"The sea," continues he, "therefore, covered the whole earth, and from thence it proceeds that all the beds of stone in the plains are horizontal and parallel; fish must have also been the most ancient inhabitants of the globe, as there was no sustenance for either birds or terrestrial animals." But how did the sea retire into these vast basins which it at present occupies? What presents itself the most natural to the mind is, that the earth, at least at a certain depth, was not entirely solid, but intermixed with some great vacuums, whose vaults were supported for a time, but at length, sunk in suddenly: then the waters must have fallen into these vacancies, filled them, and left naked a part of the earth's surface, which became an agreeable abode to terrestrial animals and birds. The shells found in quarries perfectly agree with this idea, for only the bony parts of fish could be preserved till now. In general, shells are heaped up in great abundance in certain parts of the sea, where they are immovable, and form a kind of rock, and could not follow the water, which suddenly forsook them: this is the reason that we find more shells than bones of the fish, and this even proves a sudden fall of the sea into its present basins. At the same time as our supposed vaults gave way, it is very possible that other parts of the globe were raised by the same cause, and that mountains were placed on this surface with quarries already formed, but the beds of these quarries could not preserve the horizontal direction they before had, unless the mountains were raised precisely perpendicular to the surface of the earth, which could happen but very seldom: so also, as we have already observed, in 1705, the beds of stone in mountains are always inclined to the horizon, though parallel with each other; for they have not changed their position with respect to each other, but only with respect to the surface of the earth[205:A].