Starlight Detectives

by Alan Hirshfeld

  To dispel the problematic, off-kilter distribution of globular clusters in space, Shapley asserted that these starry aggregations constitute a spheroidal halo symmetric about the galactic core; it is our solar system that lies in the Galaxy’s outskirts. Situated as many of them are, above or below the dusty galactic plane, globular clusters can be seen to much greater distances than stars within the disk. “To the measurer of the sidereal universe,” Shapley posited, “star clusters are beacon lights. They point the way to the center of the Galaxy and to its edges . . . The globular clusters are a sort of framework—a vague skeleton of the whole Galaxy—the first and still the best indicators of its extent and orientation.” By estimating the distances of globular clusters via their Cepheids, Shapley computed the locus of their distribution in space. And with that central point established, the Galaxy’s overall size and Earth’s placement within it followed.

  In his 1919 paper, “Remarks on the Arrangement of the Sidereal Universe,” Shapley proposed a breathtaking, tenfold increase in the Milky Way’s diameter, to three hundred thousand light-years. The solar system, he maintained, lies some sixty thousand light-years from the nucleus of this lenticular array of matter. Shapley acknowledged the precariousness of his results: “It is probable that the further accumulation of observations will modify to some extent the views outlined . . . The present data may in some cases be susceptible of alternative interpretation, or possibly the conclusions may be questioned in the belief that the material is insufficient. But the greater part of the hypothesis proposed is merely the most direct and simple reading of the observations.” While time would indeed scale down Shapley’s numbers, the bijou Kapteyn galaxy was now arrayed against a formidable competitor.

  Shapley believed that his Big Galaxy subsumed both the globular star clusters and the spiral nebulae—in particular, that spirals are subordinate in scale to the Milky Way: “From the new point of view, our galactic universe appears as a single, enormous, all-comprehending unit. . . . The adoption of such an arrangement of sidereal objects leaves us with no evidence of a plurality of stellar ‘universes.’” Shapley’s model was substantially fortified by the photographic work of his Mount Wilson associate, Adriaan van Maanen, who claimed to have detected rotation in comparing plates of several brighter spirals. As Shapley realized, if spirals are Big Galaxy analogs seen at great distance, their purported angular spin would translate into absurdly large stellar velocities toward their periphery, approaching or even exceeding the speed of light. Basking in the logical certainty that spiral nebulae must therefore lie within the Milky Way, Shapley dashed off a congratulatory note to van Maanen: “Between us we have put a crimp in the island universes, it seems,—you by bringing the spirals in and I by pushing the Galaxy out. We are indeed clever, we are.” Further affirmation of the Big Galaxy concept followed from a host of arguments based on what were subsequently proven to be wrongly categorized stars or misinterpreted observations.

  Evidence against Shapley’s megagalaxy-as-universe model was readily found. His edifice rested on a rickety foundation: that stars in globular clusters are identical to similar-looking stars in the disk of the Milky Way. In particular, Shapley asserted that Cepheid variables in clusters follow the same period-luminosity relation as those elsewhere in the Galaxy. If untrue, then the huge distances he attributed to them are incorrect. (Indeed, it would later be proven that the stellar population of globular clusters—including the Cepheids—is older than that found in the Milky Way’s disk, and have different observational properties.)

  Furthermore, eruptive stars, or novae, had been photographed in almost a dozen spiral nebulae, and were invariably much fainter than novae within the Milky Way. This marked diminishment in apparent brightness—on average, one ten-thousandth the radiance of a galactic nova—signaled the remoteness of the spirals. The Andromeda Nebula, considered one of the nearer spirals, was pegged at a distance of a million light-years on this basis. At the very least, the occurrence of novae within otherwise diffuse spiral nebulae infers the presence of stars, rendered irresolvable by distance.

  Also seemingly counter to Shapley’s single-galaxy universe were the hyperkinetic movements of spiral nebulae. Vesto Melvin Slipher, at the Lowell Observatory in Flagstaff, Arizona, had measured galactic radial velocities with a twenty-four-inch Clark refractor and Brashear spectrograph. His presentation to the American Astronomical Society in August 1914 drew a standing ovation, and ultimately, medals from the Paris Academy, the Royal Astronomical Society, and the Astronomical Society of the Pacific. (First-semester graduate-student Edwin Hubble was in the audience.) The acclaim stemmed in part from Slipher’s astonishing news—spirals are streaking through space at up to six hundred miles per second, much faster than any galactic star or globular cluster—but also from collegial admiration of the technological feat underlying those measurements.

  The Lowell telescope and spectrograph had been optimized for planetary observations, where light was plentiful and centrally condensed. Faint, diffuse objects, such as spiral nebulae, required exceedingly long exposures to register on the photographic plate; recording their dispersed spectra was far more challenging. Through arduous experimentation, Slipher settled on a single-prism spectrograph, sacrificing dispersion for a gain in transmitted light. He added a super-fast camera lens and chemically boosted the photographic emulsions to reduce exposure times. The result was a 450-pound analytical engine capable of revealing nebular spectral lines, as well as their Doppler shifts.

  Vesto Melvin Slipher, pictured in 1932.

  Even with Slipher’s upgrades, seven hours might be required to capture a measurable image of the spectrum of the Andromeda Nebula. Fainter spirals entailed exposures of twenty to forty hours spanning several nights. As Slipher’s list of three- and then four-figure radial velocities grew from its initial fifteen to twenty-five by 1917, astronomers pondered whether such fleet objects could be confined by the gravitational field of the Milky Way, even one as generously proportioned as Shapley’s. Shapley countered by suggesting that spirals are discrete, gaseous outflows from our galaxy, sailing through space on the gentle pressure of starlight.

  Spiral nebulae also display a peculiar avoidance of the Milky Way’s otherwise highly populated disk; instead, they occupy two broad cones of space perpendicular to the galactic plane. In an address to the American Association for the Advancement of Science in December 1816, Lick Observatory Director W. W. Campbell points out that, were spirals internal to our galaxy, as Shapley’s model places them, their observed distribution is perverse: “spirals live close to the right of us and close to the left of us, but . . . they avoid getting between us and the Milky Way structure.” Much more likely, Campbell surmises with prophetic insight, is that obscuring material in the galaxy’s disk extinguishes the light of remote nebulae along those lines of sight. As to the impasse over the nature of spiral nebulae, he adds, “We are not certain how far away they are; we are not certain what they are. However, the hypothesis that they are enormously distant bodies, that they are independent systems in different degrees of development, is one which seems to be in best harmony with known facts.”

  On April 20, 1920, astronomers’ divergent impressions on the scale of the cosmos received a formal airing at the annual meeting of the National Academy of Sciences in Washington, DC. The event, billed as a debate but more a pair of sequential lectures, pitted Mount Wilson’s rising star, Harlow Shapley, against Lick’s island-universe advocate, Heber D. Curtis. In the running for the directorship at Harvard Observatory (whose representatives were in attendance), Shapley plugged his own contributions and aspirations in the research arena. He spoke at length about the distribution of globular clusters and the size of the Milky Way, and gave short shrift to extragalactic matters. Taking the rostrum, Curtis laid out a highly technical, point-by-point rebuttal of Shapley’s assumptions and cited a string of evidence in support of the multigalaxy universe.

  Little notice was taken of the event at the time
. No minds were changed, no breakthrough discoveries announced. Only after the publication in 1921 of conjoined articles by Shapley and Curtis did the scientific community come to recognize the symbolic import of the so-called Great Debate. Implicit in the pages of observational data and esoteric arguments was the weighty issue at hand: humanity’s conception of the universe. As in their long-ago wrangle over the geocentric and heliocentric cosmologies, astronomers stood on either side of a scientific divide. Each camp believed in the facts as they saw them, each was aware that their respective models were mutually exclusive: only one of the cosmic constructions cohered with reality. Yet as fervent as their arguments, all participants recognized that the ultimate arbiter of the verity of their ideas was better observational evidence—nature’s testimony, more clearly rendered.

  Although the Great Debate failed to resolve the core cosmological issues, it did cast a spotlight on the blistering pace of technological achievement over the previous decades. Astrophysical methods only recently embryonic now probed once inaccessible realms of space. Where photographic pioneers had been hard-pressed to capture the amalgamated glow of a globular cluster, their successors were recording subtle light variations of its constituent stars. Where spectroscopists had once struggled to tease out Doppler shifts in the spectra of bright stars, they were now measuring line displacements in the dispersed light of spiral nebulae. Where astronomers had previously inspected the heavens with telescopes the size of a rolled-up carpet, now they deployed sixty-inch and one-hundred-inch mountaintop reflectors on a nightly basis. By 1920, photography, spectroscopy, and modern megatelescopes had merged into a unitary instrument of enormous analytical power. And where technology leaps ahead, discovery is apt to follow.

  Chapter 27

  A NIGHT TO REMEMBER

  What are galaxies? No one knew before 1900. Very few people knew in 1920. All astronomers knew after 1924.

  —Allan Sandage, introduction to the Hubble Atlas of Galaxies, 1961

  AN ASTRONOMER’S OBSERVING NOTEBOOK is a window onto a scientist’s working habits. Every pertinent detail is recorded: date and time; the celestial target’s designation, sky coordinates, angle from the meridian; if imaged, the photographic plate identifier (nowadays, a computer file name) plus duration of exposure; description or enumeration of the “seeing,” the cutely nontechnical term for the clarity of the sky. Errors are crossed out, never erased or obliterated, corrections scrawled in nearby.

  Seemingly little can be gleaned about the astronomer’s state of mind from such a telegraphic spreadsheet. Family stresses, concerns over promotion, physical or mental fatigue intrude nowhere in these coldly rational, alphanumeric rows and columns. Yet there is a personal dimension implicit in this ritualistic recording of nights’ passings under the stars: the supreme devotion to exploration, the relentless, sleep-deprived, frostbite-be-damned yearning to commune with the dimly lit wonders of deep space, each an astral Juliet beckoning the observer from an impossibly high balcony.

  Every line of the astronomer’s observing notebook includes space for remarks, queries, and out-of-the-ordinary circumstances: whether the camera has been modified, whether the target nebula resembles a cat’s eye, whether a star appears brighter than it did the previous night. On rare occasions, the terse prose of the remarks sections gives way to an entry whose wording is more detailed than usual, more attentive to the vagaries of interpretation—no longer a note to self, but a declaration to posterity. Here, in a more deliberately rendered hand, is where the heightened pulse of the observer is almost palpable. Here is the moment of discovery.

  On the evening of October 5, 1923, Edwin Hubble opened the dome of Mount Wilson’s one-hundred-inch reflector, unaware that tonight’s observing run would have momentous consequences, much less spark national headlines. It was his tenth allocation of time on the big telescope this year, the intervening nights spent on the sixty-inch and smaller instruments. With the shutters parted, the view of the sky through the vertical breach was unpromising: on a “seeing” scale of 1 to 5, Hubble jotted “1” in his observing notebook—barely worth opening the dome at all.

  The previous four years had been rife with developments in Hubble’s career, as well as those of his Mount Wilson compatriots. Hubble was a veteran observer by now, fully integrated into the research juggernaut fostered by George Ellery Hale. Having suffered his fourth mental breakdown, Hale had relinquished his directorship to his capable second, Walter S. Adams, and left with his family for an extended tour of Europe and the Middle East. Big Galaxy crusader Harlow Shapley had absconded for Harvard, taking with him a barely concealed antipathy toward Hubble’s Oxfordian manner, conservative politics, and dawning celebrity. (Both men were native Missourians.)

  Hubble kept his eye on the sky conditions as he prepared the camera for the evening’s initial target, NGC 6729, a fan-shaped emission nebula in Corona Australis. He could reflect, with considerable satisfaction, on his accomplishments of the past two years. With near-fanatical thoroughness, he had amassed a photographic database of nebulae: a visual menagerie as seemingly diverse as any in the biological realm. His 1922 paper, “A General Study of Diffuse Galactic Nebulae,” summarized the nature, form, and distribution of these deep-space clouds, and noted any apparent physical relation to proximate stars. Hubble’s nebula classification scheme was widely adopted in the astronomical community.

  His “galactic” types are clearly associated with stars, and include both planetaries—compact, often rounded forms whose moniker is a Herschel-era relic—and diffuse nebulae like the famed luminescent billow in Orion. The “nongalactic” category comprises clouds of spiral, spindle, ovate, globular, or irregular form. With the exception of the random flare-up of a nova, these nebulae lack discrete, fully resolved images of stars; to the eye, they appear misty throughout. (Hubble was noncommittal on the vexing issue of the spirals’ remoteness, specifying that his “nongalactic” does not imply “extragalactic.”)

  Several months after this landmark paper, Hubble released a tour de force analysis of the physical engine that sustains a galactic nebula’s glow. His conclusion: gaseous nebulae fluoresce under the withering glare of their embedded stars; the energy breaching the surface of these stellar powerhouses is absorbed by the surrounding gas, then reemitted in equal amount into the void. In Hubble’s model, planetary nebulae arise from the prior expulsion of a star’s atmosphere, creating a gassy cocoon energized by its parent star’s light. Sprawling Orion-like complexes are made visible by the radiation of multiple hot stars, both within and adjacent to the bodies of gas. In the absence of such stars, a diffuse nebula might yet reveal itself in sharp silhouette against more distant banks of luminous matter.

  As the October evening progressed, Hubble discerned an improvement in the sky conditions: he entered “2” in the Seeing column of his observing notebook. Slewing the telescope to the starry, ragged-bordered nebula NGC 6822, he took a sixty-minute exposure. By now, the night sky over Mount Wilson verged on its storied crystal clarity: the seeing had risen to “3+”. The next plate was reserved for the Andromeda Nebula. After midnight, now October 6, Hubble acquired a forty-five-minute portrait while the great spiral cloud closed in on the meridian.

  Not surprisingly, the quality of the developed plates reflected the improvement in sky conditions, Andromeda’s being the best. Scanning its magnified image, Hubble’s eye was drawn, in turn, to three faint stars, barely discernible amid the nebula’s light-speckled background. Still, they hailed their presence to Hubble, for whom Andromeda’s stellar field was virtually etched into memory. These were stars he had not seen before, each one a possible nova.

  Novae had been discovered in a number of spiral nebulae, their relative dimness compared to those in the Milky Way suggestive of great distance. But the physical process underlying these stellar eruptions was unknown. Nor had any nova occurred close enough to Earth that its distance, hence its absolute light output, might be ascertained. Brilliant as these starry beacons ar
e, in Hubble’s day, they could not be adopted as “standard candles”—calibrated energy emitters whose comparative apparent and absolute brightnesses allow computation of cosmic distance. Nevertheless, Hubble dutifully took a pen to the Andromeda plate, marked each nova’s location with a dash, and wrote the letter N alongside.

  Over the succeeding months, Hubble continued to monitor Andromeda with both the one-hundred-inch and the sixty-inch reflectors. He collected additional images of the nebula dating back as far as 1909 from Mount Wilson’s photographic plate archive. Two of the three suspected novae exhibited the characteristic behavior of this stellar class: a transient outburst that gradually fades from view. However, the third proved remarkably durable: it was visible in every picture, old or new. Yet its prominence, relative to the stars around it, was different from plate to plate. This was no one-shot nova, Hubble realized, but some type of cyclical variable star.

  On a sheet of graph paper, Hubble plotted the star’s light-curve, a series of points tracing out the star’s brightness with the passage of time. A distinctive, undulating pattern emerged: rapid rise to maximum brilliance, followed by gradual decline to minimum, the cycle repeating with clocklike regularity every 31.415 days. This inconspicuous mote of light was a Cepheid variable star, the first of its kind ever confirmed in a spiral nebula. Hubble had turned up an astronomical standard candle flickering in the arms of the Andromeda Nebula. He retrieved his observing notebook, paged to the line for October 5, 1923, and appended to the Remarks entry, “On this plate (H335H), three stars were found, 2 of which were novae, and 1 proved to be a variable, later identified as a Cepheid—the 1st to be recognized in M31 [the Andromeda Nebula].” Then, like a spotlight heralding discovery, Hubble drew a brawny arrow pointing at his five cramped lines of text. On the historic plate H335H (the first H standing for the Hooker telescope, the second for Hubble), he crossed out the label “N” that stood alongside the newly identified Cepheid and inked in “VAR!”