Patterns of endemism in African birds: how much does taxonomy matter?

Abstract

It has been suggested that the ‘right' biogeographic patterns will only arise with a species concept reflecting the finest splitting of terminal phylogenetic branches. The significance of taxonomic resolution was assessed using distributional data for all resident African land-birds, held at the Zoological Museum of Copenhagen. The primary database (in a one-degree geographical grid) served as a template for creating two new databases: one, where the distributional records were allocated to species according to the ‘Speciation Atlases' for African birds (Hall and Moreau 1970, Snow 1978) and two, according to the finest taxonomic splitting that has since then been suggested in the literature. With this spatial resolution, the species richness pattern is almost exactly the same whether old or new taxonomy is used. The endemism follows another pattern with marked local aggregates. The location of principal aggregates are quite robust to recent splitting, although a slightly more complex structure appears. Also some small new areas of endemism emerge, each with 2–3 narrow endemics. However, large portions of African savannah and lowland rainforest are still characterised by a total lack of narrow endemism. Based on the current understanding of diversification processes and adaptive re-distribution it is suggested that the uneven distribution of narrowly endemic and widespread species will persist even if it were possible, in the future, to define terminal taxa in a consistent way.


(Ostrich: 2003 74(1&2): 30–38)