Roberto Martins Searching for the Ether DIO 17
through the ether produced a real contraction of all moving bodies, according to the early explanation proposed by Fitzgerald and Lorentz. According to Lorentz, the principle of relativity would hold exactly for any optical or electromagnetic phenomenon, but Courvoisier did not follow Lorentz’s theory in this respect. He directly denied the principle of relativity and attempted to measure the motion of the solar system through the ether using several different techniques.
In 1921 Courvoisier published his first thoughts on the possibility of measuring the absolute velocity of the Earth through the ether.9 According to Courvoisier’s own declaration, his early calculations concerning the motion of the Earth were an outcome of routine work.10 In 1920 the Leyden Observatory published the details of a large series of observations of stars close to the North Pole that had been made between 1862 and 1874. Those measurements used an old method aiming to reduce observational errors: the stars were observed both with the meridian telescope directly pointed to them, and with the telescope pointed to the images of the stars reflected by a mercury mirror. This double assessment allowed corrections for any changes of the local vertical due to geological motions. It occurred to Courvoisier that those determinations could be used to measure the speed of the Earth through the ether.
Courvoisier assumed that the reflection of light by a mirror could undergo some influence of the motion of the mirror through the ether, even when the effect was observed relative to the proper reference system of the mirror. Any observable effect should be of the second order in v/c. It would be impossible to detect such a small effect if the speed of the Earth relative to the ether was about 10~4 c (that is, its orbital velocity), because for usual angle measurements (let us say, 60°) a difference of 10~8 would amount to only 0.002" - an effect that could not be observed. However, Courvoisier assumed that there could exist a much larger speed of the whole solar system relative to the ether, and analyzed the data published by the Leyden Observatory searching for some systematic effect.
He computed the difference z- z' between the direct zenith distance z and the reflected zenith distance z' of the stars listed in the catalogue, attempting to find a systematic effect that varied in a periodic way with the sidereal time of observations. Using a graphical method, he did find such an effect, and then he submitted the data to quantitative analysis. He derived an equation to describe the reflection of light in a moving mirror and
9 Leopold Courvoisier, “Zur Frage der Mitführung des Lichtäthers durch die Erde”, Astronomische Nachrichten, ccxiii (1921), 281-8; idem, “Über astronomische Methoden zur Prüfung der Lichtätherhypothese”, Astronomische Nachrichten, ccxiv (1921), 33-36.
10 Leopold Courvoisier, “Ergebnisse von Beobachtungen und Versuchen zur Bestimmung der ‘absoluten’ Erdbewegung”, Scientia, xlvii (1930), 165-74; French translation: “Résultats d’observations et d’expériences faites pour la détermination du mouvement ‘absolu’ de la Terre”, Scientia (supplément), xlvii (1930), 76-84.
Roberto Martins
Searching for the Ether
DIO 17
determined the relevant parameters from an analysis of the Leyden data, using the method of minimum squares. He obtained an effect corresponding to a speed of about 800 km/s in the direction of the Auriga constellation. This speed is, of course, much larger than the orbital speed of the Earth. Courvoisier interpreted it as due to the motion of the whole solar system through the ether. A few years later, Courvoisier obtained new data, using the same method (direct versus reflected direction). Using the vertical circle of the Babelsberg observatory, he made a long series of observations (1921-1922) that led to results similar to those that had been obtained from the Leyden observations.
After obtaining his first positive result, Courvoisier attempted to find other independent methods of measuring the speed of the Earth (or the solar system) relative to the ether. He conjectured that the Lorentz contraction of the Earth and of optical instruments could have some small observable influence on astronomical observations. According to Courvoisier, the motion of the Earth relative to the ether produces a contraction that transforms its spherical shape into an ellipsoid with the smaller axis in the direction of its motion. The surface of the ellipsoid, at each point, was supposed to be perpendicular to the local gravitational field. As the Earth rotates, each place on the surface of the Earth passes through different points of the ellipsoid, and the angle between the axis of the Earth and the local vertical direction should undergo a periodical change.
Of course, it is impossible to measure the angle between the local vertical and the axis of rotation of the Earth. However, since the direction of this axis is fairly constant relative to the fixed stars (for short time periods), it is possible to choose a star very close to the North celestial pole and to measure its distance to the zenith (that is, the local vertical direction). This angle, according to Courvoisier's theory, should undergo a periodical change, as a function of the sidereal time.
As a matter of fact, Courvoisier had already measured the position of a star very close to the North pole, in a long series of observations from 1914 to 1917, using the Babelsberg Observatory vertical circle.11 Those measurements were very accurate and were evenly distributed as regards the sidereal time of the observations. They were therefore suitable for looking for the influence of the Lorentz contraction on astronomical measurements.
As in the former case, Courvoisier first plotted the zenithal distances of the star against sidereal time, and found a regular fluctuation of the angle. He
11 Leopold Courvoisier, “Zenitdistanzbeobachtungen der Polarissima am Vertikalkreise der Stemwarte Berlin-Babelsberg”. Astronomische S'achnchlen. ccviii (1919), 349-64. He made this series of measurements as routine observations to ascertain the latitude of the Babelsberg observatory. The method used by Courvoisier is very precise, and was recently used for the determination of the azimuth of a transit instrument in Brazil: Ramachrisna Teixeira and Paulo Benevides Soares, “Absolute azimuth determination”, Astronomy and astrophysics, clxv (1986), 251-3.