



























APRIL 21, 1923] 
only those combinations of terms will normally give 
rise to lines for which & changes by +1. This is 
precisely what is found to occur, and the puzzling 
incompleteness of Ritz’s combination principle is 
entirely accounted for. There is, moreover, no real 
element of arbitrariness in this explanation, for the 
variation of the effective nuclear charge is of course 
due to the already bound electrons. These must lie 
in permitted orbits of definite quantum numbers which 
in with those of the spectral terms for an approxi- 
mately central field of force, not very greatly modified 
The numeri- 
cal requirements of the theory can be satisfactorily met, 
of the theory as here expounded by Bohr, but it does 
xt, I think, misrepresent it. 
spectral facts—the spectra of ionised atoms, in 
lar those of the alkaline earths, with their Ryd- 
erg constant 4N, which results naturally from the 
ns, which arise from the deviation of the atomic field 
from central symmetry leading to the introduction of a 
d quantum number; finally, the unique nature of 
e helium spectrum with its absence of inter-combina- 
s, of which an explanation in terms of a further 
alisation of the correspondence principle may at 
ast be said to be in sight. 
- These are mainly facts of which the explanation is 
Still under development, but three further complete 
‘Successes of the theory must also be recorded. The 
idea of stationary states accounts completely for the 
differences between emission and normal absorption 
: An atom in its normal state can absorb only 
alkali metal this means the principal series of doublets 
-only—for an alkaline earth the principal series of 
‘singlets. In other cases, such as the aluminium sub- 
| group, the theory leads us to expect that the normal 
state will correspond to the first principal term and that 
. the absorption spectrum will be the sharp and diffuse 
series—an expectation recently confirmed by direct 
experiment. Secondly, the phenomena of resonance 
_ spectra are fully accounted for. Thirdly, the theory 
__ assigns definite energies to the various atomic states, and 
_ this assignment can be tested directly by the study of 
electronic impacts. This is by itself a complete branch 
_ of modern physics directly inspired by the theory, 
_ which it as directly and completely confirms. 
NO. 2790, VOL. 111] 
q 
NATURE 
375 
The third essay, “‘ The Structure of the Atom and the 
Physical and Chemical Properties of the Elements ” 
(October 1921), is the most novel and important of the 
three. It differs from the others in being slightly 
revised in translation to bring it up-to-date (May 1922). 
This essay brings the whole of the available evidence— 
X-ray, chemical, optical—to bear on the specific 
question of the structure of the atom; that is, the way 
in which the planetary electrons are arranged. We 
have already seen that this is implicitly discussed, and 
a definite view reached, in connexion with the theory 
of series spectra. Other evidence merely confirms and 
crystallises this view. The goal to be attained is the 
theoretical deduction, from the principles of the 
quantum theory properly formulated, of the periodic 
table of the chemical elements, and all other atomic 
properties. Bohr shows that the fundamental process 
which must be considered is the successive binding of 
electrons one after another by a nucleus originally 
naked, and that, if we could say what would be the 
final orbit of the mth electron bound by a nucleus of 
charge Z, we could deduce the general features of the 
periodic table and other atomic properties in the 
desired manner. He shows that already we know, 
partly theoretically, partly empirically, a very great 
deal about these binding processes. The arguments 
cannot usefully be summarised. The result is that 
we can specify with considerable certainty the two 
principal quantum numbers, and &, of the orbits in 
most atoms. The orbits thus fall into a number of 
groups, and we know the number of equivalent orbits 
in each group. The groups of orbits are arranged with 
various types of spatial symmetry; they must on no 
account be thought of, as in earlier models, as forming 
coplanar rings of electrons. The systematic study of 
X-ray levels begun by Kossel in the field opened up by 
Moseley has played a leading part in this development. 
There are two crucial points to be emphasised in the 
present position of the theory, which can best be stated 
as questions. Can we deduce from the quantum theory 
the particular points at which a group of electronic orbits 
fills and a new group starts ? In particular, can we prove 
that the third electron in the lithium atom must remain 
in a new type of orbit (n=2, k=1) and not fall into an 
orbit equivalent to those of the first two electrons 
(n=1,k=1)? Secondly, can we deduce from the theory 
the regular sequences of physical and chemical properties, 
together with their occasional interruption for groups of 
homologous “elements such as the iron group or the rare 
earths ? It can scarcely be said, and Bohr, I think, does 
not claim that an unqualified “ yes ” is yet the answer 
to the first question ; but putting the question is itself 
a great advance, and the lines on which an answer 
will be forthcoming are already clear. It seems certain 
QI 

