The Allais effect is a claimed anomalous precession of the plane of oscillation of a pendulum during a solar eclipse. It has been speculated to be unexplained by standard physical models of gravitation, but recent mainstream physics publications tend rather to posit conventional explanations for the reported observations.
The effect was first reported in 1954 by Maurice Allais, a French polymath who went on to win the Nobel Prize in Economics. He reported another observation of the effect during a 1959 solar eclipse.
Prof. Allais’s explanation for this and other anomalies is that space evinces certain anisotropic characteristics, which he ascribes to motion through an aether which is partially entrained by planetary bodies. He has presented this hypothesis in his 1997 French book L’Anisotropie de l’espace.
Exotic explanations for Allais and related effects have not gained significant traction amongst mainstream scientists.
A recently published observation of what was initially claimed to be a possibly related anomalous gravitational effect (claimed variation of terrestrial gravitation as measured by a sensitive gravimeter) was by Wang et al. in 2000, for an experiment carried out in 1997 in a remote region of China during a total solar eclipse. Further observations which the same team performed in 2001 and 2002 during solar eclipses in Zambia and Australia appear to have yielded much weaker evidence of similar anomalies. However, the same authors later (2002 and 2003) published papers explaining that their observations could be explained by conventional phenomena such as temperature and pressure change caused by the eclipse. In their 2002 paper, they posited a more conventional explanation based on temperature changes causing ground tilting, but they suggested that this explanation was unlikely. In their 2003 paper (Flandern and Yang, 2003), on the other hand, the authors argued that atmospheric motion induced by temperature changes was both plausible and sufficient to explain the observed anomaly.
Another anomalous effect during a solar eclipse, an increase in the period of a torsion pendulum, was reported by Saxl and Allen in 1970, but subsequent attempts to replicate this experiment (under different eclipse geometries and with much smaller pendulum bobs) failed to observe any effect (Kuusela, 1991; Jun, 1991). Jeverdan in Romania claimed to have observed anomalous pendulum behavior during a solar eclipse in 1961 (Jeverdan, 1981) – decrease of the period by about 1 part in 2000 – the so-called ‘Jeverdan effect’, but his report was not published in a mainstream English-language scientific journal.
A recent published article on the topic in a mainstream scientific journal (Flandern, 2003) concludes that there have been "no unambiguous detections [of an Allais effect] within the past 30 years when consciousness of the importance of [experimental] controls was more widespread." This paper also suggests a mechanism that might cause slight gravitational variations during an eclipse (high speed high-altitude winds). They point out that "the gravitation anomaly discussed here is about a factor of 100,000 too small to explain the Allais excess pendulum precession… during eclipses" and from this conclude that the original Allais anomaly was due to merely poor controls.
A possible but yet controversial explanation is the so-called van Flandern-Yang hypothesis by van Flandern and Yang's article,[1] which conjecture the effect is due to the gravitational effect of an increased density spot in the upper atmosphere created by cooling during the solar eclipse.
A self-published review article by Chris Duif, which surveys the field of gravitational anomalies in general, concludes that the question remains open, and that such investigations should be pursued, in view of their relatively inexpensive nature and the enormous implications if genuine anomalies are actually confirmed – but the article has not undergone any peer review.
Eight gravimeters and two pendulums were deployed across six monitoring sites in China for the solar eclipse of July 22, 2009. It is hoped that this effort will resolve the uncertainties surrounding this problem.[2]
An automated Foucault pendulum was used by H.R. Salva and found no evidence for a precession change of the pendulum's oscillation plane (<0.3 degree / hour)[3] during the Solar eclipse of July 11, 2010.