The D3 domain and its flanking regions of 28S rRNA of four pairs of closely related sexual species (Eupelops hirtus and E. torulosus; Oribatella calcarata and O. quadricornuta; Chamobates voigtsi and Ch. borealis; Liacarus coracinus and L. subterraneus) and four pairs of closely related parthenogenetic species (Nanhermannia nana and Na. coronata; Nothrus silvestris and No. palustris; Tectocepheus sarekensis and T. minor; Camisia spinifer and Ca. segnis) of oribatid mites were sequenced to investigate (1) if the D3 region can be used as a species marker and (2) if there is genetic variation among closely related species pairs and if its magnitude is related to reproductive mode. Furthermore, we investigated the world-wide genetic variation of the D3 region from the oribatid mite species Platynothrus peltifer. There was no intraspecific genetic variation in the D3 region in any of the species studied; it was even identical in two closely related parthenogenetic species (Na. nana and Na. coronata) and two closely related sexual species (E. hirtus and E. torulosus). The genetic differences of the other species pairs indicated that both parthenogenetic and sexual lineages have various ages. On average, however, the differences between the closely related parthenogenetic species were larger than those between closely related sexual species, indicating that parthenogenetic lineages exist historically and may radiate slower than sexual species. The findings of this study support the hypothesis that some of the parthenogenetic oribatid mite taxa (Tectocepheus, Nothrus) are 'ancient asexuals'. The absence of intraspecific or intra-individual variation in the D3 region of parthenogenetic species is consistent with the presence of concerted evolution in the 28S rRNA gene. From this we infer the existence of a meiotic process, which is consistent with the automixy known from several other parthenogenetic oribatid species.