are the physical effects which result from an increase of eccentricity, we shall find that a host of physical agencies are brought into operation, the combined effect of which is to lower to a very great extent the temperature of the hemisphere whose winters occur in aphelion, and to raise to nearly as great an extent the temperature of the opposite hemisphere, whose winters of course occur in perihelion. Until attention was directed to those physical circumstances to which I refer, it was impossible that the true cause of the glacial epoch could have been discovered; and, moreover, many of the indirect and physical effects, which in reality were those that brought about the glacial epoch, could not, in the nature of things, have been known previously to recent discoveries in the science of heat.

The consideration and discussion of those various physical agencies are the chief aim of the following pages.

Abstract of the Line of Argument pursued in this Volume.—I shall now proceed to give a brief abstract of the line of argument pursued in this volume. But as a considerable portion of it is devoted to the consideration of objections and difficulties bearing either directly or indirectly on the theory, it will be necessary to point out what those difficulties are, how they arose, and the methods which have been adopted to overcome them.

Chapter IV. contains an outline of the physical agencies affecting climate which are brought into operation by an increase of eccentricity. By far the most important of all those agencies, and the one which mainly brought about the glacial epoch, is the Deflection of Ocean Currents. The consideration of the indirect physical connection between a high state of eccentricity and the deflection of ocean currents, and also the enormous influence on climate which results from this deflection constitute not only the most important part of the subject, but the one beset with the greatest amount of difficulties.

The difficulties besetting this part of the theory arise mainly from the imperfect state of our knowledge, (1st) with reference

to the absolute amount of heat transferred from equatorial to temperate and polar regions by means of ocean currents and the influence which the heat thus transferred has on the distribution of temperature on the earth's surface; and (2nd) in connection with the physical cause of ocean circulation.

In Chapters II. and III. I have entered at considerable length into the consideration of the effects of ocean currents on the distribution of heat over the globe. The only current of which anything like an accurate estimate of volume and temperature has been made is the Gulf-stream. In reference to this stream we have a means of determining in absolute measure the quantity of heat conveyed by it. On the necessary computation being made, it is found that the amount transferred by the Gulf-stream from equatorial regions into the North Atlantic is enormously greater than was ever anticipated, amounting to no less than one-fifth part of the entire heat possessed by the North Atlantic. This striking fact casts a new light on the question of the distribution of heat over the globe. It will be seen that to such an extent is the temperature of the equatorial regions lowered, and that of high temperate, and polar regions raised, by means of ocean currents, that were they to cease, and each latitude to depend solely on the heat received directly from the sun, only a very small portion of the globe would be habitable by the present order of beings. This being the case, it becomes obvious to what an extent the deflection of ocean currents must affect temperature. For example, were the Gulf-stream stopped, and the heat conveyed by it deflected into the Southern Ocean, how enormously would this tend to lower the temperature of the northern hemisphere, and raise the temperature south of the equator.

Chapters VI., VII., VIII., IX., X., and XIII., are devoted to the consideration of the physical cause of oceanic circulation. This has been found to be the most difficult and perplexing part of the whole inquiry. The difficulties mainly arise from the great diversity of opinion and confusion of ideas prevailing

in regard to the mechanics of the subject. There are two theories propounded to account for oceanic circulation; the one which may be called the Wind theory, and the other the Gravitation theory; and this diversity of opinion and confusion of ideas prevail in connection with both theories. As the question of the cause of oceanic circulation has not only a direct and important bearing on the subject of the present volume, but is further one of much general interest, I have entered somewhat fully into the matter.

The Gravitation theories may be divided into two classes. The first of these attributes the Gulf-stream and other sensible currents of the ocean to difference of specific gravity, resulting from difference of temperature between the sea in equatorial and polar regions. The leading advocate of this theory was the late Lieutenant Maury, who brought it so much into prominence in his interesting book on the "Physical Geography of the Sea." The other class does not admit that the sensible currents of the ocean can be produced by difference of specific gravity; but they maintain that difference of temperature between the sea in equatorial and polar regions produces a general movement of the upper portion of the sea from the equator to the poles, and a counter-movement of the under portion from the poles to the equator. This form of the gravitation theory has been ably and zealously advocated by Dr. Carpenter, who may be regarded as its representative. The Wind theories also divide into two classes. According to the one ocean currents are caused and maintained by the impulse of the trade-winds, while according to the other they are due not to the impulse of the trade-winds alone, but to that of the prevailing winds of the globe, regarded as a general system. The former of these is the one generally accepted; the latter is that advocated in the present volume.

The relations which these theories bear to the question of secular change of climate, will be found stated at length in Chapter VI. It will, however, be better to state here in a few words what those relations are. When the eccentricity of the

earth's orbit attains a high value, the hemisphere, whose winter solstice occurs in aphelion, has, for reasons which are explained in Chapter IV., its temperature lowered, while that of the opposite hemisphere is raised. Let us suppose the northern hemisphere to be the cold one, and the southern the warm one. The difference of temperature between the equator and the North Pole will then be greater than between the equator and the South Pole; according, therefore, to theory, the trades of the northern hemisphere will be stronger than those of the southern, and will consequently blow across the equator to some distance on the southern hemisphere. This state of things will tend to deflect equatorial currents southwards, impelling the warm water of the equatorial regions more into the southern or warm hemisphere than into the northern or cold hemisphere. The tendency of all this will be to exaggerate the difference of temperature already existing between the two hemispheres. If, on the other hand, the great ocean currents which convey the warm equatorial waters to temperate and polar regions be not produced by the impulse of the winds, but by difference of temperature, as Maury maintains, then in the case above supposed the equatorial waters would be deflected more into the northern or cold hemisphere than into the southern or warm hemisphere, because the difference of temperature between the equator and the poles would be greater on the cold than on the warm hemisphere. This, of course, would tend to neutralize or counteract that difference of temperature between the two hemispheres which had been previously produced by eccentricity. In short, this theory of circulation would effectually prevent eccentricity from seriously affecting climate.

Chapters VI. and VII. have been devoted to an examination of this form of the gravitation theory.

The above remarks apply equally to Dr. Carpenter's form of the theory; for according to a doctrine of General Oceanic Circulation resulting from difference of specific gravity between the water at the equator and at the poles, the equatorial water will

be carried more to the cold than to the warm hemisphere. It is perfectly true that a belief in a general oceanic circulation may be held quite consistently with the theory of secular changes of climate, provided it be admitted that not this general circulation but ocean currents are the great agency employed in distributing heat over the globe. The advocates of the theory, however, admit no such thing, but regard ocean currents as of secondary importance. It may be stated that the existence of this general ocean circulation has never been detected by actual observation. It is simply assumed in order to account for certain facts, and it is asserted that such a circulation must take place as a physical necessity. I freely admit that were it not that the warm water of equatorial regions is being constantly carried off by means of ocean currents such as the Gulfstream, it would accumulate till, in order to restoration of equilibrium, such a general movement as is supposed would be generated. But it will be shown that the warm water in equatorial regions is being drained off so rapidly by ocean currents that the actual density of an equatorial column differs so little from that of a polar column that the force of gravity resulting from that difference is so infinitesimal that it is doubtful whether it is sufficient to produce sensible motion. I have also shown in Chapter VIII. that all the facts which this theory is designed to explain are not only explained by the wind theory, but are deducible from it as necessary consequences. In Chapter XI. it is proved, by contrasting the quantity of heat conveyed by ocean currents from inter-tropical to temperate and polar regions with such an amount as could possibly be conveyed by means of a general oceanic circulation, that the latter sinks into insignificance before the former. In Chapters X. and XII. the various objections which have been advanced by Dr. Carpenter and Mr. Findlay are discussed at considerable length, and in Chapter IX. I have entered somewhat minutely into an examination of the mechanics of the gravitation theory. A statement of the wind theory is given in Chapter XIII.; and in Chapter XIV. is shown the relation of this theory to the