§ 246. =Detection and Separation of Nitro-Benzene from the Animal Tissues.=--It is evident from the changes which nitro-benzene may undergo that the expert, in any case of suspected nitro-benzene poisoning, must specially look (1) for nitro-benzene, (2) for aniline, and (3) for picric acid. The best general method for the separation of nitro-benzene is to shake up the liquid (or finely-divided solid) with light benzoline (petroleum ether), which readily dissolves nitro-benzene. On evaporation of the petroleum ether, the nitro-benzene is left, perhaps mixed with fatty matters. On treating with cold water, the fats rise to the surface, and the nitro-benzene sinks to the bottom; so that, by means of a separating funnel, the nitro-benzene may be easily removed from animal fats. The oily drops, or fine precipitate believed to be nitro-benzene, may be dissolved in spirit and reduced to aniline by the use of nascent hydrogen, developed from iron filings by hydrochloric acid, and the fluid tested with bleaching powder, or, the aniline itself may be recovered by alkalising the fluid, and shaking up with ether in the separation-tube (p. 156), the ether dissolves the aniline, and leaves it, on spontaneous evaporation, as an oily yellowish mass, which, on the addition of a few drops of sodic hypochlorite, strikes a blue or violet-blue--with acids, a rose-red--and with bromine, a flesh-red. It gives alkaloidal reactions with such general reagents as platinum chloride, picric acid, &c. Aniline itself may be extracted from the tissues and fluids of the body by petroleum ether, but in any special search it will be better to treat the organs as in Stas’ process--that is, with strong alcohol, acidified with sulphuric acid. After a suitable digestion in this menstruum, filter, and then, after evaporating the alcohol, dissolve the alcoholic extract in water; alkalise the aqueous solution, and extract the aniline by shaking it up with light benzoline. On separating the benzoline, the aniline will be left, and may be dissolved in feebly-acid water, and the tests before enumerated tried. Malpurgo[226] recommends the following test for nitro-benzene:--2 drops of melted phenol, 3 drops of water, and a fragment of caustic potash are boiled in a small porcelain dish, and to the boiling liquid the aqueous solution to be tested is added. On prolonged boiling, if nitro-benzene is present, a crimson ring is produced at the edges of the liquid; this crimson colour, on the addition of a little bleaching powder, turns emerald-green. [226] _Zeit. anal. Chem._, xxxii. 235. Oil of bitter almonds may be distinguished from nitro-benzene by the action of manganese dioxide and sulphuric acid; bitter almond oil treated in this way loses its odour, nitro-benzene is unaltered. To apply the test, the liquid must be heated on the water-bath for a little time. XI.--Dinitro-benzol.