The accurate reconstruction of palaeobiodiversity patterns is central to a detailed understanding of the macroevolutionary history of a group of organisms. However, there is increasing evidence that diversity patterns observed directly from the fossil record are strongly influenced by fluctuations in the quality of our sampling of the rock record; thus, any patterns we see may reflect sampling biases, rather than genuine biological signals. Previous dinosaur diversity studies have suggested that fluctuations in sauropodomorph palaeobiodiversity reflect genuine biological signals, in comparison to theropods and ornithischians whose diversity seems to be largely controlled by the rock record. Most previous diversity analyses that have attempted to take into account the effects of sampling biases have used only a single method or proxy: here we use a number of techniques in order to elucidate diversity. A global database of all known sauropodomorph body fossil occurrences (2024) was constructed. A taxic diversity curve for all valid sauropodomorph genera was extracted from this database and compared statistically with several sampling proxies (rock outcrop area and dinosaur-bearing formations and collections), each of which captures a different aspect of fossil record sampling. Phylogenetic diversity estimates, residuals and sample-based rarefaction (including the first attempt to capture 'cryptic' diversity in dinosaurs) were implemented to investigate further the effects of sampling. After 'removal' of biases, sauropodomorph diversity appears to be genuinely high in the Norian, Pliensbachian-Toarcian, Bathonian-Callovian and Kimmeridgian-Tithonian (with a small peak in the Aptian), whereas low diversity levels are recorded for the Oxfordian and Berriasian-Barremian, with the Jurassic/Cretaceous boundary seemingly representing a real diversity trough. Observed diversity in the remaining Triassic-Jurassic stages appears to be largely driven by sampling effort. Late Cretaceous diversity is difficult to elucidate and it is possible that this interval remains relatively under-sampled. Despite its distortion by sampling biases, much of sauropodomorph palaeobiodiversity can be interpreted as a reflection of genuine biological signals, and fluctuations in sea level may account for some of these diversity patterns. © 2010 The Authors. Biological Reviews © 2010 Cambridge Philosophical Society.