mode of suspension, and in particular to obtain direction of a rectangular or circular current by means of the terrestrial

globe. We must only take great care, when we employ but a single voltaic pair, not to forget, as we have already remarked in the First Part of this work, that the positive electricity sets out from the copper or platinum plate, and the negative from the zinc plate; that, consequently, it is, according to our conventional mode of describing it, from the copper or platinum plate that the current is found to start in order to traverse the conductor, to go thence to the zinc, and to return through the pair to its point of departure. In order that we may be able to put the different conductors successively into the circuit of the pair, the cork disc carries two small cylindrical capsules of wood filled with mercury, which communicate, the one with the platinum, the other with the zinc of the pair, and into which the extremities of the conductor through which the current is to be transmitted are made to plunge.

At the same time that M. Ampère took up Oersted's discovery, in order to generalise and to extend it, M. Arago showed that an electric current not only acts upon a magnetised needle, but that it also acts upon all magnetic bodies, even when they are not magnetised. Having slightly curved a copper wire of about th of an inch in diameter, he saw that, when this wire was traversed by a strong current, it acquired the property of attracting and retaining around it, under the form of a cylindrical envelope, a certain quantity of iron filings, and that immediately the current ceased to pass, the filings fell; the wire took them up again as soon as the current

passed. This experiment proves that the electric current impresses upon conductors, when it is transmitted by them, properties perfectly analogous to those of magnets, and not simply to those of magnetic bodies; in other words, that it magnetises them, and does not simply render them susceptible of being magnetised. M. Arago went further, by showing that the discharge of a Leyden jar may magnetise a steel needle placed in the interior of a helix made of wire, through which this discharge is made to pass. Davy shortly afterwards discovered that we can in like manner magnetise a sewing needle by merely rubbing it transversely against a rectilinear wire, traversed by the electric current of a pile. We shall return to this subject in the Third Chapter, which is entirely devoted to it.

Mutual Action of two electric Currents.

From the origin of these researches M. Ampère perceived that an electric current not only acts upon a magnet, but that it also exercises an action upon another electric current. He found that this action consisted in that, if two portions of rectilinear currents parallel to each other are both movable, or are the one fixed and the other movable, they are mutually attractive when they are moving in the same direction, and repulsive when they are moving in a contrary direction. The attraction in this case does not cease with contact, as occurs when we are referring to the attraction of electric bodies in static electricity; it remains so long as the current continues to traverse the conductors.

In order to demonstrate this principle by experiment, we may employ the floating pile, or Ampère's apparatus, adapting to either of them the vertical astatic conductor. We present to one of its vertical branches a parallel wire traversed by the current of a rather powerful pile. We perceive that this wire, which may simply be held in the hand, attracts the branch of the rectangle, if the two currents are moving in the same direction, namely, both in like manner from above downwards, or upwards from below, and repels it if they are

moving in a contrary direction, the one from below upwards, and the other from above downwards (Fig. 86.).

M. Ampère was not long in generalising the law of parallel currents by extending it to the case of angular currents, that

Fig. 86.

is, to the case in which the two conductors, each being traversed by a current, form an angle by being situated either in the same plane or in different planes; in this latter case, the angle formed by the currents is that which is made by the two planes in which they are situated, and it has for its height the right line that measures their shortest distance.

The following is the law that Ampère discovered in this general case; it is that the two angular currents attract

each other, when their direction is such that they both tend toward the summit of the angle, or that they both set out from it, and that they repel each other when one goes towards the summit and the other sets out from it. This law comprehends four different cases, for the understanding of which it is necessary to cast the

AAAA

Fig. 87.

eye upon Fig. 87., in which the arrows that are outside the angular lines represent the direction of the currents, and those

We may hold the wire, by which the two poles of the battery are connected, in the hand, in order to present it to the movable current, without the current ceasing to traverse it, as the metal is a conductor so superior to the human body that the two electricities traverse it exclusively in order to unite with each other.

withinside the lines the direction in which the conductors move in respect to each other. When the suspension of the movable current and the arrangement of the fixed current are such that the angle which they form together remains invariable, this action may give rise to a continuous movement of rotation. But before studying this particular form, under which it is sometimes presented, we must establish it in a more direct manner by entering upon the case in which the arrangement of the experiment permits the two currents to approach or to recede, so that the angle formed between them increases or diminishes.

With this in view, we may employ the floating pile, attaching to it the horizontal astatic conductor. We present to the upper branch a rectilinear current, in like manner horizontal, so that it forms with it an angle, the summit of which is at one of the extremities of each of the two conductors (Fig. 88.). Imme

diately, whatever this angle may be, whether acute, obtuse, or right, we perceive, if the two currents are directed so as to converge to each other towards its summit, or to diverge from it, the movable branch, carrying with it the whole of the circuit of which it forms a part, turns round this summit in order to place itself parallel to the fixed conductor, and as near to it as possible. In this position the two currents have become

VOL. I.

parallel, and are determined in the same direction with respect to each other. If, on the contrary, one of the currents is at the outset directed towards the summit of the angle, and the other sets out from it, we perceive the movable branch still turning around this summit and avoiding the fixed conductor by increasing the angle which they form between them, and tending to place itself in the extension of this conductor by forming with it an angle of 180°, that is to say, the greatest possible angle.

As the distance between the different parts of the two conductors, which are not very near to the summit of the angle, is necessarily very great, the currents must be very energetic in order that the experiment may succeed well. In order to increase this energy, we may cause the wire of which the astatic conductor is formed to make several turns, having taken the precaution to cover it with silk, in order to prevent metallic contact. But, for experiments of this kind, it is preferable to make use of Ampère's apparatus, which enables us to employ the current produced by a powerful pile. We suspend upon it the horizontal astatic conductor, and, in order to act upon its lower branch, we adapt to it a fixed horizontal conductor, which is placed beneath, but as near to this branch as possible. We may cause the same current to pass through the fixed conductor that traverses the movable conductor if we prefer it, which is the more convenient plan, as we need not employ two different piles. This observation is also applicable to the experiment in which we demonstrate with this same apparatus the action of a fixed vertical conductor upon the vertical branch of the astatic movable conductor. But in order to realise every possible case by employing only a single current which traverses successively the whole system of fixed and movable conductors, we must be able to change the direction of the current, not only in the two conductors at once, which is always easy, since it is sufficient for this to change the place of the two poles of the pile, but also in one of the conductors only, without changing it in the other, which requires a particular contrivance.

The following is consequently the manner in which the apparatus of Fig. 82. is arranged so as to serve for all experi