The Elephant in the Universe, page 6
Peebles’s experience at Los Alamos piqued Ostriker’s interest. What if they could tweak Peebles’s code a bit and use it to simulate the evolution of a disk galaxy, to study its long-term stability—or lack thereof? Given that rapidly rotating stars could be expected to deform and split up, it seemed impossible that a flat, spinning disk of billions of stars like our Milky Way galaxy could be stable at all. You’d expect the Turkish flatbread to readily distort into a submarine sandwich, just like a pumpkin-shaped star turns into a dog bone if you spin it up fast enough.
Sure enough, the very first two-dimensional numerical simulations of rotating disk galaxies, published by astronomers Richard Miller, Kevin Prendergast, and Bill Quirk in 1970 and by Frank Hohl in 1971, showed just that: the initially circular disk turns into an elongated, bar-like structure, and the galaxy’s stars end up in wildly elliptical orbits—very different from the orderly circular motions observed in the Milky Way.4 With the help of Princeton’s Ed Groth, Peebles and Ostriker developed a program that would run on the university’s computer, while adding a third dimension to the simulations. Their results agreed very well with those of Miller, Prendergast, Quirk, and Hohl. As Ostriker and Peebles wrote in The Astrophysical Journal, “Axisymmetric, flat galaxies are grossly and irreversibly unstable.”5
But their now-famous December 1973 paper went much further. It was one thing to show that orderly rotating disk galaxies are unstable; it was quite something else to explain why we still see them all around us in the universe. What enables our Milky Way to keep up its orderly appearance? What prevents it from flying apart?
Expectantly, Ostriker looks up from the notepad he’s been scribbling on, as if I have to provide the answer. It’s just simple, intuitive physics, he says—everyone could have thought of this. Spinning, low-mass galaxies are unstable; more mass would help. But if that additional mass is also located in the rotating disk, the galaxy would be just as unstable as before—after all, the simulations showed that it’s the disk shape itself that leads to instability. No, the extra mass needs to be distributed in a huge, more or less spherical halo, not taking part in the orderly rotation of the disk.
Intuition comes first; mathematics follows suit. New computer simulations, using the same code but quite a different initial distribution of test particles, confirmed the hunch: if there’s a lot of gravitating mass in a spherical halo (maybe up to two and a half times the mass in the disk), the flat, rotating galaxy remains stable and keeps its regular appearance. As Ostriker and Peebles wrote in their paper, “A massive halo seems the most likely solution for our own Galaxy.” And, of course, for other “cold”—that is, orderly rotating—disk galaxies, too.
The landmark publication, “A Numerical Study of the Stability of Flattened Galaxies: or, can Cold Galaxies Survive?” is mentioned in every anthology of dark matter research. Ostriker and Peebles, you’ll read, were the first to convincingly show that galaxies like our own Milky Way can’t be stable without huge, massive halos of dark matter. (Later research has revealed that large, random stellar motions in the cores of galaxies can also stabilize flat, rotating disks, but most astronomers believe that the original hunch was right anyway.) However, the phrase “dark matter” doesn’t appear a single time in the fourteen-page paper. If scientists have come to view halos as fonts of mysterious dark matter, Ostriker and Peebles were not willing to go that far in 1973. True, it was evident that the mass in the halo couldn’t emit a lot of light—after all, spiral galaxies are not observed to be embedded in luminous spheres. But who knows, large numbers of very faint stars might do the trick.
Artist’s impression of the invisible halo of dark matter (pictured as a diffuse cloud) surrounding a Milky Way–like spiral galaxy.
In fact, astronomers already knew of galactic halos—a term first coined in the 1920s—and also knew that these halos contained stellar inhabitants. For instance, dozens of so-called globular star clusters, each containing up to a few hundred thousand individual stars, swarm around the center of the Milky Way in a roughly spherical distribution, with a strong concentration toward the galaxy’s core. So as far as Ostriker and Peebles were concerned, there wasn’t any obvious reason why the halo couldn’t be home to countless dim dwarf stars, too, increasing the halo’s mass enough to stabilize the Milky Way. As Jan Oort had written back in 1965, “Some 5 percent of the total mass of the galaxy may be estimated to consist of [orange and red dwarf stars]. There is no way for estimating how much more mass there may be in the form of intrinsically still fainter stars. The real mass of the halo remains entirely unknown.”6
So how massive are galaxy halos? In other words, how massive are spiral galaxies? That was the topic of a second and much more concise Astrophysical Journal paper that Ostriker and Peebles wrote a year after their first, in 1974, with Israeli astrophysicist Amos Yahil, a visiting researcher at Princeton at the time.7 “In fact, it’s the more relevant of the two papers,” Ostriker says. However, the black hole meeting, elsewhere in the Pupin Building, is about to start in fifteen minutes or so, and there’s not much time left to discuss it in detail. “Just read the paper,” he urges.
It’s a bold publication, with a bold title—“The Size and Mass of Galaxies, and the Mass of the Universe”—and quite a number of bold statements. The very first line may even have come as a shock to some readers back in 1974. “There are reasons, increasing in number and quality,” the authors wrote, “to believe that the masses of ordinary galaxies may have been underestimated by a factor of 10 or more.” In just four pages, Ostriker, Peebles, and Yahil summed up and reviewed the various indications that skinny-looking spiral galaxies may in fact be obese heavyweights. Much more massive than you would guess on the basis of their looks.
You can’t put a galaxy on a scale, but there are other ways to estimate their mass. Just look at how strongly they tug on their neighbors. Our Milky Way galaxy is surrounded by dwarf galaxies. The dimensions—and relatively sharp edges—of these satellites are governed by the interplay between their own internal gravity and the Milky Way’s mass. Elsewhere, the dynamics of small groups of galaxies and of galaxy pairs, orbiting each other, provide information on galaxy masses. And wherever you look, you see the same thing: evidence for much more mass than you would expect on the basis of the amount of light you’re seeing. Or, in the language of astrophysicists, a very high mass-to-light ratio.
Talking about tugging: our Milky Way and the neighboring Andromeda galaxy provide another neat argument for huge galaxy masses. Despite the overall expansion of the universe, the two spirals are now approaching each other at a relative velocity of 110 kilometers per second, responding to their mutual gravity. Back in 1959, Franz Kahn of the University of Manchester and Leiden astrophysicist (and former Oort student) Lodewijk Woltjer concluded that the high approach velocity could only be explained if the total mass of the two galaxies and everything in between them were on the order of a trillion solar masses—again, a very high mass-to-light ratio.8
At a smaller scale, there were the brand-new radio astronomy results (described in more detail in chapter 8), which seemed to suggest that spiral galaxies also have a high mass-to-light ratio. These somewhat preliminary findings appeared to indicate that the outermost regions of spirals rotate unexpectedly fast, which suggests that the galaxies contain lots of mass. If they did not, they would break apart at such high speed. Yet the visible light output of a galaxy strongly declines at a certain distance from the center. So here, too, the amount of emitted light is out of alignment with the amount of mass that must be present.
A stronger pull, a larger extent, a higher mass. It looked like astronomers had indeed severely underestimated the sheer significance of galaxies—the gravity of the matter, so to say. And where would all this low-luminosity matter hide? Right: in the halo, which Ostriker and Peebles had shown to be necessary anyway, to explain the system’s stability in the first place. In their 1974 paper with Yahil, they still suggested that a spiral galaxy’s halo could mainly consist of faint stars (this second paper also doesn’t mention dark matter at all), but by now, it all started to feel a bit uneasy. A ten-fold increase in mass—could there really be that many faint dwarf stars?
Moreover, there was the second part of their paper’s title: the mass of the universe. If you know the average mass-to-light ratio of galaxies, and if you estimate the number of visible galaxies out to a certain distance, it’s pretty straightforward to calculate the average mass density of the local universe. (Even I could do that.) The answer that Ostriker, Peebles, and Yahil arrived at was 2 × 10–30 grams per cubic centimeter, or about one hydrogen atom per cubic meter if you would smear out all the mass in all the galaxies evenly throughout space. Writing in Nature, three Estonian astronomers, Jaan Einasto, Ants Kaasik, and Enn Saar independently reached a similar conclusion.9
But this number, incredibly small though it is, seemed impossibly large. In the early 1970s, cosmologists and nuclear physicists were starting to understand the synthesis of chemical elements during the big bang, and comparing these results with the observed amount of deuterium (heavy hydrogen) in the universe told them that the current mass density of the universe was much lower. (More on this in chapter 7.) In other words, it looked like there just weren’t enough atoms in the universe to explain the large galaxy masses that the Princeton team and the Estonians arrived at.
Matter, but not as we know it.
Ostriker needs to go. He gives me a copy of Heart of Darkness, the 2013 book he wrote with the British astronomer and science writer Simon Mitton.10 In the elevator Ostriker tells me about a talk he gave at the 1976 meeting of the National Academy of Sciences in Washington, DC, describing his work with Peebles and Yahil. “Much later, someone asked me why I hadn’t mentioned the work of Vera Rubin in that talk,” he says. I nod understandingly—wasn’t she the first to establish that the outer parts of galaxies rotate too fast? “Vera was a great astronomer,” Ostriker continues, “but at that time, she only had very preliminary results. The paper that brought her well-deserved fame wasn’t published until 1980.”
I leave the Columbia campus somewhat confused. Unfortunately, I can’t talk to Vera Rubin anymore; she died in 2016. But her collaborator, Kent Ford, should still be around somewhere—what’s his story? In a Starbucks coffeehouse across Broadway, I check my email and organize my notes. So much happened in the 1970s, almost half a century ago. So many surprising results, all pointing in the same direction: our expanding universe is governed by dark, mysterious stuff that may not even resemble the matter that stars, planets, and people are made of.
Outside, in the January cold, small groups of students, young parents with children, hurried businesspeople, and an endless stream of cars and cabs pass by. We’re all busy living our lives as best as we can, usually unaware of our place in the Milky Way galaxy, let alone of the huge, dark envelope it sits in. Completely unaware of the fact that, without this mysterious substance, we probably wouldn’t be here.
So important, and still we don’t know what it is.
I look up the last lines of the poem “Dark Matter and Dark Energy,” which Alicia Ostriker wrote in the year her husband was a corecipient of the prestigious Gruber Prize in Cosmology. Beautiful as the verses are, they too don’t offer any answers.
the way every human and every atom
rushes through space wrapped in its invisible
halo, this big shadow—that’s dark dark matter
sweetheart, while the galaxies
in the wealth of their ferocious protective bubbles
stare at each other
unable to cease
proudly
receding
5
Flattening the Curve
W. Kent Ford Jr. has a beetle named after him.
Pseudanophthalmus fordi was discovered in two of the many karst caves of rural Virginia by Tom Malabad of the Virginia Division of Natural Heritage. Since both the Russell’s Reserve Cave and Witheros Cave are located on Ford’s property, the new species was named after the retired astronomer.
A plaque featuring the naming citation and a photo of the rare beetle is among the exhibits Ford has prepared prior to my visit. On the coffee table in front of the friendly, stocky, and balding eighty-eight-year-old is a stack of books and papers. Large, mounted prints of black-and-white photographs are displayed against the wall, on the dresser, and on the sofa.1
“Here’s Vera at the plate-measuring machine at DTM,” he says, referring to the Department of Terrestrial Magnetism at the Carnegie Institution of Washington. “This is her at the telescope at Kitt Peak. Here’s a close-up of my image tube. This one is much later: we’re hugging each other upon meeting at a Carnegie colloquium.”
The centerpiece of his visual trip down memory lane is the famous plot of the rotation of the Andromeda galaxy. Together with Vera Rubin, Ford showed that the outer parts of Andromeda rotate much faster than scientists had expected. The discovery is generally hailed as the first convincing evidence for the existence of dark matter. “It was not until Rubin’s work that dark matter was confirmed,” the Carnegie Institution wrote in a press release about her passing away on December 25, 2016.
Vera Rubin at the plate-measuring machine of the Carnegie Institution’s Department of Terrestrial Magnetism in Washington, DC.
The DTM, in Washington, DC, is where Ford has spent his whole career, ever since he applied for a summer job back in 1955. It’s also where he helped develop the Carnegie Image Tube, an electronic device that enabled astronomers to study much fainter objects than they could with old-fashioned photographic plates. That was all decades ago.
Ellen Ford—eighty-one years old by the time of my visit—provides me with driving directions to the couple’s red farmhouse in the middle of nowhere: past the Millboro Mercantile and the Windy Cove Church, up a gravel road, and beyond a big horse barn. She welcomes me at the front porch, dressed in wellies and a windbreaker and wearing a button that says NO to the planned Atlantic Coast Pipeline. In the house, she prepares ham sandwiches with mustard—Kent’s favorite.
No, it’s not lonely out here, Kent Ford says, when we sit down in the living room, surrounded by the photographs from the past. But he misses the DTM lunch club, where science staff members took turns preparing meals, where hamburgers and hot dogs were allowed only once a week, and where every possible topic was up for discussion. It was during one of those lunch meetings, in 1965, that Ford and radio astronomer Bernard Burke introduced Rubin as their new colleague—the first woman on DTM’s science staff, believe it or not.
It wasn’t Rubin’s first encounter with male dominance in science. After earning her bachelor’s degree in astronomy in 1948, she wanted to go to graduate school at Princeton, but the university didn’t accept female astronomy graduate students—a blatant form of gender discrimination that continued until 1975. Instead she went to Cornell and subsequently obtained her PhD in 1954 at Georgetown, where she became an assistant professor in astronomy in 1962. Even then, it was hard for her to get observing time at the big telescopes of Palomar Observatory in Southern California. There simply had never been any female observers up there before.
Upon arriving at DTM—walking distance from her home, which was convenient because the youngest of her four children was five years old in 1965—Rubin had to choose whom she wanted to share an office with: Bernie Burke or Ford. She became enchanted by the delicate parts of Ford’s image tube spectrograph, which were scattered all over his desk. “She chose the spectrograph,” Ford says, smiling. They shared the office for fifteen years.
This image tube spectrograph—now on display in the National Air and Space Museum on the National Mall—was the device that made Rubin and Ford’s breakthrough observations possible. To study the motions of stars or nebulae, astronomers use a spectrograph, which employs either a prism or a grating to spread out light into the colors of the rainbow. Dark lines in the resulting spectrum—the “fingerprints” of various chemical elements—are slightly shifted toward the red or the blue end, depending on whether the object is receding or approaching, with the wavelength shift depending on the object’s velocity. This same Doppler technique had been applied by Vesto Slipher in 1912 to detect the apparent recession velocities of galaxies, caused by the expansion of the universe (see chapter 3).
However, to record the spectrum of a faint nebula on a photographic plate, you need extremely long exposure times: sometimes up to two nights. The Carnegie Image Tube—designed by Ford and eventually manufactured by electronics company RCA—works as an image intensifier, enabling faster recording of less luminous objects. Without going into too much technical detail, a photon of light hitting the so-called cathode end of the device releases an electron. Next, a cascading process inside the vacuum tube produces ever more electrons. Eventually, the electron beam generates a glowing pixel on a phosphor screen, much brighter than the original photon. The same technology can be found in military night-vision devices.
Using the novel apparatus, exposure times of just a couple of hours were enough to record the spectra of faint objects—a huge improvement. Slipher had been the first to obtain spectra—and to derive velocities—for whole galaxies. Now, with Ford’s spectrograph, it would be possible to do the same for individual objects within a galaxy, at least if the galaxy weren’t too far away. This would yield valuable information about the rotational velocities within spiral galaxies, as a function of distance from the galactic center. And that, in turn, would tell you about the galaxy’s mass and the way in which this mass is distributed.
