D O C . 9 2 E L E C T R O N A N D G E N E R A L R E L AT I V I T Y 1 1 1 formation is carried out. If we look at the associated components of the electric cur- rent density, , and specifically the density of electricity, , one realizes that under transformation it changes its sign, whereas according to (2) the gravitational field, and hence the (gravitating) mass remains unchanged. Therefore, if a solution exists that corresponds to a negative electron of mass μ and charge –ε, then a solution also exists that corresponds to the mass μ and the charge +ε. I sought, without success, a satisfactory way out of this problem. It may be use- ful, though, to add some remarks that are related to this.[4] 1) In a recently published paper on gravitation and electricity,[5] I believed I had overcome the mentioned difficulty by modifying the assignment of the tensor to the electromagnetic field, in that I took Then the current density must be taken to be a covariant tensor of 3rd rank, , and the density of electricity as given by . Indeed, the substitution (1) then does not change the sign of the electric field and of the electric charge density. However, now the substitution leads to the same difficulty. For with this substitution, the sign of the electric den- sity thus interpreted changes, without anything else being altered in a centrally symmetric electrostatic solution. ∂xν ∂f μν ∂x1 ∂f 41 ∂x2 ∂f 42 ∂x3 ∂f 43 + + [p. 332] fμν f23, f31, f12 as the electric field vector, f14, f24, f34 as the magnetic field vector. ∂xσ ∂fμν ∂xμ ∂fνσ ∂xν ∂fσμ + + ∂x1 ∂f23 ∂x2 ∂f31 ∂x3 ∂f12 + + t1 t = x1 x –= y1 y = z1 z =