substances in solution are not transformed into vapours in this process. Such distilled water is prepared by chemists and in laboratories by boiling water in closed metallic boilers or stills, and causing the steam produced to pass into a condenser—that is, through tubes (which should be made of tin, or, at all events, tinned, as water and its impurities do not act on tin) surrounded by cold water, and in which the steam, being cooled, condenses into water which is collected7 in a receiver. By standing exposed to the atmosphere, however, the water in time absorbs air, and dust carried in the air. Nevertheless, in distillation, water retains, besides air, a certain quantity of volatile impurities (especially organic) and the walls of the distillation apparatus are partly corroded by the water, and a portion, although small, of their substance renders the water not entirely pure, and a residue is left on evaporation.8
For certain physical and chemical researches, however, it is necessary to have perfectly pure water. To obtain it, a solution of potassium permanganate is added to distilled water until the whole is a light rose colour. By this means the organic matter in the water is destroyed (converted into gases or non-volatile substances). An excess of potassium permanganate does no harm, because in the next distillation it is left behind in the distillation apparatus. The second distillation should take place in a platinum retort with a platinum receiver. Platinum is a metal which is not acted on either by air or water, and therefore nothing passes from it into the water. The water obtained in the receiver still contains air. It must then be boiled for a long time, and afterwards cooled in a vacuum under the receiver of an air pump. Pure water does not leave any residue on evaporation; does not in the least change, however long it be kept; does not decompose like water only once distilled or impure; and it does not give bubbles of gas on heating, nor does it change the colour of a solution of potassium permanganate.
Water, purified as above described, has constant physical and chemical properties. For instance, it is of such water only that one cubic centimetre weighs one gram at 4° C.—i.e. it is only such pure water whose specific gravity equals 1 at 4° C.9 Water in a solid state forms crystals of the hexagonal system10 which are seen in snow, which generally consists of star-like clusters of several crystals, and also in the half-melted scattered ice floating on rivers in spring time. At this time of the year the ice splits up into spars or prisms, bounded by angles proper to substances crystallising in the hexagonal system.
The temperatures at which water passes from one state to another are taken as fixed points on the thermometer scale; namely, the zero corresponds with the temperature of melting ice, and the temperature of the steam disengaged from water boiling at the normal barometric pressure (that is 760 millimetres measured at 0°, at the latitude of 45°, at the sea level) is taken as 100° of the Celsius scale. Thus, the fact that water liquefies at 0° and boils at 100° is taken as one of its properties as a definite chemical compound. The weight of a litre of water at 4° is 1,000 grams, at 0° it is 999·8 grams. The weight of a litre of ice at 0° is less—namely, 917 grams; the weight of the same cubic measure of water vapour at 760 mm. pressure and 100° is only 0·60 gram; the density of the vapour compared with air = 0·62, and compared with hydrogen = 9.
These data briefly characterise the physical properties of water as a separate substance. To this may be added that water is a mobile liquid, colourless, transparent, without taste or smell, &c. Its latent heat of vaporisation is 534 units, of liquefaction 79 units of heat.11 The large amount of heat stored up in water vapour and also in liquid water (for its specific heat is greater than that of other liquids) renders it available in both forms for heating purposes. The chemical reactions which water undergoes, and by means of which it is formed, are so numerous, and so closely allied to the reactions of many other substances, that it is impossible to describe the majority of them at this early stage of chemical exposition. We shall become acquainted with