It is argued, after a short review of the problem, thatMach’s principle, in an interpretation near toMach’s original ideas, is not incorporated in general relativity. On the other hand, one can formulate a selection principle on the basis of which it is possible to divide the solutions of gravitational equations into two classes. Those of the first class do, those of the second do not contradict the reasonable content ofMach’s requirements. Thus, these latter solutions may perhaps remain permissible even in the future, when a fuller knowledge proves the others to be an unpermissible extrapolation beyond the domain of validity ofEinstein’s equations. We formulate a selection principle, which seems to be more general and flexible than previous ones.