The new New York Times article The Eclipse That Made Einstein Famous describes several events and discussions on the hundred year anniversary of the the 1919 solar eclipse and verification of the deflection of light by the Sun as predicted by General Relativity. The roles of Albert Einstein, Arthur Eddington, and others are discussed in the article, the play performed at the Royal Astronomical Society in London, in the Nature Physics article A hundred years of the first experimental test of general relativity and the book No Shadow of a Doubt; The 1919 Eclipse That Confirmed Einstein's Theory of Relativity by Daniel Kennefick.

The instruments used for the measurement are shown in the illustration and photograph below.

My question is about the astrometric technique used, and the calibration and verification.

The instruments are long and laying on their sides, viewing the eclipse and the stars near the Sun by reflection in coelostats (mirrors in front of telescopes, moved by clockwork mechanisms).

I can imagine several sources of instrumental and experimental errors due to variations in field distortion and magnification, presumably the measurement was done by comparing the eclipse images with similar images taken by the same instruments of the same field of stars in the same position in the field of view a few months later, at night, in the dark, in order to show that the tiny deflections were not due to variation from one instrument to another.

Is that actually how it was done? Was the experimental set-up left in-place between day and night measurements, or was it packed up and moved?

Was the flatness of the coelostat mirrors independently confirmed somehow?

Images from the new New York Times article The Eclipse That Made Einstein Famous

enter image description here

Einstein’s theory suggested that starlight would bend near a massive body like the sun.CreditChronicle/Alamy

In Sobral, Brazil, two coelostats with movable mirrors were used to direct images of the eclipsed sun into a pair of horizontal telescopes. Credit Royal Observatory, Greenwich/Science Museum Group

enter image description here

  • 1
    $\begingroup$ It's not that bad. You can pack things up and take a shot in another place, and the results are going to be very repeatable. Field flatness affects mostly focus and aberration, not the location of the center of the diffraction figure. The size of the image is dictated by the focal length (if imaging in prime focus) and so it's fixed for a given instrument with very high precision. Ergo, taking images X months before or after should provide exact replicas of the eclipse shots, sans the GR effects. 1.7 arcsec is quite a lot of bending; the resolving power of just 100 mm of aperture is 1 arcsec. $\endgroup$ May 29, 2019 at 4:00
  • $\begingroup$ BTW, "magnification" is a bad word among the cognoscenti. Try to avoid it. ;) It's not relevant to this discussion anyway. $\endgroup$ May 29, 2019 at 4:05
  • 2
    $\begingroup$ Dyson, Eddington, and Davidson 1920 may help. $\endgroup$
    – Mike G
    May 29, 2019 at 4:39
  • 1
    $\begingroup$ I suspect the answer is that you should probably read Kennefick’s book, but Mike G’s suggestion looks like a very good start. From a cursory read, it appears that in the two months between the eclipse observations and follow-up nighttime observations, they left everything in place except the mirrors, which they stored inside to keep them dust-free. $\endgroup$ May 30, 2019 at 0:34
  • 1
    $\begingroup$ @uhoh Knock yourself out. :) $\endgroup$
    – Mike G
    Jun 3, 2019 at 22:00


You must log in to answer this question.