If these views be sound, then it follows that no very considerable part of the mass of the atom can be accounted for by the mass of the outer negative corpuscles, because they are too few in number. If, therefore, we ad here to the view that mass is wholly electrical, it is highly probable that the mass of the atom resides mainly in the central positive corpuscles of the nucleus. As we now know the number of these, we may obtain a general estimate of their size (in accordance with the principles explained under in order that they may have the required mass. It is found, in this way, that the diameter of the positive cor puscle is only about one two-thousandth of the diameter of the negative corpuscle, and ai the diameter of the negative corpuscle is only about the fifty-thousandth of that of the atom, it is evident that the positive corpuscle is minute beyond conception. If an atom were magnified up to the size of the earth, the nega tive corpuscle would have a diameter of about 850 feet and the positive corpuscle would have a diameter of only about five inches.
It will be evident, from the data here given, that an atom is far from being a solid, im penetrable body. It is more like the solar sys tem in constitution—its bulk being mainly oc cupied by space, in which a few small cor puscles are circulating. This requires us to revise our old conception of the collision of atoms and molecules. They cannot often come together in actual physical contact, because the forces of attraction and repulsion that the com ponent corpuscles exert are exceedingly great, and atoms or molecules, as they approach con tact, must, therefore, be repelled apart with great violence by the mutual actions of their exterior negatively-electrified corpuscles; though it has not yet been made entirely clear how this tendency can be consistent with the facts of chemical combination. A particle possessing sufficient kinetic energy could penetrate an atom and pass through it without sensible perturba tion, unless it happened to strike the small cen tral nucleus; and we have photographic proof that the 'heavy, alpha-particles with which we have to deal in the study of radioactivity can pass through no • less than half a million atoms in succession, before being stopped or sensibly deflected.
The electrical theory of the constitution of matter is still confronted by certain serious fundamental difficulties, notwithstanding its numerous successes. One of the most formi dable of these difficulties is concerned with radiation. Electrical theory indicates, for ex ample, that an electron must radiate energy whenever its motion is changed, either in speed or in direction. An electron revolving with uniform speed in a circular orbit would, there fore, radiate energy continuously. Unless we are prepared to abandon the principle of the conservation of energy or can show that atoms are continuously supplied with energy from some external source, it is hard to escape from the conclusion that atoms are all trunning down)" In some cases, indeed, we have evi dence that this is true. The radioactive ele ments, for example, are undergoing gradual spontaneous decomposition, but that is appa rently due to the fact that the particular ar rangements of electrons that prevail in their atoms are not altogether stable. It is no part of our present problem to consider whether this indicates that after a time the radioactive elements will disappear forever, or whether there is some unrecognized agency at work, renewing them, so that the supply will be maintained substantially as at present. It is important for us to note, however, that most of the matter with which we com monly have to deal is composed of elements that show radioactivity either very slightly or not at all; and hence we must recog nize that the atoms of what may be called the elements are apparently stable, except as they are modified within certain limits by electrification, chemical combination or some other external influence.
In view of the difficulty just suggested, it has been assumed, by some of the leading physicists of the past few years, that the de duction that has been drawn from the theory of electricity, to the effect that an electron must radiate energy in consequence of a mere change in the direction of its motion, is (for some unknown reason) unsound and that radia tion occurs only when the speed of the electron changes. A certain amount of indirect evi dence in favor of this view is afforded by the phenomenon of permanent magnetism, which is not necessarily attended by the radiation of energy, although it is convenient to consider the magnetization to be due to the orientation of the orbits of some (or all) of the outer negative corpuscles in the atoms of the magnet. Apparently we here have a case in which the corpuscular revolutions continue indefinitely, without loss of energy by radiation.
Bohr has had marked success in accounting for radiation phenomena by assuming (1) there is no radiation from a corpuscle that is moving with uniform speed, even when the path it follows is curved; (2) that although the negative corpuscles in the outer parts of an atom normally revolve in circular orbits sug gestive of those followed by the planets in the solar system, yet they may (and do) pass some what abruptly from one of these orbits to an other; (3) that radiation from a corpuscle takes place only while the corpuscle is making a change of this kind in its orbit; and (4) that the possible stable orbits are not infinite in number, but that each stable orbit is char acterized by the possession, by the corpuscle following it, of a quantity of kinetic energy expressed by 1/2 Ihn, where I is any integer, h is Planck's constant (see RADIATION) and n is the number of orbital revolutions that the corpuscle makes per second. By means of his theory, Bohr is enabled to account for all the spectral lines of hydrogen with a wonder ful degree of precision. The same high degree of precision is not obtained when the theory is applied to other elements, though the agree ment is still close enough to be strongly sug gestive. In fairness to Bohr's theory we should remember that it was originally proposed in connection with the simple case in which the atom consists of a single negative corpuscle, revolving about a single free positive nucleus. That may very likely be the actual constitu tion of the hydrogen atom, and hence we should expect the most accurate correspondence with the facts in that case. When the atom contains a larger number of free positive and negative electrons, the fields of force that they produce will naturally overlap to a large extent and it is possible that some portion of the inaccuracy of the Bohr theory, when applied to elements other than hydrogen, may be due to this cause.
See also ATOMIC THEORY; BROWNIAN MOVE WENT; ELECTRON THEORY OR CORPUSCULAR THE ORY; GASES, KINETIC THEORY OF; RADIATION; RADIOACTIVITY; X-RAYS: ZEEMAN EFFECT. For a general popular review of the older form of the molecular theory, consult Risteen, 'Mole cules and the Molecular Theory of Matter.' For the beginnings of the electrical views, con sult also Thomson, J. J., 'Electricity and Mat ter); for later developments Comstock and Troland, 'The Nature of Matter and Electric ity'; Perrin, 'Les (now also avail able in an English translation) ; Millikan, 'The Electron' ; Stewart, A. W., 'Recent Advances in Physical and Inorganic Chemistry.'