The paper is an extension of this author's hypothesis, presented inAstrophys. Space Sci.3 (1969), 268, which explains the red shift in terms of a geometry of static space and stationary observers. The author introduces here the notions of ‘metric with an observer’, ‘observed distance’ and ‘space of observations’; he considers the problem of the equivalence of stationary observers and discusses the relation between the hypothesis, the special theory of relativity and Maxwell equations. Attention is drawn to the agreement between the hypothesis and the experimental results discussed by Shamir and Fox, and those discussed by Kennedy and Thorndike.