After the success of the special theory of relativity, which is formulated on the Minkowski space-time for inertial observers, it was natural for Einstein to look for a theory that extended its symmetry to all observers for defining the laws of physics. As the special theory considers only observers in relative motion with uniform velocities, it is natural to wonder as to what principle should be satisfied if the laws have to be invariant for observers in relative acceleration. In fact, if one looks at Newton’s theory, which followed Galilean invariance, as time was absolute, it did not matter when the forces applied induced acceleration, as observers still could relate to each other. But with special relativity, having given up the absoluteness of time, this would no longer be possible. The first question that comes to mind is how to relate observers in a field of gravity, which induces acceleration. To quote Tolman [12], ’Because of the fact that the expression of the equations of physics in a form which is independent of any coordinate system does not in general prevent a change in their numerical content, under coordinate transformations, but relate the changes to conceivable changes in the gravitational field, one is able to eliminate criteria for absolute motion’.