Marquardt,[8] like Betelli, extracts the fusel oil from alcohol by means of chloroform, and by oxidation converts it into valeric acid. From the quantity of barium valerate found he calculates the amount of amyl alcohol present in the original solution; 150 c.c. of the spirit, which has been diluted so as to contain 12 to 15 per cent. of alcohol, are shaken up thoroughly with 50 c.c. of chloroform, the aqueous layer drawn off, and shaken with a fresh portion of chloroform. This treatment is repeated several times. The extracts are then united, and washed repeatedly with water. The chloroform, which is now free from alcohol and contains all the fusel oil, is treated with a solution of 5 grammes of potassium bichromate in 30 grammes of water and 2 grammes of sulphuric acid, and then heated in a closed flask for six hours on a water bath at 85 deg.. The contents of the flask are then distilled, the distillate saturated with barium carbonate, and the chloroform distilled; the residue is evaporated to a small volume, the excess of barium carbonate filtered off, and the filtrate evaporated to dryness and weighed. The residue is dissolved in water, a few drops of nitric acid added, and the solution divided into two portions. In the first portion the barium is determined; in the second the barium chloride. The total per cent. of barium minus that of barium chloride gives the amount present as barium valerate, from which is calculated the per cent. of amyl alcohol. By this process the author has determined one part of fusel oil in ten thousand of alcohol. To detect very minute quantities of fusel oil, the chloroform extracts are treated with several drops of sulphuric acid and enough potassium permanganate to keep the solution red for twenty-four hours. If allowed to stand in a test tube, the odor of valeric aldehyde will first be noticed, then that of amyl valerate, and lastly that of valeric acid.--_Amer. Chem. Journal._ [Footnote 8: Berichte d. Deutsch. Chem. Gesellsch., xv., 1,370 and 1,663.] * * * * * ON SILICON. It is known that platinum heated in a forge fire, in contact with carbon, becomes fusible. Boussingault has shown that this is due to the formation of a silicide of platinum by means of the reduction of the silica of the carbon by the metal. MM. P. Schuetzenberger and A. Colson have produced the same phenomenon by heating to white heat a slip of platinum in the center of a thick layer of lampblack free from silica.