Identifying hybridizing taxa within the Daphnia longispina species complex : a comparison of genetic methods and phenotypic approaches

Abstract

Daphnia galeata Sars, D. longispina O. F. Müller and D. cucullata Sars (Crustacea: Cladocera) are closely related species which often produce interspecific hybrids in natural populations. Several marker systems are available for taxon determination in this hybridizing complex, but their performance and reliability has not been systematically assessed. We compared results from identifications by three molecular methods. More than 1,200 individuals from 10 localities in the Czech Republic were identified as parental species or hybrids by allozyme electrophoresis and the analysis of the restriction fragment length polymorphism of the internal transcribed spacer (ITS-RFLP); over 440 of them were additionally analyzed and identified by 12 microsatellite loci. Identification by microsatellite markers corresponded well with allozyme analyses. However, consistent discrepancies between ITS-RFLP and other markers were observed in two out of 10 studied localities. Although some marker discrepancies may have been caused by occasional recent introgression, consistent deviations between ITS-RFLP and other markers suggest a long-term maintenance of introgressed alleles. These results warn against its use as a sole identification method in field studies. Additionally, we quantitatively evaluated the discriminatory power of geometric morphometric (elliptic Fourier) analysis of body shapes based on photos of over 1,300 individuals pre-classified by allozyme markers. Furthermore, a randomly selected subset of 240 individuals was independently determined from photos by several experts. Despite a tendency for morphological divergence among parental Daphnia species, some taxa (especially D. galeata, D. longispina, and their hybrids) substantially overlapped in their body shapes. This was reflected in different determination success for particular species and hybrids in discriminant analysis based on shape data as well as from photographs.