Community assembly determines which and how many species live in a particular area. Evidence suggests that community assembly often leads to a single stable equilibrium, where environmental conditions and interspecific interactions determine which species exist. In such cases, communities with similar environmental conditions should have similar compositions. However, other evidence suggests that community assembly can lead to multiple stable equilibria, where the resulting community depends on the assembly history, even when all species have access. In these cases, communities with similar environmental conditions can have very different compositions. Both regional and local factors should determine community assembly patterns and the similarity or dissimilarity among localities with similar environments. A single equilibrium is more likely in systems with small regional species pools, high connectance, low productivity, and high disturbance. Multiple stable equilibria are more likely in systems with large regional species pools, low connectance, high productivity, and low disturbance. The author illustrates preliminary evidence from an observational study of small pond communities, showing important effects on community similarity, as well as on local and regional species richness. The debate over community assembly dates back to the discussion between Clements and Gleason on plant succession. Clements viewed community structure as deterministic, while Gleason emphasized the role of species invasion history. Recent reviews have been equivocal on the evidence for multiple stable equilibria. The author summarizes current evidence and develops predictions on when communities should exhibit multiple stable equilibria, influenced by regional and local environmental factors. The author evaluates these predictions using observational data from freshwater pond communities.Community assembly determines which and how many species live in a particular area. Evidence suggests that community assembly often leads to a single stable equilibrium, where environmental conditions and interspecific interactions determine which species exist. In such cases, communities with similar environmental conditions should have similar compositions. However, other evidence suggests that community assembly can lead to multiple stable equilibria, where the resulting community depends on the assembly history, even when all species have access. In these cases, communities with similar environmental conditions can have very different compositions. Both regional and local factors should determine community assembly patterns and the similarity or dissimilarity among localities with similar environments. A single equilibrium is more likely in systems with small regional species pools, high connectance, low productivity, and high disturbance. Multiple stable equilibria are more likely in systems with large regional species pools, low connectance, high productivity, and low disturbance. The author illustrates preliminary evidence from an observational study of small pond communities, showing important effects on community similarity, as well as on local and regional species richness. The debate over community assembly dates back to the discussion between Clements and Gleason on plant succession. Clements viewed community structure as deterministic, while Gleason emphasized the role of species invasion history. Recent reviews have been equivocal on the evidence for multiple stable equilibria. The author summarizes current evidence and develops predictions on when communities should exhibit multiple stable equilibria, influenced by regional and local environmental factors. The author evaluates these predictions using observational data from freshwater pond communities.