in volume will be nearly 119 to 30; and this estimation differs as little as might be expected from that gained by the action of mercury upon the compound.

It may fairly be concluded, that M. GAY LUSSAC's principle of the combination of gaseous bodies, in definite volumes, strictly applies to this compound, and that it really consists of four volumes of chlorine to one of azote; and the volumes likewise exactly coincide with the laws of definite proportions; and the detonating compound may be regarded as composed of one proportion of azote 26, and four proportions of chlorine 261.

I attempted a comparative experiment on the proportions in the compound, by estimating the quantity of azote produced in the decomposition of ammonia by it; but I found that this process was of no value for the purpose of analysis, for water appeared to be decomposed at the same time with the ammonia, and nitric acid formed; and, in consequence, the quantity of azote evolved was much less than it would have been, supposing the ammonia decomposed by the mere attraction of chlorine for hydrogen.

The results of the analysis of the new compound are interesting for several reasons.

They shew, what seemed probable from other facts, that there is no strict law of analogy, which regulates the combinations of the same substance with different substances. As three of hydrogen combine with one of azote, and one of hydrogen with one of chlorine, I thought it probable that the new compound would contain three of chlorine to one of azote, which is not the case.

This compound is the first instance known of one proportion

of a substance uniting to four proportions of another substance, without some intermediate compound of 1 and 1, 1 and 2, and 1 and 3; and the fact should render us cautious in adopting hypothetical views of the composition of bodies from the relations of the quantities in which they combine. Those who argue that there must be one proportion of oxygen in azote, because there ought to be six proportions in nitric acid, instead of five, which are produced from it by analysis, might with full as much propriety contend, that there must be azote in chlorine in some simple multiple of that existing in the compound.

It may be useful to shew, that many hypotheses may be framed upon the same principles; and which, consequently, must be equally uncertain. Views of this nature may be important in directing the practical chemist in his researches ; but the philosopher should carefully avoid the developement of them with confidence, and the confounding them with prac tical results.

The compound of chlorine and azote agrees with the compounds of the same substance with sulphur, phosphorus, and the metals, in being a non-conductor of electricity; and these compounds are likewise decomposable by heat, though they require that of Voltaic electricity.

Sulphur combines only in one proportion with chlorine; and hence the action of Sulphurane, or Dr. THOMSON's muriatic liquor upon water, like that of the new compound, is not a simple phenomenon of double decomposition.

It seems proper to designate this new body by some name: Azotane is the term that would be applied to it, according to my ideas of its analogy to the other bodies which contain chlorine; but I am not desirous, in the present imperfect and

fluctuating state of chemical nomenclature, to press the adoption of any new word, particularly as applied to a substance not discovered by myself.

I am, my dear Sir,

very sincerely yours,

MDCCCXIII.

LI

HUMPHRY DAVY.

XXIX. Experiments on the Production of Cold by the Evaporation of the Sulphuret of Carbon. By Alexander Marcet, M. D. F. R. S. one of the Physicians to Guy's Hospital.

I

Read July 8, 1813.

HAD the honour, at an earlier period of this session, of giving to the Society, conjointly with Professor BERZELIUS, an account of the alcohol of sulphur, or sulphuret of carbon, and of noticing the remarkable volatility of that fluid. I have, since that period, tried a variety of experiments on the subject, and having found this compound more volatile than any other known body, and capable of producing, by its evaporation, a degree of cold of proportional intensity, I have been induced, in order to render the history of the sulphuret of carbon more complete, to present to the Society a brief account of my experiments.*

If the bulb of a small spirit thermometer be closely covered with a bag of fine flannel, or still better, with a piece of fine

* I would recommend to those, who may wish to repeat these experiments, to prepare this substance by means of a large earthen tube of about one inch and a half in diameter, instead of the small porcelain tubes, which are commonly used for this purpose. The process is always a tedious one; but a much more considerable quantity of the sulphureous liquor is procured by the larger tube in an equal space of time. The same tube can scarcely ever be used twice. About half a pint of the liquor may be obtained in one process; but the operation requires almost a whole day. Fresh pieces of sulphur may be successively introduced whilst the distillation is going on, and without renovating the charcoal, as the occasional admission of air does not materially affect the operation. The process would, no doubt, admit of further improvement.

lint, and a few drops of the sulphureous liquor poured upon it, the thermometer rapidly sinks from 60° to about o; whilst if a similar experiment be made with ether, or alcohol, under the same circumstances, the temperature is reduced by the ether to only about + 20°, and by the alcohol to about 50°.* The hygrometrical state of the atmosphere appears to have some influence on these results.

If the bulb of the thermometer, thus wetted with the sulphureous liquor, be introduced into the receiver of an airpump, (by means of a brass plate, as expressed in the annexed

sketch, the plate fitting the receiver air tight when laid upon its open neck), the thermometer sinks rapidly, as the ex

⚫ I have also tried some experiments on the comparative elastic force of the vapour of these fluids, by introducing in succession a few drops of them, through mercury, into a torricellian tube; and I obtained the following results:

It would appear, therefore, that the degree of volatility of a fluid, or of cold produced

by its evaporation, are far from bearing an exact proportion to the elastic force of its