Of mechanical ones the most familiar is that of the scales. If they be accurately made, and not clogged by dirt or rust, it is impossible to keep a pair of scales perfectly balanced: eventually one scale will descend and the other ascend—they will assume a heterogeneous relation. Again, if we sprinkle over the surface of a fluid a number of equal-sized particles, having an attraction for each other, they will, no matter how uniformly distributed, by and by concentrate irregularly into one or more groups. Were it possible to bring a mass of water into a state of perfect homogeneity—a state of complete quiescence, and exactly equal density throughout—yet the radiation of heat from neighbouring bodies, by affecting differently its different parts, would inevitably produce inequalities of density and consequent currents; and would so render it to that extent heterogeneous. Take a piece of red-hot matter, and however evenly heated it may at first be, it will quickly cease to be so: the exterior, cooling faster than the interior, will become different in temperature from it. And the lapse into heterogeneity of temperature, so obvious in this extreme case, takes place more or less in all cases. The action of chemical forces supplies other illustrations. Expose a fragment of metal to air or water, and in course of time it will be coated with a film of oxide, carbonate, or other compound: that is—its outer parts will become unlike its inner parts. Usually the heterogeneity produced by the action of chemical forces on the surfaces of masses, is not striking; because the changed portions are soon washed away, or otherwise removed. But if this is prevented, comparatively complex structures result. Quarries of trap-rock contain some striking examples. Not unfrequently a piece of trap may be found reduced, by the action of the weather, to a number of loosely-adherent coats, like those of an onion. Where the block has been quite undisturbed, we may trace the whole series of these, from the angular, irregular outer one, through successively included ones in which the shape becomes gradually rounded, ending finally in a spherical nucleus. On comparing the original mass of stone with this group of concentric coats, each of which differs from the rest in form, and probably in the state of decomposition at which it has arrived, we get a marked illustration of the multiformity to which, in lapse of time, a uniform body may be brought by external chemical action. The instability of the homogeneous is equally seen in the changes set up throughout the interior of a mass, when it consists of units that are not rigidly bound together. The atoms of a precipitate never remain separate, and equably distributed through the fluid in which they make their appearance. They aggregate either into crystalline grains, each containing an immense number of atoms, or they aggregate into flocculi, each containing a yet larger number; and where the mass of fluid is