estimated, or the

rate of their formation has been underestimated, or both. If it be maintained that a foot in 14,400 years is too slow a rate of deposit, then it must be maintained that the land must have been denuded at a greater rate than one foot in 6,000 years. But most geologists probably felt surprised when the announcement was first made, that at this rate of denudation the whole existing land of the globe would be brought under the ocean in 6,000,000 of years.

The error, no doubt, consists in over-estimating the thickness of the sedimentary rocks. Assuming, for physical reasons already stated, that 100,000,000 years limits the age of the stratified rocks, and that the proportion of land to water and the rate of denudation have been on the average the same as at present, the mean thickness of sedimentary rocks formed in the 100,000,000 years amounts to only 7,000 feet.

But be it observed that this is the mean thickness on an area equal to that of the ocean. Over the area of the globe it amounts to only 5,000 feet; and this, let it be observed also, is the total mean thickness formed, without taking into account what has been removed by denudation. If we wish to ascertain what is actually the present mean thickness, we must deduct from this 5,000 feet an amount of rock equal to all the sedimentary rocks which have been denuded during the 100,000,000 years; for the 5,000 feet is not the present mean thickness, but the total mean thickness formed during the whole of the 100,000,000 years. If we assume, what no doubt most geologists would be willing to grant, that the quantity of sedimentary rocks now remaining is not over one-half of what has been actually deposited during the history of the globe, then the actual mean thickness of the stratified rocks of the globe is not over 2,500 feet. This startling result would almost necessitate us to suspect that the rate of subaërial denudation is probably greater than one foot in 6,000 years. But, be this as it may, we are apt, in estimating the mean thickness of the stratified rocks of the globe from their ascertained maximum thickness, to arrive at erroneous conclusions. There are considerations

which show that the mean thickness of these rocks must be small in proportion to their maximum thickness. The stratified rocks are formed from the sediment carried down by rivers and streamlets and deposited in the sea. It is obvious that the greater quantity of this sediment is deposited near the mouths of rivers, and along a narrow margin extending to no great distance from the land. Did the land consist of numerous small islands equally distributed over the globe, the sediment carried off from these islands would be spread pretty equally over the sea-bottom. But the greater part of the land-surface consists of two immense continents. Consequently, the materials removed by denudation are not spread over the ocean-bottom, but on a narrow fringe surrounding those two continents. Were the materials spread over the entire ocean-bed, a foot removed off the general surface of the land would form a layer of rock only five inches thick. But in the way in which the materials are at present deposited, the foot removed from the land would form a layer of rock many feet in thickness. The greater part of the sediment is deposited within a few miles of the shore.

The entire coast-line of the globe is about 116,500 miles. I should think that the quantity of sediment deposited beyond, say, 100 miles from this coast-line is not very great. No doubt several of the large rivers carry sediment to a much greater distance from their mouths than 100 miles, and ocean currents may in some cases carry mud and other materials also to great distances. But it must be borne in mind that at many places within the 100 miles of this immense coast-line little or no sediment is deposited, so that the actual area over which the sediment carried off the land is deposited is probably not greater than the area of this belt-116,500 miles long and 100 miles broad. This area on which the sediment is deposited, on the above supposition, is therefore equal to about 11,650,000 square miles. The amount of land on the globe is about 57,600,000 square miles. Consequently, one foot of rock, denuded from the surface of the land and deposited on this

belt, would make a stratum of rock 5 feet in thickness; but were the sediment spread over the entire bed of the ocean, it would form, as has already been stated, a stratum of rock of only 5 inches in thickness.

Suppose that no subsidence of the land should take place for a period of, say, 3,000,000 of years. During that period 500 feet would be removed by denudation, on an average, off the land. This would make a formation 2,500 feet thick, which some future geologist might call the Post-tertiary formation. But this, be it observed, would be only the mean thickness of the formation on this area; its maximum thickness would evidently be much greater, perhaps twice, thrice, or even four times that thickness. A geologist in the future, measuring the actual thickness of the formation, might find it in some places 10,000 feet in thickness, or perhaps far more. But had the materials been spread over the entire ocean-bed, the formation would have a mean thickness of little more than 200 feet; and spread over the entire surface of the globe, would form a stratum of scarcely 150 feet in thickness. Therefore, in estimating the mean thickness of the stratified rocks of the globe, a formation with a maximum thickness of 10,000 feet may not represent more than 150 feet. A formation with a mean thickness of 10,000 feet represents only 600 feet.

It may be objected that in taking the present rate at which the sedimentary deposits are being formed as the mean rate for all ages, we probably under-estimate the total amount of rock formed, because during the many glacial periods which must have occurred in past ages the amount of materials ground off the rocky surface of the land in a given period would be far greater than at present. But, in reply, it must be remembered that although the destruction in ice-covered regions would be greater during these periods than at present, yet the quantity of materials carried down by rivers into the sea would be less. At the present day the greater part of the materials carried down by our rivers is not what is being removed off the rocky face of the country, but the boulder clay, sand, and other

materials which were ground off during the glacial epoch. It is therefore possible, on this account, that the rate of deposit may have been less during the glacial epoch than at present.

When any particular formation is wanting in a given area, the inference generally drawn is, that either the formation has been denuded off the area, or the area was a land-surface during the period when that formation was being deposited. From the foregoing it will be seen that this inference is not legitimate; for, supposing that the area had been under water, the chances that materials should have been deposited on that area are far less than are the chances that there should not. There are sixteen chances against one that no formation ever existed in the area.

If the great depressions of the Atlantic, Pacific, and Indian Oceans be, for example, as old as the beginning of the Laurentian period-and they may be so for anything which geology can show to the contrary-then under these oceans little or no stratified rocks may exist. The supposition that the great ocean basins are of immense antiquity, and that consequently only a small proportion of the sedimentary strata can possibly occupy the deeper bed of the sea, acquires still more probability when we consider the great extent and thickness of the Old Red Sandstone, the Permian, and other deposits, which, according to Professor Ramsay and others, have been accumulated in vast inland lakes.

17

CHAPTER XXIII.

THE PHYSICAL CAUSE OF THE SUBMERGENCE AND EMERGENCE OF THE LAND DURING THE GLACIAL EPOCH.

Displacement of the Earth's Centre of Gravity by Polar Ice-cap.-Simple Method of estimating Amount of Displacement.-Note by Sir W. Thomson on foregoing Method.-Difference between Continental Ice and a Glacier.— Probable Thickness of the Antarctic Ice-cap.-Probable Thickness of Greenland Ice-sheet. The Icebergs of the Southern Ocean.-Inadequate Conceptions regarding the Magnitude of Continental Ice.

Displacement of the Earth's Centre of Gravity by Polar Ice-cap.* -In order to represent the question in its most simple elementary form, I shall assume an ice-cap of a given thickness at the pole and gradually diminishing in thickness towards the equator in the simple proportion of the sines of the latitudes, where at the equator its thickness of course is zero. Let us assume, what is actually the case, that the equatorial diameter of the globe is somewhat greater than the polar, but that when the ice-cap is placed on one hemisphere the whole forms a perfect sphere.

I shall begin with a period of glaciation on the southern hemisphere. Let W NE S' (Fig. 5) be the solid part of the earth, and c its centre of gravity. And let ES W be an icecap covering the southern hemisphere. Let us in the first case assume the earth to be of the same density as the cap. The earth with its cap forms now a perfect sphere with its

The conception of submergence resulting from displacement of the earth's centre of gravity, caused by a heaping-up of ice at one of the poles, was first advanced by M. Adhémar, in his work "Révolutions de la Mer," 1842. When the views stated in this chapter appeared in the Reader, I was not aware that M. Adhemar had written on the subject. An account of his mode of viewing the question is given in the Appendix.