remaining some time it disappeared, having gradually mixed with the atmosphere above it; but every repetition of the experiment produced the same results.
Presuming that I had now a right to consider the yellow fluid as pure chlorine in the liquid state, I proceeded to examine its properties, as well as I could when obtained by heat from the hydrate. However obtained, it always appears very limpid and fluid, and excessively volatile at common pressure. A portion was cooled in its tube to 0°; it remained fluid. The tube was then opened, when a part immediately flew off, leaving the rest so cooled, by the evaporation, as to remain a fluid under the atmospheric pressure. The temperature could not have been higher than -40° in this case; as Sir Humphry Davy has shown that dry chlorine does not condense at that temperature under common pressure. Another tube was opened at a temperature of 50°; a part of the chlorine volatilized, and cooled the tube so much as to condense the atmospheric vapour on it into ice.
A tube having the water at one end and the chlorine at the other was weighed, and then cut in two; the chlorine immediately Hew off, and the loss being ascertained was found to be 1.6 grain: the water left was examined and found to contain some chlorine: its weight was ascertained to be 5.4 grains. These proportions, however, must not be considered as indicative of the true composition of hydrate of chlorine; for, from the mildness of the weather during the time when these experiments were made, it was impossible to collect the crystals of hydrate, press and transfer them, without losing much chlorine; and it is also impossible to separate the chlorine and water in the tube perfectly, or keep them separate, as the atmosphere within will combine with the water, and gradually re-form the hydrate.
Before cutting the tube, another tube had been prepared exactly like it in form and size, and a portion of water introduced into it, as near as the eye could judge, of the same bulk as the fluid chlorine; this water was found to weigh 1.2 grain; a result, which, if it may be trusted, would give the specific gravity of fluid chlorine as 1.33; and from its appearance in and on water, this cannot be far wrong.
