But such an instrument is not delicate enough for the most refined experiments, and the improved form generally used is due to that prince of inventors, the late Lord Kelvin. He originally devised it, it is interesting to note, not for laboratory experiments, but for practical use as a telegraph instrument in connection with the early Atlantic cables. Before describing it, it may sharpen the reader's interest to mention a wonderful experiment which was made by Varley, the famous electrician, on the first successful Atlantic cable. He formed a minute battery of a brass gun-cap, with a scrap of zinc and a single drop of acidulated water. This he connected up to the cable. Probably there is not one reader of this book but would have thought, if he had been present, that the man was mad. What conceivable good could come of connecting such a feeble source of electrical pressure to the two thousand miles of wire spanning the great ocean; the very idea seems fantastic in the extreme. Yet that tiny battery was able to make its power felt even over that great distance, for the Thomson Mirror Galvanometer was there to detect it. Two thousand miles away, the galvanometer felt and was operated by the force generated in a battery about the size of one of the capital letters on this page. This wonderful instrument consisted of a magnet made of a small fragment of watch-spring, suspended in a horizontal position by means of a thread of fine silk, close to a coil of fine wire. When current flowed through the coil the magnetic field caused the watch-spring magnet to swing round, but when the current ceased the untwisting of the silk brought it back to its original position again. So far it seems to differ very little from the ordinary galvanometer previously mentioned, but the stroke of genius was in the method of reading it. With a small current the movement of the magnet was too small to be observed by the unaided eye, so it was attached to a minute mirror made of one of those little circles of glass used for covering microscope slides, silvered on the back. The magnet was cemented to the back of this, yet both were so small that together their weight was supported by a single thread of cocoon silk. Light from a lamp was made to fall upon this mirror, thereby throwing a spot of light upon a distant screen. Thus the slightest movement of the magnet was magnified into a considerable movement of the spot of light. The beam of light from the mirror to the screen became, in fact, a long lever or pointer, without weight and without friction.