After the atom of negative electricity had been isolated, in the form of cathode rays, the next suggestion was that corresponding positive electric particles might be discharged from the anode in a vacuum tube. By special methods success has been attained in showing and studying rays of positive particles. In order to separate them from the negative cathode ray particles the German scientist, Goldstein, let the positive particles pass through canals in the cathode; they are therefore called _canal rays_. The velocity of the particles is much less than that of the cathode rays, and the ratio _e_/_m_ between charge and mass is much smaller and varies according to the gas in the tube. In experiments where the tube contains hydrogen, rays are always found for which _e_/_m_, as in electrolysis, is about ¹/₂₀₀₀ of the ratio in the cathode rays. Therefore there can be scarcely any doubt that these canal rays are made up of charged hydrogen atoms or hydrogen ions. The values found with other gases indicate that the particles are atoms (or molecules sometimes) of the elements in question, with charges one or more times the elementary quantum of electricity (4·77 × 10⁻¹⁰ electrostatic units). Research in this field has also been due in particular to J. J. Thomson. From his results, as well as from those obtained by other methods, it follows that positive electricity, unlike negative, cannot appear of its own accord, but is inextricably connected to the atoms of the elements. The Nature of Electricity. The earlier conceptions of a one or two-fluid explanation of the phenomena of electricity appear now in a new light. We are led to think of a neutral atom as consisting of one mass charged with positive electricity together with as many electrons negatively charged as are sufficient to neutralize the positive. If the atom loses one, two or three electrons, it becomes positive with a charge of one, two or three elementary quanta of electricity, or for the sake of simplicity and brevity we say that the atom has one, two or three “charges.” If, on the other hand, it takes up one, two or three extra electrons it has one, two or three negative charges. Fig. 18 can give help in understanding these ideas, but it must not be thought that the electrons are arranged in the way indicated. The substances, which appear as electropositive in electrolysis—_i.e._ hydrogen and metals—should then be such that their atoms easily lose one or more electrons, while the electronegative elements should, on the other hand, easily take up extra electrons. Elements should be monovalent or divalent according as their atoms are apt to lose or to take up one or two electrons. From investigations with the vacuum tube it appears, however, that the atoms of the same element can in this respect behave in more ways than would be expected from electrolysis or chemical valence. [Illustration: FIG. 18.—Provisional representation (according to the electron theory) of