Huggins’s 1879 paper is at once educational, technological, and scientific, with a speculative coda on the physical ramifications of spectral line patterns. (In retrospect, one wishes Huggins had omitted this last.) He and Margaret adopted a bare-bones spectroscope with a single prism of Iceland spar, which was both more dispersive and more transparent to ultraviolet rays than either glass or quartz. Their primary targets were the so-called white stars: Sirius, Vega, Altair, Deneb, and Spica. The white-star spectra exhibited far fewer absorption lines than did the Sun; among the lines they had in common, some appeared more prominent in the star, others less.
The key feature that defines a white-star spectrum, declared Huggins, is its pattern of twelve ultrastrong lines, several of which were found to coincide with known lines of hydrogen. The spacing of these lines was intriguingly regular: as one proceeded from the blue toward the ultraviolet (toward shorter wavelengths), the lines grew progressively closer. Prior visual observations added two long-wavelength lines to this “grand rhythmical group,” as Huggins described them. He wondered “whether these lines are not intimately connected with each other, and present the spectrum of one substance.” (Indeed, the line array would come to be identified with hydrogen and its spacing quantified by a mathematical formula derived in 1885 by Swiss schoolteacher Johan Jacob Balmer. The unique pattern arises from the way light interacts with the hydrogen atom, whose structural regularity would be delineated by Neils Bohr in 1913.)
Huggins found that the orange star Arcturus was unlike white stars in its distinctly solar-type spectrum. Yet, unaccountably, its K line of calcium appeared much stronger than the Sun’s. Huggins’s published rendering of Arcturus’s spectrum was so line-rich as to resemble a fine-toothed comb.
Not surprisingly, given its longer gestation, Huggins’s paper is more far-reaching than Draper’s. The latter reads like a tutorial and progress report rather than a completed scientific work: there is only qualitative examination of the spectral lines plus frequent mention of observations yet to be conducted. Draper’s conclusions rest on a limited number of spectral plates of Vega, Arcturus, and Capella, obtained with his twelve-inch refractor between August 6 and October 4, 1879, plus his prior plates of Vega and Altair. (The twenty-eight-inch reflector telescope was undergoing maintenance in 1879.) Like Huggins, Draper distinguishes the spectra of white stars from those of their yellow and orange counterparts. He also cites the prominent array of absorption lines in the spectrum of Vega; these he attributes to hydrogen, but makes no mention of their regular spacing.
Two photographic exposures of the spectrum of the star Arcturus by William Huggins, June 9, 1878.
Did Henry Draper seek to establish priority over William Huggins by publishing first? Or was his paper meant to inform American astronomers about the evolution and current state of stellar spectrum photography? By October 1879, Draper had reached the end of his observing season and would not return to Hastings full-time until the following June. To sit on his observational results, provisional as they were, until the following October must have seemed pointless. While he does not explicitly claim to be the inventor of stellar spectrum photography, he yields no ground to Huggins in his account of its development. Draper writes that his own “experiments and the preparations for them have extended over more than twelve years,” that is, back to 1867 when he launched construction of the twenty-eight-inch telescope. He cites at length Huggins’s abortive attempt at stellar spectrum photography in 1863. And he acknowledges his own recent inspection of the Tulse Hill observatory, thanking Huggins for his recommendation to use gelatin dry plates.
That Henry Draper leveraged his research through Huggins’s willing aid is beyond doubt. Thus, his subsequent rush to publish before Huggins might strike modern sensibilities as a breach of collegial protocol, if not scientific ethics. In science, the publication date has always been regarded as the benchmark of priority in the case of discovery. Yet history has shown that credit for an evolving theory or field, such as stellar spectrum photography, often goes not to individuals who are first to publish, but to those who most convincingly establish the validity and worth of their results.
Not that this would have been of any comfort to William Huggins. He remained ever-vigilant to intrusion upon his scientific legacy, and was convinced that it had been undermined, at least in America, by Henry Draper. In his 1883 letter to Charles Young, Huggins suggests that “it may be possible for you after a time to bring about quietly without any direct reference to Dr. Draper a more truthful appreciation of my work on the photographic spectra of stars than exists in America if I may trust to popular prints.” Charles Young took no side in the dispute, and Huggins continued to stew in private. The matter might have rested there had not Harvard University astronomer Edward C. Pickering swept up Henry Draper’s banner.
A bulldog like William Huggins when it came to professional affairs, Edward Pickering sought to establish Harvard as the preeminent center for what he called the “physical side of astronomy.” Pickering had toured the Harvard Observatory in 1861 when he was fifteen years old. Then-director George P. Bond treated the young astronomy enthusiast to views of the Moon, the Ring Nebula, and the quadruple star system Epsilon Lyrae, an experience Pickering set down in mind-numbing detail in his diary: “This star [Epsilon Lyrae] appears double with a very low power and even, as Mr. Bond said, it can almost be seen without a telescope, through this telescope it appeared to consist of 4 stars, by pairs, each pair appearing at some distance from the other . . .” The sentence tumbles on for another forty-five words before blessedly encountering a period.
Pickering received his bachelor’s degree from Harvard in 1865, taught at the Massachusetts Institute of Technology (MIT) until 1877, then became Harvard Observatory’s director for the next forty-two years. Having soon made his mark in the wholesale visual measurement of stellar brightness—the trademark-redolent “Harvard photometry”—Pickering embarked on its photographic equivalent. His aim of quick success suggested the strategic wisdom of adding stellar spectrum photography to the institution’s research program. By the time William Huggins’s 1883 letter of grievance was steaming its way across the Atlantic, Edward Pickering had already made his initial move. He contacted Henry Draper’s widow, Anna, declaring his deep interest in her husband’s spectroscopic research. He had spoken to Henry during the National Academy gala at their home last November, only days before Henry’s death: “I urged upon him the importance of an early publication of his stellar spectra. . . . I called his attention to the similar work now in progress by Dr. Huggins, and that a delay might seriously diminish the value of the investigations.”
With Henry gone, Pickering inquired, what was to become of his spectroscopic plates, his equipment? Anna Draper replied with noble resolve that she intended to take up where her husband had left off, “yet I feel so very incompetent for the task that my courage sometimes completely fails me—I understand Henry’s plans and his manner of working, perhaps better than anyone else, but I could not get along without an assistant. . . . He wished to get the spectra of several of the winter stars, and this we intended this winter. It is so hard that he should be taken away just as he had arranged all of his affairs to have the time to do the work he really enjoyed, and in which he could have accomplished so much.”
Pickering urged Anna to publish Henry’s spectroscopic results, even if incomplete. The measuring engine at Harvard could be used to speedily determine the wavelengths of the spectral lines. In early February 1883, Anna traveled to Cambridge and handed Pickering twenty-one photographs of stellar spectra taken at Hastings. A month later, wavelengths in hand, Pickering sent Anna a draft of a paper that might be published under Henry’s name. With an eye toward England, he asked permission to present the results at the April 11 meeting of the American Academy of Arts and Sciences: “It is not necessary that the paper should be completed as long as the observations are finished. The work will take the date from the time of presentation, and
of course it is desirable that this should be as early as possible.” Anna Draper agreed to the presentation, but held out for an extended written volume that would serve as a proper legacy to her husband. Henry Draper’s Researches on Astronomical Spectrum Photography appeared in February 1884. The monograph contained seventy-eight stellar spectra and Pickering’s wavelength measurements for twenty-one. At Anna’s request, there was an introduction by Charles Young, excerpts from Henry’s observing logs, and pictures of their observatory in Hastings. Copies of the book were distributed to a roster of prominent astronomers—including William Huggins.
Huggins dashed off a diatribe to Pickering on March 12, 1884, indicating that the published wavelengths were “very wild indeed. . . . There can be no doubt whatever that your [measurements are] quite inaccurate and increasingly so as the wavelengths get smaller. . . . It would be a great drawback to the advancement of science if your measures were allowed to pass without criticism on this point, but as it is obviously of the nature of a slip, I should be so glad if you would yourself look into the matter and yourself publish the correction as soon as you find out the cause.”
Pickering forwarded the letter to Anna Draper. “Fortunately,” he assures her, “Dr. Huggins’ arguments, that results must be wrong if they do not agree with his, will not generally be regarded as conclusive unless supported by facts. . . . [W]e must conclude that American stars are more refractory or less refrangible than English ones.” On the same day, he fired off a reply to Huggins, indicating that he would re-examine the Harvard data, but also scolding Huggins for failure to publish details of his own measurement process, as Pickering himself had: “Your publication [of 1880] does not enable the reader to verify your reductions, and he has no means of checking your results. I hope you will supply this deficiency by a fuller publication and thus remove what seems to me the weakest part in your very valuable contribution to the subject.”
Undaunted, Huggins shot back in mid-April 1884, “I cannot tell whether you have been led astray by your formula or by measuring wrong lines, probably the photographs were very badly defined, or there may have been some shift in the apparatus. Some of our best men to whom I have shown your paper do not seem to think your method serious enough to make it worth while for me to take any notice of the paper. . . . I have written to you plainly indeed, because I felt you would take it as a mark of friendly feeling.” This letter, too, Pickering forwarded to Anna Draper, who wrote with evident sympathy, “I felt very sorry that you should have been subjected to such an ungentlemanly attack, through your interest in Dr. Draper’s work. If Dr. Huggins did not find the results of your measures agreed with his, there was no objection to his saying so if he had expressed himself in a more courteous manner.”
The transatlantic volleys ceased at this point, neither man willing to concede error. (According to modern measurements, both Huggins and Pickering were off the mark.) Pickering continued to develop the celestial photography program at Harvard, despite a chronic shortage of funds. In 1885, he conducted tests of an objective prism spectroscope, which situates a large prism in front of the telescope’s main lens. With such a configuration, the spectra of dozens or even hundreds of stars are captured at once on a single photographic plate. A five-minute exposure sufficed to record the spectrum of every star visible to the naked eye in a swath of sky ten degrees on a side. The spectra of forty stars in the Pleiades cluster were obtained in thirty-four minutes, a task that might take weeks for a conventional, single-star spectroscope. There was a price to the mass-collection process: the spectral images were tiny, each about half a millimeter long. (Henry Draper’s spectra stretched some ten times that length.) Nevertheless, under microscopic scrutiny, Pickering could identify all the major absorption lines and contemplate the large-scale classification of the stellar species.
Anna Draper tried to continue her husband’s work herself, but was unable to find a suitable assistant. On February 14, 1886, she gave Harvard $1,000 to support Pickering’s stellar spectrum program, with the promise of more funds in the future. Pickering promptly mailed out a circular to astronomers and scientific institutions worldwide announcing the creation of the Henry Draper Memorial, whose initial object was to photograph the spectrum of every star visible to the unaided eye from Cambridge, Massachusetts. By summer’s end, Pickering had taken a total of 224 plates depicting some six thousand spectral images of around three thousand stars. (The plates overlap in their sky coverage, so most stars appear twice.) The photographs were fine-grained enough that substantial enlargement preserved details seen in the minuscule originals. Anna Draper was thrilled with the quality of the Harvard spectra. With undisguised relish, she mused to Pickering, “I wonder what Mr. Huggins will say when he sees them.”
Pickering knew what William Huggins would say. How could he fail to be impressed, even intimidated, by the profusion of crisp spectral images? But Pickering knew, too, that his competition lay not in the garden-based facilities of zealous amateurs, but in the well-endowed observatories of academic and governmental institutions. In a letter dated January 18, 1887, he alerted Anna Draper to the likelihood that Harvard’s success would “induce other astronomers to undertake the same work.” He offered a strategy to preempt competition by expanding the Draper project to fainter stars and to the Southern Hemisphere sky. Draper replied within the week, “I quite agree with you in feeling that I should like to appropriate the entire ground that is possible.” She raised her financial stake in the project and subsequently transferred the Hastings Observatory eleven-inch refractor and twenty-eight-inch reflector to Harvard.
William Huggins learned of the Henry Draper Memorial in May 1887. Pickering’s announcement had its intended effect. “I have just received a paper from Harvard Observatory,” Huggins wrote to mathematician George Stokes at Cambridge University, “& there, through the large endowment of Mrs. Draper the photography of star spectra is to be carried on upon a magnificent scale. Three large instruments are to be kept at work all through the night by relays of photographers & photographs to be enlarged by special methods & measured by other men. The question is, is it worth my while to continue working in this direction now that it is being done under circumstances with which no zeal & perseverance on my part will enable me to be in an equal position.”
Huggins pivoted his spectroscopic research to avoid overlap with the ongoing work at Harvard. He and Margaret concentrated on the ultraviolet part of the stellar spectrum, while pursuing further studies of the Orion Nebula and the problematic nebulium line. In 1889 came the first of many Tulse Hill publications listing Margaret as coauthor. She was awarded honorary membership in the Royal Astronomical Society in 1903. (Full membership was denied women until 1915.) William and Margaret Huggins will be remembered as pioneers, if not the parents, of celestial spectroscopy. Their scientific achievements, combined with those of others, established the foundation of stellar and nebular spectrum photography.
The first substantive result of Anna Draper’s largesse was the publication in 1890 of the Draper Catalogue of Stellar Spectra, listing the position, brightness, and spectral type of 10,351 stars. By far the most extensive star compilation to date, it was the opening salvo in Pickering’s plan to monopolize large-scale photographic mapping and spectroscopy of the stellar realm. To generate the catalog, Pickering had hired a photographer to take the plates, and Williamina Fleming, his former housemaid, to measure and classify the spectra. (Fleming, a Scottish schoolteacher, immigrated to Boston in 1878 with her husband, who abandoned her and their unborn child. Pickering hired Fleming as a domestic while she was pregnant, then offered her part-time clerical work at the observatory. In 1881, he added her to the permanent staff.)
Because the photographs captured more spectroscopic detail than visual observations, Pickering and Fleming developed an alphabetic spectral classification system to replace the 1860s-era, Roman-numeral scheme of Italian spectrum pioneer Angelo Secchi. (Secchi’s I through IV became Harvard’s A through N.)
The system was revised during the 1890s by Annie J. Cannon, a Wellesley physics graduate hired to reduce the plates from Harvard’s new high-altitude observatory at Arequipa, Peru. Cannon went on to lead a team of female assistants in the completion of the Henry Draper Catalogue, published in nine volumes between 1918 and 1924, and containing more than 225,000 stars. (This sort of low-wage work was virtually the only career option for women in astronomy.) As Pickering had hoped, Harvard’s spectral classification system was officially adopted by the astronomical community; an expanded form is still in use today.
The goal of spectral classification is to seek commonalities among stars, to whittle down a vast and diverse population into a manageable number of species. And while discoveries were made during the inspection of spectra for the Draper catalog, these were incidental to the project’s purpose: to provide astronomers with an essential stellar database they might use to explore the physical nature of stars. To Pickering, Cannon, Fleming, and the other spectroscopic analysts, the various spectral line patterns were nothing more than a visual means of comparison, of sorting stars into categories based on a particular facet of their light. The underlying physics, as yet unknown, was immaterial to the work. After eyeing thousands of spectra, Annie Cannon recognized an overall progression in the intensity of certain line patterns, and within that progression, she found subtler subprogressions. On that basis, she merged and shuffled Williamina Fleming’s original alphabetical arrangement of spectral types into the seemingly random sequence O, B, A, F, G, K, and M. Each letter category was broken into subdivisions indicated by a numerical suffix 0 through 9, as in G2, the spectral type of the Sun. This purely taxonomic arrangement was later proven to parallel the range of stellar surface temperatures, from high to low. Secondary correlations with stellar luminosity were discovered in the early twentieth century, leading to the development of the Hertzsprung-Russell diagram, a key graphical tool used in exploring the physics and evolution of stars.
Starlight Detectives Page 26
