Is the Butcher-Oemler effect a function of the cluster redshift?

Using PSPC ROSAT data, we measure, the X-ray surface brightness profiles, size, and luminosity of the Butcher-Oemler (BO) sample of clusters of galaxies. The cluster X-ray size, as measured by the Petrosian r(eta=2) radius, does not change with: redshift; and is independent of X-ray luminosity. On the other hand, the X-ray luminosity increases with redshift. Considering that fair samples show no evolution, or negative luminosity evolution, we conclude that the BO sample is not formed from the same class of objects observed at different look-back times. This is in conflict with the usual interpretation of the Butcher-Oemler as an evolutionary (or redshift dependent) effect, based on the assumption that we are comparing the same class of objects at different redshifts. Other trends present in the BO sample reflect selection criteria rather than differences in look-back time, as independently confirmed by the fact that trends lose strength when we enlarge the sample with an X-ray-selected sample of clusters. The variety of optical sizes and shapes of the clusters in the Butcher-Oemler sample and the Malmquist-like bias are the reasons for these selection effects that mimic the trends usually interpreted as changes due to evolution.