Another form of experiment occurred to me still more direct and simple, that of transmitting muriatic acid in its gaseous form over ignited metals. If water be obtained in this experiment, it is a result which would prove subversive of the new doctrine; for muriatic acid gas is held to be the real acid, free from water; and the only change which can happen is that of the metal decomposing the acid attracting its chlorine and liberating its hydrogen. And the experiment is further free from the only resource which remained to the advocates of that doctrine, in the case of water being obtained from muriate of ammonia, that it might be derived from the decomposition of the elements of ammonia, regarding it as an alkali containing oxygen. If water were really obtained from the combination of muriatic acid and ammoniacal gases, it would rather indicate, it was said, the decomposition of nitrogen than the existence of water as a constituent of muriatic acid. No weight, I believe, is due to such an assumption; but if any importance were attached to it, it is precluded if water is obtained from the action of metals on muriatic acid gas.

I have executed the experiment in several forms; and in all with a more or less satisfactory result.

One hundred grains of iron filings, clean and dry, were strewed for a length of five or six inches, in a glass tube, which was placed in an iron case across a small furnace, so as to admit of being raised to a red heat. This tube, of about two feet in length, was connected with a wide tube eight inches long, containing dry and warm muriate of lime; and this was further connected at its other extremity, with a retort affording muriatic acid gas, from a mixture of supersulphate of potash and muriate of soda. The open extremity of the long tube, dipped by a slight curvature in quicksilver. On the iron being raised to ignition, and the transmission of the acid gas being conducted slowly, elastic fluid escaped from the extremity of the tube, which was found to be hydrogen; and though no trace of moisture appeared in the anterior part of the tube, it immediately condensed in that part which was cold, beyond the iron filings.

'

lime, free from water, does not absorb dry carbonic acid gas, but absorbs it rapidly if aqueous vapour be admitted, though water is not retained in the composition of carbonate of lime. And I have found that dry magnesia does not absorb muriatic acid gas, though, with the aid of water, it forms a combination from which the water can be expelled by heat. That ammoniacal salts exist without water is evident from the combination of carbonic acid gas and ammoniacal gas being effected with the greatest facility; and the circumstance that this compound is not neutral is one not depending on the peculiarity of the ammonia, and its not containing water, like other bases, but on that of the carbonic acid, which, with all the alkalies, even where water is present, has a tendency to form compounds with excess of base. The reason why the ammoniacal salts do not yield the combined water of their acids so completely as that of other salts, is, that from their volatility, or their susceptibility of decomposition, they do not bear that degree of heat which is necessary to produce it. I cannot, therefore, but consider the obseryation alluded to as one altogether unfounded, and which ought not, on a mere speculation, to have been brought forward against a positive result.”

This accumulated in globules, and at length ran into a small portion in the bottom; the sides were bedewed for a length of six inches, and a thin film of moisture appeared beyond nearly its whole length.

By the muriatic acid gas being extricated in the preceding experiment from nearly dry materials, and by its previous transmission over an extensive surface of loose muriate of lime, it was inferred that it would be free from hygrometric vapour; and that it held no moisture, was apparent from no trace of it appearing in the anterior portion of the tube. To obviate, however, entirely, any supposed fallacy from this source, the experiment was performed in the following manner. One hundred grains of clean and perfectly dry iron filings were put into a long glass tube, which was placed, as before, across a small furnace. Muriatic acid gas had been kept in contact with dry muriate of lime for three days, in a jar with a stop-cock adapted to it. This was connected by a short tube with a caoutchouc collar, with the tube containing the iron filings; and a little of the muriatic acid gas being passed through the tube to expel the air, the temperature was raised to ignition. The slow transmission of the gas was continued by the pressure of the mercury in the quicksilver trough, and fresh quantities, which had been equally with the other exposed to muriate of lime, were added, as was necessary. Water almost immediately appeared in the tube beyond the iron filings; it collected in spherules, and continued to accumulate as the gas continued to be transmitted for a length of about seven inches. A portion of the gas, which escaped from the extremity, was clouded, and deposited a film of moisture on the sides of the jar in which it was received over. quicksilver. The quantity of gas transmitted amounted to about 35 cubic inches.

There are some difficulties in conducting the experiment in the manner now described, from the consolidation of the metallic matter, and the volatilization of the product. It was also of some importance to vary the experiment. I, therefore, performed it in another mode. Metals scarcely act on muriatic acid gas at natural temperatures, but from such a degree of heat as could be applied by a small lamp, both iron and zinc were acted on; the gas suffered diminution of volume, hydrogen was formed, and a sensible production of moisture took place. The simplest mode of exhibiting this is to introduce iron or zinc filings, previously dry and warm, into a retort fitted with a stop-cock, exhausting it; then admitting dry muriatic acid gas, and applying heat by a small lamp to the filings in the under part of the body of the retort. Moisture soon appears at its curvature in small globules, and increases on successive applications of the heat with the admission of the requisite quantities of gas.

To conduct the experiment, however, on a larger scale, I employed a different apparatus. A tubulated retort, of the

capacity of 25 cubic inches, was connected with a jar, containing muriatic acid gas in contact with muriate of lime on the shelf of the mercurial trough, by a tube bent twice at right angles, and fitted by its shorter leg with a collar of caoutchouc to a stop-cock at the top of the jar, its longer leg passing into the tubulature of the retort, so as to terminate within an inch of its bottom, and the joinings being rendered air-tight. The retort is so placed that heat can be applied by a lamp to the bottom, and its neck dips, by a short curved tube, under a jar filled with quicksilver, which, by the reverted position of the retort, may be placed beside the other on the shelf of the trough. At the commencement of the experiment, the metallic filings, previously dry and warm, having been put into the retort, the atmospheric air is expelled by a moderate heat, and small portions of the muriatic acid gas are admitted until the retort is filled with the pure gas. The stop-cock is then closed, and heat is applied by a lamp to the bottom of the retort under a considerable pressure of mercury; any small portion of gas expelled at the extremity being received in the small jar. The heat can thus be successively cautiously applied, and this, as the experiment proceeds, to a greater extent, in consequence of the diminution of volume that takes place. Fresh quantities of muriatic acid gas are admitted from time to time from the jar, and the stop-cock being closed when the heat is applied, the hydrogen gas produced is expelled with any muriatic acid gas not acted on.

In the principal experiment I employed, zinc filings were used in preference to iron, from the consideration that muriate of zinc is less volatile than muriate of iron, and, therefore, would admit of a higher heat being applied to expel any water. One hundred grains of clean and dry zinc filings were introduced while warm into the retort; the air was expelled, and muriatic acid gas was admitted from the jar. On applying heat to the zinc, the retort, which was before perfectly dry, was bedimmed with moisture at its curvature, and small spherules collected at the top of the neck. These increased in size, and extended further as the experiment advanced. After a certain time, part of this disappeared in the interval of cooling, being absorbed by the deliquescent product; but when the heat was again applied, it was renewed, and this in increased quantity, until at length, at the end of four days, during which heat had been frequently applied, the whole tube of the retort, seven inches in length, was studded with small globules of fluid. When the heat had been raised high, a beautiful arborescent crystallization appeared in a thin film on the body of the retort, but no part of this reached the neck. The retort was now detached; the gas it contained was withdrawn by a caoutchouc bottle; a small receiver was adapted; and a slight heat having been applied to expel a little of the air, the joining was made close by cement. The receiver was surrounded with a freezing mixture, and heat was applied by a choffer to the VOL. XIII. N° 1.

C

retort, as far as could be done, without raising dense vapours. Globules of liquid, perfectly limpid, collected pretty copiously towards the middle and lower part of the neck, and the receiver, on being removed from the freezing mixture, was covered internally with a film of moisture. The globules in the neck of the retort were absorbed by a slip of bibulous paper, and the quantity was found to amount to 1.2 gr. The receiver being dried carefully and weighed, lost by the dissipation of the moisture within 0.4 gr. Distilled water, in which the bibulous paper was immersed, was quite acid; it gave no sensible turbidness on the addition of ammonia, or of carbonate of soda, and held dissolved, therefore, merely pure muriatic acid. The mass in the retort was of a grey colour, with metallic lustre, in loosely aggregated laminæ somewhat flexible. It weighed 114.8 gr. Adding to this increase of weight which the zinc had gained the weight of the water and the hydrogen gas expelled, it gives a consumption of muriatic acid gas of about 16.8 gr. equivalent to about 43 cubic inches. Supposing the weight of water to be doubled, or nearly so, by saturation with muriatic acid, this gives the product of water in the experiment as equal to nearly one gr.; or about 4th of the whole quantity of combined water which muriatic acid gas is calculated to contain.*

In all the preceding experiments, water has been procured from muriatic acid gas. It is obvious that such a result cannot be accounted for on the hypothesis that it is the real acid free from water, a compound merely of chlorine and hydrogen. On the opposite doctrine, as muriatic acid in its gaseous form is held to contain water, it may be supposed to afford a portion of it.:

It may be maintained, however, in this, as it was in the experiment of obtaining water from the muriate of ammonia by heat, that the water produced is derived from hygrometric vapour in the gas. To obviate this, it is sufficient to recur to the fact established by the experiments of Henry and Gay-Lussac, that muriatic acid gas contains no hygrometric vapour; and to the

* The action of the metals on the muriatic acid gas taking place in the above experiments at a heat comparatively moderate, it occurred to me that they might exert a similar action with no higher heat on the acid in muriate of ammonia, and that this might afford an easy mode of exhibiting the results. I accordingly found, that on mixing different metals with sal ammoniac in powder, previously exposed to a subliming heat, and exposing the mixture to heat by a lamp, so regulated as to be short of volatilization, the salt was decomposed, ammoniacal gas was expelled, and moisture condensed in the neck of the retort, covering a space of several inches with small globules, and at length running down. The metals I employed were iron, zinc, tin, and lead; 100, 150, or 200 gr. of each metal, dry and warm, being mixed with 100 gr. of the salt likewise newly heated. To obviate any fal lacy from common sal ammoniac being employed, I repeated the experiment with the salt formed from the combination of its two constituent gases, and obtained the same result. But although this affords an easy mode of exhibiting the production of water, it is not favourable to obtaining a perfect result, the heated ammoniacat gas carrying off a considerable portion of the water deposited; and accordingly the quantity, instead of increasing as the experiment proceeds, at length diminishes, and the ammoniacal gas deposits a portion of water in passing through mercury, as in being conveyed through a cold tube.

obvious result in the experiment that no quantity that can be assumed would be adequate to account for the quantity actually obtained. The circumstances of the experiment too are such as to preclude any such supposition; and this more peculiarly soo than in the experiment of obtaining water from the muriate of ammonia by heat; for in the present case the acid gas is alone employed, while in the other there is an additional equal volume of ammoniacal gas, which may be supposed to afford a double quantity of hygrometric vapour. In the latter, both the gases are condensed into a solid product, and any hygrometric vapour!! may be supposed to be liberated; but in the present experiment, there remains the hydrogen gas capable of containing hygrome tric vapour, while the muriatic acid gas contains none; and the quantity of it thus transmitted over the humid surface, and expelled from the apparatus, must have carried off more vapour than the other, introduced at a lower temperature, could have conveyed. These circumstances, independent of the quantity of water deposited, precluded the supposition of any deposition from the condensation of hygrometric vapour; and there is no other external source whence it can be derived. In this respect," nothing can be more satisfactory than the experiment with the zinc in the apparatus described. The muriatic acid gas rises: from dry mercury in contact with muriate of lime, passes through a narrow bent tube, 30 inches in length, without exhibiting the slightest film of moisture, is received into the retort perfectly dry; and when the action of the metal on it is excited by heat, humidity immediately becomes apparent in the curvature of the retort, and this even while the gas is warm, and of course capable of containing more water dissolved than it could do in its former state; and the quantity increases as the experiment proceeds. No arrangement can be supposed better adapted to prove that any deposition of water must be by separation from its existence in the gas in a combined state.

But though I consider this conclusion as established, there is a considerable difficulty attending the theory of the experiment. The result of water being obtained is actually different from what is to be looked for on the doctrine of muriatic acid gas containing combined water; and even when the fact is established, the theory of it is not easily assigned. On that doctrine, it must be held that in the action of metals on muriatic acid gas, the metal attracts oxygen from the water, the corresponding hydrogen is evolved, and the oxide formed combines with the real acid. No water, therefore, ought to be deposited; for none is abstracted from the acid but what is spent in the oxidation of the metal. This will be apparent by attending to the proportions in a single example from the scale of chemical equivalents: 100 gr. of iron combine with 29 of oxygen, and in this state of oxidation unite with 99 of real muriatic acid. This quantity of acid exists in 131-8 of muriatic acid gas combined with 32.8 of water; and this

1