The striped false limpet Siphonaria pectinata (Linnaeus, 1758) has traditionally been considered a pulmonate gastropod (but see current phylogenetic work for the placement of Siphonariidae; e.g. Grande, Templado & Zardoya, 2008; Jorger et al., 2010) with a limpet-like shell adapted to life on wave-swept rocky intertidal shores. With its 12 synonyms, S. pectinata is known from the Atlantic coast of Portugal, European and African Mediterranean and West African coast to Angola, Macaronesian archipelagos in the eastern Atlantic and from the Caribbean and nearby Atlantic, including eastern Florida, Texas and Mexico (Abbott, 1974; Rolan, 2005). Although its presence in the western Atlantic has been attributed to a Nineteenth Century invasion from the Mediterranean Sea (Morrison, 1963, 1972), this view was questioned by Carlton (1992) based on habitat and broad Western Atlantic distribution, and it is thus considered by some a cryptogenic species (Carlton, 1992; Baker, Baker & Fajans, 2004) – a species that may be either native or introduced. The potential invasive capacity of Mediterranean Siphonaria pectinata has recently been demonstrated by a considerable range extension to the Gulf of Tunis (northeastern Tunisia), where it was established after 2005 (Antit, Gofas & Azzouna, 2007) and Croatia (unpublished sequence data available in GenBank, accessioned October 2010). The dietary specialization of the species on soft microalgae that may serve to reduce competition with other grazers that consume harder encrusting material (Ocana & Fa, 2003) has been interpreted as one of the advantages of this species for colonizing new localities. In order to test the potential invasion of Siphonaria pectinata into (or from) the western Atlantic, we collected samples from near the extreme of the species distribution in localities in the province of Cadiz, southeastern Spain (La Caleta and el Puerto de Santa Maria), southern Cameroon (Kribi) and the Atlantic coast of Florida (Fort Pierce) (Table 1), totalling 23 individuals, collected between 1993 and 2010 (see Fig. 1 for some of the sampled specimens). The oldest specimens were preserved in 70% ethanol and transferred to 96% ethanol a few years after collection; newer specimens were preserved in 96% ethanol and stored at –808C until processed for DNA extraction. Specimens were retained as vouchers and, together with their DNA extractions, are deposited in the Museum of Comparative Zoology, Department of Invertebrate Zoology DNA collection under numbers MCZ DNA100660, DNA104633, DNA104886 and DNA105664. Total DNA was extracted from a small tissue sample from the foot using the DNeasy Tissue Kit from QIAGEN# and the protocol provided by the manufacturer. The purified DNA was used as a template for PCR amplification of fragments of the mitochondrial genes cytochrome c oxidase subunit I (COI), and 16S rRNA. A 658-bp fragment of the COI gene was amplified and sequenced using the primer pair LCO1490–HCO2198 (Folmer et al., 1994), and part of the 16S rRNA gene, between 436 and 437 bp, was amplified and sequenced using primer pair 16Sa–16Sb (Xiong & Kocher, 1991; Edgecombe, Giribet & Wheeler, 2002). PCRs (25 ml) included 1 ml of the template DNA, 1 ml of each primer, 2.5 ml EconoTaq 10X PCR buffer containing 15 mM MgCl2 (Lucigen), 0.25 ml of dNTPs 100 mM, 1.25 U of EconoTaq DNA polymerase (Lucigen). The PCRs were carried out using an Eppendorf Mastercycler epgradient thermal cycler, and involved, for COI, an initial denaturation at 958C for 2 min, followed by 36 cycles of denaturation step at 958C (45 s), annealing at 41–448C (COI) or 48–538C (16S rRNA) (1 min) and elongation at 728C (90 s). The final elongation step at 728C (4 min) and a rapid thermal ramp for 48C were applied to finalize the process. The double-stranded PCR products were visualized by agarose-gel electrophoresis (1.5% agarose), cleaned up with 2 ml of diluted (1:3) ExoSAP-IT (USB Corp., Cleveland, OH, USA) in a volume of 22 ml PCR product and performed at 378C for 30 min followed by enzyme inactivation at 808C for 15 min. Sequencing reactions were performed in a 10-ml reaction volume using 3.2 ml primer (1 mM), a 1 ml of ABI BigDye Terminator v. 3.0 (Applied Biosystems), 0.5 ml BigDye 5 Sequencing Buffer (Applied Biosystems) and 3.3 ml of cleaned PCR product. The sequencing reaction, performed by using the thermal cycler described above, involved an initial denaturation step for 3 min at 958C, 25 cycles (958C for 10 s, 508C for 5 s and 608C for 4 min) and a rapid thermal ramp to 48C. The BigDye-labelled PCR products were cleaned using Performa DTR Plates (Edge Biosystems, Gaithersburg, MD, USA). The chromatograms were visualized, edited and assembled using Sequencher (Gene Codes Corporation#1991–2006). External primers were cropped and discarded from the edited sequences. Subsequently, the sequences were edited using the sequence alignment editor Se-Al v. 2.0a11. The same program was used to translate COI sequences to ensure that stop codons were not present. The sequences generated in this study have been deposited in GenBank under accession numbers HQ386632–HQ386679. Using Siphonaria concinna G.B. Sowerby I, 1824 (GenBank accession numbers EF489378.1 and EF489300.1 for COI and 16S rRNA, respectively), S. serrata (Fischer, 1807) (GenBank accession numbers EF489380.1 and EF489302.1) and the western Atlantic species S. alternata Say, 1826 (new data for voucher MCZ DNA105833; GenBank accession numbers HQ386678 and HQ386679) as outgroups we performed a maximum-likelihood (ML) phylogenetic tree search. Phylogenetic analyses were conducted in RAxML 7.0.4 (Stamatakis, 2006). Nodal support was estimated via bootstrapping (1,000 replicates) (Stamatakis, Hoover & Rougemont, 2008). Because COI and 16S rRNA can be considered part of the same locus, all analyses were conducted for both genes independently and as a concatenated dataset. The