THEORIES OF PLANETARY EVOLUTION

We cannot doubt that the solar system, as we see it, is the result of some process of growth—that, during innumerable ages, the forces of Nature were at work upon its materials, blindly modelling them into the shape appointed for them from the beginning by Omnipotent Wisdom. To set ourselves to inquire what that process was may be an audacity, but it is a legitimate, nay, an inevitable one. For man's implanted instinct to "look before and after" does not apply to his own little life alone, but regards the whole history of creation, from the highest to the lowest—from the microscopic germ of an alga or a fungus to the visible frame and furniture of the heavens.

Kant considered that the inquiry into the mode of origin of the world was one of the easiest problems set by Nature; but it cannot be said that his own solution of it was satisfactory. He, however, struck out in 1755 a track which thought still pursues. In his Allgemeine Naturgeschichte the growth of sun and planets was traced from the cradle of a vast and formless mass of evenly diffused particles, and the uniformity of their movements was sought to be accounted for by the unvarying action of attractive and repulsive forces, under the dominion of which their development was carried forward.

In its modern form, the "Nebular Hypothesis" made its appearance in 1796.[1150] It was presented by Laplace with diffidence, as a speculation unfortified by numerical buttresses of any kind, yet with visible exultation at having, as he thought, penetrated the birth-secret of our system. He demanded, indeed, more in the way of postulates than Kant had done. He started with a sun ready made,[1151] and surrounded with a vast glowing atmosphere, extending into space out beyond the orbit of the farthest planet, and endowed with a slow rotatory motion. As this atmosphere or nebula cooled, it contracted; and as it contracted, its rotation, by a well-known mechanical law, became accelerated. At last a point arrived when tangential velocity at the equator increased beyond the power of gravity to control, and equilibrium was restored by the separation of a nebulous ring revolving in the same period as the generating mass. After a time, the ring broke up into fragments, all eventually reunited in a single revolving and rotating body. This was the first and farthest planet.

Meanwhile the parent nebula continued to shrink and whirl quicker and quicker, passing, as it did so, through successive crises of instability, each resulting in, and terminated by, the formation of a planet, at a smaller distance from the centre, and with a shorter period of revolution than its predecessor. In these secondary bodies the same process was repeated on a reduced scale, the birth of satellites ensuing upon their contraction, or not, according to circumstances. Saturn's ring, it was added, afforded a striking confirmation of the theory of annular separation,[1152] and appeared to have survived in its original form in order to throw light on the genesis of the whole solar system; while the four first discovered asteroids offered an example in which the dÉbris of a shattered ring had failed to coalesce into a single globe.

This scene of cosmical evolution was a characteristic bequest from the eighteenth century to the nineteenth. It possessed the self-sufficing symmetry and entireness appropriate to the ideas of a time of renovation, when the complexity of nature was little accounted of in comparison with the imperious orderliness of the thoughts of man. Since its promulgation, however, knowledge has transgressed many boundaries, and set at naught much ingenious theorising. How has it fared with Laplace's sketch of the origin of the world? It has at least not been discarded as effete. The groundwork of speculation on the subject is still furnished by it. It is, nevertheless, admittedly inadequate. Of much that exists it gives no account, or an erroneous one. The march of events certainly did not everywhere—even if it did anywhere—follow the exact path prescribed for it. Yet modern science attempts to supplement, but scarcely ventures to supersede it.

Thought has, in many directions, been profoundly modified by Mayer's and Joule's discovery, in 1842, of the equivalence between heat and motion. Its corollary was the grand idea of the "conservation of energy," now one of the cardinal principles of science. This means that, under the ordinary circumstances of observation, the old maxim ex nihilo nihil fit applies to force as well as to matter. The supplies of heat, light, electricity, must be kept up, or the stream will cease to flow. The question of the maintenance of the sun's heat was thus inevitably raised; and with the question of maintenance that of origin is indissolubly connected.

Dr. Julius Robert Mayer, a physician residing at Heilbronn, was the first to apply the new light to the investigation of what Sir John Herschel had termed the "great secret." He showed that if the sun were a body either simply cooling or in a state of combustion, it must long since have "gone out." Had an equal mass of coal been set alight four or five centuries after the building of the Pyramid of Cheops, and kept burning at such a rate as to supply solar light and heat during the interim, only a few cinders would now remain in lieu of our undiminished glorious orb. Mayer looked round for an alternative. He found it in the "meteoric hypothesis" of solar conservation.[1153] The importance in the economy of our system of the bodies known as falling stars was then (in 1848) beginning to be recognised. It was known that they revolved in countless swarms round the sun; that the earth daily encountered millions of them; and it was surmised that the cone of the zodiacal light represented their visible condensation towards the attractive centre. From the zodiacal light, then, Mayer derived the store needed for supporting the sun's radiations. He proved that, by the stoppage of their motion through falling into the sun, bodies would evolve from 4,600 to 9,200 times as much heat (according to their ultimate velocity) as would result from the burning of equal masses of coal, their precipitation upon the sun's surface being brought about by the resisting medium observed to affect the revolutions of Encke's comet. There was, however, a difficulty. The quantity of matter needed to keep, by the sacrifice of its movement, the hearth of our system warm and bright would be very considerable. Mayer's lowest estimate put it at 94,000 billion kilogrammes per second, or a mass equal to that of our moon bi-annually. But so large an addition to the gravitating power of the sun would quickly become sensible in the movement of the bodies dependent upon him. Their revolutions would be notably accelerated. Mayer admitted that each year would be shorter than the previous one by a not insignificant fraction of a second, and postulated an unceasing waste of substance, such as Newton had supposed must accompany emission of the material corpuscles of light, to neutralise continual reinforcement.

Mayer's views obtained a very small share of publicity, and owned Mr. Waterston as their independent author in this country. The meteoric, or "dynamical," theory of solar sustentation was expounded by him before the British Association in 1853. It was developed with his usual ability by Lord Kelvin, in the following year. The inflow of meteorites, he remarked, "is the only one of all conceivable causes of solar heat which we know to exist from independent evidence."[1154] We know it to exist, but we now also know it to be entirely insufficient. The supplies presumed to be contained in the zodiacal light would be quickly exhausted; a constant inflow from space would be needed to meet the demand. But if moving bodies were drawn into the sun at anything like the required rate, the air, even out here at ninety-three millions of miles distance, would be thick with them; the earth would be red-hot from their impacts;[1155] geological deposits would be largely meteoric;[1156] to say nothing of the effects on the mechanism of the heavens. Lord Kelvin himself urged the inadmissibility of the "extra-planetary" theory of meteoric supply on the very tangible ground that, if it were true, the year would be shorter now, actually by six weeks, than at the opening of the Christian era. The "intra-planetary" supply, however, is too scanty to be anything more than a temporary makeshift.

The meteoric hypothesis was naturally extended from the maintenance of the sun's heat to the formation of the bodies circling round him. The earth—no less doubtless than the other planets—is still growing. Cosmical matter in the shape of falling stars and aËrolites, to the amount, adopting Professor Newton's estimate, of 100 tons daily, is swept up by it as it pursues its orbital round. Inevitably the idea suggested itself that this process of appropriation gives the key to the life-history of our globe, and that the momentary streak of fire in the summer sky represents a feeble survival of the glowing hailstorm by which in old times it was fashioned and warmed. Mr. E. W. Brayley supported this view of planetary production in 1864,[1157] and it has recommended itself to Haidinger, Helmholtz, Proctor, and Faye. But the negative evidence of geological deposits appears fatal to it.

The theory of solar energy now generally regarded as the true one was enounced by Helmholtz in a popular lecture in 1854. It depends upon the same principle of the equivalence of heat and motion which had suggested the meteoric hypothesis. But here the movement surrendered and transformed belongs to the particles, not of any foreign bodies, but of the sun itself. Drawn together from a wide ambit by the force of their own gravity, their fall towards the sun's centre must have engendered a vast thermal store, of which 453/454 are computed to be already spent. Presumably, however, this stream of reinforcement is still flowing. In the very act of parting with heat, the sun develops a fresh stock. His radiations, in short, are the direct result of shrinkage through cooling. A diminution of the solar diameter by 380 feet yearly would just suffice to cover the present rate of emission, and would for ages remain imperceptible with our means of observation, since, after the lapse of 6,000 years, the lessening of angular size would scarcely amount to one second.[1158] But the process, though not terminated, is strictly a terminable one. In less than five million years, the sun will have contracted to half its present bulk. In seven million more, it will be as dense as the earth. It is difficult to believe that it will then be a luminous body.[1159] Nor can an unlimited past duration be admitted. Helmholtz considered that radiation might have gone on with its actual intensity for twenty-two, Langley allows only eighteen million years. The period can scarcely be stretched, by the most generous allowances, to double the latter figure. But this is far from meeting the demands of geologists and biologists.

An attempt was made in 1881 to supply the sun with machinery analogous to that of a regenerative furnace, enabling it to consume the same fuel over and over again, and so to prolong indefinitely its beneficent existence. The inordinate "waste" of energy, which shocks our thrifty ideas, was simultaneously abolished. The earth stops and turns variously to account one 2,250-millionth part of the solar radiations; each of the other planets and satellites takes a proportionate share; the rest, being all but an infinitesmal fraction of the whole, is dissipated through endless space, to serve what purpose we know not. Now, on the late Sir William Siemens's plan, this reckless expenditure would cease; the solar incomings and outgoings would be regulated on approved economic principles, and the inevitable final bankruptcy would be staved off to remote ages.

But there was a fatal flaw in its construction. He imagined a perpetual circulation of combustible materials, alternately surrendering and regaining chemical energy, the round being kept going by the motive force of the sun's rotation.[1160] This, however, was merely to perch the globe upon a tortoise, while leaving the tortoise in the air. The sun's rotation contains a certain definite amount of mechanical power—enough, according to Lord Kelvin, if directly converted into heat, to keep up the sun's emission during 116 years and six days—a mere moment in cosmical time. More economically applied, it would no doubt go farther. Its exhaustion would, nevertheless, under the most favourable circumstances, ensue in a comparatively short period.[1161] Many other objections equally unanswerable have been urged to the "regenerative" hypothesis, but this one suffices.

Dr. Croll's collision hypothesis[1162] is less demonstrably unsound, but scarcely less unsatisfactory. By the mutual impact of two dark masses rushing together with tremendous speed, he sought to provide the solar nebula with an immense original stock of heat for the reinforcement of that subsequently evolved in the course of its progressive contraction. The sun, while still living on its capital, would thus have a larger capital to live on, and the time-demands of the less exacting geologists and biologists might be successfully met. But the primitive event, assumed for the purpose of dispensing them from the inconvenience of "hurrying up their phenomena," is not one that a sane judgment can readily admit to have ever, in point of actual fact, happened.

There remains, then, as the only intelligible rationale of solar sustentation, Helmholtz's shrinkage theory. And this has a very important bearing upon the nebular view of planetary formation; it may, in fact, be termed its complement. For it involves the idea that the sun's materials, once enormously diffused, gradually condensed to their present volume with development of heat and light, and, it may plausibly be added, with the separation of dependent globes. The data furnished by spectrum analysis, too, favour the supposition of a common origin for sun and planets by showing their community of substance; while gaseous nebulÆ present examples of vast masses of tenuous vapour, such as our system may plausibly be conjectured to have primitively sprung from.

But recent science raises many objections to the details, if it supplies some degree of confirmation to the fundamental idea of Laplace's cosmogony. The detection of the retrograde movement of Neptune's satellite made it plain that the anomalous conditions of the Uranian world were due to no extraordinary disturbance, but to a systematic variety of arrangement at the outskirts of the solar domain. So that, were a trans-Neptunian planet discovered, we should be fully prepared to find it rotating, and surrounded by satellites circulating from east to west. The uniformity of movement, upon the probabilities connected with which the French geometer mainly based his scheme, thus at once vanishes.

The excessively rapid revolution of the inner Martian moon is a further stumbling-block. On Laplace's view, no satellite can revolve in a shorter time than its primary rotates; for in its period of circulation survives the period of rotation of the parent mass which filled the sphere of its orbit at the time of giving it birth. And rotation quickens as contraction goes on; therefore, the older time of axial rotation should invariably be the longer. This obstacle can, however, as we shall presently see, be turned.

More serious is one connected with the planetary periods, pointed out by Babinet in 1861.[1163] In order to make them fit in with the hypothesis of successive separation from a rotating and contracting body, certain arbitrary assumptions have to be made of fluctuations in the distribution of the matter forming that body at the various epochs of separation.[1164] Such expedients usually merit the distrust which they inspire. Primitive and permanent irregularities of density in the solar nebula, such as Miss Young's calculations suggest,[1165] do not, on the other hand, appear intrinsically improbable.

Again, it was objected by Professor Kirkwood in 1869[1166] that there could be no sufficient cohesion in such an enormously diffused mass as the planets are supposed to have sprung from to account for the wide intervals between them. The matter separated through the growing excess of centrifugal speed would have been cast off, not by rarely recurring efforts, but continually, fragmentarily, pari passu with condensation and acceleration. Each wisp of nebula, as it found itself unduly hurried, would have declared its independence, and set about revolving and condensing on its own account. The result would have been a meteoric, not a planetary system.

Moreover, it is a question whether the relative ages of the planets do not follow an order just the reverse of that concluded by Laplace. Professor Newcomb holds the opinion that the rings which eventually constituted the planets divided from the main body of the nebula almost simultaneously, priority, if there were any, being on the side of the inner and smaller ones;[1167] while in M. Faye's cosmogony,[1168] the retrograde motion of the systems formed by the two outer planets is ascribed—on grounds, it is true, of dubious validity—to their comparatively late origin.

This ingenious scheme was designed, not merely to complete, but to supersede that of Laplace, which, undoubtedly, through the inclusion by our system of oppositely directed rotations, forfeits its claim simply and singly to account for the fundamental peculiarities of its structure.

M. Faye's leading contention is that, under the circumstances assumed by Laplace, not the two outer planets alone, but the whole company must have been possessed of retrograde rotation. For they were formed—ex hypothesi—after the sun; central condensation had reached an advanced stage when the rings they were derived from separated; the principle of inverse squares consequently held good, and Kepler's Laws were in full operation. Now, particles circulating in obedience to these laws can only—since their velocity decreases outward from the centre of attraction—coalesce into a globe with a backward axial movement. Nor was Laplace blind to this flaw in his theory; but his effort to remove it, though it passed muster for the best part of a century,[1169] was scarcely successful. His planet-forming rings were made to rotate all in one piece, their outer parts thus necessarily travelling at a swifter linear rate than their inner parts, and eventually uniting, equally of necessity, into a forward-spinning body. The strength of cohesion involved may, however, safely be called impossible, especially when it is considered that nebulous materials were in question.

The reform proposed by M. Faye consists in admitting that all the planets inside Uranus are of pre-solar origin—that they took globular form in the bosom of a nearly homogeneous nebula, revolving in a single period, with motion accelerated from centre to circumference, and hence agglomerating into masses with a direct rotation. Uranus and Neptune owe their exceptional characteristics to their later birth. When they came into existence, the development of the sun was already far advanced, central force had acquired virtually its present strength, unity of period had been abolished by its predominance, and motion was retarded outward.

Thus, what we may call the relative chronology of the solar system is thrown once more into confusion. The order of seniority of the planets is now no easier to determine than the "Who first, who last?" among the victims of Hector's spear. For M. Faye's arrangements, notwithstanding the skill with which he has presented them, cannot be unreservedly accepted. The objections to them, thoughtfully urged by M. C. Wolf[1170] and Professor Darwin,[1171] are grave. Not the least so is his omission to take account of an agency of change presently to be noticed.

A further valuable discussion of the matter was published by M. du LigondÈs in 1897.[1172] His views are those of Faye, modified to disarm the criticisms they had encountered; and special attention may be claimed for his weighty remark that each planet has a life-history of its own, essentially distinct from those of the others, and, despite original unity, not to be confounded with them. The drift of recent investigations seems, indeed, to be to find the embryonic solar system already potentially complete in the parent nebula, like the oak in an acorn, and to relegate detailed explanations of its peculiarities to the dim pre-nebular fore-time.

We now come to a most remarkable investigation—one, indeed, unique in its profession to lead us back with mathematical certainty towards the origin of a heavenly body. We refer to Professor Darwin's inquiries into the former relations of the earth and moon.[1173]

They deal exclusively with the effects of tidal friction, and primarily with those resulting, not from oceanic, but from "bodily" tides, such as the sun and moon must have raised in past ages on a liquid or viscous earth. The immediate effect of either is, as already explained, to destroy the rotation of the body on which the tide is raised, as regards the tide-raising body, bringing it to turn always the same face towards its disturber. This, we can see, has been completely brought about in the case of the moon. There is, however, a secondary or reactive effect. Action is always mutual. Precisely as much as the moon pulls the terrestrial tidal wave backward, the tidal wave pulls the moon forward. But pulling a body forward in its orbit implies the enlargement of that orbit; in other words, the moon is, as a consequence of tidal friction, very slowly receding from the earth. This will go on (other circumstances remaining unchanged) until the lengthening day overtakes the more tardily lengthening month, when each will be of about 1,400 hours.[1174] A position of what we may call tidal equilibrium between earth and moon will (apart from disturbance by other bodies) then be attained.

If, however, it be true that, in the time to come, the moon will be much farther from us, it follows that in the time past she was much nearer to us than she now is. Tracing back her history by the aid of Professor Darwin's clue, we at length find her revolving in a period of somewhere between two and four hours, almost in contact with an earth rotating just at the same rate. This was before tidal friction had begun its work of grinding down axial velocity and expanding orbital range. But the position was not one of stable equilibrium. The slightest inequality must have set on foot a series of uncompensated changes. If the moon had whirled the least iota faster than the earth spun she must have been precipitated upon it. Her actual existence shows that the trembling balance inclined the other way. By a second or two to begin with, the month exceeded the day; the tidal wave crept ahead of the moon; tidal friction came into play, and our satellite started on its long spiral journey outward from the parent globe. This must have occurred, it is computed, at least fifty-four million years ago.

That this kind of tidal reactive effect played its part in bringing the moon into its present position, and is still, to some slight extent, at work in changing it, there can be no doubt whatever. An irresistible conjecture carried the explorer of its rigidly deducible consequences one step beyond them. The moon's time of revolution, when so near the earth as barely to escape contact with it, must have been, by Kepler's Law, more than two and less than two and a half hours. Now it happens that the most rapid rate of rotation of a fluid mass of the earth's average density, consistent with spheroidal equilibrium, is two hours and twenty minutes. Quicken the movement but by one second and the globe must fly asunder. Hence the inference that the earth actually did fly asunder through over-fast spinning, the ensuing disruption representing the birth-throes of the moon. It is likely that the event was hastened or helped by solar tidal disturbance.

To recapitulate. Analysis tracks backward the two bodies until it leaves them in very close contiguity, one rotating and the other revolving in approximately the same time, and that time certainly not far different from, and quite possibly identical with, the critical period of instability for the terrestrial spheroid. "Is this," Professor Darwin asks, "a mere coincidence, or does it not rather point to the break-up of the primeval planet into two masses in consequence of a too rapid rotation?"[1175]

We are tempted, but are not allowed to give an unqualified assent. Mr. James Nolan of Victoria has made it clear that the moon could not have subsisted as a continuous mass under the powerful disruptive strain which would have acted upon it when revolving almost in contact with the present surface of the earth; and Professor Darwin, admitting the objection, concedes to our satellite, in its initial stage, the alternative form of a flock of meteorites.[1176] But such a congregation must have been quickly dispersed, by tidal action, into a meteoric ring. The same investigator subsequently fixed 6,500 miles from centre to centre as the minimum distance at which the moon could have revolved in its entirety; and he concluded it "necessary to suppose that, after the birth of a satellite, if it takes place at all in this way, a series of changes occur which are quite unknown."[1177] The evidence, however, for the efficiency of tidal friction in bringing about the actual configuration of the lunar-terrestrial system is not invalidated by this failure to penetrate its natal mystery. Under its influence the principal elements of that system fall into interdependent mutual relations. It connects, casually and quantitatively, the periods of the moon's revolution and of the earth's rotation, the obliquity of the ecliptic, the inclination and eccentricity of the lunar orbit. All this can scarcely be accidental.

Professor Darwin's first researches on this subject were communicated to the Royal Society, December 18, 1879. They were followed, January 20, 1881,[1178] by an inquiry on the same principles into the earlier condition of the entire solar system. The results were a warning against hasty generalisation. They showed that the lunar-terrestrial system, far from being a pattern for their development, was a singular exception among the bodies swayed by the sun. Its peculiarity resides in the fact that the moon is proportionately by far the most massive attendant upon any known planet. Its disturbing power over its primary is thus abnormally great, and tidal friction has, in consequence, played a predominant part in bringing their mutual relations into their present state.

The comparatively late birth of the moon tends to ratify this inference. The dimensions of the earth did not differ (according to our present authority) very greatly from what they now are when her solitary offspring came, somehow, into existence. This is found not to have been the case with any other of the planets. It is unlikely that the satellites of Jupiter, Saturn, or Mars (we may safely add, of Uranus or Neptune) ever revolved in much narrower orbits than those they now traverse; it is practically certain that they did not, like our moon, originate very near the present surfaces of their primaries.[1179] What follows? The tide-raising power of a body grows with vicinity in a rapidly accelerated ratio. Lunar tides must then have been on an enormous scale when the moon swung round at a fraction of its actual distance from the earth. But no other satellite with which we are acquainted occupied at any time a corresponding position. Hence no other satellite ever possessed tide-raising capabilities in the least comparable to those of the moon. We conclude once more that tidal friction had an influence here very different from its influence elsewhere. Quite possibly, however, that influence may be more nearly spent than in less advanced combinations of revolving globes. Mr. Nolan concluded in 1895[1180] that it still retains appreciable efficacy in the several domains of the outer planets. The moons of Jupiter and Saturn are, by his calculations, in course of sensible retreat, under compulsion of the perennial ripples raised by them on the surfaces of their gigantic primaries. He thus connects the interior position of the fifth Jovian satellite with its small mass. The feebleness of its tide-raising power obliged it to remain behind its companions; for there is no sign of its being more juvenile than the Galilean quartette.

The yielding of plastic bodies to the strain of unequal attractions is a phenomenon of far-reaching consequence. We know that the sun as well as the moon causes tides in our oceans. There must, then, be solar, no less than lunar, tidal friction. The question at once arises: What part has it played in the development of the solar system? Has it ever been one of leading importance, or has its influence always been, as it now is, subordinate, almost negligible? To this, too, Professor Darwin supplies an answer.

It can be stated without hesitation that the sun did not give birth to the planets, as the earth has been supposed to have given birth to the moon, by the disruption of its already condensed, though viscous and glowing mass, pushing them then gradually backward from its surface into their present places. For the utmost possible increase in the length of the year through tidal friction is one hour; and five minutes is a more probable estimate.[1181] So far as the pull of tide-waves raised on the sun by the planets is concerned, then, the distances of the latter have never been notably different from what they now are; though that cause may have converted the paths traversed by them from circles into ellipses.

Over their physical history, however, it was probably in a large measure influential. The first vital issue for each of them was—satellites or no satellites? Were they to be governors as well as governed, or should they revolve in sterile isolation throughout the Æons of their future existence? Here there is strong reason to believe that solar tidal friction was the overruling power. It is remarkable that planetary fecundity increases—at least so far outward as Saturn—with distance from the sun. Can these two facts be in any way related? In other words, is there any conceivable way by which tidal influence could prevent or impede the throwingoff of secondary bodies? We have only to think for a moment in order to see that this is precisely one of its direct results.[1182]

Tidal friction, whether solar or lunar, tends to reduce the axial movement of the body it acts upon. But the separation of satellites depends—according to the received view—upon the attainment of a disruptive rate of rotation. Hence, if solar tidal friction were strong enough to keep down the pace below this critical point, the contracting mass would remain intact—there would be no satellite-production. This, in all probability, actually occurred in the case both of Mercury and Venus. They cooled without dividing, because the solar friction-brake applied to them was too strong to permit acceleration to pass the limit of equilibrium. The complete destruction of their relative axial movement has been rendered probable by recent observations; and that the process went on rapidly is a reasonable further inference. The earth barely escaped the fate of loneliness incurred by her neighbours. Her first and only epoch of instability was retarded until she had nearly reached maturity. The late appearance of the moon accounts for its large relative size—through the increased cohesion of an already strongly condensed parent mass—and for the distinctive peculiarities of its history and influence on the producing globe.

Solar tidal friction, although it did not hinder the formation of two minute dependents of Mars, has been invoked to explain the anomalously rapid revolution of one of them. Phobos, we have seen, completes more than three revolutions while Mars rotates once. But this was probably not always so. The two periods were originally nearly equal. The difference, it is alleged, was brought about by tidal waves raised by the sun on the semi-fluid spheroid of Mars. Rotatory velocity was thereby destroyed, the Martian day slowly lengthened, and, as a secondary consequence, the period of the inner satellite, become shorter than the augmented day, began progressively to diminish. So that Phobos, unlike our moon, was in the beginning farther from its primary than now.

But here again Mr. Nolan entered a caveat. Applying the simple test of numerical evaluation, he showed that before solar tidal friction could lengthen the rotation-period of Mars by so much as one minute, Phobos should have been precipitated upon its surface.[1183] For the enormous disparity of mass between it and the sun is so far neutralised by the enormous disparity in their respective distances from Mars that solar tidal force there is only fifty times that of the little satellite. But the tidal effects of a satellite circulating quicker than its primary rotates exactly reverse those of one moving, like our moon, comparatively slowly, so that the tides raised by Phobos tend to shorten both periods. Its orbital momentum, however, is so extremely small in proportion to the rotational momentum of Mars, that any perceptible inroad upon the latter is attended by a lavish and ruinous expenditure of the former. It is as if a man owning a single five-pound note were to play for equal stakes with a man possessing a million. The bankruptcy sure to ensue is typified by the coming fate of the Martian inner satellite. The catastrophe of its fall needs to bring it about only a very feeble reactive pull compared with the friction which the sun should apply in order to protract the Martian day by one minute. And from the proportionate strength of the forces at work, it is quite certain that one result cannot take place without the other. Nor can things have been materially different in the past; hence the idea must be abandoned that the primitive time of rotation of Mars survives in the period of its inner satellite.

The anomalous shortness of the latter may, however, in M. Wolf's opinion,[1184] be explained by the "traÎnÉes elliptiques" with which Roche supplemented nebular annulation.[1185] These are traced back to the descent of separating strata from the shoulders of the great nebulous spheroid towards its equatorial plane. Their rotational velocity being thus relatively small, they formed "inner rings," very much nearer to the centre of condensation than would have been possible on the unmodified theory of Laplace. Phobos might, in this view, be called a polar offset of Mars; and the rings of Saturn are thought to own a similar origin.

Outside the orbit of Mars, solar tidal friction can scarcely be said to possess at present any sensible power. But it is far from certain that this was always so. It seems not unlikely that its influence was the overruling one in determining the direction of planetary rotation. M. Faye, as we have seen, objected to Laplace's scheme that only retrograde secondary systems could be produced by it. In this he was anticipated by Kirkwood, who, however, supplied an answer to his own objection.[1186]

Sun-raised tides must have acted with great power on the diffused masses of the embryo planets. By their means they doubtless very soon came to turn (in lunar fashion) the same hemisphere always towards their centre of motion. This amounts to saying that even if they started with retrograde rotation, it was, by solar tidal friction, quickly rendered direct.[1187] For it is scarcely necessary to point out that a planet turning an invariable face to the sun rotates in the same direction in which it revolves, and in the same period. As, with the progress of condensation, tides became feebler and rotation more rapid, the accelerated spinning necessarily proceeded in the sense thus prescribed for it. Hence the backward axial movements of Uranus and Neptune may very well be a survival, due to the inefficiency of solar tides at their great distance, of a state of things originally prevailing universally throughout the system.

The general outcome of Mr. Darwin's researches has been to leave Laplace's cosmogony untouched. He concludes nothing against it, and, what perhaps tells with more weight in the long run, has nothing to substitute for it. In one form or the other, if we speculate at all on the development of the planetary system, our speculations are driven into conformity with the broad lines of the Nebular Hypothesis—to the extent, at least, of admitting an original material unity and motive uniformity. But we can see now, better than formerly, that these supply a bare and imperfect sketch of the truth. We should err gravely were we to suppose it possible to reconstruct, with the help of any knowledge our race is ever likely to possess, the real and complete history of our admirable system. "The subtlety of nature," Bacon says, "transcends in many ways the subtlety of the intellect and senses of man." By no mere barren formula of evolution, indiscriminately applied all round, the results we marvel at, and by a fragment of which our life is conditioned, were brought forth; but by the manifold play of interacting forces, variously modified and variously prevailing, according to the local requirements of the design they were appointed to execute.