Page:Scientific Memoirs, Vol. 2 (1841).djvu/22

10 currents generated by the same movement; but such a state can never occur to magnetism at rest.

There is probably also another state of equilibrium which may be attributed in part to magneto-electric currents. In the numerous experiments which I have made, I remarked, conformably to M. Ohm's formula, that the action of a voltaic current is much more constant when a rather long conducting wire is employed. I communicated some time back to Professor Moser of Königsberg, the idea, that under similar conditions the action of a pair ought to be of higher constancy, if long conducting wire be wound spirally round a soft iron bar, than when this same wire is extended or inverted on itself. The usual decrease in the energy of the battery would give rise to a magneto-electric current of the same direction as the voltaic current, and serving partly to re-establish this latter. I tried this delicate experiment, but have not yet been able to arrive at incontestable results. I shall recommence these researches, which appear to me to be of importance for the validity of the theory, and fertile in its practical applications.

The following experiment, which I owe in part to Professor Moser, appears to me to throw a great light on the intimate nature of magnetism. The helix of one arm of the horse-shoe (Art. 9) was brought into contact with a voltaic pair, while the helix of the other arm could be united to the multiplier of a very sensible galvanometer. This latter circuit being completed, there is, as is well known, a very energetic deviation of the needle, as soon as the voltaic circuit is completed or interrupted. But after the establishment of the voltaic circuit, if 4 to 5, or even 7 seconds, are allowed to elapse before completing the magneto-electric circuit, there will still be a sensible deviation of the needle. The same observation may be made, but in an inverse manner, if the voltaic circuit be interrupted some seconds previous to the establishment of the magneto-electric circuit. The deviation weakened in proportion to the length of time which has elapsed between the two operations. These effects are much more decided, if an armature be connected with the extremities of the magnet, and are probably still more so when a soft iron ring is employed.