Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Three in One—Multiple Faunal Elements within an Endangered European Butterfly Species

  • Marius Junker,

    Affiliation Department of Biogeography, Trier University, Trier, Germany

  • Marie Zimmermann,

    Affiliation Université de Tours, CNRS, UMR 6035 –IRBI, Avenue Monge, Parc Grandmont, Tours, France

  • Ana A. Ramos,

    Affiliation Centre of Marine Sciences, CCMAR/CIMAR Associate Laboratory, University of Algarve, Gambelas, Faro, Portugal

  • Patrick Gros,

    Affiliation Haus der Natur, Museum für Natur und Technik, Museumsplatz 5, Salzburg, Austria

  • Martin Konvička,

    Affiliation School of Biological Sciences, University South Bohemia, Branisovska 31, Ceske Budejovice, Czech Republic

  • Gabriel Nève,

    Affiliation Institut Méditerranéen de Biodiversité et d’Ecologie marine et continentale, Aix Marseille Université, CNRS, IRD, Avignon Université, Case 36, 3 place Victor Hugo, Marseille Cedex 3, France

  • László Rákosy,

    Affiliation Faculty of Biology, University Babes-Bolyai, Str. Clinicilor 5–7, Cluj, Romania

  • Toomas Tammaru,

    Affiliation Institute of Ecology and Earth Sciences, University of Tartu, Vanemuise 46, Tartu, Estonia

  • Rita Castilho ,

    Contributed equally to this work with: Rita Castilho, Thomas Schmitt

    Affiliation Centre of Marine Sciences, CCMAR/CIMAR Associate Laboratory, University of Algarve, Gambelas, Faro, Portugal

  • Thomas Schmitt

    Contributed equally to this work with: Rita Castilho, Thomas Schmitt

    thomas.schmitt@senckenberg.de

    Affiliations Department of Biogeography, Trier University, Trier, Germany, Senckenberg German Entomological Institute, Eberswalder Straße 90, Müncheberg, Germany, Zoology, Institute of Biology, Faculty Natural Science I, Martin Luther University Halle-Wittenberg, Halle (Saale), Germany

Three in One—Multiple Faunal Elements within an Endangered European Butterfly Species

  • Marius Junker, 
  • Marie Zimmermann, 
  • Ana A. Ramos, 
  • Patrick Gros, 
  • Martin Konvička, 
  • Gabriel Nève, 
  • László Rákosy, 
  • Toomas Tammaru, 
  • Rita Castilho, 
  • Thomas Schmitt
PLOS
x

Abstract

Ice ages within Europe forced many species to retreat to refugia, of which three major biogeographic basic types can be distinguished: "Mediterranean", "Continental" and "Alpine / Arctic" species. However, this classification often fails to explain the complex phylogeography of European species with a wide range of latitudinal and altitudinal distribution. Hence, we tested for the possibility that all three mentioned faunal elements are represented within one species. Our data was obtained by scoring 1,307 Euphydryas aurinia individuals (46 European locations) for 17 allozyme loci, and sequencing a subset of 492 individuals (21 sites) for a 626 base pairs COI fragment. Genetic diversity indices, F statistics, hierarchical analyses of molecular variance, individual-based clustering, and networks were used to explore the phylogeographic patterns. The COI fragment represented 18 haplotypes showing a strong geographic structure. All but one allozyme loci analysed were polymorphic with a mean FST of 0.20, supporting a pronounced among population structure. Interpretation of both genetic marker systems, using several analytical tools, calls for the recognition of twelve genetic groups. These analyses consistently distinguished different groups in Iberia (2), Italy, Provence, Alps (3), Slovenia, Carpathian Basin, the lowlands of West and Central Europe as well as Estonia, often with considerable additional substructures. The genetic data strongly support the hypothesis that E. aurinia survived the last glaciation in Mediterranean, extra-Mediterranean and perialpine refugia. It is thus a rare example of a model organism that combines attributes of faunal elements from all three of these sources. The observed differences between allozymes and mtDNA most likely result from recent introgression of mtDNA into nuclear allozyme groups. Our results indicate discrepancies with the morphologically-based subspecies models, underlining the need to revise the current taxonomy.

Introduction

Climatic oscillations with ice ages and interglacial periods have had strong impacts on the distribution of European animal and plant species (e.g. [1,2]). In this context, glacial refugia with suitable habitat conditions were pivotal for survival during glacial cycles [3]. Considering the geographic location of these refugia and the respective postglacial range changes, molecular analyses show that European species may be divided into three major biogeographic types: Mediterranean, continental and arctic/alpine [2]:

  1. “Mediterranean” species usually had important glacial differentiation centres at least in one of the three major Mediterranean peninsulas, but also in the Maghreb and Asia Minor [4,5]. However, further substructures in these refugia have been postulated since the 1950s [6], and recent phylogeographic analyses support these old postulates well [2].
  2. “Continental” species are supposed to have survived during ice ages in extra-Mediterranean refugia [7] with more buffered climatic conditions (e.g. the Carpathian Basin) [8].
  3. “Arctic” and/or “Alpine” species often survived ice ages in perialpine refugia and retreated to alpine and/or arctic areas when temperatures increased after their inter-/postglacial deglaciation [9,10].

This classification is simplistic and often fails to explain the rather complex phylogeography of European species with a currently wide range of latitudinal and altitudinal distribution. Thus, a number of continental elements also had important glacial refugia in the Balkan Peninsula (e.g. several butterfly species [11,12]; the adder Vipera berus [13]; the slug Arion fuscus [14]). Mediterranean and extra-Mediterranean refugia existed in close geographic proximity in the Balkan Peninsula so that these continental species may have occurred in extra-Mediterranean refugia of this region [8]. Several alpine species apparently have co-occurred with continental elements in several ice age retreats in the vicinity of high mountain systems [8,10]. Furthermore, Mediterranean elements turned out to have had more diverse ice age refugia than previously thought, and many examples demonstrate extra-Mediterranean refugia in addition to the classical Mediterranean retreats [8,15]. This underlines the great biogeographic importance of the cryptic extra-Mediterranean refugia that although geographically smaller in size than the Mediterranean refugia are a reservoir of genetic variation and often play a leading edge colonization function [16,17]. Various species combine two of the biogeographic patterns mentioned above [8], but the most complex possible combination of Mediterranean, extra-Mediterranean and perialpine refugia has not been observed so far.

Therefore, we selected a particularly widespread and ecologically diverse species as study object, the fritillary Euphydryas aurinia (Rottemburg, 1775) (Nymphalidae: Melitaeinae). This species is a univoltine butterfly found from Mediterranean shrub-lands at sea level to alpine meadows, and from the European Atlantic coast throughout temperate Asia [18]. It colonizes a great variety of different habitats, e.g. open Quercus woodlands in the Iberian Peninsula (E. aurinia beckeri (Herrich-Schäffer, 1851)) [19,20], dry or damp calcareous or acidophilic grasslands and mires in Central Europe and the UK (E. aurinia aurinia Rottemburg, 1775) [2124] as well as nutrient-poor alpine meadows in the Pyrenees (E. aurinia debilis Oberthür, 1909) and the Alps (E. aurinia glaciegenita Verity, 1928) [18,25,26]. Furthermore, different larval food plants are used regionally, including the genera Lonicera, Succisa, Scabiosa and Gentiana [27,28]. Due to this wide variation of ecological constraints and morphological adaptations, many subspecies have been described in the Palaearctic region[29]; recent reviews mention 12 subspecies for the western Palaearctic [30], with six subspecies for France alone [28]

Euphydryas aurinia is highly threatened in Central Europe due to dramatic damage of its habitats during the last few decades [3133] and hence was listed in Annex II of the European Habitats Directive. Since then, important efforts have been made to collect knowledge that will aid conservation, especially of the populations in western and northern Europe [23,3436]. By contrast, populations in Iberia and southern France are considered mostly stable [28,37].

Due to these strong ecological adaptations, it seemed possible that E. aurinia may represent, within a single species, all three major biogeographic groups known for Europe (cf. [2]). By applying mitochondrial and nuclear markers to populations sampled over most of the European range of E. aurinia, we analyse the differentiation patterns within this species; hereby we aim at identifying the location of glacial refugia and the postglacial range changes.

Material and Methods

Sampling

We sampled 1,307 E. aurinia individuals (46 populations) distributed over major parts of Europe for allozyme electrophoresis (Table 1; Fig 1). One sample of Euphydryas desfontainii (30 individuals) from southern Portugal was included as outgroup (sister species of E. aurinia [38]). The sample sizes ranged from 5 to 44 individuals (mean 28.4 ± 10.3 SD); only eight samples contained less than 20 individuals. After netting in the field, butterflies were immediately frozen in liquid nitrogen and stored in this medium until analysis. 492 of these individuals and from two additional sites were used for mtDNA sequencing.

thumbnail
Fig 1. Localities of the 48 sampling sites of Euphydryas aurinia and one sampling site of Euphydryas desfontainii (DP1, outgroup) for genetic analyses.

For abbreviations see Table 1.

https://doi.org/10.1371/journal.pone.0142282.g001

thumbnail
Table 1. Sampling locations of Euphydryas aurinia and E. desfontainii for mtDNA and allozyme analyses; presented in alphabetic order of countries.

The mentioned subspecies are those recognised by Tshikolovets [30].

https://doi.org/10.1371/journal.pone.0142282.t001

Specific permission for this study were granted by: Instituto da Conservacao da Natureza Lissabon for Portugal, Nature Protection Agency of the Province of Extremadura, Nature Protection Agency of the Province of Aragon, Nature Protection Agency of the Province of Catalonia, Directory of the National Parc of Majella (Italy), Naturschutzbehörde des Landes Salzburg, Obere Naturschutzbehörde des Saarlandes, Ministère de l'Environnement (Paris). Several samples were provided by local collectors or were collected together with them. No rights on private land were violated.

Laboratory procedures

Total genomic DNA was extracted from butterfly legs using an adapted Chelex-100 (Bio-Rad) protocol [39]. A partial fragment (626 bp) of the mitochondrial cytochrome c oxidase subunit I (COI) gene was amplified by PCR using the universal primers HCO2198 and LCO1490 [40]. This region of the COI gene is the most variable according to previously published work [41]. Amplifications were performed in 25 μl reactions containing 1x colorless GoTaq Flexi Buffer, 2.0 mM MgCl2, 0.2 mM of each dNTP, 0.4 μM of each primer, 1U GoTaq Flexi DNA polymerase (Promega) and 50 ng template DNA. PCR cycling conditions were as follows: initial denaturation of 5 min at 95°C; followed by 40 cycles of 1 min at 95°C, 1 min at 51°C, 1 min at 72°C; and a final extension step of 10 min at 72°C. PCR products were purified by ethanol/sodium acetate precipitation. Sequencing was performed with the primer LCO1490 on an ABI 3130xl automated sequencer (Applied Biosystems—CCMAR, Portugal). Sequences were checked, manually edited and aligned using Geneious ver. 5.4 (http://www.geneious.com/). All sequences were deposited in GenBank (Accession Numbers: KT896753—KT897244).

For allozyme electrophoresis of 1,307 E. aurinia individuals, the whole abdomen of each imago was homogenized with ultrasound in Pgm-buffer [42] and centrifuged 3 min at 10,000 g. We used cellulose acetate plates, applying standard protocols [43,44]. A total of 17 allozyme loci were analysed (S3 Table).

Statistical analyses

Information on mtDNA haplotype diversity, nucleotide diversity and frequency of each haplotype was extracted using DnaSP 5.10 [45] and Arlequin 3.5 [46]. The best model of nucleotide substitution was determined using jModelTest [47]. According to the Akaike Information Criteria (AIC) [48], the Tamura-Nei (I = 0.9270) [49] model fitted the data best (AIC 3898.0546). This model was used in further analyses as appropriate. Haplotype networks based on COI mtDNA data to depict relationships among haplotypes were performed with NETWORK v4.6 ([50], available at fluxus-engineering.com). Median-joining networks [50] that contained all possible equally short trees were simplified by running the maximum parsimony calculation option to eliminate superfluous nodes and links [51]. Pairwise location estimates of genetic differentiation plus 95% bootstrapped confidence intervals (10,000 replicates) were estimated using Weir and Cockerham’s [52] FST estimator and Jost’s [53] Dest. Calculations were executed in the Diversity R package [54].

Allozyme alleles were labelled according to their relative mobility during electrophoresis. We used G-stat [55] to compute allele frequencies and parameters of allozyme genetic diversity (i.e. mean number of alleles per locus A, expected heterozygosity He, observed heterozygosity Ho, total percentage of polymorphic loci Ptot and percentage of polymorphic loci with the most common allele not exceeding 95% P95). We calculated allozyme allelic richness (AR) with Fstat [56] to allow for the different sample sizes from different locations. Populations FR3 and IT1 were excluded from this analysis because they were represented by too few individuals (N < 7). Allozyme locus-by-locus analyses of molecular variance (Amova), hierarchical genetic variance analyses, test of Hardy-Weinberg equilibrium and linkage disequilibrium were performed with Arlequin 3.5 [46]. Nei´s standard genetic distances [57], neighbour-joining phenograms [58] and bootstraps based on 1,000 interactions were calculated with Phylip 3.5.c [59].

Spatial genetic structure was assessed using two different approaches. First, we used the Spatial Analysis of Shared Alleles (SAShA) method [60] to establish non-panmixia on both mtDNA and allozyme data sets. This method is based on the premise that if groups are evolving locally then the same alleles or haplotypes are expected to co-occur in the same location more often than expected by chance. Second, a bayesian clustering algorithm of population assignment implemented in the R package Geneland 2.0 [61] produced a map that consolidated genetic and geographic data. To determine the number of genetic clusters, independent runs were implemented using 1,000,000 MCMC iterations with a burn-in period of 100 and a thinning value of 1,000. The value of K was set from 1 to 21 clusters on a correlated frequency model. We inferred the number of clusters from the modal value of K with the highest posterior probability.

Results

Mitochondrial DNA

We sequenced 492 individuals of E. aurinia from 21 sites in Europe for 626 bp of the COI gene. No pseudogenes were present as evidenced by the absence of stop codons, the prevalence of synonymous substitutions or low pairwise divergence. A total of 18 haplotypes were identified, including two non-synonymous and 17 parsimony informative sites (Fig 2). The limited numbers of nucleotide substitutions separating haplotypes suggest that the haplotypes are closely related and shared haplotypes were the most frequently detected haplotypes. Mean haplotype diversity was low (0.147 ± 0.038 SD) with haplotype numbers ranging from one to three in each location, and average nucleotide diversity was also low (0.038% ± 0.012% SD). Overall haplotype and nucleotide diversities were low (0.757 ± 0.014 SD and 0.34% ± 0.01% SD, respectively).

thumbnail
Fig 2. Median Joining Network indicating the relationship among COI haplotypes of Euphydryas aurinia.

Each pie chart represents a different haplotype, made up of collection sites labelled by colour in which that haplotype occurs. Haplotypes connected by a line differ in sequence by one base pair unless otherwise indicated. The size of each pie chart is proportional to the relative frequency of the haplotype within the entire sample. The size of the collection site slices is influenced by both the frequency of the haplotype across the sites and the sample size for each site. For code abbreviations see Table 1.

https://doi.org/10.1371/journal.pone.0142282.g002

The spatial analysis of shared alleles found that the arrangement of COI haplotypes was not equidistributed and was statistically different from that expected under panmixia (observed mean 678 km, expected mean 1,146 km, p < 0.0001) (Fig 3A). Pairwise differentiation was significant in 174 pairwise comparisons (92%) when measured using Dest and in 159 (84%) when using FST (S1 Table).

thumbnail
Fig 3. Spatial analysis of shared COI mtDNA haplotypes (A) and allozyme alleles (B) distribution of Euphydryas aurinia.

Histograms represent the frequency of alleles between locations distance classes. Expected means and significance value were calculated with 1,000 randomized permutations of the data set. Vertical lines represent the mean of frequencies. Triangles and circles are the cumulative frequency of alleles at increasing distance. P value is the likelihood that the observed mean is greater than the expected.

https://doi.org/10.1371/journal.pone.0142282.g003

Geneland analyses detected nine major clusters (mt1 –mt9) that correspond to Iberia (mt1), Pyrenees (mt2), Provence (mt3), Alps (including Central Apennines) (mt4), Brittany (mt5), Europe north of the Alps and the Pyrenees including Northern Europe (mt6), northern Apennines (mt7), Slovenia (mt8) and eastern Carpathian Basin (mt9) (Fig 4A). However, Dest and FST pairwise comparisons between clusters mt4 and mt6 were not significant (S2 Table).

thumbnail
Fig 4. Spatial Bayesian clustering of Euphydryas aurinia.

A. Mitochondrial DNA sequences, circles represent both locations and haplotypes found. B. Allozyme dataset, black dots represent locations. Spatial clustering suggests 19 distinct clusters across geographical regions.

https://doi.org/10.1371/journal.pone.0142282.g004

Allozymes

All 17 loci analysed were polymorphic and showed banding pattern consistent with known quaternary structures [43]. The polymorphisms in Fum were restricted to the sibling E. desfontainii, as all E. aurinia populations were monomorphic for this locus. The number of alleles per locus ranged from two (Mdh1) to ten (Pgi, Pgm, PepPhe-Pro) (mean: 5.4 ± 2.7 SD). Allele frequencies can be obtained from the authors on personal request.

The mean number of alleles per locus (A) ranged from 1.41 to 2.53, with a mean of 2.00 (± 0.26 SD), while allelic richness (AR) ranged from 1.30 to 1.76 with a mean of 1.51 (± 0.10 SD) (Table 2). Excluding populations with respectively less than ten and 20 individuals yielded results between A and AR (AR10 mean: 1.60 ± 0.11 SD; AR20 mean: 1.90 ± 0.17 SD). The highest values for allelic richness were found in populations from Provence, the western Alps and the Carpathian Basin. For the polymorphic loci with the most common allele not exceeding 95% (P95), frequencies ranged from 23.5% to 52.9%, mean: 42.1% (± 8.5% SD), while the total percentage of polymorphic loci (Ptot) ranged from 35.3% to 82.4% with a mean of 57.9% (± 10.2% SD). The mean expected heterozygosity (He) was 14.8% (± 2.7% SD), ranging from 9.1% to 22.9%, and the mean observed heterozygosity (Ho) was estimated at 14.6% (± 2.6% SD), varying from 9.5% to 22.4%. The genetic diversity of the outgroup population E. desfontainii was considerably lower than the average of E. aurinia.

thumbnail
Table 2. Parameters of genetic diversity of allozymes analysed for 46 populations of Euphydryas aurinia and one population of E. desfontainii (DP1); sorting follows the twelve allozyme consensus groups al1 to al12.

https://doi.org/10.1371/journal.pone.0142282.t002

None of the loci/population combinations showed any significant deviation from Hardy-Weinberg equilibrium, after Benjamini-Hochberg procedure for multiple testing [62,63]. Therefore, we performed further analyses using standard population genetic approaches.

The total genetic variance of all 46 E. aurinia populations was 1.585 with 0.326 genetic variance among populations (FST = 0.206; p < 0.001) and 0.020 genetic variance among individuals within populations (FIS = 0.0158; p < 0.005) (Table 3). Excluding populations with less than 10 individuals did not change this result. The unbiased genetic distances [57] among all 46 samples ranged from 0.0021 to 0.1648 with a mean of 0.0502 (± 0.0300 SD). Based on these distances, a neighbour-joining dendrogam was created (Fig 5A; Fig 5C shows the rooting of the dendrogram with E. desfontainii as outgroup).

thumbnail
Fig 5. Neighbour-joining dendrogram of 46 populations of Euphydryas aurinia and one population of E. desfontainii based on genetic distances [57].

A. Dendrogram with bootstrap values >40. For population codes see Table 1. B. Differentiation into twelve groups within the E. aurinia cluster (supported by bootstrap values and mtDNA patterns revealed by Geneland). The different group names are indicated for the allozyme groups and the mitochondrial groups. Populations otherwise not visible in the phenogram are indicated by arrows. C. Genetic distance among populations of E. aurinia and E. desfontainii (DP1; outgroup). Populations used for mtDNA sequencing are given in bold in A and B.

https://doi.org/10.1371/journal.pone.0142282.g005

thumbnail
Table 3. Results of non-hierarchical variance analyses of allozyme data for different groups of Euphydryas aurinia and E. desfontainii in Europe.

https://doi.org/10.1371/journal.pone.0142282.t003

The spatial analysis of shared alleles found that the arrangement of allozyme alleles was not equidistributed and was statistically different from the expectation under panmixia (observed mean 976 km, expected mean 982 km, p = 0.001) (Fig 3B). However, this difference although being highly significant was rather limited.

Information obtained from mtDNA and allozyme allele frequencies are geographically mostly consistent and in combination supported the existence of twelve genetic groups within E. aurinia (Fig 5B): western Iberia including SW France (i.e. FR4) (al1), Pyrenees and adjoining NE Iberia (al2), Provence (al3), SW Alps (al4), western Central Alps (al5), eastern Alps (al6), western Europe (al7), Central Europe (al8), Slovenia (al9), Carpathian Basin (al10), Italy (al11) and Estonia (al12). The mean genetic diversities of these twelve groups are given in Table 4. The samples of each of these twelve groups belong to one recognised subspecies; E. aurinia beckeri: groups al1 and al2; E. aurinia provincialis: groups al3 and al11; E. aurinia glaciegenita: groups al4, al 5 and al6; E. aurinia aurinia: all other groups. The only exception is population FR4 which is within the range of E. aurinia aurinia, in the western Massif Central, France, but is grouped with E. aurinia beckeri populations in group al1.

thumbnail
Table 4. Means of the analysed genetic diversity parameters of the twelve consensus groups of Euphydryas aurinia (al1 to al12) and the outgroup E. desfontainii.

https://doi.org/10.1371/journal.pone.0142282.t004

Hierarchical variance analyses of allozyme data strongly confirm these twelve genetic groups (variance among groups: 0.304; FCT = 0.1888; p < 0.001; variance within groups: 0.049; FSC = 0.0378; p < 0.001; Table 5). Furthermore, subsequent hierarchical variance analyses based on regionalised parts of the entire data set (Table 5) supported the obtained structure among groups in the neighbour joining tree. Note that the tests presented here were mostly restricted to geographically neighbouring groups, to find whether patterns are supported or should be rejected. FST values within these groups (Table 3) are consistent with the genetic distances [57] within them. The Bayesian clustering of allozymes with GENELAND even distinguished 19 groups (best K = 19) within E. aurinia (Fig 4B); these groups were mostly congruent with the twelve groups outlined above as consensus between mtDNA and allozyme data, but revealed an even more fine-grained geographic pattern, which might however be biogeographically of limited significance in some of these cases.

thumbnail
Table 5. Results of hierarchical variance analyses of allozyme data among different groups of Euphydryas aurinia and E. desfontainii in Europe.

https://doi.org/10.1371/journal.pone.0142282.t005

Within the twelve consensus groups, the western Iberian cluster (al1) showed the strongest differentiation among populations (variance among populations: 0.083; FST = 0.0704; p < 0.001; variance among individuals within populations: 0.0037; FIS: 0.037; n.s.; Table 3). Also the Central European group (al8) (FST = 0.0543, p < 0.001) and the Italian group (al11) (FST = 0.0488, p < 0.001) showed remarkable within-group differentiation, while five other groups had low FST values ranging from not significant to 0.0246 (Table 3).

Hierarchical variance analyses showed a strong differentiation between the samples of E. aurinia and E. desfontainii (variance among groups: 0.02019; FCT = 0.562; p < 0.001; variance within groups: 0.00326; FSC = 0.207; p < 0.001; Table 5). Three taxon-specific alleles within the population of E. desfontanii were observed (Idh1: Allele 7; Fum: Allele 1 and 3). The mean genetic distance [57] between both taxa amounted 0.397 (± 0.080 SD).

Discussion

The combination of the results from both allozymes and mtDNA supports the existence of twelve genetic groups. Different groups indicate different core areas in these regions; these are located in Iberia (2), Italy, Provence, Alps (3), Slovenia, Carpathian Basin, lowlands of West and Central Europe as well as Estonia, often with considerable additional substructures. However, despite the mostly consistent genetic pattern found between allozymes and mtDNA, differences were detected in peninsular Italy, the Alps, West and Central Europe as well as Estonia. In spite of the wide geographic sampling undertaken in the present study, the main caveat of this publication is the lack of samples from the Balkan Peninsula. This prevents us clarifying the importance for E. aurinia of the Ponto-Mediterranean refugium with its putative substructures (cf. [2]). Furthermore, the precise distribution of groups and detailed information on contact zones between them are not available for all twelve genetic groups.

Genetic diversity

The FST values obtained here for E. aurinia are higher than those given previously for this species in nationwide surveys [64,65]. However our low value for the Provence area (FST = 0.0246) is much lower than that given in a previous study of this area with more samples (FST = 0.113 [65]). The parameters of genetic diversity (mtDNA and allozymes) of the E. aurinia populations are similar to many other representatives of the Nymphalidae (e.g. [6670]) and exceed the values for most relict species or species with restricted distributions (e.g. [66,71,72]). However, the genetic diversity of E. aurinia was lower than in some very common Satyrinae [7375] or Lycaenidae [70,76,77] and is also lower than in the North American mountain butterfly Parnassius smintheus [78] and in the mountain endemic Erebia palarica [71]. In general, the genetic diversity of E. aurinia thus matches the values of many moderately common and widespread butterfly species. The strong decline of E. aurinia populations in West and Central Europe during the last decades has so far not led to a remarkable loss of genetic diversity in a pan-European context.

Genetic differentiation and biogeography

A spatially explicit Bayesian clustering method (Fig 4B) detecting small amounts of genetic differentiation [79] supported 19 allozyme clades. However, just twelve groups (joining several of these 19 clades) can be seen as biogeographically informative as a consensus between nuclear and mitochondrial genetic information. This structure of twelve groups is well reflected in the neighbour joining analysis of the allozyme polymorphisms (Fig 5B). However, two major discordances (central Italy, Estonia) exist between these two marker systems; these are discussed in detail below. The degree of genetic differentiation among the twelve distinct consensus groups of E. aurinia (Fig 5B), the values of hierarchical variance analyses and the differentiation at the mtDNA level are typical of butterfly species with strong intraspecific genetic structures [67,69,77,8082]).

The clustering of the studied samples strongly suggests that recognised subspecies do not correspond to phylogenetic groups. E. aurinia provincialis should be split between its French populations (group al3) and its Italian populations (group al11). The name E. aurinia aurunca Turati, 1910 [29] may apply to the latter. Similarly, groups al4, al5 and al6 would all be in subspecies E. aurinia glaciegenita according to their distribution, while the clustering shows that al4 and al5 together belong to a different cluster than al6, which positions itself within E. aurinia aurinia groups; subspecies E. aurinia glaciegenita therefore does not correspond to a single cluster. Its eastern component, group al6 from high altitude Austria, may have had genetic exchange with lower altitude populations, without substantial changes in its phenotype (i.e. small size and dark colouring) and ecological requirements (i.e. it feeds on Gentiana), which are under strong selection in Alpine habitats. Our results clearly show that the Alpine populations sometimes known as Euphydryas glaciegenita belong to Euphydryas aurinia, and should not be given specific status [30].

The high degree of genetic differentiation of E. aurinia within Europe argues for the existence of several refugia during the last glaciation, if not before this. A similar assumption was made by Varga [83] who considered E. aurinia a polycentric species with holo-Palaearctic distribution. However, this author provided no details about the number and locations of refugia.

The three major Mediterranean peninsulas represented important refugia during the last ice age for many different animal and plant species [1,8486] and might have been of importance also for E. aurinia. Our analyses revealed independent genetic groups in south-western Europe (al1, al2) and Italy (al11) which respectively indicate an atlanto-Mediterranean and an adriato-Mediterranean glacial refugium for this species. There might also be several subrefugia in Iberia and Central Italy (Fig 6). The genetic structure of the neighbour joining tree based on allozyme data and the mtDNA haplotype network support the survival of E. aurinia in at least two subrefugia in western Iberia (al1) and south of the Pyrenees (al2) (Fig 6) during the last ice age. Such an East-West discontinuity has repeatedly been observed for Iberia [8791]. Further, more subtle regional substructuring of the western Iberian group (al1) during the last ice age is conceivable based on the remarkable allozyme structure throughout this region. Such refugia-within-refugia structures have been repeatedly demonstrated in Iberia for different taxonomic groups [9296]. Additionally, it is likely that the remarkable ecological differentiation of Iberian E. aurinia, e.g. the larvae feeding on Lonicera species thus linking the species to hedge structures [19,20], and not Succisa pratensis and Scabiosa columbaria as in most of the other lowland populations [2124], is the result of long-lasting allopatry, perhaps in combination with different selective pressures acting in the Iberian and the other refugia.

thumbnail
Fig 6. Hypothetical last glacial distribution patterns and postglacial expansion routes of Euphydryas aurinia in Europe.

Striped pattern: putative continuous refugia; dashed pattern: putative structured refugia; solid arrows: proposed postglacial expansion routes; dashed arrows: supposed mtDNA introgression.

https://doi.org/10.1371/journal.pone.0142282.g006

Similar events might have taken place in Central Italy (al11), although no clear conclusions can be drawn concerning the location of different subrefugia in this area because only four populations were analysed. However, polycentricity of E. aurinia in Italy is supported by phylogeographic structures of many other species showing often numerous genetic groups and hence core areas scattered over Italy [97102].

Furthermore, the more southern Italian population IT3 shares its haplotype with populations from the Alps and has no haplotypes endemic to peninsular Italy, unlike the more northern population IT2. This discrepancy between nuclear and mitochondrial genes is most likely due to mitochondrial introgression from a more northern perialpine centre of survival (see below) along the Apennines to the South.

A ponto-Mediterranean refugium of E. aurinia could not currently be verified due to lack of samples. However, the descriptions of independent subspecies from the Balkans (E. aurinia balcanica Schawerda, 1908; E. aurinia bulgarica Fruhstorfer, 1916) are evidence for further centres of genetic differentiation in this area. Furthermore, refugia in the Balkan Peninsula are the rule for widespread species of Mediterranean origin [13], thus making it probable that this is also the case in E. aurinia.

Another Mediterranean glacial refugium in southern France seems to be most likely for E. aurinia due to the remarkable genetic differentiation of the populations from Provence for both genetic marker sets (al3, mt3) (Fig 6). The much higher than average genetic diversities of allozymes of the respective populations provide further support for a Provence refugium. Similar refugia in this area were assumed for e.g. Zerynthia polyxena [103], Carabus auronitens [104], Natrix natrix [102], Vipera aspis [105], Quercus suber [106] and Fagus sylvatica [17].

Beside these Mediterranean glacial refugia, the importance of extra-Mediterranean refugia has been revealed for many temperate European species (see [8] for a recent review). Such refugia might also be the most probable explanation for an independent western and eastern genetic group of E. aurinia in West and Central Europe. We suggest that the origin of the western group (al7) lies in an extra-Mediterranean refugium in Central France (Fig 6). The most probable location of this extra-Mediterranean refugium seems to be in the area north of the French Massif Central, a region with frequent indications of such refugia [8], from where postglacial expansion started towards the North and West of France as well as in an easterly direction towards western Bohemia (westernmost Czech Republic) (Fig 6). Additional analyses including samples from eastern Germany (e.g. Saxony, eastern Germany) might help to establish in more detail the contact zone between the western and eastern genetic group of E. aurinia in Central Europe.

The Central European group of E. aurinia (al8) might have had its origin east/south-east of the Alps, also an important area of extra-Mediterranean glacial refugia for many animal and plant species (e.g. [2,11,13,17,80,107,108]). Postglacial expansion from this refugium seems to run mainly towards the North and, at a limited scale, in a westerly direction (Fig 6). Hence, this genetic group of E. aurinia reached the southern parts of Sweden, while no populations of this group are known west of Salzburg (eastern Alps) and Bohemia (western Czech Republic). This refugium may have been in close contact with a Slovenian refugium, which might have been just a subrefugium of the former, in which stronger postglacial introgression processes led to its populations’ uniqueness.

Genetic differentiation at the nuclear (al10) and mitochondrial (mt9) level speaks for an independent extra-Mediterranean glacial refugium in the Carpathian Basin. The extraordinarily high genetic diversities of the here analysed populations from Transylvania support the high importance of this Basin as a centre for extra-Mediterranean survival with particularly favourable conditions [8], as also supported for several other species (e.g. [1113,107]).

Expansions from south-eastern Europe up to the Baltic States (as e.g. in the brown bear [109] or the butterfly Coenonympha arcania [110]) seem unlikely due to the strong genetic differentiation of the Estonian populations at the nuclear level (al12), with several endemic allozyme alleles. In fact, the Baltic populations of E. aurinia might have their origin in the East (possibly in the southern Ural, cf. [111]). However, the mtDNA haplotype obtained from Estonia is not different from the ones of Central Europe. Therefore, we assume that mitochondrial introgression of the Central European group has substituted a putative former eastern haplotype, but did not (or only marginally) influence the nuclear genetic information.

All populations from the Alps belong to the distinct mtDNA group mt4. These Alpine populations consistently show a different phenotype, which is considerably smaller than all ordinary lowland populations, and the general colouring of the wings is much darker; furthermore, these individuals have a different mode of flight (much closer to the ground and also quicker), use different larval food plants (i.e. Gentiana species) and do not show protandry like lowland populations do [18,25,26,112]. These ecological shifts might have happened during isolation in perialpine refugia (ice ages) and in their non-refugial environments (always during interglacials). As most of these differences represent specific adaptations to survive the harsh conditions of high mountain ecosystems [26], they might be the result of directed selection and not of random drift.

At the nuclear level, however, our study also revealed remarkable genetic differentiation within the Alps (FST = 0.1199), so that three Alpine genetic groups can be distinguished (al4, al5, al6) (Tables 3 and 5; Fig 5). This calls for the existence of not just one but several perialpine glacial refugia, at least during the last ice age. The two genetic groups in the western Alps (al4, al5) show moderate genetic differentiation from each other (FCT = 0.0663), indicating that both groups arose by allopatric differentiation during presumably one ice age. This relatively short time of divergence accords with the lack of mtDNA differentiation among the Alpine populations. The respective refugia might have been located near to the foot of the Cottian Alps (al4) and in the area south of the Upper Italian lakes (al5) (Fig 6). Especially the refugium of the western Central Alps group (al5) might have been of particular importance, as underlined by the rather high allozyme diversity of the populations analysed from this group. Evidence for similar refugia in these areas is not only found in various Alpine plant species [113120], but also the beetle Oreina elongata [121].

The populations of E. aurinia in the Hohe Tauern (eastern Alps, al6) show strong genetic differentiation from those of the western and central Alps (FCT = 0.1530). This result indicates a longer isolation of the populations of the eastern Alps (but note the missing differentiation at the mtDNA level) and an independent refugium in this region. Especially the area south of the Carnian and Julian Alps apparently offered suitable environmental conditions for many species, often resulting in higher genetic diversities of the genetic groups in the eastern Alps [10,82,122124]. We therefore assume that the mentioned region also represented an important glacial refugium for E. aurinia. Postglacial expansion might have been mainly in a northern direction into the higher Alps, as in the Slovenian populations (al9), but with the latter showing some introgression from the eastern Alps and the Apennines group at the mtDNA level.

Conclusion

Euphydryas aurinia shows a large variety of morphological as well as ecological adaptations within Europe, which have been the cause of continuous taxonomic debate. However, the results of the present study revealed that the ecological adaptations are most likely the result of strong intraspecific differentiation processes and not a result of the existence of a super-species complex. Hence, E. aurinia represents a rare model organism that combines attributes of Mediterranean, continental as well as alpine faunal elements within one single species. The complex phylogeography of E. aurinia might therefore be an example of more complicated patterns of differentiation in species showing adaptations to various climatic and altitudinal conditions, as well as host plants, across the western Palaearctic.

Supporting Information

S1 Table. Pairwise-location Dest (estimated) and FST values for mtDNA sequences Euphydryas aurinia.

Site abbreviations defined in Table 1.

https://doi.org/10.1371/journal.pone.0142282.s001

(PDF)

S2 Table. Pairwise-group Dest (estimated) and FST values for mtDNA sequences Euphydryas aurinia.

Groups defined in Table 2.

https://doi.org/10.1371/journal.pone.0142282.s002

(PDF)

S3 Table. Electrophoresis conditions for the different enzymes analysed for Euphydryas aurinia.

https://doi.org/10.1371/journal.pone.0142282.s003

(PDF)

Acknowledgments

We thank all responsible authorities in the respective European countries for the permits to collect butterflies. Our gratitude also goes to Dr Anna Cassel-Lundhagen (Swedish University of Agricultural Sciences, Uppsala, Sweden) for collecting important sampling material in Sweden as well as to Dr Steffen Caspari (St. Wendel, Germany), Dr Leonardo Dapporto (Prato, Italy), Dr Stanislav Gomboc (Kranj, Slovenia), Dr Constantí Stefanescu (Granollers Museum, Catalonia, Spain) and Dr Norbert Zahm (Schmelz, Germany) for friendly help in finding sampling localities.

Author Contributions

Conceived and designed the experiments: MJ RC TS. Performed the experiments: MJ AAR. Analyzed the data: MJ AAR RC TS. Contributed reagents/materials/analysis tools: MJ MZ AAR PG MK GN TT RC TS. Wrote the paper: MJ MZ AAR PG MK GN LR TT RC TS.

References

  1. 1. Hewitt GM (2000) The genetic legacy of the Quaternary ice ages. Nature 405: 907–913. pmid:10879524
  2. 2. Schmitt T (2007) Molecular biogeography of Europe: Pleistocene cycles and postglacial trends. Front Zool 4: 11. pmid:17439649
  3. 3. Bhagwat SA, Willis KJ (2008) Species persistence in northerly glacial refugia of Europe: a matter of chance or biogeographical traits? J Biogeogr 35: 464–482.
  4. 4. De Lattin G (1949) Beiträge zur Zoogeographie des Mittelmeergebietes. Verh dt zool Ges, 1949:143–151.
  5. 5. Hewitt GM (2004) Genetic consequences of climatic oscillation in the Quaternary. Philos Trans R Soc Lond B Biol Sci 359: 183–195. pmid:15101575
  6. 6. Reinig W (1950) Chorologische Voraussetzungen für die Analyse von Formenkreisen. Syllegomena Biologica, Festschrift O. Kleinschmidt: 364–378.
  7. 7. Steward JR, Lister AM (2001) Cryptic northern refugia and the origins of the modern biota. Trends Ecol Evol 16: 608–613.
  8. 8. Schmitt T, Varga Z (2012) Extra-Mediterranean refugia–the role and not the exception? Front Zool 9: 22. pmid:22953783
  9. 9. Schönswetter P, Stehlik I, Holderegger R, Tribsch A (2005) Molecular evidence for glacial refugia of mountain plants in the European Alps. Mol Ecol 14: 3547–3555. pmid:16156822
  10. 10. Schmitt T (2009) Biogeographical and evolutionary importance of the European high mountain systems. Front Zool 6: 9. pmid:19480666
  11. 11. Schmitt T, Rakosy L, Abadjiev S, Müller P. (2007) Multiple differentiation centres of a non-Mediterranean butterfly species in south-eastern Europe. J Biogeogr 34: 939–950.
  12. 12. Gratton P, Konopinski MK, Sbordoni V (2008) Pleistocene evolutionary history of the Clouded Apollo (Parnassius mnemosyne): genetic signatures of climate cycles and a 'time-dependent' mitochondrial substitution rate. Mol Ecol 17: 4248–4262. pmid:18986502
  13. 13. Ursenbacher S, Carlsson M, Helfer V, Tegelström H, Fumagalli L (2006) Phylogeography and Pleistocene refugia of the adder (Vipera berus) as inferred from mitochondrial DNA sequence data. Mol Ecol 15: 3425–3437. pmid:16968280
  14. 14. Pinceel J, Jordaens K, Pfenninger M, Backeljau T (2005) Rangewide phylogeography of a terrestrial slug in Europe: evidence for Alpine refugia and rapid colonization after the Pleistocene glaciations. Mol Ecol 14: 1133–1150. pmid:15773941
  15. 15. Provan J, Bennett KD (2008) Phylogeographic insights into cryptic glacial refugia. Trends Ecol Evol 23: 564–571. pmid:18722689
  16. 16. Magri D, Vendramin GG, Comps B, Dupanloup I, Geburek T, Gomory D, et al. (2006) A new scenario for the Quaternary history of European beech populations: palaeobotanical evidence and genetic consequences. New Phytol 171: 199–221. pmid:16771995
  17. 17. Magri D (2008) Patterns of post-glacial spread and the extent of glacial refugia of European beech (Fagus sylvatica). J Biogeogr 35: 450–463.
  18. 18. Tolman T, Lewington R (1997) Butterflies of Britain & Europe. Harper Collins Publishers, London.
  19. 19. Munguira ML, Martin J, García-Barros E, Viejo JL (1997) Use of space and resources in a Mediterranean population of the butterfly Euphydryas aurinia. Acta Oecol 18: 597–612.
  20. 20. Junker M, Schmitt T (2010) Demography, dispersal and movement pattern of Euphydryas aurinia (Lepidoptera: Nymphalidae) at the Iberian Peninsula: an alarming example in an increasingly fragmented landscape? J Insect Conserv 14: 237–246.
  21. 21. Konvicka M, Hula V, Fric Z (2003) Habitat of pre-hibernating larvae of the endangered butterfly Euphydryas aurinia (Lepidoptera: Nymphalidae): What can be learned from vegetation composition and architecture? Eur J Entomol 100: 313–322.
  22. 22. Hula V, Konvicka M, Pavlicko A, Zdenek F (2004) Marsh Fritillary (Euphydryas aurinia) in the Czech Republic: monitoring, metapopulation structure, and conservation of an endangered butterfly. Entomol Fenn 15: 231–241.
  23. 23. Betzholtz PE, Ehrig A, Lindeborg M, Dinnétz P (2007) Food plant density, patch isolation and vegetation height determine occurrence in a Swedish metapopulation of the marsh fritillary Euphydryas aurinia (Rottemburg, 1775) (Lepidoptera, Nymphalidae). J Insect Conserv 11: 343–350.
  24. 24. Zimmermann K, Blazkova P, Cizek O, Fric Z, Hula V, Kepka P, et al. (2011) Adult demography in the Marsh fritillary butterfly, Euphydryas aurinia (Rottenburg, 1775) in the Czech Republic: patterns across sites and seasons. Eur J Entomol 108: 243–254.
  25. 25. Schweizer Bund für Naturschutz (SBN) (1987) Tagfalter und ihre Lebensräume, Band 1. Verlag K. Holliger, Egg.
  26. 26. Junker M, Wagner S, Gros P, Schmitt T (2010) Changing demography and dispersal behaviour: ecological adaptations in an alpine butterfly. Oecologia 164: 971–980. pmid:20652595
  27. 27. Mazel R (1986) Structure et évolution du peuplement d'Euphydryas aurinia Rottemburg (Lepidoptera) dans le sud-ouest européen. Vie Milieu 36: 205–225.
  28. 28. Lafranchis T, Jutzeler D, Guillosson JY, Kan P, Kan B (2015) La vie des papillons, écologie, biologie et comportement des Rhopalocères de France. Diatheo, Paris.
  29. 29. Higgins LG (1950) A descriptive catalogue of the Palaearctic Euphydryas (Lepidoptera: Rhopalocera). Trans R Entomol Soc Lond 101: 435–487.
  30. 30. Tshikolovets VV (2011) Butterflies of Europe & the Mediterranean area. Tshikolovets publications, Pardubice.
  31. 31. Van Swaay C, Warren M (1999) Red data book of European butterflies (Rhophalocera). Council of Europe, Nature and Environment, No. 99, Strasbourg.
  32. 32. Schtickzelle N, Choutt J, Goffart P, Fichefet V, Baguette M (2005) Metapopulation dynamics and conservation of the marsh fritillary butterfly: population viability analysis and management options for a critically endangered species in Western Europe. Biol Conserv 126: 569–581.
  33. 33. Bos F, Bosveld M, Groenendijk D, van Swaay C, Wynhoff I (2006) De Dagvlinders van Nederlands. Nederlandse Fauna 7.
  34. 34. Saarinen K, Jantunen J, Valtonen A (2005) Resumed forest grazing restored a population of Euphydryas aurinia (Lepidoptera: Nymphalidae) in SE Finland. Eur J Entomol 102: 683–690.
  35. 35. Bulman C, Wilson R, Holt A, Bravo L, Early R, Warren M, et al. (2007) Minimum viable metapopulation size, extinction debt, and the conservation of a declining species. Ecol Appl 17: 1460–1473. pmid:17708221
  36. 36. Sigaard P, Pertoldi C, Madsen AB, Sogaardm B, Loeschke V (2008) Patterns of genetic variation in isolated Danish populations of the endangered butterfly Euphydryas aurinia. Biol J Linn Soc Lond 95: 677–687.
  37. 37. Maravalhas E (2003) As borboletas de Portugal. Apollo Books, Stenstrup.
  38. 38. Zimmermann M, Wahlberg N, Descimon H (2000) Phylogeny of Euphydryas Checkerspot butterflies (Lepidoptera: Nymphalidae) based on mitochondrial DNA sequence data. Ann Entomol Soc Am 93: 347–355.
  39. 39. Lai BCG, Pullin AS (2004) Phylogeography, genetic diversity and conservation of the large copper butterfly Lycaena dispar in Europe. J Insect Conserv 8: 27–36.
  40. 40. Folmer OM, Black W, Hoeh R, Lutz R, Vrijenhoek R (1994) DNA primers for amplification of mitochondrial cytochrome c oxidase subunit I from diverse metazoan invertebrates. Mol Mar Biol Biotechnol 3: 294–299. pmid:7881515
  41. 41. Lunt D, Zhang DX, Szymura J, Hewlett O (1996) The insect cytochrome oxidase I gene: evolutionary patterns and conserved primers for phylogenetic studies. Insect Mol Biol 5: 153–165. pmid:8799733
  42. 42. Harris H, Hopkinson DA (1978) Handbook of enzyme electrophoresis in human genetics. North-Holland, Amsterdam.
  43. 43. Richardson BJ, Baverstock PR, Adams M (1986) Allozyme electrophoresis. A handbook for animal systematics and population studies. Academic Press, San Diego.
  44. 44. Hebert PDN, Beaton MJ (1993) Methodologies for allozyme analysis using cellulose acetat electrophoresis. Helena Laboratories, Beaumont.
  45. 45. Librado P, Rozas J (2009) DnaSP v5: A software for comprehensive analysis of DNA polymorphism data. Bioinformatics 25: 1451–1452. pmid:19346325
  46. 46. Excoffier L, Lischer H (2010) Arlequin suite ver 3.5: A new series of programs to perform population genetics analyses under Linux and Windows. Mol Ecol Resour 10: 564–567. pmid:21565059
  47. 47. Posada D (2008) jModelTest: Phylogenetic model averaging. Mol Biol Evol 25: 1253–1256. pmid:18397919
  48. 48. Akaike H (1974) A new look at the statistical model identifications. IEEE Trans Automat Contr 19: 716–723.
  49. 49. Tamura K, Nei M (1993) Estimation of the number of nucleotide substitutions in the control region of mitochondrial DNA in humans and chimpanzees. Mol Biol Evol 10: 512–526. pmid:8336541
  50. 50. Bandelt H-J, Forster P, Röhl A (1999) Median-joining networks for inferring intraspecific phylogenies. Mol Biol Evol 16: 37–48. pmid:10331250
  51. 51. Polzin T, Daneschmand SV (2003) On Steiner trees and minimum spanning trees in hypergraphs. Op Res Lett 31: 12–20.
  52. 52. Weir BS, Cockerham CC (1984) Estimating F-statistics for the analysis of population structure. Evolution 38: 1358–70.
  53. 53. Jost L (2008) GST and its relatives do not measure differentiation. Mol Ecol 17: 4015–4026. pmid:19238703
  54. 54. Keenan K, McGinnity P, Cross TF, Crozier WW, Prodöhl PA (2013). diveRsity: An R package for the estimation and exploration of population genetics parameters and their associated errors. Methods Ecol Evol 4:782–788.
  55. 55. Siegismund HR (1993) G-Stat, ver. 3. Genetical statistical programs for the analysis of population data. The Arboretum, Royal Veterinary and Agricultural University, Horsholm, Denmark.
  56. 56. Goudet J (1995) FSTAT (Version 1.2): A computer program to calculate F-statistics. Heredity 86: 485–486
  57. 57. Nei M (1972) Genetic distances between populations. Am Nat 106: 283–291.
  58. 58. Saitou N, Nei M (1987) The neighbor-joining method: A new method for reconstructing phylogenetic trees. Mol Biol Evol 4: 406–425. pmid:3447015
  59. 59. Felsenstein J (1993) PHYLIP (Phylogeny Inference Package) Ver. 3.5.c. Department of Genetics, University of Washington, Seattle, Washington.
  60. 60. Kelly RP, Oliver TA, Sivasundar A, Palumbi SR (2010) A method for detecting population genetic structure in diverse, high gene-flow species. J Hered 101: 423–436. pmid:20219885
  61. 61. Guillot G, Mortier F, Estoup A (2005) Geneland: a computer package for landscape genetics. Mol Ecol Notes 5: 712–715.
  62. 62. Benjamini Y, Hochberg Y (1995) Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing. J R Stat Soc B 57: 289–300.
  63. 63. Thissen D, Steinberg L, Kuang D (2002) Quick and Easy Implementation of the Benjamini-Hochberg Procedure for Controlling the False Positive Rate in Multiple Comparisons. J Educ Behav Stat 27: 77–83.
  64. 64. Joyce DA, Pullin AS (2003) Conservation implications of the distribution of genetic diversity at different scales: a case study using the marsh fritillary butterfly (Euphydryas aurinia). Biol Cons 114: 453–461.
  65. 65. Descimon H, Zimmermann M, Cosson E, Barascud B, Nève G. (2001). Diversité génétique, variation géographique et flux géniques chez quelques Lépidoptères Rhopalocères français. Genet Select Evol 33: S223–S249.
  66. 66. Pelz V (1995) Biosystematik der europäischen Arten des Tribus Melitaeini (Newman, 1870). Oedippus 11: 1–62.
  67. 67. Hammouti N, Schmitt T, Seitz A, Kosuch J, Veith M (2010) Combining mitochondrial and nuclear evidences: a refined evolutionary history of Erebia medusa (Lepidoptera: Nymphalidae: Satyrinae) in Central Europe based on the CO1 gene. J Zool Syst Evol Res 48: 115–125.
  68. 68. Habel JC, Lens L, Rödder D, Schmitt T (2011) From Africa to Europe and back: refugia and range shifts cause high genetic differentiation in the Marbled White butterfly Melanargia galathea. BMC Evol Biol 11: 215. pmid:21777453
  69. 69. Vila M, Marí-Mena N, Guerrero A, Schmitt T (2011) Some butterflies do not care much about topography: a single genetic lineage of Erebia euryale (Nymphalidae) along the northern Iberian mountains. J Zool Syst Evol Res 49: 114–132.
  70. 70. Nève G (2009) Population genetics of European butterflies. In Settele J, Shreeve T, Konvicka M, Van Dyck H (Eds.) Ecology of Butterflies in Europe. Cambridge University Press, Cambridge: 107–129.
  71. 71. Vila M, Björklund M (2004) The utility of the neglected mitochondrial Control Region for evolutionary studies in Lepidoptera (Insecta). J Mol Evol 58: 280–290. pmid:15045483
  72. 72. Gompert Z, Nice CC, Fordyce JA, Forister ML, Shapiro AM (2006) Identifying units for conservation using molecular systematics: the cautionary tale of the Karner blue butterfly. Mol Ecol 15: 1759–1768. pmid:16689896
  73. 73. Schmitt T, Röber S, Seitz A (2005) Is the last glaciation the only relevant event for the present genetic population structure of the meadow brown butterfly Maniola jurtina (Lepidoptera: Nymphalidae)? Biol J Linn Soc 85: 419–431.
  74. 74. Habel JC, Dieker P, Schmitt T (2009) Biogeographical connections between the Maghreb and the Mediterranean peninsulas of southern Europe. Biol J Linn Soc 98: 693–703.
  75. 75. Besold J, Huck S, Schmitt T (2008) Allozyme polymorphisms in the small heath, Coenonympha pamphilus: recent ecological selection or old biogeographical signals? Ann Zool Fenn 45: 217–228.
  76. 76. Schmitt T, Varga Z, Seitz A (2005) Are Polyommatus hispana and Polyommatus slovacus bivoltine Polyommatus coridon (Lepidoptera: Lycaenidae)? The discriminatory value of genetics in the taxonomy. Org Divers Evol 5: 297–307.
  77. 77. Schmitt T, Seitz A (2001) Allozyme variation in Polyommatus coridon (Lepidoptera: Lycaenidae): identification of ice-age refugia and reconstruction of post-glacial expansion. J Biogeogr 28: 1129–1136.
  78. 78. DeChaine EG, Martin AP (2004) Historic cycles of fragmentation and expansion in Parnassius smintheus (Papilionidae) inferred using mitochondrial DNA. Evolution 58: 113–127. pmid:15058724
  79. 79. François O, Durand E (2010) Spatially explicit Bayesian clustering models in population genetics. Mol Ecol Resour 10: 773–784. pmid:21565089
  80. 80. Schmitt T, Seitz A (2001) Intraspecific allozymatic differentiation reveals the glacial refugia and the postglacial expansions of European Erebia medusa (Lepidoptera: Nymphalidae). Biol J Linn Soc 74: 429–458.
  81. 81. Schmitt T, Hewitt GM, Müller P (2006) Disjunct distributions during glacial and interglacial periods in mountain butterflies: Erebia epiphron as an example. J Evol Biol 19: 108–113. pmid:16405582
  82. 82. Schmitt T, Haubrich K (2008) The genetic structure of the mountain forest butterfly Erebia euryale unravels the late Pleistocene and postglacial history of the mountain coniferous forest biome in Europe. Mol Ecol 17: 2194–2207. pmid:18266631
  83. 83. Varga Z (1977) Das Prinzip der areal-analytischen Methode in der Zoogeographie und die Faunenelement-Einteilung der europäischen Tagschmetterlinge (Lepidoptera: Diurna). Acta Biol Debr 14: 223–285.
  84. 84. Hewitt GM (1996) Some genetic consequences of ice ages, and their role in divergence and speciation. Biol J Linn Soc 58: 247–276.
  85. 85. Hewitt GM (1999) Post-glacial re-colonization of European biota. Biol J Linn Soc 68: 87–112.
  86. 86. Taberlet P, Fumagalli L, Wust-Saucy AG, Cosson JF (1998) Comparative phylogeography and postglacial colonization routes in Europe. Mol Ecol 7: 453–464. pmid:9628000
  87. 87. Batista V, Harris DJ, Carretero MA (2004) Genetic variation in Pleurodeles waltl Michaelles, 1830 (Amphibia: Salamandridae) across the strait of Gibraltar derived from mitochondrial DNA sequences. Herpetozoa 16: 166–168.
  88. 88. Carranza S, Arnold EN (2004) History of west Mediterranean newts, Pleurodeles (Amphibia: Salamandridae), inferred from old and recent DNA sequences. Syst Biodiv 1: 327–337.
  89. 89. Carranza S, Wade E (2004) Taxonomic revision of Algero-Tunisian Pleurodeles (Caudata: Salamandridae) using molecular and morphological data. Revalidation of the taxon Pleurodeles nebulosus (Guichenot, 1850). Zootaxa 488: 1–24.
  90. 90. Martínez-Solano I (2004) Phylogeography of Iberian Discoglossus (Anura: Discoglossidae). J Zool Syst Evol Res 42: 223–233.
  91. 91. Carranza S, Harris DJ, Arnold EN, Batista V, Gonzalez de la Vega JP (2006) Phylogeography of the lacertid lizard Psammodromus algirus in Iberia and across the Strait of Gibraltar. J Biogeogr 33: 1279–1288.
  92. 92. Alexandrino J, Froufe E, Arntzen JW, Ferrand N (2000) Genetic subdivision, glacial refugia and postglacial recolonization in the golden-striped salamander, Chioglossa lusitanica (Amphibia: Urodela). Mol Ecol 9: 771–781. pmid:10849293
  93. 93. Alexandrino J, Arntzen JW, Ferrand N (2002) Nested clade analysis and the genetic evidence for population expansion in the phylogeography on the golden-striped salamander, Chioglossa lusitanica (Amphibia: Urodela). Heredity 88: 66–74. pmid:11813109
  94. 94. Paulo OS, Dias C, Bruford MW, Jordan WC, Nichols RA (2001) The persistence of Pliocene populations though the Pleistocene climatic cycles: evidence from the phylogeography of an Iberian lizard. Proc R Soc Lond B 268: 1625–1630.
  95. 95. García-París M, Alcobendas M, Buckley D, Wake DB (2003) Dispersal of viviparity across contact zones in Iberian populations of fire salamanders (Salamandra) inferred from discordance of genetic and morphological traits. Evolution 57: 129–143. pmid:12643573
  96. 96. Martínez-Solano I, Teixeira J, Buckley D, García-París M (2006) Mitochondrial DNA phylogeography of Lissotriton boscai (Caudata, Salamandridae): evidence for old, multiple refugia in an Iberian endemic. Mol Ecol 15: 3375–3388. pmid:16968276
  97. 97. Podnar M, Mayer W, Tvrtkovic N (2005) Phylogeography of the Italian wall lizard, Podarcis sicula, as revealed by mitochondrial DNA sequences. Mol Ecol 14: 575–588. pmid:15660947
  98. 98. Canestrelli D, Cimmaruta R, Costantini V, Nascetti G (2006) Genetic diversity and phylogeography of the Apennine yellow-bellied toad Bombina pachypus, with implications for conservation. Mol Ecol 15: 3741–3754. pmid:17032271
  99. 99. Canestrelli D, Cimmaruta R, Nascetti G (2008) Population genetic structure and diversity of the Apennine endemic stream frog, Rana italica–insights on the Pleistocene evolutionary history of the Italian peninsular biota. Mol Ecol 17: 3856–3872. pmid:18643880
  100. 100. Canestrelli D, Salvi D, Maura M, Bologna MA, Nascetti G (2012) One Species, three Pleistocene evolutionary histories: Phylogeography of the Italian Crested Newt, Triturus carnifex. PLOS ONE 7: e41754. pmid:22848590
  101. 101. Canestrelli D, Sacco F, Nascetti G (2012) On glacial refugia, genetic diversity, and microevolutionary processes: deep phylogeographical structure in the endemic newt Lissotriton italicus. Biol J Linn Soc 105: 42–55.
  102. 102. Kindler C, Böhme W, Corti C, Gvoždík Jablonski D, Jandzik D, Metallinou M, et al. (2013) Mitochondrial phylogeography, contact zones and taxonomy of grass snakes (Natrix natrix, N. megalocephala). Zool Scr 42: 458–472.
  103. 103. Zinetti F, Dapporto L, Vovlas A, Chelazzi G, Bonelli S, Balletto E, et al. (2013) When the rule becomes the exception. No evidence of gene flow between two Zerynthia cryptic butterflies suggests the emergence of a new model group. PLOS ONE 8: e65746. pmid:23755277
  104. 104. Assmann T, Nolte O, Reuter H (1994) Postglacial colonization of middle Europe by Carabus auronitens as revealed by population genetics (Coleoptera, Carabidae). In Desender K, Dufrêne M, Loreau M, Luff ML, Maelfait JP (Eds.) Carabid Beetles, Ecology and Evolution. Kluwer, Dordrecht: 3–9.
  105. 105. Ursenbacher S, Conelli A, Golay P, Monney J-C, Zuffi MAL, Thiery G, et al. (2006) Phylogeography of the asp viper (Vipera aspis) inferred from mitochondrial DNA sequence data: Evidence for multiple Mediterranean refugial areas. Mol Phylogenet Evol 38: 546–552. pmid:16213755
  106. 106. Magri D, Fineschi S, Bellarosa R, Buonamici A, Sebastiani F, Schirone B, et al. (2007) The distribution of Quercus suber chloroplast haplotypes matches the palaeogeographical history of the western Mediterranean. Mol Ecol 16: 5259–5266. pmid:17995923
  107. 107. Babik W, Branicki W, Sandera M, Litvinchuk S, Borkin LJ, Irwin JT, et al. (2004) Mitochondrial phylogeography of the moor frog, Rana arvalis. Mol Ecol 13: 1469–1480. pmid:15140091
  108. 108. Ujvárosi L, Bálint M, Schmitt T, Mészáros N, Ujvárosi T, Popescu O. (2010) Divergence and speciation in the Carpathians area: patterns of morphological and genetic diversity of the crane fly Pedicia occulta (Diptera, Pediciidae). J N Am Benthol Soc 29: 1075–1088.
  109. 109. Taberlet P, Bouvet J (1994) Mitochondrial DNA polymorphism, phylogeography, and conservation genetics of the brown bear (Ursus arctos) in Europe. Proc R Soc Lond B Biol Sci 255: 195–200.
  110. 110. Besold J, Schmitt T, Tammaru T, Cassel-Lundhagen A (2008) Strong genetic impoverishment from the centre of distribution in southern Europe to peripheral Baltic and isolated Scandinavian populations of the pearly heath butterfly. J Biogeogr 35: 2090–2101.
  111. 111. Cassel A, Tammaru T (2003) Allozyme variability in central, peripheral and isolated populations of the scarce heath (Coenonympha hero: Lepidoptera, Nymphalidae); implications for conservation. Conserv Genet 4: 83–93.
  112. 112. Nève G., Singer MC (2008) Protandry and postandry in two related butterflies: conflicting evidence about sex-specific tradeoffs between adult size and emergence time. Evol Ecol 22: 701–709.
  113. 113. Schönswetter P, Tribsch A, Barfuss M, Niklfeld H. (2002) Several Pleistocene refugia detected in the high alpine plant Phyteuma globulariifolium Sternb. & Hoppe (Campanulaceae) in the European Alps. Mol Ecol 11: 2637–2647. pmid:12453246
  114. 114. Alvarez N, Manel S, Schmitt T, the IntraBioDiv Consortium (2012) Contrasting diffusion of Quaternary gene pools across Europe: The case of the arctic–alpine Gentiana nivalis L. (Gentianaceae). Flora 207: 408–413.
  115. 115. Schönswetter P, Tribsch A, Niklfeld H (2003) Phylogeography of the high alpine cushion-plant Androsace alpina (Primulaceae) in the European Alps. Plant Biol 5: 623–630.
  116. 116. Schönswetter P., Tribsch A., Schneeweiss G.M. & Niklfeld H. (2003b) Disjunctions in relict alpine plants: phylogeography of Androsace brevis and A. wulfeniana (Primulaceae). Bot J Linn Soc 141: 437–446.
  117. 117. Schönswetter P, Tribsch A, Stehlik I, Niklfeld H (2004) Glacial history of high alpine Ranunculus glacialis (Ranunculaceae) in the European Alps in a comparative phylogeographical context. Biol J Linn Soc 81: 183–195.
  118. 118. Stehlik I, Schneller JJ, Bachmann K (2002) Immigration and in situ glacial survival of the low-alpine Erinus alpinus (Scrophulariaceae). Bot J Linn Soc 77: 87–103.
  119. 119. Tribsch A, Schönswetter P, Stuessy TF (2002) Saponaria pumila (Caryophyllaceae) and the ice-age in the Eastern Alps. Amer J Bot 89: 2024–2033.
  120. 120. Thiel-Egenter C, Holderegger R, Brodbeck S, Intrabiodiv Consortium, Gugerli F (2009) Concordant genetic breaks, identified by combining clustering and tessellation methods, in two co-distributed alpine plant species. Mol Ecol 18: 4495–4507. pmid:19769690
  121. 121. Margraf N, Verdon A, Rahier M, Naisbit RE (2007) Glacial survival and local adaptation in an alpine leaf beetle. Mol Ecol 16: 2333–2343. pmid:17561894
  122. 122. Pauls SU, Lumbsch HT, Haase P (2006) Phylogeography of the montane cadddisfly Drusus discolor: evidence for multiple refugia and periglacial survival. Mol Ecol 15: 2153–2169. pmid:16780432
  123. 123. Haubrich K, Schmitt T (2007) Cryptic differentiation in alpine-endemic, high-altitude butterflies reveals down-slope glacial refugia. Mol Ecol 16: 3643–3658. pmid:17845437
  124. 124. Schmitt T, Habel JC, Rödder D, Louy D (2014) Effects of recent and past climatic shifts on the genetic structure of the high mountain Yellow-spotted ringlet butterfly Erebia manto (Lepidoptera, Satyrinae): a conservation problem. Global Change Biol 20: 2045–2061.