The Hubble Space Telescope has shown that in the sky, in an angular dimension equal to that under which we see the moon, there are millions of galaxies. With many measurements of this type it is possible to estimate that there are about 100 billion galaxies in the Universe and that each one of these is made up on average of about 100 billion stars! However, this enormous amount of matter is in fact not much when we compare it with dark matter, whether it be "ordinary" dark matter or "exotic".

The amount of ordinary or baryonic (i.e. made up of protons and neutrons) matter in the Universe, whether visible or dark, can be estimated on the basis of the relative quantity of deuterium and helium present today and that which formed three minutes after the Big Bang. If there was a lot of baryonic matter at that time, collisions first between nucleons and afterwards between nuclei would have been very probable in the first moments of the Universe, and the percentage of deuterium should now be very small because deuterium nuclei give rise to helium; if instead there was little baryonic matter, the quantity of deuterium should be relatively greater.
From the most recent measurements of current quantities of deuterium and helium, it can be deduced that the baryonic matter present in the Universe is only about one seventh of that needed to keep stars in their galaxies and galaxies in their clusters. In addition, by using indirect methods, astrophysicists have estimated that there is about nine times more baryonic matter which does not emit visible light than that which emits visible light.
What is it made of? It is enormous clouds of gas in the large cluters of galaxies, it is " black holes " arising from the collapse of massive stars and also massive black holes at the centre of the galaxies, it is "dead " stars ( white dwarfs, neutron stars, ...), it is objects the size of planets, known generally as MACHO (MAssive Compact Halo Objects), ecc..
From what has been said so far, it is clear that the idea of the existence of non-baryonic dark matter needs to be taken seriously.

It is thought that the Universe contains a large number of neutrinos ,  (particles which are well-known in the field of elementary particle physics), produced in the first moments of the life of the Universe. They are rather like "ghost particles" perché because they do not have an electric charge and rarely interact with ordinary matter.
Precision measurements made with particle accelerators have shown the existence of three different types of neutrinos.

	Fig. 2: Artist's impression of the production of atmospheric neutrinos by cosmic rays . (Credit: INFN-Notizie N.1 giugno 1999)

The Kamiokande e Superkamiokande experiments in Japan, the Macro experiment at the Gran Sasso, and the Soudan 2 experiment in the United States have presented results regarding the so-called "atmospheric neutrinos". These results strongly suggest that a neutrino of one type can transform into a neutrino of another type (neutrino oscillations); this can only happen if the neutrinos are not massless. The same conclusion is reached also for neutrinos coming from the sun (the Homestake experiment in the USA, GALLEX at the Gran Sasso, SNO in Canada, ...). The estimated mass for neutrinos is very small, however; if we multiply it by the huge number of neutrinos present in the Universe, we obtain a contribution to the total mass of the Universe which is slightly less than that from visible matter. So the contribution of neutrinos with mass does not change the situation. And their effect is reduced even further by the fact that, since they have a speed very similar to the speed of light, they can play a role only in keeping together really enormous clusters of galaxies, not normal clusters, and definitely not stars in a galaxy. Although it has not been possible to show this through experiments, it is believed that their average energy is high compared in their mass, so their speed would be about the speed of light. It is said that neutrinos form part of the hot dark matter.

We also need to consider the possibility that, in the halos of the galaxies, there are particles of relatively large mass which travel with a typical galactic speed of about one thousandth of the speed of light (this is called cold dark
matter).
For some time now, physicists have been trying to produce such particles in high energy accelerators but up to date they have not observed any, which could imply that they are incredibly massive; this research will continue at future accelerators which will have much greater energy.
In addition, there are astroparticle physicists looking for such particles in the cosmic radiation, using sophisticated detectors where these particles should interact. Given the rarity of these possible interactions, every type of background noise, such as cosmic rays and environmental radioactivity, must be reduced to the minimum. Therefore the detectors must be located in underground laboratories.

But what could these mysterious, slow-moving massive particles be? Theoretical physicists have come up with various hypotheses.
The strangest is perhaps the neutralino which would be the supersymmetric electrically neutral particle with the lowest mass. The idea of the existence of super-particles (s-particles) is based on a possible boson- fermion, symmetry, which says that, corresponding to each particle with a half-integer spin (like the electron, quarks, i neutrinos, ...) there must be an s-particle having a whole-integer spin (s-electron, s-quark, s-neutrino …). In the same way, corresponding to each ordinary particle with a whole-integer spin (eg. the photon) there must be an s-particle with a half-integer spin (eg. the  photino). The most credit-worthy candidate to explain cold dark matter is the neutralino, an s-particle which can be considered as a mixture of s-particles which are the supersymmetric partners of the photon, the boson  boson Z ⁰ and two  Higgsbosons.

Another possibility could come from nuclearites, combinations of u, d and s quarks. Ordinary atomic nuclei are formed of protons and neutrons, which are in turn formed of u and d quarks. We would not see the equivalent of protons and neutrons, in nuclearites, but the u, d and s quarks would be free to move around inside each nuclearite. The nuclearites would have a higher density than ordinary nuclei and would be stable, even for masses much greater than those of the uranium nuclei.

Another possibility could come from magnetic monopoles, hypothetical particles with a magnetic charge. These can be accelerated in the galactic magnetic field and thus would have a velocity spectrum.

There are other possible candidates: for example, axions, hypothetical particles witch could remove a problem in the Standard Model of Elementary Particles. Axions would have a very small mass, there would be a vast amount of them, and they would have a speed of about a thousandth of the speed of light; scientists look for them with detectors which act as microwave receivers.