Figure 1: A plot of the speed of stars in the Milky Way vs their distance from the galactic center. The top line represents the actual speed distribution of stars, while the bottom line represents what the motion should be based upon the distribution of normal, visible matter. (Source: University of Virginia)
In 1687 Isaac Newton published a three-volume work entitled Philosophiae Naturalis Principia Mathematica, (The Mathematical Principles of Natural Philosophy). In it Newton presented his laws of motion and his universal law of gravity. He demonstrated that these laws governed not only the motion of things on Earth, but that it also predicted the motion of the planets. Newton's laws still form the foundation of modern physics, and are used for everything from bridge construction to landing rovers on Mars.
When applied to the motion of the planets in our solar system, Newton's laws worked incredibly well. By the mid-1800s it had been used to predict the existence of Neptune based on Neptune's gravitational effect on Uranus. When astronomers applied the same laws to the motion of stars in galaxies, however, they found it didn't work well at all. Either something very strange was going on, or Newton's law of gravity was terribly wrong.
Most of the mass in our solar system is contained in our Sun. This means that planets close to the Sun—like Mercury and Venus—must move quickly in their orbits to overcome the Sun's strong gravity. For distant planets—like Uranus and Neptune—the Sun's gravity is much weaker, and so move much more slowly in their orbits. This is exactly as Newton predicted. Galaxies have a similar structure, with most of their visible matter located in near the galactic center. One would expect then that stars closer to the center would move much faster than stars near the edge of the galaxy.
Figure 2: A false color image of dark matter. Astronomers measured how the light is deflected by gravity, and calculated the mass necessary for this deflection. The result (in blue) is superimposed on the original image. (Source: Hubble Site)
What astronomers actually found was that the outer stars of galaxies move almost as fast as stars near the galactic center. The difference between the expected motion and the actual motion is huge (as seen in Figure 1) and it has been seen in every galaxy we have measured. It means that either our understanding of gravity is wrong, or about 70% of our galaxy's mass consists of an unseen dark matter.
It is tempting to presume that our understanding of gravity is wrong. After all, Einstein demonstrated that Newton's gravity wasn't quite exact, and we have only sent probes to the edge of our solar system. Perhaps gravity is different on galactic scales. Some alternative models have been proposed, such as modified Newtonian dynamics. But when we add Einstein's corrections to gravity, we find even stronger evidence for dark matter.
One prediction Einstein made about gravity is that gravity can change the direction of light, causing a beam of light to bend as it approaches a large mass. Eddington first observed this effect with starlight near our Sun (as I'll talk about next time). Since then astronomers have observed the effect with galaxies. Since dark matter makes up the majority of a galaxy's mass, one would expect the galaxy to bend the light of more distant background galaxies, thus distorting our view of them. This is exactly what we observe. It is even possible to map the distribution of dark matter around a galaxy by measuring exactly how much light is gravitationally deflected, as seen in Figure 2.
This evidence was further strengthened recently when astronomers observed the motion of stars in the Sagittarius dwarf galaxy. This galaxy collided with our Milky Way galaxy long ago, and its stars are now spread around ours in a diffuse stream. Astronomers measured the speed of these stars and again detected the effect of dark matter. Since these stars are spread all around our galaxy, astronomers could measure the distribution of the dark matter in our galaxy. The found that our galactic dark matter forms an asymmetrical squashed sphere. This clear lack of symmetry means it cannot be accounted for by modifying our gravitational theory. Dark matter is real, and it makes up the majority of mass in our galaxy.
Figure 3: An artist's view of the Sagittarius dwarf galaxy, around the Milky Way. This dwarf galaxy collided with ours long ago, spreading itself around our Milky Way galaxy. (Source: BBC News)
Still, there are many unanswered questions. Although we now know that dark matter exists, we have no idea what it is made of. We know that it cannot be regular matter as we know it. If it were simply dark gas and dust, it would cloud our view of distant stars. Dark matter is completely transparent to light. There have been several ideas put forward as to what dark matter could be, but so far these are simply ideas. There are, however, several experiments attempting to detect dark matter directly. Perhaps in the next few years we will begin to understand the dark materials that make up most of our galaxy.
Law, D., Majewski, S., & Johnston, K. (2009). EVIDENCE FOR A TRIAXIAL MILKY WAY DARK MATTER HALO FROM THE SAGITTARIUS STELLAR TIDAL STREAM The Astrophysical Journal, 703 (1) DOI: 10.1088/0004-637X/703/1/L67