Davy observes, it was easy to imagine that such magnetic effects could not be exhibited by the electrical wire, without its being capable of permanently communicating them to steel; and that, in order to ascertain whether such was the fact, he fastened several steel needles, in different directions, to the uniting wire, when those parallel to it were found to act like the wire itself, while each of those placed across it acquired two poles. Such as were placed _under_ the wire, the positive end of the battery being east, had north poles on the south of the wire, and south poles to the north. The needles _above_ were in the opposite direction; and this was constantly the case, whatever might be the inclination of the needle to the wire. On breaking the connexion, the steel needles, placed _across_ the uniting wire, retained their magnetism,[74] while those placed _parallel_ to it lost it at the moment of disunion. The most extraordinary circumstances, however, connected with these experiments were, first, that _contact_ with the uniting wire was not found necessary for the production of the effect,--indeed, it was even produced, though thick glass intervened; and, secondly, that a needle which had been placed in a transverse direction to the wire, merely for an instant, was found as powerful a magnet as one that had been long in communication with it. [74] M. Arago also, nearly at the same time, succeeded in communicating magnetism to the needle; but, at the suggestion of M. Ampère, it was effected in a different manner. A copper wire, by being rolled round a solid rod, was twisted into a spiral, so as to form a _helix_. It was easy, by passing the wire round the rod, in one direction or the other, to form a _dextrorsal_ helix, proceeding from the right hand towards the left, as in the tendrils of many plants; or a _sinistrorsal_, or left helix, proceeding downwards from the left hand to the right above the axis. Into the cavity of a spiral thus formed, connecting the two poles of a battery, a steel needle wrapped in paper was introduced; and, in order to exclude all influence of the magnetism of the earth, the conchoidal part of the wire was kept constantly perpendicular to the magnetic meridian. In a few minutes the needle had acquired a sufficiently strong dose of magnetism; and the position of the north and south poles exactly agreed with M. Ampère's notion, that the electric current traverses the connecting wire in a direction from the zinc extremity of the pile to the copper extremity. The distance to which magnetism is communicated by electricity, and the fact of its taking place equally through conductors and non-conductors, are circumstances which, in the opinion of Davy, are unfavourable to the idea of the identity of electricity and magnetism.