[Illustration: Fig. 204. A A A A. The rectangular wire covered with silk and varnished, one end of which being pointed, rests on the little cup B, connected with a covered wire passing down the centre of the brass support to the binding screw C let into ivory. D. The other extremity of the rectangular wire; this being covered and varnished, is not in metallic contact with the end B, but is likewise pointed, and dips into the mercury contained in the large cup E E. The upper and lower cups do not touch, and are separated by ivory, marked by the shaded portion, and the cup E E is in metallic communication with the brass pillar, and is connected with the negative pole of the battery at F, whilst C is connected with the positive pole of the battery, and the electricity circulates round the wire in the direction of the arrows. When a bar magnet, N, is brought towards the wire, the latter is immediately set in motion, and by alternately presenting the opposite poles of the magnet, the rectangular wire rotates freely round the cup B.] These curious movements of a magnetized needle, and rotations of wires and magnets, brought about by the agency of an active current of electricity, have induced Sir David Brewster to advance his admirable theory, which supposes the affection of the mariner's compass needle, and all other suspended pieces of steel, to be due to the agency of _electrical currents_ continually _circulating_ around the globe; and Mr. Barlow contrived the following experiment in illustration of Brewster's theory. A wooden globe, sixteen inches in diameter, was made hollow, for the purpose of reducing its weight, and while still in the lathe, grooves one-eighth of an inch deep and broad were cut to represent an equator, and parallels of latitude at every four and a half degrees each way from the equator to the poles. A groove of double depth was also cut like a meridian from pole to pole, but only half round. The grooves were cut to receive the copper wire covered with silk, and the laying on was commenced by taking the middle of a length of ninety feet of wire one-sixteenth of an inch in diameter, which was applied to the equatorial groove so as to meet in the transverse meridian; it was then made to pass round this parallel, returned again along the meridian to the next parallel, and then passed round this again, and so on, till the wire was thus led in continuation from pole [Page 214] to pole. The length of wire still remaining at each pole was returned from each pole along the meridian groove to the equator, and at this point, each wire being fastened down with small staples, the wires from the remaining five feet were bound together near their common extremity, when they opened to form separate connexions for the poles of a voltaic battery. When the battery was connected, and magnetic needles placed in different positions, they behaved precisely as they would do on the surface of the