From Integrative Taxonomy to Species Description: one Step Beyond: Species delimitation and description

International audienceIntegrative taxonomy was formally introduced in 2005 as a comprehensive framework to delimit and describe taxa by integrating information from different types of data and methodologies (Dayrat 2005; Will et al. 2005). Even if debate remains about the hierarchy of the types of characters and criteria to use for species delimitation (Schlick-Steiner et al., 2009; Padial et al., 2010; Yeates et al., 2011), most, if not all taxonomists agree that objectively evaluating several lines of evidence within a formalized framework is the most efficient and theoretically-grounded approach to defining robust species hypotheses (Samadi and Barberousse 2006; de Queiroz 2007).The last ten years have seen a renewal of taxonomy, illustrated by the increasing number of published articles related to species concepts, species delimitation methodology and its application.In the early 90s, many systematists began to suspect that the majority of species would remain undescribed (Costello et al. 2013a; Erwin 1982; Mora et al. 2011 – but see Costello et al. 2013b) and that some of them will probably go extinct before we have a chance to describe them (Barnosky et al., 2011; Leakey and Lewin, 1995; Pimm et al., 2006). The use of molecular data, and in particular molecular barcoding (Hebert et al., 2003), was presented as one answer to this “taxonomic impediment” (as defined in Rodman and Cody, 2003), and welcomed as such by taxonomists. It thus adds to the toolkit of taxonomy, which continues its development as a synergic discipline involving morphological taxonomists, field ecologists, naturalists, and statisticians (Knapp 2008). Integrative taxonomy, used for many decades by taxonomists but only recently formalized concomitantly with the molecular revolution, is organised following a three-step workflow (see also Evenhuis 2007): first, we need to accumulate data on numerous specimens (from various types of data: DNA, morphology, ecology…); second, we need to circumscribe groups of organisms using concepts that ensure that these groups correspond to species (this second step may be coupled with the first, as biological data are continuously accumulated and species hypotheses re-discussed); and third, we need to provide a species description, i.e. a diagnosis and a name for the species recognized as new.Naming new species is a fundamental step when describing biodiversity and is the only way to ensure that scientists are talking about the same entity, and that all the data linked to conspecific specimens but produced by different researchers (or amateurs) can be associated in a comparative analysis (Patterson et al., 2010; Satler et al., 2013; Schlick-Steiner et al., 2007). Not linking biological data (should they be molecular, morphological, or ecological) to a formal species name will result in these data losing tremendous value (Goldstein and DeSalle 2011). Indeed, when authors publish data on entities that are not defined within the framework of a referencing system (e.g. solely identified by an alphanumeric label), they make it very difficult for other authors to build on these data. The best example is the need for taxa to be named to have a chance to be listed in an endangered species list and to benefit from a conservation program: no name, no surviving (Mace 2004). Beyond the need for communication among scientists, names are also key to communicating with non-scientist audiences. While it is now widely recognized that integrating several lines of evidence is the most efficient and theoretically grounded way to delimit new species (e.g. de Queiroz, 2007; Schlick-Steiner et al., 2009; Yeates et al., 2011), the formal naming of new entities may have become decoupled from species delimitation. Indeed, we noted that in several cases new delimited species were not accompanied by formal species description (see also Goldstein and DeSalle 2011). The aim of this article is therefore to test the hypothesis that integrative taxonomy, as defined in 2005 (Dayrat 2005; Will et al. 2005), and in particular the use of molecular data, helped to alleviate the taxonomic impediment by delimiting and describing new species. We reviewed part of the “integrative taxonomy” literature of the last eight years (2006-2013) and tested if authors that delimit new species also name them. We also looked at how the number and type of characters used, across different taxa, varies across articles