Journal of Biogeography (J. Biogeogr.) (2012) 39, 2279–2291 ORIGINAL ARTICLE Genetically diverse but with surprisingly little geographical structure: the complex history of the widespread herb Carex nigra (Cyperaceae) Pedro Jiménez-Mejı́as1,2*, Modesto Luceño1, Kåre Arnstein Lye3, Christian Brochmann2 and Galina Gussarova2,4 1 Botany Area, Department of Molecular Biology and Biochemical Engineering, Pablo de Olavide University, 41013 Seville, Spain, 2 National Centre for Biosystematics, Natural History Museum, University of Oslo, NO-0318 Oslo, Norway, 3Department of Ecology and Natural Resource Management, Norwegian University of Life Sciences, NO-1432 Ås, Norway, 4Department of Botany, St. Petersburg State University, 199034 St Petersburg, Russia ABSTRACT Aim Based on extensive range-wide sampling, we address the phylogeographical history of one of the most widespread and taxonomically complex sedges in Europe, Carex nigra s. lat. We compare the genetic structure of the recently colonized northern areas (front edge) and the long-standing southern areas (rear edge), and assess the potential genetic basis of suggested taxonomic divisions at the rank of species and below. Location Amphi-Atlantic, central and northern Europe, circum-Mediterranean mountain ranges, central Siberia, Himalayas. Methods A total of 469 individuals sampled from 83 populations, covering most of the species’ range, were analysed with amplified fragment length polymorphism (AFLP) and chloroplast DNA (cpDNA) markers. Bayesian clustering, principal coordinates analysis, and estimates of diversity and differentiation were used for the analysis of AFLP data. CpDNA data were analysed with statistical parsimony networks and maximum parsimony and Bayesian inference of phylogenetic trees. Results Overall genetic diversity was high, but differentiation among populations was limited. Major glacial refugia were inferred in the Mediterranean Basin and in western Russia; in addition, there may have been minor refugia in the North Atlantic region. In the southern part of the range, we found high levels, but geographically quite poorly structured genetic diversity, whereas the levels of genetic diversity varied among different areas in the north. North American populations were genetically very similar to the European populations. *Correspondence: Pedro Jiménez-Mejı́as, Botany Area, Department of Molecular Biology and Biochemical Engineering, Pablo de Olavide University, Carretera de Utrera km 1, 41013 Seville, Spain. E-mail: pjimmej@upo.es ª 2012 Blackwell Publishing Ltd Main conclusions The data are consistent with extensive gene flow, which has obscured the recent history of the taxon. The limited differentiation in the south probably results from the mixing of lineages expanding from several local refugia. Northward post-glacial colonization resulted in a leading-edge pattern of low diversity in the Netherlands, Belgium, Scotland and Iceland, whereas the observed high diversity levels in Fennoscandia suggest broad-fronted colonization from the south as well as from the east. The patterns found in the American populations are consistent with post-glacial colonization, possibly even with anthropogenic introduction from Europe. Our data also suggest that the tussock-forming populations of C. nigra, often referred to as a distinct species (Carex juncella), represent an ecotype that has originated repeatedly from different populations with creeping rhizomes. Keywords Boreal–temperate herb, Europe, extensive gene flow, glacial refugia, rpl32– trnLUAG, trans-Atlantic dispersal, vicariance, ycf6–psbM. http://wileyonlinelibrary.com/journal/jbi doi:10.1111/j.1365-2699.2012.02740.x 2279 P. Jiménez-Mejı́as et al. INTRODUCTION Only three studies have explored the phylogeography of widespread temperate herbs across their entire geographical range in Europe (Melica nutans and Carex digitata, Tyler, 2002a,b; Carex pilosa, Rejzková et al., 2008). High levels of genetic variation, high dispersal ability and an overall lack of strong geographical structuring were inferred, but more species of this kind should be examined to establish whether or not the pattern identified is a general one. In widely distributed species, different parts of the range can be affected by different historical events and, thus, represent distinct genetic patterns. The Pliocene–Pleistocene climatic oscillations strongly affected the geographical ranges of species and ecosystems. Unsuitable habitat conditions caused species ranges to contract and split, leading to isolation and vicariant divergence (Kropf et al., 2006), whereas secondary contact during colonization/recolonization in warmer periods often led to the re-establishment of gene flow, counteracting complete speciation (Comes & Kadereit, 1998; Kadereit et al., 2004). In refugial populations, long-term isolation drives genetic differentiation, and in the southernmost ones the rear-edge reduction of suitable habitats can force population decrease and the loss of genetic diversity (Hampe & Petit, 2005). In recently colonized territories, loss of diversity may be induced by successive founder effects at the colonization front (Hewitt, 1996, 1999; Schönswetter et al., 2003; Puşcaş et al., 2008). However, increased genetic variation in recently colonized areas can be found in central parts of the distribution range, where different colonization fronts meet, resulting in contact zones (Konnert & Bergmann, 1995; Taberlet et al., 1998; Hewitt, 1999; Petit et al., 2003). Carex nigra (L.) Reichard is one of the most widespread sedges in Europe (Fig. 1a). This species occurs almost continuously in peat bogs and lakesides in lowlands across central and northern Europe. To the south, it occurs in scattered locations in the high mountains of the four Mediterranean peninsulas, and in Corsica, Sicily and north-western Africa. To the east, C. nigra is found in central Siberia, the Himalayas and the Caucasus (Noltie, 1994; Egorova, 1999). In the west, it shows a typical amphi-Atlantic distribution in Iceland, Greenland and north-eastern North America. Carex nigra is wind-pollinated and reported to be more or less selfincompatible (Faulkner, 1973). It belongs to Carex sect. Phacocystis Dumort., a taxonomically complex group with many species characterized by clonal growth through creeping rhizomes and widespread interspecific hybridization (Faulkner, 1973). Hybrids have been reported between C. nigra and almost all co-occurring taxa in the section (Sylvén, 1963; Chater, 1980; Jermy et al., 2007). The fruits and utricles (perigynia) of C. nigra lack any special dispersal adaptations, in contrast to what is seen in other Carex species (Allessio Leck & Schütz, 2005). However, viable Carex seeds have been recovered from bird gut contents (Schmid, 1986; Mueller & van der Valk, 2002), and endozoochory might be an important dispersal mechanism in C. nigra. The species has a variable somatic chromosome number (2n = 80–88; Roalson, 2008), Figure 1 Global distribution range of Carex nigra, with sampling locations and genetic diversity patterns in Europe, eastern North America and western Asia. (a) Global distribution range (shaded, modified from Hultén, 1958). (b) Sampling locations with cpDNA haplotypes; numbers refer to the network in Fig. 2a. (c) Genetic (AFLP) groups as identified by structure, with circle size proportional to the number of individuals sampled from each population. (d) Amplified fragment length polymorphism gene diversity and rarity (DW). Circle size represents gene diversity within each population, colours represent DW values, and partitions match interquartile intervals. Three circles with stars represent populations consisting of a single clone, for which DW could not be calculated. Scale-bar measurements are in kilometres. 2280 Journal of Biogeography 39, 2279–2291 ª 2012 Blackwell Publishing Ltd Phylogeography of Carex nigra with the variation not following geographical or taxonomic relationships. The taxonomy of C. nigra s. lat. is highly problematic, and there is little consensus among treatments. Wide variation in vegetative characters has led to the description of numerous taxa at the rank of species and below. The main intraspecific division accepted by most botanists is between the more common plants with creeping rhizomes (‘var. nigra’) and plants forming conspicuous tussocks [C. nigra var. juncea (Fr.) Hyl.], which have also been recognized as a distinct subspecies [C. nigra subsp. juncella (Fr.) Lemke] or even species [C. juncella (Fr.) Th. Fr.]. These two growth forms occur sympatrically in northern Europe and western Siberia (Sylvén, 1963; Chater, 1980; Egorova, 1999): ‘var. juncea’ typically occurs in seasonally flooded places, and ‘var. nigra’ occurs mostly in year-round humid habitats or in habitats that appear dry but have shallow groundwater levels. Carex nigra subsp. intricata (Tineo) Rivas Mart. (Carex intricata Tineo) is a name applied in local floristic treatments to the dwarfed, densely tufted plants with wide leaves that grow in the high mountains of Corsica, Sicily, southern Spain and North Africa (Maire, 1957; Vicioso, 1959; Chater, 1980; Pignatti, 1982). Other alpine forms more similar to typical C. nigra have been reported as C. nigra subsp. alpina (Gaudin) Lemke from northern and central European ranges (Chater, 1980). The name C. nigra subsp. dacica (Heuff.) Soó has been erroneously used (it applies strictly to C. bigelowii; cf. Egorova, 1999) to emphasize the distinctiveness of the plants from the Balkan Peninsula and Turkey (Chater, 1980; Nilsson, 1985), based mainly on differences in the colour of the basal sheaths. Finally, C. transcaucasica T.V. Egorova (Egorova, 1999) and C. nigra subsp. drukyulensis Noltie (Noltie, 1994) are C. nigra-like taxa described from the Caucasus and Bhutan, respectively. This intricate taxonomy reflects the complex morphological and ecological variation patterns found in C. nigra s. lat. across its wide geographical range. Based on extensive range-wide sampling, we address the phylogeographical history of C. nigra s. lat. based on AFLP markers and chloroplast DNA (cpDNA) sequences. In particular, because this species occurs more or less continuously in northern and central Europe but has a scattered, seemingly relictual, distribution in the Mediterranean, we consider whether its genetic structure differs between the recently colonized northern areas (front edge) and the long-standing zones (rear edge) in the south. We also aim to contribute to the taxonomy of this complicated species, especially in assessing whether the tussock-forming plants form a separate genetic group, representing a distinct taxon, or whether they group with rhizomatous plants from different geographical areas. MATERIALS AND METHODS Sampling Leaf material was collected in the field and stored in silica gel. In addition, a few samples were taken from herbarium material Journal of Biogeography 39, 2279–2291 ª 2012 Blackwell Publishing Ltd (E, UPOS) and used for cpDNA sequencing (see below). A total of 469 individuals from 83 distinct populations were included in the study, covering most of the distribution range (Fig. 1, and see Appendix S1 in Supporting Information). The sample size in each population varied from 1 (in 12 populations) to 10 (in 16 populations) individuals. Special effort was made to include the full range of morphological variation reported in C. nigra s. lat., especially populations described as distinct taxa. We sampled in the type localities of C. intricata, C. nigra var. juncea and C. nigra s. str., as well as plants from Bulgaria and Greece assignable to C. nigra subsp. dacica, plants from Bhutan (Himalayas) assignable to C. nigra subsp. drukyulensis, and plants from Armenia and Iran assignable to C. transcaucasica. As C. nigra can spread vegetatively by creeping rhizomes, samples were taken as far from each other as possible to minimize clonal sampling. Voucher specimens from populations sampled in the field were deposited in the herbaria of the Botanical Museum, Oslo (O, Norway) and Pablo de Olavide University in Seville (UPOS, Spain). Carex bigelowii subsp. rigida was included as an outgroup (cf. Dragon & Barrington, 2009). DNA isolation, polymerase chain reaction amplification and AFLP fingerprinting Total DNA was extracted from dried plant tissue using DNeasy plant extraction kits (Qiagen, Valencia, CA). The variability in 10 cpDNA regions was tested using a subset of plants. The most variable ones were ycf6–psbM and rpl32–trnLUAG, which were chosen for full analysis. Primers and protocols were as described in Shaw et al. (2005, 2007). A single individual per population was included, except for the Iranian population, from which two samples were sequenced, giving a total of 84 individuals for cpDNA analyses. In total, 459 individuals from 73 of the populations were successfully analysed using AFLPs (Appendix S1). The laboratory procedure followed Gaudeul et al. (2000) as modified by Schönswetter et al. (2006). A pilot study was performed on eight individuals from four populations, including one replicate of each. This pilot study included a selection of AFLP primers successfully used in Carex (Schönswetter et al., 2006, 2008; Nakamatte & Lye, 2007; Jiménez-Mejı́as et al., 2011). We chose primer combinations that retrieved the highest number of informative characters after reproducibility was tested: 6-FAM–EcoRI+AGT/MseI+AGC, NED–EcoRI+ACC/MseI+ACC and VIC–EcoRI+AGG/MseI+CA. Selective polymerase chain reaction (PCR) products were run on a capillary sequencer (ABI PRISM 3100; Applied Biosystems, Foster City, CA) with the internal size standard GeneScan ROX 500 (Applied Biosystems). Data collection and fragment sizing were performed using the program GeneMapper 3.7 (Applied Biosystems). Fragments in the range 50–500 bp were automatically scored with GeneMapper 3.7 and manually revised. The results were exported as a presence/absence (1/0) matrix. Negative controls and replicates were included throughout the process. Reproducibility was estimated based on 69 2281 P. Jiménez-Mejı́as et al. replicated samples (15% of the sampling) as the average proportion of correctly replicated bands (Bonin et al., 2004). Markers with low reproducibility were excluded. Rare presences and absences with a frequency lower than the error rate were carefully checked and used only if they were unambiguous. Linked alleles were removed from the matrix. Data analyses CpDNA chromatograms were visualized and edited using SeqEd (Applied Biosystems). The sequence matrix was easily aligned manually in a text editor, because of the few variable characters observed. Gap coding was applied using the ‘simple indel coding’ method of Simmons & Ochoterena (2000), implemented in SeqState (Müller, 2005). A statistical parsimony network analysis was conducted using tcs (Clement et al., 2000). Bootstrap support values for the tcs network were obtained using SplitsTree4 (Huson & Bryant, 2006). Maximum parsimony analyses were conducted in paup* 4.0b10 (Swofford, 2002). Random trees were used as starting points, with 100 replicates. The tree bisection–reconnection (TBR) algorithm was used for branch swapping, without the steepest descent option; no more than 20 trees of score (length) greater than or equal to 1 were saved in each replicate. Branches were collapsed if the maximum branch length was zero. The MulTrees option was in effect, and no topological constraints were enforced. In order to obtain corrected values of the consistency index, it was recalculated with uninformative characters excluded. Parsimony bootstrap support values were obtained from 1000 pseudo-replicates with four random sequence additions using heuristic search options of paup* including both informative and uninformative characters. Bayesian phylogenetic analyses were conducted in MrBayes 3.1.2 (Huelsenbeck & Ronquist, 2001; Ronquist & Huelsenbeck, 2003). The Akaike information criterion with an empirical correction for small sample sizes (AICc), as implemented in MrAIC (Nylander, 2004) together with PhyML (Guindon & Gascuel, 2003), were used for substitution model selection. Bayesian analyses included running one cold and three heated Markov chains for 6,000,000 generations during two simultaneous runs under the F81 model, with a sample recorded every 1000 generations, and 1000 generations of burn-in taken to reach stationarity. The priors were set according to the output of MrAIC. Coded indels were included as a separate data partition using the F81-like model implemented in MrBayes for binary data. To test whether the Markov chains converged, we monitored the standard deviation of split frequencies (SDSF), which should fall below 0.01 when comparing two independent runs. The results of the analysis were summarized as 50% majority rule consensus trees. The Bayesian analyses were run twice to confirm the topology. The repeated Bayesian analyses converged to identical topologies with minor variation in posterior probability values. For the AFLP data, a Bayesian clustering analysis was performed in structure 2.2 (Pritchard et al., 2000) using the Bioportal computer service of the University of Oslo (http:// 2282 www.bioportal.uio.no). structure estimates the number of genetically homogeneous groups in the data set (K). The admixture model, in which shared group membership is allowed for an individual, was considered the most appropriate for our data set, as recommended by Ehrich et al. (2008). In order to assess hierarchical structure, each identified group was also analysed separately. Similarity among the runs and stabilization of K likelihood scores were the main criteria used to choose the optimal number of groups; they were calculated using the R script Structure-sum (Ehrich, 2006; updated version 2009). Similarity among multi-locus AFLP genotypes was visualized using principal coordinates analysis (PCoA) based on Jaccard’s coefficient as implemented in ntsys (Rohlf, 1997). The ‘frequency-down-weighted marker’ (DW) value was calculated as a rarity measure (Schönswetter & Tribsch, 2005) using the R script AFLPdat (Ehrich, 2006; updated version 2009), which accounts for differences in sample size as explained in Ehrich et al. (2008). Gene diversity of the populations (after removing clones) was evaluated according to Nei’s formula for haplotype diversity (Nei, 1978), also as implemented in AFLPdat. Analyses of molecular variance (AMOVAs) to assess the level of genetic differentiation among populations and groups, and Nei’s gene diversity H (Nei, 1973) were computed using Arlequin 3.5.1.2 (Excoffier & Lischer, 2010). For AMOVA, eight distinct groupings were tested for the whole data set: populations, structure groups from K = 2 and K = 4 (see Results), Old World versus New World, cpDNA haplotypes, comparison of haplotype H9 with all other haplotypes together (see Results), and taxonomic assignment to varieties ‘juncea’ or ‘nigra’. In addition, seven partial data sets were analysed: plants from glaciated areas, plants from unglaciated areas, and five distinct cpDNA haplotype partitions (H1 alone, H9 alone, H1–H5, H6–H8, H1–H8). For DW, gene diversity and AMOVA analyses, all samples from Newfoundland and from Nova Scotia populations were merged into one set representing each geographical region, to minimize the effects of sample sizes. RESULTS CpDNA sequences The complete matrix for the two cpDNA regions contained 1129 bp (rpl32–trnLUAG: 715 bp; ycf6–psbM: 414 bp). Three indels, of 7, 5 and 1 bp, were coded as additional characters, the last one being autapomorphic. Nine haplotypes (labelled H1–H9; Fig. 1b, Appendix S1) were identified in C. nigra. The statistical parsimony network analysis retrieved a single network (Fig. 2a). Rooting with C. bigelowii suggested that the ancestral haplotype of C. nigra was unsampled or extinct (Schaal et al., 1998). Among the identified C. nigra haplotypes, H6 and H9 appeared closest to the ancestral sequence. A split of two mutational steps divided the C. nigra haplotypes into two groups: one with the Iberian haplotype H9, and one with all other haplotypes. The most widespread haplotype was H1, Journal of Biogeography 39, 2279–2291 ª 2012 Blackwell Publishing Ltd Phylogeography of Carex nigra Figure 2 Phylogenetic tree and cpDNA haplotype network of Carex nigra based on the 84 combined ycf6–psbM and rpl32– trnLUAG sequences. (a) Statistical parsimony network comprising the nine cpDNA haplotypes. Small black circles represent extinct or unsampled haplotypes; each line between haplotypes represents a mutation step. Circle size is proportional to the number of individuals with this haplotype, except for H1, which is shown 10 times smaller. Numbers near branches show bootstrap support values obtained in SplitsTree. (b) 50% majority rule consensus tree from maximum parsimony analysis of the cpDNA sequences. Bootstrap support (%) and posterior probability values (obtained in MrBayes) are given above and below the branches respectively. Labelling of the samples includes growing habit (Jun, tussockforming ‘var. juncea’; Nig, creeping-rhizome ‘var. nigra’), country [following TDWG botanical countries nomenclature (Brummitt, 2001): AUT, Austria; BGM, Belgium; BUL, Bulgaria; COR, Corsica; DEN, Denmark; EHM, Eastern Himalayas; FIN, Finland; FRA, France; GER, Germany; GRB, Great Britain; GRC, Greece; GRN, Greenland; ICE, Iceland; IRN, Iran; ITA, Italy; MOR, Morocco; NET, the Netherlands; NFL, Newfoundland; NOR, Norway; NSC, Nova Scotia; POR, Portugal; RUW, Northwest European Russia; SIC, Sicily; SPA, Spain; SWE, Sweden; TRC, Transcaucasus; TUR, Turkey; YUG-CN, Montenegro; YUG-SE, Serbia]. Iberian group with 94% bootstrap support (BS) and 1.00 posterior probability (PP). All the remaining sequences were placed in the other clade (64% BS, 0.99 PP). This latter clade showed a polytomy consisting of the unresolved haplotypes H6 and H7 (southern/eastern group), the western Mediterranean haplotype H8 (western Mediterranean group; 61% BS, 1.00 PP), and one subclade with haplotypes H1–H5 (widespread group; 65% BS). This subclade was recovered in a polytomy together with the haplotypes H6 and H7 in the Bayesian analysis. AFLPs which occupied a central position within its group together with haplotype H6. The latter occurred in a scattering of sites across the Mediterranean and also showed a disjunct occurrence in the Himalayas. Maximum parsimony and Bayesian phylogenetic analyses revealed two main clades (Fig. 2b), from which four cpDNA haplotype groups can be defined. One of the clades contained only haplotype H9, forming a NW–SE Journal of Biogeography 39, 2279–2291 ª 2012 Blackwell Publishing Ltd The geographical coverage of the AFLP analysis was more limited than that in the cpDNA analysis, because herbarium material and some other samples could not be included. The final AFLP matrix consisted of 459 individuals and 180 polymorphic markers. The final error rate was 1.43%. The structure analysis under the admixture model (Fig. 1c) suggested K = 2 as the optimal number of groups in the total data set. The results were identical among the 10 replicate runs, and only a few individuals showed mixed ancestry. Group 1 comprised mainly populations from the western and central Mediterranean and the Caucasus, while Group 2 was widespread across most of the distribution area of C. nigra. Some populations contained a single individual assigned to another group, while five populations from Spain and Morocco showed more or less equal numbers of individuals assigned to different groups (see Appendix S1). In separate structure analyses for each group, no internal structure was revealed in Group 2, but Group 1 was further divided into four subgroups. One of these was excluded as only low scores from a few individuals were assigned to it; the other three subgroups 2283 P. Jiménez-Mejı́as et al. sorted plants mostly according to geographical area. Subgroup 1A comprised mainly central–northern Iberian populations, subgroup 1B was almost totally confined to the Sierra Nevada and north-western Spain, and subgroup 1C comprised populations mainly from Atlas, Corsica, Sicily and the Caucasus. In addition, a few individuals belonging to these subgroups were also found scattered in populations from central and northern Europe and North America. Most of the Group 1 populations had haplotypes H6–H9, whereas most of the Group 2 populations had haplotypes H1–H5 (Fig. 1b,c). PCoA revealed no clear splits within the data set, and the first two axes accounted for only 9.97% of the variation (Fig. 3). The two main structure groups were separated along axis 1, but with a wide overlap (Fig. 3a). The four main cpDNA haplotype groups, defined as H1–H5 (widespread group), H6–H7 (southern/eastern group), H8 (western Mediterranean group) and H9 (NW–SE Iberian group), could not be distinguished based on the AFLP data (Fig. 3b). The samples belonging to ‘var. juncea’ and ‘var. nigra’ were fully intermixed in the PCoA plot (Fig. 3c). Only some Atlantic European populations (Belgium and the Netherlands) appeared somewhat distinct (Fig. 3d). Individuals from some southern populations were placed peripherally in the plot (Sierra Nevada, Sicily, Rif and Atlas; Fig. 3e). The North American, Greenland and Iceland populations were found in different parts of the plot (Fig. 3f). The distribution of the number of pairwise differences among AFLP genotypes within populations was distinctly bimodal (Fig. 4), suggesting the existence of clones in the data set. Taking into account the error rate (1.43%), individuals differing by up to 2.6 markers (rounded up to 3) were considered to represent the same clone. Although precautions had been taken during sampling, putative clones were found in 46 of the 71 populations analysed. Putative clones were even identified within populations of tussock-forming plants (‘var. juncea’). The average gene diversity was 0.089 (SD = 0.044) after removing the 146 identified clones. The populations from the Netherlands, Belgium and Germany (north-western Europe), Moroccan Rif and Sicily were among the least diverse (Fig. 1d; Appendix S1). There was little difference in average diversity between tussock-forming populations and populations with creeping rhizomes (0.062 vs. 0.058) or between glaciated and unglaciated areas (0.061 vs. 0.056). Surprisingly, the Old World populations contained less average diversity (0.058) than the New World populations (0.071). Genetically, the most distinctive populations (i.e. with high DW; Fig. 1d; Appendix S1) originated from the southern part of the range (e.g. from the Austrian Alps, Spain, Corsica, Atlas, France and Greece), but some populations from Scandinavia, Russia and Greenland also had high DW values. The lowest DW values were observed in northern European populations, and in a single central Pyrenean population. In the AMOVA of the total data set (Table 1), most of the variation (65.47%) was found within populations, 34.53% was found among populations, 8.52% between the two main structure groups, and 15.44% among the four structure 2284 groups/subgroups. Only 6.26% of the variation was found among cpDNA haplotypes, and when the populations with the divergent haplotype H9 were compared with all the remaining populations, differentiation was only 8.75%. The variation between the rhizomatous (‘var. nigra’) and the tussockforming (‘var. juncea’) groups was only 1.15%. The Old and New World populations were not differentiated overall ()2.19%). In the AMOVAs of partial data sets (Table 2), the populations containing the southern/eastern haplotypes H6–H8 were more differentiated (42.55%) than those with haplotypes H1–H5 (31.21%). However, when populations with haplotypes H1–H8 were all tested together, differentiation was 33.33%. The populations with the haplotype H9 were also differentiated from the rest for AFLPs (37.50%). DISCUSSION Poor differentiation suggests extensive gene flow According to our cpDNA and AFLP diversity data, the Mediterranean is a centre of diversity for C. nigra. Both markers, however, show poor geographical structure and high overall genetic diversity, implying a historical scenario with vicariance and multiple secondary contacts as the main drivers that have shaped the current genetic structure. Low levels of differentiation have also been reported in the closely related C. bigelowii (Schönswetter et al., 2008), in line with the expectation of high intrapopulational variation and low differentiation in wind-pollinated out-crossers (Hamrick & Godt, 1990). The continuity of the range of C. nigra north of the Mediterranean has clearly promoted extensive gene flow, preventing differentiation and increasing genetic variation (cf. O’Brien & Freshwater, 1999; Stenström et al., 2001). Interestingly, some Scandinavian populations show an admixture of different Mediterranean AFLP groups, suggesting gene flow as a result of long-distance dispersal. This is not unexpected, as the ability to colonize at enormous geographical scales has been previously reported in Carex (e.g. Schönswetter et al., 2008; Escudero & Luceño, 2009). Glacial refugia and post-glacial colonization In contrast to what is expected in regions colonized postglacially, glacial refugia are typically characterized by high genetic distinctiveness (DW) (Tribsch et al., 2002; Schönswetter & Tribsch, 2005), as well as by high genetic diversity (Taberlet et al., 1998; Hewitt, 1999). The high distinctiveness and diversity observed in the C. nigra populations from southern Europe (mainly the Mediterranean basin) and western Russia (Tver Oblast) suggest that these regions served as refugia during the last glaciation, as proposed for other plants (e.g. Konnert & Bergmann, 1995; Taberlet et al., 1998; Petit et al., 2003; Schönswetter et al., 2003; Magri et al., 2006). Our data also suggest possible minor refugia in the northern part of the distribution range: the populations from Sweden and Greenland have unexpectedly high DW values, and a rare cpDNA Journal of Biogeography 39, 2279–2291 ª 2012 Blackwell Publishing Ltd Phylogeography of Carex nigra Figure 3 Principal coordinates analysis (PCoA) of amplified fragment length polymorphism data for Carex nigra. The same data are shown in each plot: (a) genetic groups identified by structure; (b) main inferred cpDNA haplotype groups; (c) creeping-rhizome growth form (‘var. nigra’) versus tussock-forming growth form (‘var. juncea’); (d) placement of the populations from north-western Europe (Belgium, Germany, the Netherlands, Scotland and Iceland; remaining populations displayed in light grey); (e) placement of the southern peripheral populations (Atlas, Rif, Sicily and Sierra Nevada; remaining populations displayed in light grey); (f) placement of the populations from Iceland, Greenland and North America (remaining populations displayed in light grey). Journal of Biogeography 39, 2279–2291 ª 2012 Blackwell Publishing Ltd 2285 60 0 20 40 Frequency 80 100 P. Jiménez-Mejı́as et al. 0 5 10 15 20 25 30 Pairwise differences Figure 4 Distribution of the number of pairwise differences among Carex nigra samples within populations. haplotype (H4) was detected only in southern Norway. In situ glacial survival in northern Europe and/or in North Atlantic areas has recently been proposed for some tree species (Willis et al., 2000; Palmé et al., 2003; Magri et al., 2006) and also for a few herbs (Tyler, 2002a,b; Rejzková et al., 2008; Schönswetter et al., 2008; Westergaard et al., 2011b). However, Grouping compared and source of variation Populations Among populations Within populations structure K = 2 Among groups Within groups structure K = 4 Among groups Within groups Old World versus New World Among groups Among populations Within populations Haplotypes Among groups Among populations Within populations H9 versus all other haplotypes Among groups Among populations Within populations C. nigra ‘var. nigra’ versus ‘var. Among groups Among populations Within populations 2286 Variance components the northern populations of C. nigra mentioned above were poorly differentiated overall (see Figs 1c & 3), suggesting that the rare markers enhancing their DW values are of recent (post-glacial) origin (e.g. local introgression events; Jonsson & Prentice, 2000). A recent mutation may explain the occurrence of the rare haplotype H4 in C. nigra in Norway; a similar explanation was suggested for Betula by Palmé et al. (2003). Our results suggest that post-glacial colonization in C. nigra fits the ‘southern richness versus northern purity’ paradigm (Hewitt, 2001). The Mediterranean region harbours most of the genetic diversity, in contrast to northern areas (Fig. 1b,c). The among-population variation is also higher (37.45%) in unglaciated than in formerly glaciated areas (29.80%) (Table 2). In the Mediterranean, vicariance has been proposed as the main process to have driven differentiation (Zhang et al., 2001; Kropf et al., 2006; Martı́n-Bravo et al., 2010), promoted by populations being split among different mountains during warmer periods and by subsequent isolation. It appears that in C. nigra some populations remained quite isolated and genetically differentiated, whereas others expanded and met during colder periods, leading to extensive mixing of more or less differentiated gene pools. It is possible that two main range shifts have shaped the overall phylogeographical structure of C. nigra (Figs 1b & 2a,b): one early expansion reaching the Himalayas, and a more recent expansion of the currently widespread lineage, reaching the Caucasus and North America and establishing secondary contacts with all other groups. d.f. Sum of squares Percentage of variation 66 392 1592.070 2053.817 2.76305 5.23933 34.53 65.47 1 457 117.701 3528.186 0.71882 7.72032 8.52 91.48 3 455 249.616 3396.271 1.36341 7.46433 15.44 84.56 1 65 392 16.475 1575.595 2053.817 )0.17169 2.77622 5.23933 )2.19 35.39 66.80 6 60 392 252.718 1339.352 2053.817 0.52054 2.55653 8.31640 6.26 30.74 63.00 1 65 392 juncea’ 1 65 392 79.237 1512.833 2053.817 0.75709 2.65168 5.23933 8.75 30.66 60.58 39.188 1552.882 2053.817 0.09287 2.73402 5.23933 1.15 33.89 64.95 Table 1 Analyses of molecular variance for amplified fragment length polymorphism genotypes of Carex nigra based on the entire data set. Journal of Biogeography 39, 2279–2291 ª 2012 Blackwell Publishing Ltd Phylogeography of Carex nigra Table 2 Analyses of molecular variance for amplified fragment length polymorphism genotypes of Carex nigra based on partial data sets. Data set, grouping compared and source of variation Glaciated areas Among populations 39 Within populations 195 Unglaciated areas Among populations 26 Within populations 197 H1 Among populations 54 Within populations 300 H1–H5 (widespread group) Among populations 56 Within populations 312 H6–H8 (southern/eastern group plus western Among populations 5 Within populations 48 H1–H8 Among populations 62 Within populations 360 H9 (NW–SE Iberian group) Among populations 3 Within populations 32 The effect of isolation on rear-edge populations can be observed as so-called centrifugal differentiation. Isolated peripheral populations of C. nigra, that is, those from the Atlas, Rif, Sicily and Sierra Nevada, were little influenced by the secondary contacts (Figs 1 & 3e). Genetic drift may have led to loss of genetic diversity in the populations from Rif and Sicily. In these areas, C. nigra forms tiny patches and may have experienced peripheral depauperation in marginal habitats (Lönn & Prentice, 2002; Johannesson & André, 2006). Climatic marginality, such as very dry summers in lowelevation mountains (lower than 2500 m in the Rif, and 1850 m in Sicily), could play a role (cf. Puşcaş et al., 2008). In contrast, populations from the Sierra Nevada and Atlas mountains showed average genetic diversity, which can be explained by the availability of suitable environments at elevations over 3000 m, allowing C. nigra to form large populations. Northward migration of C. nigra seems to have had different genetic consequences on either side of the colonization front. The loss of diversity in Fennoscandia appears to be modest, whereas other parts of north-western Europe are genetically more depauperate (Fig. 1d; Appendix S1). Repeated bottlenecks during a rapid expansion process (leading-edge model) could explain the low diversity found in the populations from Belgium, northern Germany, the Netherlands and Scotland (Hewitt, 1999; Wróblewska & Brzosko, 2006), resulting in reduced genetic diversity and interpopulational genetic homogenization (Hewitt, 1996; Petit et al., 2003). Genetic drift might have operated strongly enough in such cases to result in the differentiation of some populations (Fig. 3d), as suggested for Saxifraga cernua in Scotland (Westergaard et al., 2008). In contrast, the higher Journal of Biogeography 39, 2279–2291 ª 2012 Blackwell Publishing Ltd d.f. Sum of squares Variance components Percentage of variation 682.569 979.942 2.13365 5.02534 29.80 70.20 843.438 1073.875 3.26376 5.45114 37.45 62.55 1131.703 1561.294 2.44905 5.20431 32.00 68.00 1156.068 2.38493 1639.894 5.25607 Mediterranean group) 198.469 3.84068 248.864 5.18466 31.21 68.79 42.55 57.45 1413.975 1888.758 2.62243 5.24655 33.33 66.67 98.858 165.058 3.09467 5.15807 37.50 62.50 diversity found in Fennoscandia can be explained by the accumulation of variation in a suture zone between westward and northward colonization fronts (Konnert & Bergmann, 1995; Hewitt, 1999; Petit et al., 2003). This suture zone may also have contributed to gene flow among source areas (Tyler, 2002a,b). It is also possible that the colonization process in Fennoscandia may have been broad-fronted enough to maintain genetic variability (Eidesen et al., 2007). Our results suggest that the trans-Atlantic distribution pattern of C. nigra can be explained by recent cross-oceanic dispersal from Europe to North America, probably occurring more than once. Carex nigra is found in the New World along the north-eastern Atlantic coast, from Greenland south to New York, where it is generally considered native (Cayouette & Morisset, 1986; Standley et al., 2002). The North American populations included in our analyses were very similar to the European ones, but shared much of the European variation, suggesting multiple colonizations (Fig. 3; Table 1). This finding is in line with other recent studies suggesting that the North Atlantic Ocean represents a less severe barrier to plant migration than is traditionally envisioned (Abbott & Brochmann, 2003). Trans-Atlantic dispersal has also been proposed for other Carex species (Schönswetter et al., 2008; Westergaard et al., 2011a). In C. nigra, colonization of North America from Europe was suggested by Dragon & Barrington (2008). However, we cannot exclude the possibility that the species was introduced into North America at least partly by humans. The American presence of genetic group 1b (otherwise spread through the Iberian Peninsula and Norway; Fig. 1c) and the negative AMOVA values obtained in the New World–Old World comparison (Table 1) reveal that some American plants are genetically more similar to Old World 2287 P. Jiménez-Mejı́as et al. plants than to other American plants, even those from the same location. Role of interspecific hybridization All specimens included in this study were examined to exclude possible interspecific hybrids. In agreement with our morphological observations, genetic variation patterns detected in selected ‘pure’ C. nigra did not indicate extensive interspecific introgression. In particular, the genetically depauperate populations from Belgium, northern Germany, the Netherlands and Scotland revealed by AFLPs could imply that out-crossing with other sympatric taxa is not sufficient to increase the overall genetic diversity of C. nigra against the negative effect of the successive bottlenecks during post-glacial colonization. In north-western Europe, C. nigra broadly co-exists with six taxa from the same section (C. acuta L., C. aquatilis Wahlenb., C. bigelowii Torr. ex Schwein., C. cespitosa L., C. elata All. and C. trinervis Degl.). Although all possible hybrids with C. nigra have been reported (cf. Schultze-Motel, 1968–1969; Jermy et al., 2007), they do not seem to contribute much to the total genetic variation within C. nigra. In an expanded sampling including 11 additional European and Mediterranean Carex species from sect. Phacocystis (Jiménez-Mejı́as, 2011), H9 was found to be shared with three allopatric taxa, namely C. buekii Wimm. (eastern Europe), C. randalpina B.Walln. (central Europe) and C. trinervis Degl. (western Europe). This haplotype sharing can be explained by hybridization or incomplete lineage sorting. However, the external position of H9 in the network together with the absence of H9 progenitor haplotypes in C. nigra as also in C. buekii, C. randalpina and C. trinervis (Jiménez-Mejı́as, 2011) provide a strong indication of a past hybridization event (e.g. Bänfer et al., 2006; Pleines et al., 2009). Taxonomic implications Our genetic data have several important taxonomic implications. We found no support for a taxonomic distinction between the typical creeping-rhizome ‘var. nigra’ and the tussock-forming ‘var. juncea’. The latter was formerly considered a distinct species (C. juncella; e.g. Sylvén, 1963), but lately – with rare exceptions (e.g. Egorova, 1999) – has been regarded as conspecific with C. nigra. The different growth forms are, however, maintained in cultivation, suggesting a genetic basis for these growth-form differences, although regular chromosome pairing has been observed in artificial hybrids between them (Faulkner, 1973). As the two growth forms appeared completely intermingled in our multi-locus AFLP analysis (Fig. 3c), it is therefore likely that they differ only at one or a few loci or have transcriptome-level differences not detectable by AFLP, and that ‘var. juncea’ can be considered to represent an ecotype that has originated repeatedly from different populations with creeping rhizomes. Carex nigra subsp. intricata was described from Sicily and reported from the southernmost part of the C. nigra distri2288 bution area. Our sampling covered all the mountain ranges from which C. intricata has been cited: Atlas, Corsica, Sicily and Sierra Nevada. The results clearly show that the populations from these areas are genetically heterogeneous (Figs 1b, c, 2b & 3e). In our total material, the populations from Atlas, Sicily and Sierra Nevada are genetically the most divergent, providing a basis for them to be treated as two or three separate taxa within C. nigra. However, the close morphological similarity between these populations and typical C. nigra s. str., as well as our detection of introgression among the various genetic groups, suggests that further studies must be carried out before taxonomic decisions are made. In particular, additional morphological comparisons are necessary to re-evaluate taxonomic boundaries in the circum-Mediterranean material before assigning any taxonomic rank. ACKNOWLEDGEMENTS The Spanish Ministry of Innovation and Science financed the first author (P.J.-M.) at the National Centre for Biosystematics at the University of Oslo through a pre-doctoral stay grant within the Formación de Profesorado Universitario (FPU) program. Laboratory expenses were covered by projects supported by the Spanish Ministry of Science (CGL200909972), the Regional Andalusian Government (P06RMM-02148) and a National Centre for Biosystematics (NCB) research visitor grant at the Natural History Museum, University of Oslo. The authors thank M. Mı́guez, F.J. Fernández, V. Mirré, M. Pimentel and R. Piñeiro for technical support and helpful assistance; A.J. Chaparro, M. Escudero and S. Martı́n-Bravo from Pablo de Olavide University for collecting C. nigra; M. Amini-Rad, J. Dragon, K.I. Flatberg, G. Konechnaya, P. Volkova and K.B. Westergaard for providing important collections; and the curator of the Edinburgh Herbarium (E) for letting us study the materials on loan and granting DNA extraction permission. REFERENCES Abbott, R.J. & Brochmann, C. (2003) History and evolution of the arctic flora: in the footsteps of Eric Hultén. Molecular Ecology, 11, 299–313. Allessio Leck, M. & Schütz, W. (2005) Regeneration of Cyperaceae, with particular reference to seed ecology and seed banks. Perspectives in Plant Ecology, Evolution and Systematics, 7, 95–133. Bänfer, G., Moog, U., Fiala, B., Mohamed, M., Weising, K. & Blattner, F.R. (2006) A chloroplast genealogy of myrmecophytic Macaranga species (Euphorbiaceae) in Southeast Asia reveals hybridization, vicariance and long-distance dispersals. Molecular Ecology, 15, 4409–4424. Bonin, A., Bellemain, E., Bronken Eidesen, P., Pompanon, F., Brochmann, C. & Taberlet, P. (2004) How to track and assess genotyping errors in population genetics studies. Molecular Ecology, 13, 3261–3273. Journal of Biogeography 39, 2279–2291 ª 2012 Blackwell Publishing Ltd Phylogeography of Carex nigra Brummitt, R.K. (2001) World geographical scheme for recording plant distributions, 2nd edn. Hunt Institute for Botanical Documentation, Carnegie Mellon University, Pittsburgh, PA. Cayouette, J. & Morisset, P. (1986) Chromosome studies on Carex paleacea Wahl., C. nigra (L.) Reichard, and C. aquatilis Wahl. in northeastern North America. Cytologia (Tokyo), 51, 857–884. Chater, A.O. (1980) Carex L. Flora Europaea, Vol. 5 (ed. by T.G. Tutin, V.H. Heywood, N.A. Burges, D.H. Valentine, S.M. Walters and D.A. Webb), pp. 290–323. Cambridge University Press, Cambridge, UK. Clement, M., Posada, D. & Crandall, K.A. (2000) TCS: a computer program to estimate gene genealogies. Molecular Ecology, 9, 1657–1659. Comes, H.P. & Kadereit, J.W. (1998) The effect of Quaternary climatic changes on plant distribution and evolution. Trends in Plant Science, 3, 432–438. Dragon, J.A. & Barrington, D.S. (2008) East vs. west: monophyletic clades within the paraphyletic Carex acuta complex, section Phacocystis (Cyperaceae). Sedges: uses, diversity, and systematics of the Cyperaceae (ed. by R.F.C. Naczi and B.A. Ford), pp. 215–226. Monographs in Systematic Botany, 108. Missouri Botanical Garden Press, St Louis, MO. Dragon, J.A. & Barrington, D.S. (2009) Systematics of the Carex aquatilis and C. lenticularis lineages: geographically and ecologically divergent sister clades of Carex section Phacocystis (Cyperaceae). American Journal of Botany, 96, 1896–1906. Egorova, T.V. (1999) The sedges (Carex L.) of Russia and adjacent states. Missouri Botanical Garden Press, St Louis, MO. Ehrich, D. (2006) aflpdat: a collection of R functions for convenient handling of AFLP data. Molecular Ecology Notes, 6, 603–604. Ehrich, D., Alsos, I.G. & Brochmann, C. (2008) Where did the northern peatland species survive the dry glacials: cloudberry (Rubus chamaemorus) as an example. Journal of Biogeography, 35, 801–814. Eidesen, P.B., Alsos, I.G., Popp, M., Stensrud, Ø., Suda, J. & Brochmann, C. (2007) Nuclear vs. plastid data: complex Pleistocene history of a circumpolar key species. Molecular Ecology, 16, 3902–3925. Escudero, M. & Luceño, M. (2009) Systematics and evolution of Carex sects. Spirostachyae and Elatae (Cyperaceae). Plant Systematics and Evolution, 279, 163–189. Excoffier, L. & Lischer, H.E.L. (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. Faulkner, J.S. (1973) Experimental hybridization of north-west European species in Carex section Acutae (Cyperaceae). Botanical Journal of the Linnean Society, 67, 233–253. Gaudeul, M., Taberlet, P. & Till-Bottraud, I. (2000) Genetic diversity in an endangered alpine plant, Eryngium alpinum L. (Apiaceae), inferred from amplified fragment length polymorphism markers. Molecular Ecology, 9, 1625–1637. Journal of Biogeography 39, 2279–2291 ª 2012 Blackwell Publishing Ltd Guindon, S. & Gascuel, O. (2003) A simple, fast, and accurate algorithm to estimate large phylogenies by maximum likelihood. Systematic Biology, 52, 696–704. Hampe, A. & Petit, R.J. (2005) Conserving biodiversity under climate change: the rear edge matters. Ecology Letters, 8, 461–467. Hamrick, J.L. & Godt, M.J.W. (1990) Allozyme diversity in plant species. Plant population genetics, breeding, and genetic resources (ed. by A.H.D. Brown, M.T. Clegg, A.L. Kahler, and B.S. Weir), pp. 43–63. Sinauer, Sunderland, MA. Hewitt, G.M. (1996) Some genetic consequences of ice ages, and their role in divergence and speciation. Biological Journal of the Linnean Society, 58, 247–276. Hewitt, G.M. (1999) Post-glacial re-colonization of European biota. Biological Journal of the Linnean Society, 68, 87–112. Hewitt, G.M. (2001) Speciation, hybrid zones and phylogeography – or seeing genes in space and time. Molecular Ecology, 10, 537–549. Huelsenbeck, J.P. & Ronquist, F. (2001) MRBAYES: Bayesian inference of phylogenetic trees. Bioinformatics, 17, 754–755. Hultén, E. (1958) The amphi-Atlantic plants and their phytogeographical connections. Kungliga Svenska Vetenskapsakademien Handlingar, series 4, 7. Almquist and Wiksell, Stockholm. Huson, D.H. & Bryant, D. (2006) Application of phylogenetic networks in evolutionary studies. Molecular Biology and Evolution, 23, 254–267. Jermy, A.C., Simpson, D.A., Foley, M.J.Y. & Porter, M.S. (2007) Sedges of the British Isles, 3rd edn. Botanical Society of the British Isles, London. Jiménez-Mejı́as, P. (2011) Taxonomı́a y sistemática de las secciones Ceratocystis y Phacocystis del género Carex en Europa y la cuenca mediterránea. PhD Thesis, Pablo de Olavide University, Seville. Jiménez-Mejı́as, P., Escudero, M., Guerra-Cárdenas, S., Lye, K.A. & Luceño, M. (2011) Taxonomic delimitation and drivers of speciation in the Ibero-North African Carex sect. Phacocystis river-shore group (Cyperaceae). American Journal of Botany, 98, 1855–1867. Johannesson, K. & André, C. (2006) Life on the margin: genetic isolation and diversity loss in a peripheral marine ecosystem, the Baltic Sea. Molecular Ecology, 15, 2013–2029. Jonsson, B.O. & Prentice, H.C. (2000) Allozyme diversity and geographic variation in the widespread coastal sedge, Carex arenaria. Diversity and Distributions, 6, 65–80. Kadereit, J.W., Griebeler, E.M. & Comes, H.P. (2004) Quaternary diversification in European alpine plants: pattern and process. Philosophical Transactions of the Royal Society B: Biological Sciences, 359, 265–274. Konnert, M. & Bergmann, F. (1995) The geographical distribution of genetic variation of silver fir (Abies alba, Pinaceae) in relation to its migration history. Plant Systematics and Evolution, 196, 19–30. Kropf, M., Comes, H.P. & Kadereit, J.W. (2006) Long-distance dispersal vs vicariance: the origin and genetic diversity of 2289 P. Jiménez-Mejı́as et al. alpine plants in the Spanish Sierra Nevada. New Phytologist, 172, 169–184. Lönn, M. & Prentice, H.C. (2002) Gene diversity and demographic turnover in central and peripheral populations of the perennial herb Gypsophila fastigiata. Oikos, 99, 489–498. Magri, D., Vendramin, G.G., Comps, B., Dupanloup, I., Geburek, T., Gömöry, D., Latałowa, M., Litt, T., Paule, L., Roure, J.M., Tantau, I., van der Knaap, W.O., Petit, R.J. & de Beaulieu, J.-L. (2006) A new scenario for the Quaternary history of European beech populations: palaeobotanical evidence and genetic consequences. New Phytologist, 171, 199–221. Maire, R.C.J.E. (1957) Flora de l’Afrique du Nord, Vol. 4. Paul Lechevalier, Paris. Martı́n-Bravo, S., Valcárcel, V., Vargas, P. & Luceño, M. (2010) Geographical speciation related to Pleistocene range shifts in the western Mediterranean mountains (Reseda sect. Glaucoreseda, Resedaceae). Taxon, 59, 466–482. Mueller, M.H. & van der Valk, A.G. (2002) The potential role of ducks in wetland seed dispersal. Wetlands, 22, 170– 178. Müller, K. (2005) SeqState: primer design and sequence statistics for phylogenetic DNA data sets. Applied Bioinformatics, 4, 65–69. Nakamatte, E. & Lye, K.A. (2007) AFLP-based differentiation in north Atlantic species of Carex sect. Phacocystis. Nordic Journal of Botany, 25, 318–328. Nei, M. (1973) Analysis of gene diversity in subdivided populations. Proceedings of the National Academy of Sciences USA, 70, 3321–3323. Nei, M. (1978) Estimation of average heterozygosity and genetic distance from a small number of individuals. Genetics, 89, 583–590. Nilsson, Ö. (1985) Carex. Flora of Turkey, Vol. 9 (ed. by P. Davis), pp. 73–158. Edinburgh University Press, Edinburgh. Noltie, H.J. (1994) Flora of Bhutan, Vol. 3. Royal Botanical Garden, Edinburgh. Nylander, J.A.A. (2004) MrAIC.pl. Program distributed by the author. Evolutionary Biology Centre, Uppsala University, Uppsala. O’Brien, D.L. & Freshwater, D.W. (1999) Genetic diversity within tall form Spartina alterniflora Loisel. along the Atlantic and Gulf coasts of the United States. Wetlands, 19, 352–358. Palmé, A.E., Su, Q., Rautenberg, A., Manni, F. & Lascoux, M. (2003) Postglacial recolonization and cpDNA variation of silver birch, Betula pendula. Molecular Ecology, 12, 201–212. Petit, R.J., Aguinagalde, I., de Beaulieu, J.-L., Bittkau, C., Brewer, S., Cheddadi, R., Ennos, R., Fineschi, S., Grivet, D., Lascoux, M., Mohanty, A., Müller-Starck, G., DemesureMusch, B., Palmé, A., Martı́n, J.P., Rendell, S. & Vendramin, G.G. (2003) Glacial refugia: hotspots but not melting pots of genetic diversity. Science, 300, 1563–1565. Pignatti, S. (1982) Flora d’Italia, Vol. 3. Edagricole, Bologna. 2290 Pleines, T., Jakob, S.S. & Blattner, F.R. (2009) Application of non-coding DNA regions in intraspecific analyses. Plant Systematics and Evolution, 282, 281–294. Pritchard, J.K., Stephens, M. & Donnelly, P. (2000) Inference of population structure using multilocus genotype data. Genetics, 155, 945–959. Puşcaş, M., Choler, P., Tribsch, A., Gielly, L., Rioux, D., Gaudeul, M. & Taberlet, P. (2008) Postglacial history of the dominant alpine sedge Carex curvula in the European Alpine System inferred from nuclear and chloroplast markers. Molecular Ecology, 17, 2417–2429. Rejzková, E., Fér, T., Vojta, J. & Marhold, K. (2008) Phylogeography of the forest herb Carex pilosa (Cyperaceae). Botanical Journal of the Linnean Society, 158, 115–130. Roalson, E.H. (2008) A synopsis of chromosome number variation in the Cyperaceae. Botanical Review, 74, 209–393. Rohlf, F.J. (1997) NTSYS-pc: numerical taxonomy and multivariate analysis system, version 2.0. Exeter Software, Setauket, NY. Ronquist, F. & Huelsenbeck, J.P. (2003) MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics, 19, 1572–1574. Schaal, B.A., Hayworth, D.A., Olsen, K.M., Rauscher, J.T. & Smith, W.A. (1998) Phylogeographic studies in plants: problems and prospects. Molecular Ecology, 7, 465–474. Schmid, B. (1986) Colonizing plants with persistent seeds and persistent seedlings (Carex flava group). Botanica Helvetica, 96, 19–26. Schönswetter, P. & Tribsch, A. (2005) Vicariance and dispersal in the alpine perennial Bupleurum stellatum L. (Apiaceae). Taxon, 54, 725–732. Schönswetter, P., Paun, O., Tribsch, A. & Niklfeld, H. (2003) Out of the Alps: colonization of Northern Europe by East Alpine populations of the Glacier Buttercup Ranunculus glacialis L. (Ranunculaceae). Molecular Ecology, 12, 3373–3381. Schönswetter, P., Popp, M. & Brochmann, C. (2006) Central Asian origin of and strong genetic differentiation among populations of the rare and disjunct Carex atrofusca (Cyperaceae) in the Alps. Journal of Biogeography, 33, 948–956. Schönswetter, P., Elven, R. & Brochmann, C. (2008) TransAtlantic dispersal and large-scale lack of genetic structure in the circumpolar, arctic–alpine sedge Carex bigelowii s.l. (Cyperaceae). American Journal of Botany, 95, 1006–1014. Schultze-Motel, W. (1968–1969) Carex L. Illustrierte Flora von Mittel-Europa, Vol. 2 (ed. by G. Hegi, H.J. Conert, U. Hamann, W. Schultze-Motel and G. Wagenitz), pp. 96–274. Verlag Paul Parey, Berlin. Shaw, J., Lickey, E.B., Beck, J.T., Farmer, S.B., Liu, W., Miller, J., Siripun, K.C., Winder, C.T., Schilling, E.E. & Small, R.L. (2005) The tortoise and the hare II: relative utility of 21 noncoding chloroplast DNA sequences for phylogenetic analysis. American Journal of Botany, 92, 142–166. Shaw, J., Lickey, E.B., Schilling, E.E. & Small, R.L. (2007) Comparison of whole chloroplast genome sequences to choose noncoding regions for phylogenetic studies in Journal of Biogeography 39, 2279–2291 ª 2012 Blackwell Publishing Ltd Phylogeography of Carex nigra angiosperms: the tortoise and the hare III. American Journal of Botany, 94, 275–288. Simmons, M.P. & Ochoterena, H. (2000) Gaps as characters in sequence-based phylogenetic analyses. Systematic Biology, 49, 369–381. Standley, L.A., Cayouette, J. & Bruederle, L. (2002) Carex sect. Phacocystis Dumort. Flora of North America north of Mexico, Vol. 23 (ed. by P.W. Ball and A.A. Reznicek), pp. 379–401. Oxford University Press, New York, NY. Stenström, A., Jonsson, B.O., Jónsdóttir, I.S., Fagerström, T. & Augner, M. (2001) Genetic variation and clonal diversity in four clonal sedges (Carex) along the Arctic coast of Eurasia. Molecular Ecology, 10, 497–513. Swofford, D.L. (2002) PAUP*: phylogenetic analysis using parsimony (and other methods), 4.0 beta. Sinauer Associates, Sunderland, MA. Sylvén, N. (1963) The Carices Distigmaticae of the Scandinavian flora district. Opera Botanica, 8, 1–161. Taberlet, P., Fumagalli, L., Wust-Saucy, A.-G. & Cosson, J.-F. (1998) Comparative phylogeography and postglacial colonization routes in Europe. Molecular Ecology, 7, 453–464. Tribsch, A., Schönswetter, P. & Stuessy, T.F. (2002) Saponaria pumila (Caryophyllaceae) and the Ice Age in the European Alps. American Journal of Botany, 89, 2024–2033. Tyler, T. (2002a) Large-scale geographic patterns of genetic variation in Melica nutans, a widespread Eurasian woodland grass. Plant Systematics and Evolution, 236, 73–87. Tyler, T. (2002b) Geographical distribution of allozyme variation in relation to post-glacial history in Carex digitata, a widespread European woodland sedge. Journal of Biogeography, 29, 919–930. Vicioso, C. (1959) Estudio monográfico sobre el género Carex en España. Boletı́n del Instituto Forestal de Investigaciones y Experiencias, 79, 1–205. Westergaard, K.B., Alsos, I.G., Ehrich, D., Eidesen, P.B., Hollingsworth, P.M. & Brochmann, C. (2008) Genetic diversity and distinctiveness in Scottish alpine plants. Plant Ecology & Diversity, 1, 329–338. Westergaard, K.B., Alsos, I.G., Engelskjøn, T., Flatberg, K.I. & Brochmann, C. (2011a) Trans-Atlantic genetic uniformity in the rare snowbed sedge Carex rufina. Conservation Genetics, 12, 1367–1371. Westergaard, K.B., Alsos, I.G., Popp, M., Engelskøn, T., Flatberg, K.I. & Brochmann, C. (2011b) Glacial survival may Journal of Biogeography 39, 2279–2291 ª 2012 Blackwell Publishing Ltd matter after all: nunatak signatures in the rare European populations of two west-arctic species. Molecular Ecology, 20, 376–393. Willis, K.J., Rudner, E. & Sümegi, P. (2000) The full-glacial forests of central and southeastern Europe. Quaternary Research, 53, 203–213. Wróblewska, A. & Brzosko, E. (2006) The genetic structure of the steppe plant Iris aphylla L. at the northern limit of its geographical range. Botanical Journal of the Linnean Society, 152, 245–255. Zhang, L.-B., Comes, H.P. & Kadereit, J.W. (2001) Phylogeny and Quaternary history of the European montane/alpine endemic Soldanella (Primulaceae) based on ITS and AFLP variation. American Journal of Botany, 88, 2331–2345. SUPPORTING INFORMATION Additional Supporting Information may be found in the online version of this article: Appendix S1 Studied populations and associated data. As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer-reviewed and may be re-organized for online delivery, but are not copy-edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors. BIOSKETCH The authors are interested in the phylogeography, evolution and taxonomy of boreo-temperate plants, and especially in inferring patterns of post-glacial dispersal and speciation. Author contributions: P.J.M., M.L. and K.A.L. conceived the ideas and collected the materials; P.J.M., G.G. and C.B. analysed and interpreted the data; all authors contributed to the writing. Editor: Christine Maggs 2291