2 5 4 D O C U M E N T 2 4 4 A P R I L 1 9 2 6 false in the physical sense because it adds nothing of physical meaning at all to the old th. of rel. but rather only represents it, in the way that a Lorentz transformation can be described as a rotation in Minkowski space. Conceived this way, I believe I can counter your objections. 1) The assumption for is not only arbitrary but even artificial but why shouldn’t one be permitted to do that? Considered purely logically, one can define what one wants here there- fore, one may also frame the assumption in such a way that it agrees with the al- ready known law of motion of charged point masses. 2) You say that no metric corresponds to my but it is precisely the opposite here. Eddington first takes a field as primary and derives from it a field , which he splits up into a symmetric part and an antisymmetric one. However, one can just as well take a field as primary and derive a field from it this is logically equivalent. The type of space used by me can be characterized in the following manner. In Rie- mann space the displacement operation is arranged specifically to yield a length comparison that is the same as what the metric yields. From this requirement, the Riemann values for the follow only if it is additionally assumed that the are symmetric in μ and ν. If this assumption is dropped one has somewhat more general available. The then only define a comparison of direction, not of length, because the latter is relinquished to the metric. This comparison of direction can then be arranged such that a charged point mass moves along the straightest line, which is then not coincidental with the shortest line. In this way one obtains a perfect geometric visualization of the law of motion. (3) My law of mo- tion is valid not only for a particular ratio of charge and mass but for any charge [6] only the mass must be = 1. From the point of view of geometric visualization, this shortcoming is no worse than when, e.g., not all yardsticks, but just the one of length 1 defines . But I do see that I must present the matter completely differently if others are not to misunderstand me! Physicists basically cannot imagine at all that in such things there is not at least a secret physical intention! If even you understand it this way, how else are the others going to understand it! Maybe my intention will become clearer to you if I tell you how I hit upon this matter. At the present time I am writing a comprehensive philosophical exposition on the problem of space and for it I had to present a description of the Weyl space.[7] Thereby I was pondering about what the geometric description of electric- Γμν τ Γμν τ Γμν τ Gμν Gμν Γμν τ μν τ ¯ ¿ ® ¾ ­ ½ Γμν τ Γμν τ Γμν τ Γμν τ ds2