Search

We recover the luminosity distributions over a wide range of absolute magnitude (−24.5 < MR< −16.5) for a sample of seven rich southern galaxy clusters. We find a large variation in the ratio of dwarf to giant galaxies, DGR: 0.8 ≤ DGR ≤ 3.1. This variation is shown to be inconsistentwith a ubiquitous cluster luminosity function. The DGR shows a smaller variation from cluster to cluster in the inner regions (r≲ 0.56 Mpc). Outside these regions we find the DGR to be strongly anticorrelated with the mean local projected galaxy density, with the DGR increasing towards lower densities. In addition, the DGR in the outer regions shows some correlation with Bautz–Morgan type. Radial analysis of the clusters indicates that the dwarf galaxies are less centrally clustered than the giants, and they form a significant halo around clusters. We conclude that measurements of the total cluster luminosity distribution based on the inner core alone are likely to be severe underestimates of the dwarf component, the integrated cluster luminosity and the contribution of galaxy masses to the cluster's total mass. Further work is required to quantify this. The observational evidence that the unrelaxed, lower density outer regions of clusters are dwarf‐rich adds credence to the recent evidence and conjecture that the field is a predominantly dwarf‐rich environment, and that the dwarf galaxies are under‐represented in measures of the local field luminosity function.

We present detailed simulations of long‐exposure CCD images. The simulations are used to explore the validity of the statistical method for reconstructing the luminosity distribution of galaxies within a rich cluster, i.e., by the subtraction of field number‐counts from those of a sight‐line through the cluster. In particular, we use the simulations to establish the reliability of our observational data to be presented in Paper III. Based on our intended CCD field‐of‐view (6.5 × 6.5 arcmin2) and a 1σ detection limit of 26 mag arcsec−2, we conclude that the luminosity distribution can be robustly determined over a wide range of absolute magnitude (−23 < MR< −16) provided: (a) the cluster has an Abell richness 1.5 or greater; (b) the redshift of the cluster lies in the range 0.1 < z< 0.3; (c) the seeing is better than FWHM 1.25 arcsec, and (d) the photometric zero‐points are accurate to within Δm= ± 0.12. If these conditions are not met, then the recovered luminosity distribution is unreliable. Finally, although the method clearly has limitations, within these limitations the technique represents an extremely promising probe of galaxy evolution and environmental dependences.