Everything we can see in the universe, every star, planet, gas cloud and galaxy, is only a small part of what is there. Measurements of how galaxies move, how they bend light, and how the early universe looked all point the same way: there is about five times more matter than we can see. Nobody knows what it is. It is called dark matter.
Zwicky’s missing mass
In 1933 the Swiss astronomer Fritz Zwicky measured how fast galaxies move inside the Coma Cluster. They were moving far too fast to be held together by the gravity of the visible galaxies alone. Something unseen had to be adding mass. He called it dunkle Materie, dark matter. His estimate of how much was overstated, partly because the expansion rate of the universe was poorly known then, but the basic finding has stood. Others, including Lord Kelvin and Henri Poincaré, had raised the idea of dark bodies earlier.
Galaxies that spin too fast
The idea became hard to ignore in the 1970s. Vera Rubin and Kent Ford measured how fast stars and gas orbit in spiral galaxies, starting with Andromeda. Far from the centre, where there is little visible matter, the orbits should slow down, the way the outer planets of the Solar System move more slowly. They did not. Stars at the edges orbit almost as fast as those further in. Each galaxy must sit inside a much larger, invisible halo of mass.
The Bullet Cluster
In 2006 Douglas Clowe and colleagues studied the Bullet Cluster, two clusters of galaxies that collided. The hot gas, which holds most of the ordinary matter, was slowed by the collision and glows in X-rays in the middle. But by mapping how the clusters bend the light of more distant galaxies, the team found that most of the mass had passed straight through, ahead of the gas. Matter that does not collide and does not shine: exactly what dark matter should do.
What could it be?
The leading candidates are particles that barely interact with anything: WIMPs, weakly interacting massive particles, or much lighter particles called axions. Primordial black holes from the early universe are another idea. Huge detectors deep underground, such as LZ in South Dakota and XENONnT in Italy, use tonnes of liquid xenon to catch the rare collision of a dark matter particle. So far none has been seen.
A minority of physicists think the problem lies instead with our theory of gravity, and have proposed modified gravity such as MOND, put forward by Mordehai Milgrom in 1983. It works well for single galaxies but struggles with galaxy clusters. ESA’s Euclid telescope, launched in 2023, is mapping a third of the sky to chart where dark matter lies.