x l i i I N T R O D U C T I O N T O V O L U M E 1 4 angles. A disadvantage of Compton’s setup had been that the shift in wavelength could only be measured for a fixed angle of scattering, and that the intensity of the two beams differed greatly. Einstein and Mark tried to reproduce Compton’s effect with a setup that made possible the observation of the scattered beam at different angles, and to determine its wavelength—albeit only indirectly. The idea was to use X-rays whose wavelength is close to an absorption edge of a suitable substance, and to observe a change in transmissivity when the shifted wavelength of the scat- tered beam passed through the absorption edge of the filter substance.[10] As we learn from Mark’s detailed account of their experiment (Docs. 119, 125), their idea was to use X-rays from the rhodium Kα-line with a wavelength of 0.6121Å and filter them through a molybdenum foil with an absorption edge at Å. As the scattering angle is increased, the wavelength is redshift- ed. Consequently, the scattered X-rays no longer have sufficient energy to induce ionizations in the foil, and the absorption shows a distinct fall-off. Initial tests showed the expected qualitative effect, but further improvements of the setup and observations seemed desirable. A major impediment was the difficulty of secur- ing the needed amounts of rhodium and molybdenum (Doc. 119). For the second experiment, most likely intended to test the Compton effect op- tically for ultraviolet light of mercury’s Å line reflected off a metal vapor or metallic mirror, Einstein collaborated with Peter Pringsheim. Details of their experiment are discussed in Docs. 147, 148, and 156. Pringsheim began to set up an experiment involving mercury fluorescence radiation (presumably us- ing a special mercury lamp that he had purchased earlier with KWIP funds see Doc. 156, note 1). But he subsequently found out from James Franck that a student of his was looking at the reflection of mercury fluorescence light with the aim of understanding the differences between diffuse reflection off metallic vapor and mirror-like reflection off a metallic surface. The conclusion must have been that any Compton-like shift of the mercury line reflected off mercury atoms would not be observable within the usual Doppler width of the lines. There are indications that further suggestions for experiments on the Compton effect were being explored in Berlin. For example, Leo Szilard, at the time a post- doctoral researcher on a KWIP stipend at the Kaiser-Wilhelm-Institute for the Chemistry of Synthetic Fibers, and Johann Böhm, another Assistent there, reported on planned experiments using a new design of X-ray tubes (Doc. 198). None of these experiments, carried out under Einstein’s guidance in Berlin be- tween August 1923 and early 1924, and intended to provide independent confirma- tion of Compton’s discovery, resulted in a publication.[11] During that time, the Compton effect was also discussed intensely at the Meeting of the Deutsche Physi- kalische Gesellschaft, held 16–23 September in Bonn, which Einstein attended (Einstein et al. 1923 [Doc. 116]). λ 0.618 = λ 2536.5 =