Veins have been filled with substances foreign to them, which have probably been disseminated in atoms in the adjacent rocks or by sublimation. Nothing can be more certain than that the minute particles of matter are constantly in motion from the action of heat, mutual attraction, and electricity. Prismatic crystals of salts of zinc are changed in a few seconds into crystals of a totally different form by the heat of the sun: casts of shells are found in rocks, from which the animal matter has been removed, and its place supplied by mineral matter; and the excavations made in rocks diminish sensibly in size in a short time if the rock be soft, and in a longer time when it is hard— circumstances which show an intestine motion of the particles, not only in their relative positions, but in space, which there is every reason to believe is owing to electricity—a power which, if not the sole agent, must at least have co-operated essentially in the formation and filling of mineral veins.[93] The magnetism of the earth is presumed to be owing to electrical currents circulating through its surface in a direction at right angles to the magnetic meridians. Mr. Fox, so well known in the scientific world, has long since shown, from observations in the Cornish mines, that such currents do flow through all metallic veins. Now, as the different substances of which the earth is composed are in different states of electro-magnetism, and are often interrupted by non-conducting rocks, the electric currents, being stopped in their course, act chemically on all the liquids and substances they meet with. Hence, Mr. Fox has come to the conclusion that not only the nature of the deposits must have been determined by their relative electrical conditions, but that the direction of the metallic veins themselves must have been influenced by the direction of the magnetic meridians; and, in fact, almost all the metallic deposits in the world are in parallel veins or fissures tending from east to west, or from north-east to south-west. Veins at right angles to these are generally non-metalliferous, and, if they do contain metallic ores, they are of a different kind. In some few cases both contain the same ore, but in very different quantities, as in the silver-mine at Pasco, in the Andes, and both veins are richer near the point of crossing than elsewhere.