If the ocean could ever have been above its present level sufficiently to have covered all the sedimentary rocks, we might assume that the height of mountains has not been changed. But the level of the ocean cannot be subject to much variation. The total amount of water on the globe is always the same. If the continents and mountains were all submerged at once, and the waters were expanded by the highest temperature consistent with the liquid form, there would not be a change of level of more than two hundred and fifty feet. We may assume, then, that the ocean level has always been essentially the same that it now is. We must therefore conclude that the sedimentary rocks, and the mountains of which they form a part, have been elevated to their present position from the bed of the sea. Different mountain ranges have been elevated at different periods. The silurian and carboniferous formations were deposited before the Alleghany Mountains, which they contributed to form, were elevated; while the new red sandstone and the cretaceous and tertiary formations were deposited subsequently to the upheaval. They are accordingly found at the base of the range, nearly horizontal, and have risen above the level of the ocean only as the continent generally has risen. The Pyrenees were elevated after the deposition of the cretaceous rocks, and have carried them up so that they appear at a high angle, while the tertiary rocks at the base are horizontal, as in the United States. The Andes have carried up the tertiary rocks with them, and their elevation must therefore belong to a recent period. It appears that they are even yet rising. It has recently been shown that the Alps have been subjected to upheaval at several distinct periods. At the close of the silurian period they formed a cluster of islands. At the commencement of the tertiary period they became a mountain range, and at the close of that period they were thrown up some two thousand feet higher, to their present position. Nearly the same things will probably be found true of other mountain ranges, when their structure has been minutely studied. The elevation of contiguous parallel ridges will necessarily leave intervening _valleys of elevation_. As mountain ranges generally consist of several such ridges, valleys of this description are numerous, and they are often of great extent.