Other methods, based upon the action of ultraviolet light, and upon radioactivity and phenomena of various other kinds, have been used for determining this ratio, and from the general agreement among the results obtained by different methods and different experiment ers it has become evident that although the velocity with which the corpuscles move de pends upon the circumstances under which they are liberated or set in motion, the electric charge of a corpuscle per unit of its mass,l, is always the same, no matter what the condi tion of the corpuscle is, or from what source it is obtained. It is evidence of this kind that has led physicists to conclude that the negative corpuscle is a fundamental and omnipresent constituent of matter of every kind. The best value of — that has been obtained up to the present time is certainly Bucherer's. He found m ==1:767 X if m is measured in grammes and e in absolute electromagnetic units. This is believed to be correct to within about one-half of 1 per cent. It applies only to slowly-moving corpuscles, however, because, as we shall presently see, the apparent mass of a corpuscle increases with the speed of the corpuscle, while the charge presumably remains unchanged. (Bucherer's value of — is 5.299 X 10", if the electric charge ?X is expressed in absolute electrostatic units.) The fact that the ratio — is nearly times as great as the charge per unit mass observed in connection with the hydrogen atom in electrolysis shows, most conclusively, that one of two things must be true: Either (1) the charge on the negative corpuscle is far greater than the charge accompanying an ion in electrolysis, or (2) the mass of a nega tive corpuscle is far less than the mass of any ion or atom previously known to us. Of course these may both be true, but certainly one of them is true, and as soon as this fact was recognized, it was also recognized that the discovery of the negative corpuscle was an event of fundamental importance in the history of physics.
In order to find out which alternative must be adopted, J. J. Thomson undertook to de termine the electric charge on a single corpus cle — and hence also the mass of the corpuscle, since the ratio of the two was known. More accurate values of these quantities have since been obtained by other means, but Thomson, it should be remembered, was a pioneer in a new field, and the work that he did in solving his problem has justifiably been called by Sir Oliver Lodge °one of the most brilliant things recently done in experimental physics." We can only outline his method in a rough way. It depends (1) on the fact, discovered by Aitken in 1880, that condensation of aqueous vapor in air does not occur, even when the air is supersaturated, unless there are nuclei of some sort for the mist-particles to form about; (2) on the fact, demonstrated by Lord Kelvin in 1870, that the surface tension of small droplets of water, suspended in the air, tends to cause evaporation even though the degree of saturation is enough to cause con densation on a water-surface that is flat, or that has a large radius of curvature; (3) on the fact, announced by J. J. Thomson himself
in 1888, • that the electrification of such a drop let tends to neutralize the effect of the surface tension, so that condensation can take place on a water droplet, or on any other curved sur face of exceedingly short radius, if this droplet or surface is electrified, even though no such condensation could take place in the absence of the electrification; (4) on the investigation, by Sir George Stokes in 1849, of the limiting speed at which small spherical bodies will fall, by their own weight, through a fluid of known viscosity; and (5) on the method devised by Mr. C. T. R. Wilson, in 1887, for by adiabatic expansion, a definitely-known quantity of aqueous vapor in the form of mist, from saturated air.
Thomson's experiment consisted (1) in partially ionizing, in a closed vessel and by means of X-rays or ultra-violet light, air con taining a suitable quantity of water vapor; (2) in causing the deposition of droplets of mist, by Wilson's method of quick adiabatic expansion, about the ions thus set free; (3) in observing the rate at which the mist thus formed subsides — a process which really con sists in the falling of the individual droplets through the air; (4) in calculating, by means of Stokes' formula, the diameter (and subse quently the weight) of the spherical droplets constituting the mist— this being made possible by the fact that he knew the viscosity of the air and had observed the rate of fall of the droplets; (5) in calculating the total mass (or weight) of water precipitated, in accordance with Wilson's method, from the known degree of expansion of the air; and (6) in dividing the total weight of precipitated water by the weight of a single droplet, and thereby determining the number of droplets. The number of droplets produced being assumed to be the same as the number of available ions about which condensa tion was theoretically possible, the experiment manifestly gave the total number, N, of the ions present in each cubic centimeter of the air, under the conditions prevailing in the ex perimental apparatus. In one experiment N was found to be 30,000.
The total aggregate charge of the ions was determined by means of a pair of parallel metallic plates in the vessel in which the mist was produced— one of them being insulated and connected with an electrometer. If the space between the plates contained positive ions, for example, then by suddenly Ltmmunicating a strong positive charge to the non-insulated plate these ions could be quickly repelled against the insulated plate, to which they would give up their charges; and the aggregate charge that they were carrying could then be measured by the electrometer. By means of this princi ple the total charge on the ions in a cubic centi meter of the air in the mist-chamber was de termined; and by dividing this total charge by N the number of ions in a cubic centimeter of the air, the charge on one individual ion became known.