E. The radio, X-ray and -ray skies
The 1950's was a great era for cataloguing radio sources, much of the work being done at Cambridge in England (with the 2C, 3C, etc. surveys) and at Parkes in Australia. The surveys were done at considerably different frequencies and gave disparate views of the source counts. This had a strong influence on the Steady State versus Big Bang debate, each survey being used to support a different cosmological hypothesis.
The sources in early surveys were randomly distributed over the sky (for instance, Holden (1966) on the Third Cambridge Catalog and Payne (1967) on the southern counterpart). This remained true for later surveys at low frequencies, which found, for the most part, intrinsically very bright sources at somewhat larger distances (for instance, Webster (1976) analyzing the Fourth Cambridge and Greenbank surveys, and Masson (1979) on the Sixth Cambridge Catalog). Indeed it remains true down to the present day (Trimble and Aschwanden, 2001), for the low-frequency surveys that pick out large, bright, steep-spectrum, extended double sources: Artyukh (2000), Venturi et al. (2000) reported that they did not even identify the Shapley concentration). What this means is that, on average, there is only one of these sources in each of the largest-scale structures to be found in the local universe. The absence of clustering is, therefore, in some sense evidence for the existence of "largest structures," though Artyukh and Venturi et al. note that mergers of small groups into large clusters and superclusters may well turn off fainter radio sources that would otherwise reveal intermediate structure.
In contrast, higher frequency surveys that yield intrinsically fainter radio galaxies find that they are clustered very much like radio-quiet galaxies of the same Hubble types (Cress et al. (1996) on the Faint Images of the Radio Sky at Twenty-cm (FIRST) survey from the Very Large Array (VLA), and Magliocchetti et al. (1998) a further analysis of FIRST, showing that the distribution of those radio sources in space is consistent with their having grown by gravitational instabilities from Gaussian initial conditions). Returning to the Shapley concentration, Venturi et al. (2002) found no fewer than 124 radio sources there.
Distant radio sources (of which quasars are an important sort) are rather sparsely distributed throughout the Universe and are consequently not good indicators of large scale structure. It is therefore not surprising that radio source catalogs provide little evidence for the large scale clustering.
Galaxy clusters are prominent features of the X-ray sky that can provide a good measure of the large scale clustering. X-ray selected samples of clusters are less prone to bias than catalogs for clusters selected from maps of the galaxy distribution. One problem, however, is that the selection criteria for galaxy clusters selected from X-ray surveys (Borgani and Guzzo, 2001) are quite different from the selection criteria for clusters selected from optically scanned photographic plates (Dalton et al., 1997) and it is not so easy to relate studies based on the two sources of data.
The REFLEX (ROSAT-ESO Flux Limited X-ray) cluster survey contains 449 clusters, covering an area of 4.24 steradians in the southern hemisphere ( < 2.5°). It is complete at 90%, down to a nominal flux limit of 3 × 10-12 erg s-1 cm-2 in the 0.1 - 2.4 keV band. REFLEX, as other cluster samples, shows unambiguously very large-scale inhomogeneities that appear when the clustering power is measured and compared with that of galaxies at the same scales (Guzzo, 2002).