Biogeography is one of the most synthetic of biological undertakings; it requires placing a substantiated phylogenetic model in a geological, climatological, and ecological context, all of which shift through time. In the past two decades, the application of cladistic and molecular techniques has diversified our grasp on phylogeny. This has allowed the formulation of hypotheses of past distribution patterns based on samples from available living material and algorithms for their interpretation. Fossils contribute to these cladistic approaches by adding morphological checkpoints to character associations in time and by providing a basis for estimates of rates of divergence. However, fossils also check these hypotheses by direct occurrence (does a fossil of the taxon occur where predicted?) and by ecological suitability (is it reasonable that the taxon could occur in the predicted environment?). The first test is straightforward if difficult, requiring the finding of a specific fossil in a specific place and time. The second is based on the assumption of physiological uniformitarianism—that fossil and modern taxa united by a common morphology possess similar physiologies. If correct, then hypotheses of past distribution must accord with the predicted physiological tolerances of the taxon in question. Application of physiological uniformitarianism to phylogeographic hypotheses, together with new paleontological data, suggest (a) the validity of many established phylogeographic hypotheses; (b) the need to reevaluate others; and (c) the recognition that the North Atlantic land bridge likely functioned as a link between the Old and New Worlds into the Later Tertiary, contrary to this author's earlier papers.