Dinophilidae (Annelida) is most likely not a progenetic Eunicida: evidence from 18S and 28S rDNA.

Dinophilidae (e.g., Dinophilus O. Schmidt, 1848) is a typical meiofaunal polychaete taxon with a seemingly simple organisation, e.g., no parapodia, only a few segments (Rouse and Pleijel, 2001). Dinophilidae is thought to be an eunicidan taxon and is often presented as the classical example of progenetic evolution within annelids (Westheide, 1987). The retention of ancestral juvenile characters by adult stages of descendants (paedomorphosis) can arise either by a retardation of somatic development (neoteny) or by an acceleration of the sexual maturation (progenesis) (Gould, 1977). Due to convincing similarities to developmental stages of larger eunicidans and their relatively small size, the progenetic origin of Dinophilidae and other dorvilleids (e.g., Parapodrilus) within Eunicida has been repeatedly assumed (Fig. 1) (see Eibye-Jacobsen and Kristensen, 1994; Westheide, 1987). Based on the possession of a ventral pharyngeal organ with speciWc jaw elements, Eunicida is a well-deWned annelid taxon currently comprising “Dorvilleidae,” Eunicidae, Hartmanniellidae, Histriobdellidae, Lumbrineridae, Oenonidae, and Onuphidae (Rouse and Pleijel, 2001). However, some meiofaunal taxa like Parapodrilus and a parasitic genus Biborin lack jaw apparatuses. Eunicidan species comprise both some of the smallest polychaetes (e.g., Neotenotrocha, 250!m) and the largest polychaete up to 6 m in length (e.g., Eunice) (Eibye-Jacobsen and Kristensen, 1994; Rouse and Pleijel, 2001). The most convincing evidence for a close relationship between dinophilids and eunicidans is from Åkesson’s (1977) experiments demonstrating reciprocal infection with coelomic coccidia of the genus Grellia, parasites that are generally thought to be host-speciWc. In contrast, deWnitive morphological synapomorphies are lacking. For example, the ventral pharyngeal organs of Dinophilidae and Eunicida are diVerent and most likely not homologous (Purschke, 1985, 1987). In a recent 18S rDNA study (Struck et al., 2002) Dinophilidae was not the sistergroup to any eunicidan taxon. However, data were not able to clearly refute Dinophilidae as derived eunicidans as judged by nodal support and statistical tests. Furthermore, in two of their analyses Dinophilidae were in close vicinity to Lumbrineridae and the dorvilleid Pettiboneia urciensis, the latter two taxa are usually closely related to each other in molecular analyses (e.g., Struck and Purschke, 2005). Thus, their possible progenetic origin within Eunicida is still controversial. To address the phylogenetic relationship of Dinophilidae relative to Eunicida, the nuclear 28S rDNA was chosen as an additional molecular marker. Based on 18S rDNA results (Bleidorn et al., 2003; Struck and Purschke, 2005) concerning other possible placements for Dinophilidae, partial sequences of 28S rDNA and 18S rDNA of sabellidan, eunicidan, and dinophilid species were determined in this study. Combined analyses of the two genes (approximately 4 kb of data) were performed.