Many descriptive models of chromosomal evolution have been formulated over the past two decades. We review the assumptions and predictions of these models and consider some pertinent evidence. One model of adaptive divergence, the canalization model, is difficult to test, and where predictions are explicitly stated, empirical data generally do not support the model. However, few adequate data sets are available to test this hypothesis, and it deserves further study. Several process-oriented models of chromosomal speciation are reviewed with respect to genetic isolating mechanisms, the fitness components of individual rearrangements, and the permissible population structures. From the last we attempt to estimate qualitatively the expected genetic correlates of each model to maximize its testability. Theoretical and empirical inadequacies are discussed and suggestions are made for improvement. Both phyletic and speciation models of chromosomal evolution have often been based on broad comparative studies of large taxonomic groups, but more precise testing will require multidisciplinary populationlevel studies of organisms hypothesized to be undergoing some form of chromosomally-mediated speciation or adaptive radiation.