An accurate analysis of a water is much too complicated for the manufacturer; it is sufficient for him to convince himself of the absence of certain substances. Water which, some time after the addition of a little tannic acid solution, acquires a clear green or a bluish to black shade contains much iron, and is useless. Water which coagulates a large quantity of a solution of soap in alcohol is very rich in carbonate or sulphate of lime. In order to decide approximately in what relative proportion these salts are present, a solution of barium chloride is added to the water so long as a precipitate forms. If this disappears completely on the addition of nitric acid, the water contains only carbonate of lime; if it only partly dissolves, sulphate of lime is also present. The presence of chlorine is shown by a considerable turbidity on acidifying the water with nitric acid, boiling and adding silver nitrate. If the precipitate obtained on the addition of a lead salt is not pure white, but discoloured, the water contains sulphuretted hydrogen, which has formed black lead sulphide. In order to test the water for organic substances, about a litre is evaporated to dryness in a porcelain dish, and the residue heated to redness; if it turns brown and black, and possibly gives off a smell of burnt feathers, the water contains much organic matter. Pure water is coloured permanently red by a solution of potassium permanganate; but if it contains organic matter, the solution is decolourised after some time and a brown precipitate is deposited at the bottom. From the amount of this precipitate an idea of the quantity of organic matter present may be obtained. It is only necessary to be very scrupulous concerning the quality of the water when it is to be used for the solution of salts or the extraction of dye-woods. For washing precipitates, which requires a large volume of water, there can generally be used, without detriment, water containing much lime, but it must be free from iron and sulphuretted hydrogen. The latter is particularly harmful to most of the lead colours, which would lose in beauty by washing with water containing this substance. It is hardly necessary to say that the water used in colour making must be quite clear. Muddy river water must in every case be completely freed from the solid particles contained in it, either by settling or by filtering. Filters filled with well washed sand give good results for this purpose. =Chlorine=, Cl = 35·5.—For some operations in colour making it is necessary to employ chlorine. This is a greenish yellow gas at ordinary temperatures, which is characterised by a suffocating smell and the energy with which it unites with most elements. On account of its injurious effects on man certain precautions have to be observed in preparing chlorine, and it is advisable to erect the apparatus necessary for its production in a separate room, so that the workmen are not injured by the gas.