Books by Levenson, Thomas

Levenson, Thomas. The Hunt for Vulcan. New York: Random House, 2015. ISBN 978-0-8129-9898-6.

The history of science has been marked by discoveries in which, by observing where nobody had looked before, with new and more sensitive instruments, or at different aspects of reality, new and often surprising phenomena have been detected. But some of the most profound of our discoveries about the universe we inhabit have come from things we didn't observe, but expected to.

By the nineteenth century, one of the most solid pillars of science was Newton's law of universal gravitation. With a single equation a schoolchild could understand, it explained why objects fall, why the Moon orbits the Earth and the Earth and other planets the Sun, the tides, and the motion of double stars. But still, one wonders: is the law of gravitation exactly as Newton described, and does it work everywhere? For example, Newton's gravity gets weaker as the inverse square of the distance between two objects (for example, if you double the distance, the gravitational force is four times weaker [2² = 4]) but has unlimited range: every object in the universe attracts every other object, however weakly, regardless of distance. But might gravity not, say, weaken faster at great distances? If this were the case, the orbits of the outer planets would differ from the predictions of Newton's theory. Comparing astronomical observations to calculated positions of the planets was a way to discover such phenomena.

In 1781 astronomer William Herschel discovered Uranus, the first planet not known since antiquity. (Uranus is dim but visible to the unaided eye and doubtless had been seen innumerable times, including by astronomers who included it in star catalogues, but Herschel was the first to note its non-stellar appearance through his telescope, originally believing it a comet.) Herschel wasn't looking for a new planet; he was observing stars for another project when he happened upon Uranus. Further observations of the object confirmed that it was moving in a slow, almost circular orbit, around twice the distance of Saturn from the Sun.

Given knowledge of the positions, velocities, and masses of the planets and Newton's law of gravitation, it should be possible to predict the past and future motion of solar system bodies for an arbitrary period of time. Working backward, comparing the predicted influence of bodies on one another with astronomical observations, the masses of the individual planets can be estimated to produce a complete model of the solar system. This great work was undertaken by Pierre-Simon Laplace who published his Mécanique céleste in five volumes between 1799 and 1825. As the middle of the 19th century approached, ongoing precision observations of the planets indicated that all was not proceeding as Laplace had foreseen. Uranus, in particular, continued to diverge from where it was expected to be after taking into account the gravitational influence upon its motion by Saturn and Jupiter. Could Newton have been wrong, and the influence of gravity different over the vast distance of Uranus from the Sun?

In the 1840s two mathematical astronomers, Urbain Le Verrier in France and John Couch Adams in Britain, working independently, investigated the possibility that Newton was right, but that an undiscovered body in the outer solar system was responsible for perturbing the orbit of Uranus. After almost unimaginably tedious calculations (done using tables of logarithms and pencil and paper arithmetic), both Le Verrier and Adams found a solution and predicted where to observe the new planet. Adams failed to persuade astronomers to look for the new world, but Le Verrier prevailed upon an astronomer at the Berlin Observatory to try, and Neptune was duly discovered within one degree (twice the apparent size of the full Moon) of his prediction.

This was Newton triumphant. Not only was the theory vindicated, it had been used, for the first time in history, to predict the existence of a previously unknown planet and tell the astronomers right where to point their telescopes to observe it. The mystery of the outer solar system had been solved. But problems remained much closer to the Sun.

The planet Mercury orbits the Sun every 88 days in an eccentric orbit which never exceeds half the Earth's distance from the Sun. It is a small world, with just 6% of the Earth's mass. As an inner planet, Mercury never appears more than 28° from the Sun, and can best be observed in the morning or evening sky when it is near its maximum elongation from the Sun. (With a telescope, it is possible to observe Mercury in broad daylight.) Flush with his success with Neptune, and rewarded with the post of director of the Paris Observatory, in 1859 Le Verrier turned his attention toward Mercury.

Again, through arduous calculations (by this time Le Verrier had a building full of minions to assist him, but so grueling was the work and so demanding a boss was Le Verrier that during his tenure at the Observatory 17 astronomers and 46 assistants quit) the influence of all of the known planets upon the motion of Mercury was worked out. If Mercury orbited a spherical Sun without other planets tugging on it, the point of its closest approach to the Sun (perihelion) in its eccentric orbit would remain fixed in space. But with the other planets exerting their gravitational influence, Mercury's perihelion should advance around the Sun at a rate of 526.7 arcseconds per century. But astronomers who had been following the orbit of Mercury for decades measured the actual advance of the perihelion as 565 arcseconds per century. This left a discrepancy of 38.3 arcseconds, for which there was no explanation. (The modern value, based upon more precise observations over a longer period of time, for the perihelion precession of Mercury is 43 arcseconds per century.) Although small (recall that there are 1,296,000 arcseconds in a full circle), this anomalous precession was much larger than the margin of error in observations and clearly indicated something was amiss. Could Newton be wrong?

Le Verrier thought not. Just as he had done for the anomalies of the orbit of Uranus, Le Verrier undertook to calculate the properties of an undiscovered object which could perturb the orbit of Mercury and explain the perihelion advance. He found that a planet closer to the Sun (or a belt of asteroids with equivalent mass) would do the trick. Such an object, so close to the Sun, could easily have escaped detection, as it could only be readily observed during a total solar eclipse or when passing in front of the Sun's disc (a transit). Le Verrier alerted astronomers to watch for transits of this intra-Mercurian planet.

On March 26, 1859, Edmond Modeste Lescarbault, a provincial physician in a small town and passionate amateur astronomer turned his (solar-filtered) telescope toward the Sun. He saw a small dark dot crossing the disc of the Sun, taking one hour and seventeen minutes to transit, just as expected by Le Verrier. He communicated his results to the great man, and after a visit and detailed interrogation, the astronomer certified the doctor's observation as genuine and computed the orbit for the new planet. The popular press jumped upon the story. By February 1860, planet Vulcan was all the rage.

Other observations began to arrive, both from credible and unknown observers. Professional astronomers mounted worldwide campaigns to observe the Sun around the period of predicted transits of Vulcan. All of the planned campaigns came up empty. Searches for Vulcan became a major focus of solar eclipse expeditions. Unless the eclipse happened to occur when Vulcan was in conjunction with the Sun, it should be readily observable when the Sun was obscured by the Moon. Eclipse expeditions prepared detailed star charts for the vicinity of the Sun to exclude known stars for the search during the fleeting moments of totality. In 1878, an international party of eclipse chasers including Thomas Edison descended on Rawlins, Wyoming to hunt Vulcan in an eclipse crossing that frontier town. One group spotted Vulcan; others didn't. Controversy and acrimony ensued.

After 1878, most professional astronomers lost interest in Vulcan. The anomalous advance of Mercury's perihelion was mostly set aside as “one of those things we don't understand”, much as astronomers regard dark matter today. In 1915, Einstein published his theory of gravitation: general relativity. It predicted that when objects moved rapidly or gravitational fields were strong, their motion would deviate from the predictions of Newton's theory. Einstein recalled the moment when he performed the calculation of the motion of Mercury in his just-completed theory. It predicted precisely the perihelion advance observed by the astronomers. He said that his heart shuddered in his chest and that he was “beside himself with joy.”

Newton was wrong! For the extreme conditions of Mercury's orbit, so close to the Sun, Einstein's theory of gravitation is required to obtain results which agree with observation. There was no need for planet Vulcan, and now it is mostly forgotten. But the episode is instructive as to how confidence in long-accepted theories and wishful thinking can lead us astray when what might be needed is an overhaul of our most fundamental theories. A century hence, which of our beliefs will be viewed as we regard planet Vulcan today?

January 2016