Deciphering the products of evolution at the species level: The need for an integrative taxonomy

Abstract

Progress in molecular techniques together with the incorporation of phylogenetic analyses of DNA into taxonomy have caused an increase in the number of species' discoveries in groups with morphological characters that are difficult to study or in those containing polytypic species. But some emerged criticisms plead for a taxonomic conservatism grounded either on the requirement of providing evidences of morphological distinctiveness or reproductive barriers to erect new species names. In a case study of taxonomic research on Neotropical frogs, we combine several lines of evidence (morphological characters, prezygotic reproductive isolation and phylogenetic analyses of mitochondrial DNA) to test the status of 15 nominal species and to assess the degree of agreement of the different lines of evidence. Our study reveals that morphology alone is not sufficient to uncover all species, as there is no other single line of evidence independently. Full congruence between lines of evidence is restricted to only four out of the 15 species. Five species show congruence of two lines of evidence, whereas the remaining six are supported by only one. The use of divergence in morphological characters seems to be the most conservative approach to delineate species boundaries because it does not allow the identification of some sibling reciprocally monophyletic species differing in their advertisement calls. The separate analysis of differences in advertisement calls (evidence of reproductive isolation) or of phylogenetic data alone also shows limitations, because they do not support some morphological species. Our study shows that only an integrative approach combining all sources of evidence provides the necessary feedback to evaluate the taxonomic status of existing species and to detect putative new ones. Furthermore, the application of integrative taxonomy enables the identification of hypotheses about the existence of species that will probably be rejected or changed, and those that can be expected to persist. © 2009 The Norwegian Academy of Science and Letters.