Population genetic markers are increasingly being used to study the diversity, ecology and evolution of Symbiodinium, a group of eukaryotic microbes that are often mutualistic with reef-building corals. Population genetic markers can resolve individual clones, or strains, from samples of host tissue; however, samples may comprise different species that may confound interpretations of gene flow and genetic structure. Here, we propose a method for resolving species from population genetic data using tests for genetic recombination. Assigning individuals to genetically recombining populations prior to further analyses avoids critical errors in the interpretation of gene flow and dispersal. To demonstrate the effectiveness of the approach, we first apply this method to a simulated data set. We then use the method to resolve two species of host generalist Symbiodinium that commonly co-occur in reef-building corals collected from Indo-West Pacific reefs. We demonstrate that the method is robust even when some hosts contain genotypes from two distinct species. Finally, we examine population genetic data sets from two recently published papers in Molecular Ecology. We show that each strongly supports a two species interpretation, which significantly changes the original conclusions presented in these studies. When combined with available phylogenetic and ecological evidence, the use of population genetic data offers a robust method for unambiguously delimiting morphologically cryptic species.Keywords: dispersal, microsatellites, population genetics, species boundaries, Symbiodinium Received 10 July 2015; revision received 9 October 2015; accepted 10 November 2015
IntroductionPopulation genetic data are increasingly generated in ecological studies of the dinoflagellate genus, Symbiodinium, a diverse group of dinoflagellates often associated with reefbuilding corals and other Cnidaria (Andras et al. 2011;Pettay & LaJeunesse 2013;Baums et al. 2014;Thornhill et al. 2014). The analysis of these data generally requires alleles to be assigned to individuals and assumes that all samples analysed together are members of the same species. This allows the data to be organized into multilocus genotypes (MLGs). These MLGs can then be analysed under the assumptions of population genetic models. A number of factors must be considered, however, before meeting these basic criteria for analysis.The genus Symbiodinium is widely recognized as ecologically and genetically diverse; however, there are relatively few described species (Trench & Blank 1987;Hansen and Daugbjerg 2009, LaJeunesse et al. 2012;LaJeunesse et al. 2014), making initial species identification of samples problematic. Ribosomal DNA (rDNA) has often been used to provisionally resolve 'species' diversity (LaJeunesse 2001) because Symbiodinium with fixed differences in rDNA exhibit distinct ecological, biogeographical and physiological attributes (Rodriguez-Lanetty et al. 2004;Sampayo et al. 2007;Frade et al. 2008;LaJeunesse et al. 2010). However, sometimes rDNA can fail to...