Starlight Detectives

by Alan Hirshfeld

  Williamina Fleming.

  Annie J. Cannon.

  Pickering’s entire enterprise hinged on the successful transition of stellar spectroscopy from a visual to a photographic art. That transition occurred with the introduction of the gelatin dry plate. Whereas spectral classification at the eyepiece requires an individual trained in the complexities—and inured to the discomforts—of telescopic observing, photographic classification can be conducted by workday office staff, whose expertise is channeled toward a single goal: spectral classification. Specimens are available for reinspection, communal consultation, and reproduction. English astronomer Normal Lockyer summed up the powerful synergy of the plate and the prism: “I do think we have in photography not only a tremendous ally of the spectroscope, but a part of the spectroscope itself. Spectroscopy, I think, has already arrived at such a point . . . in connection with the heavenly bodies, that it is almost useless unless the record is a photographic one.”

  Chapter 20

  A SPECTACLE OF SUNS

  Astronomy paints its picture in the brighter colors. A star, regarded as a center of attraction, or as a reference point from which to measure celestial motions, awakens little enthusiasm in the popular mind; but a star regarded as a sun, pouring out floods of light and heat as a consequence of its own contraction, torn by conflicting currents and fiery eruptions, shrouded in absorbing vapors or perhaps in vast masses of flame, appeals at once to the popular imagination.

  —James E. Keeler, “The Importance of Astrophysical Research and the Relation of Astrophysics to Other Physical Sciences,” 1897

  THE ANALYSIS AND CLASSIFICATION of stellar spectra paved the way for scientists to determine the physical properties of stars. Yet it failed to rouse the interest of number-crunching classical astronomers, for whom stars remained points of light speckling a celestial vault. To this cohort of scientists, photography held promise as a vehicle of mass measurement of stellar positions, as evidenced by Lewis Rutherfurd’s plates of star clusters and the vast Carte du Ciel mapping project. And in chronological photographs of star fields, taken across decades, lay the potential to detect the glacial drift of stars across the night sky, a clue to their actual movements through space.

  As adept as they were at quantifying such two-dimensional metrics of position and motion, nineteenth-century astronomers found it hard to pin down the scale of the cosmic third dimension, radially away from Earth. The distances of a few dozen stars had been deduced from visual estimates of their annual parallax, the tiny shift in a star’s position when viewed from opposite extremes of Earth’s orbit. These generated a sparsely populated map of our stellar neighborhood. (Photography would swell the number of parallax measurements into the thousands during the early twentieth century.) Thus, the spectroscope was judged irrelevant to the work of classical astronomy—except for one prospect: Christian Doppler’s supposition that stellar radial velocities might be gauged from spectral-line shifts.

  By the mid-1880s, the visual spectroscopy of William Huggins, Angelo Secchi, and the Royal Greenwich Observatory had failed to produce a single reliable Doppler shift. No matter how steadfast the observer and well built the instrument, the errors of measurement were comparable to the stellar velocities being measured. Then, with stunning rapidity, the camera-enabled spectroscope, or spectrograph, accomplished what its eye-based predecessor could not. And in that success, even the most committed visual adherents came to see the potential of the spectroscope in their line of work.

  Hermann Carl Vogel at Potsdam Astrophysical Observatory outside Berlin was well versed in the merits and pitfalls of the spectroscope. In the early 1870s, while directing a private observatory near Kiel, he tried unsuccessfully to discern the Doppler shifts of stellar spectral lines. However, in observing the limb of the Sun, he found that lines at the solar disk’s eastern and the western edges were shifted in opposite fashion: the eastern limb, toward shorter wavelengths; the western limb, toward longer wavelengths. The Dopplerian conclusion is that luminous matter at the Sun’s eastern limb is moving toward the observer, while the matter at the western limb is moving away from the observer. That is, the Sun rotates. Vogel’s computed solar-rotation velocity agreed with the value derived from the movement of sunspots across the Sun’s face. Furthermore, Vogel confirmed that the Sun does not rotate as a solid body: Its rotation rate varies with solar latitude, fastest at the equator, progressively slower toward the poles. At a time when scientists still debated whether the Doppler effect observed for sound waves applies as well to light waves, Vogel’s solar observations provided definitive evidence that it does.

  In 1887, Vogel embarked on a four-year-long program to determine the radial velocities of stars by photography. Having tried his hand at visual detection of Doppler shifts, he hoped that time exposures would smooth out the skittery spectral images seen by eye. Potsdam’s eleven-inch, wooden tube refractor was a mule of a telescope: generally capable, but hardly suited to the rigors of ultraprecise work. Vogel knew that generating the required high-dispersion spectra demanded the utmost stability of the entire optical–mechanical system against expansion, contraction, and flexure. The frame of his dual-prism spectrograph was fashioned out of steel, with structural bracing inside and out. The camera, too, was made of steel, its plate holder of brass. The comparison spectrum was provided by a hydrogen-gas discharge tube located within the telescope itself; so placed, the starlight and the hydrogen light formed parallel rays. Any flexure of the instrument shifted the stellar and comparison spectra in unison. A tiny auxiliary scope allowed Vogel to sight the star’s beam on the spectrograph slit, and keep it stationary for the duration of the hour-long exposures.

  Vogel’s brand of spectrographic work was far afield from either cosmochemical analysis or stellar spectral classification. Instead of photographing a broad swath of spectrum displaying dozens or hundreds of lines, he recorded only a narrow interval around each star’s Fraunhofer G line, plus the proximate gamma line of hydrogen from his discharge tube. The separation of these lines was measured under a microscope, then compared to their separation in a photograph of the solar spectrum. The deviation of the star’s G line from its solar position revealed the star’s Doppler shift and, via a mathematical formula, its line-of-sight motion.

  In 1892, Vogel published the radial velocities of fifty-one stars, streaking through space at up to thirty miles per second—more than one hundred thousand miles an hour—relative to Earth. The average measurement uncertainty was less than two miles per second, a tenfold improvement over that of Greenwich Observatory’s long-running visual program. In a welcome confluence of the new and the old, Vogel’s radial velocities for Arcturus, Aldebaran, and Betelgeuse were in close accord with those only recently measured by eye by James Keeler using Lick Observatory’s thirty-six-inch refractor. The consistency between the photographic and visual results lent further credence to Vogel’s work. That an eleven-inch refractor, fitted with a camera, matched the capability of the world’s largest mountaintop telescope highlighted the potential of photography to revolutionize the practice of astronomical observation.

  Vogel also released complete specifications of the apparatus and methods used at Potsdam, inspiring similar spectrographic programs in France, Russia, England, and the United States. The most far-reaching of these was at Lick Observatory, where William Wallace Campbell coupled a three-prism spectrograph to the giant refractor in 1896. (A one-time civil engineering student at the University of Michigan, Campbell changed his career path after reading Simon Newcomb’s Popular Astronomy.) After surveying the Northern Hemisphere sky, Campbell secured funds to erect a telescope in Chile, extending the survey to the southern sky. Overall, more than fifteen thousand plates were taken by thirty-one observers and measured by fifty-eight assistants. By the 1920s, Campbell and his team had expanded Vogel’s original page-long roster of radial velocities into a true, all-sky catalogue, containing entries for 2,771 stars. With this extensive compilation, twice as precise as Vog
el’s, Campbell quantified the Sun’s progression through space relative to its stellar neighbors: about twelve miles per second toward the constellation Hercules.

  Like his East Coast contemporary Edward Pickering, W. W. Campbell was an astronomical entrepreneur: he envisioned a large-scale research program, developed the methods, designed the instruments, raised the money, assembled a team, and managed the work. What Pickering had accomplished for stellar spectral classification with the Henry Draper project, Campbell had accomplished for stellar radial velocities with the Lick catalog. Their works provided expansive datasets against which astronomers could test theoretical models of stellar and galactic motion, as well as stellar energy production and evolution. Conceived in the nineteenth century and gestated in the twentieth, the catalogs stand as milestones in new modes of astronomical observation. Pickering, Vogel, and Campbell strove to lay the foundations upon which astronomers might build a more comprehensive picture of the physical universe. As incremental and seemingly prosaic their sort of work, they remained ever alert to the possibility of discovery.

  William Wallace Campbell, pictured around 1890 with Lick Observatory’s thirty-six-inch refractor, here fitted with a spectroscope by John Brashear.

  Of the Big Dipper’s seven stars, the most historic lies second from the end of the handle. Zeta Ursae Majoris, also known by its Arabic name, Mizar, forms a wide pair with dimmer Alcor, the dyad easily perceived by observers of antiquity. Mizar itself is a telescopic double star, the first ever photographed, by Harvard’s George Bond in 1857. Its components, designated A and B, take at least five thousand years to circle each other. Mizar A is the 116,656th entry in the Henry Draper Catalogue, its spectrum an unremarkable smudge among the dozens that crowd the objective-prism photographs of that part of the sky. However, upon microscopic inspection of plates taken between 1887 and 1889, Edward Pickering found a startling anomaly that set apart Mizar A’s spectrum from the rest: on some exposures, the Fraunhofer K line was double, while on others, it was single. And between these extremities of appearance, the K line was fuzzy, like an unresolved line-pair. Mizar A’s numerous other lines were either too broad or too faint to exhibit duality.

  To complete the spectral analysis of Mizar A, Pickering hired Antonia Maury, Henry Draper’s niece and a Vassar graduate in mathematics and chemistry. From measurements of seventy spectrographic plates, Maury confirmed Pickering’s hunch that the transition between single and double lines is periodic. Pickering concluded that, although it appears solitary in a telescope, Mizar A is a binary system. (English photographic pioneer William Henry Fox Talbot had proposed in 1871 that the orbital motion of a binary star might reveal itself in such periodic behavior of its spectral lines.)

  Locked in a gravitational embrace, the stars of Mizar A swing alternately toward and away from Earth with clockwork regularity. “When one component is approaching the earth,” Pickering explained, “all the lines in its spectrum will be moved toward the blue end, while all the lines in the spectrum of the other component will be moved by an equal amount in the opposite direction, if their masses are equal. Each line will thus be separated into two. When the motion becomes perpendicular to the line of sight [thus, no Doppler shift], the spectral lines recover their true wavelength and become single.”

  From the measured Doppler shifts and the time interval between successive line-doublings, Pickering computed the orbital velocity, combined mass, and miles of separation of Mizar A’s components. On November 13, 1889, Pickering reported his results to a gathering of the National Academy of Sciences in Philadelphia. Within a month, he informed his patron, Anna Draper, that the star Beta Aurigae was also a spectroscopic binary. The crucial spectrum had been obtained during Draper’s recent visit to the observatory, he reminded her, then added, “[I]s it not a sufficient argument in favor of your coming oftener?” (Pickering’s initial findings have been updated: the stars of Mizar A trace out a highly elliptical orbit about the size of Mercury’s every twenty and a half days. Mizar B and Alcor were each determined to be binary as well. With the recent discovery that Mizar and Alcor are gravitationally bound, the Mizar-Alcor “pair” is actually a sextuple star system.)

  Unknown to Pickering, Potsdam’s radial-velocity expert, Hermann Vogel, had simultaneously homed in on the spectrum of the most ill-omened luminary in the heavens: Algol, the so-called Demon Star. Algol shines prominently in the constellation Perseus, whose mythological namesake slew the dreaded Medusa. Within the array of stars that comprise Perseus, grasped in what is held to be his outstretched hand, is the Gorgon’s serpent-covered head. And at the hub of that little starry spray lies Algol, the glinting eye of Medusa herself. Algol’s gloomy reputation is said by some to stem from the variability of its light, which fades threefold, then swiftly recovers, every sixty-nine hours. Eighteenth-century observer John Goodricke speculated that the periodic dimming arises from eclipses of the star by an unseen orbital partner. However, more than a century of visual and photographic scrutiny revealed no such star. If Algol was a binary system, the fainter member was lost in the glare of its brighter companion.

  In December 1889, one month after Pickering’s report on Mizar, Vogel announced that Algol’s spectral lines shift in synchrony with its fluctuating brightness. Before Algol dims, the lines are redshifted from their nominal positions: the star is receding from us. Upon its return to normal brightness, the lines shift blueward: the star is approaching. During the event itself, the lines appear in their proper places in the spectrum. Vogel concluded that Algol is indeed part of an eclipsing binary system, as Goodricke had suspected.

  From the spectral data and the timing of the eclipses, Vogel estimated the diameter, mass, separation, and orbital velocity of Algol’s component stars. Although his parameters, like Pickering’s, proved to be erroneous, astronomers at the time were electrified by these twin discoveries. (Modern observations reveal Algol to be a luminous blue-white star, paired with a much dimmer orange star, about 2.9 and 0.8 times the Sun’s mass, respectively. The main eclipse occurs when the companion passes in front of Algol, a secondary eclipse when their positions are switched.)

  The photographic spectrum of the bright star Spica, in the constellation Virgo, also presented a cyclic wobbling of its spectral lines. Here, again, Vogel inferred a binary system, revolving about a common center of mass. Unlike Algol’s telltale eclipses, Spica had displayed no prior sign of duplicity. In the “new” Spica, Lick Observatory’s James Keeler saw the spectrograph’s power to inflate a luminous speck in the night sky into a physical entity in faraway space:

  Translating the mathematical formulae of Professor Vogel into the ideas which they represent, a wonderful picture of stellar motion is presented to our mind, and one to which the whole visible universe, as revealed to us by our greatest telescopes, offers no parallel. The spectacle of two great suns like our own, revolving around each other in only four days at a distance no greater than that which separates the sixth satellite of Saturn from its primary, is one which the inadequacy of our optical powers will probably forbid us from actually beholding, but the indirect evidence that such extraordinary circumstances of motion exist, is so complete that we must admit their reality.

  By now, astronomers were confident that a spectrum can reveal the elemental constituents of a star. The work of Pickering and Vogel further taught them that the spectrograph, in the periodic Doppler-dance of stellar spectral lines, can disclose once-invisible properties of a star. The advance of technology had birthed a new observational species: the spectroscopic binary. In the observation and computation of binary star orbits, the interests of classical, astrometric astronomers and their astrophysical counterparts neatly meshed. The latter provided the former with previously unobtainable data for the mathematical analysis of stellar motions. Every star might now be subject to spectroscopic surveillance. For if Spica, one of the brightest and most-observed stars in the night sky, had been unmasked as a spectroscopic binary, how many other hidden pairs aw
aited exposure by the spectrograph?

  As the twentieth century approached, with the classical mode of astronomy on the wane, the very definition of a research observatory began to change. Where once the human eye was deemed the optimal receptor of telescopic light, an observatory without a camera and a spectrograph was increasingly regarded as incomplete. And as the march of technology rendered direct vision superfluous, if not antithetical, to the act of cosmic exploration, so too the many amateur astronomers who had helped foster that technology. Prior to 1900, amateurs had participated fully in the meetings and governance of scientific societies. Their lack of formal academic training was no hindrance to ascension in the field. Their research papers appeared in journals alongside those of professionals. They received honors in equal measure to their institutional colleagues. But when astrophysics accelerated into the new century, its amateur acolytes found themselves rudely left behind.

  Chapter 21

  THE CLOUD THAT WASN’T THERE

  The born astronomer . . . is moved to celestial knowledge by a passion which dominates his nature. He can no more avoid doing astronomical work, whether in the line of observations or research, than the poet can chain his Pegasus to earth.

  —Simon Newcomb, at the dedication of Yerkes Observatory, 1897.

  SINCE THE MID-NINETEENTH CENTURY, amateur astronomers had been key participants, if not prime movers, in the development of observational astrophysics. Photography and spectroscopy would not have reached their heights as quickly as they did had not the path been blazed by a worldwide cadre of volunteer scientists and inventors. Few academic astronomers had been willing to hitch their professional prospects to these emerging technologies.