Starlight Detectives

by Alan Hirshfeld

  Into the 1890s, amateurs and professionals enjoyed equal status in the astronomical hierarchy, both in the United States and in Great Britain. They contributed to research and technological developments, published papers, participated in conferences, belonged to and even headed major scientific organizations, and received awards in like numbers. In Britain especially, the wealthy-amateur tradition remained strong, with “backyard” observatories comprising half of the nation’s astronomical research facilities; with the exception of the Royal Observatories at Greenwich and the Cape of Good Hope, there was virtually no government or philanthropic support for astronomy.

  In the 1890s, relations between academically trained astronomers and their amateur brethren changed. As the performance requirements of astrophysical observation became more stringent and the analysis of photographic and spectroscopic data more mathematical, institutional practitioners moved to redefine their field. Previously, anyone (more correctly, any man) who possessed sufficient skills was welcomed into the research fraternity. However, by the start of the twentieth century, entry required academic credentials, specialized graduate training, and absolute rigor in scientific thought and practice.

  As the years progressed, the cost of research-grade telescopes, auxiliary equipment, and personnel skyrocketed beyond the means of wealthy amateur observers. Astrophysical observatories were no longer one-person, domestic operations, but complexes sited on remote mountaintops and attended by full-time staff. Though finding themselves marginalized in research, American amateurs remained a force in fostering public awareness of astronomy’s wonder and value. Presumably, this active societal interest and its attendant media coverage gave even the most self-absorbed benefactors reason to consider astronomy as a route to secure their legacy.

  A quiet tension between professional and amateur observers emerged as the cohort of academically trained astronomers prospered. Where once the two groups had shared a common platform, distinctions were being drawn between them in terms of academic credentials, mathematical acumen, and most devastatingly to the amateurs, the scientific merit of their work. The door to amateur participation began to ease shut when institutional astrophysicists revised their organizations and their journals to foster the professionalization of the field. Professional astronomers were expected to keep up with the published literature and research protocols in their area of specialization, and to understand the relevance of their work within the overall context of cosmical studies.

  Research methods and results were subjected to unprecedented scrutiny. Astrophysical measurements were now held to the same rigorous standards as traditional astrometric quantities, the wavelength of a spectral line and the position of a star regarded with an equal eye to error. Artifacts of astrophysical observation were valued in proportion to their scientific worth, no longer their aesthetic appeal. In sum, the expectations of what constituted a publishable work were raised, as the gatekeepers of astrophysical journals reflected the ascendance of the academic astronomer.

  Professional astronomers’ concerns about their amateur colleagues were vented publicly at conferences and in the pages of research journals. In 1898, longtime amateur astronomer Frank McClean released two volumes of stellar-spectrum images obtained from his home in England and at the Cape of Good Hope. The work received a lukewarm review in the Astrophysical Journal from Yerkes Observatory’s Edwin Frost, one of America’s premier spectroscopists. Frost alludes to the many imperfections in McClean’s images, as well as to his omission of dates, exposure times, and even the barest description of his optical system. He characterizes the work as “pictorial and qualitative rather than metrical and quantitative,” that is, more family photo album than scientific document. On McClean’s discovery of oxygen lines in one star’s spectrum, Frost withheld judgment—until the observation had been confirmed by a professional astronomer.

  A nastier face-off erupted at the May 1895 meeting of the Royal Astronomical Society with a presentation by the preeminent celestial photographer Isaac Roberts. Entirely self-taught, Roberts had established his bona fides with an acclaimed series of images of celestial nebulae, most notably the giant spiral cloud in Andromeda. At the meeting, Roberts projected a lantern-slide of a three-hour exposure that displayed the nebulous surroundings of the star 15 Monocerotis. He followed this slide with an image of the same region by Lick Observatory’s Edward Emerson Barnard.

  A high-strung workaholic and passionate comet hunter—he spotted sixteen during his lifetime—Barnard made his name in 1892 with his discovery of Jupiter’s fifth moon, Amalthea. He became an expert in wide-field celestial photography, using portrait-camera lenses to capture the magnificent landscape of the Milky Way. (The photographic prints for his 1927 sky atlas had to be remade after a stray bullet from a Chicago mobster’s gun pierced the originals in the print shop.) Of his time with Barnard, astronomer Philip Fox recalled, “One could always tell how the night had been by his reaction from it. If clouds or bad seeing had marred the observing, his unconscious sighs were clearly audible. If the sky had been kind, his spirit was gay with song.”

  Although the pair of images offered by Isaac Roberts depicted identical regions of the night sky, they looked distinctly different. The stars in Roberts’s photograph, taken through his twenty-inch telescope, were virtual pinpoints; those in Barnard’s picture, taken with the low-magnification sky camera, appeared as bloated disks. Roberts acknowledged that he had enlarged Barnard’s photograph fivefold to bring its scale up to that of his own. But he had done so to make a point. The nebulosity depicted in Barnard’s picture, he asserted, was illusory; it stemmed from the poor spatial resolution of Barnard’s wide-field camera, which captured broad swaths of sky at a highly pinched scale. Through his own twenty-inch reflector, he continued, the purported knots and swirls of nebulosity are seen for what they are: aggregations of faint, barely resolved stars. And scattered among these, a profusion of even dimmer stars, entirely beyond the light grasp of Barnard’s minuscule instrument. The luminous veil around 15 Monocerotis, Roberts told his audience, awaits only a larger telescope to confirm its underlying stellar nature.

  Edward Emerson Barnard, circa 1885.

  A salvo had been fired from an amateur astronomer’s station south of London into the heart of a professional research institution in California. The return volley arrived with unsurprising swiftness in the December 1895 pages of the Monthly Notices of the Royal Astronomical Society, where Barnard points out the absurdity of Roberts’s conclusions. Comparing a five-times enlargement of one photograph to an unenlarged version of another reveals nothing. Roberts’s truncated field of view isolated the nebulosity from its overall visual context. Wide-field camera views are meant to be studied in the original format, with diaphanous nebulae standing out against their dark surround. “They are intended to be looked at and studied as pictures of the regions they show,” Barnard writes, “and are not to be examined microscopically. It is unjust to use an enlargement such as Dr. Roberts used, because it necessarily puts these pictures at a disadvantage. My picture was simply spoiled by this, while Dr. Roberts’ retained its original qualities, not being enlarged.”

  In his broadside, Barnard further points out a counterintuitive aspect of nebular observing that the amateur Roberts had failed to grasp. True, a large-aperture telescope has more light-gathering power than one with a small aperture, and will therefore reveal fainter stars. But unlike the light of a star, which is essentially point-like in the sky, the light of a diffuse nebula is spread out. The parameter that governs its visibility is its surface brightness, that is, its luminous energy per unit area of sky. For a nebula, telescope aperture is a mixed blessing: a large telescope renders an interstellar cloud in more detail, yet no brighter than a small telescope. The additional light that is collected is diluted in equal proportion by magnification, that is, the nebula’s surface brightness remains the same. What brings a low-contrast nebula to visibility are a telescope’s optical efficiency (the ratio of its
focal length to its aperture), the clarity of the night sky, the photographic exposure time, and the plate sensitivity. Indeed, Barnard’s wide-field camera was an ideal instrument to capture the species of extended, wispy nebulae that night-sky observers had long noted by eye.

  To underline his dismay, Barnard included three high-quality prints of the 15 Monocerotis region, which highlight the cramped scale of Roberts’s telescopic field of view. Examining these pictures, he tells the Royal Astronomical Society, “it will be seen that Dr. Roberts’ reasoning is decidedly wrong. This diffused light is in nowise confined to the star areas. It will be also readily seen that it spreads over a large region where there are essentially no stars at all—even where Dr. Roberts’ reflector can show no stars. That this is real diffused nebulosity there is no reason whatever to doubt.”

  Roberts refused to yield the point. A large-aperture telescope, he maintained, picks up fainter stars and therefore possesses an equal capacity to reveal fainter nebulae—which, as anyone can see, are simply not evident. To buttress his argument, Roberts mounted a five-inch, portrait-lens camera alongside his twenty-inch reflector and tried his own hand at wide-field photography. By 1896, the next battle erupted, this one regarding the presence of nebulosity surrounding the Pleiades star cluster. Here again, Barnard insisted on the reality of the extensive, diffuse corona, Roberts on its absence. Barnard dashed off an exasperated note to the Royal Astronomical Society. Three professional astronomers, he informs them, have taken unequivocal wide-field images of the supposedly fictitious nebulae around the Pleiades. Nowhere does he mention Roberts by name, but it is clear to whom he is speaking when he writes, “These nebulosities . . . have been amply verified (if such a verification were at all necessary). . . . It would therefore appear that a failure to show these remarkable features with an ordinary portrait lens and an exposure anything like 4 or 5 hours must be attributed to something else than their nonexistence.”

  The final exchange between Isaac Roberts and E. E. Barnard—and the one that most acutely highlights the escalating rigor demanded in astronomical observation—took place in 1903, this time in the pages of America’s Astrophysical Journal. The subject of Roberts’s article was William Herschel’s venerable catalog of faint, extended nebulosity. Over the past six years, Roberts had photographed all fifty-two of Herschel’s regions, yet confirmed nebulosity in only four; evidently, one of history’s keenest visual observers had conjured the rest. So extraordinary was this conclusion—as well as Roberts’s apparent ignorance of others’ work in this area—that the editors of the Astrophysical Journal engaged Barnard, now at Yerkes Observatory, to write a companion piece to address the matter.

  Feel the frustration flowing from his pen when Barnard asks his fellow astronomers whether it is not “a little unreasonable to suppose that Herschel, who made so few blunders compared with the wonderful and varied work that he accomplished, should be so palpably mistaken in forty-eight out of fifty-two observations of this kind.” (In fact, working at the very limit of visual perception, Herschel had recorded a number of spurious celestial objects.) Barnard points out that Roberts’s standard exposure time of ninety minutes is insufficient to show fainter interstellar clouds; his own camera shutter often remains open for four or five hours.

  A seven-hour exposure of a nebulous region in the constellation Scorpius, taken by Edward Barnard on June 21 and 22, 1895.

  As one might tutor a neophyte, Barnard returns to the fundamental differences between stellar and nebular photography. To demonstrate the irrelevance of telescope aperture on nebular visibility, he describes a picture he took of Herschel’s region No. 27 in Orion through a cheap lantern lens, a mere 1.6-inches across. “Most of the great curved nebula is clearly shown,” he writes about the ghostly arc, since dubbed Barnard’s Loop, “especially the region described by Herschel. . . . There is therefore no question but that this nebulosity exists where Herschel saw it.”

  In closing, Barnard thrusts a verbal dagger at Roberts’s competence, reminding readers that it was “with the same instruments described in his present paper that Dr. Roberts failed to get any traces of the exterior nebulosities of the Pleiades, which have been shown by four observers with four different instruments not only to exist, but to be not at all difficult objects.”

  A simultaneous assault on Roberts’s disavowal of Herschel’s nebulae swept in unexpectedly from the Continent. Attendees at the March 1903 meeting of the Royal Astronomical Society were treated to an exquisite, wide-field image of Barnard’s Loop, taken by Heidelberg University astronomer Max Wolf through the observatory’s new sixteen-inch photographic telescope. According to the minutes of the meeting, the picture “showed, besides the great Orion Nebula and the nebula around [the star] Zeta Orionis, considerable masses of detailed nebulosity joining these, and also connecting the great nebula with the head of the snake-like nebulosity (first photographed by Prof. Barnard), which winds through a large part of the constellation.”

  Where, fifteen years earlier, Isaac Roberts had astounded the audience with his own breakthrough image of the Andromeda Nebula, he was now confronted with an equally vivid portrait of a celestial object he claimed did not exist. At a stout sixteen-inches aperture, Wolf’s telescope was neither the small toy nor child’s lantern-lens that Roberts had called Barnard’s diminutive cameras. And Wolf’s exposure time of six hours and fifteen minutes eclipsed Roberts’s own ninety-minute standard. (Wolf’s pièce de résistance would come in 1911 with a twenty-five-hour exposure of the spectrum of the Andromeda Nebula accumulated over twenty nights. As one astronomical luminary put it, “Difficulty seems to have a peculiar attraction to him.”)

  Faced with seemingly incontrovertible evidence of the nebula’s reality, Roberts grew defensive. Perhaps, he suggested, counter to their long experience, Wolf and Barnard had each targeted the wrong object. If not that, then the apparition on their plates might have arisen from atmospheric glare or optical imperfections—again unlikely, given the multiplicity of cameras and observing sites. Had Roberts accepted the verdict of his own eyes, he would have recognized (as most others already had) why extended nebulosities did not reveal themselves on his plates. A ninety-minute exposure under moisture-laden, semitransparent English skies is frequently insufficient to capture a low-contrast space cloud. Roberts had used three- and four-hour exposures with great success for his Andromeda plates. Yet he clung to his willful misconception about the essential role of surface brightness on nebular imaging: if he exposed long enough to show the faint stars William Herschel had seen, then Herschel’s faint nebulae should likewise appear. Then in his seventies, Roberts might simply have lacked the energy to push his mammoth project to its required limits. (In fact, Barnard announced that he did find traces of nebulosity on Roberts’s plates, a claim Roberts himself denied.)

  Another reason Roberts’s photographs differ from Wolf’s was voiced at the 1903 meeting by Cambridge University astronomer Arthur Hinks. Echoing Barnard’s standing criticism, Hinks reminded the audience that Roberts’s plates span a mere two degrees, whereas Wolf’s cover a virtually panoramic ten degrees. Extended nebulae overrun the borders of Roberts’s plates, causing at most a perceptible brightening of the entire field. The same nebulae sit isolated within Wolf’s pictures, their feathered edges made manifest against the blackness of the night sky.

  The nebular controversy ended only with Isaac Roberts’s death in 1904. At first glance, the conflict might be taken to epitomize the professional–amateur divide in astronomy at the turn of the century. But, as the self-taught Barnard might be considered something of an über-amateur himself, it is more properly seen as the consequence of rising scientific standards. In his passion to become an effective researcher, Barnard aligned his working methods and his critical judgment (if not his prickly personality) to those of university-trained astronomers. His knowledge base was largely empirical, but filtered by an acute sense of the scientific constraints one learns in the halls of academia.

&
nbsp; When Barnard criticized Isaac Roberts for not keeping up with the technical literature, for shrugging off gaps in his ken of pertinent physics, for plowing ahead without regard to advancements in the field, he was enumerating the hallmarks of an unprofessional scientist. That Roberts was an amateur was not the point. A decade earlier, Barnard had taken umbrage at the mediocre observing skills of his boss, one-time Lick Observatory director Edward Holden, who tied up the thirty-six-inch refractor two nights a week with his own desultory observations of planets. Of course, Holden viewed Barnard’s grumbling in a wholly different light, remarking to a colleague, “I shall not be sorry to have him go—for in his place we shall get a professional—and I must say that Amateurs are hard to get along with!”

  As the twentieth century proceeded, amateurs evaporated from the ranks of professional scientific societies. By 1909, they constituted 12 percent of the American Astronomical Society, but contributed only 1 percent of the research papers. Amateurs had already formed their own associations, where members enjoyed the social bond of shared purpose and wonderment, minus the overtly mathematical aspects of the subject. An amateur-led incarnation of the American Astronomical Society flourished in Brooklyn between 1883 and 1888. Although several local professionals signed on, the group was roundly condemned by Simon Newcomb, dean of American classical astronomy, for its arrogance in portraying itself as a national body.

  In 1889, impressed by the joint efforts of Lick Observatory and the Pacific Coast Amateur Photographic Association during a recent solar eclipse, Edward Holden moved to found the Astronomical Society of the Pacific (ASP), the first national astronomical organization in the United States. The ASP’s liberal membership requirements are stated in its recruiting circular of that year: “The new Society is designed to be popular in the best sense of the word. We wish to count in our membership every person on the Pacific Coast who takes a general interest in Astronomy, whether he has made special studies in this direction or not.” By the end of its first year, the ASP had recruited 178 professionals and amateurs, and thrives today with an international membership base.