5.1.2. Individual stellar X-ray sources
Katz (1976) suggested that the X-ray emission from clusters was produced by a large number of individual cluster stellar X-ray sources, located either in galactic haloes or in intracluster space. He suggested that these sources might be similar to the binary X-ray sources found in our galaxy, and assumed that the X-ray luminosity to virial mass ratio would be the same for our galaxy and for clusters of galaxies. This predicts too little X-ray emission from clusters (see also Felten et al., 1966); however, globular clusters within our galaxy have a much higher X-ray luminosity to mass ratio, and so they might be used as a model for cluster X-ray sources (Fabian et al., 1976). In fact, M87 and some other central dominant galaxies are known to possess very extensive systems of globular star clusters.
Katz argued that the observed X-ray luminosities of clusters were consistent (in 1976) with a fixed X-ray luminosity to virial mass ratio; the current data would appear to rule out such a correlation (Section 4.6). It is difficult to see how this model could produce the variation in X-ray emission properties of clusters, or the optical - X-ray and radio - X-ray correlations. Of course, if the properties of the individual stellar X-ray sources are not constrained in any manner, any X-ray observations could be explained but the model has no predictive power. The granularity in the cluster X-ray emission produced by these individual sources should have been detected in M87/Virgo, but has not been seen (Schreier et al., 1982). Finally, we note that luminous stellar X-ray sources are generally very optically thick. As a result, they do not show any X-ray line emission except for the fluorescence line of low ionization iron (which is probably produced by reprocessing of X-ray radiation in other parts of the binary star system). In particular, they do not show lines from highly ionized iron (Fe+24 and Fe+25) or from lighter elements (for which the fluorescent yields are low). Clusters, on the other hand, do show strong lines from Fe+24 and Fe+25, as well as lines from lighter elements (Sections 4.3.2 and 4.3.3).
In short, there is no real evidence in favor of the individual stellar source model, and considerable evidence against it.