The normal solution or test-liquor is prepared by adding to an aqueous solution of sulphate of copper, pure ammonia water, until the precipitate, which at first forms, is just redissolved, carefully avoiding excess. Or better, by adding a rather strong solution of sulphate of copper, to a quantity of a rather strong solution of ammonia containing exactly 17 gr., or one equiv. of pure ammonia, as long as the precipitate which forms is redissolved on agitation; the resulting liquid being afterwards diluted with pure distilled water, until it accurately measures 1000 water-grains, or fills 100 divisions of an acidimeter, at 60° Fahr. In either case, the strength of the resulting ‘test-solution’ must be carefully determined by means of standard sulphuric acid, and adjusted, if necessary. This method answers well with all the stronger acids (excepting oxalic acid), even when dilute; and it has the advantage of not being affected by the presence of a neutral metallic salt with an acid reaction, as sulphate of copper, or of zinc. Besides this process a solution of lime in sugar may be used, as proposed by M. Peligot, and made as follows:—— Pure caustic lime is carefully slaked by sprinkling with water, and 50 grains (or grammes), made up by water to a milky solution, and 100 grains of pure sugar candy dissolved in 1000 grains of water, are added, and the whole well shaken. It is allowed to settle in a closed bottle, and the clear solution poured off and diluted, until 1000 grains neutralise exactly 100 grains of pure hydrochloric acid of sp. gr. 1·1812. Of course it only answers with acids whose calcium salts are readily soluble in water. _b._ GRAVERMETRICALLY:—— The test-liquors or standard solutions of the above methods are made up so as to _weigh_ exactly 1000 grains, instead of to ‘measure’ 100 acidimeter divisions. Every grain of the test-liquor thus represents 1/10th gr. of alkali; and every 10 gr., 1 gr. of alkali; or respectively, 1/10th per cent. and 1 per cent. The vessel used for containing the solutions is carefully weighed whilst empty, and 1000 gr. being placed in the opposite scale, the test-solution, containing exactly one equivalent of base, is poured in, and the whole made up with distilled water (if necessary) so as to restore the balance to an equilibrium. After the process of neutralisation, the acidimeter, with its contents, is again placed in the scales; its previous weight still remaining there. The number of grains required to restore the equilibrium of the balance (_i.e._, the loss of weight), gives the exact weight of the test-liquor consumed. In all other respects the process is the same as in the ‘volumetrical method’ already described.