Another po'nt of resemblance between a-rays and canal rays has recently been established, namely that the a-rays can suffer changes of charge during their course. Along with the helium atoms of double positive charge a smaller number of helium atoms with only one positive charge can he observed, and a very few neutral atoms, but no negatively charged atoms. The smaller the velocity of the a-rays the greater the proportion of singly charged atoms ; the neutral atoms are found only at very low velocities.
Another class of mass rays is constituted by the very much slower atoms obtained by evaporating a molten metal in a vacuum so high that the mean free path of the atoms leaving the metal surface is much greater than the dimensions of the vessel. By means of a suitable system of diaphragms pierced with slits or holes it can be arranged that the flying atoms take the form of a beam. If the temperature of the glass wall opposite the metal surface is suitable the beam of atoms forms a thin deposit where it strikes, but it must be observed that the necessary temperature is far below the solidification point of the metal. In cases where the deposit is too thin to be seen it can be "developed" by de positing silver on it from a silver nitrate solution, by a special technique. The velocity of such atomic rays, as they are termed, can be measured by rotating the vessel at a known speed and observing the consequent displacement of the line deposit made by a flat beam. The velocity found by this method agrees within experimental error with that to be anticipated on the kinetic theory of gases from the temperature of the molten metal, and this constitutes the first direct measurement of the velocity of atoms due to heat agitation. The velocity of a beam of silver
atoms derived from molten silver at about absolute is 6X cm./sec., as compared with about cm./sec. for the atoms in canal rays and 2X cm./sec. for a-particles.
The atomic rays have found an important application in the experiments of Stern and Gerlach on the magnetic moment of atoms, described in the article MAGNETISM.
Finally, a word may be said on corpuscular radiations reaching the earth across space. The aurora borealis is attributed to the action of rays from the sun on the rarefied gases of the upper atmosphere at heights of ioo kilometres and more. The magnetic action of the earth on the charged particles constituting these rays concentrates the action in the polar regions. The data available do not suffice to establish with certainty whether these radiations consist of electrons, and so resemble 3- and cathode rays, or whether they consist of positively charged atoms, and so resemble the mass rays obtained in discharge tubes and a-rays. Vegard states that positive rays of hydrogen and helium with velocities of the order of cm./sec. may produce an essential part of the aurora, but at the same time asserts that swift elec trons combine a sufficiently great magnetic deflectibility and penetrating power to explain the distribution of luminosity and the height at which the aurora is produced. The corpuscular cosmic rays, influenced by the earth's magnetic field, must be carefully distinguished from the penetrating electromagnetic cos mic rays, discussed in the first part of this article.
"Joining the Infra-red and Electric Wave Spectra," Proceedings National Academy of Sciences, 9, 211, 1923. (E. N. DA C. A.)