Time-Calibrated Phylogenies of Hummingbirds and Hummingbird-Pollinated Plants Reject a Hypothesis of Diffuse Co-Evolution

Neotropical ecosystems house levels of species diversity that are unmatched by any other region on Earth. One hypothesis to explain this celebrated diversity invokes a model of biotic interactions in which interspecific interactions drive diversification of two (or more) lineages. When the impact of the interaction on diversification is reciprocal, diversification of the lineages should be contemporaneous. Although past studies have provided evidence needed to test alternative models of diversification such as those involving abiotic factors (e.g., Andean uplift, shifting climatological regimes), tests of the biotic model have been stymied by lack of evolutionary time scale for symbiotic partners. In this study, we infer timescales for diversification of hummingbirds and a species-rich plant lineage that is ,50% hummingbird pollinated, Ruellia (Acanthaceae). Results demonstrate that hummingbirds originated about 20 million years before New World Ruellia and that all but one major hummingbird clade was extant before the plant group originated. Thus, the classic model of ‘‘diffuse co-evolution’’ between hummingbirds and this group of plants is rejected by our data. However, together with the observation that the Neotropical clade of Ruellia (,350 species) is far more species rich than its Old World sister group (,75 species), our results are consistent with the hypothesis that plant diversification in the Neotropics has been facilitated in part by a pre-existing diversity of hummingbirds. This hypothesis may find support in other lineages of Neotropical plants that similarly exhibit asymmetrical partitioning of species diversity in the Paleovs. Neotropics.