Popular Science Monthly/Volume 2/December 1872/Drifting of the Stars



FROM time to time, during the last three years, I have brought before the readers of this magazine the various arguments and considerations on which I have based certain new views respecting the constitution of the sidereal universe. In so doing I have had occasion to deal chiefly with facts already known, though not hitherto viewed in that particular light in which I sought to place them. Indeed, it is an essential part of my general argument that much that is contained in observations already made has been escaping us. In the eagerness of astronomers to ascertain new facts, they have been neglecting the interpretation of facts already ascertained.

But I have long felt that it would greatly tend to advance the new views which I have advocated, if some process of research, pursued by one of those astronomers of our day who possess the requisite means and leisure for prolonged inquiries, should confirm in a clear and decisive way some definite point of my new theories. Thus, if new observational evidence should be found in favor of my theory that the nebulæ are not external to our galaxy, or if new evidence should be obtained to show that the stars are aggregated in certain regions within our system and segregated from others; or, again, if my theory of star-drift should be confirmed by new and striking evidence, I felt that a greater measure of confidence in my analysis of former evidence would thenceforward be accorded. I had no occasion, indeed, to complain of cavil or opposition; in fact, a degree of attention had been given to the new opinions I advocated which was certainly much greater than I had looked for. But there must always be such an inertia in the general weight of opinion in favor of accepted views, that only a steady reiteration of reasoning during a long period, or else some striking and impressive discovery, can cause the weight of opinion to tend in the contrary direction.

I cannot but regard myself as most fortunate in finding the first confirmation of my views (1) coming from one of the most eminent astronomers and physicists of the day, (2) bearing upon one of the most definite and positive of my vaticinations, and (3) relating to one of the most interesting subjects in the whole range of recent astronomical research.

It will be in the remembrance of many readers of this magazine that, nearly four years ago, Dr. Huggins succeeded in showing that the bright star Sirius is travelling at an enormously rapid rate away from us. In other words, besides that rapid thwart-motion which is shifting the place of this star upon the heavens, the star has a rapid motion of recession. In the paper called "Are there any Fixed Stars?" in the Popular Science Review for October, 1868, the nature of the means by which this discovery was effected was fully described and explained. It may be permitted to me to mention, also, that while Dr. Huggins's researches were still unannounced (or rather incomplete) I was so far fortunate as to indicate the possibility of employing the very method of research which Dr. Huggins was then engaged (unknown to me) in applying to Sirius. I propose here briefly to describe and explain the method, referring the reader, who desires fuller information on these preliminary points, to the paper of October, 1868, mentioned above. I am the more desirous of doing this, because I find the principle of the method not readily grasped, and that I conceive the explanation I am about to offer may remove certain difficulties not uncommonly experienced.

Conceive that a person, standing on the edge of a steadily-flowing stream, throws corks into it at regular intervals—say one cork per second. These would float down the stream, remaining always separated by a constant distance. Thus, if the stream were flowing three feet per second, the corks would be a yard apart (supposing, for convenience of illustration, that each cork was thrown with exactly the same force and in exactly the same direction). Now, if a person a mile or so down the stream saw these corks thus floating past, he could infer that they had been thrown in at regular intervals; and, moreover, if he knew the rate of the stream, and that the corks were thrown in by a person standing at the river's edge, he would know that the interval between the throwing of successive corks was one second. But, vice versa, if he knew the rate of the stream, and that the corks were thrown in at intervals of one second, he could infer that the person throwing them was standing still. For let us consider what would happen, if the cork-thrower sauntered up-stream or down-stream while throwing corks at intervals of one second. Suppose he moved up-stream at the rate of a foot per second; then, when he has thrown one cork, he moves a foot up-stream before he throws the next; and the first cork has floated three feet down-stream; hence the second cork falls four feet behind the first. Thus the common distance between the corks is now four feet instead of three feet. Next, suppose he saunters down-stream at the rate of a foot per second; then, when he has thrown one cork, he moves a foot down-stream before he throws the next; and the first cork has floated three feet down-stream; hence the second cork falls only two feet behind the first. Thus the common distance between the corks is now two feet instead of three feet. It is clear, then, that the person standing a mile or so down-stream, if he knows that the stream is flowing three feet per second, and that his friend up-stream is throwing one cork in per second, can be quite sure that his friend is standing still if the corks come past with a common interval of three feet between them. Moreover, he can be equally sure that his friend is sauntering up-stream, if the corks come past with a common interval exceeding three feet; and that he is sauntering down-stream, if the common interval is less than three feet. And, if, by some process of measuring, he can find out exactly how much greater or how much less than three feet the interval is, he can tell exactly how fast his friend is sauntering up-stream or down-stream. It would not matter how far down-stream the observer might be, so long as the stream's rate of flow remained unchanged; nor, indeed, would it matter, even though the stream flowed at a different rate past the observer than past the cork-thrower, so long as neither of these two rates was liable to alteration.

Now, we may compare the emission of light-waves by a luminous object to the throwing of corks in our illustrative case. The rate of flow for light-waves is indeed infinitely faster than that of any river, being no less than 185,000 miles per second. The successive light-waves are set in motion at infinitely shorter time-intervals, since for extreme red light there are no less than 458,000,000,000,000 undulations per second, and for extreme violet no less than 727,000,000,000,000; but these specific differences do not affect the exactness of the illustration. It is obvious that all that is necessary to make the parallel complete is that the flow of light-waves shall reach the observer at a constant rate (which is the actual case), and that he shall know, in the case of any particular and distinguishable kind of light, what is the rate at which the wave-action is successively excited, and be able to compare with this known rate the rate at which they successively reach him. If they come in quicker succession than from a luminous body at rest, he will know that the source of light is approaching, as certainly as our observer down-stream would know that his friend was sauntering toward him if the corks came two feet apart instead of three feet. If, on the contrary, the light-waves of a particular kind come in slower succession than from a body at rest, the observer will know that the source of light is receding, precisely as the river-side observer would know that his friend was travelling away from him if the corks came past him four feet apart instead of three.

Now, the stellar spectroscopist can distinguish, among the light-waves of varied length which reach him, those which have a particular normal length. He analyzes star-light with his spectroscope, and gets from it a rainbow-tinted streak crossed by dark lines. These dark lines belong to definite parts of the spectrum; that is, to such and such parts of its red, or orange, or yellow, or green, or blue, or indigo, or violet portion. Thus they correspond to light having a particular wave-length. And many of these lines in stellar spectra are identifiable with the lines due to known elements. For instance, in the spectrum of Sirius there are four strong dark lines corresponding to the known bright lines of the spectrum of hydrogen. Thus the wave-length corresponding to any one of these dark lines is perfectly well known to the spectroscopist from what he has already learned by examining the bright lines of hydrogen. Now, if Sirius were receding very rapidly, the wave-length corresponding to one of these lines would be lengthened; it would correspond, in fact, to a part of the spectrum nearer the red end, or the region of longer light-waves, and thus the dark line would be shifted toward the red end of the spectrum; whereas, on the contrary, if Sirius were very rapidly approaching, the dark line would be shifted toward the violet end of the spectrum. All that would be necessary would be that the rate of approach or recession should bear an appreciable proportion to the rate at which light travels, or 185,000 miles per second. For, reverting to our cork-thrower, it is clear that, if he travelled up-stream or down-stream at a rate exceedingly minute compared with the stream's rate of flow, it would be impossible for the observer down-stream to be aware of the cork-thrower's motion in either direction, unless, indeed, he had some very exact means of measuring the interval between the successive corks.

Now, the spectrum of a star can be made longer or shorter, according to the dispersive power employed. The longer it is, the fainter its light will be; but, so long as the dark lines can be seen, the longer the spectrum is, the greater is the shift due to steller recession or approach; and, therefore, the more readily may such recession or approach be detected. But, with the instrument used by Dr. Huggins four years ago, it was hopeless, save in the case of the brilliant Sirius (giving more than five times as much light as any other star visible in our northern heavens), to look for any displacement due to a lower rate of recession than some hundred miles per second (little more than the two-thousandth part of the velocity of light). What was to be done, then, was to provide a much more powerful telescope, so that the stellar spectra would bear a considerably greater degree of dispersion. With admirable promptitude, the Royal Society devoted a large sum of money to the construction of such an instrument, to be lent to Dr. Huggins for the prosecution of his researches into stellar motions of approach and recession. This telescope, with an aperture of fifteen inches, and a light-gathering power somewhat exceeding that usual with that aperture, was accordingly completed, and provided with the necessary spectroscopic appliances. Many months have not passed since all the arrangements were complete.

In the mean time, I had arrived at certain inferences respecting the proper motion of the stars, on which Dr. Huggins's researches by the new method seemed likely to throw an important light.

More than three years ago, I had expressed my conviction that, whenever the recorded proper motions of the stars were subject to a careful examination, they would confirm the theory I had enunciated, that the stars are arranged in definite aggregations of various forms—star-groups, star-streams, star-reticulations, star-nodules, and so on. Making leisure, in the summer of 1869, for entering upon such an examination, I was led to several results, which not only confirmed the above-mentioned theory, but suggested relations which I had not hitherto thought of. Some of these results are discussed in the article called "Are there any Fixed Stars?" already referred to; others are presented in an article called "Star-drift," in the Student for October, 1870. The special results on which Dr. Huggins's recent discoveries throw light, were first publicly announced in a paper read before the Royal Society, on January 20, 1870.

I had constructed a chart in which the proper motions of about 1,200 stars were pictured. To each star a minute arrow was affixed, the length of the arrow indicating the rate at which the star is moving on the celestial vault, while the direction in which the arrow pointed shows the direction of the star's apparent motion. This being done, it was possible to study the proper motions much more agreeably and satisfactorily than when they were simply presented in catalogues. And certain features, hitherto unrecognized, at once became apparent. Among these was the peculiarity which I have denominated "Star-drift;" the fact, namely, that certain groups of stars are travelling in a common direction.[1] This was indicated, in certain cases, in too significant a manner to be regarded as due merely to chance distribution in these stellar motions; and I was able to select certain instances in which I asserted that the drift was unmistakable and real.

Among these instances was one of a very remarkable kind. The "seven stars" of Ursa Major—the Septentriones of the ancients—are known to all. For convenience of reference, let us suppose these seven divided as when the group is compared to a wagon and horses. Thus, there are four wagon-wheels and three horses. Now, if we take the wagon-wheels in sequence round their quadrilateral (beginning with one of the pair farthest from the horses), so as to finish with the one which lies nearest to the horses, these are named by astronomers, in that order, Alpha, Beta, Gamma, and Delta, of the Great Bear. Thus, Alpha and Beta are the well-known pointers (Alpha nearest the pole), and Delta is the faintest star of the Septentrion set. The three horses are called in order Epsilon, Zeta, and Eta; Epsilon being nearest to Delta. Now, when the proper motions of these seven stars had been mapped, I found that, whereas Alpha and Eta are now moving much as they would if the sun's motion were alone in question, the other five are all moving at one and the same rate (on the star-sphere, that is) in almost the exactly opposite direction. Moreover, a small star close by Zeta (the middle horse), a star known to the Arabian astronomers as the "Test," because to see this star was held a proof of good eyesight, is moving in the same direction and at the same rate as Zeta and the rest of this set. And besides this star (which has also been called Jack by the middle horse), Zeta has a telescopic companion which also accompanies him in his motion on the celestial sphere.

After a careful consideration of these circumstances, and an analysis of the probabilities in favor of and against the theory that the concurrence of apparent motion was merely accidental, I came to the conclusion that the five large stars and the two smaller ones form a true drifting set. I found, on a moderate computation, that the odds were upward of half a million to one against the concurrence being accidental; and, since I had recognized other instances of concurrence not less striking, I felt that it was morally certain that these stars belong to one star-family.

The reader will perhaps not be surprised to learn, however, that before publishing this conclusion I submitted it (in July, 1869) to one who was, of all men, the best able to pronounce upon its significance—the late Sir John Herschel. I have the letter (dated August 1, 1869), which he sent in reply, before me as I write. The part relating to my discovery runs as follows: "The considerations you adduce relative to the proper motions of the stars are exceedingly curious and interesting. Of late years catalogues have gone into much detail, and with such accuracy that these motions are of course much better known to us than some twenty or thirty years ago. The community of proper motion over large regions (of which you give a picture in Gemini and Cancer) is most remarkable, and the coincidence of proper motion in Beta, Gamma, Delta, Epsilon, and Zeta Ursæ Majoris, most striking. Your promised paper on this subject cannot fail to be highly interesting."[2]

In a letter written on May 11, 1870, and referring not to another letter of mine, but to my "Other Worlds," Sir John Herschel remarked, "The cases of star-drift such as that in Ursa Major are very striking, and richly merit further careful examination."

My first public expression of opinion respecting the star-drift in Ursa Major was conveyed in the following terms: "If these five stars indeed form a system (and I can see no other reasonable explanation of so singular a community of motion), the mind is lost in contemplating the immensity of the periods which the revolutions of the components of the system must occupy. Mädler had already assigned to the revolution of Alcor around Mizar (Zeta Ursæ) a period of more than 7,000 years. But, if these-stars, which appear so close to the naked eye, have a period of such length, what must be the cyclic periods of stars which cover a range of several degrees upon the heavens?" (From Zeta to Beta is a distance on the heavens of about 19°.) "The peculiarities of the apparent proper motions of the stars," I added, "lend a new interest to the researches which Dr. Huggins is preparing to make into the stellar proper motions of recess and approach."

But a few months later, in a lecture delivered at the Royal Institution, I pointed out more definitely what result I expected from Dr. Huggins's researches. "Before long," I said, "it is likely that the theory of star-drift will be subjected to a crucial test, since spectroscopic analysis affords the means of determining the stellar motions of recess and approach. The task is a very difficult one, but astronomers have full confidence that in the able hands of Dr. Huggins it will be successfully accomplished. I await the result with full confidence that it will confirm my views."

It will be manifest that if the five large stars in Ursa are really travelling in the same direction, then, when Dr. Huggins applied the new method of research, he would find that, so far as motion in the line of sight was concerned, these stars were either all receding or all approaching at the same rate, or else that they were all alike in showing no signs of any motion, either of recess or approach.

But in the mean time there was another kind of evidence which the spectroscope might give, and on which I formed some expectations. If these stars form a single system, it seemed likely that they would all be found to be constituted alike—in other words, that their spectra would be similar. Not, indeed, that associated stars always display such similarity. Indeed, the primary star of a binary system not unfrequently exhibits a spectrum unlike that of the small companion. But the five large stars in Ursa, being obviously primary members of the scheme they form, might be expected to resemble each other in general constitution. Moreover, since the stars not included in the set viz.,—Alpha and Eta—might be regarded as probably very much nearer or very much farther away, it was to be expected (though not so confidently) that these two stars would have spectra unlike the spectrum common (on the supposition) to the five stars.

Now, Secchi announced that the stars of the Great Bear, with the exception of Alpha, have spectra belonging to the same type as the spectrum of the bright stars Sirius, Vega, Altair, Regulus, and Rigel. This result was in very pleasing accordance with the anticipations I had formed, except that I should rather have expected to find that the star Eta had a spectrum unlike that of the remaining five stars of the Septentriones. Moreover, as the stars belonging to this particular type are certainly in many cases, and probably in all, very large orbs[3] (referring here to real magnitude, not to apparent brilliancy), the inference seemed fairly deducible that the drifting five stars are not nearer than Alpha, and therefore (since we have seen that it is unlikely that all the Septentriones lie at nearly the same distance) the inference would be that the drifting stars lie much farther away than the rest.

It remained, however, that the crucial test of motion-measurement should be applied.

In the middle of May last I received a letter from Dr. Huggins announcing that the five are all receding from the earth. In all, the hydrogen line called F is "strong and broad." In the spectrum of Alpha the line F is "not very strong" (so faint, indeed, Dr. Huggins afterward informed me, that he preferred to determine the star's motion by one of the lines due to magnesium in the star's atmosphere). He found that Alpha is approaching. As to Eta, Dr. Huggins remarked that the line at F is "not so strong or so broad" as in the spectrum of "the five." He was uncertain as to the direction of motion, and mentioned that "the star was to be observed again." He subsequently found that this star is receding. But, whereas all the five are receding at the enormous rate of thirty miles per second, Eta's recession was so much smaller that, as we have seen, Dr. Huggins was unable to satisfy himself at a single observation that the star was receding at alL

It will be seen that my anticipations were more than fulfilled. The community of recessional motion was accompanied by evidence which might very well have been wanting—viz., by the discovery that neither Eta nor Alpha shared in the motion. Moreover, the physical association between the five stars was yet further evidenced by the close resemblance found to exist between the spectra of the five stars. Dr. Huggins remarked in his letter: "My expectation had nothing to do with the above results. At the moment, I thought Alpha was included in the group, and was therefore a little disappointed when I found Beta going the opposite way."

We have at length, then, evidence, which admits of no question—so obviously conclusive is it—to show not only that star-drift is a reality, but that subordinate systems exist within the sidereal system. We moreover recognize an unquestionable instance of a characteristic peculiarity of structure in a certain part of the heavens. For, though star-drift exists elsewhere, yet every instance of star-drift is quite distinct in character—the drift in Cancer unlike that in Ursa, and both these drifts unlike the drifts in Taurus, and equally unlike the drift in Aries or Leo. Much more, indeed, is contained in the fact now placed beyond question, than appears on the surface. Rightly understood, it exhibits the sidereal system itself as a scheme utterly unlike what has hitherto been imagined. The vastness of extent, the variety of structure, the complexity of detail, and the amazing vitality, on which I have long insisted, are all implied in that single and, as it were, local feature which I had set as a crucial test of my theories. I cannot but feel a strong hope, then, that those researches which my theories suggest, and which I have advocated during the last few years, will now be undertaken by willing observers. The system of star-gauging, which the Herschels did little more than illustrate (as Sir W. Herschel himself admitted), should be applied with telescopes of different power to the whole heavens,[4] not to a few telescopic fields. Processes of charting, and especially of equal surface charting, should be multiplied. Fresh determinations of proper motions should be systematically undertaken. All the evidence, in fine, which we have, should be carefully examined, and no efforts should be spared by which new evidence may be acquired. Only when this has been done will the true nature of the galaxy be adequately recognized, its true vastness gauged, its variety and complexity understood, its vitality rendered manifest. To obtain, indeed, an absolutely just estimate of these matters, may not be in man's power to compass; but he can hope to obtain a true relative interpretation of the mysteries of the stellar system. If any astronomer be disposed to question the utility or value of such researches, let him remember that Sir W. Herschel, the greatest of all astronomers, set "a knowledge of the constitution of the heavens" as "the ultimate object of his observations."—Popular Science Review.


  1. I include this among "features hitherto unrecognized," though Michell had already noted the fact that the stars are arranged into systems. "We may conclude," he said, "that the stars are really collected together in clusters in some places, where they form a kind of systems; while in others there are few or none of them, to whatever cause this may be owing, whether to their mutual gravitation or to some other law or appointment of the Creator."
  2. He proceeds as follows (the passage is removed from the main text, as relating to a different branch of the subject): "I cannot say that I am at all surprised at its being found that the average proper motions of stars of small magnitudes are not less than those of large, considering (as I have always done) that the range of individual magnitude (i. e., lustre) must be so enormous that multitudes of very minute stars may in fact be our very near neighbors." Compare my paper on "The Sun's Journey through Space," above referred to, which paper also deals with the point touched on in the next sentence of Sir John Herschel's letter: "Your remark on the conclusion I have been led to draw, relative to the small effect of the correction due to the sun's proper motion, will require to be very carefully considered, and I shall of course give it every attention."
  3. Sirius demonstrably gives out much more light than our sun, and according to the best determinations of his distance he must (if his surface is of equal intrinsic lustre) be from 2,000 to 8,000 times larger than the sun. Vega, Altair, and Rigel, are also certainly larger and may be very much larger than our sun.
  4. This is a work in which telescopes of every order of power would be useful. The observations, also, would be very easily made and would tell amazingly.