that, if what has already been advanced in regard to the origin of the sun's heat be correct, it will follow that the argument for the recent age of the earth, based upon the assumption that the sun could have derived its store of heat only from the condensation of its mass, must be wholly abandoned, and that, in so far as this argument is concerned, there is no known limit to the amount of heat which the sun may have possessed, or to the time during which it may have illuminated the earth.

Argument from Tidal Retardation.-It is well known that, owing to tidal retardation, the rate of the earth's rotation is slowly diminishing; and it is therefore evident that if we go back for many millions of years, we reach a period when the earth must have been rotating much faster than now. Sir William's argument is,* that had the earth solidified several hundred millions of years ago, the flattening at the Poles and the bulging at the Equator would have been much greater than we find them to be. Therefore, because the earth is so little flattened, it must have been rotating, when it became solid, at very nearly the same rate as at present. And as the rate of rotation is becoming slower and slower, it cannot be so many millions of years back since solidification took place. A few years ago I ventured to point outt what appeared to be a very obvious objection to this argument, viz., that the influence of sub-aërial denudation in altering the form of the earth had been entirely overlooked. It has been proved, as we have seen, that the rocky surface of our globe is being lowered, on an average, by sub-aërial denudation, at the rate of about 1 foot in 6000 years. It follows as a consequence, from the loss of centrifugal force resulting from the retardation of the earth's rotation occasioned by the friction of the tidal wave, that the sea-level must be slowly sinking at the Equator and rising at the Poles. This, of course, tends to protect the polar regions and expose equatorial regions to sub-aërial denudation. Now it is perfectly obvious that unless the sea-level at the Equator has, in consequence of tidal retardation, been sinking during past ages at a greater rate than 1 foot in 6000 years, it is physically impossible the form of our globe could have been very much different from what it is at present, whatever may have been its form when it consolidated, because sub-aërial denudation would have lowered the Equator as rapidly as the sea sank. But in equatorial regions the rate of denudation is no doubt much greater than I foot in 6000 years, because there the

* Trans. Geol. Soc. of Glasgow, vol. iii., p. I.

Nature, August 21, 1872. Climate and Time, p. 335.

rainfall is greater than in the temperate regions. It has been shown that the rate at which a country is being lowered by subaërial denudation is mainly determined not so much by the character of its rocks as by the sediment-carrying power of its river systems. Consequently, other things being equal, the greater the rainfall the greater will be the rate of denudation. We know that the basin of the Ganges, for example, is being lowered by denudation at the rate of about 1 foot in 2300 years; and this is probably not very far from the average rate at which the equatorial regions are being denuded. It is therefore evident that sub-aërial denudation is lowering the Equator as rapidly as the sea-level is sinking from loss of rotation, and that consequently we cannot infer from the present form of our globe what was its form when it solidified. In as far as tidal retardation can show to the contrary, its form, when solidification took place, may have been as oblate as that of the planet Jupiter. There is another circumstance which must be taken into account. The lowering of the Equator, by the transference of materials from the Equator to the higher latitudes, must tend to increase the rate of rotation, or, more properly, it must tend to lessen the rate of tidal retardation.

The argument may be shown to be inconclusive from another consideration. The question as to whether the earth's axis of rotation could ever have changed to such an extent as to have affected the climate of the Poles is at present exciting a good deal of attention. The subject has recently been investigated with great care by Professor Haughton, Mr. George Darwin,† the Rev. J. F. Twisden,‡ and others, and the general result arrived at may be expressed in the words of Mr. G. Darwin :-" If the earth be quite rigid no re-distribution of matter in new continents. could ever have caused the deviation of the Pole from its present position to exceed the limit of about 3°."

*

66

Mr. Darwin has shown that, in order to produce a displacement of the Pole to the extent of only 1° 46′, an area equal to one-twentieth of the entire surface of the globe would have to be elevated to the height of two miles. The entire continent of Europe elevated two miles would not deflect the Pole much over half a degree. Assuming the mean elevation of the continents of Europe and Asia to be 1000 feet, Prof. Houghton calculates that their removal would displace the Pole only 1994 miles.

Proc. Roy. Soc., vol. xxvi., p. 51.

Proc. Roy. Soc., vol. xxv., p. 328.

Paper read before the Geological Society, February 21st, 1877.

It may now be admitted as settled that if the earth be perfectly rigid the climate of our globe could never possibly have been affected by any change in the axis of rotation. But it is maintained that if the earth can yield as a whole, so as to adapt its form to a new axis of rotation, the effects may be cumulative, and that a displacement of the Pole as much as 10° or 15° is possible.*

But then if the earth be able to adapt its form to a change in the axis of rotation, there is no reason why it may not be able to adapt its form to a change in the rate of rotation, and, if so, the flattening at the Poles and the bulging at the Equator would diminish as the rate of rotation diminished, even supposing there were no denudation going on.

Argument from the Secular Cooling of the Earth.-The earth, like the sun, is a body in the process of cooling, and it is evident that if we go back sufficiently far we shall reach a period when it was in a molten condition. Calculating by means of Fourier's mathematical theory of the conductivity of heat, Sir William Thomson has endeavoured to determine how many years must have elapsed since solidification of the earth's crust may have taken place. This argument is undoubtedly the most reliable of the three. Nevertheless, the data on the subject are yet very imperfect, so that no definite and trustworthy result can be arrived at by this means as to the actual age of the earth. In fact this is obvious from the very wide limits assigned by him within which solidification probably took place. "We must," quoting Sir William's own words on the subject, "allow very wide limits on such an estimate as I have attempted to make; but I think we may, with much probability, say that the consolidation cannot have taken place less than 20,000,000 years ago, or we should have more underground heat than

* A displacement of the Pole of less than 15° or 20° would be of very little service in accounting for the warm climate of Greenland during the Miocene and other periods. But a displacement to that extent, even supposing we admit the earth to be yielding, demands a condition of things which few geologists would be willing to grant. When it becomes generally recognised to what an enormous extent the temperature of the Arctic regions is dependent upon ocean currents, the difficulties in understanding how those regions have once enjoyed a temperate climate will disappear. Were the ice removed from Greenland that region would at present enjoy a warm summer, suitable for plant and animal life. It is the presence of ice rather than a positive deficiency of heat that makes Greenland so cold and barren (see Climate and Time, Chap. IV.). An increase in the quantity of heat conveyed by oceancurrents, merely sufficient to prevent the accumulation of ice, would completely transform the climate of the Arctic lands. And such an increase would take place during an Inter-glacial period when the eccentricity of the earth's orbit was at a high value and the winter solstice in perihelion,

we actually have,-nor more than 400,000,000 years ago, or we should not have so much as the least observed underground increment of temperature. That is to say, I conclude that Leibnitz's epoch of 'emergence' of the 'consistentur status' was probably within these dates,"*

III. THE GLACIAL PERIOD IN THE SOUTHERN HEMISPHERE.

By THOMAS BELT, F.G.S.

CHE tablets on which the ice of the Glacial period left its record in the southern hemisphere are probably now mostly covered by the sea, and we cannot trace its progress and extent with the same facility and certainty as in the northern temperate regions. Yet notwithstanding this, and also that the land surfaces of the South that have been glaciated have not been studied to anything like the same extent as in Europe and North America, points of resemblance are apparent, and grounds exist for the belief that both hemispheres have passed through a somewhat similar glacial experience.

In our hemisphere, I have sought to show in former papers, there were two ways in which the ice spread. One was an accumulation on mountain-chains, and a radiation from them over the surrounding country. The other, and I think by far the most important, was the gradual advance of a ridge of ice down the bed of the North Atlantic, and probably also of the North Pacific, which blocked up the drainage of the continents as far as it extended, and caused enormous lakes of fresh or brackish water and immense destruction of life amongst the animals that were caught on the plains by the rising floods. The marks left by the Atlantic ice are seen in Europe, as far as the southern extremity of Ireland, and in America, to the south of New York, and beyond these points, its further progress can be traced by the evidence of the interruption of the drainage of the continents as far as the northern slopes of the Pyrenees on one side of the Atlantic, and to the coasts of

Trans. Roy. Soc. of Edinburgh, vol. xxiii., p. 161.

Virginia on the other. The real markings on the rocks, however, have not been traced in Europe farther south than 51° N. lat., and in America than 40° N. lat., excepting those of the ice that proceeded from the mountain ranges.

In the other hemisphere, within the distance from the Southern Pole that the ice has been shown by actual markings on the rocks to have reached from the North Pole on the European coasts,—that is, to latitude 51° N.,—there are no large masses of land, excepting the extreme end of South America and the Antarctic continent. And if we take a corresponding circle in the south to the limit the ice has left its marks in America or to lat. 40°, we shall still only embrace Patagonia in South America and the Middle Island in New Zealand. Even if we take the furthest limit of the extension of the Atlantic ice, as shown by its interference with the drainage of the American continent, we only bring South America as far north as the Rio Plata, and New Zealand and Tasmania, with the southern end of Australia, within the area where we could expect to find any similar evidence on the supposition that in the Glacial period the ice extended everywhere as far from the Southern Pole as its extreme limit reached from the Northern.

The conditions are therefore very different in the two hemispheres. In the one, broad continents stretch from within the Arctic circle toward and up to the Equator; in the other, nearly the whole of the temperate zone is covered with water. If, then, there were much less evidence than there is of the glaciation of southern lands, we need not have been surprised; but of late there seems to have arisen an idea amongst some geologists that there is no evidence in the southern hemisphere of the occurrence of a Glacial period, and it may be useful if I bring together what has been described, and show how far the phenomena agree with those of the northern hemisphere.

This

Commencing in America, immediately south of the Equator, we have first to deal with the remarkable theory of Agassiz, that the great valley of the Amazon was once filled with ice flowing from the distant Andes, which left an enormous terminal moraine on the Atlantic coast. moraine he supposed blocked up for a time the waters of the great valley, and caused the deposition of various stratified deposits covered by a peculiar drift clay that rarely contains transported boulders. I have examined this deposit from Pernambuco northwards through the provinces of Ceara and Maranham, as far as Para. In some parts it is composed of small angular fragments of rock, cemented together by an