There is, moreover, good reason to believe that the general features of this arrangement are of vast antiquity; and that throughout much of the Tertiary period, at all events, the relative positions of our continents and oceans have remained the same, although they have certainly undergone some changes in their extent, and in the degree of their connection with each other. This is proved by two kinds of evidence. In the first place, it is now ascertained by actual measurement that the depths of the great oceans are so vast over wide areas, while the highest elevations of the land are limited to comparatively narrow ridges, that the mass of land (above the sea-level) is not more than ¹⁄₃₆th part of the mass of the ocean. Now we have reason to believe that subsidence and elevation bear some kind of proportion to each other, whence it follows that although several mountain ranges have risen to great heights during the Tertiary period, this amount of elevation bears no proportion to the amount of subsidence required to have changed any considerable area of what was once land into such profound depths as those of the Atlantic or Pacific Oceans. In the second place, we find over a considerable area of all the great continents fresh-water deposits containing the remains of land animals and plants; which deposits must have been formed in lakes or estuaries, and which therefore, speaking generally, imply the existence in their immediate vicinity of land areas comparable to those which still exist. The Miocene deposits of Central and Western Europe, of Greece, of India, and of China, as well as those of various parts of North America, strikingly prove this; while the Eocene deposits of London and Paris, of Belgium, and of various parts of North and South America, though often marine, yet by their abundant remains of land-animals and plants, equally indicate the vicinity of extensive continents. For our purpose it is not necessary to go further back than this, but there is much evidence to show that throughout the Secondary, and even some portion of the Palæozoic periods, the land-areas coincided to a considerable extent with our existing continents. Professor Ramsay has shown[64] that not only the Wealden formation, and considerable portions of the Upper and Lower Oolite, but also much of the Trias, and the larger part of the Permian, Carboniferous, and Old Red Sandstone formations, were almost certainly deposited either in lakes, inland seas, or extensive estuaries. This would prove that, throughout the whole of the vast epochs extending back to the time of the Devonian formation, our present continents have been substantially in existence, subject, no doubt, to vast fluctuations by extension or contraction, and by various degrees of union or separation, but never so completely submerged as to be replaced by oceans comparable in depth with our Atlantic or Pacific.