_Friction is also increased in proportion to the tendency of the surfaces to adhere_; hence, it is usually found greater between bodies of the same kind (steel on steel proving almost an exception) than between those of different kinds; it is usually greater when the surfaces have been long in contact, and at the beginning of motion, and always so, unless corrected by lubricants, between metallic surfaces so highly polished that air may be excluded from between them. The frictional resistance retarding the flow of _water_ is subject to three laws: 1. It is proportioned to the amount of surface in contact. 2. It is independent of the pressure. 3. It is proportional to the square of the velocity. It should be remembered that the laws relating to friction, between _solid bodies_ operate quite differently from what they do when applied to liquids; hence, the large mass of data relating to the general subject of friction must be disregarded in the consideration of hydromechanics and allied subjects. For all fluids, whether liquids or gaseous, the resistance is independent of the pressure between the masses in contact. This is in accordance with the second law as stated. The friction for all fluids (liquid or gaseous) is in proportion to the area of the rubbing surface; this follows from the first law and as the sectional area of a circle is the least, _pipes_ from their circular form present the smallest resistance to the flow of water. From the third law, in practice we desire the making of pipes as large as possible; experiment having proved that low speeds are preferable to moderate and still more so, as compared to high speeds in proportioning the piping of hydraulic apparatus. Friction of fluids is also independent of the nature of the solid against which the stream may flow, but dependent to some extent upon their degree of roughness. Friction _for all fluids is also proportional to the density of the fluid_, and related in some way to its adhesiveness. _Water flowing through a pipe tends to drag the pipe along with it_ on account of friction; in all actual fluids there is viscosity or internal friction, but if the relative motion is only slow enough it makes little difference whether the fluid is viscous or not. _Ordinary fluids will change in shape under the action of a force_, however small, if enough time is given for the change to take place, and the rate of change of shape is a measure of its viscosity. _The laws of friction, both for solids and liquids, have been established from experiments endlessly varied._ In investigating these principles we first proceed on the supposition that the forces in question act without any impediments. Great simplicity is attained by first bringing the subject to this ideal standard of perfection, and afterwards making suitable allowances for all these causes which operate in any given case to prevent the perfect application of the law.