which I have alluded in Art. 25., support this mode of considering the magnetic state of the inner part of concave surfaces. The electric helix, and this concave surface, are nearly in the same relation as two juxta-posed magnets, with poles of opposite name. There is no manifest effect; it is merely a disguised magnetic state.
34.
The experiments of M. Parrot, which he made only with a battery of small dimensions, led me to think at first that the development of the magnetism on the inner surface was only very weak, and that, on employing an arrangement sufficiently energetic, the magnetization would be shown in a more decided manner. This, however, was not the case; for the pile of 16 pairs (30), which sensibly heated the helix, had not the least effect. Notwithstanding this powerful current, there was still not a trace of magnetization in a soft iron wire of a millimeter in thickness, which had been connected to the external surface of a helix parallel to its axis. On detaching it, and introducing only one of its extremities into the interior, and to the centre of the helix, the wire became strongly magnetized. In this experiment, care must be taken not to submit the wire to the influence of the helix, after having completed the circuit; for the effects of induction which accompany the moments when contact is established or broken, affect the steel (37), and the soft iron made into wire, and which has partly taken the nature of steel.
35.
In electro-magnetic experiments, the question as to indefinite currents frequently occurs; an expression which, in certain cases, should not be employed without modification. In fact, this expression is inexact, for it is especially characteristic of a voltaic current to return to its origin, a condition which must never be lost sight of. Each closed circuit possesses an axis which bears the same relation to electric or magnetic elements, as the axis passing through the centre of gravity does to the mass of the body. These two axes will coincide in symmetrical and homogeneous circuits. On exposing soft iron to the inductive power of indefinite voltaic currents, we must take into account the position of the electrical axes, so as to be able to predict whether there will be an effect of magnetization or not. , , (Pl. I. fig. 2.), being any indefinite current, the bar cannot become
