The Glass Universe

by Dava Sobel

  “My adventures began before the train left Boston,” Pickering wrote in his travel diary on Saturday, August 20, 1910. “The porters could not tell me in which car was my drawing room, and in passing from one car to another I got locked in between them! As the [William] Pickerings and Professor Bailey waved me a cheerful farewell, they did not know that I was a prisoner, alone in a glass cell, from which I could not escape.”

  Everyone had benefited from the talks at the three-day Harvard meeting just completed. Six of the distinguished foreign visitors had been elected, at their express request, to membership in the society. All enjoyed the way Pickering interspersed the technical sessions with appropriate diversions, such as the group excursion to the Harvard-affiliated Blue Hill Meteorological Observatory in Milton on Wednesday afternoon, and also the trip to the Whitin Observatory at Wellesley College on Thursday. On Friday, mindful of everyone’s fatigue, Pickering moved the body only as far as the Harvard Students’ Astronomical Laboratory in the Yard. Throughout the week, at all hours, his staff had ushered interested visitors to any on-site area they wished to see, from the telescope domes to the astrophotographical library in the Brick Building. Pickering wrote in his diary that he thought he might sleep for three days on the train West, but in fact he had a full agenda of important committee work to do en route.

  Pickering’s expertise in photography and photometry allied him with two major European stellar mapping projects, one headquartered in Paris and the other in Groningen. The moment had come for each of these projects to select a standard reference for photographic magnitudes. Pickering wanted to see a single standard applied to both efforts, and he wanted that standard to be the Revised Harvard Photometry. As chairman of the Committee on Photographic Magnitudes of the Astrographic Chart Conference (the Parisian effort, also known as the Carte du Ciel), Pickering held considerable power, but other photometric standards had been devised, and the issue was to be decided by vote. Harvard’s chief competition came from Pickering’s fellow committee member Karl Schwarzschild, who had authored his own Potsdam standards of photographic photometry. As it happened, Schwarzschild was on the train with Pickering. So were committee members Herbert Hall Turner of Oxford and Oskar Backlund of Pulkovo, making a quorum. Indeed, the entire contingent of traveling astronomers was conveniently confined to two specially hired railcars.

  On Sunday, August 21, 1910, they reached “Niagara, in which no one is disappointed,” Pickering wrote. “The roar of the Falls, interrupted only by astronomical talk. Informal committee meetings whenever I sit still. In the morning a carriage ride to Goat Island, and the wonderful electric railway from which you see the entire river. In the afternoon the Maid of the Mist (steamer, not young lady) and an impressive view of the American Falls, from its base (the best I have seen). My coat got so wet that I had to keep my back to the Sun, to dry it.”

  They got to Chicago on Monday, toured the parks and the university physics laboratories, and were joined by several more astronomers when they reboarded the train at nightfall. John Stanley Plaskett of the Dominion Observatory in Ottawa, who made his own record of the trip highlights, relished the way the group “traveled across the continent in two special cars and in the eight days occupied in the journey became almost like a family party.”

  On Tuesday, August 23, after a long morning talk with Turner, Pickering called a meeting of the photographic magnitudes committee in his drawing room. “Backlund, Schwarzschild, Turner, and I discussed the matter for two hours, so busily that we did not find out that it was extremely hot until the meeting was over. Temperature 102 in the shade. The thermometer went down when you put the bulb in your mouth! At an open window the breeze was hot like that from a furnace. We all suffered, and several ladies were ill.” Many of the visiting astronomers were accompanied by their wives, and Harvard’s own well-known Mrs. Fleming was also aboard the train.

  The next day, Wednesday the twenty-fourth, Pickering worked all morning to complete his part of the committee report on photographic magnitudes before another rolling meeting convened at three. This one included an additional member, Edwin Brant Frost of the Yerkes Observatory. Pickering re-created the drama later in his journal, in the present tense: “They don’t want to come, as the thermometer is nearly 100, and point to Turner, who is asleep. I wake him up, and make them all attend the meeting in my drawing room. It is so hot they cannot contribute their portions of the report. As a result of our labors (and heavy labors, too) we all agree on a system of photographic magnitudes which will probably be the system of the world. I am repaid for my journey of two thousand miles, had I done nothing else. Astronomers very kind and complimentary, and Schwarzschild gives up his (Potsdam) system and accepts that of Harvard. My part in this will be regarded as one of the most important things I have ever done.” Thus the acceptance of the Harvard Photometry standard, one of Pickering’s top goals for the trip, became a fait accompli before the train crossed the Great Divide.

  In Flagstaff, Arizona, on Thursday, Percival and Constance Lowell guided the visiting astronomers through the Lowell Observatory, then waved them on to the scenic wonder of the American West: “Saturday, August 27. Walk in morning to another point on the rim of the Grand Canyon. Prepare six copies of a third draft of the report on photographic magnitudes with the help of the hotel typewriter. Leave in evening for Pasadena.” Pickering, flush with his success at advancing the Harvard photographic standards for photometry, could only hope that the Draper system of stellar classification by spectra would fare as well in the imminent contest for international approval.

  Over the five decades since Father Secchi grouped the stars visually by color and a few spectral lines, classification schemes had proliferated. Harvard alone had produced two—or three, depending on how one counted Miss Cannon’s modifications of Mrs. Fleming’s original Draper catalogue. A veritable babel of terminology prevailed. To make himself perfectly clear when addressing other astronomers, Pickering often translated Harvard designations into the simpler Secchi names, describing one of Miss Cannon’s F 5 G stars, for example, as belonging to Secchi’s second class (a spectrum crowded with many lines). Secchi’s system, familiar as a Latin grammar, lacked the vocabulary to describe all the spectral distinctions revealed through photography and modern analytic techniques. Astronomers knew they could improve their communications by choosing one classification system to abide by, or else by creating a new hybrid. The issue was due to arise at Mount Wilson, when the Solar Union debated broadening its mandate to include other stars.

  Wilted from the heat of the Mojave Desert, the astronomers reached Pasadena late Sunday afternoon, August 28, 1910, and checked into the Hotel Maryland. The collegial body that had banded together in Boston, then expanded in Chicago, now melded with West Coast residents and Solar Union delegates newly arrived from as far away as Japan. Eighty-seven strong, the attendees represented thirteen countries and fifty observatories at the largest gathering of astronomers ever assembled.

  “Monday, August 29. The fourth anniversary.” No doubt Pickering would continue to mark the date of Lizzie’s death until the day of his own. This year he passed the grim occasion in good company, visiting the offices, laboratory, and machine shop of the Mount Wilson Solar Observatory. The facilities occupied a one-story concrete building in town, where Hale joined the group to describe the fabrication of the unique instruments they would see over the next few days on his mountaintop. At the afternoon garden party given by Hale and his wife, Evelina Conklin Hale, the astronomers met some of Pasadena’s most influential citizens.

  It took all of Tuesday to reach the summit of Mount Wilson. A few of the astronomers, though dressed in suits, ties, and derby hats, mounted saddle horses and mules for the ascent. Others chose to walk. Most, including Pickering and Mrs. Fleming, rode up in carriages. “Several dangerous turns in the road, at one of which we all had to get out. Road so narrow that there is no chance for teams to pass. Outer wheels within a
foot of the edge (and death) for a large part of the way.” Those who could bear to look down extolled the view of orange groves and vineyards in the valley.

  At the top, Pickering, hoarse from days of talking and the alkali dust of the desert, retreated to the one-room cottage assigned to him. “Living very primitive, but comfortable. No appliances for blacking boots, which are always white with dust, instead of black. A feather duster is used in this part of the country instead of a blacking brush. My greatest wants are a cow and a bath tub. Water is scarce, and milk much more so, as there is no grass on top, and all the fodder must be hauled up the mountain.” The recently visited Lowell Observatory, in contrast, had accommodated a dairy cow named Venus.

  Most topics of discussion at the Solar Union’s plenary sessions pertained specifically to the Sun, naturally enough, in a mix of English, French, and German. Not until the afternoon of the final day, Friday, did the solar scientists vote, unanimously, in favor of extending their studies to the stars and formally considering the question of stellar classification.

  “A committee of fourteen is appointed and I am made chairman (they kindly say ‘of course’). I rise to express my thanks, and ask the members of the committee to remain after we adjourned so that we might begin work at once. A shout of laughter goes up as all have heard of our meeting with the thermometer at 100°.”

  Undaunted, everyone present who had been named to the new committee stayed put as Pickering requested, and listened to him tell the story of the Henry Draper classification. He described how the letters of its alphabet had strayed from the usual order into Miss Cannon’s arrangement, in which each category seemed to define a different stage in the life of a star. Pickering did not press for the system’s acceptance. He anticipated many more discussions before the committee, let alone the Solar Union as a whole, reached consensus on classification. For now he merely wished to acquaint his fellows with the system he knew best, and hear their ideas on how to proceed.

  The first to speak, Mount Wilson deputy director Walter Sydney Adams, testified loudly in favor of the Draper system. The ensuing discussion soon proved that most members shared his good opinion of it. “As much to my surprise as to that of the others,” Pickering marveled in his diary, “practically everyone approved our system, so that instead of an attempt to replace it, it received the strongest endorsement I could have desired.”

  CHAPTER NINE

  Miss Leavitt’s Relationship

  THE EASTBOUND TRAIN TO BOSTON, lacking any reserved cars for astronomers, afforded Pickering few opportunities for politicking. Nevertheless he managed to meet briefly between San Francisco and Denver with two members of his new Committee on the Classification of Stellar Spectra. Together they constructed a questionnaire for polling their peers about the pros and cons of the Draper system. Although the full committee favored the Draper classification, some wanted to modify it, a little or a lot, before proposing its formal adoption at the next meeting of the Solar Union, three years hence, in Bonn.

  The beauty of the Draper nomenclature lay in the richness of its data. Harvard’s Draper Memorial catalogues comprised more than thirty thousand stars, a claim that no other classification could make. The great number of stars falling into the relatively small number of categories affirmed the system’s validity. Its level of complexity struck a pleasing compromise between the minimalism of Secchi and the minutiae of Miss Maury. Moreover, it depended entirely upon observable differences, without defending a particular theory.

  Not theorizing had been a point of honor with Pickering from the outset. By 1910, however, young astrophysicists chafed to embrace theory. The ideal classification system must be rigid enough to guide and support new research, yet fluid enough to contain conflicting ideas about the dynamics, distribution, and evolution of stars.

  In November committee secretary Frank Schlesinger of the Allegheny Observatory in Pennsylvania sent out the questionnaire he had helped draft on the train. It went to all fifteen committee members and about as many nonmembers selected for their strong interest or expertise in classification—notably Annie Cannon, Williamina Fleming, Antonia Maury, and Ejnar Hertzsprung, the Danish astronomer who had so emphatically endorsed Miss Maury’s approach.

  The questionnaire began with a recapitulation of the committee’s impromptu meeting on Mount Wilson at the close of the Solar Union conference. Given the fact that all present had smiled on the Draper classification as the most useful ever proposed, the first question asked, “Do you concur in this opinion? If not, what system do you prefer?”

  Answers trickling in over the next several months overwhelmingly favored the Draper system, as might have been predicted. Even Hertzsprung endorsed it, though he called for specific improvements in response to question number two: “In any case, what objections to the Draper Classification have come to your notice and what modifications do you suggest?”

  Here a few astronomers took aim at the system’s alphabetical names. Mundane labels such as B and A, they felt, failed to conjure any helpful images. In contrast, the system fashioned by Norman Lockyer in 1899 applied the name of a typical star in each category to the category as a whole. Procyon, for example, a yellowish star in the Little Dog constellation, defined Lockyer’s Procyonian division—a cumbrous but evocative term.

  Neither Pickering nor Mrs. Fleming had viewed the alphabet letters as permanent fixtures when they introduced them, but rather as neutral symbols, easily replaced with meaningful names once meanings emerged. Even so, years of use had imbued the letters with significance. At Harvard, at least, the mention of A instantly called to mind an alpha star like Altair in the Eagle constellation, with blue-white light and a spectrum of unadulterated hydrogen lines.

  Among those astronomers content with capital letters, a few regretted the lack of alphabetical order in the Draper system. They thought the progression O, B, A, F, G, K, M looked grotesque or random, as though signifying nothing. Miss Maury roundly rejected it—not simply on aesthetic grounds, but because she had convinced herself the categories, as now arranged, represented the true course of stellar evolution. The “overwhelming predominance” of types O and B in the nebulous regions of Orion and the Pleiades, she told the committee, proved that stars were born from gaseous nebulae in a blue-white heat. As stars aged, they cooled, faded to white, then yellow, finally ending their days in red senescence. The letters or numbers affixed to each stage, therefore, should reflect the seamless flow of stellar life.

  Astronomers who shared Miss Maury’s evolutionary view commonly spoke of “early stars” when they meant white ones, and called the red stars “late.” Those opposed clung to the color words and cautioned against hitching classification to an evolutionary theory. Henry Norris Russell of Princeton, the youngest member of Pickering’s committee, envisioned a different evolutionary path from the one Miss Maury described. Russell thought stars might start off red, warm up to yellow or white, and then cool down to red again. He further theorized that stars led different lives depending on their birth weight, and that only the most massive ever achieved the highest temperatures.

  “The Draper Classification seems to me all the better because the letters are not in alphabetical order,” Russell declared. “This helps to keep the novice from thinking that it is based on some theory of evolution.” Apparently the alphabet could flout its own order and still remain effective—or even improve its utility—as a labeling scheme. Pickering could see that much on his typewriter’s QWERTY keyboard.

  The third of the questionnaire’s five questions contained three parts: “Do you think it would be wise for this committee to recommend at this time or in the near future any system of classification for universal adoption? If not, what additional observations or other work do you deem necessary before such recommendations should be made? Would you be willing to take part in this work?”

  The mixed reactions to this question crossed party lines. Some of the most outspo
ken boosters of the Draper system hesitated to press for its formal adoption, fearing the time not yet ripe. The Draper classification surely surpassed all competition, but perhaps something grander might yet be created in its place.

  Committee member Edwin Frost of the Yerkes Observatory had long dreamt of a classification system modeled on the ones for plants and animals, dividing the kingdom of the heavens into phyla, classes, genera, and species, all with Latin names. He still hoped astronomers would set that sort of system as a future goal. For the time being, however, Frost thought it foolhardy to tamper with the Draper classification, especially given Pickering’s personality. “With his habitual courteous consideration for the views of others,” Frost warned in his questionnaire response, “Director Pickering might adopt those of the suggestions made with some unanimity, and then we should have still another classification to add to the present confusion.”

  Question four concerned a single detail: “Do you think it desirable to include in the classification some symbol that would indicate the width of the lines, as was done by Miss Maury in Annals of the Harvard College Observatory, Volume 28?” This question, too, elicited oddly divided opinions. Both Miss Cannon and Mrs. Fleming gave a qualified yes. Miss Cannon pointed out that such distinctions applied to only a small fraction of stars studied. Mrs. Fleming welcomed any symbol that would obviate the need for extensive remarks.

  The open-ended final question asked, “What other criteria for classification would you suggest?” Answers ranged widely, but the most common answer was no answer.

  When Pickering apprised Mrs. Draper of the growing acclaim for her husband’s namesake system, she declared the case “a triumph”—the well-deserved response to the years of labor the director had devoted to the classification and the thought he had expended upon it. She was happy for him, she said, and happy, too, for the sake of Henry’s memory.