The formation of the azo compound takes place slowly on the addition of the dimethylaniline, but the speed of the reaction is greatly increased when the hydrogen ion concentration is lowered by the addition of the sodium acetate. It is nevertheless necessary to allow the reaction mixture to stand a long time; if the product be filtered off after only twenty-four hours, a further quantity of dye will separate from the filtrate on standing. The hydrochloride of methyl red is only sparingly soluble in cold water, and is apt to separate in blue needles if the acidity is not sufficiently reduced. The alcoholic filtrate, obtained on digesting and washing the crude methyl red, contains a more soluble red by-product which gives a brownish-yellow solution in alkali. The methyl alcohol may be recovered with very little loss by distillation; it is, however, impracticable to attempt to recover any methyl red from the residue, owing to the tarry nature of the by-product. The proportion of this by-product is greatly increased if the temperature of the mixture is allowed to rise too soon after the addition of the sodium acetate. Methyl red is described as crystallizing in needles from glacial acetic acid; on recrystallization from toluene it separates in plates. When the methyl red is crystallized from toluene, it sometimes separates in the form of bright-red lumps, probably on account of too rapid crystallization. Under these conditions it is advisable to crystallize again, using a somewhat larger amount of toluene. It is advisable to titrate the crude anthranilic acid with standard alkali and phenolphthalein before starting the experiment. In checking these directions, an 80 per cent anthranilic acid was used; it gave a correspondingly lower yield of methyl red (650-700 g.). The yield of methyl red is about 65 to 70 per cent based on the dimethylaniline actually used up, but only 58-63 per cent based on the anthranilic acid actually present in the technical anthranilic acid employed. 3. Other Methods of Preparation Methyl red was first prepared[1] by diazotization of anthranilic acid in alcoholic solution, the product being allowed to react with dimethylaniline in the same solvent. It has been stated[2] that this process does not work satisfactorily and yields a different product, of brownish-red color. The preparation of methyl red in aqueous solution has been described by two workers, one of whom[3] gives but few details and claims a nearly quantitative yield; the other[4] gives fuller details and states the yield to be 43.1 per cent of the theory. The recrystallization of methyl red from toluene is stated[5] to yield a product melting at 183'0. [1] Ber. 41, 3905 (1908). [2] Chem. Zentr. 1910, (1), 960; 1910, (11), 1561. [3] J. Chem. Soc. 97, 2485 (1910). [4] C. A. 14, 3406 (1920) [5] J. Chem. Soc. 99, 1334 (1911). XIII _p_-NITROBENZOIC ACID (_p_)NO2C6H4CH>3s> + 3O(Na2Cr2O7 + H2SO4)--> (_p_)NO2C6H4CO2H + 3H2O