made evident. It has been pointed out that the best thought of political science requires that authority and responsibility must be combined in administration; that departments must be functionalized, that is, that there must be a grouping of related activities into single sections, divisions, and departments according to the functions performed; that in charge of each such units of administration there must in most cases be a single executive with complete authority over it and full responsibility to the next highest executive, the increasing degree of authority and responsibility culminating in the supreme executive or governor, whose power is made complete by virtue of his control of the tenure of all subordinates; and that the process of consolidation must be carried far enough so that for consultative, advisory, and co-ordinative purposes, the heads of the primary departments of administration may constitute a small compact group or cabinet. When the state administration has been reorganized according to these principles, the problem of effecting co-ordination for the conscious recognition and encouragement of scientific research will not be difficult to solve. Within each main department the work will be so ordered and arranged that the research function will find its proper place, and the department will be of sufficient size to support adequately a properly appointed research laboratory. Moreover, concerning the state administration as a whole, it can be said that while the process of consolidation demands the placing of executive power in a few hands, it does not preclude, but on the contrary makes more easy, the introduction of the advisory function. The purpose of reducing the state administration to a small number of primary departments is to make possible the creation of a body for giving advice to the chief executive and providing consultation and co-operation among its members. No sound principle is violated if for certain other purposes the heads of departments or some of them constitute councils of various sorts for special consultation, or if additional members be introduced upon such councils for the purpose of giving representation to lay or expert interests in an advisory capacity. It will be recalled that such a council for the representation of the labor interests of the government was included in the recently proposed plan of reorganization in California. 2. ORGANIZATION OF A STATE SCIENTIFIC COUNCIL Granted the creation of a departmentalized state administration, the ideal plan for the organization of scientific research within the state government in the light of the above conclusion would seem to be the creation of a Scientific Council or Council on Scientific Research which should be composed first, of the heads of those departments in which functions of a scientific nature are or might be carried on; second, representatives of the great academic institutions and of the various scientific societies within the state; third, representatives of the great industrial enterprises within the state which would expect to benefit from the scientific conduct of the state departments and which in turn could aid such work by their intelligent support. While the Council should necessarily require the service of an expert bureau for the administration of its functions it should not constitute a central research laboratory but should rely upon the separate research divisions of the various departments for the actual solution of scientific problems. The advantage of such a form of oganization would be that while. it would imply a conscious recognition of the field of science, it would not in any way produce a rigidity of system and formality of procedure. Leaving freedom of action to departments to work out their own problems, it would definitely co-ordinate the efforts of all in a single direction. The Council would serve as a great clearing house between departments, between the state and councils in other states, and between the state and the nation. It would pool facilities and avoid duplication. It would place the research equipment of the whole country at the disposal of a single state department. Its mere existence would be an earnest of the vital importance of the scientific method in the whole field of human relations. OF THE NATIONAL RESEARCH COUNCIL Vol. 2, Part 6 JULY, 1921 'Number 14 A GENERAL SURVEY OF THE PRESENT STATUS OF THE ATOMIC STRUCTURE PROBLEM Report of the Committee on Atomic Structure of the *This committee of the Division of Physical Sciences of the National Research Council consists of the following members: F. A. Saunders, Chairman, P. W. Bridgman, Irving Langmuir, G. N. Lewis, R. A. Millikan, Leigh Page, D. L. Webster. The late Professor H. A. Bumstead was chairman of the committee as originally constituted and much of the work preliminary to the preparation of the report was done under his leader PART I THE PRESENT CONCEPTION OF ATOMIC STUCTURE BY DAVID L. Webster PROBLEMS And MethodS A hundred years ago, when Dalton first used the idea of atoms, there was no question of atomic structure. The very name "atom" meant "indivisible," and an atom was supposed to be the last stage in the division of matter. At that time there was no way of knowing how small an atom was. Since then, it has been found to be about ten thousand times as small as the smallest particle whose structure we can see at all, even with a microscope. Nevertheless it appears that the atom has a structure, and not only that, but that along with the problem of atomic structure we now have the problem of atomic dynamics as well. These problems date, practically, from the discovery of the electron. Maxwell,' in 1869, had explained the dispersion of light in a prism by the forced vibrations of minute electrified particles in the material of the prism. And in 1897, Sir J. J. Thomson2 proved that cathode rays were indeed just such electrified particles as Maxwell had spoken of, and that they were smaller and lighter than any atoms. Then it appeared, from the study of cathode rays, that these minute particles were all exactly alike, no matter what sort of material they were extracted from, and that the mass of each one of them was only about one two thousandth part of that of a hydrogen atom. So they were given a new name, "corpuscles," or "electrons," and they were assumed to be the most important units of which all atoms are built up,-not the only units, however. For the electrons are always charged negatively, and an atom is often as a whole electrostatically neutral, or even positive. So there must be positive electricity in it also. But the only positively charged particles that have ever been isolated, as electrons have, are positively charged ions, and they may be identified by their much greater masses as practically whole atoms. So the electrons, or negative corpuscles, being so light and mobile, are more in evidence in most phenomena than the positive electricity. Thus the first problem of atomic structure is how many of these electrons there are in any atom, and the next is what sort of positions they occupy. 1J. C. Maxwell, Math. Tripos Exam., 1869. On the number of electrons, we have, fortunately, some reliable information in the fact predicted by Thomson, that the scattering of X-rays is done primarily by individual electrons, without cooperation. Thus the number of electrons per atom may be counted by measuring the scattering power per atom. This has been done by Barkla,' who found that, for light elements at least, the number of electrons is probably equal to the ordinal number of the element in Mendeleyeff's table, thus: Hydrogen has 1, helium 2, lithium 3, carbon 6, fluorine 9, neon 10, sodium 11, chlorine 17, argon 18, potassium 19, etc. This important discovery is confirmed by all subsequent work, especially by Moseley's' discovery of the great importance of this "atomic number" as the determining factor in X-ray spectra. With this point established we can attack the next important problem, that of the positions of the electrons. Since we must get our information indirectly, by inference from observed phenomena, the question arises, what phenomena are the most ⚫ reliable guides in this matter? All the phenomena we can observe depend on the structure of the atoms concerned in them, and because the electrons are light and mobile, the presumption is that any phenomenon we choose is likely to be due to them. Practically, however, most of our evidence to date has been derived from two classes of data, those of chemical statics, and those of spectroscopy. The theory finally accepted must of course fit both these classes of data and many others, but strange to say, the results obtained from these two classes to date have been very discordant. Thus the first problem is to consider the reliability of these classes of evidence, not as to details of the final model, but for a mere rough outline of the structure. For this purpose data of the chemical type have several distinct advantages: First, the chemical data relate to phenomena involving a few electrons at a time and give some evidence on all of them, while the spectroscopic data deal primarily with only one electron at a time, but always involve essential and complex corrections due to all the rest. This makes the chemical data give a more satisfactory picture of the relations between electrons. Second, the chemical data are much more numerous than the spectroscopic, and the chemical properties of the elements differ in quality as well as quantity, and so their significance is much more readily grasped by our minds, although they may actually be no more important in the final analysis; thus the chemical data are to the atom what a photograph is to a man, while the spectroscopic data are its Bertillon measurements; neither class alone gives a complete identification, but the photograph conveys an impression of his appearance more readily than the measurements. 1C. G. Barkla, Phil. Mag., 21, 648, 1911. H. J. G. Moseley, Phil. Mag., 27, 703, 1914. |