sippi is also more likely to afford us a fair test of ordinary denudation, because, unlike the St. Lawrence and its tributaries, there are no great lakes in which the fluviatile sediment is thrown down and arrested on its way to the sea. In striking a general average we have to remember that there are large deserts in which there is scarcely any rainfall, and tracts which are as rainless as parts of Peru; and these must not be neglected as counterbalancing others in the tropics where the quantity of rain is in excess. If then, argues Mr. Geikie, we assume that the Mississippi is lowering the surface of the great basin which it drains at the rate of one foot in 6,000 years, 10 feet in 60,000 years, 100 feet in 600,000 years, and 1,000 feet in 6,000,000 years, it would not require more than about 4,500,000 years to wear away the whole of the North American continent if its mean height is correctly estimated by Humboldt at 748 feet; and if the mean height of all the land now above the sea throughout the globe is 1,000 feet, as some geographers believe, it would only require 6,000,000 years to subject a mass of rock equal in volume to the whole of the land to the action of sub-aërial denudation. It may be objected that the annual waste is partial, and not equally derived from the general level of the country, inasmuch as plains, watersheds, and level ground at all heights remain comparatively unaltered; but this, as Mr. Geikie has well pointed out, does not affect our estimate of the sum total of denudation. The amount remains the same; and if we allow too little for the loss from the surface of table-lands, we only increase the proportion of the loss sustained by the sides and bottoms of the valleys, and vice versâ.”

every

We may note, in passing, that, adopting the estimated rate of denudation here indicated, the actual time required for the entire submergence of the present continents, if no vulcanian forces were at work to prevent submergence, would not necessarily be even approximately represented by the period of 6,000,000 years mentioned above. At the outset the rate of submergence would be greater than the mere rate of denudation, since foot removed from the surface of the continents would cause a rise of about 4 inches in the level of the sea; so that at first the surface of continents would be lowered on the average not one foot only in 6,000 years, with respect to the sea-level, but 1 foot 43 inches. On the other hand, as the continents became greatly reduced in extent, it is probable that the average annual rate of denudation would be diminished, the portions still remaining above the sea-level being of harder and more durable material than those which had been removed. We need not inquire further, however, into the question here raised, which, though suggested by our subject, does not, strictly speaking, belong to it; moreover, in nature the process considered cannot take place, the earth's internal forces constantly restoring the balance between land and water by the upheaval of submerged regions.

For the purpose of our present inquiry the action of the earth's vulcanian energies need not be considered; because we are concerned

only with the question how long would be the period of time required for the removal of a stratum so many hundreds or thousands of feet in thickness. We know, certainly, that, in the special cases we have to deal with, strata of such and such thickness were removed; and it matters little whether, as the process of removal went on, they were being steadily raised by the earth's subterranean action, or whether the original dislocation was followed by the sudden raising of the strata at one side of the fault and their equally sudden lowering on the other side. However the difference was brought about, it is certain that the raised strata were worn down eventually by the steady action of the same causes which wear down the general surface of the large continents. Having ascertained the average rate at which these causes work, we can apply the result to determine how long they would be in producing the observed levelling down of the upraised strata in faults. There is no reason for supposing that in the remote past the process would go on more quickly than at the present time. And we have seen that even if it did, that would imply a greater activity in the solar energies to which these processes are all in reality due, so that our difficulty would be in no way diminished by any such assumption. The time required would be reduced by a few millions of years perhaps; but the difficulty we are dealing with is not a question of time at all. We are inquiring now into the amount of the total expenditure of solar energies in past ages; and the time-intervals indicated by the earth's crust are only of importance in so far as they show how vast that expenditure of energy has been. Doubtless, in considering other questions, the length of these time-intervals is a question of great interest, but it does not directly concern us here.

Let us, however, follow Mr. Croll in recognising the possibility that, in some of the cases we have to deal with, the rate of denudation may have been greater than the average rate inferred from the consideration of river drainage. To prevent the possibility of over-estimating the periods necessary to effect the observed denudation, let us assume the rate to have been double the average rate, or equal to one foot in 3,000 years.

At this rate a thickness of three miles which (at the very least) has been swept away in South Scotland since the old red sandstone period would require 45 million years!

But, older than the old red sandstone rocks, the silurian formations have been denuded in places to depths of thousands of feet before the old red sandstone strata were deposited. And these ancient formations were themselves deposited in the ocean by the slow denudation of the Cambrian rocks. These in turn had been formed from the earlier Laurentian strata. And lastly (so far as the researches of geologists at present extend), the Laurentian rocks themselves were built up from the ruins of other rocks which were themselves sedimentary rocks, not the actual primary rocks of our globe. We should almost certainly

under-estimate the period required for these processes of denudation preceding the old red sandstone era, if we assumed that it was only equal in length to the period which has elapsed since that era. But making this assumption, and assuming also (which is also almost certainly an under estimate) that the interval which has elapsed since the old red sandstone era is 45 million years, we find a total of 90 million years for the stratified rocks. In other words, we find at least 90 million years for the period during which rain has fallen on the earth as at present. During that time, therefore, the sun has poured his rays upon the ocean, raising up their waters by evaporation to be carried by winds (also generated by the sun) over the continents, and there discharged in the form of rain and snow.

It may be noticed, in passing, that Sir William Thomson infers, from the observed underground temperature of our earth, that the consolidation of the crust cannot have taken place less than 20 million years ago, or the underground temperature would have been greater than it is; nor more than 400 million years ago, or the underground temperature would have been less than it is. The limits are rather wide; but a value well within these limits would accord with Mr. Croll's estimate of 90 million years as the interval since the earliest stratified rocks were deposited.

Now, the difficulty thus raised is this:-At present we know of no way in which the sun could have emitted the same amount of heat as at present for anything like this period of 90 million years, without having shrunk to much smaller dimensions than he at present has.

It is generally admitted by physicists and astronomers that the solar heat has had its origin in the main, almost wholly in fact, in processes of contraction; and that it is maintained by such processes. In other words, the gravitation of the sun's mass has given birth to all, or very nearly all, the heat which the sun has emitted in the past, and will continue to emit till the end of his career as a sun. It was once supposed that meteoric downfall on the sun's surface produced the chief share of the solar heat. The idea has now been generally abandoned for reasons into which we need not here enter. But, practically, it is of no importance whether we consider the sun's heat to have been generated by the downfall of masses on his surface (continually fed by such downfall) or by the gradual contraction of the entire mass now constituting his globe, till it had assumed its present dimensions. This is the accepted form of the gravitation theory of the solar heat.

But manifestly, the greatest possible amount of heat which could have been generated in this way would be that produced by the contraction of a great nebulous mass containing all the sun's present substance, from an original extension throughout an infinitely large space to the present dimensions of the sun. It might be supposed, perhaps, that the result of such a process of contraction would be the generation of an infinite amount of heat. But in reality this is not the

case, any more than it is the case that a meteoric mass allowed to fall from an infinite distance upon the sun would strike his surface with infinite velocity (after a journey lasting an infinite time-which, however, is a mere detail). We know, on the contrary, the rate at which such a mass would strike the sun-namely, about 360 miles per second. And precisely as we can calculate the velocity of such a mass after being subjected to the sun's pull over an infinitely long journey, so we can calculate the total amount of heat which would result from the contraction of the sun's mass to its present dimensions from a former extension throughout infinite space. We find that it corresponds only to about 20 million years' supply at his present rate of emission.

Thus, while the earth seems to tell us that the sun has been pouring his rays upon her at the same rate as at present during at least 90 million years, the sun seems to tell us that he has not been pouring out heat at that rate for more than 20 million years.

Even if we reject the earth's evidence, or if we endeavour to show that the rainfalls by which the earth's surface has been again and again denuded were not always due to solar heat, but may have been generated by the earth's own heat, we scarcely find our difficulty removed. For it seems utterly absurd to suppose that the mighty central orb of the solar system only attained its present activity during the comparatively recent years of the history of our earth, one of the smaller and shorter-lived members of the sun's family. Sir W. Thomson has shown, by the most satisfactory of the three methods he employed to shorten the estimates formed respecting the earth's duration, that more than 20 million years must have elapsed since her crust was formed-a time which certainly followed by many millions of years the actual genesis of the earth as a gaseous mass. Many physicists reject even the 400 million years given by Thomson as the superior limit, doubting whether the formula and data he employed could be relied upon as confidently as the various processes of mathematical calculation which he applied to them. But even if we accept his minimum result-certainly the very least which science will permit us to accept-it would still follow that the sun's present emission of light and heat could not have continued throughout the time of our earth's existence as a planet; if the sun's heat had its origin entirely or chiefly in those processes of contraction combined with meteoric indraught in which astronomers and physicists at present believe, and if the space into which the sun's mass has contracted is really that which the sun we see appears to occupy.

Mr. Croll, who passes over the latter consideration with the remark that if the sun's density increases towards the centre the supply of solar heat might be somewhat greater, suggests, as the true explanation of the difficulty, that the sun may have derived a portion of his energies in another way than merely through the process of contraction. "In proving that the antiquity of our habitable globe may be far greater than 20 or 30 million years, we prove," he says, "that there must have

been some other source in addition to gravity from which the sun derived his store of energy."

:

He goes back to the initial state of things conceived by Laplace in presenting what is usually called the nebular hypothesis of the solar system. According to this, the whole of the solar system was formerly a great gaseous mass; but whether cold or hot Laplace did not say. As Helmholtz remarks, "The chemical forces must have been present, and ready to act; but, as these forces could only come into operation by the most intimate contact of the different masses, condensation must have taken place before the play of chemical forces began whether a still further supply of force in the shape of heat was present at the commencement we do not know." Mr. Croll, who regards the chemical forces as equivalent only to a few thousand years' supply of heat, and, therefore, as comparatively insignificant, thinks we may safely infer that the original nebulous mass was intensely heated, and that in such intense heat we may find the explanation of the problem before us. "It is evident," he says, "that if we admit that the nebulous mass was in a state of incandescence prior to condensation, it will really be difficult to fix any limit either to the age of the sun, or to the amount of heat which it may have originally possessed. The 20 million years' heat obtained by condensation may in such a case be but a small fraction of the total quantity possessed by the mass."

He

But then the question arises, Whence did the nebulous mass derive its heat? Mr. Croll considers that we may find a satisfactory answer to this question in the assumption that the nebulous mass was formed by the collision of two bodies, each of half the mass of the sun, rushing full tilt upon each other with a velocity of nearly 500 miles per second. Their concussion would generate enough heat to last more than 50 millions of years, which we should have to add to the 20 millions of years provided for by the subsequent condensation of the mass. asks: " Why may not the sun have been composed of two such bodies? and why may not the original store of heat possessed by him have all been derived from the concussion of these two bodies? Two such bodies coming into collision with that velocity would be dissipated into vapour and converted into a nebulous mass by such an inconceivable amount of heat as would thus be generated; and when condensation on cooling took place, a spherical mass like that of the sun would result."

It will be asked, Mr. Croll says (and certainly it seems likely), "Where did the two bodies get their velocity?" It may as well be asked, he answers, "Where did they get their existence? It is just as easy to conceive that they always existed in motion as to conceive that they always existed at rest." At first sight it might seem a fair rejoinder to this to say that, if we are free to assign these enormous velocities to bodies in space, we must be free also to assign to them other properties such as matter can possess-heat, for instance: so that we might solve our problem at once by saying that the nebulous mass was originally