552 SCIENCE PROGRESS 



no theory or hypothesis has been put forward by them which 

 is capable of explaining at one and the same time, not only the 

 motion of the perihelion of Mercury, but also the deflection of 

 light rays in a gravitational field. It is easy to put forward 

 possible alternative explanations of the latter phenomenon, 

 but then these " explanations " are merely qualitative, and 

 are not quantitative predictions, as was Einstein's : further, 

 they leave unexplained the movement of Mercury's perihelion, 

 which has long remained an outstanding anomaly in gravi- 

 tational astronomy, and of which Einstein's theory alone 

 gives a rational explanation. Other theories to account for 

 this anomaly have, indeed, been previously put forward, but in 

 no instance were they convincing. A few hypotheses have been 

 put forward which have appeared qualitatively to explain both 

 phenomena ; thus Sir Oliver Lodge has suggested that the 

 dielectric constant of the aether may have a value in the 

 neighbourhood of the sun (or other matter) different from 

 that in free space. Prof. Lindemann has pointed out, however 

 (Observatory, 43, 42, 1920), that the value deduced from the 

 motion of the perihelion of Mercury is K = 6, whilst that 

 deduced from the deviation of light from the stars is K = 4. 

 The two results are mutually inconsistent. In a similar way 

 do other explanations fall to the ground. 



It may be objected that, in spite of the ability of Einstein's 

 theory to account for the motion of Mercury and the deflec- 

 tion of light rays in a gravitational field, the spectroscopic 

 evidence is against it. The position as regards the existence 

 or not of a displacement of solar spectral lines towards the red 

 is at present somewhat peculiar. As far as observation is 

 concerned, it may be stated that the existence of the shift 

 has been neither definitely proved nor disproved, although 

 the balance of evidence appears against it. The spectro- 

 scopists at the Mount Wilson Observatory and elsewhere are 

 doubtless busy at the present time collecting further evidence 

 on this matter. Theoretically there are several interesting 

 possibilities ; Sir Joseph Larmor (Proc. R. S., ser. A, 96, 360, 

 1 919) considers that Einstein's theory does not require a 

 spectral displacement, and Cunningham (Nature, 104, 395, 

 1 91 9) is inclined to support this conclusion. The question is 

 very carefully discussed by Prof. Eddington in his Report re- 

 ferred to above, and he decides that there is " little chance 

 of evading the conclusion that a displacement of the Fraun- 

 hofer lines is a necessary and fundamental condition for the 

 acceptance of Einstein's theory ; and that if it is really non- 

 existent, under conditions which strictly accord with those 

 here postulated, we should have to reject the whole theory 

 constructed on the principle of equivalence." This agrees 



