Résumé : Imaging data from the Sloan Digital Sky Survey are used to characterize the population of galaxies in groups and clusters detected with the MaxBCG algorithm. We investigate the dependence of Brightest Cluster Galaxy (BCG) luminosity, and the distributions of satellite galaxy luminosity and satellite color, on cluster properties over the redshift range 0.1 < z < 0.3. The size of the dataset allows us to make measurements in many bins of cluster richness, radius and redshift. We find that, within r_200 of clusters with mass above 3e13 h-1 M_sun, the luminosity function of both red and blue satellites is only weakly dependent on richness. We further find that the shape of the satellite luminosity function does not depend on cluster-centric distance for magnitudes brighter than ^{0.25}M_i - 5log(h) < -19. However, the mix of faint red and blue galaxies changes dramatically. The satellite red fraction is dependent on cluster-centric distance, galaxy luminosity and cluster mass, and also increases by ~5% between redshifts 0.28 and 0.2, independent of richness. We find that BCG luminosity is tightly correlated with cluster richness, scaling as L_{BCG} ~ M_{200}^{0.3}, and has a Gaussian distribution at fixed richness, with sigma_{log L} ~ 0.17 for massive clusters. The ratios of BCG luminosity to total cluster luminosity and characteristic satellite luminosity scale strongly with cluster richness: in richer systems, BCGs contribute a smaller fraction of the total light, but are brighter compared to typical satellites. This study demonstrates the power of cross-correlation techniques for measuring galaxy populations in purely photometric data.