employed, it did not till now occur that the time of transmission would vary with the calibre of the tube.) I had not at this time any narrow tubes in my possession. Considering, however, that if any non-conducting substance-such as silk, or paper-were moistened with water, that those substances could have no more conducting power than what was imparted to them by the moisture; my first experiment was with a single thread of sewing silk, about four inches long, well moistened by drawing it between my lips. This thread was made a part of the circuit between the inside and the outside of a charged jar. At another part of the circuit an interruption was made between the extremities of two wires; and at this interruption was placed some gunpowder. On discharging the jar the gunpowder ignited. I repeated the experiment several times. with the same success. I afterwards varied the experiment, by using the same thread and a smaller jar, and succeeded in igniting gunpowder with about thirty inches of charged surface. I must here observe that when the thread was very wet, I never succeeded with this small jar, owing, as I suppose, to the quantity of water contained in the thread being too great to retard the small quantity of electric fluid contained in the jar. For by squeezing out some of the moisture, the thread became a worse conductor, and then I always succeeded.

succeed with the first jar, For this purpose a piece This twine, however, con

I next tried how far it was possible to and augmenting the quantity of water. of twine was used, well soaked in water. ducted the electric fluid with too much facility to ignite the gunpowder; but when some of the moisture was squeezed out it answered very well. Thus by proportioning the one with the other, I always succeeded. The same results were obtained by using moistened paper.

Having satisfied myself on this point, I next endeavoured to ascertain if the electric fluid undergoes any change in its physical character by passing through water; or if the ignition of gunpowder depends entirely on the time occupied by the fluid to pass through

it.

For this purpose I employed two jars, which for distinction we will call A and B. I charged A positively and B negatively, and connected their outsides by water. On exploding A into B, through water, both jars became neutralized. A was again charged positively, and afterwards partly discharged through water into B. On discharging both jars separately there appeared no difference in the explosions. A was once more charged positively, and again partly discharged through water into B. On discharging B through a moistened thread, gunpowder was ignited in the circuit. These experiments were reversed by charging the jar A with negative instead of positive electricity, and the results were similar. Hence I concluded that the ignition of gunpowder by the electric fluid depends on the time occupied by the latter in passing through it, and not on any change in the physical character of the fluid.

Hence also, by the foregoing experiments (when the intensity of the charge is constant), the time occupied by any given quantity of the electric fluid, in passing any one point in the circuit, will be in some reciprocal proportion to the thickness of the column of water employed in that circuit.

It was observed in these experiments, that the moistened thread soon became nearly dry. Hence the electric fluid had either decomposed the water or caused it to evaporate. The former effect I suppose to have taken place; which, if true, perhaps the decomposition of water by this agent may be facilitated by reducing the diameter of the column employed. I have not yet had time to ascertain this particular satisfactorily, therefore it remains a mere supposition.

I have frequently discharged a jar through my own body without any other inconvenience than a burning sensation at the extremities of the fingers; and have ignited gunpowder in the same circuit.

I

So modified is the electrical discharge by being transmitted through aqueous conductors that the effect of an intense charge of the most powerful battery may be reduced to almost any degree. have discharged eight feet of charged surface through my own body without feeling the least shock. But the burning sensation was very severe.

In medicine this modification of electricity can hardly fail to be useful; for it may be administered to any particular part of the body without affecting any other part. It may be applied to the skin of the most delicate patient; and, without the least danger of. giving a shock, a most powerful stream of electricity may be poured on the part affected.

It has been also ascertained, that the force of the electric discharge through metal, is always proportioned to the thickness of the wire through which it is transmitted; or, that the same quantity and intensity of electric fluid acts with a greater force when transmitted through a thick than through a thin wire.

This law, of course, has a limit; for if the conducting wire be sufficiently stout or capacious to transmit the fluid without interruption, a wire of larger dimensions can give no more facility to the transmission. I am of opinion, however, that thick wire facilitates the transmission of the electric fluid to a greater degree than is generally suspected. Now it is evident that as thin wire has the property of diminishing the intensity of an electrical discharge, the fluid during its transmission through such a wire must necessarily be

This proportion may vary either as the diameter, or as the square of the diameter of the column: according as the electric fluid occupies the surface, or the whole body of the water; and if the velocity of each individual particle of the electric fluid be affected, the transmission of the whole mass will also vary on that account.

+ Eight feet of lining and eight of coating.

B

drawn out (as it were) into a longer stream, than if it were transmitted through a thick one. In the same manner that a certain quantity of water would be drawn out into a longer stream by passing through a narrow than through a wide tube; or as a piece of metal would be drawn into a longer wire by passing through a small than through a large hole in the plate.

Hence it became a curious question, What length of wire of a given dimension does a certain quantity of the electric fluid (say a jar charged to a certain intensity) occupy at any moment during its transmission? I supposed that this might probably be ascertained by placing some gunpowder at an interruption near to the positive side of the jar, and the moistened thread near to the negagative side, having a long copper wire between them. For if the wire was of sufficient dimensions to contain all the fluid at once, the latter would not meet with any resistance till it arrived at the moistened thread; and the time of its transmission through that part of the circuit where the gunpowder was placed would not be lengthened; upon which supposition the gunpowder ought not to be ignited.

With such arrangements I have separated the gunpowder and moistened thread by a copper wire, No. 16, of different lengths, from one to twenty yards; yet with all this length of good conducting substance beyond the gunpowder, the latter never failed to be ignited.

I consider this a very curious circumstance, and the inquiry important. Not so much because the gunpowder always ignited in these experiments (for it is possible that with longer and stouter wire, and a smaller charge, the ignition might not take place); but, because, if it could be proved by experiment that the electric fluid would, by intervening capacious good conductors, ignite gunpowder at the negative, and not at the positive side of the moistened thread, such experiment would prove to demonstration the truth of the Franklinian hypothesis.

I am well convinced that if the electric fluid ever passed the gunpowder without interruption, the latter could not be ignited by any recoil of the fluid into the jar from the interrupting moistened thread; because if ever it passed through the gunpowder with violence, it would scatter, or blow the latter substance away, so that none would remain in the circuit to be ignited at the time of the fluid's return.

Several experiments on the ignition of gunpowder by the electric fluid were exhibited, and demonstrated on the foregoing principles, in a lecture which I had the honour to deliver before the members of the Western Literary and Scientific Institution, on Monday evening last (May 8th), in the Concert Room, King's Theatre, Haymarket. In one experiment the gunpowder was ignited at one extremity of the moistened thread; and in another experiment four guns were fired in the same circuit. I have frequently fired six

guns by one discharge of a jar; and so instantaneous is the ignition at the several guns, that their united reports appear like the report of one gun only.

I am at present engaged in other experiments on this branch of electricity, and shall not fail to communicate to the public the results of such as appear worthy of notice.

Artillery Place, Woolwich,
May 15, 1826.

I am, Sir, your's faithfully,

W. STURGEON.

On the Inflammation of Gunpowder and other Substances by Electricity: with a Proposal to employ the Term Momentum as expressive of a certain Condition of the Electric Fluid. By Mr. W. STURGEON.

To the Editor of the Philosophical Magazine and Journal.

SIR,-Since the publication of my paper on the ignition of gunpowder by the electric fluid, I have had an opportunity of perusing Mr. Woodward's paper* on the same subject. But I find that, like all other authors I have yet read, Mr. W. has left us entirely in the dark with respect to the law that is necessary to be observed in varying the experiment. For although the dimensions of his jar and tube are given, no mention whatever is made why it is necessary to observe those dimensions, or why other jars and tubes of various other dimensions may not answer as well. Indeed, so little does Mr. W.'s hypothesis concur with the conclusions deduced from my experiments, that he supposes the water derives its retarding property solely from its confinement in tubes. For it is stated in Mr. W.'s paper: "If the tube be filled with ether or alcohol, and placed in the circuit, the powder will be inflamed. If it be filled with sulphuric acid, which is a better conductor, the powder will be scattered and not inflamed: but the dispersion will not be so great as when metals only form the circuit.

"The same effect will be produced by transmitting the charge through the animal economy, or through water not inclosed in tubes, in which case the water does not appear to oppose a sufficient resistance to the passage of the fluid."

It is evident from this statement of Mr. Woodward's, that he never varied the experiment by employing tubes of various diameters; his experiments being with different fluids in the same tube. Had Mr. W. taken into consideration the necessity of varying the

This paper is one of those you so obligingly pointed out to me for perasal; and appears at page 283, vol. vii, of the new series of the Annals of Philosophy. The two papers in vol. viii, contain nothing material on the subject.

diameter of the tube with the nature of the charge, I suspect he would have found little difficulty in igniting gunpowder by transmitting the electric fluid through sulphuric acid, it being necessary to observe nothing more than to augment the charge, or reduce the diameter of the tube.

That the retarding property of the water does not depend on its confinement, is evident from the success of my experiments with a moistened thread exposed to the open air. Indeed, so certain it is, that confinement does not impart to water this property, that gunpowder may be ignited in the electrical circuit, by transmitting the fluid through a sufficiently narrow streak of water, drawn on the surface of a piece of flat glass.

I should, however, be extremely sorry to detract any thing from the merits of any man; and although Mr. Woodward's explanation of igniting gunpowder by the electric fluid, when transmitted through water, does not exactly agree with the principles on which I suppose the action to depend, I nevertheless should be wanting in candour, were I not to express my sentiments on the highly interesting nature of his experiments: and I feel a pleasure in acknowledging that Mr. W. has preceded me in some that are detailed in my paper (although I did not know it at the time); viz., those wherein gunpowder was ignited on both the positive and negative side of a long conducting wire.

Mr. Leuthwaite has likewise conducted a number of interesting experiments on this subject; but I believe they were intended chiefly to ascertain the conducting power of different fluids, in order to make choice of the most eligible for the purpose of igniting gunpowder by the electric fluid.

Hence, so far it appears that the idea had not been entertained (prior to the institution of my experiments), that the diameter of the tube, and the nature of the charge, would any way affect the result of the experiment. Neither does it appear that the ignition of gunpowder, by means of transmitting the electric fluid through unconfined water, was ever attempted prior to those experiments detailed in my former paper on this subject. The most extraordinary method of effecting the ignition of gunpowder by the electric fluid, that I have yet heard of, is that stated by Mr. Howldy, in the Philosophical Magazine, vol. lxviii, p. 173. I have been induced to pay some attention to this method, "especially as it saves the experimenter time, labour, and power," circumstances highly important and necessary to be understood.

I think, however, it is to be regretted, that Mr. Howldy has not mentioned the hygrometrical state of that part of the table ("four inches") between the extremity of the chain and the outside of the jar as it is possible, that a variation in that particular may vary the result of the experiment. But as Mr. H. has practised this method for several years, and with uniform success, it is to be expected that such a circumstance could hardly escape the notice of so accurate an