IT鈥橲 embarrassing. We don鈥檛 know the answer to a question that could hardly be more fundamental: what is the universe made of? The question has arisen because astronomers really can鈥檛 explain why stars and galaxies are moving around so fast. The Milky Way, for example, rotates once every 200 million years, and although that may not sound dizzying, it ought be enough to tear the whole thing apart. The only force holding our galaxy together is gravity, and gravity is too weak to rein in that rotation, unless there is a lot more matter out there than we can see, matter that exerts the extra gravity needed to glue the galaxy together.
Similarly, the rapid motion of whole galaxies within clusters makes a lot more sense if they are held together by vast clouds of unseen matter. There are some alternative explanations: that our understanding of gravity is flawed, for example. But most physicists鈥� favourite theory is that there is something out there that we know only as 鈥渄ark matter鈥�. No one has a clue what it actually is 鈥� but it makes up 90 per cent of the universe.
The reason we need such an exotic explanation is that ordinary dark junk 鈥� burnt out stars, for example 鈥� just won鈥檛 do, because we already know how much ordinary matter there is. A few minutes after the big bang, protons and neutrons were packed so close that some of them fused together into heavier nuclei, such as helium. If there had been much more of the stuff, it would have been packed even more tightly and fused into heavier elements still. Stars and interstellar gas would contain much more helium, boron and neon, for example, than astronomers see.
Advertisement
So whatever gives us the missing mass, it needs to be some kind of material that won鈥檛 undergo nuclear fusion. The most popular candidate is called cold dark matter. CDM is a gas of heavy subatomic particles, individually called WIMPs (standing for 鈥渨eakly interacting massive particles鈥�), which affect ordinary matter and each other by the weak force and gravity. WIMPs don鈥檛 feel electromagnetic forces, or the strong nuclear force that binds proton and neutrons together in nuclei, so they are immune to nuclear fusion and effectively 鈥渃ollisionless鈥�, passing straight through each other.
WIMPs are the simplest kind of particle that astrophysicists think can do the job. What鈥檚 more, many tentative new theories of particle physics, developed to try to unify the forces of nature, predict the existence of heavy, weakly interacting particles. So particle physicists agree that WIMP-like particles might indeed exist.
Cosmic sculptors
Most of all, WIMPs are favoured because they make able sculptors of the universe. Computer simulations show their pull does a good job of clumping matter into galaxy-sized lumps, and that these galaxies then congregate into a web of walls, knots and voids that looks a bit like a half-melted honeycomb, and a lot like the large-scale structure of the real universe. Quite a triumph for a hypothetical material.
But while the CDM lobby was congratulating itself on having dreamed up a particle that could solve the dark matter problem, Ben Moore of the University of Durham in the UK was taking a closer look. He noticed that computer simulations showed cold dark matter piling up in the hearts of galaxies. It would mean that if you were to plot the density of dark matter as you travel from one side of the galaxy to the other, you鈥檇 get a curve with a sharp point called a cusp. The curve rises faster and faster as you near the centre of the galaxy until you reach a peak right in the middle 鈥� a dark matter Matterhorn (see Diagram).
Moore reasoned that there ought to be evidence of these cusps in data on galactic hydrogen gas gathered from radio telescopes. Hydrogen molecules emit a radio signal at a characteristic wavelength of 21 centimetres. But if the gas is moving relative to us, the wavelength we observe is shifted by the Doppler effect. Measuring the Doppler shift tells you how fast the gas is moving; faster gas usually means more gravity, and therefore more matter in the vicinity. A sharp concentration of dark matter in each galaxy should therefore show up in the form of increasingly frenetic hydrogen near the centre.
However, radio observations of many nearby galactic cores revealed that the hydrogen is pretty sluggish, implying that there is a plateau in the density of dark matter, rather than a cusp 鈥� more Table Mountain than Matterhorn. The observations seem to contradict the theory of cold dark matter.
Data from new optical telescopes that boast much better spatial resolution adds weight to the case. These telescopes looked mainly at hydrogen and carbon monoxide gas inside small galaxies called dwarfs, where there is relatively little in the way of ordinary luminous matter such as nebulae and stars. Here astronomers can be pretty sure that the mass they trace is mostly dark matter. And spectral lines emitted by the gas showed that it was moving slowly. Where the theory said there should be heaps of cold dark matter, observations said otherwise. 鈥淚 find the evidence very convincing,鈥� says Jerry Sellwood of Rutgers University in New Jersey.
This crisis has seeded a whole forest of hypotheses about the nature of dark matter (see 鈥淭he right stuff?鈥�). There is warm dark matter and strongly interacting dark matter. There is annihilating dark matter, whose particles explode on contact, decaying dark matter, which simply falls apart. And there is even fuzzy dark matter made of particles thousands of light years across.
Each of these crazy flavours of dark matter has what it takes to get around the cusp problem: one way or another they avoid congregating densely in galactic centres. And they all generate the right kind of large-scale structure, too. But they all have drawbacks. Each of the solutions invokes some new property of dark matter particles, governed by a parameter that has to be tuned to fit the observations, such as the mass or lifetime of the particle.
Theorists don鈥檛 like having to introduce extra free parameters, and many are disarmingly sceptical about their own speculations. Some of them are so put off by the untidiness of their theories that they are inclined to question the experimental data. 鈥淭here鈥檚 no observation that says we definitely have to go beyond conventional cold dark matter,鈥� says Wayne Hu of the Massachusetts Institute of Technology. 鈥淚t鈥檚 still possible that it鈥檚 just WIMPs.鈥� Paul Steinhardt of Princeton University agrees: 鈥淚 cannot claim that CDM 鈥� that is, dark matter consisting of cold, collisionless particles 鈥� is ruled out.鈥�
Off target
Carlos Frenk, at the University of Durham, doesn鈥檛 think that there is a crisis at all. He believes that the observations contain systematic errors, such as telescopes aimed just off a galaxy鈥檚 centre, that weaken them fatally. 鈥淎ny mistake would give you the impression that there is no cusp,鈥� he says. Out of 24 dwarf galaxies that Stacey McGaugh of the University of Maryland in College Park has observed, Frenk claims that only three have good enough data to rule out a cusp. 鈥淚f 10 per cent of your sample are anomalous, should you throw away your theory, or look more carefully at those examples?鈥� He feels that cold dark matter is backed by such strong evidence that we should hold onto it.
But McGaugh is unconvinced. 鈥淭his apparent controversy is just a good old-fashioned case of pounding the square peg into a round hole,鈥� he says. 鈥淭he data disagree with theory, so the theorists blame the data.鈥� Sellwood agrees: the dark matter theorists are now clutching at straws, he says. 鈥淭hey鈥檙e getting pretty fanciful at this point 鈥� wheels within wheels within wheels.鈥�
Sellwood and McGaugh are so disillusioned by these efforts that they prefer an entirely different approach, one first suggested about 20 years ago by Mordehai Milgrom, then at the Institute for Advanced Study in Princeton, New Jersey. Milgrom鈥檚 proposal is a modification of Newton鈥檚 300-year-old law of gravity. In the years since Milgrom formulated his revised gravitational law, it has never been proved wrong. Nevertheless, such an arbitrary upheaval of physics remains an unpopular option.
While Milgrom is playing dark matter鈥檚 nemesis, Moore seems just plain dispirited. 鈥淭here have been about 10,000 papers written on this topic at a cost of more than a billion dollars, and we still know almost nothing about the nature of dark matter,鈥� he says.
NASA鈥檚 James Webb Space Telescope, the successor to the Hubble Space Telescope, could help by killing many of the newer theories after its launch in 2010. But could there be a third way 鈥� one that doesn鈥檛 need funny new particles or a radical rethink of gravity? Jeremiah Ostriker of the University of Cambridge thinks an old, discarded suggestion should be picked up and polished. Maybe most of the matter in the universe is in the form of huge black holes, each more than a million times the mass of our sun. Some believe these may have been born in the first split second after the big bang. If so, they would spiral in towards the centres of galaxies, congregating in swarms. In the densest part of the swarm, the black holes would take part in a complicated and violent dance that would eventually fling many of them out of the galaxy altogether 鈥� putting a ceiling on the swarms鈥� density and thus solving the cusp problem. 鈥淣ow that massive black holes have been found in all galactic nuclei, the possibility seems even more attractive,鈥� Ostriker says. And, of course, black holes have no problem meeting the criteria for dark matter. They鈥檝e got plenty of gravity, and they鈥檙e as dark as it gets.