I N T R O D U C T I O N T O V O L U M E 1 5 x l i i i line-by-line comments in which he pointed out typographical errors, weaknesses in the argument, and ways to improve it. In his three letters to Einstein (Abs. 93, 114, and 214), Alexander asked him to address Rainich’s comments, and whether any changes to the original paper should be made no answer from Einstein is extant. Soon afterward, Rainich wrote to Einstein directly (Doc. 96). Quoting from the opening remarks of Einstein’s latest attempt at a unified field theory of gravity and electromagnetism (Einstein 1925t, Doc. 17), in which Einstein had stated that a convincing theory had yet to be found, Rainich enclosed a paper of his own from the year before (Rainich 1925a) in which he believed he had shown that there is a way to provide the sought-after unification within the framework of Einstein’s original theory of general relativity. This time Einstein answered swiftly (Doc. 106). But rather than engaging with Rainich’s paper, he elaborated on how much it disturbed him that in the context of the 1915 Einstein equations with an electromagnetic source term, the electromagnetic and the gravitational field enter as separate entities. In response, Rainich summarized his results on how and to what extent one could unify gravity and electromagnetism within the framework of general relativ- ity, i.e., within a pseudo-Riemannian manifold subject to a version of the Einstein equations. His account relied on a decomposition of the Riemann tensor into a sym- metric and an antisymmetric tensor he related the former to gravity and the cos- mological constant, and the latter to electromagnetism (Doc. 126). Instead of being captured by this possibility, Einstein latched on to Rainich’s mathematical result of decomposing the Riemann tensor in this way, and to the fact that Rainich had managed to relate both the gravitational field and the cosmologi- cal constant to the symmetric part of the curvature tensor. This must have prompted Einstein to return to his own modified field equations of 1919 (Einstein 1919a, Vol. 7, Doc. 17) in which he had first suggested an alternative to his field equations of late November 1915 with the total energy-momentum tensor given by that of the electromagnetic field. The alternative consisted in replacing the Einstein tensor by the trace-free tensor . This allowed him to obtain the cosmological constant as a constant of integration and to reconceptualize it as a constant negative pressure term. He had hoped to find solutions to the modified field equations that would be capable of representing elementary particles, with the cosmological constant thus reinterpreted playing a role in ensuring the stability of the particles. At the same time, Einstein explicitly did not think of the 1919 equa- tions as a unification of the gravitational and electromagnetic fields, as expressed at the time in a letter to Kaluza of 29 May 1919 (Vol. 9, Doc. 48). Rμν 1 2 -- - gμνR – Rμν 1 4 -- - gμνR –