Around 23% of the mass of the universe is believed to exist in this form. Determining the nature of dark matter is also known as the dark matter problem or the missing mass problem, and is one of the most important problems in modern cosmology.
Much of the evidence for dark matter comes from the study of galaxy clusters. Many of these appear to be roughly static and fairly uniform, so by the virial theorem the total kinetic energy should be half the total gravitational binding energy of the galaxies. Experimentally, however, it is found to be much greater, often off by an order of magnitude or so, and assuming that the visible material makes up only a small part of the cluster is the most straightforward way of accounting for this.
With gravitational theory and new computer analyses, astronomers have now been able to work out where the dark matter is. It is just what you would expect if dark matter and galaxies are clustered in exactly the same way. Galaxies themselves also show signs of being composed largely of dark matter - for instance the rotation curves in and indeed the very existence of our galaxy's disc indicates the presence of a large extended halo.
Knowing where the dark matter is, also reveals how much of it exists. About seven times as much as ordinary matter (that is only one quarter of what is necessary to slow down the universe's expansion to a halt).
Since it cannot be detected via optical means, the composition of dark matter remains speculative. Large masses like galaxy-sized black holes can be ruled out on the basis of lensing data. Possibilities involving normal baryonic matter include brown dwarfs or perhaps small, dense chunks of heavy elements; such objects are known as massive compact halo objects, or "MACHOs." The possible amount of baryonic dark matter is, however, restricted by big bang nucleosynthesis. At present, though, the most common view is that dark matter is made of elementary particles other than the usual electrons, protons, and neutrons, such as neutrinos, axions[?], or hypothetical particles known as weakly interacting massive particles (or "WIMPs").
See also strange matter.
An alternative to dark matter is to suppose that gravitational forces become stronger than the Newtonian approximation at great distance. For instance, this can be done by assuming a negative value of the cosmological constant (the value of which is believed to be positive based on recent observations) or by assuming modified Newtonian dynamics. Another approach, proposed by Finzi (1963) and again by Sanders (1984), is to replace the gravitational potential with the expression
where B and ρ are adjustable parameters. However, all such approaches run into difficulties explaining the different behavior of different galaxies and clusters, whereas one can easily describe such differences by assuming different quantities of dark matter.
For a deeper discussion of this subject, see Modified Newtonian dynamics.
Data from galaxy rotation curves indicate that around 90% of the mass of a galaxy cannot be seen. It can only be detected by it's gravitational effect. There are several types of dark matter postulated to exist.
Hot dark matter consists of particles that travel with relativistic velocities. The best candidate for hot dark matter is the neutrino. Neutrino's have negligible mass, do not partake in either the electromagnetic or the strong nuclear force and so are incredibly difficult to detect. This is why they are such good candidates for hot dark matter.
Hot dark matter cannot explain how individual galaxies formed from the big bang. The microwave background radiation as measured by the COBE satellite is very smooth and fast moving particles cannot clump together on this small scale from such as smooth initial clumping. To explain small scale structure in the universe it is necessary to invoke cold dark matter. Hot dark matter therefore is nowadays always discussed as part of a a mixed dark matter theory.