Abstract
Identifying accurately species is critical for our understanding of patterns of diversity and speciation. However, for many organisms with simple and variable morphological traits, the characters traditionally used by taxonomists to identify species might lead to a considerable under appreciation of their diversity. Recent advances in molecular-data based computational methods have considerably improved our ability to identify and test species limits. Here, we use an integrative approach to delineate species in a complex of sea cucumbers. We used a three-step approach to show that “Holothuria impatiens”, a common, shallow-water species, occurring across the Indo-Pacific, the Western Atlantic and the Mediterranean Sea, targeted locally by fisheries, is a complex of at least 13 species. (1) We used the Generalized Mixed Yule Coalescent (GMYC) model to identify putative species without a priori hypotheses. In the process, we also show that the number of putative species estimated with GMYC can be affected considerably by the priors used to build the input tree. (2) We assessed based on coloration patterns and distributional information, the most relevant hypothesis. This approach allowed us to identify unambiguously 9 species. However, some of the lineages consistently assigned to belong to different species using GMYC, are occurring in sympatry and are not differentiated morphologically. (3) We used Bayes factors to compare competing models of species assignment using the multispecies coalescent as implemented in *BEAST. This approach allowed us to validate that the species identified using GMYC were likely reproductively isolated. Estimates of the timing of diversification also showed that these species diverged less than 2 Ma, which is the fastest case of closely related species occurring in sympatry for a marine metazoan. Our study demonstrates how clarifying species limits contribute to refining our understanding of speciation.
