We are acquainted with the striking and perfect analogy existing between magnets and electro-dynamic cylinders. There is a very beautiful experiment performed by Dr. Roget, and which might serve as a confirmation to these arguments; I relate it in his own words: "It occured to me, soon after hearing of Ampère's discovery of the attraction of electrical currents, that it might be possible to render the attraction between the successive turns of a heliacal coil very sensible, if the wires were sufficiently flexible and elastic; and, with the assistance of Mr. Faraday, this conjecture was put to the test of experiment, in the laboratory of the Royal Institution. A slender harpsichord wire, bent into a helix, being placed in a voltaic current, instantly shortened itself whenever the electrical current was sent through it, but recovered its former dimensions the moment the current was intermitted." I hope soon to be able to communicate to the Academy the result of the experiments I shall institute on this subject, conjointly with its illustrious member M. Struve, who has promised me his assistance, these experiments requiring the exactitude and delicacy of micrometrical observations. When we consider the electro-dynamic cylinder, its total effect increases by the reciprocal attraction of separate coils, whose action becomes less oblique. The same is the case with soft iron, the magnetism of which increases up to a certain limit, by the action of contraction. Heat everywhere presents itself as the enemy of magnetism; perhaps it is because the two forces encounter each other in opposed molecular actions. For the present I abstain from following up these discussions, and all connected with them.
32.
Although the remarkable effects occurring at the instant when voltaic contact is completed or interrupted, bear a striking analogy to the actions of magneto-electric currents, there exists, however, a marked difference. If a conducting wire, covered with silk, be bent back on itself, so that the direction of the current is opposed in the adjacent parts, (Faraday, 9th Series, Art. 1096) there is neither discharge nor spark at the moment of the disjunction, the opposite currents of induction arranging themselves in equilibrium to annul these effects. If we remove the adjacent parts, the discharge and spark take place. This experiment made me think that it was also possible to destroy the effects of a magneto-electric current, by
