QUANTIZATION IN SPACE
89
vi. 3]
3. Later Experiments
Recently experiments have been carried out in America by
J. B. Taylor.* * * § He has modified the apparatus of Gerlach and
Stern so as to be able to determine the magnetic moments of
sodium and potassium atoms. He first repeated the experiment
of Stern and Gerlach for silver atoms, using a silver-plated
tungsten coil heated electrically as a source of the atoms instead
of the furnace employed by the first experimenters. This
simplified the procedure as there is considerable difficulty in
ridding the furnace of gas. The slits employed were too wide to
show more than a decided broadening of the image with a trace
of division in some of the plates. In the experiments on sodium
and potassium the slits were made narrower (0-03 mm.).
“ The metals were evaporated into the apparatus at 345 0 C.
and 245 0 C. respectively, and the images formed by the deposition
of the atomic rays on cool glass strips were rendered visible by
immersing the strips in hydrochloric acid gas, whereupon films
of opaque chloride were formed. Both metals were found to
possess an atomic moment of one Bohr magneton, within the
limits of experimental error, which were about ten per cent.”
The question as to whether molecules as a whole assume
quantum orientations in a magnetic field has been brought into
prominence by the experiments of Glaser, f Although the so-
called Glaser anomaly has not been confirmed by later experi
menters, some reference to the question seems called for. He
investigated the magnetic susceptibility of the gases hydrogen,
nitrogen and carbon dioxide at low pressures by means of a
very sensitive form of Curie balance. On the classical theory
the susceptibility should be proportional to the pressure of the
gas. This was not in agreement with the experiments. Glaser
suggested that his results, which cannot be explained by the
classical theory, depend on the relative values of the times taken
by the molecules to turn in the direction of the field and the
times between molecular collisions, the ratio between these
two times depending on the pressures and fields employed.
On the other hand Lehrer $ has found proportionality between
volume susceptibility and pressure both for carbon dioxide and
for argon ; Hammar § finds no evidence of a magnetic anomaly,
though with imperfectly dried gases he obtains curves not unlike
those of Glaser.
* Taylor, Phys. Rev., vol. 28, p. 576, 1926.
t Glaser, Ann. d. Physik, vol. 75, p. 459» 1924.
% Lehrer, Zeits. f. Physik, vol. 37, p. 155, 1926 ; Ann. d. Physik,
vol. 81, p. 229, 1926.
§ Hammar, Proc. Nat. Acad. Sci., vol. 12, p. 594, 1926.