The conversion of the igneous liquid surface into solid matter, could only have taken place in successive shells or concentric layers; hence would arise a stratified character. And as the cooling proceeded, lowering the mean temperature of the whole mass, a consequent diminution of bulk must have taken place, according to the well known law of expansion by heat and contraction on cooling. Such diminution in bulk must have broken the strata into fragments, through the fissures of which, according to the laws of hydrostatics, the fluid mass beneath would rise until the equilibrium of rotation would have been obtained, and the strata, originally concentric, would be dislocated and turned in every possible direction, pierced with veins and dikes of all possible magnitude, from slender threads to mountain masses, caused by the cooling and consolidation of the rising fluid, and occasionally spreading in overlying currents, congealed and fixed in ridges and chains. These veins and dykes would present different characters, according to the dates of their elevation. If raised at a period when the surface was still of high temperature, they must have crystallized slowly, and in a perfect manner; at diminished temperatures, the crystallization would be less complete; if raised into the mass of ocean, they would assume one character; if coming in contact with air, another. A breaking of the bed of the ocean, and bringing its waters in contact with the liquid mass beneath, might produce consequences extending in their action to districts of the globe, the most remote from those in which the convulsion occurred; for the water, rising into vapour, would tend to extend itself in one uniform atmosphere over the whole surface of the globe, and might be precipitated in unusual abundance wherever causes of condensation existed. Thus, partial, or even total deluges, may have occurred, great portions of the ocean being hurried in vapour from its bed, and precipitated upon the land whose temperature is not affected by the distant catastrophe.
The waters might, in some cases, flow directly back to the ocean, in others might accumulate in basins and form lakes, fresh at first, and gradually becoming saline. These in turn might burst their bounds, carrying ruin and devastation in their course, or might by evaporation be dried up, and be again filled by a recurrence of the original cause of supply.
Such violent and rapid action would finally be exhausted by the gradual cooling of the earth, but the outer crust would still press on the igneous fluid beneath, and although far less liable to rupture, its fluid action might yet enable it to force its way occasionally to the surface, but at distant intervals, and with diminished energy. Now, a new series of phenomena must occur, similar to the more familiar of those we see acting at present; at first more intense, but finally, when the state of equilibrium of temperature is reached, exactly such as we now find them both in kind and in energy.
To see how far such a view of what might have occurred, under the action of well known causes, in case of a certain original order of things, is correct, let us examine the appearances our globe actually presents.
To a systematized and general examination, it presents the appearance of a great ocean, covering about three-fourths of its whole surface, and surrounding two great, and a number almost infinite of smaller islands. The two great islands are the old and the new continents; the largest of those that remain is New-Holland. To exhibit this great ocean in its most general aspect, take an artificial globe, raise the south pole 50° above the horizon, and bring New-Zealand to the meridian. The hemisphere above the horizon will now be wholly of water, with the exception of the southern part of South America on the one side, and New-Holland, with the Indian archipelago, on the other. These bear, when united, but a small proportion to the entire hemisphere. The opposite hemisphere contains more land than water; and when it is in its turn placed above the horizon, the Atlantic will be seen lying almost wholly on the western side of the meridian, and forming, with the Arctic ocean, a species of channel, narrowing from the latitude of the Cape of Good Hope towards the northern pole, and communicating with the great ocean which lies principally in the opposite hemisphere by Behring's straits. On this hemisphere are also seen parts of the Pacific and Indian oceans, which are considerably more than equal in surface to the lands which project into the opposite one.
If we turn our attention to the land, we find it unequal in its surface; and although compared with the whole diameter of the earth, the inequalities be very small, yet, compared with our own stature, they often present an imposing magnitude. These greater elevations are mountains; and we find them sometimes united in chains, sometimes isolated, and at other times uniting to form elevated plains or table lands. These table lands sometimes slope outwards, at others they are surrounded by eminences that prevent the efflux of the waters, or only admit them to pass through apertures made by their own action. Upon our continent, table lands of the latter description are to be found of great magnitude, entering as parts of the great system of the Cordilleras or Andes; in Europe they are rare, but in Tartary, Persia, and in central Africa, they occur, forming regions of great extent. In general, the greater part of the mountains of a continent appear to have a connexion more or less obvious; it has even been conceived that they form the skeleton upon which the rest of the land has been deposited, and which has determined the form of the continent. Thus we speak habitually of chains of mountains. Mountains, however, do not always present a continuous ridge, from which the peaks or more elevated summits rise, but occasionally, the groups we call chains, are composed of separate mountains divided by valleys; such are the mountains of Scotland, of Sweden, and Norway; and such is the general structure of the chain of mountains called in the state of New-York the Highlands, of whose connexion and grouping we shall hereafter speak.
This being understood, namely, that by a chain or ridge of mountains we do not necessarily intend a continuous elevation, the term may be conveniently used in order to express the configuration of mountains. These chains surround or border upon greater or less basins, which are each distinguished by the name of the principal stream that conveys its surface waters to the ocean, or they may, as has been stated, envelop a table land, whence there is no issue for the waters, or no more than a mere passage sufficient to afford them an outlet. Even if a map contain no expression of the position of mountains, we can, by mere inspection of the courses of rivers, determine the lines in which the chains are directed, and, from the size of the rivers, judge in some measure of the elevation of the district. Thus, on inspection of the map of Europe, we find four of its greatest rivers rising at no great distance from each other, the Rhine, the Rhone, the Danube, and the Po; here, then, we might infer a great elevation, and here we accordingly find its highest mountains, the Alps. In another part of this continent, we see the Dwina, the Nieper, and the Volga, diverge from points not far distant from each other, and here accordingly we find an elevated table land, two hundred miles in length by fifty in breadth, marked however by no mountain summits. In central Asia, we see a vast space inclosed by lines joining the sources of a number of mighty rivers, the Indus, the Ganges, the Barrampooter, the Irrawaddy, the Houng Ha, and Kiang Ku, the Amour, the Lena, the Yermisir, and the Oby; accordingly, here we find the greatest table land surrounded by the highest mountains of the globe. Still, however, the instance we have cited of the rivers of Russia shows, that the land whence great rivers take their rise, is not necessarily mountainous; in this case the ascent is almost imperceptible, and the summit offers the aspect of a level and marshy plain. Such also occurs in the famous boundary between the United States and Canada, where the highlands that figured in two successive treaties have disappeared, and in their supposed place has been found a series of swamps.
Attempts have been made to arrange the chains of mountains into connected systems. Of these the most successful is that of Malte-Brun.
"If we draw a line from the centre of Thibet, across Chinese Mongolia towards Ochotsk, and thence towards Cape Tchutscki, the eastern promontory of Asia, this line will in general coincide with a great chain of mountains which runs from the south-west to the north-east, and which every where descends rapidly towards the Indian and Pacific oceans, while on the contrary, it extends itself towards the Frozen ocean in high plains and secondary hills. It is probable that we may some day refer to the same rule the chain of Lapata, called the backbone of the world, in Africa; at any rate this chain runs from the Cape of Good Hope to that of Gardafui, in a direction south-east and north-west, and therefore in nearly the same direction as the great chain of Asia, but we are ignorant of the disposition of the slopes of these mountains. We may regard the mountains of the Happy Arabia, which are both steep and lofty, as the link that connects the mountains of Lapata with the table lands and mountains of Persia, which proceed from the mountains of Thibet.
"If we follow the western coasts of America, from Behring's straits, which hardly form a sensible interruption, to Cape Horn, we find an uninterrupted chain of mountains. From time to time this chain retires a little into the interior, but more frequently it immediately borders upon the great ocean, in immense cliffs, and often by frightful precipices. On the other side of it, the manner in which the lakes discharge themselves, and the direction of the great rivers, show sufficiently, that the surface of America inclines gently towards the Atlantic ocean.