The presence of alcohol or ether, or both, may also be discovered by the bichromate test, which is best applied as follows:--A few milligrammes of potassic bichromate are placed at the bottom of a test-tube with four or five drops of sulphuric acid, which liberates the chromic acid; next, a very little water is added to dissolve the chromic acid; and lastly, the chloroform. The whole is now shaken, and allowed to separate. If the chloroform is pure, the mass is hardly tinged a greenish-yellow, and no layer separates. If, however, there is anything like 5 per cent. of alcohol or ether present, the deep green of chromium chloride appears, and there is a distinct layer at the bottom of the tube. Another way to detect alcohol in chloroform, and also to make an approximate estimation of its quantity, is to place 20 c.c. of chloroform in a burette, and then add 80 c.c. of water. On shaking violently, pure chloroform will sink to the bottom in clear globules, and the measurement will be as nearly as possible the original quantity; but if anything like a percentage of alcohol be present, the chloroform is seen to be diminished in quantity, and its surface is opalescent, the diminution being caused by the water dissolving out the alcohol. The addition of a few drops of potash solution destroys the meniscus, and allows of a close reading of the volume. The supernatant water may be utilised for the detection of other impurities, and tested for sulphuric acid by baric chloride, for free chlorine and hypochlorous acid by starch and potassic iodide, and for hydrochloric acid by silver nitrate.[160] Fuchsine, proposed by Stœdeler, is also a delicate reagent for the presence of alcohol in chloroform, the sample becoming red in the presence of alcohol, and the tint being proportionate to the quantity present. The most delicate test for alcohol is, however, the iodoform test fully described in “Foods,” p. 375.[161] Dichloride of ethylene is detected by shaking up the chloroform with dry potassic carbonate, and then adding metallic potassium. This does not act on pure chloroform, but only in presence of ethylene dichloride, when the gaseous chlor-ethylene (C₂H₃Cl) is evolved. Ethyl-chloride is detected by distilling the chloroform and collecting the first portions of the distillate; it will have a distinct odour of ethyl-chloride should it be present. Methyl compounds and empyreumatic oils are roughly detected by allowing the chloroform to evaporate on a cloth. If present, the cloth, when the chloroform has evaporated, will have a peculiar disagreeable odour. Aldehyde is recognised by its reducing action on argentic nitrate; the mineral acids by the reddening of litmus paper, and the appropriate tests. Hypochlorous acid first reddens, and then bleaches, litmus-paper. [160] Neither an alcoholic nor an aqueous solution of silver nitrate causes the slightest change in pure chloroform.