The effect was first reported as an anomalous precession of the plane of oscillation of a Foucault pendulum during the solar eclipse of June 30, 1954 by Maurice Allais, a French polymath who later won the Nobel Prize in Economics.
[4] The veracity of the Allais effect remains controversial among the scientific community, as its testing has frequently met with inconsistent or ambiguous results over more than five decades of observation.
[6][7] Erwin Saxl and Mildred Allen similarly reported strong anomalous changes in the period of a torsion pendulum during the solar eclipse of March 7, 1970 and concluded that "gravitational theory needs to be modified".
[15] Results reporting the observation of the Allais and Jeverdan-Rusu-Antonescu effects during the annular solar eclipse of September 22, 2006 were presented the following year by a Romanian team, with a quantization of the behavior of the paraconical pendulum.
[17] The same teams repeated a dual experiment during the annular solar eclipse of January 26, 2009, this time outside of the umbra, with the same significant correlation between the behavior of light torsion balances and a Foucault pendulum.
[18] They also registered similar anomalies using a Foucault pendulum and a very light torsion balance, both located underground in a disused salt mine with minimal interference, during the partial solar eclipse of June 1, 2011.
NASA's Marshall Space Flight Center first inquired about experimental protocols to Maurice Allais,[5] in order to coordinate ahead of the event a worldwide effort to test the Allais effect between observatories and universities over seven countries (United States, Austria, Germany, Italy, Australia, England and four sites in the United Arab Emirates).
The lead supervisor then stated: "The initial interpretation of the record points to three possibilities: a systematic error, a local effect, or the unexplored.
"[22][23] However, after the eclipse, Allais criticized the experiments in his final NASA report, writing the period of observation was "much too short […] to detect anomalies properly".
[25] A possible yet controversial explanation was finally proposed by the same author and Tom Van Flandern which conjectured that the anomaly is due to the gravitational effect of an increased air density spot in the upper atmosphere created by cooling winds during the solar eclipse.
[30] Allais's explanation for another anomaly (the lunisolar periodicity in variations of the azimuth of a pendulum) is that space evinces certain anisotropic characteristics, which he ascribes to motion through an aether which is partially entrained by planetary bodies.
[34] Applying the Titius–Bode law to the Earth–Moon system, which he generalizes to aether, Allais calculates a "wind" of 7.95 km/s, which is comparable to the values found by the experiments of Michelson and Miller.