In commonly occurring situations, predictions made by general relativity and Newton's theory vary only slightly. This is a good thing because Newton's theory had plenty of experimental evidence to back it up. If Einstein's theory had been significantly different, it would have been sunk from the beginning. The bad news was that this made it difficult to prove that general relativity was a better theory. They needed a test for it, and Einstein himself proposed it.

General relativity predicts that the presence of the sun distorts the path of light that passes near it. Because the light from the sun overpowers light from stars in the daytime, this would normally not be possible to see. During a solar eclipse, however, you can see star light. Einstein predicted that light from a star that travels very close to the sun would be bent approximately 0.00049 of a degree. An expedition was sent to the island Principe off the coast of West Africa to test his prediction. They took pictures of a solar eclipse on May 29, 1919. For months they worked, comparing the apparent position of the star that day with it real position known from other observations at night. They announced the confirmation of Einstein's prediction at a meeting of the Royal Society and Royal Astronomical Society in England on November 6, 1919. The accuracy of this experiment was later under scrutiny, but other predictions since leave no doubt that general relativity makes more accurate predictions than Newton's universal theory of gravity.