Among Thomson's early discoveries was the fact that it was good for students to do laboratory work. So almost immediately he was appointed the professor of physics, at Glasgow, in 1846, at the age of twenty-two, he "organized a laboratory corps from volunteer students," and about 1849 he "established an incipient laboratory in the wine cellar of an old professor's house," so his successor, Professor A. Gray, tells me. In my time Thomson's laboratory consisted of one room and the adjoining coal cellar, the latter being the birthplace of the syphon-recorder. To avoid friction with the capillary glass syphon, which was moved to the right or left by the electric signals coming through the submarine cable, the end of the syphon was not allowed to touch the strip of paper, but a continuous stream of ink spurted out of the syphon on to this paper in consequence of the reservoir of ink, into which the other end of the syphon dipped, being kept highly electrified.
To find out what sort of electrical machine should be used for this purpose Thomson suggested that we should measure the efficiency of frictional electric machines. We did so, and brought him the result—viz., efficiency equals some small fraction of unity. He replied, "I can not degrade a man by asking him to use his energy so wastefully; I must design something better." And he did—viz., the influence machine; then when, by carrying out his suggestions, a fellow student and I had constructed an influence machine and got it to work, he sent us to the Glasgow Patent Office to see whether any one had thought of that principle before. And we found Varley's and other anticipations, with, however, this difference—that the earlier patentees proposed giving to the arrangement an initial charge to start its action, whereas Thomson's was a machine that worked on the compound interest law, starting with an infinitely small initial capital. This led not only to the "mouse mill" and the "replenisher," but to the class working all kinds of problems on investments at compound interest. "Now, suppose the interest is one one-thousandth per cent., paid every one one-hundredth of a second, etc.," and we who had never invested any money in our lives, indeed, possessed no money to invest, might have been mistaken for budding pupils of a stock broker had any visitor chanced to come into the lecture room.
There was no special apparatus for students' use in the laboratory,