RECOLLECTIONS AND REFLECTIONS
produced in ordinary air, saturated with water vapour, when it is cooled by a very slight expansion of its volume, no fog is produced even by very considerable expansions when the dust is filtered out of the air. The drops of water require nuclei on which they can be deposited : in ordinary-air these are supplied by the particles of dust which it contains. Wilson found that when the dust-free gas which would not give a fog was ionised, and therefore contained both negatively and positively electrified particles, a fog was produced by an expansion too small to produce one in dust-free air, and that a fog would be produced on negative particles by an expansion less than that required for positive ones. Thus with certain expansions all the drops will be negatively electrified.
Thus if the gas is ionised, say by radiation from radium, it will contain negatively electrified particles ; drops of water will be deposited on these by a supersaturation which is not sufficient to produce any drops when the particles are not electrified, or electrified only with positive electricity. This discovery has been developed into a method which has had a profound influence on the recent study of atomic physics.
If we supersaturate the gas by cooling it by a sudden expansion of known amount, we can calculate the lowering of temperature, and hence the difference in the amount of water required to saturate the air before and after expansion. This will be the amount of water deposited as drops on the negatively charged particle. These drops will fall down slowly under gravity. They will in consequence of the viscosity of the air fall with uniform velocity. This velocity depends on the size of the drops, which can be determined by Townsend's method of measuring the rate at which they fall. We know the total quantity of 342