x l v i I N T R O D U C T I O N T O V O L U M E 1 5 nonlinear field equations, and between Einstein’s program of searching for solu- tions that would be capable of representing both the interior and the exterior of electrons and Rainich’s program of deriving the properties of such particles from their exterior fields (Docs. 216, 245, 258, 293, 300). In the course of the correspondence, Einstein underwent a remarkable change. On 18 April 1926 (Doc. 258) he wrote that the “cardinal question is of course whether one should think of electricity as continuous or made up of singularities,” two options he had already offered in his 1922 Princeton lectures (Vol. 7, Doc. 71). There he had deemed the latter option a pseudosolution (Scheinlösung), favoring the former option instead, just as he did in his correspondence with Rainich. How- ever, on 6 June 1926, he wrote: “[This] is the core question: a theory is sensible only if it allows a derivation of the equations of motion of particles without any ex- tra assumptions. Whether the electrons are treated as singularities or not does not really matter in principle” (Doc. 300).[9] This line of thought culminated in the longest scientific writing in the present volume: a paper cowritten with Jakob Grommer on the problem of motion, i.e., on how to derive the equations of motion of particles directly from the gravitational field equations. The paper, Einstein and Grommer 1927 (Doc. 443), starts out by contrasting three possible approaches (Betrachtungsweisen) to the problem, two of which are serious contenders. The question is whether to start from the full Einstein equations, with the energy-momentum tensor of generic material systems as a source term, and to derive the equations of motion via the Bianchi identities or whether to start instead from the vacuum Einstein equations and allow for particles to correspond to singularities in the metric field. Einstein and Grommer opt for the latter alternative and start by investigating a special class of solutions to the vacu- um field equations, the Weyl class of static axialsymmetric solutions. Einstein had cited Weyl’s and Levi-Civita’s papers on such solutions during his correspondence with Rainich (Doc. 258), arguing that they contained a static two-body solution, exactly the kind of solution that Rainich had claimed would render general relativ- ity empirically inadequate because it would correspond to two bodies exerting gravitational fields on one another, yet not move toward each other (Rainich 1926b). But Rainich had kept insisting that a static two-body solution need not ex- ist in a nonlinear theory such as general relativity (Doc. 293). Einstein ended up agreeing with him (Doc. 300). He must have gone back to these papers by Weyl and Levi-Civita, following Rainich’s insistence, and it is plausible that this brought about the change in what constituted “the core question.”