Page:Benton 1959 The Clock Problem (Clock Paradox) in Relativity.djvu/20



A reply to McCrea's criticism, Nature 177:784-785, Apr. 28, 1956.

A statement prepared in conclusion of clock paradox argument with W. H. McCrea.

According to the restricted theory of relativity, moving measuring rods are shortened in the direction of motion, and moving clocks run slow in the rates (1-v2/c2)½: 1; this view is correct so far as measuring rods are concerned but is incorrect with respect to clocks. Circumstantial evidence for the contraction of a moving rod is given by the Michelson-Morley and other experiments; there is no evidence for the definite retardation of clocks. There is in physics, no explicit definition of a clock. By consideration of a clock of hour-glass type it is shown that the ratio of time intervals recorded by stationary and moving clocks is different according as the interval is measured by the number, by the total weight or by the total volume of similar particles falling into a receptacle. The corrections to be applied to the readings of clocks in motion in order to realize the readings of an ideal clock if it were set moving are almost as various as the constructions themselves. The transformation formula for t is a logical consequence of that for x; it is the necessary consequence if Newtonian mechanics is to hold good for uniformly moving systems despite the change in x. The Kennedy-Thorndike experiment does not afford a confirmation of the Lorenz transformation for time, but shows that light behaves as an ideal clock. The statement that relativity has caused a fusion of time and space is not true; the conformability of light to Newtonian mechanics, established by the Michelson-Morley and Kennedy-Thorndike experiments, makes it possible to define corresponding units of space and time in terms of light.

Chapter IV: Time.