I N T R O D U C T I O N T O V O L U M E 9 x x x v i i
With respect to the remaining numerical discrepancies among the final values,
skeptics and believers alike now looked forward to the eclipse of 1923 as a further
opportunity for confirming or refuting the new results.
I V
With the confirmation of the second test of general relativity, issues revolving
around its third empirical consequence, the gravitational redshift, became all the
more acute. If a light source emits light at the Sun with a frequency ν0, measured
locally, the frequency of the light received and measured at the Earth is ν, and the
relative shift in frequency is
where Φ is the (negative) gravitational potential difference between the points of
emission and detection at the surface of the Sun and the Earth, respectively, and c
is the velocity of light.
Skeptics objected that relativity could not be correct if it failed this crucial test,
a view shared by Einstein himself. As he remarked to Eddington: “If it were proved
that this effect did not exist in nature, then the whole theory would have to be aban-
doned” (Doc. 216). But whereas outspoken critics such as Silberstein were certain
that the evidence from observations of the Sun’s spectrum was overwhelmingly
against the “Einstein effect,” Einstein himself remained sanguine.
Like the perihelion advance of Mercury, the redshift of spectral lines in the Sun
had been of interest to astronomers before the advent of relativity theory. That the
majority of solar spectral lines show a distinct, but small, shift to the red when com-
pared to laboratory spectra had first been ascertained in the late 1890s by Henry A.
Rowland and Lewis E.
Jewell.[27]
A pressure-dependent line shift in the spectra of
arc lamps initially suggested some similar effect for the Sun, but subsequent re-
search showed that high pressure in the solar atmosphere could not account for the
appearance of the lines. Willem H. Julius tried to explain the shift in terms of the
anomalous dispersion of light in certain optically active substances in the solar at-
mosphere, but his theory never enjoyed wide acceptance among solar physicists,
and by 1919 most of them rejected it
outright.[28]
The other principal candidates for
an explanation of the solar redshift were Doppler shifting due to the motion of solar
gases along the line of sight, and the Einstein effect.
The Doppler effect, depending upon motion along the line of sight, was perfectly
suited to explain the solar redshift observed in gases located near the center of the
ν0 ν –
ν0
--------------
Φ
c2
----= - 2.10–6,
=