A day of total luna-see: Of all the thousands of eclipses studied by scientists, the most important was the eclipse of 1919, writes Simon Singh. It provided clinching evidence in favour of one of the most revolutionary ideas in the history of physics - Einstein's theory of general relativity.
Although general relativity was a radically new formulation of gravity, its predictions were largely consistent with Newton's theory of gravity.
However, Einstein's theory did make one or two predictions that distinguished it from Newton's, and, if true, these predictions would show that Einstein's model was closer to reality.
Einstein predicted that a gravitational field should bend rays of light much more than was expected by Newton's theory.
Although the effect was too small to be observed in the laboratory, Einstein calculated that the immense gravity of the massive sun would deflect a ray of light by 1.75 seconds of arc - less than one thousandth of a degree, but twice as large as the deflection according to Newton, and significant enough to be measured.
Einstein pictured a scenario whereby the straight line of sight between a star and an observer on earth would be just blocked by the edge of the sun. He believed that the star would still be visible because gravity would bend the rays of light around the sun and towards the earth. The sighting of a star that should have been blocked by the sun would prove Einstein right, but it is generally impossible to see starlight that passes close to the sun, because it is swamped by the brilliance of the sun. But during eclipses, the sun is blacked out by the moon, and under such conditions a gravitationally distorted star should be visible.
General relativity was born in 1915 during the first world war, and as soon as the war ended, the astronomer royal Sir Frank Dyson began preparing for the next total eclipse, which would occur on May 29, 1919, and which would be an opportunity to test Einstein's theory.
He had already recruited Arthur Eddington, Plumian professor of astronomy at Cambridge, to make the observations. The son of devout Quakers, Eddington nearly spent the war as a conscientious objector peeling potatoes in an army camp, but instead Dyson arranged for a letter of deferment, which allowed him to carry on his astronomical research. In return, Eddington promised to make the trek to the island of Principe, off the coast of West Africa, one of the best locations for observing the 1919 eclipse.
The eclipse seemed almost too good to be true. Totality would last for 410 seconds, almost seven minutes - extraordinarily long for an eclipse - providing plenty of opportunity for measurements to be taken. Also, the eclipse would occur against the rich background of the Hyades constellation, increasing the likelihood of an appropriately positioned star.
But as the vital day approached, a cloud, or rather several, loomed on the horizon. It rained every day for the 19 days before the eclipse, and as it began on May 29, the sun was obscured by clouds. For 400 seconds the eclipse was hidden from view, and throughout this period Eddington prayed. Then with only ten seconds of the eclipse remaining, the skies miraculously cleared, and he was able to take just one meaningful photograph.
Eddington compared his eclipse photo with images taken when the sun was not present and announced that the sun had caused a deflection of roughly 1.61 seconds of arc, a result that was in agreement with Einstein's prediction, thereby validating the theory of general relativity. In recent years, scientists have questioned Eddington's margin of error, arguing that his equipment was not sufficiently accurate to discriminate between the predicted effects of the rival gravitational theories. In other words, Eddington believed in Einstein's theory and wanted to prove that it was true and therefore subconsciously minimised his errors in order to get the right result. Regardless of whether or not this was the case, Eddington's result was hailed as a wondrous piece of science.
J. P. McEvoy, author of Eclipse, encapsulates its significance: "A new theory of the universe, the brainchild of a German Jew working in Berlin, was confirmed by an English Quaker on a small African island."
Next week's eclipse will resonate with the one in 1919. The last of the millennium, it will pass over Einstein's birthplace, the German city of Ulm.
Simon Singh's The Code Book - the Science of Secrecy from Ancient Egypt to Quantum Cryptography is published in September by Fourth Estate.