Identifying global zoogeographical regions: lessons from Wallace

摘要: Aim When dividing the world into zoogeographical regions, Alfred Russel Wallace stipulated a set of criteria by which regions should be determined, foremost use generic rather than species distributions. Yet, recent updates Wallace's scheme have not followed his reasoning, probably explaining in part discrepancies found. Using recently developed quantitative method, we evaluated world's following as closely possible. Location Global. Methods We subjected presence–absence data from range maps birds, mammals and amphibians to an innovative clustering algorithm, affinity propagation. We used genera our taxonomic rank, although familial ranks were also assessed, evaluate how divergence influences results. accepted argument that bats migratory birds excluded (although he was contradictory about birds) devised procedure determine optimal number eliminate subjectivity delimiting regions. Results Regions attained using (eight for six amphibians) strongly coincided with proposed Wallace. The nearly identical scheme, whereas obtained two new ‘regions’ largely coincide subregions. As argued Wallace, there are strong reasons consider these being equivalent main regions. Species distributions generated many small related contemporary climate vegetation patterns, at rank very broad. differences between all involve areas identified uncertain regionalization. Main conclusions Despite more 135 years additional knowledge distributions, shuffling concepts, development computers complex analytical techniques, appear no less valid they when them. Recent studies re-evaluating considered such because computer-based analyses, including this one, subject vagaries particular methods used.