Total solar ecplise: could the moon shield the sun's gravitational pull? Image © NASA

Allegedly, Isaac Newton got the inspiration for his description of the force of gravity from a falling apple. A few centuries later, Einstein showed that reality was a little more complicated than Newton's (apocryphally) fruit-inspired theory. Now new measurements continue the trend, implying that gravity could yet have a few surprises in store.

Despite its role in shaping the Universe, we do not really understand the force of gravity. General relativity, the theory developed by Einstein, remains a cornucopia of subtle possibilities, all of them very difficult to confirm or rule out by measurement. These difficulties are basically due to the weakness of gravity: it takes huge masses like planets and stars to generate significant forces.

Enter a huge mass: the Moon coming between the Earth and the Sun, during a total solar eclipse in March 1997 in Helongjiang Province, China. Researchers Qian-shen Wang and colleagues of the Chinese Academy of Sciences, Beijing, China, went looking for the effects of the eclipse on the 'acceleration due to gravity'. They report what they found in Physical Review D1.

Galileo (in an episode about as historically reliable as Newton's falling apple) saw that two stones dropped from the top of the Leaning Tower of Pisa fell with the same acceleration: the acceleration due to gravity, or '_g_' for short.

But Wang's group was testing the idea that the mutual gravitational attraction of two objects could actually be 'shielded' by a third object placed between them. If so, when the Moon comes between the Sun and the Earth as in a total solar eclipse, the gravitational pull between the Sun and the Earth, measured via g, should be reduced in a way not predicted by existing gravity theory.

Wang's group did indeed see a decrease in g, just at the time of the eclipse. They claim that the measured change in g, even when corrected for all the effects (like tides) predicted by existing theory, was still too big to be explained by errors in their instruments.

During the eclipse the measured g changed by just less than one millionth of a per cent of its usual value. This might sound insignificant. But that's the catch-22 of looking for quirks in gravity. The weakness of gravity means any anomal y will be very small, while at the same time it has to be big enough to be measured.

Which is why new claims of gravitational peculiarities have cropped up repeatedly over the centuries. Indeed, they are something of a cottage industry2. Reported oddities range from gyroscopes that weigh less the faster they spin, to superconductors lowering the weight of nearby objects. And, it has to be said, many of these measurements turn out to be doubtful.

Wang's team cautiously admits that their new measurements do not prove the idea of gravitational 'shielding'. According to gravity expert George Gillies of the University of Virginia, "there have been several other searches for a shielding effect -- but all with negative results." Nevertheless, the results are curious and, for now, difficult to explain.