We present an analysis of the 2-point correlation function, of the X-ray Brightest Abell-type Cluster sample (XBACs; Ebeling et al. 1998) and of the cosmological constraints that it provides. If \xi(r) is modelled as a power-law, we find r_0=26.0 +/- 4.5 Mpc/h and \gamma=2.0 +/-0.4, with errors corresponding to 2\sigma uncertainties. Only a marginal increase of the correlation amplitude is found as the flux limit is increased from 5x10^{-12} cgs to 12x10^{-12} cgs, thus indicating a weak dependence of the correlation amplitude on the cluster X-ray luminosity. Furthermore, we present a method to predict correlation functions for flux-limited X-ray cluster samples from cosmological models. The method is based on the analytical recipe by Mo & White (1996) and on an empirical approach to convert cluster fluxes into masses. We use a maximum-likelihood method to place constraints on the model parameter space. We find that the shape parameter of the power spectrum lies in the 2\sigma range 0.05<\Gamma<0.20. As for the amplitude of the power-spectrum, we find \sigma_8 in the range 0.4-0.8 for \Omega_0=1 and 0.8-2.0 for \Omega_0=0.3. This result is in agreement with, although less constraining than, results based on the local cluster abundance.

We estimate the dipole of the whole sky X-ray flux-limited sample of Abell/ACO clusters (XBACs) and compare it to the optical Abell/ACO cluster dipole. The X-ray cluster dipole is well aligned ($\le 25^{\circ}$) with the CMB dipole, while it follows closely the radial profile of its optical cluster counterpart although its amplitude is $\sim 10 - 30$ per cent lower. In view of the fact that the the XBACs sample is not affected by the volume incompleteness and the projection effects that are known to exist at some level in the optical parent Abell/ACO cluster catalogue, our present results confirm the previous optical cluster dipole analysis that there are significant contributions to the Local Group motion from large distances ($\sim 160h^{-1} $Mpc). In order to assess the expected contribution to the X-ray cluster dipole from a purely X-ray selected sample we compare the dipoles of the XBACs and the Brightest Cluster Sample (Ebeling et al. 1997a) in their overlap region. The resulting dipoles are in mutual good aggreement with an indication that the XBACs sample slightly underestimates the full X-ray dipole (by $\le 5$ per cent) while the Virgo cluster contributes about 10 - 15 per cent to the overall X-ray cluster dipole. Using linear perturbation theory to relate the X-ray cluster dipole to the Local group peculiar velocity we estimate the density parameter to be $\beta_{c_{x}} \simeq 0.24 \pm 0.05$.