© 2012 Wiley Publishing Asia Pty Ltd, ISZS and IOZ/CAS286

123456789

101112131415161718192021222324252627282930313233343536373839404142434445464748495051

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051

Integrative Zoology 2012; 7: 286–298 doi: 10.1111/j.1749-4877.2012.00305.x

ORIGINAL ARTICLE

Population structure and range expansion: the case of the invasive gastropod Cyclope neritea in northwest Iberian Peninsula

Lucía COUCEIRO, Lúa LÓPEZ, José Miguel RUIZ and Rodolfo BARREIRODepartment of Ecology, College of Science, University of A Coruña, A Coruña, Spain

Abstract Biotic invasions have a reputation for unpredictable behavior. Here, we report how slight changes in human ac-tivity responsible for the introduction and range expansion of a non-native mollusk have led to detectable differ-ences in the genetics of the invasion. Cyclope neritea is a non-predatory gastropod introduced to 2 areas of the European Atlantic: the northwest Iberian Peninsula (NWIP) and the French Atlantic coast (FAC). Shellfish sea-bed farming is intense in both areas but focuses on different commercial species. Using mitochondrial gene se-quences, the lower genetic diversity recorded along the NWIP suggests a more homogeneous range of source populations than in the FAC. Unlike FAC, genetic diversity and haplotype composition in the NWIP correlate with the date of first occurrence of C. neritea at each site rather than with geographical location. Although this pattern evokes the genetic signature expected under a serial-founder colonization model from a single initial en-clave, a comparison with samples from potential source populations suggests that the NWIP probably experi-enced several independent reintroductions. The jump dispersal pattern of C. neritea in the NWIP, together with the observation that populations established in the same year are genetically undifferentiated, point to human transport as the most plausible explanation for the current range expansion. Despite evidence for human-mediat-ed dispersal, C. neritea managed to develop a seemingly non-random genetic pattern in the NWIP. It is suggest-ed that caution must be exerted when interpreting genetic patterns in invaders.

Key words: alien species, Cyclope neritea, cytochrome oxidase I, genetic geographic pattern, population structure

Correspondence: Lúa López, Área de Ecología, Facultad de Ciencias, Campus A Zapateira s/n, 15008 A Coruña, Spain.Email: lua.lperez@udc.es

INTRODUCTION Biotic invasions and their ecology have received

growing academic attention since Elton (1958) pub-lished his seminal work. Several topics addressed by El-

ton are still the focus of current research, although new prominent themes have emerged in modern times (Rich-ardson & Pysek 2008). The fact that some non-native species exert important negative effects on native organ-isms and human economies alike explains much of this interest (Grosholz et al. 2000; Perrings et al. 2000; Pi-mentel 2002) (Gurevitch & Padilla [2004] offer a more controversial assessment of the ecological consequenc-es of invasions.) However, the interest in invasions does not exclusively derive from the challenge of solv-

287

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051

Expansion of Cyclope neritea: genetic pattern

© 2012 Wiley Publishing Asia Pty Ltd, ISZS and IOZ/CAS

ing applied issues. Species invasions are large-scale, un-planned ecological experiments that offer the potential to gain insights into ecology, evolution and biogeogra-phy (Sax et al. 2005). Like other research fields, con-siderable effort has been devoted to derive general prin-ciples for the ecology of invasions (Williamson 1996; Lockwood et al. 2007). In fact, this active discipline has already produced a remarkable number of books and manuals (see Simberloff 2004; Lockwood 2007; Davis 2009). However, there is still work to be done to attain a unified framework (Blackburn et al. 2011).

Despite these efforts, invasions seem reluctant to re-veal their general principles and to facilitate the building of appropriate predictive power (Kolar & Lodge 2001; National Research Council 2002; Honnay & Jacquemyn 2008). For example, it has been suggested that the spe-cies-level characteristics that might be predictive of suc-cessful invaders are likely to be taxa-specific or even site-specific (Hayes & Barry 2008); the fact that one of the best predictors of invasion success and invasion im-pact is the performance of the same species when in-troduced elsewhere (National Research Council 2002; Hayes & Barry 2008) probably exemplifies our poor predictive ability. Likewise, other sides of the ecolo-gy of invasions (e.g. boom and bust dynamics, lag times and the determinants of invasion spread) still require clarification (Simberloff & Gibbons 2004; Hastings et al. 2005; Sax & Gaines 2008). Drawing generalizations is possibly hindered by the existence of many potential mechanisms behind the observed patterns (Lockwood et al. 2007), as well as the rather idiosyncratic nature of some invasions.

Here, we provide a further example of the idiosyn-cratic nature of invasions. Instead of comparing success-ful versus failed introductions (Hayes & Barry 2008), we focus on the differences in the invasion genetics of 1 invader in 2 areas outside its native range. The inva-sions of these 2 areas share many similarities: the 2 oc-cur in the same marine ecoregion (South European At-lantic Shelf within the Lusitania province) (Spalding et al. 2007), the 2 are linked to the same habitat (soft-bot-tom locations on or close to shellfish beds within shel-tered embayments and estuaries) and they derive from the same vector (accidental introduction with shell-fish translocations). Despite these many similarities, we show how slight differences in the human activity more closely linked to this invasion (i.e. shellfish aquaculture) seem to have led to detectable differences in the struc-ture and genetic diversity of the non-native populations found on each area.

Cyclope neritea (Linnaeus, 1758) is a small (approx-imately 10–15 mm) burrowing nassariid gastropod na-tive to the Black Sea and the Mediterranean Sea and to the Atlantic coasts of Morocco and the southern Iberi-an Peninsula as far north as Setúbal (Portugal) (Hidalgo 1917; Nobre 1932). Historical surveys recorded its pres-ence in a few enclaves of the Cantabrian coast (north of the Iberian Peninsula) as early as 1900 (for a detailed analysis see Sauriau 1991). Whether these Cantabrian populations are native or introduced remains unclear be-cause oyster culture is historically common in the area and the presence of C. neritea has been discontinuous in time (‘unstable’ range [see Simon-Bouhet et al. 2006]). Therefore, these populations might be regarded as a case of cryptogenic introduction (see Carlton 1996). In the mid-1970s/early 1980s, C. neritea began to spread grad-ually northwards along the French Atlantic coast (FAC) up to the entrance of the English Channel (for a summa-ry of the process see Bachelet et al. 2004). Human-me-diated transport with cultivated bivalves (mostly oys-ters) from various distant sources, perhaps favored by a slight warming of the area during the period 1970 to 1990, was identified as the vector for this introduction into NWIP waters and as the most likely mechanism for its rapid range expansion along the FAC (Sauriau 1991; Bachelet et al. 2004; Simon-Bouhet et al. 2006). C. ne-ritea has also been reported as a non-native introduction into NWIP rias, where a moderately abundant popula-tion was first detected in the early 1990s on the south-ern bank of the Ría de Arousa (Rolán 1992). In NWIP, C. neritea remained largely confined until the early 2000s, when dense populations were detected at enclaves scat-tered along the south shore of the Ría de Arousa and the adjacent Ría de Pontevedra (Quintela et al. 2006). Mito-chondrial DNA (mtDNA) data identified clam transfers from the Veneto region (Adriatic Sea) as the most like-ly vector for this introduction and pointed to new rein-troductions from abroad, rather than a range expansion from an earlier introduction, as the likely cause for the recent proliferation within the region (Couceiro et al. 2008). In this regard, one particularly relevant feature of C. neritea as an invader is its presumed low vagility. This snail is a direct developer; juveniles hatch as craw-laways from single-embryo capsules that females have firmly attached to hard substrates (Gomoïu 1964). The-oretically, the lack of a planktonic stage could impose some restrictions to an unaided range expansion in non-native regions.

Cyclope neritea is typical of shallow waters on clean sand. In the Atlantic, it is restricted to sheltered habi-

288

123456789

101112131415161718192021222324252627282930313233343536373839404142434445464748495051

L. Couceiro et al.

© 2012 Wiley Publishing Asia Pty Ltd, ISZS and IOZ/CAS

tats within coastal lagoons, embayments and estuaries (Nordsieck 1968). Being a deposit feeder and opportu-nistic scavenger (Morton 1960; Southward et al. 1997), it is not anticipated that C. neritea would impact heavi-ly on the receiving fauna. Still, laboratory experiments and field observations suggest that it might outcom-pete some resident nassariids in terms of carrion detec-tion and parasitic load (Bachelet et al. 2004). Likewise, biologists working for shellfisher organizations in the NWIP have noticed that C. neritea rapidly builds very high densities in areas recently sown with clam seed and there are untested claims that those high densities might increase seed mortality (L. Conde, pers. comm.).

Here, we extend our previous mtDNA investigation of NWIP non-native populations of C. neritea to include the new enclaves discovered in 2008 that expand con-siderably its range within the region. We use the infor-mation added by these new sites to: (i) provide a more comprehensive picture of the origins and propagation of

the snail within the region; (ii) obtain a comparative da-tabase by sampling several locations along the native range; and (iii) assess the stability of haplotype frequen-cies in non-native populations by re-sampling sites in-cluded in our previous study (Couceiro et al. 2008).

MATERIALS AND METHODS Adult C. neritea were collected at 8 sites for this

study. Five samples, collected between April and No-vember 2008, included the native range (Venice, in the Veneto region), the ‘unstable’ range (Mundaka and San Vicente, in the Cantabrian coast; thereon referred to as cryptogenic following Carlton 1996) and the new re-cords for the species within the NWIP (San Simon and Ponte do Porco) (Fig. 1). These new records expand considerably the non-native range of C. neritea with-in the NWIP. The other 3 samples (Vilagarcía, A Toxa and Poio) were collected in March 2009 at sites already

Figure 1 Location and haplotype frequency for the 8 study populations of Cyclope neritea. Native distribution range denoted by dashed lines, while solid lines indicate the non-native range. The dotted line designates the cryptogenic distribution range. Data for populations A Toxa, Vilgarcía and Poio from Couceiro et al. (2008). Shading in pie charts refers to clades defined in Figure 2. For northwest Iberian Peninsula populations, the date of first documented record is indicated in parentheses. Sources: aRolán 1992; bQuintela et al. 2006; and cthis study.

289

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051

Expansion of Cyclope neritea: genetic pattern

© 2012 Wiley Publishing Asia Pty Ltd, ISZS and IOZ/CAS

included in our previous study (Couceiro et al. 2008). These 3 samples were used to assess the stability of the genetic composition of non-native populations through time. Sampling sites were always between tidal marks, on or close to shellfish beds, and substratum consisted of fine sand. Specimens were collected by hand at low tide with the aid of a bait to facilitate their detection. Animals were transported alive to the laboratory where the shell was cracked with the help of a bench vice, soft tissues were washed with water to remove traces of sed-iment and a portion of foot tissue was stored in 96% ethanol before DNA extraction.

Total DNA was extracted using Chelex-100 resin (Bio-Rad Laboratories, Hercules, USA) following Es-toup et al. (1996). We amplified a 533 bp fragment of the mitochondrial cytochrome c oxidase I (cox1) gene using the primer set and protocols detailed in Bachelet et al. (2004). Purified polymerase chain reaction prod-ucts (Shrimp Alkaline Phosphatase and Exonuclease I) were sequenced on a 3130 XL Genetic Analizer (Ap-plied Biosystems, Foster City, USA). Sequencing reac-tions were performed using the BigDye Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems, Foster City, USA), following the manufacturer’s recommendations. Sequences were aligned using CodonCode Aligner v3.0.1 (CodonCode Corporation, Dedham, USA) and compared with those available in GenBank using BLAST (Altschul et al. 1990); only traces with high quality values and no

ambiguities were used. Haplotypes were named follow-ing Simon-Bouhet (2006) and Couceiro et al. (2008). The protocol used for the research conforms to the pro-visions of the Declaration of Helsinki.

Data analysis combined the haplotype frequencies recorded in this study with those obtained in 2006 by Couceiro et al. (2008). The genetic diversity of each population was estimated using conventional statistics implemented in the program DNASP v.5 (Librado & Rozas 2009): haplotype diversity (Hd), nucleotide diver-sity (Π), number of haplotypes (H) and number of seg-regating sites (S). For the 2 estimates of diversity, the significance of the differences among populations was tested using the GT2 method for multiple unplanned comparisons (Hochberg 1974). The existence of a rela-tionship between genetic diversity (haplotype and nucle-otide diversity) and date of first recorded occurrence in NWIP sites was tested using Spearman’s rank correla-tions. To examine the phylogenetic relationship between haplotypes, the haplotype network of Simon-Bouhet et al. (2006), later extended by Simon-Bouhet (2006), was rebuilt incorporating the new haplotypes found in the NWIP. The haplotype network was built with the soft-ware NETWORK v4.5.0.2. (http://fluxus-engineering.com/) using the median-joining algorithm; the complex-ity of the resulting network was reduced using the max-imum parsimony option (Bandelt et al. 1999). Final-ly, the magnitude of the genetic differentiation between

Table 1 Sample size (N) and genetic diversity estimates for the 8 study populations of Cyclope neritea

Location Year N S H Hd (SD) ∏ × 102 (SD)Northwest Iberian Peninsula

A Toxa 1992 30 18 6 0.731 (0.046) 1.102 (0.128)Vilagarcía 2005 30 12 3 0.517 (0.088) 1.003 (0.137)Poio 2005 30 15 5 0.483 (0.096) 0.947 (0.162)San Simón 2008 33 20 3 0.119 (0.076) 0.302 (0.193)Ponte do Porco 2008 30 12 3 0.191 (0.093) 0.397 (0.188)

Cryptogenic rangeMundaka 27 12 2 0.513 (0.034) 1.155 (0.076)San Vicente 29 12 2 0.502 (0.040) 1.131 (0.089)

Native rangeVenecia 28 17 5 0.630 (0.082) 1.131 (0.124)

S, number of segregating sites; H, number of haplotypes; Hd (standard deviation), haplotype diversity; Π (standard deviation), nucleotide diversity. In northwest Iberian Peninsula sites, Year refers to the first documented record of the species in the area of the location.

290

123456789

101112131415161718192021222324252627282930313233343536373839404142434445464748495051

L. Couceiro et al.

© 2012 Wiley Publishing Asia Pty Ltd, ISZS and IOZ/CAS

populations was assessed by an analysis of molecular variance (AMOVA) (Excoffier et al. 1992) and by esti-mating pairwise ΦPT statistics, an analog of FST. Both es-timates were based on the haplotype frequencies as im-plemented in Arlequin v3.11 (Excoffier et al. 2005) and their significance was tested using a permutational ap-proach (10 100 permutations). Pairwise ΦPT statistics were also used to investigate the stability of haplotype frequencies with time by comparing samples collected in 2006 with those collected in May 2009 on a per site basis. Finally, the genetic arrangement of the 8 sampled populations was depicted by constructing a non-metric multidimensional scaling (NMDS) ordination plot based on ΦPT values as estimates of genetic distance using the software Primer v6.1.9 (Clarke & Gorley 2006).

RESULTS The 237 sequences obtained for the 8 study popula-

tions yielded a total of 11 haplotypes (Fig. 1). All es-timates of diversity revealed considerable differences between populations (Table 1). The number of haplo-types per site ranged from 2 in locations sampled with-in the cryptogenic range (Mundaka and San Vicente) to 6 in the non-native site of A Toxa. Altogether, the 5

non-native enclaves sampled in the NWIP produced 8 haplotypes. The 3 most abundant haplotypes in Venice were also the most common in the NWIP. The 2 popu-lations sampled within the cryptogenic range (Munda-ka and San Vicente) displayed the same 2 haplotypes. Likewise, the number of polymorphic bases (S) ranged from 12 (sites of the cryptogenic range and non-native locations of Ponte do Porco and Vilagarcía) to 20 (San Simon). As for haplotype diversity, the 2 sites more re-cently established in the NWIP (San Simon: 0.119 ± 0.076; Ponte do Porco: 0.191 ± 0.093) showed signif-icantly lower estimates (GT2 test, α = 0.01) than any other study population, while A Toxa (0.731 ± 0.046) and Venice (0.630 ± 0.082) yielded the significantly highest estimates; the 2 sites from the cryptogenic range formed a statistically homogeneous group of intermedi-ate haplotype diversity with Vilagarcía and Poio (range: 0.483–0.517). Nucleotide diversity revealed a similar pattern. Again, the significantly lowest values (GT2 test, α = 0.01) were obtained at the 2 more recently estab-lished colonies (San Simon: 0.302 ± 0.193; Ponte do Porco: 0.397 ± 0.188), while the significantly highest ones in-cluded A Toxa (1.102 ± 0.128) and Venice (1.131 ± 0.124). However, the estimates of nucleotide diversity at the cryptogenic range (San Vicente: 1.131 ± 0.089; Mun-

Figure 2 Median-joining network showing the phylogenetic relationships among Cyclope neritea mtDNA hap-lotypes. Shaded labels identify haplo-types found in this study; other haplo-types obtained from GenBank (sources: G1-G2 Couceiro et al. 2008; H30-H39 Simon-Bouhet 2006; H1-H29 Simon-Bouhet et al. 2006). The size of each circle is proportional to the frequency of each haplotype in our study. Shad-ing distinguishes the main clades iden-tified by Simon-Bouhet (2006). Small white dots indicate missing intermedi-ates. Each branch between 2 haplotypes (sampled or missing) represents 1 muta-tional step.

291

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051

Expansion of Cyclope neritea: genetic pattern

© 2012 Wiley Publishing Asia Pty Ltd, ISZS and IOZ/CAS

daka: 1.155 ± 0.076) were statistically indistinguish-able from the group of highest values. The 2 non-native sites where C. neritea was first recorded in 2005 yield-ed slightly, but significantly lower values (0.947 ± 0.162 and 1.003 ± 0.137, respectively). There was a significant negative correlation between genetic diversity and date of first recorded occurrence of C. neritea at NWIP sites (Spearman’s r = −0.95, P = 0.014, for both estimates of genetic diversity).

Of the 11 haplotypes, 5 dominated our dataset (37.9% H22, 13.5% H1, 12.6% H16, 8.0% H17 and 11.0% H18). The haplotype network revealed that these 5 hap-lotypes belong to 3 different lineages (Fig. 2): H22, H1 and a lineage containing H16, H17 and H18. Moreover,

2 of 6 minor haplotypes also belonged to these same lin-eages: G3 (GenBank accession no. GQ265781), a new haplotype detected in Venice, was clearly linked to the lineage of H22, while G1 was separated by 1 single mu-tational step from the more common haplotype H17. The 4 remaining minor haplotypes (H28, H29, H33 and G2) were found in only 1 individual each and belonged to 2 other genetic clades.

Haplotype identities and relative frequencies in NWIP sites largely resembled the haplotype composi-tion found in Venice, while they were conspicuously dif-ferent from samples collected in the cryptogenic range (Fig. 1). The Cantabrian locations were nearly identical to each other and they were composed by an even pres-

Table 2 Pairwise ΦPT estimates between the 8 studied populations. Significance levels obtained by permuting (10 100 permutations) haplotypes between populations

Mundaka San Vicente Venecia A Toxa Vilagarcía Poio San SimónSan Vicente −0.033NSVenecia 0.426*** 0.435***A Toxa 0.363*** 0.374*** 0.057NSVilagarcía 0.483*** 0.490*** −0.025NS 0.108***Poio 0.493*** 0.501*** 0.004NS 0.136*** −0.011NSSan Simón 0.670*** 0.698*** 0.194*** 0.412*** 0.142*** 0.123***Ponte do Porco 0.654*** 0.655*** 0.126*** 0.338*** 0.338*** 0.053NS −0.009NSNS, non-significant; * < 0.05, ** < 0.01 and *** < 0.001

Figure 3 Genetic differentiation, popu-lation structure, and haplotype composi-tion. Non-metric multidimensional scal-ing plot based on pairwise ΦPT-values for 8 populations of Cyclope neritea. Populations depicted as pie charts that represent haplotype frequencies (shad-ing follows clade codes from Fig. 2). Lines encircle populations with pair-wise distances below certain levels. The date of first documented occurrence is indicated (in parenthesis) for non-native northwest Iberian Peninsula sites.

292

123456789

101112131415161718192021222324252627282930313233343536373839404142434445464748495051

L. Couceiro et al.

© 2012 Wiley Publishing Asia Pty Ltd, ISZS and IOZ/CAS

ence of 2 haplotypes (H1, H18), none of which abound-ed in either Venice or the NWIP. In contrast, both Ven-ice and NWIP samples were dominated by haplotype H22, followed by haplotypes H17 and H16. The relative frequencies of these haplotypes varied among NWIP sites and the AMOVA estimated that a significant por-tion of the genetic variation could be attributed to inter-population differences; that is, whether the whole datas-et was included in the analysis (ΦPT = 0.356, P = 0.0000) or the analysis was restricted to NWIP sites (ΦPT = 0.171, P = 0.0000). The NMDS ordination plot based on pair-wise ΦPT values (stress value < 0.01) confirmed the considerable genetic distance between the samples col-lected within the native range in the Cantabrian coast and all other sites (Table 2 and Fig. 3). It also revealed that, within the NWIP group, samples seemed to be ar-ranged according to the date of the first recorded occur-rence of C. neritea, with the group Vilgarcía-Poio (first recorded occurrence: 2005) in an intermediate position between the oldest population of A Toxa (1992) and the recently established sites of Ponte do Porco and San Si-mon (2008). This temporal arrangement seems to be paralleled by a gradual increase in the relative frequen-cy of haplotype H22 from nearly one-third of the indi-viduals in A Toxa, to nearly three-quarters in Vilagarcía-Poio, and to approximately 90% in Ponte do Porco-San Simon. The haplotype composition of the sample from Venice was not significantly different from the non-na-tive populations of Vilagarcía, Poio and A Toxa. Final-ly, haplotype composition in NWIP non-native sites seemed notably stable with time. Pairwise ΦPT esti-mates between samples collected in 2006 and samples collected in May 2009 (A Toxa, Vilagarcía and Poio sites) were always non-significant; in fact, the 3 ΦPT es-timates were slightly negative (values between −0.013 and −0.037).

DISCUSSION

Northwest Iberian Peninsula versus French Atlantic coast

To date, non-native populations of C. neritea have been reported for the FAC (Sauriau 1991) and the NWIP coasts exclusively (Rolán 1992; Rolán et al. 2005; Quintela et al. 2006). The status of the cryptogen-ic range along the Cantabrian coast remains uncertain. Mollusc exchanges are the only known vector for the in-troduction and arrival of C. neritea into those 2 regions seems obviously linked to the fact that shellfish farming

is intense in both areas. The invasion appears to have gone through a comparable lag time in both regions (ap-proximately a decade between introduction and spread): a curious parallelism if we recall that the whole process took place nearly 2 decades later in the NWIP (introduc-tion in the early 1990s and spread in the early 2000s; Rolán 1992; Rolán et al. 2005; Quintela et al. 2006) than in France (introduction in 1976 and spread from 1983 onwards; Bachelet et al. 1980; Sauriau 1991). De-spite these similarities, differences also exist. Oysters represent a major industry in French sites, while NWIP seabed farming is focused on clams (several species). This discrepancy conceivably implies differences in the sources of the transferred shellfish that, given the strong phylogeographic signal displayed by C. neritea along its native range (Simon-Bouhet et al. 2006), seem to have left an imprint on the genetic makeup of the resulting non-native populations. In this regard, our results reveal obvious differences with those obtained for FAC popu-lations (Simon-Bouhet et al. 2006). The number of hap-lotypes per population is comparable in both regions (Mann–Whitney–Wilcoxon test, P = 0.444). However, genetic diversity, either based on haplotypes or on nucle-otides, is significantly higher in French sites (P = 0.037 and P = 0.022, respectively) because: (i) NWIP populations tend to be dominated by 1 single haplotype (H22), while French populations always include 3 or more moderate-ly frequent (i.e. < 50%) halotypes, and (ii) haplotypes along the FAC typically belong to 3 different gene lin-eages, while only 2 are commonly found in NWIP sites colonized during the 2000s.

Unlike previous assumptions, increased or similar levels of genetic diversity in comparison to native popu-lations seem rather common in successful aquatic inva-sions linked to high propagule vectors and/or to multi-ple introductions (reviewed in Roman & Darling 2007). This seems to be the case of species accidentally trans-ported with shellfish transplantations. In this regard, the high genetic diversity of FAC populations was in-terpreted as evidence of multiple introductions from several native areas of low local genetic diversity (Si-mon-Bouhet et al. 2006). By the same logic, the low-er diversity detected in NWIP populations, particular-ly in more recently settled sites (2005, 2008), reveals a less diversified range of sources. Several haplotypes common in French sites are rare, or altogether absent, in the NWIP, suggesting that several of the donor areas that contributed to the presence of C. neritea in French coasts have played a negligible role, if any, in the NWIP. Haplotype H1, for example, despite being characteristic

293

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051

Expansion of Cyclope neritea: genetic pattern

© 2012 Wiley Publishing Asia Pty Ltd, ISZS and IOZ/CAS

of the Iberian section of the native range of C. neritea, has never been detected in the NWIP (Simon-Bouhet et al. 2006). Likewise, H18 is extremely rare in the NWIP, but it abounds in the 3 southernmost FAC sites, proba-bly because H18 is native of oyster-culture lagoons in the French Mediterranean coast, with a long tradition of commercial oyster supplies to FAC sites (Sauriau 1991). Interestingly, both H1 and H18 were the only haplo-types found in the 2 sites sampled along the cryptogenic range in this study (Mundaka and San Vicente). There-fore, our data for the 2008 sites reinforce the previous conclusion that the genetic ancestry of the individuals living in NWIP sites invaded from 2000 onwards must likely be placed in a single region, the Veneto (Couceiro et al. 2008). The Veneto is the only native area where the 3 haplotypes that dominate NWIP populations (H22, H17 and H16) are likewise common (Simon-Bouhet et al. 2006; Venice sample in this study), while the haplo-types that dominate FAC sites are related to various re-gions along the native range of the species. As already noted (Couceiro et al. 2008), this conclusion is consis-tent with the fact that the Veneto region is a traditional source for shellfish (mainly Manila clams) imported into NWIP seabeds.

Range expansion in northwest Iberian Peninsula and implications for the management of the invasion

Between 1976 and 2003, C. neritea expanded its range nearly 700 km along the FAC (see records in Sau-riau 1991 and Bachelet et al. 2004). This value amounts to a mean rate of expansion of 25.9 km/year that com-pares surprisingly well with estimates obtained in other regions for marine invaders irrespective of whether they are direct developers, like C. neritea, or they go through some planktonic phase during their development (Gro-sholz 1996; Leppäkoski & Olenin 2000). Extrapolat-ed to the NWIP, these estimates imply that C. neritea might be found all across the region in a few years un-less appropriate control measures are put into effect. In fact, the discovery of 2 new colonies in the short time span elapsed since our previous study (Couceiro et al. 2008) suggests that C. neritea is already at its spread stage in NWIP rias. The new populations (Ponte do Por-co and San Simón) are located far from each other and quite distant from any previous sighting of the snail, at least Ponte do Porco. Furthermore, they are genetically undifferentiated from each other, suggesting a common origin, but they significantly deviate from the haplotype composition found in older enclaves located in-between

them. Altogether, these observations reveal a jump dis-persal pattern that seems beyond the reach of C. neritea, a snail believed to be a poor disperser because it com-bines direct development (Kisch 1950; Gomoïu 1964) with a need for sheltered, soft bottom habitats that com-monly appear as isolated enclaves along the shoreline (Nordsieck 1968). Instead, the detected jump dispersal pattern suggests the involvement of some external dis-persal vector and human transport appears as the most plausible option. Alternative non-human vectors (e.g. birds and fishes) have been occasionally proposed for the long-distance dispersal of molluscs (Darwin 1878); examples even include other non-native, direct-develop-ing gastropods (Leppäkoski & Olenin 2000). However, the frequency of clam transfers in the NWIP, together with the observation that C. neritea colonies are always close to shellfish beds, suggests that human mechanisms already offer a great deal of opportunities to disper-sal without the need to invoke natural substitutes. The relevance of human-mediated transport for long-dis-tance dispersal and its prevalence over alternative nat-ural vectors has been thoroughly demonstrated in other well-studied aquatic invasions (e.g. Johnson & Carlton 1996).

Initially, a spread with human intervention rather than by natural mechanisms could facilitate its control be-cause human routes should be more amenable to sur-veillance. However, some life-history traits of C. neritea might render this task more cumbersome than anticipat-ed. These traits might even promote human-mediated dispersal and subsequent successful establishment. First, direct development implies that newly hatched juveniles might not disperse much, thus boosting the growth of recently settled colonies (Johannesson 1988). Second, C. neritea lays small, single-embryo capsules firmly at-tached to hard substrates that, importantly, include the shells of live clams and/or conspecifics (Gomoïu 1964). This strategy might be useful in the propagation of the snail between clam beds when clams from C. neritea-infested sites are restocked into new areas or transferred to the shellfish depuration facilities that abound in the NWIP. Indeed, clam exchanges might represent an ide-al vector for dispersal because they must ensure that clams (and, presumably, accompanying fauna) arrive in good physiological condition at destiny. In this re-gard, clams from shellfish beds around our Poio location have been relocated to various sites in the Ría de Arou-sa (where C. neritea displays its widest presence) by the tens to the hundreds of metric tons annually in the last decade. To our knowledge, there has been no attempt

294

123456789

101112131415161718192021222324252627282930313233343536373839404142434445464748495051

L. Couceiro et al.

© 2012 Wiley Publishing Asia Pty Ltd, ISZS and IOZ/CAS

to check whether those clams might also carry egg cap-sules on their shells. However, this possibility, if proven true, might constitute a notable vector for the range ex-pansion of C. neritea within the region given the huge amount of clams involved in those restockings. Accord-ing to our own observations, adults of C. neritea collect-ed in March 2009 from various NWIP sites (Vilagarcía, Poio, A Toxa) regularly wore several capsules with live juveniles attached to their shells.

Genetic geographic pattern despite human-mediated spread and several reintroductions

It has been argued that human-mediated and natural dispersal are likely to leave different genetic signatures across populations of invasive species at the regional level, with human-mediated transport more likely lead-ing to a lack of correlation between genetic and geo-graphical distances (Dupont & Viard 2003). In contrast with these expectations, we obtained a seemingly non-random genetic pattern even in a scenario of human-me-diated propagation. As shown in Fig. 3, genetic distanc-es seem to depend on the known date of introduction into each location rather than on geographic position; populations established simultaneously (2005, 2008) are genetically similar despite considerable geographic sep-aration. Moreover, populations settled in 2005 occupy a position halfway between the older colony of A Toxa and those more recently detected in 2008. The pattern also involves an increase in genetic diversity with popu-lation age. Population age correlates with distance from the initial colony of A Toxa. Hence, the gradual change in genetic differentiation and genetic diversity with the year of population founding might be interpreted as driven by that correlation.

The pattern found in the NWIP evokes some of the ge-netic signatures expected under a serial-founder effect starting at a single origin (A Toxa), where each consec-utive colonization step would have involved a sampling from the previous subset of the original population. Ram-achandran et al. (2005) show that such a stepping stone colonization model can generate a gradual increase in ge-netic differentiation with increasing distance from the ini-tial source population and a concurrent decrease in genet-ic diversity. Because the serial-founder model provides a nonequilibrium alternative to explanations based on equi-librium theories (e.g. Kimura & Weiss 1964), it appears particularly well-suited for biological invasions (an ex-treme example of serial reduction in the genetic diver-sity of an introduced plant can be found in Amsellem et al. 2000). However, many studies of biological in-

troductions have failed to detect an increase in genet-ic differentiation with distance (e.g. Martel et al. 2004b; Therriault et al. 2005; Marrs et al. 2008; Dupont et al. 2009). Moreover, the serial-founder hypothesis has been explicitly considered, but rejected, in another aquat-ic invader (Herborg et al. 2007). Likewise, other details of our results strongly argue against a stepwise expan-sion from A Toxa as a plausible explanation for our pat-tern. Thus, the striking genetic similarity between the 2 sites settled in 2005 and the sample from Venice (a common source for Manila clam importations) suggests that those sites more probably derived from new rein-troductions from abroad (Couceiro et al. 2008). More-over, whenever 2 populations were founded on the same year, both in 2005 and in 2008, they resulted in a similar haplotype composition irrespective of their geographi-cal location, suggesting that the main determinant of the structure is time of settlement rather that distance from a hypothetical initial centre. Herborg et al. (2007) pro-pose a mechanism to explain the correlation between genetic differentiation and diversity with population age in another aquatic invader, the Chinese mitten crab. In their interpretation, new populations display a found-er bottleneck-associated genetic differentiation that lat-er cumulative gene flow gradually reduces. However, their mechanism does not seem applicable to C. neritea either. In contrast with the Chinese mitten crab, where genetic differentiation decreases with population age, recent populations of C. neritea in the NWIP are genet-ically similar provided that they were founded in the same year. The Chinese mitten crab was first observed in Europe in 1912 and the youngest population includ-ed in the aforementioned study dates from 1988. In con-trast, the arrival of C. neritea to NWIP waters seems too recent to invoke a mechanism where a gradual genetic homogenization plays a central role. In fact, our re-sam-pling of several NWIP sites reveals that haplotype fre-quency has remained unaltered, at least on a short-time basis (3 years). This stability contrasts with the notable genetic drift among cohorts found in the Chinese mitten crab (Herborg et al. 2007). Hence, and while we cannot exclude that a gradual homogenization might occur in the future, we have found no evidence that the current genetic structure could be the outcome of an analogous process.

In our interpretation, the sites founded in 2005 prob-ably derived from a new reintroduction from Venice. Whether this reintroduction also added new individuals to the older population of A Toxa is unknown, but, irre-spective of the scenario, it seems obvious that its influ-

295

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051

Expansion of Cyclope neritea: genetic pattern

© 2012 Wiley Publishing Asia Pty Ltd, ISZS and IOZ/CAS

ence on A Toxa did not erode the significant genetic dif-ferentiation between this older colony and more recently settled ones. C. neritea greatly expanded its range with-in the NWIP between 2006 (Couceiro et al. 2008) and 2008. The significant genetic similarity between the newest colonies suggests that this range expansion pos-sibly also derived from a single introduction event. Un-fortunately, we cannot resolve whether this new intro-duction event represented another reintroduction from abroad or just a secondary introduction within the re-gion. In contrast with 2005 sites, haplotype composition in the newest sites does not particularly resemble any of the populations studied along the native range (Simon-Bouhet et al. 2006; this study). Yet, Simon-Bouhet et al. (2006) show that even within the Veneto region, local populations of C. neritea display significant differences in haplotype composition. Our inability to conclusive-ly determine the source of the individuals that settled the 2008 sites is disappointing (see below). However, it does not dwarf the observation that even in a scenar-io with at least 2 independent reintroductions, C. neritea has developed a region-wide genetic pattern. A regional-scale genetic arrangement is not new to non-native C. neritea. FAC populations have been reported to fit to an isolation-by-distance (IBD) scheme (Simon-Bouhet et al. 2006). In fact, the hypothesis of a cryptogenic or-igin for this species along the FAC came, in part, from an understandable difficulty to accept that random hu-man-mediated exchanges between shellfish culture sites might have developed a spatial genetic pattern (Simon-Bouhet et al. 2006). As noted for other aquatic invaders (Herborg et al. 2007), our results suggest that a gradu-al increase in genetic differentiation with distance does not necessarily involve a migration-drift equilibrium. As shown above, FAC and NWIP invasions show nota-ble differences. However, it is interesting to note that as C. neritea invaded French coasts gradually from south to north, geographical distances between French sites might be paralleled by differences in the date of intro-duction (detailed in Bachelet et al. 2004). It might be in-teresting to investigate whether a process analogous to the one observed in NWIP sites (isolation-by-date-of-in-troduction paralleled by geographic distance) has had had some influence on the occurrence of IBD among FAC sites.

CONCLUSION The introduction of non-native molluscs into NWIP

waters has been steadily growing from the late-1980s

onwards (Bañón et al. 2008). As for C. neritea, most of these exotics seem linked to unintentional co-transport with commercial bivalves from the Adriatic Sea (refer-ences in Bañón et al. 2008). However, and unlike C. ne-ritea, they include many predatory gastropods with a potential for negative impacts on the receiving system. A better understanding of the sources and routes of these introductions, even if obtained from a non-predatory ex-otic such as C. neritea, could provide some guidance for their effective management. Shellfish culture has played a huge role in spreading nonindigenous marine organ-isms worldwide (Ribera Siguán 2003). In the NWIP, the shellfish industry is run by many different players: semi-cooperative shellfisher organizations, private con-cessioners of small shellfish bed pots and shellfish dep-uration facilities. As a result, shellfish translocations are not always as tightly monitored by regional fisher-ies and environment authorities as would be desirable. In particular, private concessioners often run their busi-ness with minimal concern about conservation issues. The resultant network of shellfish translocations, in an example of ‘small decision effects’ (Odum 1982), prob-ably explains the introduction and later range expansion of C. neritea within the region. Against this background, the contrast between the number and diversity of sourc-es detected in the NWIP and those reported for French sites, where multiple introductions are reported for C. neritea as well as for other gastropods also co-transport-ed with commercial shellfish (Martel et al. 2004a; Viard et al. 2006), evidences that shellfish culture practices impose different invasion pressures on each region. The scarcity of donor areas inferred for the NWIP recalls the geographically restricted origins of some transocean-ic introductions also linked to shellfish transplantations (e.g. Miura et al. 2006). From an applied point of view, it seems reasonable to anticipate that pre-emptive mea-sures against further reintroductions should be easier to implement in the NWIP than in France if, as predicted, a narrower range of sources is accompanied by a narrow-er range of routes of introduction. The great majority of pathways for non-native invasions are associated with the expansion of global trade, and the impetus of this expansion acts against the reduction of pathways for in-vasions (Ruiz & Carlton 2003). In this regard, and in an effort to counteract this trend at least on a regional scale, further clam transfers from the Adriatic to the NWIP, or from intermediate sites that might have received Adri-atic clams themselves, should be carefully screened for the presence of C. neritea before approval.

296

123456789

101112131415161718192021222324252627282930313233343536373839404142434445464748495051

L. Couceiro et al.

© 2012 Wiley Publishing Asia Pty Ltd, ISZS and IOZ/CAS

ACKNOWLEDGMENTS Financial support for this work was provided by

the Spanish Ministerio de Educación y Ciencia grant CTM2004-04496/MAR (partially co-founded by Fon-do Europeo de Desarrollo Regional) and the Xunta de Galicia grant PGIDT05PXIC10302PN. The authors are indebted to Marco Sigovini (CNR-ISMAR) for kindly providing the sample from Venice, Italy.

REFERENCES Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ

(1990). Basic local alignment search tool. Journal of Molecular Biology 215, 403–10.

Amsellem L, Noyer JL, Le Bourgeois T, Hossaert-McK-ey M (2000). Comparison of genetic diversity of the invasive weed Rubus alceifolius Poir. (Rosaceae) in its native range and in areas of introduction, us-ing amplified fragment length polymorphism (AFLP) markers. Molecular Ecology 9, 443–55.

Bachelet G, Cazaux C, Gantès H, Labourg PJ (1980). Contribution à l’étude de la faune marine de la région d’Arcachon, IX. Bulletin du Centre d’Etudes et de Recherches Scientifiques Biarritz 13, 45–64.

Bachelet G, Simon-Bouhet B, Desclaux C et al. (2004). Invasion of the eastern Bay of Biscay by the nassar-iid gastropod Cyclope neritea: origin and effects on resident fauna. Marine Ecology Progress Series 276, 147–59.

Bandelt HJ, Forster P, Rohl A (1999). Median-joining networks for inferring intraspecific phylogenies. Mo-lecular Biology and Evolution 16, 37–48.

Bañón R, Rolán E, García-Tasende M (2008). First re-cord of the purple dye murex Bolinus brandaris (Gastropoda: Muricidae) and a revised list of non na-tive molluscs from Galician waters (Spain, NE Atlan-tic). Aquatic Invasions 3, 331–4.

Blackburn TM, Pysek P, Bacher S et al. (2011). A pro-posed unified framework for biological invasions. Trends in Ecology and Evolution 26, 333–9.

Carlton JT (1996). Biological invasions and cryptogenic species. Ecology 77, 1653–5.

Clarke KR, Gorley RN (2006). PRIMER v6: User Man-ual/Tutorial. PRIMER-E, Plymouth.

Couceiro L, Míguez A, Ruiz JM, Barreiro R (2008). In-troduced status of Cyclope neritea (Gastropoda, Nas-sariidae) in the NW Iberian Peninsula confirmed by mitochondrial sequence data. Marine Ecology Prog-ress Series 354, 141–6.

Davis MA (2009). Invasion Biology. Oxford Biology, New York.

Darwin C (1878). Transplantation of shells. Nature 18, 120–21.

Dupont L, Viard F (2003). Isolation and characteriza-tion of highly polymorphic microsatellite markers from the marine invasive species Crepidula fornica-ta (Gastropoda: Calyptraeidae). Molecular Ecology Notes 3, 498–500.

Dupont L, Viard F, Dowell MJ, Wood C, Bishop JDD (2009). Fine- and regional-scale genetic structure of the exotic ascidian Styela clava (Tunicata) in south-west England, 50 years after its introduction. Molec-ular Ecology 18, 442–53.

Elton CS (1958). The Ecology of Invasions by Animals and Plants. Methuen, London.

Estoup A, Largiader CR, Perrot E, Chourrout D (1996). Rapid one-tube DNA extraction for reliable PCR de-tection of fish polymorphic markers and transgenes. Molecular Marine Biology and Biotechnology 5, 295–298.

Excoffier L, Laval G, Schneider S (2005). Arlequin ver. 3.0: an integrated software package for population genetics data analysis. Evolutionary Bioinformatics Online 1, 47–50.

Excoffier L, Smouse PE, Quattro JM (1992). Analysis of molecular variance inferred from metric distances among DNA haplotypes: application to human mito-chondrial DNA restriction data. Genetics 21, 479–91.

Gomoïu MT (1964). Biologishes Studium der Arten Nassa reticulata (L.) und Cyclonassa neritea (L.) im schwarzen Meer (rumänisher Küstenbereich). Revue Roumaine de Biologie. Série de Zoologie 9, 39–49.

Grosholz ED (1996). Contrasting rates of spread for in-troduced species in terrestrial and marine systems. Ecology 77, 1680–86.

Grosholz ED, Ruiz GM, Dean CA, Shirley KA, Maron JL, Connors PG (2000). The impacts of a nonindige-nous marine predator in a California bay. Ecology 81, 1206–24.

Gurevitch J, Padilla DK (2004). Are invasive species a major cause of extinctions? Trends in Ecology & Evolution 19, 470–74.

Hastings A, Cuddington K, Davies KF et al. (2005). The spatial spread of invasions: new developments in the-ory and evidence. Ecology Letters 8, 91–101.

Hayes KR, Barry SC (2008). Are there any consistent predictors of invasion success? Biological Invasions 10, 483–506.

297

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051

Expansion of Cyclope neritea: genetic pattern

© 2012 Wiley Publishing Asia Pty Ltd, ISZS and IOZ/CAS

Herborg L-M, Weetman D, Van Oosterhout C, Hanfling B (2007). Genetic population structure and contem-porary dispersal patterns of a recent European invad-er, the Chinese mitten crab, Eriocheir sinensis. Mo-lecular Ecology 16, 231–42.

Hidalgo JG (1917). Fauna malacológica de España, Por-tugal y las Islas Baleares. Trabajos del Museo Nacio-nal de Ciencias Naturales 30, 1–752.

Hochberg Y (1974). Some generalizations of the T-method in simultaneous inference. Journal of Multi-variate Analysis 4, 244–34.

Honnay O, Jacquemyn H (2008). A meta-analysis of the relation between mating system, growth form and ge-notypic diversity in clonal plant species. Evolution-ary Ecology 22, 299–312.

Johannesson K (1988). The paradox of Rockall – why is a brooding gastropod (Littorina saxatilis) more wide-spread than one having a planktonic larval dispersal stage (Littorina littorea). Marine Biology 99, 507–13.

Johnson LE, Carlton JT (1996). Post-establishment spread in large-scale invasions: dispersal mechanisms of the zebra mussel Dreissena polymorpha. Ecology 77, 1686–90.

Kimura M, Weiss GH (1964). The stepping stone mod-el of population structure and the decrease of genetic correlation with distance. Genetics 49, 561–76.

Kisch BS (1950). Growth stages of Cyclope neritea (L.). Journal of Conchology 23, 108.

Kolar CS, Lodge DM (2001). Progress in invasion biol-ogy: predicting invaders. Trends in Ecology & Evolu-tion 16, 199–204.

Leppäkoski E, Olenin S (2000). Non-native species and rates of spread: lessons from the brackish Baltic Sea. Biological Invasions 2, 151–63.

Librado P, Rozas J (2009). DnaSP v5: a software for comprehensive analysis of DNA polymorphism data. Bioinformatics 25, 1451–2.

Lockwood JL, Hoopes MF, Marchetti MP (2007). Inva-sion Ecology. Blackwell Publishing, Malden, MA.

Marrs RA, Sforza R, Hufbauer RA (2008). Evidence for multiple introductions of Centaurea stoebe micran-thos (spotted knapweed, Asteraceae) to North Ameri-ca. Molecular Ecology 17, 4197–208.

Martel C, Viard F, Bourguet D, Garcia-Meunier P (2004a). Invasion by the marine gastropod Ocinebrel-lus inornatus in France: I. Scenario for the source of introduction. Journal of Experimental Marine Biolo-gy and Ecology 305, 155–70.

Martel C, Viard F, Bourguet D, Garcia-Meunier P (2004b). Invasion by the marine gastropod Ocine-brellus inornatus in France. II. Expansion along the Atlantic coast. Marine Ecology Progress Series 273, 163–72.

Miura O, Torchin ME, Kuris AM, Hechinger RF, Chiba S (2006). Introduced cryptic species of parasites ex-hibit different invasion pathways. Proceedings of the National Academy of Sciences of the United States of America 103, 19818–23.

Morton JE (1960). The habits of Cyclope neritea, a style-bearing stenoglossan gastropod. Proceedings of the Malacological Society of London 34, 96–105.

National Research Council (2002). Predicting Invasions of Nonindigenous Plants and Plant Pests. National Academy Press, Washington, DC.

Nobre A (1932). Moluscos Marinhos de Portugal. Insti-tuto Zoológico da Universidade do Porto, Imprensa Portuguesa, Porto.

Nordsieck F (1968). Die europäischen Meeres-Gehäu-seschnecken. (Prosobranchia). Vom Eismeer bis Kap-verden und Mittelmeer. G. Fischer Verlag, Stuttgart.

Odum WE (1982). Environmental degradation and the tyranny of small decisions. Bioscience 32, 728–9.

Perrings C, Williamson MH, Dalmazzone S (2000). The Economics of Biological Invasions. Edward Elgar, Cheltenham, UK.

Pimentel D (2002). Biological Invasions: Economic and Environmental Costs of Alien Plant, Animal, and Mi-crobe Species. CRC Press, Boca Raton, Florida.

Quintela M, Couceiro L, Ruiz JM, Barreiro R (2006). Discovery of imposex in the gastropod Cyclope ne-ritea now invading Galicia (north-west Spain). Jour-nal of the Marine Biological Association of the Unit-ed Kingdom 86, 1171–3.

Ramachandran S, Deshpande O, Roseman CC, Rosen-berg NA, Feldman MW, Cavalli-Sforza LL (2005). Support from the relationship of genetic and geo-graphic distance in human populations for a seri-al founder effect originating in Africa. Proceedings of the National Academy of Sciences of the United States of America 102, 15942–7.

Ribera Siguán MA (2003). Pathways of biological inva-sions of marine plants. In: Hewitt CL, Campbell ML, Gollasch S, eds. Invasive Species: Vectors and Man-agement Strategies. Island Press, Washington, DC, pp. 183–226.

298

123456789

101112131415161718192021222324252627282930313233343536373839404142434445464748495051

L. Couceiro et al.

© 2012 Wiley Publishing Asia Pty Ltd, ISZS and IOZ/CAS

Richardson DM, Pysek P (2008). Fifty years of invasion ecology – the legacy of Charles Elton. Diversity and Distributions 14, 161–8.

Rolán E (1992). Dos especies más de moluscos medi-terráneos introducidos en la bahía de O Grove. Thal-assas 10, 135.

Rolán E, Horro J, Quintela M (2005). Cyclope neritea (Gastropoda, Nassariidae) en Galicia. Noticiario SEM 44, 57–9.

Roman J, Darling JA (2007). Paradox lost: genetic di-versity and the success of aquatic invasions. Trends in Ecology & Evolution 22, 454–64.

Ruiz GM, Carlton JT (2003). Invasive Species: Vectors and Management Strategies. Island Press, Washing-ton, DC.

Sauriau P-G (1991). Spread of Cyclope neritea (Mol-lusca: Gastropoda) along the northern-eastern Atlan-tic coasts in relation to oyster culture and to climatic fluctuations. Marine Biology 109, 299–309.

Sax DF, Gaines SD (2008). Species invasions and ex-tinction: the future of native biodiversity on islands. Proceedings of the National Academy of Sciences of the United States of America 105, 11490–97.

Sax DF, Stachowicz JJ, Gaines SD (2005). Species In-vasions: Insights into Ecology, Evolution, and Bioge-ography. Sinauer Associates, Sunderland, MA.

Simberloff D (2004). A rising tide of species and litera-ture: a review of some recent books on biological in-vasions. Bioscience 54, 247–54.

Simberloff D, Gibbons L (2004). Now you see them, now you don’t – population crashes of established in-troduced species. Biological Invasions 6, 161–72.

Simon-Bouhet B (2006). Expansion d’aire et processus d’introductions biologiques en milieu marin: le cas de Cyclope neritea (Nassariidae) sur les côtes fran-çaises. (PhD thesis). Université de La Rochelle, La Rochelle.

Simon-Bouhet B, Garcia-Meunier P, Viard F (2006). Multiple introductions promote range expansion of the mollusc Cyclope neritea (Nassariidae) in France: evidence from mitochondrial sequence data. Molecu-lar Ecology 15, 1699–711.

Southward AJ, Southward EC, Dando PR et al. (1997). Behaviour and feeding of the nassariid gastropod Cy-clope neritea, abundant at hydrothermal brine seeps off Milos (Aegean Sea). Journal of the Marine Bio-logical Association of the United Kingdom 77, 753–71.

Spalding MD, Fox HE, Allen GR et al. (2007). Ma-rine ecoregions of the world: a bioregionalization of coastal and shelf areas. Bioscience 57, 573–83.

Therriault TW, Orlova MI, Docker MF, MacIsaac HJ, Heath DD (2005). Invasion genetics of a freshwater mussel (Dreissena rostriformis bugensis) in eastern Europe: high gene flow and multiple introductions. Heredity 95, 16–23.

Viard F, Ellien C, Dupont L (2006). Dispersal abili-ty and invasion success of Crepidula fornicata in a single gulf: insights from genetic markers and lar-val-dispersal model. Helgoland Marine Research 60, 144–52.

Williamson M (1996). Biological Invasions. Chapman & Hall, London.