220 
THE QUANTUM [ X v L 2 
spectra. This analogy, which has been the subject of much 
discussion, is difficult to explain satisfactorily on the basis of a 
relativity effect. If, however, we adopt the hypothesis of the 
spinning electron, an immediate explanation is obtained. 
The well-known doublets in spectra of the alkaline type, of 
which the D-lines of sodium form a conspicuous example, are 
now to be regarded as regular spin-doublets. This view enables 
us to explain the dependence of the doublet separation on the 
effective nuclear charge and the quantum numbers describing 
the orbit, as well as the rules of combination. 
In the atom of an alkali a single electron describes an orbit 
round an atomic residue containing only completed electronic 
groups, which are magnetically inert. But as the electron has a 
magnetic moment, its axis will experience a precession because 
of the couple due to its motion in the electric field. There will 
also be a corresponding secular perturbation of the plane of the 
orbit. It can then be shown that the energy of a single orbit 
splits up into a set of energies, the differences between which 
vary as Z 4 as required by the observations. L. H. Thomas has 
obtained a separation formula giving correct numerical values. 
When we pass to atoms in which several atoms revolve round 
a magnetically inert core, the results are more complicated as 
there are other directing influences on the axis of each spinning 
electron. Multiplets of higher complexity now make their appear 
ance, and although it may be difficult to give a quantitative 
explanation of the phenomena observed, we can account for 
them in a general way by means of the new hypothesis. 
(3) The Zeeman Effect.—The introduction of the spinning 
electron was primarily suggested by the anomalous Zeeman effect. 
The simple Lorentz resolution demanded by Larmors theorem is 
found only for lines of singlet systems. In practice the com 
ponents of multiplet structures show anomalous effects which 
formerly gave rise to grave difficulties in the attempt to explain 
them. The more complicated structures usually observed can 
be described by introducing a “ splitting factor.” This anomaly 
could be explained formally if the magnetic moment of the atom 
arose from two sources, one of these being the orbital angular 
momentum. According to Uhlenbeck and Goudsmit the diffi 
culties disappear at once when it is assumed that the electron 
has a spin, for the spinning electron provides a second source of 
magnetic moment, and it has been shown that a qualitative 
explanation can then be given. 
On the basis of the restricted theory of relativity the kine 
matics of an electron with an axis has been discussed by Thomas.* 
* L. H. Thomas, Nature, vol. 117, p. 514, 1926; Phil. Mag., vol. 3, 
p. 1, 1927.