A comparison between optically and X-ray selected quasars.

A method which allows the estimation of the distribution of the ratio of X-ray to optical fluxes for the entire quasar population starting from X-ray-selected samples of objects is presented. It does not require any knowledge of either the luminosity function or the cosmological evolution (if any) of the objects under study. The only external input to the method is the optical number-magnitude relationship. This method is applied to the Einstein Observatory Medium Survey sample of X-ray-selected active galactic nuclei, and it is found that the underlying distribution of the ratio of X-ray to optical fluxes computed from this sample is significantly different from that observed in a comparison sample of optically selected quasars. In particular, the fraction of objects with large ratios of X-ray to optical fluxes which are derived from the X-ray-selected sample is much smaller than what is observed in the optically selected sample. Various effects which could, in principle, contribute to this difference are discussed. While none of them appears to be fully satisfactory, there is, however, some indication that long-term variability may have produced a significant broadening in the observed distribution of the ratio of X-ray to optical fluxes in the optically selected sample. This result suggests that some of the previous estimates of the quasar contribution to the X-ray background, based on the available X-ray data for optically selected quasars, are likely to have overestimated it.