Starlight Detectives

by Alan Hirshfeld

  Using Harlow Shapley’s published calibration of the period-luminosity law for Cepheids in the Milky Way, Hubble computed the distance to Andromeda’s lone stellar milepost: one million light-years. He weighed the various factors that might alter or even nullify this gross inflation of the cosmic landscape. There was scant evidence of significant light extinction along the line of sight, which would make the star appear dimmer, hence, farther, than it really is. Furthermore, regions of sky adjacent to the Andromeda Nebula were devoid of Cepheids; thus, the newfound variable most likely resided within Andromeda and was not a stellar interloper projected against it. (Subsequent observations confirmed this assertion.)

  Edwin Hubble’s photographic negative H335 of the Andromeda nebula, dated October 6, 1923. The label “VAR!” marks his discovery of a Cepheid variable star.

  At base, Hubble’s distance estimate rested on the belief that Cepheids in Andromeda are equivalent to their Milky Way compeers—that Shapley’s period-luminosity calibration holds wherever Cepheids are found. Like numerous others before him, Hubble postulated a uniformity of nature, which, he explains, “seems to rule undisturbed in this remote region of space. This principle is the fundamental assumption in all extrapolations beyond the limits of known and observable data, and speculations which follow its guide are legitimate until they become self-contradictory.”

  Hubble’s distance estimate for Andromeda was a thunderous blow to proponents of the single-galaxy cosmos. The night sky’s most prominent spiral nebula, he asserted, lies well beyond the borders of the Milky Way, even Shapley’s hyper-extended Big Galaxy version. Thus, Andromeda is a galaxy unto itself, subsumed by no other, its dimensions and stellar population comparable to, if not larger than, our own. Can the thousands of fainter spirals be anything other than independent Milky Ways and Andromedas strewn throughout the vastness of space?

  In late February 1924, Hubble alerted Shapley to his discovery of a Cepheid in the Andromeda Nebula. Included in the envelope was a copy of the star’s telltale light-curve. Sitting in his office with his graduate student, Shapley lamented, “Here is the letter that has destroyed my Universe.” Nevertheless, he replied with a critique of Hubble’s assumptions and measurements. The purported Cepheid, he insisted, must be anomalous, if it exists at all; spurious variable stars can appear whenever plates of different exposure are compared.

  Hubble dismissed Shapley’s allegations of error, yet he understood that his controversial distance estimate could not rest on the measurement of a single star. He continued to mine Andromeda for variables, and simultaneously delved more intensively into the loosely wound spiral Messier 33 in Triangulum and the irregular nebula NGC 6822 in Sagittarius. By year’s end, having hardly tapped Andromeda’s rich vein of variables, he had raised the Cepheid count to a dozen; Messier 33 eventually yielded up thirty-five Cepheids, and NGC 6822 another eleven. Without exception, the stars confirmed the extragalactic nature of their host systems. The corresponding luminosities he derived for other key stellar components, such as novae and prominent blue stars, cohered with those found in the Milky Way and the Magellanic Clouds. This mutual consistency girded the reasonableness of the island-universe model.

  While Hubble doggedly searched for extragalactic Cepheids, word of his transformative measurements flooded the professional grapevine. Even before publication, Princeton’s Henry Norris Russell, dean of American stellar astrophysics and a lapsed Shapley ally, pronounced Hubble’s work “undoubtedly among the most notable scientific advances of the year.” The New York Times alerted its readers on November 23, 1924, that “‘Island Universes’ Similar to Our Own” had been discovered by Mount Wilson’s “Dr. Hubbell.” Yet astronomers who scoured the journals wondered alike: When was Edwin Hubble going to publish his findings?

  In a congratulatory note, Russell encouraged Hubble to present his results at the end-of-the-year meeting of the American Astronomical Society in Washington, DC. He added that the paper would likely garner the annual $1,000 research prize of the Association for the Advancement of Science. Yet as the meeting neared its conclusion, there was no correspondence from Hubble, who remained at Mount Wilson. “Well, he is an ass,” Russell muttered to an associate. “With a perfectly good thousand dollars available he refuses to take it.” On December 31, 1924, as Russell prepared a last-ditch missive to his recalcitrant colleague, he spied a large envelope bearing his name behind the hotel’s front desk. The much-anticipated paper had arrived. The following day, Russell stood before the assembled scientists—including Harlow Shapley—and read aloud Hubble’s momentous report, “Cepheids in Spiral Nebulae.” (The paper shared the AAAS research prize with a study of protozoans in the digestive tracts of termites.)

  Hubble’s hesitancy to publish, he soon confessed, stemmed from a reluctance to confront his senior Mount Wilson colleague Adriaan van Maanen, whose observations of spiral-nebula rotation were irreconcilable with the Cepheid-based distances. The rotation data had garnered weighty support among astronomers, as it derived from straightforward plate measurements and bore a lesser burden of assumptions than Cepheid-derived results. Hubble knew that publication of his work would be a scientific slap of the gauntlet: an open charge that van Maanen’s measured rotations were spurious. True, Lowell Observatory’s Vesto Slipher and others had observed the spectroscopic signature of rotation in a few inclined spirals: a telltale tilt of spectral lines along each nebula’s long axis. And spirals’ swept-back arms might reflect a whirling action. But given the enormous distances Hubble had assigned the spiral nebulae, such roundabout displacements are all but immeasurable on photographic plates, even over a span of centuries. Before going public with his incendiary results, Hubble had wanted to nail down the source of what he already believed were illusory signs of rotation.

  Adriaan van Maanen.

  Later in 1925, Hubble bolstered the island-universe model with his seven hundred thousand light-year distance to the Magellanic-Cloud analog NGC 6822. The following year, in a paper knowingly titled “A Spiral Nebula as a Stellar System,” he situated Messier 33 approximately 850,000 light-years from Earth. Hubble applied the full resolving power of the one-hundred-inch telescope to the system’s profuse, star-like condensations, which some astronomers classed as stellar-nebular hybrids, perhaps gassy protostars. The vivid exposures showed beyond doubt that these were ordinary stars, their photographic images as small and round as any previously recorded. Hubble’s summary report on the Andromeda Nebula appeared in 1929, with data from forty Cepheids sustaining its status as a far-flung star system.

  On the other hand, the inferred diameters of both Andromeda and Messier 33 were a mere one-tenth that of Shapley’s Milky Way. In consequence, Shapley offered a multitiered arrangement, with a swollen, central Milky Way accompanied by subordinate systems like globular clusters, the Magellanic Clouds, and spiral nebulae. The counterpunch to this “meta-galaxy” concept would land in incremental blows, as Hubble identified ever-more-remote spiral nebulae.

  Hubble had earlier urged Harlow Shapley to abandon his opposition to the island-universe model, suggesting that “the straws are all pointing in one direction and it will do no harm to begin considering the various possibilities involved.” Now through his latest results, he appeared to be informing both Shapley and van Maanen that the battle of the universes was over. In fact, the controversy spilled into the next decade, with increasingly strained explanations of the systematic movements that nobody, it seemed, but van Maanen could detect. The issue was settled only after Hubble published his own thumbs-down reassessment of spiral rotation, using some of van Maanen’s own plates. Left unsaid, but clearly implied: van Maanen had seen what his own preconceptions had led him to see. In a companion paper, a chastened van Maanen conceded that his prior results should be viewed “with reserve.”

  In his autobiography, Harlow Shapley attributes his lingering intransigence to his friendship with van Maanen and his misplaced confidence in the Big Galaxy model. Had he not left Mount
Wilson for a telescope-disadvantaged Harvard in 1921, he might have expanded his variable-star search from globular clusters to spiral nebulae—and he might have basked in the acclaim accorded Hubble for the discovery of Andromeda’s Cepheids. Shapley long harbored regret over his lost opportunity, declaring in 1969, “The work that Hubble did on galaxies was very largely using my methods. . . . He never acknowledged my priority, but there are people like that.”

  From the 1930s onward, astronomers fashioned a Procrustean bed of research-derived constraints that shrank Shapley’s Big Galaxy and stretched Hubble’s cosmic distances. The dimming of starlight by dust in the galaxy’s disk was found to be more pronounced than had been assumed, requiring various degrees of correction to stellar distances. Of equal import, the Cepheid species was found to be bipartite, based on the circumstances of their origin. Older Cepheids, which inhabit globular clusters and the bulging nucleus of spiral galaxies, formed from a different brew of chemical elements than their younger counterparts in the galactic disk. (This spatial segregation also holds for the general stellar population in spiral galaxies; evidently the old stars in the halo and nucleus reflect the initial sphericity of a protogalaxy before its gas condenses into a flattened disk.) In other words, the Cepheids observed by Hubble in the disks of extragalactic systems are a significant variant of those observed by Shapley in globular clusters: For a given period of variation, Hubble’s Cepheids are fully four times more luminous than Shapley’s. Thus, there are two period-luminosity laws, each of which was substantially refined and recalibrated over the years.

  The twin revelations about galactic light extinction and the Cepheid dichotomy became the agents of change that brought conflicting evidence into concordance. The diameter of Shapley’s Milky Way saw a threefold decrease to a more moderate one hundred thousand light-years. Meanwhile, Hubble’s distances to extragalactic star systems doubled, with a commensurate increase in their size. The Andromeda Galaxy (formerly, Nebula) is more than two million light-years away, its diameter now in full measure worthy of the galactic appellation.

  Hubble continued to range over the extragalactic spacescape during the mid-1920s, developing a morphological classification system for galaxies that is still in use today. (In his own writings, Hubble eschewed the term “galaxy,” preferring the more traditional “extra-galactic nebula.”) Honors were bestowed upon him, audiences crowded his public lectures, newspapers featured his eminently heroic accomplishments. In 1928, Hubble and his wife Grace toured Europe for five months, visiting tourist sites and scientific institutions. Hubble arrived home, energized by exchanges with colleagues at an astronomical conference in Holland.

  One discussion harked back to the memorable meeting Hubble had attended in 1914 when Vesto Slipher debuted his measures of the radial velocities of spiral nebulae—motions that far outpace even the fastest stars in the Milky Way. As startling as their breakneck speeds was the curious fact that thirteen of the fifteen spirals in Slipher’s sample display redshifted spectral lines: the radial component of their motion is directed away from Earth, as though in collective flight from our planet. The skewed velocity spread was sustained in Slipher’s follow-up report of 1917, with twenty-one of twenty-five spirals in recession. The data also suggested a possible correlation between a spiral’s faintness and its redshift, as though more distant spirals are systemically receding at greater velocity. The inference could not be proven at the time; there was no reliable way to gauge extragalactic distances prior to Hubble’s discovery of Cepheids in spiral nebulae.

  In this hint of a coordinated flight of galaxies, Hubble sensed opportunity—a timely convergence of his own interests, experience, and resources. By now, Slipher had painstakingly accumulated the radial velocities of more than forty spirals, having extended the reach of Lowell Observatory’s spectrograph as far into space as it could go. But the twenty-four-inch refractor perceived only the collective glow of these systems, and was too small to resolve their constituent stars. To detect Cepheids and other all-important standard candles required a larger telescope. With Mount Wilson’s one-hundred-inch reflector, Hubble had the means to complete the motional characterization of extragalactic systems: to wed each of Slipher’s radial velocities to its corresponding distance. In late 1928, following Vesto Slipher’s leading thrust, Major Hubble marshaled his resources and launched an all-out assault on the realm of the galaxies.

  Chapter 28

  OCULIS SUBJECTA FIDELIBUS

  It is . . . an enormous advantage to be in the right place at the right time and to know that one is in that enviable position.

  —Malcolm Longair, “History of Astronomical Discoveries,” 2009

  HUMANITY INHABITS A SPINNING GLOBE, coursing round a central star, variously tugged by a nearby moon and neighboring planets, all moving through space among a multitude of stars that themselves bob and weave as they circle a disk-shaped galaxy. The notion of cosmic place is a complex construct, amenable only to sophisticated measurement and subject to an agreed-upon definition of a reference frame. Immobile as human senses make this earthly station appear, only by the determination of positions and movements of celestial objects do we infer our planet’s relative location, course, and speed.

  For centuries, the goal was to situate Earth within the solar system and ascertain its orbital parameters, along with those of its companion planets. During the 1800s, astronomers extended their reach into space, assembling over many decades a crude, three-dimensional map of the solar neighborhood. Photographers tracked the movement of stars in the plane of the sky, while spectroscopists used the Doppler shift to obtain the corresponding radial velocities. Astronomers were thus able to animate their once-fixed stellar map.

  With estimable sagacity, Harlow Shapley used globular star clusters to lay out the form and dimensions of our galaxy, and situate the solar system within it. Astronomers discovered a telltale streaming of stars in our vicinity, confirming that objects in the disk of the Milky Way revolve according to the same Keplerian principles as do the Sun’s family of planets. They believed that galaxies might form an external reference frame against which to refine the solar system’s movement through space: if galaxies are free-floating motes in the void, their helter-skelter motions would average to zero, establishing a three-dimensional coordinate grid as practical as an array of fixed points. In essence, the Earth-centric viewpoint was abandoned for an imaginary aerie that overlooks the realm of the galaxies. Astronomers learned to their surprise that, while each galaxy exhibits a degree of random motion, galaxies collectively move in coordinated fashion: their radial velocity increases with distance. Thus, while assessing Earth’s location and heading from an extragalactic perspective, observers stumbled upon a cosmic mystery with far-reaching implications.

  Empowered by theoretical developments in physics and mathematics during the early twentieth century, scientists came to realize that time and space are inseparable: a cosmic fourth dimension exists, accessible only in a notional sense, yet real. The galaxies in this vision became luminous surveyor’s stones that traced out the gravitational contours of deep space and led, ultimately, to a purely mathematical representation of the extragalactic landscape. With newfound conceptual tools, astronomers could probe beyond the limits of their instruments—indeed, beyond the limits of any conceivable instrument—and ponder the universe in its entirety. In more than the physical sense, gravity pulled it all together.

  Isaac Newton’s mathematical theory of gravitation, derived in the late seventeenth century, proved extremely successful in accounting for the motions of comets, moons, planets, and stars. In Newton’s conception, celestial bodies inhabit a three-dimensional universe governed by the prosaic rules of Euclidean geometry. A unique and uniform flow of time reigns over the entirety of space. That gravity’s origin and mechanism of transmission remained a mystery was of no consequence: Newton’s celestial mechanics generated results that cohered with the observed movements of bodies in the cosmos. However, while the model
worked well for planetary and stellar systems (for example, the orbits of binary stars), it prompted a troubling conundrum when applied on the more expansive cosmological scale. Newton found that an infinite space uniformly sown with matter is unstable against the self-gravity of its material contents. It will invariably draw itself together. Our universe—if not infinite, then sufficiently large that Newton’s analysis applies—cannot remain static; absent any countervailing force, it must be in a state of contraction or expansion. Thus, in Newtonian physics, there was no explanation as to how the observed and presumably infinite universe remained stable.

  Albert Einstein’s General Theory of Relativity, whose initial development culminated in 1917, provides a pathway to a solution of the cosmological quandaries posed by Newtonian mechanics. Einstein abandoned the Euclidean constraints under which Newton had labored and instead applied more modern, non-Euclidean geometries to the notions of space and time—a radical inquiry that would have profound scientific and philosophical ramifications.

  In Einstein’s world, parallel lines might intersect or diverge, and the angles in a triangle might sum up to more or less than the standard 180 degrees. Time forms the basis of a fourth dimension that stands on an equal footing with the three familiar dimensions of space; hence, the introduction of the catch-all term space-time. To Einstein, gravity is not a force emanating from a massive body, but the result of a warp in the invisible continuum of space–time surrounding the body. A moving object that strays into such a region follows a predetermined path dictated by the local contour of space–time. The impression of a gravitational pull is an artifact of this distorted geometry, which is the true agent that compels the object to deviate from its intended path. Denser bodies deform their space–time environment to a greater degree, imparting the sensation of a stronger gravitational force. The field equations of Einstein’s General Theory of Relativity provide the mathematical means to compute the shape of space–time around a given configuration of matter. Conversely, the observed path of, say, a comet or a satellite reveals the local distribution of matter that governs its movement.