a very small jar through the first circuit, to make instantly visible at all points of interruption a spark which seems everywhere to have the same strength and lustre. My first researches have been directed with the end of establishing the direction of the different currents, induced and inducing at the same time. I commenced with two pairs of spirals, and employing two galvanometers in communication with the second and with the fourth spiral. My experiments have been made with the battery of twelve jars, from 10° up to 40° of tension. The pairs of spirals were only separated by a plate of very thin glass. The following table presents the results of my experiments, which I have repeated several times with M. Pacinotti, my colleague, who has considerably aided me in these researches. (Fig. 1). When we employ three pairs of spirals in place of two, the results with the same discharges are produced in the same direction. (Fig. 2). We see very easily that when the induced current becomes an inductor, the current which it develops is always directed as it would be if the induction was produced by a voltaic current which originates it; that is to say, that the direction is the inverse of that of the inducing current. This is precisely the contrary of that which happens to the first current of induction which the discharge produces: the current of induction has the same direction as the inducing current; it acts, consequently, as a voltaic current which intermits.

I wished to study the relations of intensity of these different currents of induction. I always give the numerical results, still being a little diffident of the correctness. At first the needles of the galvanometer lose much of their degree of magnetism by the effect of these currents, and there is always some little spark which explodes. It is for this reason that in all these experiments I have never taken notice of any result without ascertaining whether, in reversing the position of the extremities of the galvanometer, the direction of the deflection was equally reversed. By employing the same galvanometer successively united to the spirals 2, 4, 6, and with the same charge of 10°, I obtained 7° ex., 3°, 1°. These numbers give the intensity for the current of induction of the first, second, and third order.

It was important to study the direction of the current when the spark was given out in the secondary circuit. Having made some trial with the galvanometer, I found that the deviation was hardly sensible in this case, and that even when discharging my jars directly, the deviation was not so great as with my currents of induction. I employed, in place of the galvanometer, a very simple process which was suggested to me by M. Pacinotti. It is that of the hole which the electric spark always makes in paper near to the negative point. The apparatus is very simple, very sensitive, and constant in its results. By interrupting my circuit of the second and third order, I had sparks of induction which yet made a very visible perforation. I took two plates of pewter terminating in a point. I pasted them, one above and the other underneath a morsel of common paper, leaving the two points at two millimetres distance between each other

When the spark was given out a hole was found at the negative extremity. Ordinarily we obtain another sign equally constant. It is that of a black stain near the positive point. In a very great number of experiments that I have made, the phenomenon of the hole at the negative point has always been constant; the other sign very rarely fails. I then disposed several of these pieces of apparatus in my spirals; that is to say, at the second, at the fourth, and at the sixth spiral. We see in every place, with a discharge of 10° of the battery of twelve jars, that the hole is formed, and the direction of the different currents of induction may be seen very readily in the following tables. (Figs. 3 & 4). The law of the direction of the currents of induction produced by the discharge of the battery, and of currents of induction which the induced currents develope, is evident. When the secondary circuit is open, and there is a spark, the secondary current is directed in an inverse direction to the primary current, and this result is verified for the same current of the jar. In fact, the first current of induction has a direction opposite to that which it has when the circuit is closed. The induction is then subject to the same law as voltaic currents at the commencement of their action. We can see what happens when we compare closed circuits with those which are broken. We find everywhere verified that which we saw occur to the first secondary current: if the circuit is closed, the secondary current has the same direction as the current of the jar. I here give several tables which evidently prove it, and the results of which have been many times confirmed. (Figs. 5 and 6). We see, then, that whichever of the secondary currents it is that we take, the current which is developed by induction is always in the same direction as the inducing current, if its circuit is closed, the other being open; or if its circuit be open, the other being closed. If we hold the galvanometer in the open circuit, we obtain very small deflections, and hardly sensible to the instrument I employed. I have always found that when the two ends were very near each other, the needle deviated, giving the same indication as the hole. If we remove the points more remote from each other, the indication of the galvanometer will appear reversed. This is a subject which merits a profound study, and for which I have already disposed apparatus with M. Pacinotti. Perhaps the galvanometer falls short, in this case, by the inductive actions which are always contrary to the current which is commencing, and similar to the current which is ceasing, not being separated. This is a question which we may perhaps resolve when we can obtain signs of induction by a discharge from a battery slowly produced, in such a manner as to see, when in darkness, a continued current of light. Hitherto all the attempts which have been made to accomplish this have been useless. I am persuaded that it is necessary further to insulate the wire of the galvanometer and that of the spirals.

I have only made a small number of experiments on the action of interposed plates between the spirals. At first I said that if we oblige the primitive current of the battery to produce two secondary

currents, by placing it in the middle of two spirals, the current of induction which it developes in each of these spirals is always directed as when there is only a single secondary spiral, and the intensity of the current is equal to that which it developes when we suppress one of these secondary spirals. One of these secondary spirals has not any effect on the enfeebling of the secondary current, if its circuit is not closed. On the contrary, if we put in contact with the secondary spiral another spiral closed, in such a manner as to place it between the spiral which is traversed by the discharge from the battery and another precisely similar, the secondary current has the same direction as it would have without this spiral, and its intensity appears augmented.

The action of non-conducting plates interposed is nothing. If we employ metallic plates, the secondary current is considerably enfeebled, but its direction is not changed. It is thus that a plate of zinc of 1 millimeters of thickness destroys the current of induction from the discharge of eight jars at 40°, and at the distance of 0TM, 01, between the two spirals. A very thin plate of pewter has no influence on this discharge; it requires five, one on the other, to reduce the deviation to 5°, from 9°, when it was without plates. A plate of silver and copper, very thin, hindered the production of the secondary current. It appears, then, that in order to reduce from a given quantity the property which the current possesses of producing induction, it is necessary to interpose a metallic plate whose thickness shall be in inverse ratio to its conductibility.

I was desirous of trying if it was possible to obtain any signs from the galvanometer, by cutting the plate of pewter, and making the two points of the cut disc touch the two wires of the galvanometer. I could obtain nothing with the strongest discharges. This is sufficient to prove the great superiority of the process of magnetism. I prepared a disc of pewter whose diameter was four times greater than that of my spirals. By soldering the wires united with the cylindrical spirals in different parts of this disc, I obtained, with very feeble discharges, a strong magnetization on the steel needles which were introduced into my spirals.

The process of magnetizing merits a more profound study. Hitherto I have not been able to obtain any signs of action in the galvanometer with the discharge from a small bottle which constantly gives a secondary current capable of magnetizing, as a voltaic current would do, directed in a contrary direction to the current from the jar or bottle. This fact is constant.

Pisé au Cabinet d' Phys. del Université,

10 Fevrier, 1841.

[NOTE. The tables and figures referred to, are not in the original; and therefore some mistake must have occurred in the printing of it.-EDIT.]

C

On some peculiar Phenomena exhibited in a Solution of Nitrate of Silver, traversed by an Electric Current. By M. C.

MATTEUCCI.*

IN a former memoir I have described the following fact. When an electric current is transmitted through a solution of nitrate of silver, a black flocculent matter is deposited on the negative platinum terminal; but which becomes instantaneously transformed into a silvery appearance whenever it falls from the terminal into the liquid, or when the current ceases to be transmitted.

The production of these phenomena can be accomplished only under certain circumstances. If the solution be strong, the silver is always liberated in a very well crystallised state: and it is only by employing a very dilute solution that we succeed in producing the above described phenomena. If the solution be of 100 water to 1 of nitrate of silver, the black flocculent matter is readily produced but when the water is in greater proportion, it no longer appears. Its production also depends upon the power of the current, which must be stronger as the solution is less charged with the salt. Whether the silver be liberated in the flocculent or in the crystalline state, the quantity does not vary to a sensible extent; being equal to that obtained by calculation founded on the quantity of water decomposed in another part of the same electric current. When a portion of the transformed silvery matter is placed between the platinum terminals in a solution of nitrate of silver, those parts of it near to the positive terminal become blackened, but suffer no change by the current when similarly situated in liquids containing no nitrate of silver; though a single drop of the nitrate in them would cause the blackness to appear.

The black matter is transformed to the silvery white state whenever the platinum terminal to which it is attached is shaken in the liquid; but returns to its blackness by letting fall upon it a stream of a similar nitrate solution. If, whilst still attached to the negative platinum on which it was formed, the black flocculent matter be transferred to a strong solution of nitrate of silver, it suffers no change by the new current, but becomes covered with crystallised silver, which forms upon it in the same manner as it would appear on the platinum terminal alone.

The formation of the black matter occurs whether the terminals in the solution be four inches or only one inch apart. If a slip of glass be placed between them the deposit will be white: but becomes black when the glass is removed. This fact, I think, is attributable to the different degrees of intensity which the current assumes by the presence or absence of the glass; for, as the deposited matter which forms is black, with great distances between the terminals, the absence of all colour when the glass interposes, cannot be attri

Les Archives de L'Electricité.

buted to any action which the polar products exercise on one another; which, under such circumstances, could not happen.

Acetate of silver, when dissolved in a large proportion of water, affords the same phenomena as the nitrates: and they as uniformly appear when to the solution an excess of the nitric and sulphuric acid has been added. We might suppose that the black deposited matter is composed of oxide of silver which is protected by the transmission of the current; but which, when the current ceases to flow, immediately assumes the metallic state. This, however, is an hypothesis which must be considered as perfectly gratuitous, for I must confess that, hitherto, I am unable to account satisfactorily for the phenomena.

Pisa, July, 1871.

Historical Sketch of Professor Schoenbein's Experiments on a peculiar Voltaic Condition of Iron.*

The first series of these experiments on record that we are acquainted with, are described in a letter to Dr. Faraday, dated Bâle, May 17, 1836. They are as follow:

"If one of the ends of an iron wire be made red hot, and after cooling be immersed in nitric acid, sp. gr. 1.35, neither the end in question nor any other part of the wire will be affected, whilst the acid of the said strength is well known to act rather violently upon common iron. To see how far the influence of the oxidized end of the wire goes, I took an iron wire of 50' in length and 0""'-5 in thickness, heated one of its ends about 3" in length, immersed it in the acid of the strength above mentioned, and afterwards put the other end into the same fluid. No action of the acid upon the iron took place. From a similar experiment made upon a cylindrical iron bar of 16' in length and 4"" diameter, the same result was obtained. The limits of this protecting influence of oxide of iron with regard to quantities I have not yet ascertained; but as to the influence of heat, I found that above the temperature of about 75° the acid acts in the common way upon iron, and in the same manner also, at common temperatures, when the said acid contains water beyond a certain quantity, for instance, 1, 10, 100, and even 1000 times its volume. By immersing an iron wire in nitric acid of sp. gr. 1.5 it becomes likewise indifferent to the same acid of 1.35.

"But by far the most curious fact observed by me is, that any number of iron wires may be made indifferent to nitric acid by the following means. An iron wire with one of its ends oxidized is made to touch another common iron wire; both are then introduced into nitric acid of sp. gr. 1.35, so as to immerse the oxidized end of the one wire first into the fluid, and to have part of both wires above the level of the acid. Under these circumstances no chemical action upon the wires will take place, for the second wire is, of course, but • Phil. Mag.