The small subunit (SSU) 18S rRNA gene is one of the most frequently used genes in phylogenetic studies (see below) and an important marker for random target PCR in environmental biodiversity screening . In general, rRNA gene sequences are easy to access due to highly conserved flanking regions allowing for the use of universal primers . Their repetitive arrangement within the genome provides excessive amounts of template DNA for PCR, even in smallest organisms. The 18S gene is part of the ribosomal functional core and is exposed to similar selective forces in all living beings . Thus, when the first large-scale phylogenetic studies based on 18S sequences were published - first and foremost Field et al.'s  phylogeny of the animal kingdom - the gene was celebrated as the prime candidate for reconstructing the metazoan tree of life. And in fact, 18S sequences later provided evidence for the splitting of Ecdysozoa and Lophotrochozoa [, see also ], thus contributing to the most recent revolutionary change in our understanding of metazoan relationships. Methodological innovation within the last years came from the incorporation of secondary structure into phylogenetic analyses. In particular RNA specific substitution models considering paired sites in rRNA genes have been shown to outperform standard DNA models [7–12].
During recent years and with increased numbers of taxa included into molecular phylogenies, however, two problems became apparent. First, there are prevailing sequencing impediments in representatives of certain taxa, such as the mollusk classes Solenogastres and Tryblidia [13, 14], selected bivalve taxa (pers. comment H. Dreyer), and the enigmatic crustacean class Remipedia (pers. comment H. Glenner). Failure to obtain 18S sequences of single taxa is considered a common phenomenon but is rarely ever reported. Secondly, in contrast to initially high hopes, 18S cannot resolve nodes at all taxonomic levels and its efficacy varies considerably among clades. This has been discussed as an effect of rapid ancient radiation within short periods . Multigene analyses are currently thought to give more reliable results for tracing deep branching events in Metazoa but 18S still is extensively used in phylogenetic analyses.
Considering the wide range of studies based on the 18S gene as a molecular marker, both sequencing problems and the applicability of 18S for phylogenetic inferences need to be scrutinized. To address these questions, we focus on the Mollusca, the second largest animal phylum. There are eight higher taxa (classes) defined within Mollusca: the aplacophoran Solenogastres (= Neomeniomorpha) and Caudofoveata (= Chaetodermomorpha), two small clades with about 250 and 150 currently described species; Polyplacophora (ca. 920 species); and the conchiferan clades Tryblidia (= Monoplacophora; 29 species described), Scaphopoda (ca. 520 species), Cephalopoda (ca. 1,000 species), Bivalvia (ca. 30,000 species), and Gastropoda (40,000-150,000 species). The monophyly of the phylum is well established based on morphological characters, but 18S phylogenies often show the Mollusca as polyphyletic or paraphyletic with low resolution of the deeper nodes [e.g: [16–19]]. One of the main deficiencies of all published studies on mollusk phylogeny is the underrepresentation of the minor taxa Solenogastres, Caudofoveata, and Monoplacophora. The aplacophoran Solenogastres and Caudofoveata together with Polyplacophora (chitons) are traditionally thought to represent early branching taxa within Mollusca, but their relative position is still controversially discussed [for review see [20, 21]]. Despite of their key position in mollusk phylogeny and evolution, the number of molecular phylogenetic studies including any aplacophoran mollusk species is low: Winnipeninckx et al. ; Okusu et al. ; Lindgren et al. ; Passamaneck et al ; Giribet et al. ; Dunn et al. ; Wägele et al. ; Wilson et al. . To date there are no more than four caudofoveate and five solenogaster (partial) 18S sequences published in Genbank, the taxon sampling spanning no more than two genera of Caudofoveata and Solenogastres, respectively.
In solenogasters, Okusu and Giribet  described severe contamination issues caused by cnidarian prey. Here, we specifically address contamination issues and point out technical problems hampering amplification during PCR. This is important not only for prompting representative taxonomic sampling for phylogenetic analyses, but also for avoiding under- or overestimation of biodiversity in environmental screening programs. Moreover, we evaluate the usefulness of the 18S gene for phylogenetic inferences by combining our new authentic solenogaster 18S data with published molluscan sequences and - based on an extended taxon sampling - analyze sequence divergence and compositional heterogeneity within the phylum.