A rock is said to undergo a chemical change, when one or more of its component parts passes from one state of combination to another--as, for instance, when a mineral absorbs oxygen, and, from the condition of a protoxide, changes into that of a peroxide; or when, parting with its silicic acid, it takes an equivalent amount of carbonic acid, and in place of a silicate becomes a carbonate. Now these, and similar metamorphisms, are chiefly produced by water permeating through the rocky mass, and thus no sooner does the old combination cease, than the new one which replaces it is dissolved by the slowly filtering water, and carried away either to greater depths, or to the surface. Every drop of water, therefore, that finds its way through the rock, carries away an infinitesimal portion of the mineral matter, and the stone is consequently undergoing a continual decay. This condition of things may go on either at some depth in the earth's crust, or on the surface. In the former case, springs and percolating water are the agents in effecting the change; in the latter, it is produced chiefly by rain and streams. But wherever the process goes on, the results, unless where counteracted by some opposite agency, are ultimately the same. It may be of use to look at some examples of these changes, and, by dividing them in a rough way, into underground and surface actions, we shall be enabled to mark more clearly their effects. A common source of the decay of rocks arises from the percolation through them of water charged with carbonic acid. Decomposing vegetation gives off a large amount of this gas, which is readily absorbed by rain-water. The water sinks into the ground filtering through cracks and fissures in the rocks, whence it afterwards re-emerges in the form of springs. Now wherever, in its passage through these subterranean rocks, the water meets with any carbonate, the carbonic acid contained in the liquid immediately begins to dissolve out the mineral matter, and carries it eventually to the surface. There the amount of evaporation is often sufficient to cause a re-deposit of the mineral in solution. If it be lime, a white crust gathers along the sides of the stream, delicately enveloping grass-stalks, leaves, twigs, snail-shells, and other objects, which it may meet with in its progress. Such "petrifying" springs, as they are popularly termed, occur abundantly in our limestone districts. It should be borne in mind, however, that they only produce an incrustation round the organic nucleus, and do not petrify it. That alone is a true petrifaction where the substance is literally fossilized, or turned into stone. A familiar instance of a similar chemical process may be seen under many a bridge, and along the vaulted roofs of many an old castle. Numerous tapering stalactites hang down from between the joints of the masonry, resembling, so to speak, icicles of stone, often of a dazzling whiteness.