ON ERRORS IN THE DETERMINATION OF THE DENSITIES OF MIXED VAPOURS.—In the Compt. Rend., Ixxxiii., Messrs. Troost and Hautefeuille record some experiments made by them to discover the error which occurs in determining the vapour-density of substances by the application of the law of Dalton on the tension of mixed gases, and Boyle and Gay-Lussac's laws, as applied to a mixture of the vapour examined, either with air or with some other vapour. They examined a mixture of carbon and silicium chlorides, using a modified form of Gay-Lussac's vapour-density apparatus. On increasing the amount of carbon chloride, the tension of the silicium chloride diminished. The vapour-density of silicium chloride alone varied only from 5.94 to 6.0, but in presence of carbon chloride was found to increase 6.27 to 8.2.