Gay-Lussac 'a volumetric method of estimating silver by a standard solution of chloride of sodium, has already been described under ASSAYING. The converse is of course equally easy, namely, the deter mination of the amount of chlorine in any chloride by a standard solution of nitrate of silver. Mohr has shown that this method admits of very general application : thus ammonia may be estimated by adding excess of hydrochloric acid, evaporating to dryness. and determining the amount of chlorine in the residue; the resulting chloride of ammonium being a definite salt, it is only necessary to know the amount of chlorine in the residue and then a simple calculation will give the quantity of ammonia experimented on. Nitrogen may be determined in a similar manner after conversion into ammonia by soda lime. Carbonic acid may be absorbed in a mixture of chloride of barium and ammonia, the carbonate of barytes collected on a filter, washed, dissolved in hydrochloric acid, the solution evaporated to dryness, and the chlorine estimated as before ; the weight of the latter being, of course, transformed by calculation into an equivalent quantity of carbonic acid. Carbonates or nitrates are converted into chlorides, the chlorine estimated, and the equivalent of nitrate or carbonate calculated. Salts of organics acids may be altered to carbonates by ignition, and the carbonates treated as already described. Chlorates, iodates, and bromates are reduced to chlorides, iodides, and bromides, and also acted upon in the way indicated.
Permanganate of potash is a re-agent of considerable value in the volumetric analysis of substances which readily absorb definite quau tities of oxygen. Possessed of an intense purple colour, its solution freely gives up its oxygen and becomes almost colourless. It is best standarised by noting the quantity necessary to be added to a known amount of a definite protosalt of iron, such as the ammonio-protosul phate, before a permanent purple tint is produced : this of course only occurs when the protoxide of iron is wholly converted into peroxide.
After the strength of the permanganate of potash solution has been thus fixed, it can be used for estimating an unknown amount of pro toxide of iron in a weighed quantity of a mixture of the two salts and for other similar operations.
In all the above described processes of volumetric analysis, as well as in those only referred to, a balance and weights or saturated solu tions of definite strength are necessary at some stage of the operations. Quantitative determinations can, however, be made without the use of weights, with volumetric solutions of unknown strength, and tho strength of which need nut be ascertained or regulated. For example, let it be required to know the amount of carbonate of soda in a specimen of crude soda-ash. Place on one pan of a balance a piece of chemically pure carbonate of soda and counterpoise it by the crude ash; dissolve in separate quantities of water, tinge the solutions by litmus, then add dilute acid from a graduated measure until both aro saturated ; suppose, now, that twenty-two volumes of acid were required for the pure soda and sixteen for the impure, then as 22 : 16 : : 100 : x, a calculation that shows the crude ash to contain per cent. of pure carbonate of soda. Many other determinations may obviously be made on the same principle. The substance to be determined need not even be available iu the pure state; one of its compounds must, however, be attainable in that condition, and both that compound and the impure body must be capable of being similarly acted upon by the test-liquid employed. For instance, in the example above worked out, pure carbonate of lime may be substituted for carbonate of soda, but the number of volumes of test-liquid used must be added to or subtracted from according as the equivalent of the substituting body is higher or lower than that of the body substituted.