Studies on marine invertebrates have shown that the rock record (outcrop area, number of localities etc.) has a strong influence on diversity patterns (e.g. Smith, 2001), but similar corrections have not been applied to terrestrial clades. We compiled diversity curves for dinosaurs as a whole (including Mesozoic birds) and various component clades. Taxic diversity curves were obtained for species and genera: these were corrected using ghost lineages obtained from phylogenetic analyses. Statistical tests (Pearson�s and Spearman�s correlations) revealed strong positive relationships between the species/genus curves and the taxic/phylogenetic curves. This indicates that the same signals were being sampled in each case. A model of expected species-richness was then constructed for each clade using the number of dinosaur bearing formations (DBFs) present in each time interval as a proxy for outcrop area. Comparisons between the models and the diversity curves indicate that apparent patterns in ornithischian and theropod diversity can be explained largely as a result of changes in the number of DBFs. Conversely, sauropodomorph diversity bears little resemblance to that expected from the model, suggesting that at least some features of this curve result from genuine evolutionary signals.