When the compressibility of a liquid or gas is measured at very high pressures, the compression vessel has to be enclosed in a strong cylinder of metal, and thus it must be made, in some way, selfregistering. I first used indices, prevented from slipping by means of hairs. Sir W. Thomson's devices for sounding, at small depths, by the compression of air, in which he used various physical and chemical processes for recording purposes, led me to devise and employ a thin silver film which was washed off by a column of mercury. Much of my work connected with the Challenger Thermometers was done by the help of this process. Till quite recently I was unaware that it had been devised and employed by Cailletet in 1873, only that his films were of gold.