2 3 8 D O C U M E N T 2 4 5 J U L Y 1 9 2 8 I thank you warmly for communicating the general method of integration for the approximation equations that belong to the second Hamilton invariant (density),[3] The whole problem is, however, considerably more difficult than I had thought. Mr. Grommer’s doubts were indeed justified in part.[4] As you know, I had interpreted as the electromagn. potential,[5] and that these quantities in the first ap- proximation satisfy the Maxwell relations[6] and . On the other hand, it had not been proven that every solution of these equations cor- responds to a system , as must be required.[7] In particular, the total solution of appears not to exist. Thus, according to this theory, there seem to be no electrical masses in the sense of Maxwell’s theory in its present form.[8] This is one of the reasons why I have not yet succeeded at all in finding a physical interpretation of the theory up to now. One must, however, keep in mind that according to Bohr’s results, the relationship between the motion of the singularities and the wave field is not at all as simple as it should be according to Maxwell.[9] One thus should not slavishly follow Maxwell, but rather ask quite without prejudices: What kind of singularities are allowed by the theory? How do they behave? What is the nature of the interactions between the singularities and the field? In all of this, I actually imagine the theory to be free of singularities. Singulari- ties must be introduced only when one makes use of approximation equations and forgoes an analysis of the structure of matter.[10] When I return,[11] we should discuss this matter in detail. Of course, one could simply declare the whole of our efforts to be utopian. But a theory that is mathe- matically so natural is worth serious consideration, especially in view of the cur- rently disparate situation of theoretical physics.[12] Best regards, yours truly, A. Einstein. hggg = x --------- 0 = 0 = ha 1 2 3 0 = = = 4 j r =