Fine-scale genetic structure across a New Zealand disjunction for the direct-developing intertidal whelk Cominella maculosa (Gastropoda: Buccinidae)

Fine-scale genetic structure across a New Zealand disjunction for the direct-developing... Abstract The spotted whelk, Cominella maculosa, is a common intertidal species endemic to New Zealand with crawl-away juveniles that hatch from benthic egg capsules. The aim of this study was to determine the fine-scale pattern of genetic differentiation and over what distances this direct-developing species can form isolated populations, across a region that includes a known genetic disjunction. Whelks were collected from seven locations, 10–26 km apart, along 125 km of a linear coastline, and DNA sequences from the mitochondrial cytochrome c oxidase subunit I (COI) gene were compared. Four northern sites were genetically differentiated from each other and from the group of three southern sites. The three southern sites showed no genetic differentiation, consistently clustered together in a single group, shared a dominant haplotype and were characterized by a low number of private haplotypes and low haplotype and nucleotide diversity. The two sites located ~10 km north of the southern group lacked private haplotypes and contained high frequencies of both dominant northern and southern haplotypes. The lack of an isolation-by-distance pattern of structure, and a leapfrog-like dispersal grouping of non-neighbouring sites, suggests sporadic long-distance dispersal events occur in this species, even if rarely. INTRODUCTION Dispersal ability is a key factor that mediates the level of population connectivity (Selkoe & Toonen, 2011). In the marine environment, species with planktonic larvae that are in the water column for days to weeks, or longer, have a high dispersal potential. Gene flow is therefore expected to be high and population genetic differentiation low (Scheltema, 1971, 1986; Burton & Feldman, 1982; Collin, 2001), although at relatively large geographical scales some studies have shown only a weak relationship between population structure and the duration of the planktonic larval stage. Species that lack planktonic larvae and hatch from brooded or benthic eggs with direct development of juveniles typically have lower dispersal potential. These species are more likely to form genetically differentiated populations (Lee & Boulding, 2009; Karl & Hayes, 2012; Dawson et al., 2014), which could be either discrete groups or a collection of sites that form a pattern of isolation by distance (IBD). However, the relationship between geographical and genetic distance is not always linear because the distribution of suitable habitat can be patchy, and environments are often heterogeneous and variable. IBD can be significant at large scales but may not appear at smaller scales (Hoffman et al., 2013; Giles et al., 2015). Parts of a distribution can contain populations that have an IBD pattern, and yet contain surprisingly abrupt genetic disjunctions between neighbouring populations. These types of genetic breaks have been reported for intertidal snails on the west coast of North America (Kyle & Boulding, 2000) and in two species at the same latitude along the Chilean coastline (Sánchez et al., 2011; Brante, Fernández & Viard, 2012). Observations of genetic disjunctions, particularly in the absence of obvious geographical barriers, make it difficult to establish the general causes of isolation among populations. There is a need for more studies that attempt to document the fine-scale genetic structure of direct-developer species, which will enable a comparative approach to identify common patterns and processes underlying the genetic structuring of coastal marine species with lower dispersal potential (Pelc, Warner & Gaines, 2009). New Zealand is an archipelago that includes two main islands (North and South) with long, mostly linear, coastlines that span the temperate range from the subtropics to the subantarctic (Fig. 1A, B). Major current systems from the west split as they hit the continental shelf and flow around the North and South Islands, forming a complex system of coastal and offshore currents and eddies (Heath, 1982; de Lange et al., 2003; Laing & Chiswell, 2003). Previous studies have shown that planktonic dispersers and direct-developers exhibit a general pattern of clustering into northern and southern populations (reviewed by Ross et al., 2009), with the Cook Strait region dividing them (Fig. 1B). However, most studies have used broad sets of sample sites to cover as much area as possible. Thus, there is a lack of fine scale genetic sampling that can be used to describe the patterns in and around genetic breaks. Figure 1. View largeDownload slide A. The position of New Zealand, within the dotted region, at the global scale. B. New Zealand, with the two mainland islands (North and South Island) divided by Cook Strait. The dotted region defines the southern tip of the North Island. C. Southern tip of the North Island, including Wellington Harbour to the west and the Wairarapa coast to the east. Sample sites are marked on the map; see Table 1 for site codes and scale bar for distance. Pie charts show the distribution of COI haplotypes for Cominella maculosa at sites sampled along the Wairarapa coast; circle size represents sample size. White colour represents private haplotypes; all other haplotypes are shared between two or more sites. Figure 1. View largeDownload slide A. The position of New Zealand, within the dotted region, at the global scale. B. New Zealand, with the two mainland islands (North and South Island) divided by Cook Strait. The dotted region defines the southern tip of the North Island. C. Southern tip of the North Island, including Wellington Harbour to the west and the Wairarapa coast to the east. Sample sites are marked on the map; see Table 1 for site codes and scale bar for distance. Pie charts show the distribution of COI haplotypes for Cominella maculosa at sites sampled along the Wairarapa coast; circle size represents sample size. White colour represents private haplotypes; all other haplotypes are shared between two or more sites. Cominella maculosa Martyn, 1784, the spotted whelk, is a direct-developing buccinid whelk endemic to New Zealand. It is a common carnivorous scavenger found in the rocky intertidal zone throughout the North Island, and the northern part of the South Island (Morton & Miller, 1968; Powell, 1979). Females communally lay masses of egg capsules on hard substrata over the summer, with an average of seven hatchlings emerging from each capsule 9 weeks later (Carrasco & Phillips, 2014). A recent phylogeographical study of C. maculosa by Fleming et al. (2018) found discrete population structure with an IBD pattern throughout the North Island distribution; however, the study also showed three genetic disjunctions that each occurred over a relatively small geographical scale (< 130 km). One of these disjunctions occurs along the Wairarapa coast of the North Island between two sites, Cape Palliser and Castlepoint, ~125 km apart, along a linear stretch of coastline where rocky outcrops are interspersed with sandy beaches (Fig. 1C). This section of the east coast of the North Island has not been well sampled in other studies of population genetic structure, especially for direct-developing species. Nevertheless, there is limited evidence that there is no disjunction along this coast for species with planktonic larvae (snakeskin chiton: Veale & Lavery, 2011; cockle: Ross et al., 2012). The aim of this study was to use fine-scale sampling (i.e. from sites tens of kilometres apart) to investigate the genetic disjunction along the Wairarapa coast of New Zealand (Fig. 1C) for the direct-developing whelk C. maculosa. METHODS Sampling and DNA extraction Cominella maculosa individuals were collected in 2015 from rocky shore tide pools that had been baited with dead mussels at low tide. Whelks > 10 mm were collected over a 2-h period and either placed directly in 80% ethanol or transported in seawater and frozen before processing. A total of 250 specimens were collected from five locations between Castlepoint (CA) and Cape Palliser (CP). From north to south, these sites included Riversdale (RI), Flat Point (FP), Honeycomb Rock (HR), Glendhu (GL), and Tora (TO) (Fig. 1C, Table 1). Distances between neighbouring sites varied from ~10 to 26 km. Foot tissue was dissected and preserved in 95% ethanol. Table 1. Sample information, genetic diversity indices and neutrality tests for Cominella maculosa sampled along the Wairarapa coast of the North Island of New Zealand Sample information  Diversity indices  Neutrality/tests  Site code  Sample site  Latitude/l/ongitude  N  S  H  pH  h (SD)  π (SD)  k (SD)  Fu’s FS (P)  Tajima D (P)  CA  Castlepoint  −40.865833/176.234500  47  9  4  1  0.68 (0.04)  0.005 (0.003)  3.19 (1.68)  5.65 (0.97)  1.60 (0.95)  RI  Riversdale  −41.076663/176.086164  56  3  4  2  0.23 (0.07)  0.000 (0.001)  0.24 (0.28)  −2.47 (0.03)  −1.27 (0.08)  FP  Flat Point  −41.246383/175.958717  53  9  3  0  0.53 (0.02)  0.005 (0.003)  3.17 (1.67)  7.90 (0.99)  1.66 (0.95)  HR  Honeycomb Rock  −41.346634/175.825988  47  6  2  0  0.26 (0.07)  0.003 (0.001)  1.55 (0.95)  5.67 (0.98)  0.37 (0.68)  GL  Glendhu  −41.400731/175.716665  59  1  2  1  0.10 (0.05)  0.000 (0.000)  0.10 (0.17)  −0.57 (0.14)  −0.70 (0.22)  TO  Tora  −41.504460/175.528460  35  1  2  1  0.06 (0.05)  0.000 (0.000)  0.06 (0.13)  −1.34 (0.05)  −1.14 (0.13)  CP  Cape Palliser  −41.600000/175.283333  27  1  2  1  0.14 (0.09)  0.000 (0.000)  0.14 (0.21)  −0.35 (0.17)  −0.73 (0.21)  All      324  16  10  6  0.61 (0.02)  0.006 (0.003)  3.40 (1.74)  3.75 (0.87)  0.86 (0.84)  Sample information  Diversity indices  Neutrality/tests  Site code  Sample site  Latitude/l/ongitude  N  S  H  pH  h (SD)  π (SD)  k (SD)  Fu’s FS (P)  Tajima D (P)  CA  Castlepoint  −40.865833/176.234500  47  9  4  1  0.68 (0.04)  0.005 (0.003)  3.19 (1.68)  5.65 (0.97)  1.60 (0.95)  RI  Riversdale  −41.076663/176.086164  56  3  4  2  0.23 (0.07)  0.000 (0.001)  0.24 (0.28)  −2.47 (0.03)  −1.27 (0.08)  FP  Flat Point  −41.246383/175.958717  53  9  3  0  0.53 (0.02)  0.005 (0.003)  3.17 (1.67)  7.90 (0.99)  1.66 (0.95)  HR  Honeycomb Rock  −41.346634/175.825988  47  6  2  0  0.26 (0.07)  0.003 (0.001)  1.55 (0.95)  5.67 (0.98)  0.37 (0.68)  GL  Glendhu  −41.400731/175.716665  59  1  2  1  0.10 (0.05)  0.000 (0.000)  0.10 (0.17)  −0.57 (0.14)  −0.70 (0.22)  TO  Tora  −41.504460/175.528460  35  1  2  1  0.06 (0.05)  0.000 (0.000)  0.06 (0.13)  −1.34 (0.05)  −1.14 (0.13)  CP  Cape Palliser  −41.600000/175.283333  27  1  2  1  0.14 (0.09)  0.000 (0.000)  0.14 (0.21)  −0.35 (0.17)  −0.73 (0.21)  All      324  16  10  6  0.61 (0.02)  0.006 (0.003)  3.40 (1.74)  3.75 (0.87)  0.86 (0.84)  Analyses are based on 610-bp sequences from the COI gene. Latitude and longitude are in decimal degrees. N, sample size; S, segregating sites; H, number of haplotypes; pH, number of private haplotypes; h, haplotype diversity; π, nucleotide diversity; k, nucleotide differences. View Large A small amount of foot tissue was digested for 2 h at 60 °C in 600 μL extraction buffer (12 mm Tris-HCl pH 8.0, 47.5 mm NaCl, 9.5 mm EDTA, 0.19% SDS, 0.5 mg proteinase-K). DNA was isolated using a modified phenol–chloroform method (Sambrook, Fritsch & Maniatis, 1989). The digested samples were twice mixed with 600 µL of phenol/chloroform/isoamyl alcohol (25:24:1), followed by the addition of chloroform/isoamyl alcohol (24:1). DNA was precipitated using 2.5 volumes of ethanol and 3 m sodium acetate and centrifuged for 30 min at 4 °C. Pelleted DNA was washed using 70% ethanol and then dried in a vacuum centrifuge (Eppendorf Concentrator 5301) for 5–10 min. Purified DNA was re-suspended in 30–70 µL of TE buffer (10 mm Tris-HCl pH 8.0, 1 mm EDTA) and its concentration and quality was determined using a Nanodrop spectrophotometer (Thermo Fisher Scientific). PCR sequencing and alignment A portion of the mitochondrial cytochrome c oxidase subunit I (COI) gene was amplified by PCR using primers LCO1490 and HCO2198 (Folmer et al., 1994). The PCR was conducted in 25-µL volumes that contained 67 mm Tris-HCl pH 8.8, 16 mm (NH4)2SO4, 3.0 mm MgCl2, 0.6 mg/mL bovine serum albumin, 0.05 units Taq DNA polymerase (Bioline), 0.4 mm of each dNTP, 0.1 μm of each primer and 120 ng/μL of template DNA. Thermocycling was carried out on a Biometra TProfessional Thermocycler under the following conditions: 5 min at 95 °C, 40 cycles of 35 s at 95 °C, 35 s at 50 °C and 45 s at 72 °C, followed by a final 10-min extension at 74 °C. PCR products were prepared using ExoSAP-It (Amersham Parmacia Biotech) and the DNA sequences were determined using a 3730xl DNA Analyzer (Applied Biosystems) at the Macrogen sequencing service (Macrogen Inc.). The 710-bp DNA sequences were quality checked and edited by eye before being trimmed to 610 bp using Geneious 8.1.6 (Kearse et al., 2012). The new DNA sequences reported here have been deposited in GenBank under accession numbers MG583718–MG583721. From Fleming et al. (2018), 28 sequences from Castlepoint and 27 sequences from Cape Palliser were added to the dataset. Another 19 Castlepoint sequences were extracted from donated specimens and added to the dataset. All 324 sequences were aligned using the Geneious alignment algorithm under default settings. Statistical analyses Haplotypes were identified using DnaSP 5.10.01 (Librado & Rozas, 2009). Genetic diversity was determined using Arlequin 3.5.2.2 (Excoffier & Lischer, 2010). Measures of diversity included the number of segregating sites (S), transitions, transversions, nucleotide differences (k), number of haplotypes (H), haplotype diversity (h), nucleotide diversity (π) and nucleic acid composition. Historical demographic changes were tested in Arlequin 3.5.2.2 using Tajima’s D statistic (Tajima, 1989) and Fu’s FS statistic (Fu, 1997). Due to the nature of the Fu’s FS statistic, only P-values below 0.02 are significant (Fu, 1997; Excoffier & Lischer, 2015). Haplotype relationships were visualized with a minimum spanning haplotype network (Bandelt, Forster & Röhl, 1999) created in PopART (http://popart.otago.ac.nz, accessed 23 August 2017). Differentiation between populations was estimated in Arlequin 3.5.2.2 (Excoffier & Lischer, 2010) using haplotype frequency to calculate the fixation index, ɸST, and 10000 permutations to test for significance. To test for a pattern of IBD, a Mantel test was performed under the one-dimensional model of habitat (Rousset, 1997) using Slatkin’s linearized genetic differentiation and untransformed distance. Mantel tests were conducted in Arlequin 3.5.2.2 with 10000 permutations (Excoffier & Lischer, 2010). Distances between populations were determined as the shortest distance through water using Google Earth v7.1.5 (Google Inc., 2015). Due to the linear nature of the Wairarapa coast, the shortest distance through water is also nearly equal to the coastal distance. Analysis of molecular variance (AMOVA) was used to partition total molecular variance into three hierarchical components, each with an associated fixation index (Excoffier, Smouse & Quattro, 1992). These indices describe variation within all populations (ɸST), between populations within a predefined group (ɸSC) and between all predefined groups (ɸCT) (Excoffier et al., 1992). The best group structure is determined by maximizing the between-group variation (i.e. largest significant ɸCT). Three a priori groupings were tested with AMOVA in Arlequin 5.2.2.2 (Excoffier & Lischer, 2010) using 10000 permutations of haplotype frequency data. The first two groupings tested for a break between northern and southern locations. The first scenario included HR in the northern group while the second scenario included HR in the southern group. The third grouping tested the hypothesis that two or more neighbouring populations formed distinct groups. This third scenario combined CA with RI, FP with HR, and the three southern sites together (GL, TO, CP). While AMOVA is useful for testing hypotheses of population structure, it requires a priori assignment of groups. A priori bias can be removed by conducting spatial analysis of variance (SAMOVA), which clusters populations into a predefined number of groups with the aim of maintaining geographical homogeneity while maximizing between-group variation (Dupanloup, Schneider & Excoffier, 2002). The optimum number of groups is determined by identifying the largest significant ɸCT from iterative analysis using two to N − 1 groups, where N is the total number of populations. Iterative runs of SAMOVA were performed for two to six groups with settings to retain geographical homogeneity using SAMOVA 2.0 (Dupanloup et al., 2002). The best SAMOVA grouping was identified as the configuration producing the largest significant ɸCT statistic. To accurately compare SAMOVA groupings to a priori groups, the best SAMOVA grouping was analysed by AMOVA in Arlequin 3.5.2.2 (Excoffier & Lischer, 2010). RESULTS Sequences from a total of 324 whelks [55 sequences from Fleming et al. (2018) and 269 new for this study] were analysed from seven locations along 125 km of Wairarapa coastline (Fig. 1C; Table 1). Ten haplotypes were identified with 16 segregating sites, 13 of which were transition-type changes. Most of the segregating sites were synonymous substitutions, but one site (T to C; bp 270) created a non-synonymous change (valine to alanine) in the haplotype shared by CA and RI. Five of the seven locations contained at least one private haplotype, but the central locations of HR and FP exhibited no private haplotypes (Table 1). Overall haplotype diversity was 0.61. The highest levels of diversity (0.68 and 0.53) were found in two northerly locations (CA and FP, respectively). All other locations had much lower haplotype diversity, ranging from 0.06 to 0.26 (Table 1). Nucleotide diversity within locations showed the same pattern, with overall diversity being similar to the highest location-level values (CA and FP; Table 1). Even these ‘high’ values are quite low (0.005), but values for the other locations were even lower (0.0001 for TO). The average number of nucleotide differences ranged from 3.19 for the most diverse location (CA) to 0.06 for the least diverse location (TO; Table 1). None of the locations deviated significantly from neutrality under Fu’s FS or Tajima’s D (Table 1). The minimum spanning network produced an overall ‘dumbbell’ shape that is often associated with two isolated populations undergoing recent population expansions (Supplementary Information, Fig. S1; Avise, 2000). Pairwise differentiation was significant between all locations except the three southern sites (CP, TO and GL; Table 2). Tests for IBD were insignificant (R2 = 0.18, P = 0.1797). A priori assignments suggested a break between northern and southern (N–S) groups between GL and HR, based on to the higher AMOVA support of HR in the northern cluster (Table 3). Between-group variation became insignificant if HR was included in the southern cluster and the significant support for three groups, with a break between GL and HR and between FP and RI, was weaker than the N–S grouping (Table 3). The best SAMOVA result was obtained by retaining the southern grouping (CP, TO and GL) while breaking up the northern group of locations (Table 3). Despite setting the analysis to retain geographical heterogeneity, the only two populations to be grouped together in the north (HR and RI) were separated by FP (Table 3). The relatively small spatial scale between sites could explain why the algorithm allowed this clustering. The four groups identified by SAMOVA were better supported than the a priori N–S grouping (Table 3). Table 2. Pairwise comparison (ɸST) for Cominella maculosa sampled along the Wairarapa coast of the North Island using 610-bp fragments of the COI gene   CA  RI  FP  HR  GL  TO  CP  CA    ***  ***  ***  ***  ***  ***  RI  0.23504    ***  *  ***  ***  ***  FP  0.22124  0.33674    ***  ***  ***  ***  HR  0.24023  0.03914  0.22315    ***  ***  ***  GL  0.63132  0.83649  0.42356  0.80073    NS  NS  TO  0.60096  0.8381  0.40042  0.80083  0.00072    NS  CP  0.54459  0.7999  0.34103  0.75275  0.01136  0.01155      CA  RI  FP  HR  GL  TO  CP  CA    ***  ***  ***  ***  ***  ***  RI  0.23504    ***  *  ***  ***  ***  FP  0.22124  0.33674    ***  ***  ***  ***  HR  0.24023  0.03914  0.22315    ***  ***  ***  GL  0.63132  0.83649  0.42356  0.80073    NS  NS  TO  0.60096  0.8381  0.40042  0.80083  0.00072    NS  CP  0.54459  0.7999  0.34103  0.75275  0.01136  0.01155    Site codes are given in Table 1. NS, non-significant. * P<0.05. ** P<0.01. *** P<0.001. View Large Table 3. Analysis of molecular variance (AMOVA) and spatial analysis of molecular variance (SAMOVA) using 610-bp fragments of the COI gene for Cominella maculosa sampled along the Wairarapa coast Number of groups  Group composition  Within groups  Within all populations  Between groups    ɸSC  % Variation  ɸST  % Variation  ɸCT  % Variation  AMOVA in Arlequin 3.5.2.2  2  CP-GL, HR-CA  0.22303***  33.37  0.66633***  9.58  0.57054*  57.05  2  CP-FP, RI+CA  0.45942***  38.57  0.61427***  32.78  0.28645NS  28.65  3  CP-GL, HR+FP, RI+CA  0.19094***  39.9  0.60096***  9.42  0.50679*  50.68  SAMOVA in SAMOVA 2.0  2  CP-GL, HR-CA  0.26393***  23.77  0.76227***  8.52  0.67702*  67.7  3  CP-GL, FP, HR+RI+CA  0.10981***  27.8  0.72202***  3.43  0.68773*  68.77  4  CP-GL, HR+RI, FP, CA  0.02219***  30.34  0.69663***  0.69  0.68974**  68.97  5  CP-GL, HR, FP, RI, CA  −0.02241*  31.73  0.68272***  −0.7  0.68967*  68.97  6  CP, TO+GL, HR, FP, RI, CA  −0.02053NS  33.09  0.66909***  −0.67  0.67574*  67.57  Best SAMOVA grouping analysed using AMOVA in Arlequin 2.5.2.2  SAMOVA  CP-GL, HR+RI, FP, CA  0.00262*  40.54  0.59457***  0.11  0.59351**  59.35  Number of groups  Group composition  Within groups  Within all populations  Between groups    ɸSC  % Variation  ɸST  % Variation  ɸCT  % Variation  AMOVA in Arlequin 3.5.2.2  2  CP-GL, HR-CA  0.22303***  33.37  0.66633***  9.58  0.57054*  57.05  2  CP-FP, RI+CA  0.45942***  38.57  0.61427***  32.78  0.28645NS  28.65  3  CP-GL, HR+FP, RI+CA  0.19094***  39.9  0.60096***  9.42  0.50679*  50.68  SAMOVA in SAMOVA 2.0  2  CP-GL, HR-CA  0.26393***  23.77  0.76227***  8.52  0.67702*  67.7  3  CP-GL, FP, HR+RI+CA  0.10981***  27.8  0.72202***  3.43  0.68773*  68.77  4  CP-GL, HR+RI, FP, CA  0.02219***  30.34  0.69663***  0.69  0.68974**  68.97  5  CP-GL, HR, FP, RI, CA  −0.02241*  31.73  0.68272***  −0.7  0.68967*  68.97  6  CP, TO+GL, HR, FP, RI, CA  −0.02053NS  33.09  0.66909***  −0.67  0.67574*  67.57  Best SAMOVA grouping analysed using AMOVA in Arlequin 2.5.2.2  SAMOVA  CP-GL, HR+RI, FP, CA  0.00262*  40.54  0.59457***  0.11  0.59351**  59.35  Site codes are given in Table 1. Groupings that maximize between-group variation (ɸCT) are in bold type. NS, non-significant. * P<0.05. ** P<0.01. *** P<0.001. View Large DISCUSSION Fleming et al. (2018) found an IBD pattern for C. maculosa around the North Island of New Zealand. Neighbouring sites shared at least one haplotype. However, a distinct genetic break was identified on the Wairarapa coast where no haplotypes were shared between Castlepoint (CA) and Cape Palliser (CP). The northern CA site shared haplotypes with sites up to 300 km north along the East Coast, but did not share a single haplotype with its neighbouring southern site. Here, we sampled at an additional five locations along the 125 km between these two sites, and found a mixed pattern of strong and weak genetic differentiation across these populations. The three southern locations (CP, TO and GL) span 43 km and formed a genetically homogeneous group that was differentiated from each of the northern sites. This grouping was also significant in the AMOVA, which had the highest support for a southern group (CP, TO and GL) and a northern group (HR, FP, RI and CA). While genetic breaks between coastal populations of direct-developing gastropods have been reported in other regions of the Pacific (Kyle & Boulding, 2000; Sánchez et al., 2011; Brante et al., 2012), to the best of our knowledge this is the first study to use large sample sizes and fine-scale sampling to describe the location of the break and structure of populations around it. There were two sites (HR and FP) in the middle of the sampling range that contained high frequencies of both dominant southern and northern haplotypes. These two sites appeared to be either a contact zone between the south and north, or co-founded by source populations from the north and south. Interestingly, the two sites lacked any private haplotypes, which have been found in nearly all other C. maculosa populations (Fleming et al., 2018). The absence of private haplotypes could indicate the two sites have been recently founded and new mutations have not yet had time to accumulate. Alternatively, private haplotypes may be present in the two populations at a very low frequency and a much larger sample size is needed to detect the rare variants. The mix of southern and northern haplotypes at sites HR and FP suggest they could have been colonized from multiple sources, but the source populations cannot be inferred from the available data. If colonization had occurred from the north and/or the south by crawling along the coastline, a pattern of IBD might have been likely. The grouping of non-neighbouring sites HR and RI by the SAMOVA shows a ‘leapfrog’ dispersal pattern (Thiel & Haye, 2006). A similar clustering of non-neighbours has been observed for the direct-developing C. lineolata along the east coast of Australia, for which algal rafting was proposed as the mechanism that enabled gene flow (Hoskin, 1997). While rafting dispersal has been implicated in the dispersal of some direct-developing gastropods in New Zealand (Cumming et al., 2014), these species are commonly associated with large algae. Cominella maculosa whelks are are often found under rocks or buried in gravel rather than on algae or wood, which would greatly reduce their opportunities for rafting on algae. Laboratory studies conducted with C. maculosa hatchlings indicate drifting of juveniles in the water column is unlikely to contribute to a significant amount of long-distance dispersal between populations (Dohner, 2016). However, other mechanisms might transport egg capsules to other sites, for example if females attached egg capsules to drift wood, or if clumps of intact egg capsules became dislodged and drifted in the northbound, nearshore Wairarapa Current (Chiswell, 2000). Because each capsule contains an average of seven hatchlings (Carrasco & Phillips, 2014), it would take only a couple of sporadic rafting or drifting events to transport a reasonable quantity of dispersers northwards. Although rare, these events could be important over evolutionary timescales as only a few migrants per generation are required to homogenize genetic diversity (Ovenden, 2013). Assuming low post-hatching mortality, colonization by a few drifting egg capsules could create a ‘founder takes all’ situation (Waters, Fraser & Hewitt, 2013). Such colonization events could also lead to secondary contact, which might explain the presence of two highly divergent haplotypes coexisting at Castlepoint (Fig. S1). Other studies have found differentiation among populations of direct-developers closer together than those in this southern group. For example, populations of Antarctic topshells (Margarella antarctica) located less than 6 km apart (Hoffman et al., 2013), and populations of mudsnails (Zeacumanths subcarinatus) within 5 km of each other (Keeney et al., 2009) are genetically distinct. Within our own study there was a much higher level of differentiation between northern sites that have a similar amount of spatial separation to the southern group. The lack of differentiation between the three southern sites could indicate strong genetic connectivity among them, due either to demographic continuity or to recolonization after local extinctions (Kyle & Boulding 2000). However, recolonization or dispersal among sites via crawling adults is likely to be very low because rocky reefs along this coast are not contiguous, but are separated by patches of unsuitable habitat due to large sandy beaches and freshwater inputs. Alternatively, the three southern sites might be genetically similar because they were founded from the same source population. Further support for this hypothesis is that these southern Wairarapa locations, and the other southern sites sampled by Fleming et al. (2018), all have low levels of genetic diversity which can occur from low effective population sizes, such as founder events or bottlenecks. Low genetic diversity in direct-developing gastropods at high latitudes has been attributed to recolonization from lower latitudes after a population contraction while restricted to ice age refugia (Kojima et al., 2004; Marko, 2004; Keeney et al., 2009). It is possible that the southern populations in our study are a product of founding from a genetically homogenous Last Glacial Maximum refugium. Two C. maculosa samples (Donald et al., 2015; NCBI reference numbers KP694145 and KP694146) from the Kaikoura Peninsula on the South Island, 300 km away and across the Cook Strait (Fig. 1B), were compared to the current dataset. From the 552 bp that could be aligned, both individuals were identified as the same dominant haplotype found in the three southern locations of our study. Walton (2017) found that 135 of 140 C. maculosa collected from seven South Island sites also matched the dominant southern haplotype reported in the present study. The overall findings from the two studies suggest that the land gap of the Cook Strait has not historically been a strong barrier to dispersal for this species. However, this conclusion is based on data from maternally inherited mitochondrial DNA sequences, which means if only a few females were in the founder population all of the hatchlings would be genetically similar. The populations should be further investigated using bi-parentally inherited nuclear DNA markers that might allow for a more detailed study of the recent barriers to dispersal. Moreover, studies of other direct-developing intertidal species could be used to corroborate our findings, if they also displayed a similar pattern of genetic structure along the Wairarapa coast and across the Cook Strait. Overall, the genetic pattern of C. maculosa at large and small scales suggests that it is made up of discrete populations and there are low levels of gene flow, similar to patterns found for other direct-developing species (Lee & Boulding, 2009; Barbosa et al., 2013; Pálsson et al., 2014). Discrete population genetic structure is generally evidence that populations exist in demographic isolation and are unlikely to form a metapopulation that can quickly recolonize after local population extinction. Conversely, high levels of genetic connectivity do not necessarily reflect demographic connectivity (Ovenden, 2013). The decoupling of genetic and demographic connectivity in ‘crinkled’ populations such as the southern group in this study can be caused by founder events, migration into large populations or the appearance of a recent barrier to gene flow. Further fine-scale studies should be conducted across other genetic breaks to determine how the interaction of life history traits, historical events and environmental factors impacts population structure. ACKNOWLEDGMENTS We would like to thank Glenburn, Glendhu and Pahaoa Stations for the use of their roads to access several field sites. Thanks to Kerry Walton for the donation of Castlepoint whelks. Thanks also to the whelk collection volunteers: Ali Duncan, Jana Wold, Gustav Kessel, Kerry Walton, Juliette Champagnat, Angela Fleming and John van der Sman. Thanks to Angela Fleming for support in the genetics lab and providing sequence data for Castlepoint and Cape Palliser whelks. This work was improved by comments from two anonymous reviewers. Funding was provided by Victoria University of Wellington. SUPPLEMENTARY INFORMATION Additional Supporting Information may be found in the online version of this article at the publisher’s website: Figure S1. Minimum spanning network of COI haplotypes for Cominella maculosa sampled along the Wairarapa coast of the North Island of New Zealand. The size of each circle reflects the total number of samples detected for each haplotype, with colours reflecting how many individuals from each site contribute to the sample size. Hash marks represent one mutational change between haplotypes. Black dots represent theoretical haplotypes. REFERENCES Avise JC. 2000. Phylogeography: The history and formation of species . Cambridge: Harvard University Press. Bandelt HJ, Forster P, Röhl A. 1999. Median-joining networks for inferring intraspecific phylogenies. Molecular Biology and Evolution  16: 37– 48. Google Scholar CrossRef Search ADS PubMed  Barbosa SS, Klanten SO, Puritz JB, Toonen RJ, Byrne M. 2013. Very fine-scale population genetic structure of sympatric asterinid sea stars with benthic and pelagic larvae: influence of mating system and dispersal potential. Biological Journal of the Linnean Society  108: 821– 833. Google Scholar CrossRef Search ADS   Brante A, Fernández M, Viard F. 2012. Phylogeography and biogeography concordance in the marine gastropod Crepipatella dilatata (Calyptraeidae) along the southeastern Pacific coast. Journal of Heredity  103: 630– 637. Google Scholar CrossRef Search ADS PubMed  Burton RS, Feldman MW. 1982. Population genetics of coastal and estuarine invertebrates: does larval behavior influence population structure? In: Kennedy VS, ed. Estuarine comparisons . New York: Academic Press, 537– 551. Google Scholar CrossRef Search ADS   Carrasco SA, Phillips NE. 2014. Encapsulation and development of three New Zealand neogastropods with contrasting embryo packaging and maternal provisioning. New Zealand Journal of Zoology  41: 171– 186. Google Scholar CrossRef Search ADS   Chiswell SM. 2000. The Wairarapa Coastal Current. New Zealand Journal of Marine and Freshwater Research  34: 303– 315. Google Scholar CrossRef Search ADS   Collin R. 2001. The effects of mode of development on phylogeography and population structure of North Atlantic Crepidula (Gastropoda: Calyptraeidae). Molecular Ecology  10: 2249– 2262. Google Scholar CrossRef Search ADS PubMed  Cumming RA, Nikula R, Spencer HG, Waters JM. 2014. Transoceanic genetic similarities of kelp-associated sea slug populations: long-distance dispersal via rafting? Journal of Biogeography  41: 2357– 2370. Google Scholar CrossRef Search ADS   Dawson MN, Hays CG, Grosberg RK, Raimondi PT. 2014. Dispersal potential and population genetic structure in the marine intertidal of the eastern North Pacific. Ecological Monographs  84: 435– 456. Google Scholar CrossRef Search ADS   de Lange W, Bell RG, Gorman R, Reid S. 2003. Physical oceanography of New Zealand waters. In: Goff J, Nichol S, Rouse H, eds. The New Zealand coast: Te Tai o Aotearoa . Palmerston North: Dunmore Press, 59– 78. Dohner M. 2016. Crawling to connectivity? The direct-developing journey of the spotted whelk (Cominella maculosa) . Unpublished Master’s of Science Thesis, Victoria University of Wellington. Donald KM, Winter DJ, Ashcroft AL, Spencer HG. 2015. Phylogeography of the whelk genus Cominella (Gastropoda: Buccinidae) suggests long-distance counter-current dispersal of a direct developer. Biological Journal of the Linnean Society  115: 315– 332. Google Scholar CrossRef Search ADS   Dupanloup I, Schneider S, Excoffier L. 2002. A simulated annealing approach to define the genetic structure of populations. Molecular Ecology  11: 2571– 2581. Google Scholar CrossRef Search ADS PubMed  Excoffier L, Lischer HE. 2010. Arlequin suite ver 3.5: a new series of programs to perform population genetics analyses under Linux and Windows. Molecular Ecology Resources  10: 564– 567. Google Scholar CrossRef Search ADS PubMed  Excoffier L, Lischer H. 2015. Arlequin Ver 3.5: an integrated software package for population genetics data analysis . Berne: Swiss Institute of Bioinformatics. Excoffier L, Smouse PE, Quattro JM. 1992. Analysis of molecular variance inferred from metric distances among DNA haplotypes: application to human mitochondrial DNA restriction data. Genetics  131: 479– 491. Google Scholar PubMed  Fleming AM, Dohner MM, Phillips NE, Ritchie PA. 2018. Genetic connectivity among populations of two congeneric direct-developing whelks varies across spatial scales. New Zealand Journal of Marine and Freshwater Research  52: 100– 117. Google Scholar CrossRef Search ADS   Folmer O, Black M, Hoeh W, Lutz R, Vrijenhoek R. 1994. DNA primers for amplification of mitochondrial cytochrome c oxidase subunit I from diverse metazoan invertebrates. Molecular Marine Biology and Biotechnology  3: 294– 299. Google Scholar PubMed  Fu YX. 1997. Statistical tests of neutrality of mutations against population growth, hitchhiking and background selection. Genetics  147: 915– 925. Google Scholar PubMed  Giles EC, Saenz-Agudelo P, Hussey NE, Ravasi T, Berumen ML. 2015. Exploring seascape genetics and kinship in the reef sponge Stylissa carteri in the Red Sea. Ecology and Evolution  5: 2487– 2502. Google Scholar CrossRef Search ADS PubMed  Heath RA. 1982. A review of the physical oceanography of the seas around New Zealand. New Zealand Journal of Marine and Freshwater Research  19: 79– 124. Google Scholar CrossRef Search ADS   Hoffman JI, Clarke A, Clark MS, Peck LS. 2013. Hierarchical population genetic structure in a direct developing antarctic marine invertebrate. PLoS ONE  8: e63954. Google Scholar CrossRef Search ADS PubMed  Hoskin MG. 1997. Effects of contrasting modes of larval development on the genetic structures of populations of three species of prosobranch gastropods. Marine Biology  127: 647– 656. Google Scholar CrossRef Search ADS   Karl SA, Hayes KA. 2012. Extreme population subdivision in the crown conch (Melongena corona): historical and contemporary influences. Journal of Heredity  103: 523– 532. Google Scholar CrossRef Search ADS PubMed  Kearse M, Moir R, Wilson A, Stones-Havas S, Cheung M, Sturrock S, Buxton S, Cooper A, Markowitz S, Duran C, Thierer T, Ashton B, Meintjes P, Drummond A. 2012. Geneious Basic: an integrated and extendable desktop software platform for the organization and analysis of sequence data. Bioinformatics (Oxford, England)  28: 1647– 1649. Google Scholar CrossRef Search ADS PubMed  Keeney DB, King TM, Rowe DL, Poulin R. 2009. Contrasting mtDNA diversity and population structure in a direct-developing marine gastropod and its trematode parasites. Molecular Ecology  18: 4591– 4603. Google Scholar CrossRef Search ADS PubMed  Kyle CJ, Boulding E. G. 2000. Comparative population genetic structure of marine gastropods (Littorina spp.) with and without pelagic larval dispersal. Marine Biology  137: 835– 845. Google Scholar CrossRef Search ADS   Kojima S, Hayashi I, Kim D, Iijima A, Furota T. 2004. Phylogeography of an intertidal direct-developing gastropod Batillaria cumingi around the Japanese Islands. Marine Ecology Progress Series  276: 161– 172. Google Scholar CrossRef Search ADS   Laing A, Chiswell SM. 2003. The ocean medium. In: Andrew NL, Francis MP, eds. The living reef: the ecology of New Zealand’s rocky reefs . Nelson: Craig Potton Publishing, 24– 31. Lee HJ, Boulding EG. 2009. Spatial and temporal population genetic structure of four northeastern Pacific littorinid gastropods: the effect of mode of larval development on variation at one mitochondrial and two nuclear DNA markers. Molecular Ecology  18: 2165– 2184. Google Scholar CrossRef Search ADS PubMed  Librado P, Rozas J. 2009. DnaSP v5: a software for comprehensive analysis of DNA polymorphism data. Bioinformatics (Oxford, England)  25: 1451– 1452. Google Scholar CrossRef Search ADS PubMed  Marko PB. 2004. ‘What’s larvae got to do with it?’ Disparate patterns of post-glacial population structure in two benthic marine gastropods with identical dispersal potential. Molecular Ecology  13: 597– 611. Google Scholar CrossRef Search ADS PubMed  Morton JE, Miller MC. 1968. The New Zealand sea shore . Liverpool: Tinling and Co. Ovenden JR. 2013. Crinkles in connectivity: combining genetics and other types of biological data to estimate movement and interbreeding between populations. Marine and Freshwater Research  64: 201– 207. Google Scholar CrossRef Search ADS   Ovenden JR, Berry O, Welch DJ, Buckworth RC, Dichmont CM. 2015. Ocean’s eleven: a critical evaluation of the role of population, evolutionary and molecular genetics in the management of wild fisheries. Fish and Fisheries  16: 125– 159. Google Scholar CrossRef Search ADS   Pálsson S, Magnúsdóttir H, Reynisdóttir S, Jónsson ZO, Örnólfsdóttir EB. 2014. Divergence and molecular variation in common whelk Buccinum undatum (Gastropoda: Buccinidae) in Iceland: a trans-Atlantic comparison. Biological Journal of the Linnean Society  111: 145– 159. Google Scholar CrossRef Search ADS   Pelc RA, Warner RR, Gaines SD. 2009. Geographical patterns of genetic structure in marine species with contrasting life histories. Journal of Biogeography  36: 1881– 1890. Google Scholar CrossRef Search ADS   Powell AWB. 1979. New Zealand Mollusca: marine, land and freshwater shells . Auckland: Collins. Ross PM, Hogg ID, Pilditch CA, Lundquist CJ. 2009. Phylogeography of New Zealand’s coastal benthos. New Zealand Journal of Marine and Freshwater Research  43: 1009– 1027. Google Scholar CrossRef Search ADS   Ross PM, Hogg ID, Pilditch CA, Lundquest CJ, Wilkins RJ. 2012. Population genetic structure of the New Zealand estuarine clam Austrovenus stutchburyi (Bivalvia: Veneridae) reveals population subdivision and partial congruence with biogeographic boundaries. Estuaries and Coasts  35: 143– 154. Google Scholar CrossRef Search ADS   Rousset F. 1997. Genetic differentiation and estimation of gene flow from F-statistics under isolation by distance. Genetics  145: 1219– 1228. Google Scholar PubMed  Sambrook J, Fritsch EF, Maniatis T. 1989. Molecular cloning: a laboratory manual . New York: Cold Spring Harbor Press. Sánchez R, Sepúlveda RD, Brante A, Cárdenas L. 2011. Spatial pattern of genetic and morphological diversity in the direct developer Acanthina monodon (Gastropoda: Mollusca). Marine Ecology Progress Series  434: 121– 131. Google Scholar CrossRef Search ADS   Selkoe KA, Toonen RJ. 2011. Marine connectivity: a new look at pelagic larval duration and genetic metrics of dispersal. Marine Ecology Progress Series  436: 291– 305. Google Scholar CrossRef Search ADS   Scheltema RS. 1971. Larval dispersal as a means of genetic exchange between geographically separated populations of shallow-water benthic marine gastropods. Biological Bulletin  140: 284– 322. Google Scholar CrossRef Search ADS   Scheltema RS. 1986. On dispersal and planktonic larvae of benthic invertebrates: an eclectic overview and summary of problems. Bulletin of Marine Science  39: 290– 322. Tajima F. 1989. The effect of change in population size on DNA polymorphism. Genetics  123: 597– 601. Google Scholar PubMed  Thiel M, Haye PA. 2006. The ecology of rafting in the marine environment. III. Biogeographical and evolutionary consequences. Oceanography and Marine Biology  44: 323– 429. Google Scholar CrossRef Search ADS   Veale AJ, Lavery SD. 2011. Phylogeography of the snakeskin chiton Sypharochiton pelliserpentis (Mollusca: Polyplacophora) around New Zealand: are seasonal near-shore upwelling events a dynamic barrier to gene flow? Biological Journal of the Linnean Society  104: 552– 563. Google Scholar CrossRef Search ADS   Waters JM, Fraser CI, Hewitt GM. 2013. Founder takes all: density-dependent processes structure biodiversity. Trends in Ecology & Evolution  28: 78– 85. Google Scholar CrossRef Search ADS PubMed  Walton K. 2017. Phylogeography of the New Zealand whelks Cominella maculosa and C. virgata . Unpublished Master’s of Science Thesis, Victoria University of Wellington. © 2018 The Linnean Society of London, Biological Journal of the Linnean Society http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Biological Journal of the Linnean Society Oxford University Press

Fine-scale genetic structure across a New Zealand disjunction for the direct-developing intertidal whelk Cominella maculosa (Gastropoda: Buccinidae)

Loading next page...
 
/lp/ou_press/fine-scale-genetic-structure-across-a-new-zealand-disjunction-for-the-tXmB7jnRX9
Publisher
The Linnean Society of London
Copyright
© 2018 The Linnean Society of London, Biological Journal of the Linnean Society
ISSN
0024-4066
eISSN
1095-8312
D.O.I.
10.1093/biolinnean/blx167
Publisher site
See Article on Publisher Site

Abstract

Abstract The spotted whelk, Cominella maculosa, is a common intertidal species endemic to New Zealand with crawl-away juveniles that hatch from benthic egg capsules. The aim of this study was to determine the fine-scale pattern of genetic differentiation and over what distances this direct-developing species can form isolated populations, across a region that includes a known genetic disjunction. Whelks were collected from seven locations, 10–26 km apart, along 125 km of a linear coastline, and DNA sequences from the mitochondrial cytochrome c oxidase subunit I (COI) gene were compared. Four northern sites were genetically differentiated from each other and from the group of three southern sites. The three southern sites showed no genetic differentiation, consistently clustered together in a single group, shared a dominant haplotype and were characterized by a low number of private haplotypes and low haplotype and nucleotide diversity. The two sites located ~10 km north of the southern group lacked private haplotypes and contained high frequencies of both dominant northern and southern haplotypes. The lack of an isolation-by-distance pattern of structure, and a leapfrog-like dispersal grouping of non-neighbouring sites, suggests sporadic long-distance dispersal events occur in this species, even if rarely. INTRODUCTION Dispersal ability is a key factor that mediates the level of population connectivity (Selkoe & Toonen, 2011). In the marine environment, species with planktonic larvae that are in the water column for days to weeks, or longer, have a high dispersal potential. Gene flow is therefore expected to be high and population genetic differentiation low (Scheltema, 1971, 1986; Burton & Feldman, 1982; Collin, 2001), although at relatively large geographical scales some studies have shown only a weak relationship between population structure and the duration of the planktonic larval stage. Species that lack planktonic larvae and hatch from brooded or benthic eggs with direct development of juveniles typically have lower dispersal potential. These species are more likely to form genetically differentiated populations (Lee & Boulding, 2009; Karl & Hayes, 2012; Dawson et al., 2014), which could be either discrete groups or a collection of sites that form a pattern of isolation by distance (IBD). However, the relationship between geographical and genetic distance is not always linear because the distribution of suitable habitat can be patchy, and environments are often heterogeneous and variable. IBD can be significant at large scales but may not appear at smaller scales (Hoffman et al., 2013; Giles et al., 2015). Parts of a distribution can contain populations that have an IBD pattern, and yet contain surprisingly abrupt genetic disjunctions between neighbouring populations. These types of genetic breaks have been reported for intertidal snails on the west coast of North America (Kyle & Boulding, 2000) and in two species at the same latitude along the Chilean coastline (Sánchez et al., 2011; Brante, Fernández & Viard, 2012). Observations of genetic disjunctions, particularly in the absence of obvious geographical barriers, make it difficult to establish the general causes of isolation among populations. There is a need for more studies that attempt to document the fine-scale genetic structure of direct-developer species, which will enable a comparative approach to identify common patterns and processes underlying the genetic structuring of coastal marine species with lower dispersal potential (Pelc, Warner & Gaines, 2009). New Zealand is an archipelago that includes two main islands (North and South) with long, mostly linear, coastlines that span the temperate range from the subtropics to the subantarctic (Fig. 1A, B). Major current systems from the west split as they hit the continental shelf and flow around the North and South Islands, forming a complex system of coastal and offshore currents and eddies (Heath, 1982; de Lange et al., 2003; Laing & Chiswell, 2003). Previous studies have shown that planktonic dispersers and direct-developers exhibit a general pattern of clustering into northern and southern populations (reviewed by Ross et al., 2009), with the Cook Strait region dividing them (Fig. 1B). However, most studies have used broad sets of sample sites to cover as much area as possible. Thus, there is a lack of fine scale genetic sampling that can be used to describe the patterns in and around genetic breaks. Figure 1. View largeDownload slide A. The position of New Zealand, within the dotted region, at the global scale. B. New Zealand, with the two mainland islands (North and South Island) divided by Cook Strait. The dotted region defines the southern tip of the North Island. C. Southern tip of the North Island, including Wellington Harbour to the west and the Wairarapa coast to the east. Sample sites are marked on the map; see Table 1 for site codes and scale bar for distance. Pie charts show the distribution of COI haplotypes for Cominella maculosa at sites sampled along the Wairarapa coast; circle size represents sample size. White colour represents private haplotypes; all other haplotypes are shared between two or more sites. Figure 1. View largeDownload slide A. The position of New Zealand, within the dotted region, at the global scale. B. New Zealand, with the two mainland islands (North and South Island) divided by Cook Strait. The dotted region defines the southern tip of the North Island. C. Southern tip of the North Island, including Wellington Harbour to the west and the Wairarapa coast to the east. Sample sites are marked on the map; see Table 1 for site codes and scale bar for distance. Pie charts show the distribution of COI haplotypes for Cominella maculosa at sites sampled along the Wairarapa coast; circle size represents sample size. White colour represents private haplotypes; all other haplotypes are shared between two or more sites. Cominella maculosa Martyn, 1784, the spotted whelk, is a direct-developing buccinid whelk endemic to New Zealand. It is a common carnivorous scavenger found in the rocky intertidal zone throughout the North Island, and the northern part of the South Island (Morton & Miller, 1968; Powell, 1979). Females communally lay masses of egg capsules on hard substrata over the summer, with an average of seven hatchlings emerging from each capsule 9 weeks later (Carrasco & Phillips, 2014). A recent phylogeographical study of C. maculosa by Fleming et al. (2018) found discrete population structure with an IBD pattern throughout the North Island distribution; however, the study also showed three genetic disjunctions that each occurred over a relatively small geographical scale (< 130 km). One of these disjunctions occurs along the Wairarapa coast of the North Island between two sites, Cape Palliser and Castlepoint, ~125 km apart, along a linear stretch of coastline where rocky outcrops are interspersed with sandy beaches (Fig. 1C). This section of the east coast of the North Island has not been well sampled in other studies of population genetic structure, especially for direct-developing species. Nevertheless, there is limited evidence that there is no disjunction along this coast for species with planktonic larvae (snakeskin chiton: Veale & Lavery, 2011; cockle: Ross et al., 2012). The aim of this study was to use fine-scale sampling (i.e. from sites tens of kilometres apart) to investigate the genetic disjunction along the Wairarapa coast of New Zealand (Fig. 1C) for the direct-developing whelk C. maculosa. METHODS Sampling and DNA extraction Cominella maculosa individuals were collected in 2015 from rocky shore tide pools that had been baited with dead mussels at low tide. Whelks > 10 mm were collected over a 2-h period and either placed directly in 80% ethanol or transported in seawater and frozen before processing. A total of 250 specimens were collected from five locations between Castlepoint (CA) and Cape Palliser (CP). From north to south, these sites included Riversdale (RI), Flat Point (FP), Honeycomb Rock (HR), Glendhu (GL), and Tora (TO) (Fig. 1C, Table 1). Distances between neighbouring sites varied from ~10 to 26 km. Foot tissue was dissected and preserved in 95% ethanol. Table 1. Sample information, genetic diversity indices and neutrality tests for Cominella maculosa sampled along the Wairarapa coast of the North Island of New Zealand Sample information  Diversity indices  Neutrality/tests  Site code  Sample site  Latitude/l/ongitude  N  S  H  pH  h (SD)  π (SD)  k (SD)  Fu’s FS (P)  Tajima D (P)  CA  Castlepoint  −40.865833/176.234500  47  9  4  1  0.68 (0.04)  0.005 (0.003)  3.19 (1.68)  5.65 (0.97)  1.60 (0.95)  RI  Riversdale  −41.076663/176.086164  56  3  4  2  0.23 (0.07)  0.000 (0.001)  0.24 (0.28)  −2.47 (0.03)  −1.27 (0.08)  FP  Flat Point  −41.246383/175.958717  53  9  3  0  0.53 (0.02)  0.005 (0.003)  3.17 (1.67)  7.90 (0.99)  1.66 (0.95)  HR  Honeycomb Rock  −41.346634/175.825988  47  6  2  0  0.26 (0.07)  0.003 (0.001)  1.55 (0.95)  5.67 (0.98)  0.37 (0.68)  GL  Glendhu  −41.400731/175.716665  59  1  2  1  0.10 (0.05)  0.000 (0.000)  0.10 (0.17)  −0.57 (0.14)  −0.70 (0.22)  TO  Tora  −41.504460/175.528460  35  1  2  1  0.06 (0.05)  0.000 (0.000)  0.06 (0.13)  −1.34 (0.05)  −1.14 (0.13)  CP  Cape Palliser  −41.600000/175.283333  27  1  2  1  0.14 (0.09)  0.000 (0.000)  0.14 (0.21)  −0.35 (0.17)  −0.73 (0.21)  All      324  16  10  6  0.61 (0.02)  0.006 (0.003)  3.40 (1.74)  3.75 (0.87)  0.86 (0.84)  Sample information  Diversity indices  Neutrality/tests  Site code  Sample site  Latitude/l/ongitude  N  S  H  pH  h (SD)  π (SD)  k (SD)  Fu’s FS (P)  Tajima D (P)  CA  Castlepoint  −40.865833/176.234500  47  9  4  1  0.68 (0.04)  0.005 (0.003)  3.19 (1.68)  5.65 (0.97)  1.60 (0.95)  RI  Riversdale  −41.076663/176.086164  56  3  4  2  0.23 (0.07)  0.000 (0.001)  0.24 (0.28)  −2.47 (0.03)  −1.27 (0.08)  FP  Flat Point  −41.246383/175.958717  53  9  3  0  0.53 (0.02)  0.005 (0.003)  3.17 (1.67)  7.90 (0.99)  1.66 (0.95)  HR  Honeycomb Rock  −41.346634/175.825988  47  6  2  0  0.26 (0.07)  0.003 (0.001)  1.55 (0.95)  5.67 (0.98)  0.37 (0.68)  GL  Glendhu  −41.400731/175.716665  59  1  2  1  0.10 (0.05)  0.000 (0.000)  0.10 (0.17)  −0.57 (0.14)  −0.70 (0.22)  TO  Tora  −41.504460/175.528460  35  1  2  1  0.06 (0.05)  0.000 (0.000)  0.06 (0.13)  −1.34 (0.05)  −1.14 (0.13)  CP  Cape Palliser  −41.600000/175.283333  27  1  2  1  0.14 (0.09)  0.000 (0.000)  0.14 (0.21)  −0.35 (0.17)  −0.73 (0.21)  All      324  16  10  6  0.61 (0.02)  0.006 (0.003)  3.40 (1.74)  3.75 (0.87)  0.86 (0.84)  Analyses are based on 610-bp sequences from the COI gene. Latitude and longitude are in decimal degrees. N, sample size; S, segregating sites; H, number of haplotypes; pH, number of private haplotypes; h, haplotype diversity; π, nucleotide diversity; k, nucleotide differences. View Large A small amount of foot tissue was digested for 2 h at 60 °C in 600 μL extraction buffer (12 mm Tris-HCl pH 8.0, 47.5 mm NaCl, 9.5 mm EDTA, 0.19% SDS, 0.5 mg proteinase-K). DNA was isolated using a modified phenol–chloroform method (Sambrook, Fritsch & Maniatis, 1989). The digested samples were twice mixed with 600 µL of phenol/chloroform/isoamyl alcohol (25:24:1), followed by the addition of chloroform/isoamyl alcohol (24:1). DNA was precipitated using 2.5 volumes of ethanol and 3 m sodium acetate and centrifuged for 30 min at 4 °C. Pelleted DNA was washed using 70% ethanol and then dried in a vacuum centrifuge (Eppendorf Concentrator 5301) for 5–10 min. Purified DNA was re-suspended in 30–70 µL of TE buffer (10 mm Tris-HCl pH 8.0, 1 mm EDTA) and its concentration and quality was determined using a Nanodrop spectrophotometer (Thermo Fisher Scientific). PCR sequencing and alignment A portion of the mitochondrial cytochrome c oxidase subunit I (COI) gene was amplified by PCR using primers LCO1490 and HCO2198 (Folmer et al., 1994). The PCR was conducted in 25-µL volumes that contained 67 mm Tris-HCl pH 8.8, 16 mm (NH4)2SO4, 3.0 mm MgCl2, 0.6 mg/mL bovine serum albumin, 0.05 units Taq DNA polymerase (Bioline), 0.4 mm of each dNTP, 0.1 μm of each primer and 120 ng/μL of template DNA. Thermocycling was carried out on a Biometra TProfessional Thermocycler under the following conditions: 5 min at 95 °C, 40 cycles of 35 s at 95 °C, 35 s at 50 °C and 45 s at 72 °C, followed by a final 10-min extension at 74 °C. PCR products were prepared using ExoSAP-It (Amersham Parmacia Biotech) and the DNA sequences were determined using a 3730xl DNA Analyzer (Applied Biosystems) at the Macrogen sequencing service (Macrogen Inc.). The 710-bp DNA sequences were quality checked and edited by eye before being trimmed to 610 bp using Geneious 8.1.6 (Kearse et al., 2012). The new DNA sequences reported here have been deposited in GenBank under accession numbers MG583718–MG583721. From Fleming et al. (2018), 28 sequences from Castlepoint and 27 sequences from Cape Palliser were added to the dataset. Another 19 Castlepoint sequences were extracted from donated specimens and added to the dataset. All 324 sequences were aligned using the Geneious alignment algorithm under default settings. Statistical analyses Haplotypes were identified using DnaSP 5.10.01 (Librado & Rozas, 2009). Genetic diversity was determined using Arlequin 3.5.2.2 (Excoffier & Lischer, 2010). Measures of diversity included the number of segregating sites (S), transitions, transversions, nucleotide differences (k), number of haplotypes (H), haplotype diversity (h), nucleotide diversity (π) and nucleic acid composition. Historical demographic changes were tested in Arlequin 3.5.2.2 using Tajima’s D statistic (Tajima, 1989) and Fu’s FS statistic (Fu, 1997). Due to the nature of the Fu’s FS statistic, only P-values below 0.02 are significant (Fu, 1997; Excoffier & Lischer, 2015). Haplotype relationships were visualized with a minimum spanning haplotype network (Bandelt, Forster & Röhl, 1999) created in PopART (http://popart.otago.ac.nz, accessed 23 August 2017). Differentiation between populations was estimated in Arlequin 3.5.2.2 (Excoffier & Lischer, 2010) using haplotype frequency to calculate the fixation index, ɸST, and 10000 permutations to test for significance. To test for a pattern of IBD, a Mantel test was performed under the one-dimensional model of habitat (Rousset, 1997) using Slatkin’s linearized genetic differentiation and untransformed distance. Mantel tests were conducted in Arlequin 3.5.2.2 with 10000 permutations (Excoffier & Lischer, 2010). Distances between populations were determined as the shortest distance through water using Google Earth v7.1.5 (Google Inc., 2015). Due to the linear nature of the Wairarapa coast, the shortest distance through water is also nearly equal to the coastal distance. Analysis of molecular variance (AMOVA) was used to partition total molecular variance into three hierarchical components, each with an associated fixation index (Excoffier, Smouse & Quattro, 1992). These indices describe variation within all populations (ɸST), between populations within a predefined group (ɸSC) and between all predefined groups (ɸCT) (Excoffier et al., 1992). The best group structure is determined by maximizing the between-group variation (i.e. largest significant ɸCT). Three a priori groupings were tested with AMOVA in Arlequin 5.2.2.2 (Excoffier & Lischer, 2010) using 10000 permutations of haplotype frequency data. The first two groupings tested for a break between northern and southern locations. The first scenario included HR in the northern group while the second scenario included HR in the southern group. The third grouping tested the hypothesis that two or more neighbouring populations formed distinct groups. This third scenario combined CA with RI, FP with HR, and the three southern sites together (GL, TO, CP). While AMOVA is useful for testing hypotheses of population structure, it requires a priori assignment of groups. A priori bias can be removed by conducting spatial analysis of variance (SAMOVA), which clusters populations into a predefined number of groups with the aim of maintaining geographical homogeneity while maximizing between-group variation (Dupanloup, Schneider & Excoffier, 2002). The optimum number of groups is determined by identifying the largest significant ɸCT from iterative analysis using two to N − 1 groups, where N is the total number of populations. Iterative runs of SAMOVA were performed for two to six groups with settings to retain geographical homogeneity using SAMOVA 2.0 (Dupanloup et al., 2002). The best SAMOVA grouping was identified as the configuration producing the largest significant ɸCT statistic. To accurately compare SAMOVA groupings to a priori groups, the best SAMOVA grouping was analysed by AMOVA in Arlequin 3.5.2.2 (Excoffier & Lischer, 2010). RESULTS Sequences from a total of 324 whelks [55 sequences from Fleming et al. (2018) and 269 new for this study] were analysed from seven locations along 125 km of Wairarapa coastline (Fig. 1C; Table 1). Ten haplotypes were identified with 16 segregating sites, 13 of which were transition-type changes. Most of the segregating sites were synonymous substitutions, but one site (T to C; bp 270) created a non-synonymous change (valine to alanine) in the haplotype shared by CA and RI. Five of the seven locations contained at least one private haplotype, but the central locations of HR and FP exhibited no private haplotypes (Table 1). Overall haplotype diversity was 0.61. The highest levels of diversity (0.68 and 0.53) were found in two northerly locations (CA and FP, respectively). All other locations had much lower haplotype diversity, ranging from 0.06 to 0.26 (Table 1). Nucleotide diversity within locations showed the same pattern, with overall diversity being similar to the highest location-level values (CA and FP; Table 1). Even these ‘high’ values are quite low (0.005), but values for the other locations were even lower (0.0001 for TO). The average number of nucleotide differences ranged from 3.19 for the most diverse location (CA) to 0.06 for the least diverse location (TO; Table 1). None of the locations deviated significantly from neutrality under Fu’s FS or Tajima’s D (Table 1). The minimum spanning network produced an overall ‘dumbbell’ shape that is often associated with two isolated populations undergoing recent population expansions (Supplementary Information, Fig. S1; Avise, 2000). Pairwise differentiation was significant between all locations except the three southern sites (CP, TO and GL; Table 2). Tests for IBD were insignificant (R2 = 0.18, P = 0.1797). A priori assignments suggested a break between northern and southern (N–S) groups between GL and HR, based on to the higher AMOVA support of HR in the northern cluster (Table 3). Between-group variation became insignificant if HR was included in the southern cluster and the significant support for three groups, with a break between GL and HR and between FP and RI, was weaker than the N–S grouping (Table 3). The best SAMOVA result was obtained by retaining the southern grouping (CP, TO and GL) while breaking up the northern group of locations (Table 3). Despite setting the analysis to retain geographical heterogeneity, the only two populations to be grouped together in the north (HR and RI) were separated by FP (Table 3). The relatively small spatial scale between sites could explain why the algorithm allowed this clustering. The four groups identified by SAMOVA were better supported than the a priori N–S grouping (Table 3). Table 2. Pairwise comparison (ɸST) for Cominella maculosa sampled along the Wairarapa coast of the North Island using 610-bp fragments of the COI gene   CA  RI  FP  HR  GL  TO  CP  CA    ***  ***  ***  ***  ***  ***  RI  0.23504    ***  *  ***  ***  ***  FP  0.22124  0.33674    ***  ***  ***  ***  HR  0.24023  0.03914  0.22315    ***  ***  ***  GL  0.63132  0.83649  0.42356  0.80073    NS  NS  TO  0.60096  0.8381  0.40042  0.80083  0.00072    NS  CP  0.54459  0.7999  0.34103  0.75275  0.01136  0.01155      CA  RI  FP  HR  GL  TO  CP  CA    ***  ***  ***  ***  ***  ***  RI  0.23504    ***  *  ***  ***  ***  FP  0.22124  0.33674    ***  ***  ***  ***  HR  0.24023  0.03914  0.22315    ***  ***  ***  GL  0.63132  0.83649  0.42356  0.80073    NS  NS  TO  0.60096  0.8381  0.40042  0.80083  0.00072    NS  CP  0.54459  0.7999  0.34103  0.75275  0.01136  0.01155    Site codes are given in Table 1. NS, non-significant. * P<0.05. ** P<0.01. *** P<0.001. View Large Table 3. Analysis of molecular variance (AMOVA) and spatial analysis of molecular variance (SAMOVA) using 610-bp fragments of the COI gene for Cominella maculosa sampled along the Wairarapa coast Number of groups  Group composition  Within groups  Within all populations  Between groups    ɸSC  % Variation  ɸST  % Variation  ɸCT  % Variation  AMOVA in Arlequin 3.5.2.2  2  CP-GL, HR-CA  0.22303***  33.37  0.66633***  9.58  0.57054*  57.05  2  CP-FP, RI+CA  0.45942***  38.57  0.61427***  32.78  0.28645NS  28.65  3  CP-GL, HR+FP, RI+CA  0.19094***  39.9  0.60096***  9.42  0.50679*  50.68  SAMOVA in SAMOVA 2.0  2  CP-GL, HR-CA  0.26393***  23.77  0.76227***  8.52  0.67702*  67.7  3  CP-GL, FP, HR+RI+CA  0.10981***  27.8  0.72202***  3.43  0.68773*  68.77  4  CP-GL, HR+RI, FP, CA  0.02219***  30.34  0.69663***  0.69  0.68974**  68.97  5  CP-GL, HR, FP, RI, CA  −0.02241*  31.73  0.68272***  −0.7  0.68967*  68.97  6  CP, TO+GL, HR, FP, RI, CA  −0.02053NS  33.09  0.66909***  −0.67  0.67574*  67.57  Best SAMOVA grouping analysed using AMOVA in Arlequin 2.5.2.2  SAMOVA  CP-GL, HR+RI, FP, CA  0.00262*  40.54  0.59457***  0.11  0.59351**  59.35  Number of groups  Group composition  Within groups  Within all populations  Between groups    ɸSC  % Variation  ɸST  % Variation  ɸCT  % Variation  AMOVA in Arlequin 3.5.2.2  2  CP-GL, HR-CA  0.22303***  33.37  0.66633***  9.58  0.57054*  57.05  2  CP-FP, RI+CA  0.45942***  38.57  0.61427***  32.78  0.28645NS  28.65  3  CP-GL, HR+FP, RI+CA  0.19094***  39.9  0.60096***  9.42  0.50679*  50.68  SAMOVA in SAMOVA 2.0  2  CP-GL, HR-CA  0.26393***  23.77  0.76227***  8.52  0.67702*  67.7  3  CP-GL, FP, HR+RI+CA  0.10981***  27.8  0.72202***  3.43  0.68773*  68.77  4  CP-GL, HR+RI, FP, CA  0.02219***  30.34  0.69663***  0.69  0.68974**  68.97  5  CP-GL, HR, FP, RI, CA  −0.02241*  31.73  0.68272***  −0.7  0.68967*  68.97  6  CP, TO+GL, HR, FP, RI, CA  −0.02053NS  33.09  0.66909***  −0.67  0.67574*  67.57  Best SAMOVA grouping analysed using AMOVA in Arlequin 2.5.2.2  SAMOVA  CP-GL, HR+RI, FP, CA  0.00262*  40.54  0.59457***  0.11  0.59351**  59.35  Site codes are given in Table 1. Groupings that maximize between-group variation (ɸCT) are in bold type. NS, non-significant. * P<0.05. ** P<0.01. *** P<0.001. View Large DISCUSSION Fleming et al. (2018) found an IBD pattern for C. maculosa around the North Island of New Zealand. Neighbouring sites shared at least one haplotype. However, a distinct genetic break was identified on the Wairarapa coast where no haplotypes were shared between Castlepoint (CA) and Cape Palliser (CP). The northern CA site shared haplotypes with sites up to 300 km north along the East Coast, but did not share a single haplotype with its neighbouring southern site. Here, we sampled at an additional five locations along the 125 km between these two sites, and found a mixed pattern of strong and weak genetic differentiation across these populations. The three southern locations (CP, TO and GL) span 43 km and formed a genetically homogeneous group that was differentiated from each of the northern sites. This grouping was also significant in the AMOVA, which had the highest support for a southern group (CP, TO and GL) and a northern group (HR, FP, RI and CA). While genetic breaks between coastal populations of direct-developing gastropods have been reported in other regions of the Pacific (Kyle & Boulding, 2000; Sánchez et al., 2011; Brante et al., 2012), to the best of our knowledge this is the first study to use large sample sizes and fine-scale sampling to describe the location of the break and structure of populations around it. There were two sites (HR and FP) in the middle of the sampling range that contained high frequencies of both dominant southern and northern haplotypes. These two sites appeared to be either a contact zone between the south and north, or co-founded by source populations from the north and south. Interestingly, the two sites lacked any private haplotypes, which have been found in nearly all other C. maculosa populations (Fleming et al., 2018). The absence of private haplotypes could indicate the two sites have been recently founded and new mutations have not yet had time to accumulate. Alternatively, private haplotypes may be present in the two populations at a very low frequency and a much larger sample size is needed to detect the rare variants. The mix of southern and northern haplotypes at sites HR and FP suggest they could have been colonized from multiple sources, but the source populations cannot be inferred from the available data. If colonization had occurred from the north and/or the south by crawling along the coastline, a pattern of IBD might have been likely. The grouping of non-neighbouring sites HR and RI by the SAMOVA shows a ‘leapfrog’ dispersal pattern (Thiel & Haye, 2006). A similar clustering of non-neighbours has been observed for the direct-developing C. lineolata along the east coast of Australia, for which algal rafting was proposed as the mechanism that enabled gene flow (Hoskin, 1997). While rafting dispersal has been implicated in the dispersal of some direct-developing gastropods in New Zealand (Cumming et al., 2014), these species are commonly associated with large algae. Cominella maculosa whelks are are often found under rocks or buried in gravel rather than on algae or wood, which would greatly reduce their opportunities for rafting on algae. Laboratory studies conducted with C. maculosa hatchlings indicate drifting of juveniles in the water column is unlikely to contribute to a significant amount of long-distance dispersal between populations (Dohner, 2016). However, other mechanisms might transport egg capsules to other sites, for example if females attached egg capsules to drift wood, or if clumps of intact egg capsules became dislodged and drifted in the northbound, nearshore Wairarapa Current (Chiswell, 2000). Because each capsule contains an average of seven hatchlings (Carrasco & Phillips, 2014), it would take only a couple of sporadic rafting or drifting events to transport a reasonable quantity of dispersers northwards. Although rare, these events could be important over evolutionary timescales as only a few migrants per generation are required to homogenize genetic diversity (Ovenden, 2013). Assuming low post-hatching mortality, colonization by a few drifting egg capsules could create a ‘founder takes all’ situation (Waters, Fraser & Hewitt, 2013). Such colonization events could also lead to secondary contact, which might explain the presence of two highly divergent haplotypes coexisting at Castlepoint (Fig. S1). Other studies have found differentiation among populations of direct-developers closer together than those in this southern group. For example, populations of Antarctic topshells (Margarella antarctica) located less than 6 km apart (Hoffman et al., 2013), and populations of mudsnails (Zeacumanths subcarinatus) within 5 km of each other (Keeney et al., 2009) are genetically distinct. Within our own study there was a much higher level of differentiation between northern sites that have a similar amount of spatial separation to the southern group. The lack of differentiation between the three southern sites could indicate strong genetic connectivity among them, due either to demographic continuity or to recolonization after local extinctions (Kyle & Boulding 2000). However, recolonization or dispersal among sites via crawling adults is likely to be very low because rocky reefs along this coast are not contiguous, but are separated by patches of unsuitable habitat due to large sandy beaches and freshwater inputs. Alternatively, the three southern sites might be genetically similar because they were founded from the same source population. Further support for this hypothesis is that these southern Wairarapa locations, and the other southern sites sampled by Fleming et al. (2018), all have low levels of genetic diversity which can occur from low effective population sizes, such as founder events or bottlenecks. Low genetic diversity in direct-developing gastropods at high latitudes has been attributed to recolonization from lower latitudes after a population contraction while restricted to ice age refugia (Kojima et al., 2004; Marko, 2004; Keeney et al., 2009). It is possible that the southern populations in our study are a product of founding from a genetically homogenous Last Glacial Maximum refugium. Two C. maculosa samples (Donald et al., 2015; NCBI reference numbers KP694145 and KP694146) from the Kaikoura Peninsula on the South Island, 300 km away and across the Cook Strait (Fig. 1B), were compared to the current dataset. From the 552 bp that could be aligned, both individuals were identified as the same dominant haplotype found in the three southern locations of our study. Walton (2017) found that 135 of 140 C. maculosa collected from seven South Island sites also matched the dominant southern haplotype reported in the present study. The overall findings from the two studies suggest that the land gap of the Cook Strait has not historically been a strong barrier to dispersal for this species. However, this conclusion is based on data from maternally inherited mitochondrial DNA sequences, which means if only a few females were in the founder population all of the hatchlings would be genetically similar. The populations should be further investigated using bi-parentally inherited nuclear DNA markers that might allow for a more detailed study of the recent barriers to dispersal. Moreover, studies of other direct-developing intertidal species could be used to corroborate our findings, if they also displayed a similar pattern of genetic structure along the Wairarapa coast and across the Cook Strait. Overall, the genetic pattern of C. maculosa at large and small scales suggests that it is made up of discrete populations and there are low levels of gene flow, similar to patterns found for other direct-developing species (Lee & Boulding, 2009; Barbosa et al., 2013; Pálsson et al., 2014). Discrete population genetic structure is generally evidence that populations exist in demographic isolation and are unlikely to form a metapopulation that can quickly recolonize after local population extinction. Conversely, high levels of genetic connectivity do not necessarily reflect demographic connectivity (Ovenden, 2013). The decoupling of genetic and demographic connectivity in ‘crinkled’ populations such as the southern group in this study can be caused by founder events, migration into large populations or the appearance of a recent barrier to gene flow. Further fine-scale studies should be conducted across other genetic breaks to determine how the interaction of life history traits, historical events and environmental factors impacts population structure. ACKNOWLEDGMENTS We would like to thank Glenburn, Glendhu and Pahaoa Stations for the use of their roads to access several field sites. Thanks to Kerry Walton for the donation of Castlepoint whelks. Thanks also to the whelk collection volunteers: Ali Duncan, Jana Wold, Gustav Kessel, Kerry Walton, Juliette Champagnat, Angela Fleming and John van der Sman. Thanks to Angela Fleming for support in the genetics lab and providing sequence data for Castlepoint and Cape Palliser whelks. This work was improved by comments from two anonymous reviewers. Funding was provided by Victoria University of Wellington. SUPPLEMENTARY INFORMATION Additional Supporting Information may be found in the online version of this article at the publisher’s website: Figure S1. Minimum spanning network of COI haplotypes for Cominella maculosa sampled along the Wairarapa coast of the North Island of New Zealand. The size of each circle reflects the total number of samples detected for each haplotype, with colours reflecting how many individuals from each site contribute to the sample size. Hash marks represent one mutational change between haplotypes. Black dots represent theoretical haplotypes. REFERENCES Avise JC. 2000. Phylogeography: The history and formation of species . Cambridge: Harvard University Press. Bandelt HJ, Forster P, Röhl A. 1999. Median-joining networks for inferring intraspecific phylogenies. Molecular Biology and Evolution  16: 37– 48. Google Scholar CrossRef Search ADS PubMed  Barbosa SS, Klanten SO, Puritz JB, Toonen RJ, Byrne M. 2013. Very fine-scale population genetic structure of sympatric asterinid sea stars with benthic and pelagic larvae: influence of mating system and dispersal potential. Biological Journal of the Linnean Society  108: 821– 833. Google Scholar CrossRef Search ADS   Brante A, Fernández M, Viard F. 2012. Phylogeography and biogeography concordance in the marine gastropod Crepipatella dilatata (Calyptraeidae) along the southeastern Pacific coast. Journal of Heredity  103: 630– 637. Google Scholar CrossRef Search ADS PubMed  Burton RS, Feldman MW. 1982. Population genetics of coastal and estuarine invertebrates: does larval behavior influence population structure? In: Kennedy VS, ed. Estuarine comparisons . New York: Academic Press, 537– 551. Google Scholar CrossRef Search ADS   Carrasco SA, Phillips NE. 2014. Encapsulation and development of three New Zealand neogastropods with contrasting embryo packaging and maternal provisioning. New Zealand Journal of Zoology  41: 171– 186. Google Scholar CrossRef Search ADS   Chiswell SM. 2000. The Wairarapa Coastal Current. New Zealand Journal of Marine and Freshwater Research  34: 303– 315. Google Scholar CrossRef Search ADS   Collin R. 2001. The effects of mode of development on phylogeography and population structure of North Atlantic Crepidula (Gastropoda: Calyptraeidae). Molecular Ecology  10: 2249– 2262. Google Scholar CrossRef Search ADS PubMed  Cumming RA, Nikula R, Spencer HG, Waters JM. 2014. Transoceanic genetic similarities of kelp-associated sea slug populations: long-distance dispersal via rafting? Journal of Biogeography  41: 2357– 2370. Google Scholar CrossRef Search ADS   Dawson MN, Hays CG, Grosberg RK, Raimondi PT. 2014. Dispersal potential and population genetic structure in the marine intertidal of the eastern North Pacific. Ecological Monographs  84: 435– 456. Google Scholar CrossRef Search ADS   de Lange W, Bell RG, Gorman R, Reid S. 2003. Physical oceanography of New Zealand waters. In: Goff J, Nichol S, Rouse H, eds. The New Zealand coast: Te Tai o Aotearoa . Palmerston North: Dunmore Press, 59– 78. Dohner M. 2016. Crawling to connectivity? The direct-developing journey of the spotted whelk (Cominella maculosa) . Unpublished Master’s of Science Thesis, Victoria University of Wellington. Donald KM, Winter DJ, Ashcroft AL, Spencer HG. 2015. Phylogeography of the whelk genus Cominella (Gastropoda: Buccinidae) suggests long-distance counter-current dispersal of a direct developer. Biological Journal of the Linnean Society  115: 315– 332. Google Scholar CrossRef Search ADS   Dupanloup I, Schneider S, Excoffier L. 2002. A simulated annealing approach to define the genetic structure of populations. Molecular Ecology  11: 2571– 2581. Google Scholar CrossRef Search ADS PubMed  Excoffier L, Lischer HE. 2010. Arlequin suite ver 3.5: a new series of programs to perform population genetics analyses under Linux and Windows. Molecular Ecology Resources  10: 564– 567. Google Scholar CrossRef Search ADS PubMed  Excoffier L, Lischer H. 2015. Arlequin Ver 3.5: an integrated software package for population genetics data analysis . Berne: Swiss Institute of Bioinformatics. Excoffier L, Smouse PE, Quattro JM. 1992. Analysis of molecular variance inferred from metric distances among DNA haplotypes: application to human mitochondrial DNA restriction data. Genetics  131: 479– 491. Google Scholar PubMed  Fleming AM, Dohner MM, Phillips NE, Ritchie PA. 2018. Genetic connectivity among populations of two congeneric direct-developing whelks varies across spatial scales. New Zealand Journal of Marine and Freshwater Research  52: 100– 117. Google Scholar CrossRef Search ADS   Folmer O, Black M, Hoeh W, Lutz R, Vrijenhoek R. 1994. DNA primers for amplification of mitochondrial cytochrome c oxidase subunit I from diverse metazoan invertebrates. Molecular Marine Biology and Biotechnology  3: 294– 299. Google Scholar PubMed  Fu YX. 1997. Statistical tests of neutrality of mutations against population growth, hitchhiking and background selection. Genetics  147: 915– 925. Google Scholar PubMed  Giles EC, Saenz-Agudelo P, Hussey NE, Ravasi T, Berumen ML. 2015. Exploring seascape genetics and kinship in the reef sponge Stylissa carteri in the Red Sea. Ecology and Evolution  5: 2487– 2502. Google Scholar CrossRef Search ADS PubMed  Heath RA. 1982. A review of the physical oceanography of the seas around New Zealand. New Zealand Journal of Marine and Freshwater Research  19: 79– 124. Google Scholar CrossRef Search ADS   Hoffman JI, Clarke A, Clark MS, Peck LS. 2013. Hierarchical population genetic structure in a direct developing antarctic marine invertebrate. PLoS ONE  8: e63954. Google Scholar CrossRef Search ADS PubMed  Hoskin MG. 1997. Effects of contrasting modes of larval development on the genetic structures of populations of three species of prosobranch gastropods. Marine Biology  127: 647– 656. Google Scholar CrossRef Search ADS   Karl SA, Hayes KA. 2012. Extreme population subdivision in the crown conch (Melongena corona): historical and contemporary influences. Journal of Heredity  103: 523– 532. Google Scholar CrossRef Search ADS PubMed  Kearse M, Moir R, Wilson A, Stones-Havas S, Cheung M, Sturrock S, Buxton S, Cooper A, Markowitz S, Duran C, Thierer T, Ashton B, Meintjes P, Drummond A. 2012. Geneious Basic: an integrated and extendable desktop software platform for the organization and analysis of sequence data. Bioinformatics (Oxford, England)  28: 1647– 1649. Google Scholar CrossRef Search ADS PubMed  Keeney DB, King TM, Rowe DL, Poulin R. 2009. Contrasting mtDNA diversity and population structure in a direct-developing marine gastropod and its trematode parasites. Molecular Ecology  18: 4591– 4603. Google Scholar CrossRef Search ADS PubMed  Kyle CJ, Boulding E. G. 2000. Comparative population genetic structure of marine gastropods (Littorina spp.) with and without pelagic larval dispersal. Marine Biology  137: 835– 845. Google Scholar CrossRef Search ADS   Kojima S, Hayashi I, Kim D, Iijima A, Furota T. 2004. Phylogeography of an intertidal direct-developing gastropod Batillaria cumingi around the Japanese Islands. Marine Ecology Progress Series  276: 161– 172. Google Scholar CrossRef Search ADS   Laing A, Chiswell SM. 2003. The ocean medium. In: Andrew NL, Francis MP, eds. The living reef: the ecology of New Zealand’s rocky reefs . Nelson: Craig Potton Publishing, 24– 31. Lee HJ, Boulding EG. 2009. Spatial and temporal population genetic structure of four northeastern Pacific littorinid gastropods: the effect of mode of larval development on variation at one mitochondrial and two nuclear DNA markers. Molecular Ecology  18: 2165– 2184. Google Scholar CrossRef Search ADS PubMed  Librado P, Rozas J. 2009. DnaSP v5: a software for comprehensive analysis of DNA polymorphism data. Bioinformatics (Oxford, England)  25: 1451– 1452. Google Scholar CrossRef Search ADS PubMed  Marko PB. 2004. ‘What’s larvae got to do with it?’ Disparate patterns of post-glacial population structure in two benthic marine gastropods with identical dispersal potential. Molecular Ecology  13: 597– 611. Google Scholar CrossRef Search ADS PubMed  Morton JE, Miller MC. 1968. The New Zealand sea shore . Liverpool: Tinling and Co. Ovenden JR. 2013. Crinkles in connectivity: combining genetics and other types of biological data to estimate movement and interbreeding between populations. Marine and Freshwater Research  64: 201– 207. Google Scholar CrossRef Search ADS   Ovenden JR, Berry O, Welch DJ, Buckworth RC, Dichmont CM. 2015. Ocean’s eleven: a critical evaluation of the role of population, evolutionary and molecular genetics in the management of wild fisheries. Fish and Fisheries  16: 125– 159. Google Scholar CrossRef Search ADS   Pálsson S, Magnúsdóttir H, Reynisdóttir S, Jónsson ZO, Örnólfsdóttir EB. 2014. Divergence and molecular variation in common whelk Buccinum undatum (Gastropoda: Buccinidae) in Iceland: a trans-Atlantic comparison. Biological Journal of the Linnean Society  111: 145– 159. Google Scholar CrossRef Search ADS   Pelc RA, Warner RR, Gaines SD. 2009. Geographical patterns of genetic structure in marine species with contrasting life histories. Journal of Biogeography  36: 1881– 1890. Google Scholar CrossRef Search ADS   Powell AWB. 1979. New Zealand Mollusca: marine, land and freshwater shells . Auckland: Collins. Ross PM, Hogg ID, Pilditch CA, Lundquist CJ. 2009. Phylogeography of New Zealand’s coastal benthos. New Zealand Journal of Marine and Freshwater Research  43: 1009– 1027. Google Scholar CrossRef Search ADS   Ross PM, Hogg ID, Pilditch CA, Lundquest CJ, Wilkins RJ. 2012. Population genetic structure of the New Zealand estuarine clam Austrovenus stutchburyi (Bivalvia: Veneridae) reveals population subdivision and partial congruence with biogeographic boundaries. Estuaries and Coasts  35: 143– 154. Google Scholar CrossRef Search ADS   Rousset F. 1997. Genetic differentiation and estimation of gene flow from F-statistics under isolation by distance. Genetics  145: 1219– 1228. Google Scholar PubMed  Sambrook J, Fritsch EF, Maniatis T. 1989. Molecular cloning: a laboratory manual . New York: Cold Spring Harbor Press. Sánchez R, Sepúlveda RD, Brante A, Cárdenas L. 2011. Spatial pattern of genetic and morphological diversity in the direct developer Acanthina monodon (Gastropoda: Mollusca). Marine Ecology Progress Series  434: 121– 131. Google Scholar CrossRef Search ADS   Selkoe KA, Toonen RJ. 2011. Marine connectivity: a new look at pelagic larval duration and genetic metrics of dispersal. Marine Ecology Progress Series  436: 291– 305. Google Scholar CrossRef Search ADS   Scheltema RS. 1971. Larval dispersal as a means of genetic exchange between geographically separated populations of shallow-water benthic marine gastropods. Biological Bulletin  140: 284– 322. Google Scholar CrossRef Search ADS   Scheltema RS. 1986. On dispersal and planktonic larvae of benthic invertebrates: an eclectic overview and summary of problems. Bulletin of Marine Science  39: 290– 322. Tajima F. 1989. The effect of change in population size on DNA polymorphism. Genetics  123: 597– 601. Google Scholar PubMed  Thiel M, Haye PA. 2006. The ecology of rafting in the marine environment. III. Biogeographical and evolutionary consequences. Oceanography and Marine Biology  44: 323– 429. Google Scholar CrossRef Search ADS   Veale AJ, Lavery SD. 2011. Phylogeography of the snakeskin chiton Sypharochiton pelliserpentis (Mollusca: Polyplacophora) around New Zealand: are seasonal near-shore upwelling events a dynamic barrier to gene flow? Biological Journal of the Linnean Society  104: 552– 563. Google Scholar CrossRef Search ADS   Waters JM, Fraser CI, Hewitt GM. 2013. Founder takes all: density-dependent processes structure biodiversity. Trends in Ecology & Evolution  28: 78– 85. Google Scholar CrossRef Search ADS PubMed  Walton K. 2017. Phylogeography of the New Zealand whelks Cominella maculosa and C. virgata . Unpublished Master’s of Science Thesis, Victoria University of Wellington. © 2018 The Linnean Society of London, Biological Journal of the Linnean Society

Journal

Biological Journal of the Linnean SocietyOxford University Press

Published: Mar 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off