The Hunt for Vulcan

by Thomas Levenson

  So it happened, on the morning of March 20, 1862, a “Mr Lummis, of Manchester” stole a few minutes to peer at the sun through a small telescope. As the formal report in The Astronomical Register told it, Lummis was watching “between the hours of 8 and 9 A.M., when he was struck by the appearance of a spot possessed of a rapid proper motion.” The object was startling enough that Lummis called for a witness, and they “both remarked on its sharp circular form.” Lummis tracked the spot for twenty minutes before being called in to his day’s work. By the time he returned to his telescope, the object was gone, “but he has not the slightest doubt of the matter.” Radau and a colleague repeated the by-now familiar exercise, constructing the elements of an orbit from incomplete observations, and they found that Lummis’s potential Vulcan was at least compatible with Lescarbault’s, even if there wasn’t enough data to settle the matter once and for all.

  There were doubters. Two professional astronomers, the American Christian H. F. Peters and the German Gustav Spörer dismissed Lummis’s “discovery” as a mere sunspot. But for many others, Le Verrier among them, the ongoing identification of plausible Vulcans, in sightings that allowed for at least rough estimates of consistent trajectories, made an ultimate validation seem inevitable. By the mid-1860s, The Astronomical Register itself seemed to view the matter as settled, listing Vulcan (without stating whether it was Lescarbault’s object or some other) as the innermost body in its “Descriptive Account of the Planets.”

  —

  Matters soon grew more complicated, though. Reports of sightings continued to arrive, some from reputable observers, others from unknowns. In 1865, an otherwise completely obscure M. Coumbary wrote to Le Verrier with a detailed account of an observation he made in the city that he—an unreconstructed Byzantine, apparently—referred to as Constantinople. With his telescope in Istanbul he watched as a black spot separated itself from a group of sunspots and appeared to move independently. He continued to track the object for forty-eight minutes, until it vanished over the limb of the sun. Le Verrier endorsed Coumbary’s report, noting that though he didn’t know his correspondent, his information seemed to him to be marked by a combination of “exactitude and sincerity.” In 1869, a group of four eclipse mavens at St. Paul’s Junction, Iowa (one a lady, as contemporary records took pains to mention), saw “with the naked eye what they termed a little brilliant at a distance about equal to the Moon’s diameter from the Sun’s limb”—an object that at least two others (one equipped with a small telescope) seem to have noted as well.

  To those for whom the logical necessity of Vulcan was overwhelming, this spray of messages was comforting, not proof in and of itself, but an ongoing accumulation of information building on an already established pattern. The lack of a pure Neptune moment must have been frustrating, but given the inherent difficulty of the problem, such momentary glimpses gained significance each time another letter from some sincere and precise stranger reached Paris. As The New York Times put it, “a little scrap of positive evidence overbears an immense amount of negative.” But despite a growing heap of such hopeful wisps, Vulcan remained almost maliciously elusive when confronted by a systematic search.

  Benjamin Apthorp Gould had a perfect Boston pedigree: son of the headmaster of the Boston Latin School, grandson of a Revolutionary War veteran, he graduated from Harvard College—where else?—in 1844, all of nineteen years old. Then, having paid his debt to ancestry, he kicked over the traces. Heading to Europe, he took work at the Greenwich, Paris, and Berlin observatories just as Neptune made its (perceived) solar system debut. He studied math at the University of Göttingen, and in 1848 became the first American to receive a Ph.D. in astronomy—still only twenty-three! On returning to Boston in 1849, he was appalled by the primitive state of research in his home country, and took it on himself to transform American astronomy. Most important for the future of the discipline as a whole, in the 1860s he became one of the first investigators skilled in the new technique of astrophotography, the marriage of a camera to a telescope.

  Gould brought his cameras with him when he traveled to observe the same 1869 eclipse at which the amateurs had spied a possible Vulcan. He set up in the town of Burlington, Iowa, working on the right bank of the Mississippi River. His goal: to study the solar corona—the sun’s atmosphere, visible only during totality—and to survey the region close to the sun as precisely as possible, looking for whatever might reveal itself within the orbit of Mercury. He and his assistants made forty-two photographs during the eclipse. Gould also examined many of what he estimated were four hundred images made by others along the path of totality. In all those pictures, he saw—nothing.

  Astrophotographers from the Canadian Eclipse Party, at their observing station in Iowa in August 1869.

  Gould sent his findings to Yvon Villarceau at the Paris Académie. He began with a baseline estimate: in the shadow of the eclipse, a planet or planets substantial enough to account for Mercury’s motion should shine about as brightly as Polaris, the North Star, a second magnitude object—easily seen by the naked eye.* His photographic equipment, Gould wrote, was sensitive enough to detect any object down to the limit of unaided human perception, well below what he considered the plausible threshold for the discovery of Vulcan. Thus, he concluded, “I am convinced that this investigation dispenses with the hypothesis that the movement of the perihelion of Mercury results from the effects of one or many small interior planets.” I’ve looked, he said, and Vulcan ain’t there.

  Not so fast, though: Villarceau added a note of his own to the published version of Gould’s letter. It wasn’t necessary to accept the American’s conclusion as absolute, he argued. There were configurations of asteroids, for example, that could both provide the necessary gravitational influence on Mercury and evade detection. In other words: the problem remained. Mercury still wobbled, and in Newton’s cosmos, its motion still demanded something like a Vulcan. Absence of evidence, to invoke what has become a cliché, could not be taken as evidence of absence.

  Others agreed. William F. Denning was by general agreement Victorian Britain’s greatest amateur astronomer. He had made his reputation with the first comprehensive analysis of the motion of the Perseid meteor shower, still to be seen from late July to its peak in mid August, and meteors remained his primary obsession. Vulcan, though, was a sufficiently pressing problem to draw his attention. He was an obligate organizer, and he used his influence to launch a systematic search for solar transits during the next likely window: March and April of 1869. He persuaded fifteen other sky-watchers to put the sun “continually under observation, when visible…with a view of rediscovering the suspected intra-Mercurial planet Vulcan.”

  Vulcan obstinately refused to appear.

  Denning tried again the next year, recruiting a team of twenty-five to chase the elusive planet during the spring transit season in 1870, and yet once more with a plea to collaborators in 1871. As he gathered his volunteers, he had declared that his aim was to settle the issue once and for all. “There is every reason,” he wrote, “to suppose that the search will end satisfactorily, if not successfully.” End it did. After three conscientious attempts at locating the missing planet, he seems to have concluded that there was nothing more to be done. He did not repeat his call for aid on the search, and those fellow amateurs of the sky who had responded to him were released to their prior ambitions.

  —

  After what was to that point the largest systematic search for the object since word of Lescarbault’s sighting first spread, Denning’s null result left Vulcan in a predicament. An explanation for Mercury’s errant motion remained necessary. On one side of the ledger, there was the blunt fact of Le Verrier and his genuine abilities. No one doubted his calculation, and no one should have—a restudy of Mercury’s perihelion advance in the 1880s confirmed and slightly enlarged the very real anomaly he identified. Glimpse after glimpse of possible candidate planets offered tantalizing hints—yet a decade into the search, the
most rigorous observers kept coming up empty. What could be done?

  A way out was obvious to the more mathematically sophisticated Vulcan hunters. People simply could have gotten their sums wrong. There were enough imprecise assumptions about the elements of a putative Vulcan’s orbit so that calculations for transits could just be wrong. Princeton’s Stephen Alexander told his fellow members of the National Academy of Sciences that he had reworked Vulcan’s elements to arrive at the conclusion that there should be “a planet or group of planets at a distance of about twenty-one million miles from the sun, and with a period of 34 days and 16 hours.” In other words: we may have been looking in the wrong places, or at the wrong times. Vulcan could be elusive, but not absent.

  That claim seemed to be confirmed when Heinrich Weber—for once, an actual well-trained professional astronomer—sent word from northeast China that he had seen a dark circular shape transit the sun on April 4, 1876. Sunspot expert and Vulcan devotee Rupert Wolf passed word of his colleague’s sighting on to Paris, taking a bit of a victory lap as he did so. He told Le Verrier that “the interval between Lescarbault’s observation and Weber’s amounts to exactly one hundred and forty eight times the period” that Wolf had calculated so many years before.

  The news enthralled Le Verrier—and energized yet another corps of planet seekers more eager than expert. As historian Robert Fontenrose put it, “everyone with a telescope was looking for Vulcan; some found it.” For a time, Scientific American eagerly trumpeted each new “discovery”: from “B. B.” in New Jersey to a Samuel Wilde in Maryland, to W. G. Wright in San Bernardino, to witnesses from beyond the grave, in the form of a minister who remembered that Professor Joseph S. Hubbard “had repeatedly assured him he had seen Vulcan with the Yale College Telescope.” New Vulcans kept turning up that autumn in seemingly every mail delivery, until at last Scientific American cried “Uncle!” and, following its December 16, 1876, issue, declined to publish any more such happy memories. It was as if the question of Vulcan had ridden a seesaw since 1859. Occasional sightings and seemingly consistent calculations would propel it up to the top of the ride; hard-nosed attempts to verify its existence sent it crashing back down. Now, for all that the editors of Scientific American had tired of the flood of anecdotes, the teeter-totter was pointing up: between the one seemingly authoritative report from China and the sheer number, if not the quality of sky-gazer accounts, the matter of Vulcan seemed just about settled.

  The popular press certainly thought so. In late 1876, The Manufacturer and Builder said, “Our text books on astronomy will have to be revised again, as there is no longer any doubt about the existence of a planet between Mercury and the sun.” That autumn, The New York Times was even less bashful, interrupting its coverage of the Hayes-Tilden presidential election to assert that any residual doubts about the intra-Mercurian planet could be put down to simple professional jealousy: “ ‘Vulcan may possibly exist,’ said the conservative astronomers, ‘but Professor So and So never saw it…’ ”—pure us-against-them nastiness, according to the Times, adding “they would hint, with sneering astronomic smiles, that too much tea sometimes plays strange pranks with the imagination.”

  Now, such too-smart fellows were about to receive their due, the newspaper proclaimed. Why? Because, in the wake of Weber’s report, the grand old man himself, Urbain-Jean-Joseph Le Verrier, had roused himself. “The man who untied Neptune with his nose—so to speak—cannot be accused of confounding accidental flies with actual planets. When he firmly asserts that he has not only discovered Vulcan, but has calculated its elements, and arranged a transit especially for its exhibition to routing astronomers…” the Times wrote, “there is an end of all discussion. Vulcan exists…”

  The Times got at least one thing right. After shifting his attention to other problems for a few years, Le Verrier had indeed returned to the contemplation of Vulcan. Wolf’s news had fired his passion for the planet, and he began a comprehensive reexamination of everything that might bear upon its existence. Starting with yet another catalogue of claimed sightings dating back to 1820, he identified five observations spread from 1802 to 1862 that seemed to him most likely to represent repeat glimpses of a single planet. That allowed him to construct a new theory for the planet, complete with the prediction the Times had rated so high: a transit that could perhaps be observed, Le Verrier suggested, on October 2nd or 3rd.

  The headline writers would be disappointed. Vulcan did not cross the face of the sun in early October. More confounding, Weber’s revelation from China was debunked: two photographs made at the Greenwich Observatory clearly revealed his “Vulcan” to be just another sunspot. Scientific American called this the “coup de grace” for this latest “discovery,” but, as usual in the annals of Vulcan, its real impact was more deflating than destructive. Le Verrier’s calculation turned on earlier observations, not Weber’s, and there was a way to explain away the missed transit, by positing an orbit for Vulcan that was much more steeply inclined than previously assumed. Thus Le Verrier hedged his bets: there might be a chance to see Vulcan against the face of the sun in the spring of 1877, but given the full range of possible orbits this insufferably errant planet might occupy, it might be five years or more before the next transit would occur.

  —

  No transits occurred that March. Le Verrier said nothing more in public about Vulcan. He had turned sixty-six on March 11, and he was tired to the bone. As the year advanced, he found he couldn’t drag himself to the weekly meetings of the Académie, nor to his daily post at the Observatory. Time off seemed to help—he returned to his desk in August—but fatigue masked his real trouble: liver cancer.

  On the evidence, Le Verrier was not a religious man. He did accept communion in late June on the urging of a much more committed Catholic colleague, but that seems to have been the limit of his willingness to acknowledge conventional pieties. By summer’s end, he could no longer mistake his illness. The end came on September 23rd—forty-one years to the day since young Johann Gottfried Galle had sought and found Neptune in the night sky above Berlin.

  Le Verrier left the solar system larger than he found it—one both better and less completely understood. Of Vulcan itself, though—surely, given all the fully satisfactory explanations for the behavior of every other astronomical object derived from the Newtonian synthesis, the fault, it seemed so nearly certain, must lie not in the stars, but in some human failure to crack this one particular mystery.

  * * *

  * The magnitude scale for celestial objects dates back to Greek astronomers, and was based on a rough visual distinction between the brightness of different stars. It was originally calibrated on Polaris, which was assigned a magnitude of exactly two. Smaller (and negative) numbers are brighter. The sun as seen from Earth has a magnitude of –26. In the modern definition, a magnitude one star—like Antares or Spica—is one hundred times brighter than the magnitude of six objects that lie at the edge of human naked-eye perception.

  “SO LONG ELUDING THE HUNTERS”

  July 24, 1878, Rawlins, Wyoming.

  The man from New Jersey had heard stories about the mythical West, but this was his first chance to compare the tales with the real thing—at least, as much of it as he could now observe it from the comfort of a rail carriage. His trip to Rawlins, on the Union Pacific line that ran to and through the little town, had thus far revealed only a sightseer’s version of the frontier. “The country at that time was rather wild,” he wrote. “Game was in great abundance and could be seen all day long from the car window, especially antelope.”

  It all got a little more up-close-and-personal at the hotel. He and his roommate had settled down for the night when “A thundering knock on the door awakened us; on opening the door a tall handsome man in western style entered the room.” That visitor—on inspection, not entirely sober—introduced himself as Texas Jack. The hotel owner arrived and tried to persuade Jack to keep it down. He was bounced across the hall for his pains. C
almly, Jack “explained he was the top pistol shot of the West…and then suddenly pointing to a weather vane on the freight depot, pulled out a Colt, revolved and fired through the window, hitting the vane.”

  Other guests swarmed into the room to see who’d just been killed. With no corpse in sight, calm, of a sort, soon returned, and it became clear that the Texan just wanted to talk. At last, by promising to make time for him in the morning, Jack was persuaded to pack it in, and the two strangers returned to their beds.

  But not to sleep. They left this first true encounter with the West of legend “rather scared,” and, understandably, unsure “what would be the result of this interview.” Nothing that had followed reassured them, to the point that they became “so nervous we did not sleep any that night.” The next morning the travelers were relieved to find out that around town Texas Jack “was not one of the ‘badmen’ of whom they had a good supply.” Thus reassured, the traveler from New Jersey could focus on the business that had brought him west. The great eclipse of 1878 would be coming to Rawlins in five days’ time, and a tribe of visiting scientists were already racing to prepare. Among them, there to test his latest invention, came the man Texas Jack couldn’t wait to meet: Mr. Thomas Edison.

  —

  The eclipse of July 29, 1878, followed a route that stretched from Siberia across the Bering Strait into Alaska, then down through western Canada. From there it traversed the United States through the northern Rockies across Wyoming, before heading southeast into the Gulf of Mexico, finally coming to an end just east and south of Haiti. The total phase of the eclipse—the time the moon completely blocks the surface of the sun, thus revealing both the wispy, achingly beautiful corona and any faint celestial bodies close to the solar limb—had a maximum duration of about three minutes, eleven seconds, achieved in Siberia.*1 In Rawlins, totality would extend just two minutes and fifty-six seconds, but the site had one critical advantage: the decade-old transcontinental railroad ran neatly along the eclipse route, which meant astronomers and their bulky, awkward equipment could ride in unprecedented luxury to (hopefully) perfect vantage points.