per with solution of sea salt; the copper becomes negative, and the solution positive. This result explains why a plate of copper in contact with zinc or tin, as lately ascertained by Sir H. Davy, is less acted on by the sea-water than when not in contact with an electro-positive metal. It cannot be denied, that two substances at the moment they combine are in different electrical states, and that there is a certain relation between those states and the chemical affinities. Now if we can modify those electrical states, it is almost certain that we shall also modify the play of affinities; but we have seen that a plate of copper, by contact with a solution of sea-salt, becomes negative; it follows that if we touch the same metal with an electro-positive metal, the copper will be placed between two bodies, each tending to impart the same kind of electricity, a condition which we know will tend to annul the electro-motive action of the copper on the solution of sea-salt. Thus, according to the electro-chemical theory, circumstances are so arranged as to weaken the chemical action of the solution of sea-salt on the copper."

The memoir concludes with pointing out a method by means of electricity, of ascertaining the changes which some solutions experience by contact with the air.

Dissolve iron in nitric acid; filter the solution, and immerse into it two lamina of platina, each communicating with one of the extremities of the wire of the galvanometer; leave one of the wires in the solution, withdraw the other, and again immerse it; it will be positively electrified.

The nitrates of copper and lead give similar results for a short time; nitrate of zinc produces no such effect.

When the experiment is made in an atmosphere of hydrogen, no electrical current is established, though all circumstances, except the want of contact with the atmosphere, are precisely similar in both experiments.

"Hence the contact of atmospheric air is indispensable to the production of the electrical current by the immersion of platina laminæ in several fresh-prepared nitrates; but what is the modification that is instantaneously effected in the liquid adhering to the surface of the lamina withdrawn from the solution? We can, to a certain extent, explain this: The solution of a metal in nitric acid gives rise to several compounds: take iron, for instance; first deutoxide of azote is formed, and soon after nitrous acid, a protonitrate and a deutonitrate; by degrees the deutoxide of azote passes to the state of nitrous acid, the protonitrate to that of deutonitrate, and, after a certain time, only deutonitrate remains in the liquid. According to this statement, when we withdraw one of the platina laminæ from the solution, the liquid which adheres to it immediately, in consequence of the thinness of the stratum, experiences changes from

the action of the air, which do not take place till after several hours in the bulk of the solution. It follows, therefore, that when we re-immerse the lamina, we bring in contact two dissimilar liquids, and nothing in that case opposes the production of an electrical current.

"On the other hand, since the immersion of platina lamina in a solution of nitrate of zinc does not produce any current, although it contains deutoxide of azote and nitrous gas, it is probable that this may be owing to the nitrate not suffering any change by contact with the air, in consequence of the metal being capable of forming only one oxide."

ARTICLE IX.

On a deoxidating Property of the Vapour of Water.
By C. H. Pfaff*

It was remarked by Hermbstadt, while making the experiments from which he deduced the existence of a peculiar colouring principle in sea-water, and its superincumbent atmosphere,† that if that liquid be boiled in a retort, and if, by means of a glass tube attached to the beak of the retort, the gases and aqueous vapour evolved be made to pass through a solution of nitrate of silver, the latter by degrees assumes the colour of red wine, and at the end of 24 hours, a brownish-yellow coloured sediment is deposited. I observed the same appearance, on repeating this experiment with sea-water from the bay at Kiel. As, however, I had reasons for suspecting that this change is not occasioned by any peculiar gaseous constituent, I prepared an artificial mixture of solutions of the muriates of magnesia and soda, in the proportions which constitute sea-water,' and on making the experiment with this, I still obtained the same result. I observed also that the colour imparted to the solution of nitrate of silver at the beginning, and towards the conclusion of the experiment, is different: at first it is a weak violet, but after the experiment has gone on for some time (provided there be a sufficient quantity of the metallic salt in the vessel through which the vapours pass), it has a considerable infusion of reddish-brown. Hence I considered it not unlikely, that in these experiments there are two distinct causes which produce discoloration. This induced me to repeat the experiment in a variety of ways; which at last conducted me to the following very interesting results.

The experiments themselves are extremely simple. The

⚫ Schweigger's Journal für Chemie und Physik, xxxvi. 68.

+ This opinion was refuted by Pfaff in a small pamphlet, entitled Das Kieler Seebad verglichen mit andern Seebädern an der Ostsee und Nordsee, Kiel, 1822.

liquids were boiled in clean glass retorts, and great care was taken to prevent any portion of them from being carried over mechanically in the state of drops. The solutions on which the action of the gases and vapours expelled by boiling was to be determined, were contained in Woulfe's bottles, and, in general, the vapours passed through two or three of these in succession. The tubes of communication between the retort and the first bottle, and between the bottles themselves, were plunged to a sufficient depth in the solutions.

1. The mere vapour of pure distilled water when passed through a transparent solution of nitrate of silver, has the property of communicating to it a discoloration, in proportion as it heats the solution to the temperature of ebullition; and the intensity of this discoloration varies from yellow to dark-brown, according to the concentration of the solution, and to the length of time during which it has been exposed to the action of the vapour.

2. This discoloration is inconsiderable, so long as the solution of nitrate of silver remains under the boiling point, but it becomes exceedingly striking, the instant ebullition commences. The colour which first appears is yellow, but it rapidly becomes darker. The colour of the solution of nitrate of silver, when sufficiently diluted, has a close resemblance to red wine.

3. In the same manner, the solutions of nitrate of silver in the remaining bottles may be discoloured; because the vapour, after heating the liquid contained in one bottle, passes over into the next, and raises its temperature also to the point of ebullition.

4. This discoloration is caused by a deoxidation of the nitrate of silver, and except that it takes place with much greater rapidity, it is similar in all respects to the effect produced by light. În proof of this we may mention, a. The similarity of the colours with those produced by the action of mere light. b. The complete removal of the colour, and restoration of transparency, by the addition of nitric acid. c. The similar deoxidizing effect of the vapour of water upon other metallic solutions, which are easily deoxidized, either by light, or by some chemical action. d. The disengagement of oxygen gas during the process.

5. The most convincing of all proofs is furnished by a solution of muriate of gold. A solution of this salt so much diluted as scarcely to retain a shade of yellow, when heated to the boiling point by a stream of the vapour of water, acquires a beautiful blue colour, exactly similar to the colour produced in it by tincture of nutgalls, oxalic acid, &c.

than

6. After the expulsion of the atmospheric air, I collected the gas which was disengaged in a constant stream of minute bubbles, from the boiling-hot solution of nitrate of silver. It proved to contain a considerably greater quantity of oxygen common air: 100 volumes of it mixed with 100 volumes of nitrous gas, sustained a diminution amounting to 91, whereas common air sustained a diminution of only 80.

7. Of the other solutions whose colour is changed by deoxidation, I examined acetate of silver, which became discoloured, like the nitrate, but more feebly; and muriate of platinum, which underwent no alteration.

8. Sea-water and solutions of common muriate of soda and of muriate of magnesia, when boiled, and when the disengaged vapour is passed through a solution of nitrate of silver, occasion appearances of a more complicated nature. In this case, the metallic solution is decomposed, not merely by aqueous vapour, but by the muriatic acid which is disengaged from the boiling liquid; and a quantity of chloride of silver is formed, which the action of the aqueous vapour subsequently renders violetcoloured, provided it at the same time raises the temperature of the solution of nitrate of silver to the boiling point. Should any portion of the nitrate of silver be left undecomposed, it is deoxidized by the vapour of water, and a yellow or brown colour is developed, which mixes with the violet, and imparts to it various modifications of shade. My original opinion, therefore, that the change of colour is caused by the muriatic acid which distils over along with the water, is confirmed, but at the same time restricted, by this experiment. Muriatic acid is also disengaged from a boiling solution of the common muriate of soda, although in much smaller quantity, than from a solution of muriate of magnesia. Whether in the case of muriate of soda, the acid proceeds from a small residue of muriate of magnesia, or muriate of iron, I do not venture to determine. Neither of the bases (soda, magnesia) appears to pass over: at least, the distilled water leaves no residue when evaporated.

9. This yellow, reddish-brown, or dark-brown coloured solution of nitrate of silver, produced in so remarkable a manner by the deoxidizing action of aqueous vapour, retains its colour for a long time unaltered, but it at last deposits a dark-brown oxide of silver.

10. If previously to the introduction of aqueous vapour the solution of nitrate of silver be raised to the boiling point by the immediate application of heat, it does not completely retain its transparency after having been exposed as usual to a current of the vapour, but the discoloration which it sustains is greatly

inferior.

The deoxidizing property of aqueous vapour, demonstrated beyond a doubt by the foregoing experiments, deserves to be still further investigated, and would, perhaps, already admit of some practical applications. It is my intention to continue my experiments on the subject. Before concluding, I may observe, that the vapour of water does not appear to produce any change upon a solution of corrosive sublimate, or upon solutions of the protoxide or peroxide of mercury in nitric acid.

ARTICLE X.

On the Transmission of Electricity through Tubes of Water, &c. By Mr. C. Woodward.

SIR,

(To the Editor of the Annals of Philosophy.)

June 5, 1824.

ON perusing the last number of your Annals, I observed a letter signed T. J., informing me" that the experiment of firing loose gunpowder by passing the charge of a leyden phial through tubes filled with water, and also on the conducting power of alcohol, ether, and acids, were made by a Mr. Lewthwaite, in May 1821; and are published in the eleventh volume of the Institution Journal:" the natural inference of which, I apprehend, is, either that I published some experiments as new, which were not so; or that I gave as my own, the experiments of another.

If T. J. will refer to my letter, I think he will perceive that my object was to offer a theory in explanation of a singular phenomenon, and not merely to state the fact of the inflammation of loose gunpowder, or the conducting power of alcohol, ether, and acids. For this purpose I introduced the subject as briefly as I could, and then enumerated the experiments, which led me to conclude that the theory I offered was the cor

rect one.

"It was (observes T. J.) from reading this letter that I became acquainted with the experiment.' This, however, was not the case with myself, as the effects of electricity on loose gunpowder, when transmitted through tubes of water, were communicated by me to Mr. Lewthwaite some time previous to the publication of his letter.

It is extremely painful to speak of one's self; therefore, in my last communication, I avoided any allusion to what I had done elsewhere; but considering myself now called upon to explain, allow me, through you, to inform T. J. that I introduced the experiment in my concluding lecture on Electricity, at the Surry Institution, in December 1820; observing at the same time, that I could not offer any theory in explanation, the experiment having been but a few hours communicated to me by my much esteemed friend, Mr. Knight Spencer, the Secretary to the Institution.

Early in 1821 I instituted a series of experiments to ascertain the cause; and, although I had then no explanation to offer, my experiments would have been presented to the public through the medium of one of the philosophical journals, had not the appearance of Mr. L.'s letter in the Institution Journal