D O C U M E N T 3 3 6 O C T O B E R 1 9 2 4 5 2 9 —Quite so. It will be difficult to do altogether without a certain continuity of matter and we shall believe in the existence of a certain medium between the var- ious points of matter. Call it ether or whichever you like. But in a physical theory, such a concept will be meaningless. The undulatory hypothesis will have to be dropped and a new theory of emission sought for. We are now in a position similar to that which faced the physicists when they had to decide between Coulomb’s stat- ic theory and Faraday’s electrical field. Coulomb held that electricity is a mere property of matter, while Faraday, and then Maxwell, taught that the electric field is a reality of an independent existing. The issue between the two views was decid- ed by certain experiments where everything could not be explained by static electricity. —Am I to understand thereby that Eddington’s remarkable synthesis is hope- lessly condemned? The unification of the gravitational field and the electromagnet- ic field, that is to say the assimilation of physical phenomena to an all-embracing system of geometry, is essentially incompatible and unworkable with a discontin- uous matter. —It may be disappointing for those who believe that the unity of physical forces is nearing its mathematical expression but I am sorry to say that so far as the elec- tron theory is concerned, the Theory of Relativity has failed to give an adequate account of it. Eddington’s generalization of the Theory of Relativity is the most wonderful achievement of the new doctrines since their inception. Eddington went considerably further than Weyl and myself. But his mathematical description of na- ture which I very deeply admire, unfortunately clashes with the most recent exper- iments in physics. We must not forget that the Theory of Relativity is a theory of principle, and that Eddington changed nothing to the original position I had taken when I first proposed these new ideas. While, on the other hand, the electron theory is essentially a matter of unceasing experiment. A clash between reason and reality is therefore not at all impossible. To give you a concrete instance, I shall say that I find this to happen in the very formulae Eddington gives of the electron when he describes it mathematically with reference to the potentials of gravitation. This mathematical description contains an effective term which I discover nowhere in the electron experiments. To make the two agree together, this term must be con- sidered as infinitely small if not practically neglected. But there is no experimental reason why it should be taken in that way so that the discrepancy between reason and reality becomes evident here. At first I was reluctant at abandoning Edding- ton’s conclusions. I may also confess that I did my best to make them agree with the electron experiments. I tried first the method of Hamiltonian integrals, which gave nothing, then the method of superdetermining the equations, as I call it. You know, for instance, that the ten potentials of gravitation are characterized, in my