_Friction is in effect an equivalent force exerted in a direction opposite to that in which the sliding occurs._ Its whole amount is the product of two factors: the first of these, which sums up the effect of the nature and condition of the surfaces, is called _the coefficient of friction_; the second, which is the sum of all pressures, as weight strain, and the adhesion due to magnetism (when employed), which act to urge the two bodies together, _i. e._, perpendicularly to the surface of contact, is called _the normal pressure_. But this law holds only where, with dry surfaces, the pressure is not enough to indent or abrade either; or, with wet surfaces, not enough to force out the unguent. In either of these cases, the friction increases more rapidly than the ratio of normal pressure. _No surfaces can be made absolutely hard or smooth_; when one surface is made to slide over another, the slight roughness of the one interlock with those of the other, so that the surfaces must be separated or the points abraded to allow of the motion; but if one surface roll upon another, the prominent points are successively raised, without the need of complete lifting of the body or wearing off those points. _Hence, there are two kinds of friction, the sliding and the rolling._ The former of these in amount greatly exceeds the latter; it is a leading element in the stability of structures and fabrics of all lands, and the most important resistance and source of waste in all machinery, and is therefore a chief object of regard in the arts of construction and the science of engineering. Sliding friction increases with the roughness of the surfaces in contact; hence, it is in practice diminished as these surfaces become worn, also by polishing, and by the use of lubricants, which smooth the rubbing surfaces by filling their depressions. It increases, almost universally, in exact proportion with the entire pressure, owing to weight or other causes, with which the two surfaces are held together; but at very great pressures, somewhat less rapidly. Consequently, in all ordinary cases, so long as the entire weight or pressure remains the same, the friction is, in general, entirely independent of the extent of the surfaces in contact. The exceptions are, some increase when the rubbing surfaces under the same total pressure are very greatly extended, or when either surface is comparatively soft; and considerable lessening of friction when the bodies are very small, as in the runners of skates upon ice. For ordinary rates of motion, the total friction within a given space or distance is in like manner entirely independent of the velocity with which one surface is caused to move over the other; but in very slow motions it is increased, and in very rapid motions perceptibly diminished.