Species-Level Paraphyly and Polyphyly: Frequency, Causes, and Consequences, with Insights from Animal Mitochondrial DNA

▪ Abstract Many uses of gene trees implicitly assume that nominal species are monophyletic in their alleles at the study locus. However, in well-sampled gene trees, certain alleles in one species may appear more closely related to alleles from different species than to other conspecific alleles. Such deviations from species-level monophyly have a variety of causes and may lead to erroneous evolutionary interpretations if undetected. The present paper describes the causes and consequences of these paraphyletic and polyphyletic patterns. It also provides a detailed literature survey of mitochondrial DNA studies on low-level animal phylogeny and phylogeography, results from which reveal the frequency of nonmonophyly and patterns of interpretation and sampling. This survey detected species-level paraphyly or polyphyly in 23% of 2319 assayed species, demonstrating this phenomenon to be statistically supported, taxonomically widespread, and far more common than generally recognized. Our findings call for increased attention to sampling and the interpretation of paraphyletic and polyphyletic gene trees in studies of closely related taxa by systematists and population geneticists alike and thus for a new tradition of “congeneric phylogeography.”