Water is chiefly used in tanneries for soaking and washing hides and skins, for making the limes, the bates, and the tanning liquors, for steam boilers, and in dyeing. For all these purposes it should be as free as possible from impurities, but since water is the most universal solvent in Nature, it is never found pure, but always contains mineral matter derived from the rocks and soil through which it has flowed, as well as organic impurities from decaying animal and vegetable matter. Associated with the latter are usually living organisms of putrefaction (_bacteria_) which may affect the quality of the water for tanning even more seriously than the mineral impurities. The purest natural waters are those which have flowed only over hard sandstones and volcanic rocks. Water sufficiently pure for laboratory use can only be obtained by distillation. The steam-water from heating pipes usually contains large quantities of dissolved iron, and often also volatile organic matters from the oil, etc., which finds its way into the boiler. It may sometimes be made fit for use by boiling (which precipitates the ferrous carbonate present), and subsequent settling or filtration. The use of steam-water containing iron is a frequent source of stains and discolorations in the tannery which more than counterbalances the advantage of its softness. The “hardness” of natural waters is mostly due to the salts of lime and magnesia which they contain, which precipitate soap in the form of insoluble stearates and oleates, which are useless for washing. It is commonly estimated by determining the amount of a standard alcoholic soap solution which must be added in order to produce a permanent froth on shaking. Theoretically about 12 parts of soap (sodium stearate or oleate) are destroyed by 1 part of calcium carbonate or an equivalent quantity of other lime salts, with formation of insoluble lime soaps (calcium stearate or oleate). Really, the reaction is much more complicated, owing to the dissociation of the soap into free alkali and acid-salts on solution in water. Teed[61] estimates that ¹⁄₃ to ¹⁄₂ more is required than the theoretical quantity, and more in hot water than cold. This uncertainty is partially overcome by testing the soap solution against a known solution of calcium chloride. The presence of magnesia also complicates the test and leads to discrepant results. [61] Journ. Soc. Chem. Ind., 1889, p. 256. Cp. also Allen, ibid. 1888, p. 795. The methods of determining hardness originated by Hehner (see L.I.L.B., p. 19) are simpler and more accurate than the soap-test, and are to be preferred, except for direct determination of the suitability of a water for scouring with soap. “Degrees” of hardness in England are calculated as parts of CaCO₃ per 100,000, or sometimes grains per gallon (70,000 grains). Hardness is of two kinds, “temporary” and “permanent”; the former being removed by boiling, while the latter is not so removed.