The Limits of the Stellar Universe

WHAT is the number of the stars, and their distribution in space ? The extent and probable arrangement of the luminous bodies which compose the stellar universe ?

Astronomers have been studying these questions since the days of Galileo; and the general conclusions which some of the greatest observers and thinkers have deduced from data gathered during the past century may not be devoid of interest to the readers of popular scientific literature.

To grasp the problem satisfactorily, we should recall that the exploration of the heavens to date represents three historical stages : (1.) The naked-eye study of the sky, which comprises the observations and speculations of the Greeks and Romans, and of modern philosophers who lived prior to the year 1610. (2.) The telescopic study of the heavens since the days of Galileo, especially augmented during the nineteenth century. (3.) The photography of the celestial sphere, developed wholly within the last fifty years.

Great telescopes and the applications of photography have recently given the astronomer enormous power in gathering observational data ; and if the entire sky were surveyed, it is estimated that he might now perceive with the eye or on the photographic plate about 100,000,000 stars. Accordingly, we shall briefly trace the steps involved in this unparalleled development, and in the end point out some of the most remarkable discoveries yet wrought by the human mind.

When we look at the heavens on a cloudless and moonless night, we get the impression that the stars visible to the naked eye are numbered by tens of thousands. The fact is that the number of points of light actually noted by direct vision is much smaller than most persons suppose. For oblique vision, such as we obtain by a mere motion of the eye, enables us to see stars as faint as the seventh magnitude ; and thus we glimpse more objects than we can locate by direct searching. Unless the atmosphere is rare and very clear, it is difficult for the average person to see stars fainter than the sixth magnitude.

This view of the case is confirmed by several circumstances. In the first place, the catalogues of Hipparchus, Ptolemy, Al Sûfi, and Ulugh Beigh, all formed before the invention of the telescope, contained objects as faint as the sixth magnitude, and therefore, presumably, the greater part of the stars which can be easily located without a telescope ; and none of these catalogues contained much over 1000 stars. Moreover, Argelander, who made a critical study of the brightness of all the stars north of 36° south declination, concluded that there are in this region 3256 stars visible to the naked eye. Allowing 844 stars for the remaining canopy near the south pole, we see that in the whole celestial sphere the total number of stars recognizable by normal vision is about 4100. Of this number of stars visible to the naked eye in the entire heavens, probably not more than 2000 could be observed at one time, except perhaps by indirect vision, in a dry climate, where the atmosphere is excessively clear and transparent. For only half of the starry sphere is visible at any given instant, and the regions near the horizon are obscured by the density of the atmosphere, and the fainter stars are thus cut off. Nevertheless, it is perhaps not unreasonable to suppose that if a person be endowed with extraordinary power of vision (such a person usually has an alert mind and other senses as keen as his sight) he might see at one time as many as 4000 stars, so that to him 8000 such lucid points would appear in the entire heavens.

Prior to the invention of the telescope, opinion relative to the constitution of the heavens was necessarily of an indefinite character. The brightest portion of the starry sphere, known as the Milky Way, has always borne a name similar to that which we use to-day. The Greeks called the stupendous arch of light which spans the heavens the Galaxias, while the Romans named it the Via Lactea; and we may infer that all civilized nations have understood it to be due to uncounted multitudes of stars, too small and too dense to be seen individually. This opinion, indeed, was expressed by several ancient philosophers, but the doctrine first became established through the discovery of the telescope by Galileo in 1610. We can hardly realize what a revolution of opinion these early telescopic discoveries wrought; their effect was like lifting a curtain from things which had been hidden from mortal eye for thousands of years. Galileo’s announcement of his discoveries is not devoid of interest; the wording of his message shows how these revelations impressed that great philosopher. He says : " It is truly a wonderful fact that to the vast number of fixed stars which the eye perceives, an innumerable multitude, before unseen, and exceeding more than tenfold those hitherto known, have been rendered discernible. Nor can it be regarded as a matter of small moment that all disputes respecting the nature of the Milky Way have been brought to a close, and the nature of the zone made manifest not to the intellect only, but to the senses.” After Galileo and his successors had dissolved the cloud forms of the Milky Way, and shown them to consist of uncounted thousands of stars, speculation relative to the distribution of the stars in space naturally began to develop. We can here mention only a few of the more prominent of these speculations.

Kepler, who was a contemporary of Galileo, followed Copernicus in placing the sun in the centre of the universe, and assumed that an equal number of stars are distributed in successive equidistant spheres ; the first sphere he assumed to contain twelve stars, the next twelve more, and so on. By this arrangement, the body of stars would soon become so remote that they would cease to shine from mere faintness of their light. And as Kepler foresaw that in some regions of the heavens stars of equal brightness are denser than this theory required, he surmised that they are in reality much closer to one another than they are to the sun. This line of thought was of course largely arbitrary, and could not well stand thorough analysis.

Numbers like 12, used by Kepler, were obviously fixed upon from mystical considerations, which so frequently appear in the writings of this extraordinary man. In spite of this mysticism, however, he saw the significance of the arrangement of the Milky Way, and suggested that our sun is near the centre of the great band which encircles the sky. Nor did he fail to place the stars at an immense distance, where they would appear as points, and exhibit no measurable parallax. Yet he was inclined to reconcile his novel conceptions with the old theories of crystalline spheres, and even to find beyond these spheres the firmament and waters of the Pentateuch. It is supposed that these last views were accommodations which he thought to be in the interest of science, at a time when most astronomers rejected the Copernican system as subversive of ancient doctrines.

The views of Huyghens, given in his Cosmotheoros, indicate a full realization that our sun is an ordinary fixed star, and the opinion is put forward that other stars are centres of planetary systems similar to our own. Despairing of ever measuring the distance of the fixed stars by direct processes, this acute philosopher proposed an indirect photometric process, by which the light of the stars may be compared to that of our sun experimentally, and their distances deduced on certain hypotheses. Applying his method to Sirius, he found that this brilliant object is 28,000 times the distance of the sun, which is now known to be about one eighteenth part of the actual distance. Huyghens assumed that Sirius and the sun give the same amount of light; but as modern research shows that Sirius is some sixty times the more brilliant of the two objects, his substitution of a body having the same luminosity as the sun, at one eighteenth of the distance of the more brilliant body, is equivalent to placing Sirius at one eighth of its actual distance, which must be considered a very remarkable approximation for a Dutch philosopher of the seventeenth century.

When we come down to Kant, we meet with a philosopher who outlined many of the grand theories held to-day. In the introduction to his Natural History and Theory of the Heavens, published in 1755, he proposes “to discover the arrangement which connects the grand parts of creation, in all their infinite extent, and to deduce by the aid of mechanical laws the formation of the celestial bodies, and the origin of their motions, from the primitive condition of nature.”

As Kant beheld the stupendous arch of the Milky Way, and noticed how dense the stars appear to be near its centre, and how they fade away with increasing distance from its fundamental plane, he rightly concluded that the stars extend much farther in that direction than in the direction perpendicular to the plane of the Galactic circle. He attributed this interpretation of the universe to Thomas Wright, of Durham, England, whose view is sometimes called the “ grindstone theory.” Our sun is supposed to be near the centre, and when we look edgewise along the stratum we see the immense number of stars in the plane of the Milky Way ; but when we examine the regions remote from this plane we find very few stars. This thin stratum into which the stars are crowded suggested to Kant a resemblance to the solar system, in which all the planets are confined near a fundamental plane.

In our time, the resemblance is made still more striking on account of the swarm of nearly 500 Asteroids discovered during the nineteenth century. Led on by this analogy, the immortal philosopher concluded that the stars too are confined to the fundamental plane of the Milky Way, and are moving in orbits under the attraction of some great central body, which he conjectured may be the dog star Sirius. As the distance of the stars was known to be immense, it was held that motion about a centre would necessarily be very slow, and hence he concluded that, though previously undetected, such motion must eventually come to light. Halley’s recent researches had established the existence of proper motion for only a few of the brighter stars, and to dispute on observational grounds the truth of these grand schemes of creation was then impossible. At the present day, when hundreds and even thousands of proper motions are known, from researches made during the nineteenth century, and the stars are found, not to be tracing orbits along the Milky Way, but darting indiscriminately in all directions, except that a tendency to widen out appears in the region of Hercules, and a closing up at the opposite point, which is attributed to the secular motion of the whole solar system, we may unhesitatingly affirm that Kant’s grand system is not in accord with actual nature. In order to insure stability of his system, he assumed that the stars are controlled by a central mass ; but as not even the planetary system, in its admirable symmetry and harmony, can claim such a quality, it has long been conceded by astronomers that the sidereal system is not eternal.

With the flight of ages the majestic arch constituting the Milky Way will gradually undergo permanent changes, owing to the continued action of the clustering power first noticed by the elder Herschel. This curdling tendency has already given the Milky Way the appearance of a vast aggregation of clusters rather than that of a continuous band of uniform light, and in time it must entirely alter the aspect of the Galaxy, and leave nothing but individual groups of stars, with little of the continuous appearance now observed.

Though Kant supposed all the stars which compose the Galaxy and stud our heavens to belong to one gigantic system dominated by the mutual gravitation of its parts, he did not suppose this to be the whole universe, but saw in the nebulæ other systems, so immensely remote that the combined light of their millions of stars merely made the impression of a faint cloud even when viewed with the telescope. This somewhat grandiose view has been little credited since the time of Laplace and Sir William Herschel; and since the invention of the spectroscope, about forty years ago, has been directly disproved, as many of the nebulæ were thus shown by the bright lines in their spectra to be masses of self-luminous gas, though it is not yet quite clear how their light is maintained at the low temperature of the celestial spaces.

The speculations of Lambert, published a few years after those of Kant, supposed the universe to be arranged in systems of different orders. Satellites moving about planets constitute the most elementary of these systems ; the planets revolving round the sun and the similar bodies attending the fixed stars, the next in order ; and as the planetary system carries with it sub-systems of satellites, so the planetary system itself is a subsystem to a star cluster, and a grand arrangement of star clusters makes up the Galaxy.

The central bodies in the solar system have masses which are predominant, and a like supremacy is ascribed to the central bodies elsewhere ; but as these bodies cannot be seen among the stars, Lambert is careful to suppose them to be opaque and dark. All the systems are held to obey the law of gravitation. Unfortunately for this theory, there is no evidence that obscure bodies of immense mass, such as Lambert assumes, really exist ; and hence, while this grand scheme is not positively disproved by known facts, it has never been seriously debated by men of science. It is conceded that in all probability the heavens are literally filled with dark bodies of various sizes ; but they are assumed to be stars like our sun, and smaller bodies like the planets and comets. However numerous such dark stars may be, there is no reason to consider them larger than those stellar bodies which are still luminous; on the contrary, being already burnt out, there is some reason to think that these dark masses may be smaller than average stars.

The speculations of the English philosopher John Michell are chiefly important for the applications which he makes of the theory of probability. In this way he is led to conclude that groups of closely packed stars are connected into physical systems of binary, triple, quadruple, or multiple stars and clusters. These conjectures, advanced in 1767, have been amply confirmed by the experience of the past century, and now we have a large number of such known binary, or physical, systems in all parts of the sky.

None of the great astronomers who studied the structure of the universe left so profound an impression on human thought as did Sir William Herschel, to whom it was given to penetrate into remoter regions of creation than had ever before been unveiled to mortal eye. By many years of arduous labor he had gradually developed the powers of the reflecting telescope, and, after attaining this unprecedented instrumental means, set for himself the problem of exploring the structure of the heavens. By the plan of star-gauging he hoped to fathom the depths of the universe. He had a twenty-inch reflector, on which he used a magnifying power of 120, and this gave him a field of view one fourth as large as the disk of the moon; such a region, extending from the observer’s eye to infinity, includes all the stars within a solid conical space, increasing in volume as the cube of the distance of the observer from the base of the cone. Thus, if the stars are uniformly distributed in space, and his telescope penetrated twice as far as former instruments, he would see eight times as many stars as were known in that region before.

Now Herschel could not count all of the stars visible in the entire sky, and hence he contented himself with surveying a wide belt extending more than halfway round the celestial sphere, and counting the number of stars seen in some 3400 fields of view. This belt, chosen in the equatorial regions, was perpendicular to the Galaxy, and Herschel discussed the results of his gauges with respect to that plane. He found that the average number of stars in a gauge rapidly increased as he approached the Milky Way. His numbers are : —

The first zone, from 90° to 75° from Galaxy, averaged 4 stars per field,

The second zone, from 75° to 60° from Galaxy, averaged 5 stars per field,

The third zone, from 60° to 45° from Galaxy, averaged 8 stars per field,

The fourth zone, from 45° to 30° from Galaxy, averaged 14 stars per field,

The fifth zone, from 30° to 15° from Galaxy, averaged 24 stars per field,

The sixth zone, from 15° to 0° from Galaxy, averaged 53 stars per field.

From his survey in the southern hemisphere, Sir John Herschel found the star numbers in the corresponding zones to be 6, 7, 9, 13, 26, 59.

From these results we see that the elder Herschel easily satisfied himself that the universe is greatly extended in the direction of the Galaxy. If the doctrine of equal distribution were adhered to, one might be led to think that some regions — as, for example, the Pleiades, Præsepe, Coma Berenices, and other clusters — represent protuberances on the general body of the universe. This manifest absurdity was probably never entertained by Herschel; but as the stars increase steadily in approaching the Galactic plane, he believed the method of gauging to give a good representation of the actual universe.

The only rational explanation of a group of stars projected into a comparatively void region is that we have a genuine group or cluster of some kind. In the Milky Way we have a great number of such clusters or “ cloud forms,” as Professor Barnard calls them, with comparatively dark spaces between them. In this case, the conclusion is obvious that we have to deal with real aggregations of stars, and not merely with a region in which the bounds of the universe are more widely extended. In treating of the actual universe, we can assume neither that the stars have equal density in different regions, nor that they are of equal intrinsic lustre. Manifestly, they are very unequally distributed, and of all shades of brightness, from intense brilliancy to dull luminosity or actual obscurity.

Herschel’s ideas of the extent of the universe, like those of Kant, were much too grandiose. They represented what might be called the transcendental stage of astronomical science. He naturally reached the conclusion that the depths of space are unfathomable, even with such a sounding instrument as his fortyfoot reflector, which, however, did not materially alter the results at which he had arrived with his smaller telescopes.

Since the time of Herschel, one of the chief cultivators of the branch of science which treats of the structure of the heavens is William Struve, the illustrious observer of double stars. His conclusions relative to the distribution of the stars in space are founded mainly on the number of stars of the several magnitudes observed by Bessel in a zone extending fifteen degrees on either side of the equator. Struve proceeded by methods similar to those of Herschel, but reached some results of a materially different character. He found that if account be taken only of stars brighter than the fifth magnitude, they are no thicker in the Milky Way than elsewhere. Those of the sixth magnitude are relatively a little thicker, and those of the seventh yet thicker, and so on ; and as a result, very small stars are extremely dense in the Milky Way. But while the density of stars is great near the central plane, and a gradual thinning out occurs in receding from it, there is no definite limit to the stratum, but merely a gradual fading away. Nor are the stars in the central parts by any means equally distributed ; in some regions they are many times denser than in others. On account of probable diversity in intrinsic brightness, it is still quite impossible to say whether certain small stars which crowd the Milky Way in great numbers are in reality very remote, or whether they are fainter and smaller than average stars, and confined accordingly within the limits of the sidereal system.

The late astronomer Richard A. Proctor, who devoted considerable study to this question, found it necessary to abandon or greatly modify the highly artificial hypotheses underlying the speculations of Herschel. In general, he held that the structure of the Galaxy is that of a series of streams or spiral wisps of stars, many of which are distorted by projection ; and that our solar system is not directly connected with the Milky Way, as an observer may infer by the well-defined edges repeatedly found in different parts of the Galaxy. In 1886 Proctor said : “ The naked-eye appearance of the Milky Way is sufficient evidence on which to ground the belief that there is a distinct ring of matter out yonder in space, and that this ring is not flattened, as Sir John Herschel thought, but is (roughly speaking) of nearly circular section throughout its length.”

The most important recent investigation of the distribution of the stars relative to the Milky Way is that of Professor Seeliger, director of the Royal Observatory at Munich. This distinguished mathematician, whose labors have included almost every field of astronomy, has discussed nearly all the observations accumulated during the past century. He divides the whole celestial sphere into nine zones of twenty degrees each, all parallel to the medial plane of the Galaxy, thus making four zones on either side ; those near the pole being of course much contracted in area, while the one central zone includes the circuit of the Milky Way. Then, examining the hundreds of thousands of stars which have been catalogued, he deduces numbers representing their density in the several regions, which are as follows : 278, 303, 354, 532, 817, 607, 371, 321, 314. The density of 817 in the medial zone, 532 and 607 in the two zones next adjacent to the Milky Way, with the sensible uniformity in the zones nearer the poles of the Galaxy, shows conclusively that the universe is much more extended in the direction of the Milky Way than in the direction of its poles, as had in fact been long ago inferred by the researches of Herschel and Struve.

Professor Celoria, who recently succeeded the illustrious Schiaparelli as director of the Royal Observatory at Milan, has also confirmed, independently, the conclusions of Seeliger by a somewhat different process, based partly upon the density of stars catalogued, and partly upon counts of great multitudes of these objects still uncatalogued. Without dwelling longer upon these investigations, it may be asserted that the stellar universe is much flattened and relatively extended in the direction of the Galaxy.

Dr. B. A. Gould and Sir John Herschel inclined to the belief that the great canopy of brilliant stars in the southern hemisphere, with a centre in Lupus or the Southern Cross, represents one or more galaxies or groups of stars superposed upon a more remote galaxy. The only other way of accounting for this brilliant southern cluster, which Proctor explains essentially in accord with Gould’s views, is to suppose our sun very eccentrically situated in the Galactic circle. Gould seems inclined to the belief that the cluster of bright stars might include the sun ; and that our eccentric situation in this group of bright stars causes us to see some of them in all directions, but a far greater number than usual in the regions of the Cross, Lupus, Centaurus, and the Ship Argo. This view of the arrangement of the universe in relation to this group of bright stars corresponds entirely with the writer’s impression. It accounts for all the facts, and is inherently probable, both on geometrical principles and on the obvious nakedeye aspect of the lucid stars of the southern hemisphere.

It is not probable that any one of these theories represents the phenomena of nature perfectly; but before we can make a distinct advance over existing theories we must extend our photographic impressions over the entire Galaxy, and study the material thus furnished. Such research on the nature and extent of the Milky Way is well worthy of the attention of our great observatories, and until carried out with exhaustive care will remain an ultimate desideratum of science. As the problems to be dealt with are among the most stupendous which present themselves to the philosopher, so are they, on that account, all the more worthy of the most supreme effort of which the human mind is capable.

While connected with the Lick Observatory, from 1887 to 1895, Professor E. E. Barnard, now of the Yerkes Observatory, was led to attempt the photography of the Milky Way, with a portrait lens of short focus and wide angular aperture. The lens employed bore the name of Willard, and had been used many years before in a portrait gallery in San Francisco. The exquisite pictures of the Milky Way secured with this lens at Mount Hamilton have rendered it the most famous photographic lens in the world, and added a permanent lustre to the photographic art of America.

With an assiduity and a perseverance scarcely equaled by Herschel himself, Professor Barnard applied this lens to all parts of the Milky Way visible in this latitude, and obtained views of the Galaxy as much superior to those previously known as those of Herschel’s great telescope had been to the views of the older telescopes of the eighteenth century. But while Herschel’s improvements in telescopic power had been achieved by increasing the size of his mirrors, Barnard’s extraordinary achievements in celestial photography were the result of diminishing the size of his instrument and widening its field of view, so that the structure of the Milky Way might be depicted in the images of hundreds of thousands of stars registered on the photographic plates. The duration of exposure varied from one to four hours, according to the object photographed. It need hardly be pointed out that almost infinite labor and patience had to be expended in watching the telescope during these long sittings with the stars, to secure a sharp picture, unblemished by any accident to the diurnal motion of the telescope on which the Willard lens was mounted. The slightest hitch in the motion of the driving clock, the least jar of the telescope, even by a gust of wind, would ruin the picture. We can more easily imagine than describe the enthusiasm of the astronomer on finding a beautiful view of the structure of creation, after a tiresome vigil extending half through the night, and unrelieved by moving the eye from the sight wire of the finder during the whole exposure.

As a result of such indefatigable labor, Barnard depicted for the first time, on a splendid scale, the wonderful cloud forms of the Milky Way in Scorpius, Ophiuchus, Scutum Sobieski, Aquila, Cygnus, Cepheus, Andromeda, Perseus, and Monoceros. Millions upon millions of stars in diverse branches and streams, all intertwined with nebulosity, and the whole arranged in the form of an immense tree or branching cloud, with occasional dense clusters and some dark lanes exhibiting almost a total absence of stars, are the characteristic appearance of these pictures, which give us without doubt the most sublime views of creation ever yet witnessed by mortal eye. Impressive, luminous, majestic masses and streams of stars in uncounted millions set in the depths of immensity are unfolded to the mind, — a spectacle grand beyond conceiving!

The stupendous arch of light which spans the heavens is thus revealed in its true nature, — a multitude of clusters, streams, wisps, and swarms of stars, which confirm only too fully the suggestion of Herschel that the Milky Way is already breaking up under the continued action of a clustering power, and will some day shine as distinct clusters rather than as a continuous band of milky light. The photographs of Barnard show to the eye at a glance how immensely the bounds of the stellar universe must be extended in the direction of the Galaxy. Through them the mind obtains an insight into the arrangement of the stars such as the naked eye would afford if it had the sensitiveness of the photographic plate under a portrait lens, exposed for three continuous hours, or a full watch of the night. Few achievements of science in the century which has just closed may be considered more wonderful than that of celestial photography in affording a revelation of the real nature of the stellar universe.

With this insight into the arrangement of the universe, let us consider the distances of the fixed stars.

It is well known that the nearest of the fixed stars is Alpha Centauri, the brilliant southern binary, which is removed from us 275,000 times farther than the sun. One of the next nearest stars is Sirius, which is about 500,000 times the sun’s distance. These distances correspond to the spaces traversed by a ray of light in four and eight years, and hence we see these two brilliant stars as they shone four and eight years ago respectively. The smallest angular magnitude which can be certainly measured in the greatest modern telescope is five one-hundredths of a second of arc, and this corresponds to the parallax of a star at a distance of sixty light-years, or the angle subtended by a human hair, assumed to have a diameter of one thousandth of an inch, at a distance of 350 feet. Hence it follows that all stars removed from us by more than sixty light-years have parallaxes too small to be detected even by the most refined methods of modern research, and we can at best merely guess at their distances. As the nearest star, Alpha Centauri, is only four light-years distant, while some of the known stars are fifteen times as remote, it seems probable that all those stars which have a measurable parallax are very close to us, compared with the distances of the more remote objects. If we suppose the average star to be fifteen times as remote as those objects having the smallest measurable parallax, the average distance would be about 900 light-years. As this estimated distance is probably too small, it seems certain that the multitude of stars are removed from us by more than 1000 light-years, or 250 times the distance of Alpha Centauri. We may reckon that in all probability the most remote regions are ten times more distant from us than the nearest portions of the Galaxy, and hence that our telescopes probably penetrate regions lying so remote that light from the most distant objects visible would not reach us in less than 10,000 years. What we see in these border regions of the universe is, not the events now transpiring there, but phenomena as they were 10,000 years ago, or before the beginning of human history !

The rays that reach our eye from different portions of the sky thus started in different ages, and may be said to disclose different phases of the development of the universe ; those from the more remote regions representing ancient, those from the nearer portions more modern cosmical history. Even if all the creation began at the same time and progressed uniformly, our view of it would be altered by the time required for the propagation of light across the immense spaces by which we are removed from other portions of the universe. And as all the stars probably did not begin to develop at the same time, it is natural that we should see all stages of world development now going on in the heavens.

In this connection, mention may perhaps be made of a method proposed by the writer of these lines for measuring the distance of the Milky Way. It is founded upon the use of the major axis of the orbit of a double star for a base line, instead of the major axis of the earth’s orbit, which is too small for use in measuring stellar distances greater than sixty light-years. The length of the axis of the stellar orbit in question is determined with the spectroscope, in miles or kilometers ; and as the astronomer knows by micrometer measurement how large this space looks in the telescope, he can compute how far away the system is. It is thought that some day the distances of stars may be determined by this process, when removed from us by at least 1000 light-years. And when orbits for the double stars in the Milky Way have been determined, the method can be applied to find the distances of the clusters which compose that stupendous arch, so remote as to be forever immeasurable by every other process.

It is worthy of remark that if we imagine a sufficiently powerful, sensitive, and perfect set of eyes placed in a cluster of the Milky Way, at the distance of 5000 light-years, and directed toward the earth, the ethereal throbs falling upon them would reveal history as it was 5000 years ago ; and if these eyes should move toward the earth, they would witness all human history as it was enacted through the successive centuries. Thus the nature of terrestrial events is forever preserved and transmitted on through the ether of infinite space.

In this connection one naturally asks, Is the universe infinite ? To answer this question, we must first examine the nature of the problem which science has to deal with. Our only means of exploring the heavens is the combination of the eye and the photograph with the telescope and spectroscope. The rays of light which reach us from distant regions can alone inform us what is there, and a study of the phenomena revealed by the waves of ether can alone make known to us the nature of the universe. Compared to cosmical ages, the life of the individual, and even of the race, is very short, and wholly confined to the small space traversed by the earth during a few years or a few centuries. Thus the available sources of information are limited, and the difficulty of the problem is tremendous. In spite of this impediment, much study has been given to the subject, and results of no inconsiderable interest have been reached.

After Sir William Herschel had attempted to sound the depths of creation by his mighty telescopes, and found nothing but world on world, with no sign of an end of space, the first man to examine the problem more critically was the illustrious William Struve. The ether of the celestial spaces had been a subject of speculation from the earliest ages of science, and Struve asked the question whether this fluid might not absorb the light of stars in the most distant regions, and thus render them forever invisible to the inhabitants of the terrestrial globe. He first showed, by an investigation based upon the theory of probability, and following the same lines of inquiry which Chéseaux and Olbers had pursued in 1744 and 1823 respectively, that if the ether be a perfect fluid, so that no light is lost in propagation, and the universe be of infinite dimensions, the stars being scattered promiscuously throughout immensity, the face of the heavens would necessarily glow like the disk of the sun ; the whole heavens would be bright like the points now occupied by the stars. As the vault of the celestial sphere is in reality comparatively dark, even in the regions occupied by the densest masses of stars, it follows either that the universe is not infinite, or that the ether is not a perfect fluid. The light of the more distant stars fails to reach us, and we thus miss the empyrean of which the poets have written.

If now we ask which of these two alternatives is indicated, we are reduced to the following answer : in the first place, it is not probable that a fluid like the ether, which transmits waves of light and electricity with a finite velocity, is a perfect fluid ; and therefore the unfathomable depths of it which fill the heavens would perhaps absorb the light of the more distant stars. Even if the universe were infinite, we could never discover this fact. Besides, we know that all space is abundantly strewn with diffused particles of gaseous or meteoric matter, cosmic dust, which here and there, agglomerated into masses, shines as nebulæ ; and hence this dark matter, scattered throughout immensity, and often wholly invisible, must absorb a small part of the light of distant stars. The more distant the stars, the greater the number of dark masses in our line of vision, and hence the greater the absorption of their light. This cosmic dust alone would finally cut off our vision of objects beyond a certain finite distance. Thus the observed absence of Struve’s empyrean may be explained by three hypotheses:

(1.) The universe is finite.

(2.) The universe is infinite, and the imperfectly elastic ether absorbs uniformly (that is, without producing coloration in) the light, and cuts down the magnitude of the more distant stars, so that the vault of the heavens appears comparatively dark even where the stars are densest.

(3.) The light of remote stars is obscured by dark cosmic matter diffused more or less abundantly throughout space.

Which of these hypotheses represents nature, if any of them does, we have no present means of determining. It is a well-known fact that the sky in many directions is not perfectly black, but somewhat brown, as if faintly illuminated by excessively tenuous nebulosity.

The constellation Microscopium, in the southern heavens, offers regions which are very striking on account of the hazy background ; other regions, in various constellations and in the Milky Way, appear perfectly black, without a trace of illumination. In view of these facts, the writer inclines strongly to the belief that hypotheses 2 and 3 offer an adequate explanation of all known phenomena; for the elasticity of the ether does not seem to be perfect, and cosmic dust is evidently widely diffused throughout the immensity of space.

About twelve years ago, Professor D. B. Brace, of the University of Nebraska, examined the transparency of the ether from a physical point of view, and in the light of the most important modern researches into the nature of this medium. Considering the effects of absorption or imperfect elasticity in frittering down ether waves of various lengths, emitted by distant bodies, he found that the more distant parts of the universe, from this cause, ought to exhibit marked coloration, in contrast to the whiter appearance presented by neighboring masses of stars. As there seems to be a total absence of increase of coloration even in the most distant clusters of the Galaxy, it follows that the percentage of light lost through the imperfect elasticity of the ether is infinitesimal. The observational evidence, therefore, gives little support to the theory of absorption by the ether proper, and would rather point to the existence of a veil of dark matter, cosmical dust, which would affect all wave lengths alike, and thus give no relative coloration in the different parts of the universe. Neglecting absorption of light by dark matter, Professor Brace concludes “that the universe must be finite, or, if infinite in extent, the average density of distribution of self-luminous bodies outside our own system must be exceedingly small, as otherwise the sky would appear of a uniform brightness approximating that of the sun.”

As the existence of dark matter in the form of extensive nebulosity diffused generally over the background of the sky must appreciably diminish the light of distant stars, these conclusions, when all causes are considered, are valid only within the premises upon which they rest.

It may occur to some persons that we cannot conceive of an end of space, and it is hardly likely that infinite space would exist without matter ; and hence that the universe necessarily is infinite. This argument proceeds upon the supposition that we can conceive all things which exist, — an admission hardly warranted by experience. For as we can conceive of many things which do not exist, so also there may exist many things of which we can have no clear conception ; as, for example, a fourth dimension to space, or a boundary to the universe.

To make this suggestion more obvious, we shall draw on an analogy sometimes used in transcendental mathematics. The surface of a sphere or an ellipsoid has no end, and yet is finite in dimensions ; and if a being be conceived as moving in the surfaces of either of these mathematical figures, it is clear that he would find no end, and yet he might start from a place and return to it by circumnavigating his universe. The space returns to itself. In like manner, though we cannot conceive of an end to our tridimensional universe, and it may have no end so far as we are concerned, it may in reality be finite, and return to itself by some process to us forever unknowable.

Thus, while our senses conceive space to be endless, it does not follow that the universe is in reality of infinite extent; much less can the absence of an empyrean prove that the cosmos is finite, even to our experience ; for this effect may be due to dust in space, or to the uniform absorption of light by the ether. In the exploration of the sidereal heavens, it is found that the more powerful the telescope, the more stars are disclosed ; and hence the practical indications are that in most directions the sidereal system extends on indefinitely. But the possible uniform extinction of light due to the imperfect elasticity of the luminiferous ether, and the undoubted absorption of light by dark bodies widely diffused in space, seem to preclude forever a definite answer to the question of the bounds of creation.

T. J. J. See.