254 RELATIVITY, THE GENERAL THEORY
of a letter to Herve Faye in which he recorded his findings [L2]. The perihelion
of Mercury advances by thirty-eight seconds per century due to 'some as yet
unknown action on which no light has been thrown* ... a grave difficulty, worthy
of attention by astronomers.' The only way to explain the effect in terms of known
bodies would be (he noted) to increase the mass of Venus by at least 10 per cent,
an inadmissible modification. He strongly doubted that an intramercurial planet,
as yet unobserved, might be the cause. A swarm of intramercurial asteroids was
not ruled out, he believed. 'Here then, mon cher confrere, is a new complication
which manifests itself in the neighborhood of the sun.'
Perihelion precessions of Mercury and other bodies** have been the subject of
experimental study from 1850 up to the present.f The value 43 seconds per cen-
tury for Mercury, obtained in 1882 by Simon Newcomb [Nla], has not changed.
The present best value is 43". 11 + 0.45 [W4]. The experimental number quoted
by Einsteintf on November 18, 1915, was 45" ± 5 [E48].
In the late nineteenth and early twentieth centuries, attempts at a theoretical
interpretation of the Mercury anomaly were numerous. Le Verrier's suggestions
of an intramercurial planet^ or planetary ring were reconsidered. Other mecha-
nisms examined were a Mercury moon (again as yet unseen), interplanetary dust,
and a possible oblateness of the sun [O2, F3]. Each idea had its proponents at one
time or another. None was ever generally accepted. All of them had in common
that Newton's 1 /r^2 law of gravitation was assumed to be strictly valid. There were
also a number of proposals to explain the anomaly in terms of a deviation from
this law. Recall that Newton himself already knew that small deviations from the
power —2 would lead to secular perturbations of planetary orbits [N2]. Two
kinds of modifications from Newton's law were considered: a slightly different,
purely static law [O3] or a 1/r^2 law corrected with velocity-dependent terms [Zl]
(Lorentz's theory of gravitation mentioned in Chapter 13 belongs to this last cat-
egory). These attempts either failed or are uninteresting because they involve
adjustable parameters. Whatever was tried, the anomaly remained puzzling. In
his later years, Newcomb tended 'to prefer provisionally the hypothesis that the
sun's gravitation is not exactly as the inverse square' [N3J.5H
Against this background, Einstein's joy in being able to give an explanation
'without any special hypothesis' becomes all the more understandable. The tech-
*'. .. du a quelque action encore inconnue, "cui theoriae lumen nundum accessed!." '
**See, for example, the table in [W4].
t A detailed list of nineteenth century references is found in [Ol].
f fEinstein took this value from a review by Freundlich [F3]. For his appreciation of Newcomb, see
[E49a].
Uln the 1870s, it was briefly thought that such a planet (it was named Vulcan) had actually been
seen.
51[For a detailed survey of Le Verrier's and Newcomb's work, see [Gl].