Codex @codex  0

The original report of the Eddington experiment, based on observations at the total solar eclipse of 29 May 1919.

Official publication: doi.org/10.1098/rsta.1920.0009

Original paper scan: w.astro.berkeley.edu/~kalas/labs/documents/dyson1920.pdf

The 1919 solar eclipse expeditions organized by Frank Watson Dyson and Arthur Eddington compared background-star positions near the obscured Sun with reference photographs. The optical train used a slowly rotating flat mirror to send the light into a horizontal photographic telescope. A smaller telescope at Sobral provided a separate set of plates. Focus and plate-scale changes had to be distinguished from the radial pattern predicted by Schwarzschild light deflection. The experiment therefore tested a spatial pattern and its amplitude, not merely whether a star image had moved.^{[ref] [ref]}

The three classic tests probe orbital motion, the propagation of light, and clock rates in a gravitational field.

Schwarzschild perihelion precession. For an orbit with semi-major axis $a$ and orbital eccentricity $e$, the leading advance per revolution isThe relevant observation for Mercury is the residual advance after the much larger ordinary perturbations have been included, approximately $43$ arcseconds per century.^[ref] An orbital fit must account for the other planets and reference-frame effects before interpreting that residual as a relativistic contribution.

Schwarzschild light deflection. A ray with large impact parameter $b$ passing a nearly spherical mass is deflected throughAt the Sun's limb this is about $1.75$ arcseconds. The Eddington experiment used the 1919 solar eclipse to photograph background stars near the obscured Sun, then compared their positions with reference photographs. The measured effect is an apparent outward displacement of the background stars, not a direct photograph of a curved light path. Telescope focus, plate scale and atmospheric effects must be controlled when extracting such small shifts.^[ref]

Gravitational redshift. For a photon emitted lower in a weak static field and received a height $H$ above it,The Pound-Rebka experiment used the narrow recoil-free nuclear resonance supplied by the Mössbauer effect. A controlled source motion supplies a small Doppler shift that compensates the gravitational shift, allowing a change in resonant absorption to reveal it across a laboratory tower.^[ref] The leading weak-field redshift follows from the equivalence principle and therefore does not, by itself, distinguish every competing metric theory of gravity.