has demonstrated together with several ground experiments, that the fluctuations in the microwave background radiation are really there at the expected level, and has thus provided a clear view of one very early stage of structure formation in the universe. Together with the galaxy counts, which trace structure on very large scales, there appears to be no simple theory to match both sets of data. The reference model, the socalled CDM (Cold Dark Matter) model
proposes that all matter, visible or not, is non-relativistic at the time of decoupling. In such a model, a small fraction of the density of the universe is baryonic, and most appears to be in a form of non-baryonic matter, of which we see only the gravitational effects, both through giant arcs in clusters of galaxies and in microlensing by compact objects in galaxy halos.
Already we have the well-known discrepancy between the baryonic fraction of the density of the universe inferred from nucleosynthesis, and that much smaller baryonic fraction derived from direct observations of stars, galaxies, and interstellar and intergalactic gas (in clusters of galaxies). The observations of the faint glow around some galaxies
as well as the micro-lensing observations
suggest that this discrepancy is due to faint stars distributed in the form of an isothermal sphere around galaxies.
The rotation curves of galaxies, as well as the lensing observations of clusters of galaxies
clearly demonstrate that clusters have a much larger mass than can be accounted for with baryonic matter
This dark matter leads to cluster masses of the order of magnitude of 1015 M. Mass determinations for clusters of galaxies are done in several ways: a) The virial theorem uses the motions of galaxies; b) the giant lensing arcs are used, which directly give the mass: and c) the X-ray observations are used to integrate the hydrostatic equilibrium equation. This last method needs an assumption about the total pressure in the gas. Allowing for a population of energetic particles and magnetic fields in clusters in analogy to the interstellar medium in galaxies basically triples the total pressure in the gas in the central region of the cluster, and also brings methods b) and c) into consistency. This then leads, for instance, to a lower limit to the density of the universe of 0.4 h50-1/2 of the critical density,
where h50 is the Hubble constant in units of 50 km s-1 Mpc-1. Allowing for hot dark matter may well push the total count of baryonic and dark matter, hot and cold,
to unity in terms of the critical density, so desired by cosmological purists.
The Hubble Space Telescope, in its very long observation of what is called the "Hubble Deep Field"
has observed a large number of young galaxies and what appear to be galaxy fragments with ISO.
Radio observations of this field with the VLA (the Very Large Array, in New Mexico, USA) as well as infrared observations with ISO have shown that many of these objects are very young, that they emit a large fraction of all their power in the infrared, and that galaxy formation appears to be a prolonged affair. Metal formation appears to peak between redshifts of 1 and 2, in agreement with CDM models.
Finally, HIPPARCOS and the recalibration of stellar distances, and so stellar ages have now demonstrated that the age of globular clusters is much lower than we used to think, only 11.5 ± 1.3 Gyr.
This allows quite possibly a convergence of the constraints for the age of the universe and the determination of the Hubble constant to numbers compatible with a negligible cosmological constant Λ. This suggests the density to be unity in terms of the critical density Ω = 1, the Hubble constant H0 to be near 65 km s-1 Mpc-1, an age of the universe near 11 Gyr and is compatible with Λ = 0. Then the baryonic fraction is in the few percent range, and little information about the dark matter is available. Dark matter around galaxies and clusters seems to follow isothermal spheres of non-interacting particles (therefore non-dissipative).
Low surface brightness galaxies are actually dominated throughout their structure by dark matter halos, while high surface brightness galaxies have an inner region where the baryonic matter dominates gravity. All early Hubble type galaxies also appear to have a central black hole of a mass proportional to the spheroidal mass with a factor of order 1/500.
should therefore be common in all galaxies with a spheroidal component throughout some of their evolution.
The reader will be well advised to check the web for the latest publications based on observations with HST, ISO, HIPPARCOS, other satellites or ground instruments. Also, many reviews are available on the web. The Space Telescope Science Institute has made the Hubble Deep Field data available
spawning large scale follow-up work which can be found under "Science Resources" at http://www.stsci.edu/. One conference at the Space Telescope Science Institute has been dedicated already to the Hubble Deep Field.