disc, and reciprocally. Finally, as it results from numerous experiments, Mr. Christie succeeded, as we shall see, in deducing from the force with which different substances draw along the magnetised needle in their rotatory movement, the conducting power of these substances for electricity.

But one of the most important facts is due to MM. Ampère and Colladon, who found that, in all the experiments on magnetism by rotation, the place of the magnet might be supplied by a helix, through which was transmitted an electric current; thus establishing a new analogy between a magnet and an assemblage of electric currents, circulating all in the same direction, in closed circuits parallel to each other, transversely to an axis. The helix traversed by a powerful electric current was movable, and a copper disc, in rotation beneath it, drew it along with it as it would have drawn along a magnetised needle.

Very numerous experiments have been also made with discs of iron and steel, and with solid and hollow iron spheres. When the iron is very soft, the results are greatly similar to those that are obtained with other metals, only they are more energetic; but a steel disc does not produce any appreciable effect upon the magnetised needle, which, after a few irregular oscillations, remains in its position of equilibrium. M. de Haldat, to whom we owe these observations, concludes from them, that the drawing force is for magnetic bodies in inverse ratio to the coercitive force. We shall see further on that this conclusion cannot be admitted, the phenomenon not being a magnetic phenomenon, properly so called, but an electric phenomenon.

Mr. Barlow studied in detail the influence that is exercised by an iron sphere in motion upon a magnetised needle. He took the precaution to neutralise, by the neighbourhood of a fixed magnet suitably placed, the influence of terrestrial magnetism upon this needle. The latter was placed sometimes tangentially to the iron sphere, sometimes parallel to its axis of rotation, which axis itself, from the construction of the apparatus, might have different directions. According to the direction in which the sphere rotated, either the north or

the south pole of the needle was seen to recede from it; the repulsion exercised upon either of the poles depended also upon the part of the iron sphere towards which the needle was placed; in other words, the direction of the deviation of the needle changed according as it was placed on the south or on the north of the moving sphere. All these effects are evidently due to the combined influence exercised upon the iron globe, both of terrestrial magnetism, and of the magnetised needle placed in its neighbourhood. It would have been curious to have proved the influence of the first cause alone by putting a needle of soft iron in place of the magnetised needle.

A very important point, that was established by Mr. Barlow, is the great difference in the action exercised by an iron sphere according as it is solid or hollow. This difference is completely null when the globe and the magnetised needle are at rest, which arises, as we have said, from the ordinary magnetic force being entirely concentrated upon the surface; but as soon as there is motion this ceases to be the case. Thus, under the same circumstances, and by employing the same needle, a solid iron cannon-ball, making 640 turns per minute, and weighing 68 pounds, and being 787 in. in diameter, determined a constant deviation of 28° 24′; whilst a hollow ball of the same diameter, but weighing one-half less, determined a deviation of only 15° 5'. These two numbers are the mean of many experiments made with great

care.

M. Poisson, who had already submitted to mathematical analysis the labours of Coulomb upon magnetism, endeavoured to explain by the same theory the phenomena of magnetism by rotation. Attributing all magnetic phenomena to two imponderable fluids subjected to the general laws of equilibrium and of motion, which attract and repel each other in the inverse ratio of the square of the distance, he established that the great difference existing between these fluids and those to which electric actions are due, is, that the latter are able to pass from one molecule to another; whilst the magnetic fluids, during magnetisation, undergo only feeble dis

placements, which are not directly appreciable. M. Poisson gave the name of " magnetic elements" to these small portions of bodies, in which the south and north magnetic fluids are able to move, and which are separated from each other by intervals impermeable to the magnetism. All bodies are susceptible of having their natural magnetism decomposed; but more or less easily according to their coercitive force. It is to this force, which is insensible in some bodies, that M. Poisson attributes the difference existing between them with regard to the actions they exercise according as they are in rest or in motion. In the state of rest, bodies whose coercitive force is null should not exercise any sensible action upon the magnetised needle, or, at least, should exercise only a very feeble action; but, in the state of motion, calculation demonstrates that things are not the same. M. Poisson also succeeded in determining in this case à priori the existence of the three components that M. Arago had discovered by experiment.

However, the theory that serves as the basis to M. Poisson's calculations was overthrown by the subsequent discoveries of Mr. Faraday, which give an entirely different explanation to the phenomena of magnetism by rotation. So that we will not detain ourselves with them longer, merely remarking that mathematical analysis may be further consulted usefully by those who, by resting on hypotheses different from those which serve for M. Poisson's starting-point, would submit them to the proof of calculation. We shall therefore now pass on to the phenomena of induction, with which Faraday enriched the science in 1832; and in which, as we shall see, are naturally included those of magnetism by rotation.

Production of Induced Currents and Explanation of Magnetism by Rotation.

In 1832, Faraday discovered that an electric current or a magnet is able by induction to develope at a distance electric currents in a conducting wire; just as a body, charged with static electricity, electrises an insulated conductor by

induction. The following is the mode by which this remarkable result is obtained.

We wind round a wooden cylinder two silk-covered wires, so as to make two perfectly similar helices, and the spirals of which are parallel, and as near to each other as possible (Fig. 141.). The two ends of one of the wires are made to

communicate with a galvanometer, and the two ends of the other with the two poles of a pile. At the moment when this latter communication is established, the first having been established previously, the needle of the galvanometer is seen to deviate; but this deviation immediately ceases, even though the current of the pile continues to circulate. As soon as this current is interrupted, the needle of the galvanometer a second time experiences a sudden and non-permanent deviation; but this deviation occurs in a contrary direction to that in which the former had occurred. Thus the voltaic current that traverses one of the wires determines in the other an instantaneous current, at the moment when it commences to pass, and determines in it a second, at the moment when it ceases to pass. These two currents are called induced currents, and the current of the pile the inducing current; the induced currents, as we see, are instantaneous: let us further add that the former has a contrary direction to that of the inducing current, and the latter a similar direction.

The experiment may be made under another form.

wind round a wooden or glass tube a single silk-covered wire, the two ends of which are in communication with the wire of a galvanometer; we then suddenly introduce into the tube an electro-dynamic cylinder, namely, a helix traversed by an electric current, and we then draw it out in the same manner. At the moment of the introduction, we obtain in the outer helix a current of induction, the movement of which is in a direction contrary to that of the current of the electrodynamic cylinder; at the moment when we withdraw the cylinder we obtain a second induced current, the movement of which is in a direction the same as its own. In order that these two currents may be sensible, the electro-dynamic cylinder must be introduced and withdrawn very suddenly. This experiment, as may be readily seen, comes to the same thing as the preceding one: in the latter, we possess the advantage of creating and destroying the electro-dynamic cylinder instantaneously by closing and opening the circuit; whilst in the other we introduce it and withdraw it,-an operation which cannot be executed with so much rapidity. We shall see further on, that there results an essential difference between the induced currents; those which are produced according to the latter mode having a duration in a slight degree sensible, whilst those that are produced according to the former mode are altogether instantaneous.

Whatever may be the case, the two experiments equally prove that, when we suddenly bring near to a part of a conductor, forming a closed circuit, a conductor traversed by a current, we determine in the former an instantaneous current, moving in a direction contrary to that of the current brought near to it; and that, when we remove it, we determine a second instantaneous current, moving in the same direction as the current that is removed.

The analogy existing between the properties of magnets and those of electro-dynamic cylinders, led Faraday to suppose that the same results would be obtained by introducing into the interior of the hollow helix of the second experiment a magnet instead of an electro-dynamic cylinder. This is what in fact happened. For this experiment we may employ the