Phylogenies contain information about the evolutionary relationships among species. Including this information when quantifying the change in species composition across geographic and environmental gradients (betadiversity) provides us with additional information potentially important in understanding the distribution of  diversity. Particularly, we expect phylogenetic information to reflect historical or evolutionary processes affecting the distribution of diversity. We measured betadiversity using data on the composition of 189 hummingbird communities in Ecuador, a region with considerable environmental and topographic complexity. We evaluated whether gradients in phylogenetic betadiversity matched gradients in compositional betadiversity and highlighted cases where there was a mismatch between the two measures. We analyzed compositional and phylogenetic betadiversity using two related methods. Both methods use a null model to isolate the amount of phylogenetic dissimilarity among communities that could be attributed to phylogeny alone.  
Results/Conclusions

The steepest gradients in compositional and phylogenetic betadiversity occurred along the elevational gradient. Despite the high species turnover found between the lowlands on both side of the Andes, phylogenetic betadiversity was much smaller than expected based on species turnover. This is consistent with the idea that these two regions were connected previously and the rise of the Andes provided a vicariant barrier promoting speciation events within the same groups, without leading to the further radiation or extinction of isolated groups on either side. On the other hand, phylogenetic betadiversity was much higher than expected based solely on species turnover along the elevational gradient. This is consistent with the idea that the Andes provided an environmental region that promoted simultaneously, the origination and further diversification of groups, and the extinction of possibly many lineages that were unable to colonize these regions. Phylogenetic information proved highly useful for understanding historical factors associated to the distribution of diversity across geographic and environmental gradients. Combined with compositional betadiversity it gave us insights into alternative mechanisms that might be responsible for current patterns of biological diversity.