§ 89. =Detection and Estimation of Nitric Acid.=--The detection either of free nitric acid or of its salts is not difficult. Free nitric acid, after preliminary estimation of the total acidity by decinormal soda, may be separated by the cinchonine process given at p. 100. On precipitation by ammonia or soda solution, the nitrate of ammonia or soda (and, it may be, other similarly combined acids) remain in solution. If free nitric acid is present in small quantity only, it may be necessary to evaporate the filtrate from the quinine nearly to dryness, and to test the concentrated liquid for nitric acid. The ordinary tests are as follows:-- (1.) Nitrates, treated with mercury or copper and strong sulphuric acid, develop nitric oxide, recognised by red fumes, if mixed with air or oxygen. (2.) A nitrate dissolved in a small quantity of water, with the addition of a crystal of ferrous sulphate (allowed to partially dissolve), and then of strong sulphuric acid--poured through a funnel with a long tube dipping to the bottom of the test-tube, so as to form a layer at the bottom--strikes a brown colour at the junction of the liquid. When the test is properly performed, there will be three layers--the uppermost being the nitrate solution, the middle ferrous sulphate, and the lowest sulphuric acid; the middle layer becomes of a smoky or black hue if a nitrate is present. Organic matter interferes much with the reaction. (3.) Nitrates in solution, treated in the cold with a zinc copper couple, are decomposed first into nitrites, and then into ammonia. The nitrites may be detected by a solution of metaphenyldiamine, which strikes a red colour with an infinitesimal quantity. Hence, a solution which gives no red colour with metaphenyldiamine, when submitted to the action of a zinc copper couple, and tested from time to time, cannot contain nitrites; therefore, no nitrates were originally present. (4.) Nitrates, on being treated with strong sulphuric acid, and then a solution of indigo carmine dropped in, decolorise the indigo; this is a useful test--not conclusive in itself, but readily applied, and if the cinchonine method of separation has been resorted to, with few sources of error.