be seen that there exists a distribution of magnetism in solid cylinders entirely distinct from that in hollow cylinders, so that not only the mass, but the geometrical form also, must be taken into account.
28.
The magneto-electric effects which take place on completing or breaking a voltaic circuit, are, as is well known, much more decided when the connecting wire is bent into a helix, and still more so, if this latter contains a soft iron nucleus. Whether this nucleus was a solid or hollow tube, there was no perceptible difference in the discharge or splendour of the spark of disjunction, even when the wire of the helix 1½ line in thickness, or a second coil of brass wire of 0·5 line diameter was employed. These experiments were performed in the dark, in order to be able to judge well of the splendour of the spark; after each experiment, the surface of the mercury, and the electrode (réophore) inserted in it, were carefully cleaned. Moreover, I ascertained the action of the pair, that constructed of zinc and silver, by the interposition of a galvanometer which only indicated a variation of from 35½ to 35, and from 24½ to 24, occurring during the course of the experiments. It is hardly necessary to add, that the spark was scarcely perceptible as soon as the nucleus had been removed.
I made various combinations, which it is unnecessary to describe more in detail, to produce a magneto-electric current by the aid of those tubes, sometimes by employing them as armatures of a horse-shoe, sometimes as the nuclei of a magneto-electric helix. The force of the current was indicated by the deviations of the needle of an astatic galvanometer of Nobili. Although these experiments were sufficiently delicate, considering that the least accidents in the position of the armature exert a certain influence on the deviation of the needle, yet I did not find any difference so notable as to be obliged to attribute it to some influence of the mass of the electro-motive cylinders. The law, therefore, of Mr. Barlow may also be admitted with respect to magneto-electric actions. But repeated experiments are necessary to learn whether there are not conditions analogous to those of Art. 27, and relative to a highly developed magnetic state. I may be allowed to add that, had there been any notable difference in these experiments, it might justly have been concluded that the magneto-electric current, or the extra-current
