Studies of the X-ray surface brightness profiles of clusters, coupled with theoretical considerations, suggest that the breaking of self-similarity in the hot gas results from an `entropy floor', established by some heating process, which affects the structure of the intracluster gas strongly in lower mass systems. Fitting analytical models for the radial variation in gas density and temperature to X-ray spectral images from the ROSAT PSPC and ASCA GIS, we derive gas entropy profiles for 20 galaxy clusters and groups. Scaling these profiles to coincide in the self-similar case, the lowest mass systems are found to have higher scaled entropy profiles than more massive systems. This appears to be due to a baseline entropy of 70-140 h50^-1/3 keV cm^2, depending on the extent to which shocks have been suppressed in low mass systems. The extra entropy may be present in all systems, but is detectable only in poor clusters, compared to the entropy generated by gravitational collapse. This excess entropy appears to be distributed uniformly with radius outside the central cooling regions. We determine the energy associated with this entropy floor, by studying the net reduction in binding energy of the gas in low mass systems, and find that it corresponds to a preheating temperature of ~0.3 keV. Since the relationship between entropy and energy injection depends upon gas density, we can combine the excesses of 70-140 keV cm^2 and 0.3 keV to derive the typical electron density of the gas into which the energy was injected. The resulting value of 1-3x10^-4 h50^1/2 cm-3, implies that the heating must have happened prior to cluster collapse but after a redshift z~7-10. The energy requirement is well matched to the energy from supernova explosions responsible for the metals which now pollute the intracluster gas.

It is widely believed that structure in the Universe evolves hierarchically, as primordial density fluctuations, amplified by gravity, collapse and merge to form progressively larger systems. The structure and evolution of X-ray clusters, however, seems at odds with this hierarchical scenario for structure formation. Poor clusters and groups, as well as most distant clusters detected to date, are substantially fainter than expected from the tight relations between luminosity, temperature and redshift predicted by these models. Here we show that these discrepancies arise because, near the centre, the entropy of the hot, diffuse intracluster medium (ICM) is higher tha$ achievable through gravitational collapse, indicating substantial non-gravitational heating of the ICM. We estimate this excess entropy for the first time, and argue that it represents a relic of the energetic winds through which forming galaxies polluted the ICM with metals. Energetically, this is onl$ possible if the ICM is heated at modest redshift ($z \ltsim 2$) but prior to cluster collapse, indicating that the formation of galaxies precedes that of clusters and that most clusters have been assembled very recently.

Abell 2218 is one of a handful of clusters in which X-ray and lensing analyses of the cluster mass are in strong disagreement. It is also a system for which X-ray data and radio measurements of the Sunyaev-Zel'dovich decrement have been combined in an attempt to constrain the Hubble constant. However, in the absence of reliable information on the temperature structure of the intracluster gas, most analyses have been carried out under the assumption of isothermality. We combine X-ray data from the ROSAT PSPC and the ASCA GIS instruments, enabling us to fit non-isothermal models, and investigate the impact that this has on the X-ray derived mass and the predicted Sunyaev-Zel'dovich effect. We find that a strongly non-isothermal model for the intracluster gas, which implies a central cusp in the cluster mass distribution, is consistent with the available X-ray data and compatible with the lensing results. At r<1 arcmin, there is strong evidence to suggest that the cluster departs from a simple relaxed model. We analyse the dynamics of the galaxies and find that the central galaxy velocity dispersion is too high to allow a physical solution for the galaxy orbits. The quality of the radio and X-ray data do not at present allow very restrictive constraints to be placed on H_0. It is apparent that earlier analyses have under-estimated the uncertainties involved. However, values greater than 50 km/s/Mpc are preferred when lensing constraints are taken into account.