104 



NATURE 



[March 24, 192 1 



Letters to the Editor. 



[The Editor does not hold himself responsible for opinions 

 expressed by his correspondents. Neither can he undertake 

 to return, or to correspond with the writers of, rejected 

 manuscripts intended for this or any other part of Nature. 

 No notice is taken of anonymous communications.^ 



Atomic Structure. 



In a letter to Nature of November 25 last Dr. 

 Norman Campbell discusses the problem of the 

 possible consistency of the assumptions about the 

 motion and arrangement of electrons in the atom 

 underlying the interpretation of the series spectra of 

 the elements based on the application of the quantum 

 theory to the nuclear theory of atomic structure, and 

 the apparently widely different assumptions which 

 have been introduced in various recent attempts to 

 develop a theory of atomic constitution capable of 

 accounting for other physical and chemical properties 

 of the elements. Dr. Campbell puts forward the 

 interesting suggestion that the apparent inconsistency 

 under consideration may not be real, but rather 

 appear as a consequence of the formal character of 

 the principles of the quantum theory, which might 

 involve that the pictures of atomic constitution used 

 in explanations of different phenomena may have a 

 totally different aspect, and nevertheless refer to the 

 same reality. In this connection he directs attention 

 especially to the so-called "principle of correspond- 

 ence," by the establishment of which it has been 

 possible — notwithstanding the fundamental difference 

 between the ordinary theory of electromagnetic radia- 

 tion and the ideas of the quantum theory — to com- 

 plete certain deductions based on the quantum theory 

 by other deductions based on the classical theory of 

 radiation. 



In so far as it must be confessed that we do not 

 possess a complete theory which enables us to 

 describe in detail the mechanism of emission and 

 absorption of radiation by atomic systems, I naturally 

 agree that the principle of correspondence, like all 

 other notions of the quantum theory, is of a some- 

 what formal character. But, on the other hand, the 

 fact that it has been possible to establish an intimate 

 connection between the spectrum emitted by an 

 atomic system^ — deduced according to the quantum 

 theory on the assumption of a certain type of motion 

 of the particles of the atom — and the constitution of 

 the radiation, which, according to the ordinary theory 

 of electromagnetism, would result from the same 

 type of motion, appears to me to afford an argument 

 in favour of the reality of the assumptions of the 

 spectral theory of a kind scarcely compatible with 

 Dr. Campbell's suggestion. On the contrary, if we 

 admit the soundness of the quantum theory of 

 spectra, the principle of correspondence would seem 

 to afford perhaps the strongest inducement to seek 

 an interpretation of the other physical and chemical 

 properties of the elements on the same lines as the 

 interpretation of their series spectra ; and in this 

 letter I should like briefly to indicate how it seems 

 possible by an extended use of this principle to 

 overcome certain fundamental difficulties hitherto 

 involved in the attempts to develop a general theory 

 of atomic constitution based on the application of the 

 quantum theory to the nucleus atom. 



The common character of theories of atomic con- 

 stitution has been the endeavour to find configura- 

 tions and motions of the electrons which would seem 

 to offer an interpretation of the variations of the 

 chemical properties of the elements with the atomic 

 number as they are so clearlv exhibited in the well- 

 known periodic law. A consideration of this law 



NO. 2682, VOL. 107] 



leads directly to the view that the electrons in the 

 atom are arranged in distinctly separate groups, each 

 containing a number of electrons equal to one of the 

 periods in the sequence of the elements, arranged 

 according to increasing atomic number. In the hrst 

 attempts to obtain a definite picture of the configura- 

 tion and motion of the electrons in these groups it 

 was assumed that the electrons within each group at 

 any moment were placed at equal angular intervals 

 on a circular orbit with the nucleus at the centre^ 

 while in later theories this simple assumption has 

 been replaced by the assumptions that the configura- 

 tions of electrons within the various groups do not 

 possess such simple axial symmetry, but exhibit a 

 higher degree of symmetry in space, it being 

 assumed, for instance, that the configuration of 

 the electrons at any moment during their motions 

 possesses polyhedral symmetry. All such theories 

 involve, however, the fundamental difficulty that no 

 interpretation is given why these configurations 

 actually appear during the formation of the atom 

 through a process of binding of the electrons by the 

 nucleus, and why the constitution of the atom is 

 essentially stable in the sense that the original con- 

 figuration is reorganised if it be temporarily dis- 

 turbed by external agencies. If we reckon with 

 no other forces between the particles except the 

 attraction and repulsion due to their electric charges, 

 such an interpretation claims clearly that there must 

 exist an intimate interaction or "coupling" between 

 the various groups of electrons in the atom which is 

 essentially different from that which might be ex- 

 pected if the electrons in different groups are assumed 

 to move in orbits quite outside each other in such a 

 way that each group may be said to form a "shell " 

 of the atom, the effect of which on the constitution 

 of the outer shells would arise mainly from the com- 

 pensation of a part of the attraction from the nucleus 

 due to the charge of the electrons. 



These considerations are seen to refer to essential 

 features of the nucleus atom, and so far to have no 

 special relation to the character of the quantum 

 theory, which was originally introduced in atomic 

 problems in the hope of obtaining a rational inter- 

 pretation of the stability of the atom. According to 

 this theory an atomic system possesses a number of 

 distinctive states, the so-called " stationarv states," in 

 which the motion can be described by ordinary 

 mechanics, and in which the atom can exist, at any 

 rate for a time, without emission of energy radiation. 

 The characteristic radiation from the atom is emitted 

 onlv during a transition between two such states, and 

 this process of transition cannot be described by 

 ordinarv mechanics, any more than the character of 

 the emitted radiation can be calculated from ^ the 

 motion bv the ordinary theory of electro-magnetism, 

 it being, in strikinf^ contrast to this theory, assumed 

 that the transition is always followed by an emission 

 of monochromatic radiation the frequency of which 

 is determined simply from the difference of energy 

 in the two states. The application of the quantum 

 theorv to atomic problems — which took its starting 

 point" from the interpretation of the simple spectrum 

 of hydrogen, for which no a priori fixation of the 

 stationary states of the atoms was needed — has in 

 recent vears been largelv extended by the development 

 of svstematic methods for fi^fing the stationary states 

 corresponding to certain general classes of mechanical 

 motions. While in this way a detailed interpretation 

 of spectroscopic results of a very different kind has 

 been obtained, so far as phenomena which depend 

 essentially on the motion of one electron^ in the 

 atom were concerned, no definite elucidation has 

 been obtained with regard to the constitution of 



