I N T R O D U C T I O N T O V O L U M E 1 2 l v
But among all these experiments, the one that gave Einstein his “strongest sci-
entific experience in years,” as he wrote to the Borns on 30 December (Doc. 345),
was the one designed to probe the process of light emission by canal ray particles.
The first mention of such an experiment is found in a letter to Born of 22 August
(Doc. 211), followed only six days later by a report that Einstein had already begun
its execution at the Physikalisch-Technische Reichsanstalt (Doc. 218). His experi-
mental collaborator there was Hans Geiger; they were later joined by Walther
Bothe.[61]
Einstein’s interest in light quanta dated back to his earliest publications, most
notably to his heuristic introduction of the concept in 1905 (Einstein 1905i [Vol. 2,
Doc. 14]). In the following years, his analysis of energy fluctuations for Planck’s
distribution law supported it (Einstein 1909b, 1909c [Vol. 56, Docs. 56, 60]), and
observations by Millikan in 1916 confirmed Einstein’s relation between energy and
emitted frequency for the photoelectric effect, although the theoretical interpreta-
tion of Millikan’s experiments remained controversial. The same year, following
his result of the directional nature of quantum emission, Einstein concluded that
“there is no emission of radiation in the form of spherical
waves.”[62]
In 1921, the
light quantum was still controversial among many of Einstein’s colleagues. At the
Solvay conference that summer, Niels Bohr in his paper stated that Einstein’s quan-
tum presented “difficulties that appear insurmountable”: the hypothesis still stood
in conflict with the familiar interference phenomena of
light.[63]
In order to probe the nature of the light emission process, Einstein proposed an
experimental arrangement with canal ray light sources. According to the wave pic-
ture account of the emission process, light emitted would be subject to the Doppler
effect, with its familiar angular dependency for the frequency. Einstein believed
that quantum theory, too, would allow for the existence of a Doppler effect (whose
existence had been demonstrated for canal ray sources by Johannes Stark as early
as 1905, see Doc. 261), yet, he expected that the Doppler shift in a quantum emis-
sion process would give very different dispersive properties of
light.[64]
To test the different perspectives, Einstein proposed to collect light emitted into
the angle of the aperture of a lens and to focus it onto a slit, and then project the
emitted light rays into a parallel bundle by a secondary lens. If the light rays enter-
ing the optical setup were to display a Doppler shift as predicted on the basis of the
wave picture, there would have to be coherent phase fronts that showed a position-
dependent frequency shift. The idea of the experiment was to test this scenario by
sending the parallel light bundle through a dispersive medium with a frequency-
dependent diffractive index. Upon traversing such a dispersive medium, Einstein
argued, a bundle of light rays with variable frequency would be bent away from a