TRANSACTIONS OF SECTION A. 297 
the double charge on the a particle, but it does not follow when two of the 
movable electrons are freed that nine are left bound with the nucleus. The 
supposition that the atomic number for H is 1 and for He is 2, means that 
there are no intermediate elements between them. But there are several con- 
siderations which point to the existence of 2. (1) Nicholson has given very 
weighty reasons for supposing that the lines observed in the corona and in 
nebulz are due to two elements whose atomic weights lie between those of 
H and He, which he has called respectively Coronium and Nebulium. Their 
nuclear charges, however, are 4 and 5, which would make He 6. (2) Rydberg 
has proposed a theory of the constitution of the periodic table which has been 
remarkably justified in one respect by Moseley’s measurements, in so far that 
it requires 32 elements between Kr and Ra-Em in place of 36 as hitherto 
supposed. The same reason which requires these 32 elements also requires 
2 between H and He. (3) In the July number of the ‘ Philosophical 
Magazine’ Rydberg has discussed Moseley’s measurements of the frequencies 
of the Barkla K and L series, and finds that if N—the atomic number—be 
based on 4 for He, the frequencies of the lines can be represented by the follow- 
ing scheme :— 
K(a) and K(8) by P(N —3)? P(N- 3:5)? 
L(a) and L(8) ,, P(N-—3x 3)? P(N —8 x 3:5)? 
L(y) and L(8) ,, P(N-4x3)? P(N-4x3°5)? 
but that such an arrangement is impossible if N be based on any other number 
than 4 for He. More exact numbers, however, are needed before these relations 
can be regarded as established. 
We already know certain definite facts as to the constitution of an atom. 
They are :— 
(1) All atoms contain electrons as a part of their constitution. Of these 
they can apparently lose a certain number without altering their chemical 
identity, whilst in the case of radioactive elements the loss of other sets 
changes them into different elements. We shall doubtless be justified in the 
assumption that the same law extends to all elements. 
(2) There exist also positively charged nuclei associated with the atomic 
mass, containing multiples of the fundamental electric charge, and the evidence 
tends to show that the chemical nature of the element is determined by this 
multiple. 
(3) In the case of a certain number of substances there are found associated 
magnetic doublets whose moments are multiples of a definite quantity, called 
by Weiss the magneton. It appears legitimate to suppose that the same 
phenomenon may exist in other elements, though whether the magneton has an 
independent existence or is a consequence of electronic motion is an open ques- 
tion. If the latter, an explanation of the multiple quality will have to be 
sought for. 
Any theory of atomic structure must, then, be a theory of the way in which 
the atom is built up of these fundamental quantities. So far there are two 
types : (1) Thomson’s theory of an extended positive nucleus within which the 
electrons revolve in Saturnian systems; (2) Rutherford’s theory of an extremely 
small nucleus with electrons in planetary or Saturnian orbits. Neither of them, 
however, has shown the slightest aptitude in explaining the series laws of 
spectra. The actual structure must be a much more complicated one than is 
assumed in either. Unfortunately the complete mathematical treatment of the 
simplest case is one of extreme difficulty. We may, however, I believe, make 
one very important first step, viz., as to the direction in which to look for the 
source of the energy emitted in spectral radiations. This energy may arise 
either from small vibrations about a stable state or from change from one 
stable state to another. In both cases the stable states must be such as to 
lose no energy, and must therefore be in static equilibrium, or their relative 
motions must be such as to produce no change in an external field relative to 
itself{—such as, for instance, a charged sphere moving with uniform velocity. 
In the first case the energy would be made up of extremely small amounts from 
all the atoms, and an increase in intensity would be due to increased ampli- 
