I N T R O D U C T I O N T O V O L U M E 1 5 l x v able session of the National Academy of Sciences at which Miller and Strömberg presented their results, Hale reported on work by Walter S. Adams, an astronomer at the Mount Wilson Observatory, that demonstrated the existence of a gravita- tional redshift in the spectrum of the white dwarf star Sirius B (Adams 1925). This result was in agreement with earlier calculations by Eddington (Eddington 1924). Skepticism about the “third test” of relativity had played, for a time, a role in sus- taining the critics of relativity (see, e.g., Vol. 9, Introduction, pp. xxxvii–xl). Eddington, who had also been contacted by Slosson, replied that Miller’s attempt to reopen the debate on special relativity came just as the empirical verdict on gen- eral relativity was beginning to look quite settled (see Doc. 12). Einstein suspected that small differences between the arms of the interferometer were to blame for Miller’s results: “Temperature differences in the air between the two beams, of the order of magnitude of 1/10 of a degree, would suffice to produce the whole fuss,” he wrote to Ehrenfest (Doc. 71). He did not find any reference in Miller’s publications as to how this error had been avoided, and suggested as much to both Miller and Piccard (Doc. 219). Miller agreed with the remark that 1/10 of a degree difference would produce the observed displacement of interference fringes, therefore “very elaborate precautions have been taken to eliminate such an effect of temperature” (Doc. 289). However, these precautions had to compete with Miller’s desire not to enclose his apparatus in stone walls. Einstein told Millikan privately that he distrusted Miller’s result but had no right to say so in public (Doc. 58). The data reaching Einstein strengthened his opinion, expressed much earlier to Ehrenfest: “Basically I think nothing of Miller’s experi- ment, based on my malicious soul, but I must not say it aloud.” He characteristical- ly invoked the Creator, whom he credited “with more elegance and intelligence than that” both to an old friend (Doc. 26) and to Ehrenfest: “the difference between Cleveland and Mount Wilson cannot be so significant, considering the grand scale on which the Old One created the world” (Doc. 49). When he had first learned of Miller’s experiments in 1921, he had expressed his distrust in a similar vein: “Subtle is the Lord, but malicious he is not” (see Vol. 12, Introduction, p. liii). Now we see exactly what he had in mind. Scientists decided upon new observations aimed at replication of results, al- though Einstein even here had doubts, desperately noting the expense involved: “What can one do now to bring some order to this epidemic? It would be a pity to spend too much money on this shady matter” (Doc. 86). Many of those planning such replication endeavors communicated with Einstein, some seeking help in rais- ing funds. While Kennedy’s experiment was certainly influential in turning opinion against Miller, Piccard’s balloon-based experiment additionally proved to be of