also made, as we have already said, a very special study of the induced discharges themselves, employing for the purpose of measuring their intensity the heat produced in the very fine platinum wire of an electric thermometer, through which they are transmitted. This mode of operating could not give M. Riess the direction of the induced discharge; but he found, by the magnetisation of a steel needle, that this direction is the same as that of the inducing discharge. However, he recognised that the duration of the magnetisation impressed upon a needle by a secondary discharge may vary with circumstances independent of the direction of this discharge, as occurs in Savary's experiments for direct discharges. Furthermore, with regard to the intensity of the induced discharge, measured by the calorific effect, it is proportional to that of the inducing discharge, and to the efficacious length of the wire by which this discharge is conducted; but it is in inverse ratio to the distance of the two wires, and independent of the conductibility of the induced wire. The influence of interposed conductors upon the calorific effect of the induced current is very sensible, whether this conductor is a metal wire, the two ends of which are connected, or a plate, that is a very good conductor, either by its nature or its thickness. The interposition of a copper disc, 16 inches in diameter and in. in thickness, completely annihilated the effect of heating in the induced wire, although the inducing discharge was produced by four highly charged jars. Riess found that in general the intensity of the current in the secondary wire is the more diminished as the thickness of the interposed plate is greater.

To sum up, it follows from all the numerous researches made upon induced discharges, that this is a very complex phenomenon, and is influenced, in all the circumstances essential to its existence, by the very mode that is employed for determining it; on which account this determination is very difficult. However, we may reduce it to this very simple form, namely, that a discharge determines in a conductor that is near to the one by which it is transmitted, two

induced discharges, having the first a contrary direction, and the second a similar direction to that of the inducing discharge; that these two opposite discharges succeed each other in an interval of time of inappreciable duration; - that they nearly neutralise each other if the induced circuit presents no resistance, but that if the circuit is interrupted either by a fine wire that is heated, or by an interval that gives rise to a spark, then one of the discharges surpasses the other, and it is generally the latter, namely, the one that travels in the same direction as the inducing discharge. The cause of this superiority is essentially due to the conditions of the circuit, which more facilitate the transmission of electricity in one direction than in the other; it is also due to the fact that the direct discharge being the second, the causes that retard in general the propagation of electricity, must act proportionately with less force upon it than upon the former. However, we must not disguise that this manner of explaining the phenomena of the induction, produced by discharges or instantaneous currents, is not that adopted by several German philosophers, and in particular M. Dove. This philosopher, and others also, especially M. Weber, hold different views of this order of phenomena, and do not think that they can be reduced to such simple laws. Probably they are not altogether wrong. Further, we may judge of this by the explanation, unfortunately but a brief one, that we are about to give of their researches.

Influence that is exercised upon Induction by metal Masses, principally magnetic, placed in the Interior of Bobbins.

It remains for us, in terminating this Chapter, to study for a few moments a point that we have merely alluded to, namely, the influence that is exercised upon discharges, as well as upon induced currents, by the nature and arrangement of the metal masses that are introduced into the interior of the helices, intended for the production of induction. This influence is manifested by the modification that it exercises over the properties of induced currents. It is further

manifested, and is explained by the production of induced currents occurring on the surface itself of these masses, when they are continuous, and not formed of isolated pieces, in a greater or less number, and particularly on the surface of magnets.

M. Dove made a very special study of this subject. With this view, he successively employed currents and discharges to produce induction, and arrived at perfectly concordant results. Laying down the position that, in a current, namely, in the neutralisation of two contrary electricities, we must distinguish two elements, the original intensity of these two electricities, and the time that their neutralising lasts; or, which comes to the same thing, the force of the current and its duration, he succeeded in recognising that, according to the nature of the effects produced, one of these elements exercises a greater influence than the other. If the magnetic, chemical, physiological, and calorific effects of an electric current depended equally on its force and on its duration, the equality recognised between currents for one of these classes of effects would have equally occurred for the three others. But things do not occur in this way. The differences that are observed between the effects of two currents, arising from the neutralisation of equal quantities of electricity, must therefore be attributed to a difference in the duration of this neutralisation. These differences are especially sensible when the galvanometric and chemical effects, that are proportional to each other, are compared with the physiological effect. Now this latter is by no means proportional to the deviation of the needle nor to the quantities of gas given by the voltameter; it is not, like the two others, a product of the duration by the force; it only depends on this latter, and increases consequently with the rapidity of the neutralisation. Thus it is that the same discharge of a Leyden jar, which violently shakes the body, and does not cause the needle to deviate, may, if it is slowly drawn off by a point, affect a galvanometer, the coils of which are well insulated, and not produce any shock upon the human body

placed in its route. It is, however, the same quantity of electricity in both cases. The property possessed by the electric current, of magnetising tempered steel, is of the same order as its physiological effect. If, therefore, there are two currents developed in the same conductor,—and that these currents produce the same deviation in the galvanometer, and that one determines a more powerful physiological effect and more vivid sparks than the other, and communicates more powerful magnetisation to steel, we must conclude from this that the same quantity of electricity is moved in less time in the former than in the latter; and reciprocally, when the physiological and magnetising effects of the two currents shall be the same, that one of the currents whose effect shall be the least upon the galvanometer will merely have existed for a shorter space of time, still being however of the same intensity.

These differences, which distinguish in a striking manner the phenomena of the electricity of common friction machines from those of voltaic electricity, are no less sensible when we compare inductive currents one with the other, and when, without making any change in the induced circuit, we confine ourselves to modifying the nature and the mechanical aggregation of the metal masses that are placed in the helix; a modification that is sufficient to produce a considerable one in the properties of the induced currents themselves.

We have already remarked that, in magnetising bundles of iron wire by the electric current traversing the wire of a helix, we obtain much more powerful shocks than when we use a cylinder of solid iron. M. Dove, in order to establish an exact comparison between the physiological action and the galvanometric action, in respect to the influence that is exercised upon each of them by the nature and the state of the iron that is employed, made use of two perfectly similar helices made with thick copper wire, and traversed successively by the same voltaic current. These helices acted, by induction, upon a helix superposed on each of them. These new helices, made with a much finer wire, communicated together by one of their extremities; so that the direction of

the inductive current in the one was opposed to that of this current in the other. The two free extremities of the system of two helices were furnished with handles, which permitted of completing the circuit by the intervention of the body or by that of the galvanometer. In both cases, the two inductive currents being inverse, neutralised each other. It was no longer the same when wires or masses of iron were introduced into the interior of each helix. M. Dove, after having successively introduced into one of the helices cylinders made of forged iron, of different kinds of cast iron, of tempered steel, placed at the same time in the other the number of wires necessary for rendering null either the galvanometric or the physiological effect. He thus determined the number of wires necessary for establishing the compensation or the equilibrium between the induced current for the bundle of wires. This number was different according as the current was appreciated by its effect upon the galvanometer, or by its effect upon sensation. Thus, with forged iron, 110 wires were not sufficient to establish galvanometric equilibrium; 15 were sufficient to establish it for sensation; with tempered steel, 28 were required in the former case, 7 in the latter; with grey cold blast iron, 27 and 11. Analogous results were obtained with a differential galvanometer, the two wires of which were traversed each by one of the inductive currents. The latter having been rendered equal with regard to their action upon the galvanometer, they were not so with regard to physiological action; the superiority for this latter kind of effect belonged to the one that came from the helix in which were the iron wires. With regard to the different kinds of iron, experiment shows that, if we class them according to their galvanometric effect, we obtain a different series from that to which we arrive in classing them according to their physiological effect. Thus this latter effect depends, not only upon the discontinuity of the mass, but also upon the nature itself of the iron. Thus, by combining the two causes, we arrive at finding soft iron wires of a certain diameter capable of compensating the action of a cylinder of a certain kind of iron as well in relation to the galvanometer as in relation to sensation.