Starlight Detectives

by Alan Hirshfeld

  The Andromeda Nebula, in a four-hour exposure by Isaac Roberts on December 29, 1888.

  An even more stunning four-hour exposure was featured in the mass-circulation magazine Knowledge in February 1889. Its editor, A. Cowper Ranyard, pointed out one detail astronomers had evidently missed: “Lines of small stars shown in Mr. Roberts’s photographs lie along the edges of the dark rifts following all their sinuosities. . . . [O]ne hardly knows whether to describe them as minute stellar points or as regions of greater nebulous brightness.” These stars, Ranyard implies, are not foreground bodies seen in projection, but clearly belong to the spiral nebula itself. At least some fraction of Andromeda’s light springs from stellar sources. Could it be that its broader glow is the “united luster of millions of stars,” as the famous eighteenth-century observer William Herschel suggested?

  The controversy over the nature of nebulae had simmered since their earliest discovery. One theory posed that nebulae are gaseous aggregations scattered among the stars of our Milky Way. Spiral nebulae, with their distinctive whirlpool form, elicited visions of matter cascading inward and condensing into new planetary systems, consistent with the so-called nebular hypothesis proposed by Pierre Simon de Laplace in 1796. The photographs of Andromeda, Isaac Roberts announced, were a clear demonstration of Laplace’s idea: “Here one might see a new solar system in process of condensation from the nebula.” Roberts bolstered his Laplacian vision in April 1889 with a crisp, four-hour exposure of the Whirlpool Nebula, the definitive face-on spiral.

  The alternative theory of the nebulae alleged that at least some of these supposed clouds of gas are “island universes,” distant Milky Ways whose remoteness renders their stars irresolvable. The observations of pioneering English spectroscopist William Huggins only fueled the dispute. Huggins confirmed in 1864 that spectra of diffuse nebulae generally mimic features seen in a spectrum of a tenuous, incandescent gas in a laboratory. However, he subsequently found that spectra of spiral nebulae, notably Andromeda, indicate a stellar origin—that is, the collective light of individual, unresolved stars. And if Andromeda could hide its spirality through a simple accident of orientation, how many other covert spirals might there be in the night sky? The competing beliefs about the spiral nebulae posed very different concepts of the universe: one, a lone assemblage of stars—the Milky Way—surrounded by a void; the other, a vast population of star systems like our own, strewn throughout what might be an infinite space.

  In 1893, Roberts published a photographic album of deep-space species, a first step in realizing Andrew Common’s dream of a new kind of “library, not of books full of descriptions and figures . . . but of pictures written on leaves of glass by the stars themselves.” Even the most ardent visual astronomers had to acknowledge the practical import of these accumulating chemical masterpieces: in probing nebulae, the eye had become the poor stepchild to the photographic plate. In his 1888 memoir of Henry Draper, University of Pennsylvania physicist George Barker writes, “[T]he facility of reproduction by photographic means so far surpasses that by drawing or sketching, and is, moreover, so much more accurate a method of delineation, that the evidence given by an untouched photograph is everywhere accepted as prima facie proof.”

  The Dumbbell Nebula, in a three-hour exposure by Isaac Roberts on October 3, 1888.

  The camera—and only the camera—had the ability to disclose Andromeda’s true form. And, going forward, the camera would be the essential agent of nebular imaging. Yet before the human eye was shoved entirely out of the picture, the institutional elite had to be convinced that photography was relevant to their work. The majority of these professionals were based at universities and government-sponsored timekeeping facilities. Like generations of forebears in Britain, France, and Germany, they were practitioners of so-called exact astronomy: the mathematical analysis of positions and movements of celestial bodies. To these number crunchers, nebular exposures were “soft” data, the antithesis of the quantitative ledgers they patiently compiled and scrutinized. To add a decimal place of precision to a star’s position was a badge of honor within this fraternity. Their commitment to mathematical rigor and clockwork accuracy was born of tradition dating back more than a century. Positional astronomers were not about to suspend ongoing visual projects and tramp over to the observatory with a camera in hand. At least not until 1882. In that year, a single photograph, taken on a whim by one of their own, made them see the light.

  Chapter 11

  THE GRANDEST FAILURE

  If we could first know where we are, and whither we are tending, we could better judge what to do, and how to do it.

  —Abraham Lincoln, Illinois Republican Convention, June 16, 1858

  DAVID GILL WAS ASTONISHED by the brilliance of the comet as it hovered above the mountaintops of False Bay, opposite the Cape of Good Hope. Nothing in his decade-long career, first as director of the Dun Echt Observatory in Scotland and now as Her Majesty’s Astronomer at the Royal Observatory in South Africa, had approached the lustrous majesty of this interplanetary visitor. Discovered in the predawn hours of September 8, 1882, by his chief assistant, William H. Finlay, the comet had brightened over the past week and a half until it was visible in broad daylight, even near the Sun’s edge. (Andrew Common independently sighted the comet on the morning of September 17 from the Northern Hemisphere, just hours before it rounded the Sun’s fiery surface.)

  David Gill.

  A watchmaker’s son, David Gill was an astronomer of the “rigorously orthodox kind,” charged with meticulous measurement of positions of celestial bodies in the southern sky. As to why delineation of the heavens is important, Gill once explained that if, “of two points marking a frontage boundary on Cornhill, one were correct, the other 10 feet in error, what a nice fuss there would be! what food for lawyers! what a bad time for the Ordnance Survey Office! Well, it is just the same in astronomy.” Like many positional astronomers, Gill obsessed over the detection and correction of telescopic deficiencies. He advised a colleague, “[H]owever perfect an instrument may be (and it is the astronomer’s business to see that it is perfect), it is the astronomer’s further business to look upon it with complete and utter mistrust.” Gill was no stone-faced automaton. He admitted that astrometry—celestial position measurement—offered “no dreamy contemplation, no watching for new stars, no unexpected or startling phenomena.” Yet he reveled in the grinding routine of the observatory, frequently cutting the monotony with a pipe or cigar. He counted himself among the intrepid observers who “betake themselves to bed, tired, but (if they are of the right stuff) happy and contented men.” Many nights, Isobel Gill recalled, her husband arrived home singing.

  David Gill’s telescope of choice was an unconventional, split-objective refractor called a heliometer, whose lens-halves could be offset by means of a screw to reveal the separations between objects in the eyepiece. The instrument’s complexity confounded all but a handful of astronomers, and Gill was acknowledged to be the world’s expert in its use. His greatest claim to fame was his 1877 determination of the Sun’s distance, which became the worldwide standard into the twentieth century.

  When not immersed in the statistical minutiae of observing and data reduction—or in the tennis and hockey contests he loved to play with the observatory’s staff—Gill would plead with the Admiralty to increase the meager funding for its Southern Hemisphere station. Yet in the morning twilight of September 18, 1882, one day after perihelion, the beck of research, money, and sport faded as the billowing comet awakened Gill’s poetic muse. “There was not a cloud in the sky,” he reported, “only a merging into a rich yellow that fringed the blackish blue of the distant mountains, and over the mountains and amongst the yellow an ill-defined mass of golden glory rose with a beauty I cannot describe. The Sun rose a few minutes afterward, but to my intense surprise the comet seemed in no way dimmed in brightness. . . . I left Simon’s Bay and hurried back to the observatory, pointing out the comet in broad daylight to the friends I
met along the way.”

  An observer in India likened the comet’s graceful arc to that of an elephant’s tusk. By October, the tail lengthened to fifteen degrees, and took an hour to fully rise above the horizon. A few enthusiasts managed to photograph the comet using tripod-mounted cameras, although the nucleus was inevitably smeared out by Earth’s rotation. (The spurious elongation happened to align with the comet’s tail, masking the flaw; nevertheless, at least one amateur retouched the diaphanous tail with a paint brush.) Hearing of these images stirred another dormant muse in David Gill. In 1869, back in his native Scotland during the wet-collodion era, he had dabbled in lunar photography with a second-hand reflector telescope. It was, in part, the acclaim following one of these pictures that had led to his appointment as director of Lord Lindsay’s Dun Echt Observatory. Now, with the spectacular comet gracing his southern sky, Gill suspended his heliometer studies for what would become a brief but momentous return to celestial photography.

  Without a camera or any practice in the new dry-plate process, Gill had no choice but to follow the example of his Harvard predecessor William Bond during the daguerreotype era: he sought the help of a professional photographer. E. H. Allis, from the nearby village of Mowbray, proved to be as able and enthusiastic a collaborator as John Whipple had been to Bond thirty-five years before. On October 19, 1882, Allis arrived at the observatory with his portrait camera, which Gill mounted on a stout board and clamped to the counterweight of a six-inch refractor: the telescope became a high-power sight for the camera. Allis operated the camera shutter, while Gill adjusted the slow-motion controls to keep the comet’s nucleus centered on a pair of crossed spider webs at the telescope’s focus. Six photographs were taken during October and November, with exposures ranging from thirty minutes to two hours and twenty minutes. The best of the pictures showed the comet in sharp focus, with structural details evident in both its tail and its nucleus. In a confluence of human invention and cosmic clockwork, a technology had matured in time to record one of the brightest comets of the modern era.

  Yet even as Gill rushed his comet portraits to colleagues in England and France, he was drawn to an incidental—and wholly unanticipated—feature of the photographs: the profusion of background stars. The faintest of these he estimated at tenth magnitude, some forty times dimmer than the minimum light detectable by the unaided eye. What amazed Gill was that the stars speckling his comet pictures had been recorded with a studio camera whose aperture was a mere two-and-a-half inches. (In fact, photographic pioneer Warren De La Rue had imaged the Pleiades star cluster with a portrait camera in 1861, during the wet-plate era.) Tenth magnitude also happened to coincide with the brightness cutoff of the famed Bonner Durchmusterung (BD), a visual census of more than 324,000 Northern Hemisphere stars compiled by midcentury German astronomers. An extension to just below the celestial equator raised that number to 458,000.

  In 1850, a relatively small number of stars had classical names or other designations; the multitude of fainter stars were anonymous. The need for unique stellar identifiers and accurate sky charts had become critical. Already a dozen asteroids had been discovered, whose orbits could be derived only by tracking their movement against a well-defined grid of stars. Variable stars had to be sorted out from their nonvariable neighbors, so they could be monitored for changes in brightness. “Stars not entered in [the BD] have no official existence,” Agnes Clerke wrote in 1888. “Should they fade and vanish, the fact cannot be attested; should they brighten into conspicuousness, we are obliged to regard them as ‘new’ for lack of previous acquaintanceship. Whatever is known of the distribution of stars in space is founded on this grand enumeration.”

  The Great Comet of September 1882, photographed by David Gill at the Royal Observatory, Cape of Good Hope.

  With his own years of telescopic experience, David Gill was no stranger to the squint-eyed fatigue of visual star-position measurement. (The BD observers were relieved every one-and-a-quarter hours.) He realized that what the German astronomers had painstakingly assembled star by star might be captured in rapid fashion through a series of wide-angle photographs. Stitched together, the mosaic of chemical images would form a map of any region of the night sky. He envisioned a photographic extension of the BD, spreading its coverage all the way to the south celestial pole.

  In late November 1882, with the great comet still prominent in the sky, Gill wrote to J. H. Dallmeyer of the noted optical firm in London to propose development of a low-distortion, wide-angle objective lens: “If we can get over the distortion in a reasonable degree, so as to get sharp . . . pictures of stars over a field of 10° square, here is a very easy way of making star maps for working purposes.” Dallmeyer delivered a four-inch photographic objective within four months and a six-inch version the following year. In 1885, the Royal Society provided Gill with a three-hundred-pound subsidy to begin photography for his proposed Cape Photographic Durchmusterung (CPD). The exposed plates were shipped to the Netherlands, where the star images were machine measured by prison inmates under the supervision of the astronomer J. C. Kapteyn at the University of Groningen. When the Royal Society withdrew its funding in 1887, Gill paid for completion of the project out of his own pocket (after consulting his wife). The CPD, the first comprehensive photographic star catalog, listing positions and magnitudes of nearly 455,000 southern stars, was published in installments between 1896 and 1900.

  Among the recipients of Gill’s 1882 comet prints was Rear Admiral Ernest Mouchez, director of the Paris Observatory. Mouchez, too, contemplated the implications of the pictures’ starry backdrops. His conclusions about the future of celestial photography would quickly be underscored by the activities of two of his staff astronomers. Paul and Prosper Henry were inseparable in their professional lives. A coworker recalled that the brothers were “so united that often at the Observatory we saw in them but one person. . . . [I]t is really impossible to discern what may belong to each one in their common work.” Fourteen asteroid discoveries are credited to the Henry name: seven to each brother, in alternate succession, between 1872 and 1882. During this period, the Henrys were immersed in the creation of a visual map of stars along the ecliptic (roughly, the plane of the solar system), where asteroids—and possibly new planets—would most likely be found. Having arrived in their survey at the point where the ecliptic crosses the Milky Way, the brothers were stymied by the sheer number of stars to be hand-plotted: up to eighteen thousand on a thirteen-inch-square chart. The new dry-plate photography, they reasoned, might provide a route through these star-choked regions.

  During their off-hours in 1884, the Henrys designed and built a six-inch refractor, mounted a camera, and trained it on the picturesque double star cluster in Perseus. The result was astonishing: star images reportedly so pinpoint that skeptical astronomers came from around Europe to inspect the plates. Seeing these finely rendered, twin celestial hives, Admiral Mouchez ordered the immediate development of a larger instrument. Within a year, the Henry brothers completed a boxlike thirteen-inch refractor capable of photographing sizable regions of the Milky Way with minimal distortion. A one-hour test exposure produced a mini-skyscape strewn with dots as small as a thousandth of an inch across: stars of magnitude fifteen. The same field, mapped sequentially by eye, would have involved months of labor at the telescope. (To the amusement of visitors, the Henrys optical “laboratory” was a rude shed on the observatory grounds.)

  Although both David Gill and Ernest Mouchez had been steeped in traditional exact astronomy, they shared a progressive—and monumental—vision of the next step in celestial imaging: a comprehensive, pole-to-pole photographic chart of the heavens. This freeze-frame of the night sky at the end of the nineteenth century was meant to be their generation’s legacy to astronomers in centuries to come. If stars vary over time, if they appear or disappear, if a faint planet lurks among its starry lookalikes—all would become evident when comparing observations from some future epoch to the archived photographs.


  Gill and Mouchez knew that no single observatory or proximate group of observatories could assemble such an all-sky map. Earth’s spherical form dictated that participants had to be recruited from around the world. Gill wrote his French counterpart in March 1886 to suggest that they convene an international gathering of astronomers to launch the vast project. For his part, Mouchez circulated among the astronomical community a “Proposal for Photographing the Heavens.” The Astrographic Congress—the first-ever international conference of astronomers—took place in Paris in April 1887. Fifty-six delegates from nineteen nations gathered for eleven days to sketch out the creation of the great star map. Procedural standards were developed, committees were formed to supervise the work and adjudicate any technical or scientific disputes, progress reports and future meetings were scheduled. Each facility would bear its own costs for equipment, personnel, and eventual publication of results.

  As to the map itself—the Carte du Ciel—there would be twenty-two-thousand long-exposure plates, each six inches on a side, covering a two-by-two-degree square on the sky and rendering of some twenty million stars down to fourteenth magnitude. Measuring the position and brightness of twenty million stars was not practicable, even if every astronomer on Earth pitched in. Therefore, a duplicate set of short-exposure plates would be obtained, providing measurement data for about two million stars to eleventh magnitude: the Astrographic Catalogue. After much wrangling over the merits of refractors versus reflectors, the delegates agreed to adopt the Henry brothers’ thirteen-inch astrographic telescope as the project’s standard.