The problem with verifying general relativity is that its effects on Earth and in the solar system—the places where we can most easily perform tests—are very small. Just as special relativity produces major departures from Newtonian mechanics only when velocities approach the speed of light, general relativity predicts large departures from Newtonian gravity only when extremely strong gravitational fields are involved—in effect, when orbit speeds and escape velocities become relativistic.

We will encounter other experimental and observational tests of general relativity elsewhere in this chapter. Here we consider just two "classical" tests of the theory. These tests are solar system observations that helped ensure acceptance of Einstein's theory. Later, more accurate measurements confirmed and strengthened these results. Bear in mind, however, that there are no known tests of general relativity in the "strong-field" regime—that part of the theory that predicts black holes, for example—so the full theory has never been experimentally tested.

At the heart of general relativity is the premise that everything, including light, is affected by gravity because of the curvature of spacetime. Shortly after he published his theory in 1915, Einstein noted that light from a star should be deflected by a measurable amount as it passes the Sun. The closer to the Sun the light comes, the more it is deflected. Thus, the maximum deflection should occur for a ray that just grazes the solar surface. Einstein calculated that the deflection angle should be 1.75"—a small, but detectable, amount. Of course, it is normally

In 1919 a team of observers, led by the British astronomer Sir Arthur Eddington, succeeded in measuring the deflection of starlight during an eclipse. The results were in excellent agreement with the prediction of general relativity. Virtually overnight Einstein became world famous. His previous major accomplishments notwithstanding, this single prediction assured him a permanent position as the best-known scientist on Earth! Recently, the high-precision Hipparcos satellite (Interlude 17-1) has observed shifts in the apparent positions of many stars, even those whose line of sight is far from the Sun. The shifts are exactly as predicted by Einstein's theory.

Another prediction of general relativity is that planetary orbits should deviate slightly from the perfect ellipses of Kepler's laws. Again, the effect is greatest where gravity is strongest—that is, closest to the Sun. Thus, the largest relativistic effects are found in the orbit of Mercury. Relativity predicts that Mercury's orbit is not a closed ellipse. Instead, its orbit should rotate slowly, as shown in the (highly exaggerated) diagram at right. The amount of rotation is very small—only 43" per century—but Mercury's orbit is so well charted that even this tiny effect is measurable.

In fact, the observed rotation rate is 574" per century, much greater than that predicted by relativity. However, when other (nonrelativistic) gravitational influences, primarily the perturbations due to the other planets, are taken into account, the rotation is in complete agreement with the foregoing prediction.