All rocks are naturally divided by joints into blocks, which are frequently more or less regular, and often of quite small size. Water and air penetrate into these cracks and decompose the surfaces of the blocks, and thus the field of their operations is enormously extended. These rock-blocks sometimes show very beautifully the progress of the decomposing agents from the outside inward by concentric layers or shells of rotten material, which, in the larger blocks, often envelop a nucleus of the unaltered rock. It is interesting to observe, too, that these concentric lines of decay cut off the angles of the original blocks, so that the undecomposed nucleus, when it is found, is approximately spherical instead of cuboidal. Both these points are well illustrated by specimen No. 2; for although now nearly spherical, it was originally perfectly angular, and has become rounded by the peeling off, in concentric layers, of the decomposed material, and in most cases several of these layers are distinctly visible, like the coats of an onion. But by stripping these off we should discover, in all the larger balls at least, a solid, spheroidal nucleus, while in the smaller balls the decomposition has penetrated to the centre. In the rocks also we find many imperfect joints and minute cracks. In cold countries these are extended and widened by the expansive power of freezing water, and thus the surfaces of decomposition become constantly greater. Nearly all rocks suffer this chemical decomposition when exposed to the weather, but in some the decay goes on much faster than in others. Diabase is one of the rocks which decay most readily; while granite is, among common rocks, one of those that resist decay most effectually. The caverns which are so large and numerous in most limestone countries are a splendid example of the solvent action of meteoric waters, being formed entirely by the dissolving out of the limestone by the water circulating through the joint cracks. The process must go on with extreme slowness at first, when the joints are narrow, and more rapidly as they are widened and more water is admitted. We get some idea, too, of the magnitude of the results accomplished by these silent and unobtrusive agencies when we reflect that almost all the loose earth and soil covering the solid rocks are simply the insoluble residue which carbon dioxide and water cannot remove. In low latitudes, where a warm climate accelerates the decay of the rocks, the soil is usually from 50 to 300 feet deep. MECHANICAL EROSION.—_On the edge of the land._—Let us trace next the _mechanical_ action of water and air upon the land. First we will consider the _edge_ of the land, where it is washed by the waves of the sea. Whoever has been on the shore must have noticed that the sand along the water’s edge is kept in constant motion by the ebb and flow of the surf.