The Perfect Machine (13 page)

Casting the mirror of the one-hundred-inch telescope had pressed plate-glass technology to the limit. The largest glass furnaces in the world had required three pours to produce enough plate glass for the one-hundred-inch mirror. There weren’t enough furnaces in France to melt the glass for a three-hundred-inch mirror all at once, and even if
a mirror that huge could be poured, it would require an astronomer’s working lifetime to anneal a plate-glass mirror that immense.

A mounting to support and point a mirror that large presented a comparable challenge. At the longest focus, the focal length of the telescope would be more than two hundred feet. A skyscraper of comparable dimensions sways inches; the tower of a suspension bridge may sway twenty feet. The telescope would have to be rigid enough to maintain the alignment of that long shaft of light on a spectrograph slit just one-thousandth of an inch wide, even as the telescope itself moved to follow the sidereal motions of the object across the heavens.

The sixty- and one-hundred-inch telescopes could be floated on mercury bearings, to provide low-friction precision motion. But there wasn’t enough mercury available to support the volume of the telescope Pease had sketched. Even a freshman engineering major could calculate that the huge roller bearings Pease included in his drawing would entail so much friction that vibration-free movement of the machine would be impossible.

Although the engineers and astronomers were aware of these limitations, the reporters refused to understand that Pease’s sketches weren’t meant to be a working model. They were a pipe dream, a concept the astronomers batted around the way managers of sports teams fervently discuss the dream team they know they can never put together. As enticing as the huge machine was, Hale and his colleagues at Mount Wilson had enough experience building large telescopes to know that every step of the process would be several orders of magnitude more difficult, more time-consuming, and more expensive than anyone had ever anticipated.

Still, the dream tantalized the astronomers and cosmologists. For even as Hubble and others were expanding the known realm by careful use of the one-hundred- and sixty-inch telescopes, theoreticians in astrophysics and cosmology had begun raising questions about the very structure of the universe that cried out for investigation, demanding observations deeper into space than any existing telescope could reach.

In the years after the publication of his General Theory of Relativity in 1915, Einstein and others began to explore the cosmological implications of his revolutionary theory of gravity. One conclusion of Einstein’s theory was that the universe could not be static: It had either to be expanding or contracting. Yet there was no observational evidence for this novel and disturbing idea. For centuries, with increasingly sophisticated and precise instruments, astronomers had measured the proper motions and radial velocities of stars. All the data, gathered with the finest equipment and the most careful analysis, suggested that the stars wandered more or less randomly through
space. Einstein concluded that there was something wrong with his theory and included a term in the field equations of gravity that he called the cosmological constant, or A, which was designed to make the radius of the universe hold steady with time. The cosmological constant was a “fudge factor,” with no known physical basis. Einstein readily admitted that it was introduced “only for the purpose of making possible a quasi-static distribution of matter, as required by the fact of the small velocities of the stars.”

In fact the inelegant fudge factor didn’t even accomplish its avowed purpose. The Russian mathematician Aleksandr Friedmann discovered that Einstein had made an algebraic error in introducing the term. Friedmann threw out the cosmological constant. Without A, a universe according to the General Theory of Relativity had to be continually expanding or contracting.

The same year that Einstein had reluctantly introduced the cosmological constant, an American astronomer named Vesto Slipher published a remarkable paper with evidence that the constant might not be needed after all. Slipher was on the staff of Percival Lowell’s observatory in Flagstaff, Arizona, where Lowell pursued his much-publicized efforts to prove there were canals on the surface of Mars. Lowell gave Slipher the assignment of taking spectra of spiral nebulae, in the hope that measurements of the rotation of the nebulae would prove that the spirals were the early stages of solar systems like our own, a theory Laplace had advanced years before and to which few astronomers other than Lowell subscribed.

Using a new spectrograph with a very fast lens on the twenty-four-inch telescope at the Lowell Observatory, Slipher found the rotations Lowell sought—which we now know are the spiral motion of the billions of stars that make up each spiral nebulae. Slipher also found that the spectral lines of most of the spiral nebulae were shifted toward the red end of the spectra. The only reasonable explanation for this displacement was a Doppler shift, the same phenomenon we notice in the pitch of a car horn as the car drives away from or toward us. As the car approaches, the pitch of the horn seems to rise; it seems to drop as the car drives away. The same shifts in color or wavelength occur to the light emitted by a body when it is moving away from or toward the observer. Astronomers had long made use of Doppler shifts to measure stellar velocities. In fact it had been Doppler shift measurements, which asserted the randomness of the motions of stars in the Milky Way that had prompted Einstein to put the cosmological constant into his equations in the first place.

But the velocities of the spiral nebulae Slipher measured were not random. Twenty-one of the twenty-five spirals for which he had spectra were displaced toward the red, indicating that they were flying away from each other and from the earth. The displacements were
enormous, suggesting that the spirals were moving at speeds that had never before been measured.
^*
By 1924 a trio of astronomers—Carl Wirtz in Germany, Knut Lundmark in Sweden, and Ludwik Silberstein in the United States—had assembled evidence that the smallest and presumably most distant spiral nebulae were receding from the Milky Way faster than the nearer ones. As provocative as this material was, the data were marginal. The velocities the astronomers claimed to have measured seemed too great to believe.

But to Edwin Hubble, who had been able to use the Cepheid variable stars he had resolved in the Andromeda and Triangulum spiral galaxies as a yardstick to estimate the distance to those galaxies, Slipher’s observations suggested a whole new line of inquiry. If these galaxies were actually receding, the universe was expanding—each galaxy or island universe moving away from the others at incredible velocities. The cosmological implications were exciting, but Hubble was not a man to shoot from the hip. What was needed was observations, data. He proposed an extensive program to measure the red shifts of galaxies within range of the one-hundred-inch telescope. His program would require a demanding schedule of observations, far more than one man could ever manage. By 1928 Hubble had a talented assistant and, ultimately, collaborator to aid him in his observations.

Milton Humason had been at Mount Wilson long before Hubble arrived. He was originally hired as a mule packer, trekking material up the steep path to the observatory. After years on the path he was promoted to busboy in the dining room and then to observatory janitor. Humason’s eighth-grade education didn’t stop him from asking questions of the observers. “I decided the only way to stay awake at night was to get interested in what was going on,” Humason said. “So I did.” Shapley and some of the others were so impressed by Humason’s intelligence that they urged his promotion to night assistant on one of the smaller telescopes. Over the years Humason was promoted to minor observational assignments on the six-inch and then the ten-inch telescope on Mount Wilson. Finally Hubble asked Humason to assist him with observations on the one-hundred-inch telescope.

Traditionally there had been a chasm of education and status between the observatory staff and the observers. Observers were professors of astronomy, usually with Ph.D.s from eastern universities. They brought tweed jackets, pipes, common-room-sherry habits, haughty accents, and the latest research to Pasadena and Mount Wilson. The observatory staff were onetime farmers, truck drivers, and
mechanics, men who were comfortable living year round on a lonely mountaintop where the winter temperatures ranged from painful to unbearable and where the isolation drove many men to drink despite the observatory bans. Humason, who had come up through the ranks, remained one of the boys. He chewed tobacco, played poker and billiards, and went fishing with the night assistants on his days off. He was also a patient and talented observer. Before long he was doing the bulk of the actual observation work on the big telescope. Hubble, his ever-present pipe clenched in his teeth, pored over the plates.

As Hubble and Humason began to accumulate data on red shifts, they attempted spectrograms of fainter and fainter nebulae, pressing the telescope to the limits of its range. A spectrograph requires that the observer keep the light of the distant object on a narrow slit. For faint nebulae the exposure needed to gather enough light might run to eight hours or more. Sometimes a single observing session wasn’t long enough to gather the light, so the spectra had to be made over several nights. To reduce the needed exposure times and extend the range of the telescope, they tried a new corrective lens on the one-hundred-inch telescope, built by W. B. Rayton at Bausch & Lomb. Working on the principles of a reversed microscope objective, the lens achieved the very fast focal ratio of
f
/0.6, making it possible to record spectra for extremely faint, distant nebulae.

The observation procedures were demanding. For some objects the observer would have to twist and contort himself to be able to see through the guidescope. For eight or more hours, the observer would struggle to keep a barely visible image centered on the slit of the spectrograph. The temperatures were often cold enough to freeze breaths to mustaches and skin. The weight-driven clock for the right ascension drive of the one-hundred-inch telescope chimed whenever Humason pushed a correction button, and the balky controls sometimes had him leaning on the telescope or even climbing onto it to muscle the tube into the exact position he wanted. But Humason had the patience and persistence of a mule driver, and he got results—tiny glass plates fifteen millimeters square, with a spectrogram, only millimeters long, of the distant spiral galaxy.

When Hubble and Humason compared the lines on these spectrograms with the same lines on a calibration spectrum, they could measure how much the spectrum of the galaxy had shifted. The magnitude of the shift corresponded to the velocity at which the galaxy was receding. By 1928 Hubble and Humason had accumulated enough data to begin the search for a relationship between the magnitude of the red shifts of these distant galaxies and their distance. As Hubble plotted each new spectrum, he discerned a linear correlation between the receding velocity of the galaxies, as measured by the red shifts, and their distance—at least for the few galaxies whose distance he had been able to determine by the identification of the brightest resolved stars. The relationship was elegantly
simple: The more distant the galaxy, the faster it was receding.

His earlier yardstick, the Cepheid variables he had discovered in nearby galaxies, required that the galaxy be close enough to resolve individual stars. Because red shifts were measured from the light of the galaxy as a whole, the reach of the telescope was suddenly extended a thousand fold. As equipment and technique improved, Hubble and Humason could record usable spectra of galaxies so faint that the galaxy could barely be resolved in the telescope. By combining the techniques—extrapolating from the distances to bright, nearby galaxies, which he could determine from the Cepheid variables, and using his linear relationship between the red shift, or speed of the receding galaxy, and its distance to measure the distance to galaxies that appeared only as a diffuse blob on photographic plates—Hubble had extended his yardstick to reach to the limits of the observable universe.

Hubble told Hale and his other colleagues about the work, but he hadn’t read much of Einstein. Although Hale had corresponded with Einstein, he confessed his ignorance: “The complications of the theory of relativity are altogether too much for my comprehension…. I fear it will always remain beyond my grasp.” Without a sound cosmological explanation for the data he had found, Hubble was reluctant to publish his findings too quickly.

Unknown to Hubble, a Belgian priest and mathematician named George Lemaitre had already drawn a connection between Slipher’s red shifts and Einstein’s theory of relativity. Lemaitre had made a tour of the United States, attending astronomy conferences. He learned of Slipher’s red-shift data and, on his return to Brussels, in 1927, wrote a paper that connected the observed red shifts with the expanding universe predicted by general relativity. But Lemaitre was an outsider. His paper was published in an obscure journal, and he got a brush-off from astrophysicists when he tried to present his views at conferences.

Cosmology hadn’t caught up yet, but Hubble was certain that he was onto a very promising direction of research. He concentrated on what he could supply—measurements of red shifts that would reach farther into the depths of the universe. By 1931 Hubble and Humason had measured red shifts of 20,000 km/h, overwhelming evidence for the linear relationship between red shifts and distance. Humason was soon training the telescope on nebulae so distant that they could not be seen at the Cassegrain focus of the telescope. Humason’s procedure for taking a spectrogram on these faint nebulae was to calibrate the slit of the spectrograph on known stars and then to move it by measured amounts (determined from long-exposure direct photographs) to the positions of the unseen nebulae. “The observations thus extend nearly to the extreme limit of existing equipment,” Hubble wrote. “No very significant advances are expected until larger telescopes are constructed.”