because we cry, so while a nova may be made by a nebula, no less may a nebula be made by a star.

Let us see how this might be brought about and what sign manuals it would present. Suppose that the two bodies actually grazed. Then the disruption would affect the star’s cuticle, first raising the outer parts, consisting rather of carbon than of the metals, since that substance is the lighter, to intense heat and the gases about it at the same time. The glowing carbon would be intensely bright, and at first its light would overpower that from the gases, and not till its great glow had partially subsided would theirs be seen. Then the gases, hydrogen, helium, and so forth, would make themselves evident. Finally only the most tenuous ones, those peculiar to a nebula, would remain visible. After which the more solid particles due to the disruption would fall together and light up again by their individual collisions. Much the same would result if without striking the stars passed close.

Now to put this theory to the proof. In the early morning of the 22d of February, 1901, Dr. Anderson, the discoverer of Nova AurigÆ, perceived that Algol had a neighbor, a star as bright as itself, which had never been there before. Within twenty-four hours of its detection the newcomer rivalled Capella, and shortly after took rank as the premier star of the northern hemisphere. Its spectrum on the 22d was found at Harvard College Observatory to be like that of Rigel, a continuous one crossed by some thirty faint dark lines. On the 24th, however, so soon as it began to wane, the bright lines of hydrogen were conspicuous with their dark correlatives, just as they had been with Nova AurigÆ and other novÆ. At the same time each particular spectral line proved a law unto itself, some shifted more than others, thus negativing motion as their only cause and indicating change of pressure or density as concerned concomitants of the affair. Blue emissions like those of Wolf-Rayet stars next made their appearance; then a band, found by Wright at the Lick to characterize nebulÆ, shone out, and finally in July the change to a nebular spectrum stood complete.

Then came what is the most suggestive feature in the whole event. On August 22 and 23 Dr. Wolf at KÖnigstahl took with his then new Bruce objective some long exposure plates of the nova, and on them found, to his surprise, wisps of nebulous matter to the southeast of the star. On September 20 Ritchey, with a two-foot mirror of his own constructing exposed for four hours, brought the whole formation to light. It turned out to be a spiral nebula encircling and apparently emanating from the star. Its connection with the nova was patent. But there was more to come. Later plates taken at the Lick on November 7 disclosed the startling fact that the nebula was visibly expanding, uncoiling outward from the star. A plate by Ritchey on November 13 confirmed this, and still later plates by him in December, January, and February showed the motion to be progressive. At the same time the star showed no parallax, and the speed of the motion seemed thus to be indicated as enormous. Kapteyn suggested to account for it that appearance, not reality, was here concerned; that the nebula had always existed, and was only shown up by the light from the conflagration travelling outward from the nova at the rate of one hundred and eighty-six thousand miles a second. This would make the catastrophe to have occurred as far back as the time of James I, of which the news more truthful but less timely than that of the morning papers had only just reached us.

But a little of that simple reasoning by which Zadig recovered the lost horses of the Sultan, and which from its unaccustomedness in the affairs of men got him suspected of having stolen them and very nearly caused his death, will show the untenableness of this idea and help us to a solution. In the first place we note that the star holds the very centre of the nebular stage, a remarkable prominence if the star has no creative right to the position. Then the same knots and patches of the nebulous configuration are visible in all the photographs, in the same relative positions, turned through corresponding angles as one will see for himself, all having moved symmetrically from one date to another. At the truly marvellous mimicry implied if different objects were concerned common sense instinctively shies, and very properly, as the chances against it are millions to one. Clearly it was not a mere matter of ethereal motion, but a very material motion of matter, which was here concerned. Something corpuscular emanating from the nova spread outward into space.

Clinching this conclusion is the result of a search by Perrine for traces of the nebula on earlier plates. For on one taken by him on March 29 (1901) he found the process already started in two close coils, its conception thus clearly dating from the time of the star’s outburst. In Nova Persei, then, we actually witnessed a spiral nebula evolved from a disrupted star.

What was this ejectum and what drove it forth? Professor Very regarded it as composed of corpuscles such as give rise to cathode rays discharged from the star under the stress of light pressure or electric repulsion. But I think we may see in it something simpler still; to wit, gaseous molecules driven off by light pressure alone—the smoke, as one may say, of the catastrophe—akin exactly to the constituents of comet’s tails. The mere light of the conflagration pushed the hydrogen molecules away. This would explain their presence and their exceeding hurry at the same time. They were started on their travels by domestic jars and kept going by the vivid after-effects of that infelicity.

The fairly steady rate of regression from the nova observed may be explained by the observed decrease in the light of the repellent source. Such combined with the retarding effect of gravity might make the regression equable. This is the more explanatory as the speed was certainly much less than that of light, though greatly exceeding any possible from the direct disruption. At the same time both the bright and the dark lines of hydrogen seen in the spectrum stand accounted for; the colliding molecules, at their starting on their travels from the star, shining through their sparser fellows farther out. An interesting biograph of the levity of light!

Nova Persei thus introduces us at its birth to one of a class of most interesting objects comparatively recently discovered and of most pregnant import,—the spiral nebulÆ.

In 1843 when Lord Rosse’s giant speculum, six feet across, was turned upon the sky, a nebula was brought to light which was unlike any ever before seen. It was neither irregular like the great nebula in Orion nor round like the so called planetary nebulÆ,—the two great classes at that time known,—but exhibited a striking spiral structure. It proved the forerunner of a remarkable revelation. For the specimen thus disclosed has turned out to typify not only the most interesting form of those heavenly wreaths of light, but by far the commonest as well. As telescopic and especially photographic means improved, the number of such objects detected steadily increased until about thirteen years ago Keeler by his systematic discoveries of them came to the conclusion that a spiral structure pervaded the great majority of all the nebulÆ visible. Their relative universality was outdone only by the invariability of their form. For they all represent spirals of one type: two coiled arms radiating diametrically from a central nucleus and dilating outward. Even nebulÆ not originally supposed spiral have disclosed on better revelation the dominant form. Thus the great nebula in Andromeda formerly thought lens-shaped proves to be a huge spiral coiled in a plane not many degrees inclined to the plane of sight.

As should happen if the spirals are unrelated, left-handed and right-handed ones are about equally common. In Dr. Roberts’ great collection of those in which the structure is distinctly discernible, nine are right-handed, ten left-handed, showing that they partake of the ambidextrous impartiality of space.

Lastly the spirals are evidently thicker near the centre, thinning out at the edge, and when the central nucleus is pronounced, it seems to have a certain globularity not shared by the arms, and more or less detached from them. This appears in those cases where they are shown us edgewise, and it has been thought perceptible in the great nebula of Andromeda. The difficulty in establishing the phenomenon comes from the impossibility of both features showing at their best together. For the globularity to come out well, the spiral must be presented to us nearly in the plane of sight; for the spirality, in a plane at right angles to it.

Much may be learnt by pondering on these peculiarities. The widespread character of the phenomenon points to some universal law. We are here clearly confronted by the embodiment of a great cosmic principle, causing the helices it is for us to uncoil. It is a problem in mechanics.

In the first place, a spiral structure denotes action on the face of it. It implies a rotation combined with motion out or in. We are familiar with the fact in the sparks of pin-wheel pyrotechnics. Any rotating fluid urged by an outward or an inward impulse must take the spiral form. A common example occurs in the water let out of a basin through a hole in the centre when we draw out the plug. Here the force is inward, and because the bowl and orifice are not perfectly symmetric, a rotation is set up in the water trying to escape, and the two combine to give us a beautiful conchoidal swirl. In this case the particles seek the centre, but the same general shape is assumed when they seek to leave it.

Another point to be noticed is that a spiral nebula could not develop of itself and subsist. To continue it must have outside help. For if it were due to internal explosive action in the pristine body, each ejectum must return to the point it started from, or else depart forever into space, for the orbit it would describe must either be closed or unclosed. If the former, it would revisit its starting-point; if the latter, it would never return. Explosion, therefore, of itself could not have produced the forms we see, unless they be ephemeral apparitions, a supposition their presence throughout the heavens seems effectually to exclude.

The form of the spiral nebulÆ proclaims their motion, but one of its particular features discloses more. For it implies the past cause which set this motion going. A distinctive detail of these spirals, which so far as we know is shared by all of them, are the two arms which leave the centre from diametrically opposite sides. This indicates that the outward driving force acted only in two places, the one the antipodes of the other. Now what kind of force is capable of this peculiar effect? If we think of the matter, we shall realize that tidal action would produce just this result. We see it daily in the case of the Moon; when it is high tide in the open ocean hereabouts, it is high tide also at the opposite end of the Earth. The reason is that the tideraising body pulls the fluid nearest it more strongly than it pulls the Earth as a whole, and pulls the Earth as a whole more than it pulls the fluid at the opposite extremity.

Suppose, now, a stranger to approach a body in space near enough; it will inevitably raise tides in the other’s mass, and if the approach be very close, the tides will be so great as to tear the body in pieces along the line due to their action; that is, parts of the body will be separated from the main mass in two antipodal directions. This is precisely what we see in the spiral nebula. Nor is there any other action that we know of which would thus handle the body. If it were to disintegrate under increased speed of rotation due to contraction upon itself, parts of its periphery should be shed continually and a pin-wheel of matter, not a two-armed spiral, be thrown off. If explosion were the disintegrating cause, disruption would occur unsymmetrically in one or more directions, not symmetrically as here.

As the stranger passed on, his effect would diminish until his attraction no longer overbalanced that of the body for its disrupted portions. These might then be controlled and forced to move in elliptic orbits about the mass of which they had originally made part. Thence would come into being a solar system, the knots in the nebula going to form the planets that were to be.

Before proceeding to what proof we have that it actually did occur in this way we may pause to consider some consequences of what we have already learned. Thus what brought about the beginning of the system may also compass its end. If one random encounter took place in the past, a second is as likely to occur in the future. Another celestial body may any day run into the Sun, and it is to a dark body that we must look for such destruction, because they are so much more numerous in space.

That any of the lucent stars, the stars commonly so called, could collide with the Sun, or come near enough to amount to the same thing, is demonstrably impossible for Æons of years. But this is far from the case for a dark star. Such a body might well be within a hundredth of the distance of the nearest of our known neighbors, Alpha Centauri, at the present moment without our being aware of it at all. Our senses could only be cognizant of its proximity by the borrowed light it reflected from our own Sun. Dark in itself, our own head-lights alone would show it up when close upon us. It would loom out of the void thus suddenly before the crash.

We can calculate how much warning we should have of the coming catastrophe. The Sun with its retinue is speeding through space at the rate of eleven miles a second toward a point near the bright star Vega. Since the tramp would probably also be in motion with a speed comparable with our own, it might hit us coming from any point in space, the likelihood depending upon the direction and amount of its own speed. So that at the present moment such a body may be in any part of the sky. But the chances are greatest if it be coming from the direction toward which the sun is travelling, since it would then be approaching us head on. If it were travelling itself as fast as the Sun, its relative speed of approach would be twenty-two miles a second.

The previousness of the warning would depend upon the stranger’s size. The warning would be long according as the stranger was large. Let us assume it the mass of the Sun, a most probable supposition. Being dark, it must have cooled to a solid, and its density therefore be much greater than the Sun’s, probably something like eight times as great, giving it a diameter about half his or four hundred and thirty thousand miles. Its apparent brightness would depend both upon its distance and upon its intrinsic brightness or albedo, and this last would itself vary according to its distance from the Sun. While it was still in the depths of space and its atmosphere lay inert, owing to the cold there, its intrinsic brightness might be that of the Moon or Mercury. As its own rotation would greatly affect the speed with which its sunward side was warmed, we can form no exact idea of the law of its increase in light. That the augmentation would be great we see from the behavior of comets as they approach the great hearth of our solar system. But we are not called upon to evaluate the question to that nicety. We shall assume, therefore, that its brilliancy would be only that of the Moon, remembering that the last stages of its fateful journey would be much more resplendently set off.

With these data we can find how long it would be visible before the collision occurred. As a very small telescopic star it would undoubtedly escape detection. It is not likely that the stranger would be noticed simply from its appearance until it had attained the eleventh magnitude. It would then be one hundred and forty-nine astronomical units from the Sun or at five times the distance of Neptune. But its detection would come about not through the eye of the body, but through the eye of the mind. Long before it could have attracted man’s attention to itself directly its effects would have betrayed it. Previous, indeed, to its possible showing in any telescope the behavior of the outer planets of the system would have revealed its presence. The far plummet of man’s analysis would have sounded the cause of their disturbance and pointed out the point from which that disturbance came. Celestial mechanics would have foretold, as once the discovery of another planet, so now the end of the world. Unexplained perturbations in the motions of the planets, the far tremors of its coming, would have spoken to astronomers as the first heralding of the stranger and of the destruction it was about to bring. Neptune and Uranus would begin to deviate from their prescribed paths in a manner not to be accounted for except by the action of some new force. Their perturbations would resemble those caused by an unknown exterior planet, but with this difference that the period of the disturbance would be exactly that of the disturbed planet’s own period of revolution round the Sun.

Our exterior sentinels might fail thus to give us warning of the foreign body because of being at the time in the opposite parts of their orbits. We should then be first apprised of its coming by Saturn, which would give us less prefatory notice.

It would be some twenty-seven years from the time it entered the range of vision of our present telescopes before it rose to that of the unarmed eye. It would then have reached forty-nine astronomical units’ distance, or two-thirds as far again as Neptune. From here, however, its approach would be more rapid. Humanity by this time would have been made acquainted with its sinister intent from astronomic calculation, and would watch its slow gaining in conspicuousness with ever growing alarm. During the next three years it would have ominously increased to a first magnitude star, and two years and three months more have reached the distance of Jupiter and surpassed by far in lustre Venus at her brightest.

Meanwhile the disturbance occasioned not simply in the outer planets but in our own Earth would have become very alarming indeed. The seasons would have been already greatly changed, and the year itself lengthened, and all these changes fraught with danger to everything upon the Earth’s face would momentarily grow worse. In one hundred and forty-five days from the time it passed the distance of Jupiter it would reach the distance of the Earth. Coming from Vega, it would not hit the Earth or any of the outer planets, as the Sun’s way is inclined to the planetary planes by some sixty degrees, but the effects would be none the less marked for that. Day and night alone of our astronomic relations would remain. It would be like going mad and yet remaining conscious of the fact. Instead of following the Sun we should now in whole or part, according to the direction of its approach, obey the stranger. For nineteen more days this frightful chaos would continue; as like some comet glorified a thousand fold the tramp dropped silently upon the Sun. Toward the close of the nineteenth day the catastrophe would occur, and almost in merciful deliverance from the already chaotic cataclysm and the yet greater horror of its contemplation, we should know no more.

Unless the universe is otherwise articulated than we have reason to suppose, such a catastrophe sometime seems certain. But we may bear ourselves with equanimity in its prospect for two mitigating details. One is that there is no sign whatever at the moment that any such stranger is near. The unaccounted-for errors in the planetary theories are not such as point to the advent of any tramp. Another is, that judged by any scale of time we know, the chance of such occurrence is immeasurably remote. Not only may each of us rest content in the thought that he will die from causes of his own choosing or neglect, but the Earth herself will cease to be a possible abode of life, and even the Sun will have become cold and dark and dead so long before that day arrives that when the final shock shall come, it will be quite ready for another resurrection.