ΑΝ

STELLAR EVOLUTION.

N interesting work on this subject was published by the late Dr. Croll, the eminent geologist, a short time before his decease. Adopting Laplace's nebular theory of the origin of the solar system, which has been either developed, criticised, or rejected by other astronomers, Dr. Croll goes a step further back in time and proceeds to consider the probable origin of the nebulous mass from which the planetary system was originally evolved. This nebulous mass he supposes to have been formed by the collision of two dark bodies moving directly towards each other in space with a high velocity. A large proportion of the energy of motion thus arrested would, of course, be instantly converted into heat, which would be sufficient to transform into the gaseous state-at least on their surface-the broken fragments of the colliding bodies. We have a familiar example of motion thus converted into heat in the case of a bullet striking an iron target, the heat caused by the concussion being sufficient to raise the temperature of the bullet to a considerable degree. In this case, however, the target being of so much greater mass than the bullet, absorbs most of the developed heat, and being at rest, the amount of heat generated in the bullet is not so great as if it met another bullet moving in the opposite direction.

The first thing which strikes us in considering Dr. Croll's theory is the enormous velocity assumed-476 miles per second! The greatest velocity we know of among the stars having large "proper motions" is that of Arcturus, which, according to a somewhat doubtful parallax, is speeding through space with the amazing velocity of 368 miles a second, and μ Cassiopeia, for which a minute parallax indicates a motion of 315 miles per second! The small star known as 1830 Groombridge, aptly termed by Professor Newcomb "a runaway star," is-if the small parallax found for it (about one-tenth of a second of arc) can be relied upon-moving with a velocity of over 200 miles per second. These are velocities at right angles to the line of sight. The stars may have also a motion in the line of sight, which would of course increase the above velocities. Since,

however, the apparent proper motion of 1830 Groombridge is the maximum known to astronomers, we must look on Dr. Croll's assumed velocity as somewhat excessive. The bodies coming into collision are assumed by Dr. Croll to be dark bodies, but of the existence of these dark bodies we have of course no positive evidence. He objects to the nebular hypothesis that it "begins in the middle of a process," but the assumption of two dark bodies coming into collision leaves unexplained the origin of these bodies, and not alone the origin of their existence, but the origin of their motions still remains a mystery. Dr. Croll, however, says, "The changes that now occur arose out of preceding changes, and these preceding changes out of changes still prior, and so on indefinitely back into the unknown past. This chain of causation-this succession of change, of consequent and antecedent-could not in this manner have extended back to infinity, or else the present stage of the universe's evolution ought to have been reached infinite ages ago. The evolution of things must therefore have had a beginning in time," evidently admitting the possibility of a creation a nihilo.

Assuming, however, the velocity adopted by Dr. Croll, he finds that for two bodies" each one-half of the mass of the sun, moving directly towards each other," the result of the collision would be the development of an amount of heat which would satisfactorily account for the "present rate of the sun's radiation" for a period of fifty millions of years. He computes that the gas developed would have a temperature of about 300 million degrees of the Centigrade thermometer, or more than 140,000 times that of the voltaic arc!" The first result of this collision would be the shattering of both bodies into a number of fragments, which, by their subsequent collisions inter se, would be reduced to smaller fragments, and these again, by the same process, into smaller fragments still, which, being acted on by the enormous heat of the generated gas, would gradually become gaseous also, so that "in the course of time the whole would assume the gaseous condition, and we should then have a perfect nebulaintensely hot, but not very luminous," occupying a space equal in volume to that of our solar system. "As the temperature diminished, the nebulous mass would begin to condense, and ultimately, according to the well-known nebular hypothesis, pass through all the different phases of rings, planets, and satellites into our solar system as it now exists." To this hypothesis Dr. Croll gives the name of the "Impact Theory," to distinguish it from the nebular theory on the one hand, and from the meteoritic and all other gravitation theories on the other.

With reference to the fragments produced by the supposed collision, Dr. Croll considers that it would be "highly improbable, if not impossible, that the whole of the fragments projected outwards with such velocity should be converted into the gaseous condition." Many of the smaller fragments would pass away into outer space, thus forming meteorites, which on this theory must be looked upon as "the offspring of sidereal masses, and not their parents, as Mr. Lockyer concludes." Comets also he considers to have had a similar origin. He admits, however, that some meteorites may have come from other systems.

A necessary assumption of Dr. Croll's theory is that the stars are moving in all directions with various velocities "in perfectly straight lines, and not in definite orbits of any kind." He says, "So far as observation has yet determined, all these conditions seem to be fulfilled." But are we justified in assuming that the stars are moving in straight lines? It is true certainly that the observed proper motion of stars is apparently in a straight line, except in a few cases, like Sirius and Procyon, in which irregularities exist, the cause of which has only been partially explained. But should we therefore assume that the motion is really rectilinear? The small arc described in the comparatively limited number of years during which observations of this kind have been made leaves it, I think, an open question whether the motion is really in a straight line, or whether the short line of motion hitherto observed is really the small arc of a gigantic orbit described round some, as yet unknown, centre.

To the stars mentioned by Dr. Croll as having large proper motions, may be added Lacaille 9352, a Southern star, which, with a proper motion of nearly 7 seconds of arc per annum, stands next in order of rapid motion to 1830 Groombridge, Gould 32416, which has an annual motion of 61 seconds, and the triple star 40 Eridani, of which the proper motion is 4.07 seconds. As Professor Asaph Hall says, "Although the parallax of the star introduces considerable uncertainty" into the computed velocities, "yet we already know enough to be sure that these velocities are very great. Some of them are comparable to that of a comet in close proximity to our sun. But in most cases there is no visible object near the one in motion to which we can ascribe an attractive force, acting according to the Newtonian law, which would produce the velocity observed, unless we assume enormous velocities."

Dr. Croll lays stress on "the enormous space occupied by nebulæ." Some of these wonderful objects are certainly of vast proportions. The large "planetary" nebula in Ursa Major, known as "The Owl Nebula," has an apparent diameter which, if placed at the distance of the nearest fixed star, would imply a real diameter of about 200 times the sun's distance from the earth. As the distance of this nebula is, however, probably much greater than that of a Centauri, its dimensions may be even still larger. Dr. Huggins finds the spectrum gaseous. It is possibly a solar system in its nebulous stage.

Dr. Croll derives some evidence in favour of his hypothesis from the relative densities of the planets composing the solar system, the interior planets being the heaviest, and the exterior the lightest. This is, however, only true in a general way, for though Mercury is the heaviest planet of the solar system, the lightest is not Neptune, but Saturn. In speaking of the satellites, Dr. Croll falls into a serious error. He says, "The satellites of Jupiter, for example, have a density of about only one-fifth of that of the planet, or about onetwenty-fifth of that of the earth; showing that when the planet was rotating as a nebulous mass, the more dense elements were in the central parts, and the less dense at the outer rim, where the satellites were being formed." As a matter of fact, however, the density of Jupiter's first satellite (that nearest to the planet) is only a little less than that of Jupiter itself; that of satellite IV. (the exterior one) is about equal in density to its primary, while the densities of satellites II. and III. are actually greater than that of the planet. This is a very different state of affairs to that supposed by Dr. Croll, the nearest being actually the lightest, and the heaviest next in order.

According to Mr. Lockyer, the temperature of the original solar nebula was as high as that of the sun at present. Dr. Croll, however, considers that "in some of its stages the nebulæ had a very much higher temperature than that now possessed by the sun." This seems very probable, for the cooling process which is now going on in the sun has, in all probability, been in action since the planets solidified, or possibly from an earlier date, so that whatever temperature we may at present assign to our central luminary, we must consider it to have been at a much higher temperature, say, twenty millions of years ago. As Dr. Croll points out, it is impossible that hydrogen or carbon could exist in the cold of stellar space in the gaseous state unless possessed of considerable heat; and we know, from the evidence of the spectroscope, that hydrogen does exist in some of the gaseous nebulæ, notably in the great nebula in Orion. The feeble light emitted by these objects is well accounted for by the fact that incandescent hydrogen, although possessing intense heat, has very little luminosity. Dr. Croll, however, ascribes the faint

1

16,248,160,

light of the nebulæ as chiefly due to the enormous space occupied by these bodies, and computes that their density does not probably exceed "that of hydrogen at ordinary temperature and pressure," a calculation which I find to be correct. The quantity, therefore, of light would be very small, "resembling very much the electric light in a vacuum tube."

Dr. Croll considers that the phenomena of temporary stars, such as those of 1572, 1604, and 1876, are due to the collision of a star with one of the dark bodies, or with a swarm of meteorites. The continuance of visibility was, however, of varying duration in the recorded examples of these wonderful objects; those of 1572 and 1604 remaining bright for over a year, while the maximum brilliancy of those of 1866 and 1876 was only sustained for a few days, or probably hours. His theory will also have to account for the remarkable fact noted by Sir John Herschel, "that all stars of this kind on record, of which the places are distinctly indicated, have occurred, without exception, in or close upon the borders of the Milky Way, and that only within the following semicircle, the preceding having offered no example of the kind." Since this was written, however, a notable exception to this rule occurred in the case of the temporary star of 1866, which so suddenly blazed out in Corona Borealis on May 12 of that year. But this star is itself an exception to the general rule, inasmuch as it was an outburst of a small star previously known to astronomers. The temporary star of 1876, however, conformed to Herschel's rule, as it appeared in the Milky Way near Cygni. The fact of this star having apparently faded into a small planetary nebula seems in favour of Dr. Croll's hypothesis.

Star clusters are explained by Dr. Croll on the supposition that, in some cases, the fragments resulting from the collision would be so "widely distributed through space" as to "prevent a nebula condensing into a single mass." The separate fragments would "gradually condense into separate stars, which would finally assume the conditions of a cluster." I presume that Dr. Croll refers more especially to the "globular clusters" rather than to those in which the components are more widely scattered, and certainly a satisfactory theory of the origin of these wonderful "balls of stars" is a desideratum in sidereal astronomy.

Sir W. Thomson's conclusion that twelve millions of years is the maximum period which can be allowed on the gravitation theory for the duration of the sun's heat in past time, and the apparent inadequacy of this period to meet the views of geologists as to the duration of life on the earth, seems also in favour of Dr. Croll's theory.