The work of which an account is given in the following pages was carried out in the winter of 1887-88, and forms a sequel to much of a similar character already published by the authors, on methyl, ethyl, and isopropyl alcohols, on ethyl oxide, on acetic acid, and on a mixture of alcohol and ether (‘Phil. Trans.' 1886, Part I., p. 123; 1887, A., p. 57; 1887, A., p. 313; 1889, A., p. 137; ‘ Chem. Soc. Trans.' 1886, p. 790 ; 1887, p. 755). The apparatus with which it was carried out has been fully described in the memoir on ether ( loc. cit .). The only important change was in the nature of the glass tubes used to contain the liquid under experiment. While, in the case of the alcohols, ether, &c., lead glass proved the best material for tubes, it is too easily attacked by water. At comparatively low temperatures it becomes etched, and it is impossible to take readings. Moreover, the water dissolves a not inconsiderable quantity of potassium silicate; tubes of green boiler-gauge glass were, therefore, substituted for tubes of lead glass. This glass consists of a silicate of calcium and potassium, containing a trace of ferrous iron (SiO 2 = 71.20; CaO=14.99; K 2 O = 13.19; Total, 99.38.) They withstand a higher pressure than lead-glass tubes, and they are attacked with difficulty. Even after an exposure of several days to liquid water at 280°C., only 0.7 per cent, of residue remained on evaporating the water. Inasmuch as some of the material of the glass is dissolved, however, the water cannot be considered to be absolutely pure, but the results may be given as the best attainable with water in contact with glass.