15urn:lsid:arphahub.com:pub:E0185C18-FE79-5ADE-9877-ED333312DD4Furn:lsid:zoobank.org:pub:EF082B8D-8FD0-41B8-BC24-3D1190FEC17FNota LepidopterologicaNL0342-75362367-5365Pensoft Publishers10.3897/nl.45.6813868138Research ArticleNymphalidaeBiogeographyCenozoicEuropeMitochondrial DNA-based phylogeography of the large ringlet Erebiaeuryale (Esper, 1805) suggests recurrent Alpine-Carpathian disjunctions during Pleistocene (Nymphalidae, Satyrinae)CupedoFransfrans@cupedo.eu1DoorenweerdCamielhttps://orcid.org/0000-0002-0418-443923Processieweg 2, 6243 BB Geulle, Netherlands; frans@cupedo.euUnaffiliatedGeulleNetherlandsBiodiversity Discovery group, Naturalis Biodiversity Center, PO Box 9517, 2300 RA Leiden, NetherlandsNaturalis Biodiversity CenterLeidenNetherlandsUniversity of Hawaii, Department of Plant and Environmental Protection Sciences, 3050 Maile way, Honolulu, 96822 Hawaii, USA; Camiel.doorenweerd@hawaii.eduUniversity of HawaiiHonoluluUnited States of America
Subject Editor: Roger Vila
2022190120224565860904222B-39CA-56AB-B89E-A201DD4F7BF6B4BD2D0B-A98C-4C2C-B965-C93C8E417E6359090190105202113122021Frans Cupedo, Camiel DoorenweerdThis is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.http://zoobank.org/B4BD2D0B-A98C-4C2C-B965-C93C8E417E63
Most species of the butterfly genus Erebia are high altitude specialists, in which territorial fragmentation is associated with distinct genetic patterns. This is also true for the large ringlet, Erebiaeuryale (Esper, 1805), a species widespread across European mountain systems. Previous molecular studies revealed four lineages: two in the Alps, coinciding with the ssp. adyte and isarica, one in the Pyrenees and Cantabria (ssp. pyraenaeicola), and one in the Carpathians and the Balkans (ssp. syrmia). Two morphological subspecies inhabiting delimited ranges in the southern Alps (ssp. pseudoadyte and kunzi) were not included in these studies. To further our understanding of the relationships between populations, both the Alpine and the extra Alpine ones, we sequenced 1,496 bp of the COI gene in 16 Alpine and Jurassian populations and analysed them in combination with published Pyrenean and Carpathian sequences. The resulting haplotype network shows five lineages, congruent with the morphologic delineation of subspecies. Based on the current distribution ranges and genetic affinities, we reconstructed a pre-Würm phylogeographic scenario. This suggests an initial split resulting in an Alpine and a Carpathian clade, probably of Carpathian origin. Within the Alps, three subspecies subsequently differentiated, probably during several glacial cycles, generating ssp. adyte, pseudoadyte and kunzi. In parallel, the Carpathian clade underwent a second Alpine–Carpathian disjunction and differentiated into ssp. euryale and syrmia in the Carpathians, and ssp. ocellaris and isarica in the eastern Alps, revealing a heterogeneous origin of the E.euryale subspecies across the Alps. The Pyrenean and Jurassian populations are a relatively young divergence in the western part of the species’ range.
High altitude organisms in general have the same insular distribution patterns as their habitats (a “continental–insular fauna”: Varga 1996). In their fragmented territory gene flow is effectively interrupted, resulting in a geographical pattern of genetic variation (Varga 1996; Varga and Schmitt 2008). This phylogeographic structure was mainly shaped during the Quaternary climate oscillations (Hewitt 1996; Comes and Kadereit 1998; Schmitt 2007). The succession of warm and cold periods caused repeated altitudinal range shifts of mountain biota. The classic idea that, during glacial periods, cold-adapted species were widespread in the plains (Reinig 1938; Holdhaus 1954; de Lattin 1967; Muster and Berendonk 2006) has been superseded (Schmitt et al. 2006; Schmitt 2007). The glaciated Alps were flanked by permafrost tundra in the north, and in the south by dry cold-steppe (van Andel and Tzedakis 1996; Tzedakis et al. 2002; Monegato et al. 2007). Both were arid and treeless (Lang 1994; Müller et al. 2003; Stojakowits et al. 2020), thus hostile to alpine species dependent on moisture and wind shelter. For these organisms, the plains were liveable only during a relatively short period in the transition to a warm climate. This enabled them to colonise distant, nowadays ecologically isolated mountain ranges (de Lattin 1967; Hewitt 1999). Under pleniglacial conditions, permanently liveable circumstances were restricted to discrete, climatologically favoured areas along the alpine borders (Stehlik 2000; Schönswetter et al. 2005; Schmitt 2009). For an increasing number of plant species these peripheral refugia have been identified through a fine scale analysis of their phylogeographic patterns (Stehlik et al. 2002; Schönswetter et al. 2003; Tribsch and Schönswetter 2003; Schönswetter et al. 2004). To the best of our knowledge, no such analysis exists of Alpine butterfly species. Nunatak survival, as supposed for some alpine plants (Stehlik 2000, 2003; Parisod and Besnard 2007), has not been evidenced in butterflies yet.
The Holarctic ringlet butterfly genus Erebia is amongst the most intensively studied alpine insect groups, due to its high species and subspecies richness across montane/alpine and boreal/arctic biomes. The genus most likely originated in Asia, and colonised Europe some 17–23 million years (My) ago (Peña et al. 2015). It underwent a broad diversification into some dozens of species during Miocene and Pliocene (Martin et al. 2000; Peña et al. 2015; Wiemers et al. 2020), followed by a fast radiation caused by the Pleistocene glacial cycles. These most recent differentiations as a rule were intraspecific (Vila et al. 2005; Schmitt et al. 2006; Schmitt et al. 2014; Hinojosa et al. 2018; Cupedo and Doorenweerd 2020), and in a few cases on species level (Martin et al. 2002; Albre et al. 2008; Wiemers et al. 2020). The large ringlet, Erebiaeuryale (Esper, 1805) is one of the most widespread and most variable species of the genus, with sixteen morphological subspecies described (Cupedo 2010). Five of these are endemic to the Alps: adyte, pseudoadyte, kunzi, isarica and ocellaris. Their territories are separated by extremely narrow intergradation zones (Rezbanyai-Reser 1991; Sonderegger 2005; Cupedo 2014), as is common in closely related Erebia taxa (Schmitt and Müller 2007; Lucek et al. 2020). In contrast, the subspecies isarica and ocellaris are separated by a hybrid zone tens of kilometres wide (Cupedo 2010). Two molecular studies addressed the genetic relationships of the central European taxa. Schmitt and Haubrich (2008) identified two allozyme based lineages in the Alps, coinciding with the morphologic ssp. adyte and isarica. The ssp. ocellaris and isarica were found genetically identical, the ssp.pseudoadyte and kunzi (both described two years later) were not sampled. Outside the Alps, their study identified a lineage in the Pyrenees (concordant with ssp.pyraenaeicola), and one in the Southern Carpathians and the Balkans (ssp.syrmia). A mtDNA based study (Vila et al. 2011) revealed that (i) the populations from the Pyrenees and Cantabria represent a single lineage, and (ii) the ocellaris and isarica samples are phylogenetically closer to syrmia than to each other. Paučulová et al. (2017) found two genetic lineages in the Carpathians and the Sudeten Mts, coinciding with the morphologic ssp.euryale and syrmia, and established the border between them. Ssp.euryale inhabits the Sudeten Mts (Jeseník, Krkonoše) and the Western Carpathians (Tatra, Fatra), whereas ssp.syrmia inhabits the Eastern and Southern Carpathians.
To date, the subspeciespseudoadyte and kunzi are only morphologically defined. Although very similar to adyte and ocellaris in their wing pattern, they exhibit consistent differences in the male genital characters. Ssp.kunzi occurs in two forms: western and eastern kunzi (Fig. 1B). Both geographically and wing-morphologically the western form is intermediate between pseudoadyte and eastern kunzi. Because of genital characters it was nonetheless regarded as kunzi, rather than of hybrid origin (Cupedo 2010). At the very few known contact sites, both between adyte and pseudoadyte and between kunzi and ocellaris, morphological intergradation zones are extremely narrow, suggesting reduced reproductive compatibility (Cupedo 2014).
Overview of the locations listed in Table 1. A. Locations outside the Alps and the Jura, sequences retrieved from GenBank. Black rectangle: area covered by Fig. 1B; B. Locations in the Alps and the Jura. Stars – sampling locations, dots – sequences mined from GenBank, coloured areas – distribution of morphologic subspecies after Willien (1985), Sonderegger (2005), Cupedo (2010) and personal observations, dotted area – transitional populations isarica/ocellaris, dark grey – lakes and coastline. Sample numbers refer to Table 1.
https://binary.pensoft.net/fig/635880
The number of genetic lineages of E.euryale in the Alps thus may vary between two and five, depending on the underlying data. In order to obtain one dataset covering all recognized and potential lineages, we extended previously published mtDNA data (Dincă et al. 2010; Vila et al. 2011; Paučulová et al. 2017) with 15 Alpine populations and a Jurassian one. We sequenced both the 5’P section of the Cytochrome C Oxidase I gene (the “barcode section”), and the 3’P part, which has proven to be more informative in some Erebia species (Vila et al. 2005; Vila et al. 2011), although not all of them (Cupedo and Doorenweerd 2020). The aim of our study is to identify the mtDNA based genetic lineages of E.euryale in the Alps, to establish their relation to morphologic subspecies, to clarify their phylogenetic relations and their connections with the Jura, Pyrenees (including Cantabria), the Carpathians and the Sudetes, and to reconstruct their phylogeographic history. With these objectives, we addressed six specific questions: (1) Are the subspeciespseudoadyte and kunzi distinct lineages, different from their look-alikes adyte and ocellaris, respectively? (2) If so, are the morphologically intermediate populations (‘western kunzi’) rightly assigned to ssp.kunzi? (3) Do the ssp.isarica and ocellaris represent one or two genetic lineages? (4) What is the genetic relation between the Alpine lineages and those from the Jura, the Pyrenees, the Sudetes and the Carpathians/Balkans? Having clarified these issues, we (5) performed a fine-scale reconstruction of the Würm glacial refugia of the Alpine lineages, as a setup for (6) a reconstruction their pre-Würm phylogeography, in conjunction with the Jurassian, Pyrenean and Carpathian lineages.
Material and methodsStudy species
Erebiaeuryale finds its ecologic optimum in edges and clearings of the timberline forest, irrespective of the forest association (conifer or beech forest), or the underlying bedrock (calcareous or siliceous). Above the timberline, it can be abundant in the green alder shrubs (Alnetumviridis) and in the alpenrose heath vegetation (Rhododendro–Vaccinietum). Locally, it descends into the montane zone or ascends into the alpine meadows (Gonseth 1987; Sonderegger 2005). The larvae feed on grasses, their development is biennial (Sonderegger 2005; Klecková et al. 2015).
Sampling
Adult individuals were collected at 16 localities (Fig. 1 and Table 1). As an outgroup, two males were sampled of E.ligea (Linnaeus, 1758), the nearest relative of E.euryale (Dincă et al. 2010; Peña et al. 2015; Dincă et al. 2021). Netted individuals were paralysed with ethyl acetate vapour, and two legs were transferred to ethanol 96% before death. Specimens included in this study are all registered on the Barcode of Life Database (Ratnasingham and Hebert 2007) with detailed information on collecting localities and identification (dataset https://doi.org/10.5883/DS-EURYALE) Vouchers are deposited in the collection of Naturalis Biodiversity Center (RMNH; Leiden, the Netherlands). Additional COI sequences were retrieved from BOLD and GenBank, from localities not represented in our samples. These covered bp 1–658 (Dincă et al. 2010), bp 668–1483 (Vila et al. 2011) or bp 305–1475 (Paučulová et al. 2017) (Table 1, Suppl. material 1: Table S1).
Sampling data of Erebiaeuryale and Erebialigea (outgroup) underlying the haplotype networks in Fig. 2. No – sample number, Voucher ID – of own samples (prefix RMNH.INS.), N – sample size, 3’P, 5’P, 305-1475 – COI segment sequenced. Samples marked with asterisk are mined from GenBank.
No
Subspecies
Author
Massif
Locality
Ctry
Coordinates (DDM)
Alt (m)
Date
Voucher ID
N
3’P
5’P
305-1475
1
adyte
(Hübner, 1822)
Cottian Alps
Risoul
FR
44°38.07'N, 006°37.95'E
5525
13/07/2012
552442–552452
11
x
x
2
adyte
(Hübner, 1822)
Penninic Alps
Cheggio
IT
46°05.84'N, 008°06.06'E
1760
09/07/2012
559813–559822
8
x
x
*3
adyte
(Hübner, 1822)
Graian Alps
Val d’Isère
FR
45°27'N, 006°59'E
2000–2100
04/08/2003
10
x
4
adyte
(Hübner, 1822)
W. Rhaetian Alps
Langtauferertal
IT
46°49.49'N, 010°40.82'E
2000
29/07/2009
544671–544685
6
x
x
5
pseudoadyte
Cupedo, 2010
Garda Pre-Alps
Passo Tremalzo
IT
45°50.78'N, 010°40.92'E
1490
27/07/2013
559888–559897
10
x
x
6
pseudoadyte
Cupedo, 2010
Adamello
Val Genova
IT
46°10.81'N, 010°38.06'E
1390
28/07/2013
556626–556637
11
x
x
7
kunzi
Heinkele, 2007
Dolomites-Latemar
Gallio
IT
46°55.58'N, 011°33.62'E
1620
21/07/2013
559865–559876
12
x
x
8
kunzi
Heinkele, 2007
Dolomites-Latemar
Passo Brocon
IT
46°07.24'N, 011°41.43'E
1620
20/07/2013
559853–559864
12
x
x
9
kunzi
Heinkele, 2007
Dolomites-Feltre chain
Passo Palughet
IT
46°10.97'N, 011°54.90'E
1850
03/08/2008
544662–544670
4
x
x
10
kunzi
Heinkele, 2007
Venetian Pre-Alps
Monte Grappa
IT
45°52.97'N, 011°47.69'E
1630
22/07/2013
559877–559887
11
x
x
11
kunzi
Heinkele, 2007
Venetian Pre-Alps
Monte Cavallo
IT
46°07.95'N, 012°31.36'E
1350
17/07/2013
556602–556613
11
x
x
12
ocellaris
Staudinger, 1861
Dolomites
Karer Pass
IT
46°25.12'N, 011°36.04'E
1780
31/07/2013
556614–556625
12
x
x
13
ocellaris
Staudinger, 1861
Dolomites
Passo Pordoi
IT
46°28.92'N, 011°47.30'E
1940
01/08/2013
559898–559907
10
x
x
*14
ocellaris
Staudinger, 1861
Dolomites
Kalkstein
AT
46°48'N, 012°19'E
1600
15/08/2005
5
x
15
isarica
Heyne, 1895
Graian Alps
Cormet de Roselend
FR
45°41.62'N, 006°39.49'E
1760
25/07/2012
559823–559828
5
x
x
16
isarica
Heyne, 1895
N Tyrol limestone Alps
Rofan Mountains
AT
47°27.82'N, 011°49.56'E
1750
05/08/2013
559829–559840
12
x
x
*17
isarica
Heyne, 1895
Nieder Tauern
Obertauern
AT
47°15'N, 013°34'E
1800–2000
18/08/2005
6
x
18
isarica
Heyne, 1895
Gurktal Alps
Turracher Höhe
AT
46°55.53'N, 013°53.05'E
1890
16/07/2013
559841–559852
12
x
x
*19
syrmia
Fruhstorfer, 1909
S. Carpathians
Bucegi Mts
RO
45°21'N, 025°31'E
1400–1600
22/07/2004
10
x
20
tramelana
Reverdin, 1818
Jura
Mijoux
FR
46°21.59'N, 006°01.03'E
1520
26/07/2012
556638–556648
11
x
x
*21
cantabricola
Verity, 1927
Cantabrian Mts
Cantabria
SP
42°N, 004°E–006°E
07/2006
36
x
*22
pyrenaeicola
v.d. Goltz, 1930
Pyrenees
La Glèbe
FR
42°40'N, 002°13'E
1500-2000
28/07/2003
10
x
*23
syrmia
Fruhstorfer, 1909
S. Carpathians
Div. loc.
RO
17
x
*24
cantabricola
Verity, 1927
Cantabria
Div. loc.
SP
6
x
*25
pyrenaeicola
v.d. Goltz, 1930
Pyrenees
Div. loc.
FR, AND
10
x
*26
syrmia
v.d. Goltz, 1930
S. Carpathians
Bucegi Mts
RO
45°24.37’N, 025°29.86’E
2012
5
x
*27
syrmia
v.d. Goltz, 1930
E. Carpathians
Bukovské Mts
SK
49°05.30’N, 022°34.05’E
2011
4
x
*28
euryale
(Esper, 1805)
W. Carpathians
Volovské Mts
SK
48°46.95’N, 020°59.32’E
2011
5
x
*29
euryale
(Esper, 1805)
W. Carpathians
Great Fatra
SK
48°54.13’N, 019°04.75’E
2011
2
x
*30
euryale
(Esper, 1805)
W. Carpathians
Čergov Mts
SK
49°13.83’N, 021°00.45’E
2011
4
x
*31
euryale
(Esper, 1805)
W. Carpathians
High Tatra
SK
49°13.77’N, 020°13.20’E
2013
4
x
*32
euryale
(Esper, 1805)
W. Carpathians
Slovak Paradise
SK
48°53.38’N, 020°20.65’E
2013
5
x
*33
euryale
(Esper, 1805)
Sudeten Mts
Hrubý Jeseník
CZ
50°08.00’N, 017°23.00’E
2009
3
x
*34
euryale
(Esper, 1805)
Sudeten Mts
Krkonoše Mts
CZ
50°44.42’N, 015°44.42’E
2013
4
x
Outgroup
E.ligea
(Linnaeus, 1758)
Garda Pre-Alps
Valle di Concei
45°55.68’N, 010°45.97’E
1590
26/07/2013
556599–556600
2
DNA extraction, PCR and sequencing
Genomic DNA was extracted from the legs with a Macherey–Nagel NucleoMag 96 Tissue magnetic bead kit on a Thermo Fisher KingFisher flex system. Polymerase chain reaction (PCR) was used to amplify a fragment of 1,496 base pairs (bp) of the Cytochrome C Oxidase subunit I gene (COI) from the mitochondrial genome, in two parts. The first part of 658 bp, near the 5’P side of the gene, also known as the DNA barcode region (Hebert et al. 2003) was amplified using a 1:1 primer mix of LepF1 (5’ ATT CAA CCA ATC ATA AAG ATA TTG G 3’) and LCO1490 (5’ GGT CAA CAA ATC ATA AAG ATA TTG G 3’) as forward primers, and a 1:1 primer mix of LepR1 (5’ TAA ACT TCT GGA TGT CCA AAA AAT CA 3’) and HCO2198 (5’ TAA ACT TCA GGG TGA CCA AAA AAT CA 3’) as reverse primers (Folmer et al. 1994; Doorenweerd et al. 2014). The second part that we amplified, from here on indicated as the 3’P fragment, has a 32 bp overlap with the 5’P fragment, and a target length of 870 bp. This fragment was amplified with forward primer SeqIntCOIf (5’CWT CWT TTT TTG AYC CAG CWG GAG 3’) and reverse primer LepLEUr (5’ CCA TTA CWT ATA RTC TGC CAT ATT 3’) (Vila and Björklund 2004). M13 forward (5’ TGT AAA ACG ACG GCC AGT 3’) or reverse (5’ CAG GAA ACA GCT ATG AC 3’) tails were attached to the forward and reverse primers respectively, and used for sequencing. Each PCR reaction included 40 cycles, consisting of 3 minutes initial denaturation at 94 °C, 15 seconds cycle denaturation at 94 °C, 30 seconds cycle annealing at 50 °C, and 40 seconds cycle extension at 72 °C. A final extension at 72 °C for 5 minutes completed the reactions. Bidirectional Sanger sequencing was outsourced to BaseClear, Leiden, The Netherlands. The resulting chromatograms were checked for quality and congruence in Geneious R6.1.8 and the resulting sequences were managed using VoSeq 1.7.4 (Peña and Malm 2012) prior to uploading to BOLD (www.barcodingoflife.org) and Genbank (www.ncbi.nlm.gov/Genbank) (accessions MT762427–MT762602).
Analyses
Specimens with missing data were removed, leaving 158 E.euryale samples for haplotype analyses. Median-joining haplotype networks were constructed using NETWORK v10.2.0.0 (www.fluxus-engineering.com), based on different sequences: (i) Our main dataset of 1495 bp, i.e., the combined 3’P and 5’P fragments. (ii) Two subsets, of the 3’P and 5’P fragments separately, both expanded with sequences mined from GenBank. (iii) A set of sequences covering bp 305–1475, retrieved from GenBank. Because the mined 3’P sequences were 816 bp long instead of 838, we truncated our 3’P sequences to bp 816 bp. The excised nucleotides contained no mutations. Because the 5’P fragment in ssp.syrmia was invariable (N=17) and obviously did not contribute to the variation, we concatenated the mined 3’P and 5’P sequences of this subspecies and included them into our 3’P+5’P dataset, which we accordingly truncated to 1,474 bp.
From the main dataset we inferred a maximum-likelihood tree using IQ-TREE v1.6.10 (Nguyen et al. 2015) with 5,000 approximate likelihood ratio test repetitions (flag -alrt) and 5,000 rapid bootstraps (flag -bb). The IQ-Tree integrated ModelFinder selected TN+I+F as the best fit model according to the Bayesian information criterion, but all models scored very similar. For Bayesian tree inference and molecular dating we used RevBayes v1.0.12 (Höhna et al. 2016). We calibrated the root of the tree, the split between E.euryale and E.ligea, at 3.49 My to match the estimate that was based on a four gene Erebia wide phylogeny (Peña et al. 2015). A study that used a supertree approach with largely the same genetic data for Erebia but different calibration points, estimated the split between E.ligea and E.euryale very similarly, at 3.58 My (Wiemers et al. 2020). We ran two independent MCMC chains in parallel for 20,000 generations, both with the GTR+G substitution model, and used Tracer v1.7.1 (Rambaut et al. 2018) to verify that this resulted in sufficiently large estimated sampling sizes (>>200).
Potential Würm glacial refugia
Reconstructing phylogeography requires insight in past area shifts. Because Würm glacial refugia form the bridge between present and pre-Würm distribution patterns, we primarily examined how the present distributions line up with potential Würm refugia. Six regions are broadly recognised as potential Würm glacial refugia: (i) In the westernmost (French pre-Alps, south of the Isère glacier) and in (ii) in the easternmost parts of the Alps (south of the Enns glacier), extended areas of unglaciated lower mountains acted as large scale refugia to alpine organisms (Schönswetter et al. 2003; Buoncristiani and Campy 2004; van Husen 2004; Schönswetter et al. 2005). (iii) Along the southern border of the Alpine arch, habitable areas were present on the south exposed slopes. These were unglaciated (Penck and Brückner 1909; Geologische Bundesanstalt (AT) 2013), and foothill topography ensured wind shelter, increased insolation, and moist conditions by orogenic precipitation and summery melt water supply, resulting in tempered micro habitats (Avigliano et al. 2002; Kropf et al. 2002; Ravazzi et al. 2004; Monegato et al. 2007). A key feature of this habitable belt is that it was heavily compartmentalized by north–south running glacial valleys (Penck and Brückner 1909; Castiglioni 2004). These valleys acted as habitat interruptions, impeding lateral dispersal. (iv) Along the northern Alpine border, a lobed ice front composed of piedmont glaciers in lateral contact, shovelled far into the foreland (Stojakowits et al. 2020). Some persisting incisions in this ice front (Keller and Krayss 2010; Reitner 2011; Heß 2013; Graf et al. 2015) offered suitable refugial conditions to sub alpine organisms (Schönswetter et al. 2005). (v) Adjacent to the eastern Alps, the Pannonian Basin has been evidenced to have had taiga like forests and even patches of temperate deciduous woodland during the Würm glacial (Rudner and Sümegi 2001; Willis and van Andel 2004; Feurdan 2005; Janovska and Pokorný 2008; Kuneš et al. 2008; Juřičková et al. 2014). (vi) The southern Pyrenean foreland offered favourable conditions even during pleniglacial periods (González-Sampériz et al. 2005). This goes to a lesser degree for the western and eastern edges of the Pyrenees (Calvet 2004). The northern Pyrenees were far less hospitable (Heinz and Barbaza 1998), and treeless (de Beaulieu et al. 1994).
Pleistocene chronology and nomenclature
For the nomenclature of Pleistocene warm and cold phases we follow the Marine Isotope Stages (MIS) system (Martinson et al. 1987; Lisiecki and Raymo 2005) (Appendix 1: Fig. A1). The main Würm glacier advance (the Last Glacial Maximum, LGM) corresponds to MIS2, the main Riss advance to MIS6. We avoid the terms Mindel and Günz, because there is no consensus on their correlation with MIS stages (cf. Head and Gibbard 2005; Häuselmann et al. 2007). Both probably cover more than one climatic cycle (Kukla 2005). Nomenclature of mountain stocks is according to Marazzi (2005).
ResultsHaplotype networks
The main network, based on 1,474 bp of COI (Fig. 2A) contains fifty-three haplotypes, determined by 51 variable sites (two variants per site), accounting for 3.4% of the sequence. Thirty-two sites were parsimony informative, 13 of which were located on the 5’P fragment and 19 on the 3’P fragment. Twenty-nine substitutions (57%) were transitions. Fifty-one haplotypes (98%) were private to a single subspecies, forty-six (88%) to a single population. Haplotypes are numbered 1 through 53 in Fig. 2A. They appeared grouped in five clusters, designated A through E. Additional haplotypes in the mined sequences of the 3’P fragment are numbered 54 through 59, (inserts in Fig. 2A). The maximum distance (outgroup excluded) was 9 mutational steps (0.61% pairwise sequence difference). The outgroup Erebialigea (haplotypes 60 and 61) was connected by 20 mutational steps, a pairwise sequence difference of 1.35%. The only loop, regarding haplotype 37, was resolved by breaking the connection to cluster C according to the criteria by Crandall and Templeton (1993). The 5’P fragment (Fig. 2B) was hardly discriminating. In the network based on bp 305–1,475 (Fig. 2C), the Sudeten and Carpathian samples retrieved from GenBank appear all in cluster E.
Median–joining haplotype networks, based on different COI segments. Circle diameters are proportional to the frequency of the haplotype. Red diamonds represent hypothetical haplotypes. A. Based on 168 3’P+5’P sequences of Erebiaeuryale. Inserts: cluster A and sub cluster E2, based on the 3’P section of the same 168 sequences, together with 77 sequences retrieved from GenBank (Vila et al. 2011). Haplotypes are numbered 1 through 61; B. Based on the 5’P section of the same 168 sequences plus 23 sequences retrieved from GenBank (Dincă et al. 2010); C. Based on bp. 305–1,475 of the same 168 sequences plus 36 sequences mined from GenBank (Paučulová et al. 2017). In Fig. 2C only the part corresponding to cluster E in Fig. 2A is depicted. Colour legends in frame B apply to Figures 2 to 4.
https://binary.pensoft.net/fig/635881
The relationship between subspecies, clusters and haplotypes is shown in Table 2. The clusters in the network turned out largely congruent with morphologic subspecies. Cluster A (Fig. 2A) includes all haplotypes of the western extra-Alpine populations: the Jurassian ssp.tramelana, the Pyrenean ssp.pyraenaeicola and the Cantabrian ssp.cantabricola.
The relationship between subspecies and haplotype, based on the sequences underlying Fig. 2A (sample 1–22 in table 1). Subsp – subspecies, syr – syrmia, tram – tramelana, cant – cantabricola, pyr – pyraenaeicola, Clus – cluster, No – sample number, corresponding to Table 1, Hpl – haplotype number, corresponding to Fig. 2A. Field numbers indicate the number of individuals. In bold and italics: discrepancy of phenotype and haplotype.
Subsp
adyte
pseudoad.
kunzi
ocellaris
isarica
syr.
tram.
cant.
pyr.
Clus
No
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
Hpl
A
1
5
30
10
2
4
3
1
4
1
55
1
56
1
57
1
58
1
59
2
B
5
3
6
10
6
2
7
2
8
1
9
1
10
1
11
1
12
1
13
1
C
14
1
15
2
16
1
5
17
4
18
1
19
1
20
1
21
1
22
1
23
1
24
2
D
25
3
10
2
26
1
27
1
28
2
29
1
30
1
31
1
32
4
33
9
34
1
35
1
E
36
8
37
1
38
1
39
3
2
8
40
1
41
1
42
1
43
1
44
2
45
1
46
1
48
5
49
1
49
1
50
1
51
2
6
6
4
1
5
10
52
5
11
53
1
54
1
Subspecies adyte makes up cluster B, but several adyte specimens are placed in cluster E.
Cluster C contains all individuals of ssp. pseudoadyte. It has the highest genetic diversity of all subspecies with 11 haplotypes found in two localities.
Cluster D contains only haplotypes of ssp. kunzi, but some kunzi are assigned to cluster E. Western and eastern kunzi are not separated. The Monte Cavallo population (sample 11) builds a separate sub cluster D1, four mutational steps apart from D2 (Fig. 2A), but morphologically indistinguishable from it.
Cluster E is private to four subspecies (euryale, syrmia, isarica and ocellaris), and contains individuals of two subspecies that have their own private clusters (adyte and kunzi). In the first group there is a clear structuring: ssp. syrmia is found around the central axis of the network, from which four subclusters, E1 through E4, branch off (Fig. 2A, C). The sub-clusters E2 and E3 are private to ssp. isarica, E4 to ssp. euryale. Ssp. ocellaris occupies E1, but is also represented in E3, together with isarica, with which it has a documented hybrid zone (Cupedo 2010). The proportion of individuals with the shared haplotype increases with decreasing distance to the hybrid zone (Table 2). The genetic dichotomy in ssp. isarica (E2 and E3) coincides with a geographic-morphologic sub-division. E3 contains the western and northern Alpine populations, E2 the eastern Alpine populations. These groups differ in wing shape and wing pattern (pers. obs. FC).
As to the subspecies with a different private cluster, we suspect that their cluster E haplotypes are due to introgression or incomplete lineage sorting. In sample 4 there is 100% mismatch of phenotype and haplotype. Phenotypically it is an adyte enclave in isarica territory (Cupedo 2010). The haplotype matches the surrounding isarica populations. This fits with previous studies in which gene-flow between these subspecies was not reflected in morphology (Geiger and Rezbanyai 1982). In ssp. kunzi, on the other hand, the cluster E individuals were found in the most peripheral samples (7 and 10, Fig. 1), far from the boundary with ocellaris, with which they share haplotype 51 (Fig. 2A). This argues against a recent introgression. The fact that both western and eastern kunzi are involved, suggests an introgression event predating their separation.
Phylogenetic trees and timing of diversification
Bayesian Inference (BI) and Maximum Likelihood (ML) analyses based on the 1,474 bp alignment (Fig. 3) show similar topologies for the haplogroups A through D, but the position and relationships of haplogroup E are not well supported in any of the trees. The clade comprising ssp. adyte and the western extra Alpine subspecies (the network clusters A+B), is well-supported in both trees. The western extra Alpine subspecies are jointly monophyletic with good support, a monophyletic ssp. adyte is not supported. There is medium support for a monophyletic ssp. pseudoadyte, private cluster C in the network. The clades D1 and D2 are well supported in both trees, the monophyly of D1+D2, however, is not well supported in the BI analysis and only moderately supported by ML. The monophyly of individual clades E1 and E2 is also well supported in both analyses, E3 somewhat less so in BI, but their positions in the trees are not well resolved. Overall, the BI with molecular dating suggests that the diversification into the present subspecies took place between ~0.75 and ~2 My ago.
Phylogenetic trees based on the haplotypes in Fig. 2A, with clusters and sub-clusters collapsed into triangles. Colour codes as in Fig. 2, cluster labels as in Fig. 2A. A. Bayesian Inference tree. Support values on the branches indicate the Bayesian posterior probabilities. Blue node bars indicate the 95% posterior density for the Bayesian age estimates. Insert: estimates of the divergence time from the RevBayes approach, based on a calibrated split at 3.5 My BP between E.euryale and E.ligea (Peña et al. 2015). The time scale is in My ago. B. Maximum Likelihood tree. Support values on the branches indicate bootstrap support (5,000 replicates).
https://binary.pensoft.net/fig/635882Discussion
Three of our specific questions have implicitly been resolved.
(i) Both in the haplotype network (Fig. 2A), in the BI tree and in the ML tree (Fig. 3) the subspecies pseudoadyte and kunzi appear as distinct lineages, different from the morphologically similar ssp. adyte and ocellaris.
(ii) Western kunzi, morphologically intermediate between pseudoadyte and nomotypical (eastern) kunzi, is genetically inseparable from the eastern form (Fig. 2A) and has correctly been ascribed to ssp. kunzi.
(iii) Western ssp. isarica and ssp. ocellaris share a single haplotype (Table 2). Eight haplotypes derived from the shared one, by one or even by two mutations, are private. In addition, ssp. ocellaris has two haplotypes in the private sub cluster E1. With one out of seven (isarica) or eight (ocellaris) haplotypes shared (Table 2), we consider isarica and ocellaris distinct genetic lineages. The conclusion that they are genetically inseparable (Schmitt and Haubrich 2008) was based on the Kalkstein population, which is inseparable from isarica by its mtDNA as well (Table 2).
Linking lineages to Würm glacial refugia
The actual distribution pattern results from Holocene range extensions, originating in the Würm glacial refugia. Overlaying the potential Würm refugia with the current subspecies distributions shows a clear matching pattern.
Ssp. adyte occupies the entire southwestern Alps (Fig. 1). Its distribution area stretches between two known biogeographical borders (Cupedo and Doorenweerd 2020): the Isère valley in the west and the Valtellina (the Adda valley) in the east, suggesting survival in refugium (i) and in the western compartments of refugium (iii).
The territory of ssp. pseudoadyte is sharply delimited by the valleys of Adda and Adige (Fig. 4), matching a single compartment of refugium (iii). Its postglacial northward expansion was halted at the Adda and Adige valleys, which currently delimit its distribution.
The southern Alps east of Lake Como, showing the relationship between the current distribution (coloured areas) and hypothesized refugial areas (coloured dots) of the subspecies of E.euryale during the maximum Würm glaciation (MIS2). Dotted area – transitional populations isarica/ocellaris. Colour legends as in Fig. 2.
https://binary.pensoft.net/fig/635883
As to ssp. kunzi, all former peripheric nunataks between Val d’Adige and Valcellina, belonging to refugium (iii), harbour relict populations of this ssp. (Cupedo 2010; Bonato et al. 2014): the Monti Lessini, Monte Grappa, Col Visentin and Monte Cavallo (Fig. 4). Sub-cluster D1 to date is confined to its refugial massif, the Monte Cavallo, and is separated from D2 by the Piave valley (Fig. 4). Only sub-cluster D2, living across the Piave valley, could invade the southern Dolomites, where it failed to cross the deeply incised Cordevole valley (Fig. 4). Eastern and western D2, only differing in wing pattern, are separated by the Cismon-Brenta valley.
The ssp. ocellaris territory is considerably larger than that of ssp kunzi, but it includes only a short section of the southern pre-Alpine chain (Figs 1, 4). Its refugium along the Alpine border most probably stretched from the Valcellina to the Tagliamento valley, matching the eastern part of refugium (iii).
Ssp. isarica has by far the largest distribution of the Alpine subspecies (Fig. 1). We suppose that the two COI sub clusters represent populations originating from different glacial refugia. The current territory of cluster E2 comprises the vast refugial area (ii) on the eastern edge of the Alps (Fig. 4), whereas cluster E3 suggests refugia along the northern outskirt of the Alps (refugium iv), the exact locations of which remain unknown.
Reconstruction of pre-Würm phylogeography
The five-fold cluster structure of the network is superimposed by a geographic tripartition. We distinguish (i) a southern Alpine group, consisting of the clusters B, C and D, (ii) a circum–Pannonian group, coinciding with cluster E, in which Carpathian populations are combined with northern and eastern Alpine ones and (iii) a western extra Alpine group, equalling cluster A.
Ad (i). In the southern Alpine group the subspecies adyte, pseudoadyte and kunzi occupy adjacent territories. They share a sharply defined, coherent territory and a considerable genetic variation. The pairwise sequence difference within the clusters ranges from 0.68% to 1.15%. This suggests a long-term isolation within the actual distributional borders. Most probably the increasing incision of the dividing valleys, by repeated glacial erosion, resulted in a permanent vicariance, enabling their differentiation.
Ad (ii). In the circum–Pannonian group, the core area of cluster E in the haplotype network is occupied by the Southern and Eastern Carpathian ssp. syrmia (Figs 1A, 2A). Sub-clusters radiate from it into the Western Carpathians (ssp. euryale, Fig. 2C) and into the Alps (ssp. isarica and ssp. ocellaris, Fig. 2A).
This division into a southern Alpine group and a circum–Pannonian group suggests an ancient Alpine–Carpathian disjunction. This is in agreement with earlier allozyme data: Schmitt and Haubrich (2008) found the split between the Carpathian and Alpine lineages (in their study: ssp. syrmia and ssp. adyte) the most ancient disjunction in the studied E.euryale populations. Such an Alpine–Carpathian disjunction is typically found in species of humid grasslands of the montane-subalpine levels (Varga and Schmitt 2008), and is associated with a Pannonian refugium (refugium v). The common ancestor could, in theory, have been resident either in the Alps or in the Carpathians. The close genetic affinity between the outgroup E.ligea and the Carpathian samples in the network (Fig. 2A) suggests a Carpathian origin, but the phylogenetic tree lacks the statistical support to corroborate this scenario.
Subsequently, i.e., during subsequent cold periods, the southern Alpine group differentiated into its three lineages (adyte, pseudoadyte and kunzi), and the circum-Pannonian group split up into a Carpathian and an eastern Alpine lineage. This required a second, more recent glacial retreat in the Pannonian plain, again followed by interglacial invasion of both the Alps and the Carpathians. Finally, the Sudeten and Carpathian populations differentiated into what is now known as ssp. euryale and ssp. syrmia, while the Alpine populations differentiated into today’s ssp. ocellaris, eastern isarica and western isarica. We presume that this occurred during MIS2, when the refugia of the Alpine populations were localised in the Alpine periphery rather than in the Pannonian plain (Fig. 4), because the Alpine glacier extension was far less than during the foregoing glaciations (Penck and Brückner 1909; Castiglioni 2004; van Husen 2004).
Ad (iii) The western extra-Alpine group. Both the network and the ML tree suggest that it derived from cluster B (Figs 2A, 3B). Its star-like configuration, with many low frequency haplotypes derived from a most common one, is typical of a relatively recent demographic expansion after a genetic bottleneck (Rogers and Harpending 1992; Avise 2000; Hwang and Cho 2018). This bottleneck probably was associated with its separation from the Alpine cluster B (Fig. 2A). Because an Alpine–Pyrenean disjunction required sub alpine conditions in the in-between Rhone basin, it must have been associated with a glacial period. Because of the identical genetic content of the Pyrenean and Jurassian populations, we exclude two independent colonisation events. We assume that either the Pyrenees and the Jura were colonised simultaneously, or the Jura was colonised secondarily from the Pyrenees.
However, nuclear data suggest a somewhat different scenario. Schmitt and Haubrich (2008) found the Pyrenean population genetically closer to isarica than to adyte. Although their data refer only to eastern isarica, this suggests a connection to the northern, rather than to the southern Alps. This mismatch between mitochondrial and nuclear data calls for further study on the origin of the western extra-Alpine group of E.euryale. In general, mitochondrial and nuclear DNA need not have responded similarly to past biogeographic events, and their patterns of variation provide complementary data on the phylogeographic history (Hinojosa et al. 2019). We expect additional genomic data to enhance the insight in the origin of the western extra-Alpine populations.
Morphology
Some morphologic taxonomic characters have been shown phylogenetically relevant, provided they are combined with molecular data. (Wahlberg and Nylin 2005; Simonsen et al. 2006; Pisani et al. 2007; Stephanović et al. 2016). In our study, the three morphologic subspecies that make up the southern Alpine group (adyte, pseudoadyte and kunzi) appeared congruent with well separated haplotype clusters (Fig. 2A). These three subspecies (together with the Apennine ssp brutiorum which is not included in this study), share a distinctive character: they have white pupiled apical ocelli on the forewing upper side. All remaining subspecies (including the boreal/tundral populations, not studied either), have uniform black ocelli. The pupiled ocelli appear to be a synapomorphy of the southern Alpine group. This supports the conclusion that the separation of the southern Alpine group was a basal split in the history of Erebiaeuryale. It further suggests that the Apennine populations are derived from the southern Alpine group, and that the boreal/tundral populations are derived from the circum-Pannonian group. Both hypotheses are consistent with geography. Moreover, it supports the alternative, allozyme based scenario for the origin of the western extra-Alpine group, in which the apical ocelli are black.
Overall timeline of diversification
The Bayesian estimate of divergence times, using an external calibration, dates the differentiation of the crown group roughly 0.75 My to 2 My ago, i.e., in the early Pleistocene (Appendix 1: Fig. A1). However, the correlation of lineage splits to glacial conditions provides an alternative calibration for the age of splitting events (Schmitt et al. 2016; Hinojosa et al. 2018). For the relatively young Pyrenean disjunction we assume at least a MIS6 age, because there is some evidence for more recent (i.e., MIS2) differentiations within this cluster: (i) genetic differences between populations at both sides of the Cantabrian watershed are a possible indication of two Würm glacial refugia in the Cantabrian region (Vila et al. 2011). (ii) Likewise, the morphological differentiation of the Pyrenean populations into ssp. antevortes in the north-western Pyrenees (not investigated here) and ssp. pyraenaeicola in its southern and eastern parts, must be younger than the Alpine–Pyrenean disjunction.
Because the differentiation of the clusters B, C and D in their southern Alpine refugia predates the Alpine–Pyrenean disjunction (Fig. 3), it is at the latest of MIS10 age. The initial Alpine–Carpathian disjunction, necessarily associated with a severe cold phase, thus most probably is not younger than MIS12. We emphasise that the ages of disjunctions reconstructed this way are always minimal ones and splitting events might prove considerably older.
The dependency of major cold phases, however, sets a maximum age too. It has been shown that tributary glaciers in the eastern Alps did not reach the main valleys during minor cold periods like MIS4 (van Andel and Tzedakis 1996; van Husen 2004), therefore subalpine organisms were not dispelled into refugial isolation. This was only the case during the more severe cold phases. The very first of those major ice ages was MIS22, ca 870–880 ka ago (Ehlers and Gibbard 2007) (Appendix 1: Fig. A1). All foregoing cold phases were in the order of MIS4 or less. Consequently, it is unlikely that the first Alpine–Carpathian disjunction took place more than 1 My ago. This conflicts with our Bayesian dating (Fig. 3A), secondary calibrated with Peña et al. (2015). It is in line, though, with recent molecular studies on the E.tyndarus group (Schmitt et al. 2016) and the E. epiphron–orientalis group (Hinojosa et al. 2018). Both calculated divergence times, in the Pliocene–Pleistocene period, that are considerably younger than those inferred from secondary calibrations based on the external calibration from Peña et al. (2015).
Conclusions
In Erebiaeuryale the 3’P section of the COI gene is far more discriminative than the 5’P section. It revealed the presence of five intraspecific clades, three of which remained hidden using the barcode section alone. The clades are congruent with morphologic subspecies. One of these, ssp. isarica, seems even composed of two genetically different lineages. That makes six lineages in the Alps, four of which were not recognised earlier.
All subspecies addressed here, i.e. all but the boreal component and the Apenninian populations, are derived from a common ancestor, supposedly resident in the Carpathians. This ancestral population split up into two main clades, an Alpine one and a Carpathian one, probably no more than 1 My ago. This dichotomy is consistent with a glacial refugium in the Pannonian plain. During mid-Pleistocene, partitioning of the southern Alps by glacial valleys led to the differentiation of three lineages: adyte, pseudoadyte and kunzi. The Carpathian branch colonised, after a second refugial retreat in the Pannonian plain, both the Carpathians and the eastern Alps. In a subsequent cold period the Alpine populations differentiated into western isarica, eastern isarica and ocellaris, whereas the Carpathian populations differentiated into euryale and syrmia. In the western Alps, a Pyrenean–Jurassian clade was separated from an Alpine clade, at the latest during MIS6. Our COI data support a split from adyte, but allozyme data endorse a western isarica rooting. Postglacially, the southern Carpathian ssp. syrmia spread into the Balkans, and the Alpine ssp. isarica and ocellaris built a broad secondary contact zone. There is, however, no doubt that additional genomic data in future work will enhance the resolution of the relationships and of the timing of splitting events. The use of nuclear genomic data would also address the possibility that the divergences we detected in COI have been influenced by Wolbachia infections, which have been detected in E.Euryale too (Ritter et al. 2013; Lucek et al. 2021).
Acknowledgements
This research was financially supported by a grant (SUB.2013.05.10) of the Uyttenboogaart–Eliasen Stichting in the Netherlands for which we are particularly indebted. We are most grateful to Kay Lucek (Basel, CH) and Vlad Dincă (Oulu, FI), whose comments contributed substantially to the final version of the manuscript.
ReferencesAlbreJGersCLegalL (2008) Molecular phylogeny of the Erebiatyndarus (Lepidoptera, Rhopalocera, Nymphalidae, Satyrinae) species group combining CoxII and ND5 mitochondrial genes: a case study of a recent radiation.47: 196–210. https://doi.org/10.1016/j.ympev.2008.01.009AviglianoRCalderoniGMonegatoGMozziP (2002) The late Pleistocene-Holocene evolution of the Cellina and Meduna alluvial fans (Friuli NE Italy).57: 133–139.AviseJC (2000) Harvard University Press, Massachusetts, 447 pp.BonatoLUlianaMBerettaS (2014) Fondazione Musei Civici di Venezia, Venezia, 391 pp.BuoncristianiJFCampyM (2004) The palaeogeography of the last two glacial episodes in France: the Alps and Jura. In: EhlersJGibbardPL (Eds) Quaternary Glaciations, extent and chronology Part I: Europe., 101–110. https://doi.org/10.1016/S1571-0866(04)80059-9CalvetM (2004) The Quaternary glaciations of the Pyrenees. In: EhlersJGibbardPL (Eds) Quaternary Glaciations, extent and chronology Part I: Europe., 119–128. https://doi.org/10.1016/S1571-0866(04)80062-9CastiglioniGB (2004) Quaternary glaciations in the eastern sector of the Italian Alps. In: EhlersJGibbardPL (Eds) Quaternary Glaciations, extent and chronology Part I: Europe., 209–214. https://doi.org/10.1016/S1571-0866(04)80072-1CohenKMGibbardPL (2019) Global chronostratigraphical correlation table for the last 2.7 million years, version 2019 QI-500.500: 20–31. https://doi.org/10.1016/j.quaint.2019.03.009ComesPKadereitJW (1998) The effect of quaternary climatic changes on plant distribution and evolution.3: 432–438. https://doi.org/10.1016/S1360-1385(98)01327-2CrandallKTempletonAR (1993) Empirical Tests of Some Predictions From Coalescent Theory With Applications to Intraspecific Phylogeny Reconstruction.134: 959–969. https://doi.org/10.1093/genetics/134.3.959CupedoF (2010) A revision of the infraspecific structure of Erebiaeuryale (Esper, 1805) (Nymphalidae; Satyrinae).33: 85–106.CupedoF (2014) Reproductive isolation and intraspecific structure in Alpine populations of Erebiaeuryale (Esper, 1805) (Lepidoptera, Nymphalidae, Satyrinae).37: 19–36. https://doi.org/10.3897/nl.37.7935CupedoFDoorenweerdC (2020) The intraspecific structure of the yellow-spotted ringlet Erebiamanto ([Denis & Schiffermüller], 1775), with special reference to the bubastis group: an integration of morphology, allozyme and mtDNA data (Lepidoptera, Nymphalidae, Satyrinae). Nota Lepidopterologica: 43–60. https://doi.org/10.3897/nl.43.47409de BeaulieuJ-LAndrieuVPonelPReilleM (1994) The Weichselian late-glacial in southwestern Europe (Iberian Peninsula, Pyrenees, Massif Central, northern Appenines).9: 101–107. https://doi.org/10.1002/jqs.3390090203de LattinG (1967) Fischer Verlag, Stuttgart, 499 pp.DincăVDapportoLSomervuoPVodăRCuvelierSGascoigne-PeesMHuemerPMutanenMHebertPDNVilaR (2021) High resolution DNA barcode library for European butterflies reveals continental patterns of mitochondrial genetic diversity. Communications Biology 4: e315. https://doi.org/10.1038/s42003-021-01834-7DincăVZakharovEVHebertPDNVilaR (2010) Complete DNA barcode reference library for a country’s butterfly fauna reveals high performance for temperate Europe. Proceedings of the Royal Society Biological Sciences Series B: 1–9. https://doi.org/10.1098/rspb.2010.1089DoorenweerdCvan HarenMSchermerMPieterseSvan NieukerkenE (2014) A Linnaeus NG (TM) interactive key to the Lithocolletinae of North-West Europe aimed at accelerating the accumulation of reliable biodiversity data (Lepidoptera, Gracillariidae).422: 87–101. https://doi.org/10.3897/zookeys.422.7446EhlersJGibbardPL (2007) The extent and chronology of Cenozoic Global Glaciation. Quaternary International 164–165: 6–20. https://doi.org/10.1016/j.quaint.2006.10.008FeurdanA (2005) Holocene forest dynamics in northwestern Romania.15: 435–446. https://doi.org/10.1191/0959683605hl803rpFolmerOBlackMHoehWLutzRVrijenhoekR (1994) DNA primers for amplification of mitochondrial cytochrome c oxidase subunit I from diverse metazoan invertebrates.3: 294–299.GeigerHRezbanyaiL (1982) Enzymelektrophoretische Untersuchungen über die Verwandtschaftsbeziehungen bei Erebia, mit besonderer Berücksichtigung der Taxa euryaleisarica Rühl und adyte Hübner (Lep.: Satyridae).92: 49–63.Geologische Bundesanstalt (AT) (2013) Der Alpenraum zum Höhepunkt der letzten Eiszeit. Rekonstruktion der maximalen Gletscherausbreitung während des Höhepunktes der letzten Eiszeit (Würm) von 26 000 bis 20 000 Jahren vor Heute. https://opac.geologie.ac.at/wwwopacx/wwwopac.ashx?command=getcontent&server=images&value=Poster_Alpenraum%20Eiszeit_opt.pdf [accessed 30-10-2019]GonsethY (1987) Schweizerischer Bund für Naturschutz, Neuchâtel, 242 pp.González-SampérizPValero-GarcésBLCarriónJSPeña-MonnéJLGarcía-RuizJMMartí-BonoC (2005) Glacial and lateglacial vegetation in northeastern Spain: New data and a review. Quaternary International 140–141: 4–20. https://doi.org/10.1016/j.quaint.2005.05.006GrafAAkçarNIvy-OchsSStraskySKubikPWChristlMBurkhardMWielerRSchlüchterC (2015) Multiple advances of Alpine glaciers into the Jura Mountains in the Northwestern Switzerland.108: 225–238. https://doi.org/10.1007/s00015-015-0195-yHäuselmannPFiebigMKubikPWAdrianH (2007) A first attempt to date the “Deckenschotter”of Penck and Brückner with cosmogenic nuclides. Quaternary International 164–165: 33–42. https://doi.org/10.1016/j.quaint.2006.12.013HeadMJGibbardPL [Eds] (2005) Geological Society, London, 329 pp.HebertPDNCywinskaABallSLde WaardJR (2003) Biological identifications through DNA barcodes.270: 313–321. https://doi.org/10.1098/rspb.2002.2218HeinzCBarbazaM (1998) Environmental changes during the Late Glacial and Post-Glacial in the central Pyrenees (France): new charcoal analysis and archaeological data. Review of Palaeobotany and Palynology.104: 1–17. https://doi.org/10.1016/S0034-6667(98)00050-5HeßM (2013) Das Allgäu in der Eiszeit und deren Auswirkungen. http://www.georgianum-lingen.de/cms/upload/Fachbereiche/Erdkunde/Allgau_in_der_Eiszeit.pdf [accessed 30.10.2019]HewittGM (1996) Some genetic consequences of ice ages, and their role in divergence and speciation.58: 247–276. https://doi.org/10.1006/bijl.1996.0035HewittGM (1999) Post-glacial re-colonization of European biota.68: 87–112. https://doi.org/10.1111/j.1095-8312.1999.tb01160.xHinojosaJCKoubínováDSzenteczkiMAPitteloudCDincăVVilaR (2019) A mirage of cryptic species: Genomics uncover striking mitonuclear discordance in the butterfly Thymelicussylvestris.28: 3857–3868. https://doi.org/10.1111/mec.15153HinojosaJCMonasterioYEscorbésRDincăVVilaR (2018) Erebiaepiphron and Erebiaorientalis: sibling butterfly species with contrasting histories.20: 1–11. https://doi.org/10.1093/biolinnean/bly182HöhnaSLandisMHeathTBoussauBLartillotNMooreBHuelsenbeckJRonquistF (2016) RevBayes: Bayesian Phylogenetic Inference Using Graphical Models and an Interactive Model-Specification Language.65: 1–11. https://doi.org/10.1093/sysbio/syw021HoldhausK (1954) Universitätsverlag Wagner, Innsbruck, 493 pp.HwangJ-YChoG-J (2018) Identification of novel haplotypes and interpretation of gene flow of mitochondrial DNA control region of Eurasian otter (Lutralutra) for the effective conservation.80: 1791–1800. https://doi.org/10.1292/jvms.17-0678JanovskaVPokornýP (2008) Forest vegetation of the last full-glacial period in the Western Carpathians (Slovakia and Czech Republic).80: 307–324.JuřičkováLHoráčkováJLožekV (2014) Direct evidence of central European forest refugia during the last glacial period based on mollusc fossils.82: 222–228. https://doi.org/10.1016/j.yqres.2014.01.015KellerOKrayssE (2010) Mittel- und spätpleistozäne Stratigraphie und Morphogenese in Schlüsselregionen der Nordschweiz.59: 88–119. https://doi.org/10.3285/eg.59.1-2.08KleckováIVrbaPKonvickaM (2015) Quantitative evidence for spatial variation in the biennial life cycle of the mountain butterfly Erebiaeuryale (Lepidoptera: Nymphalidae) in the Czech Republic.112: 114–119. https://doi.org/10.14411/eje.2015.003KropfMKadereitJWComesHP (2002) Late Quaternary distributional stasis in the submediterranean mountain plant Anthyllismontana L. (Fabaceae) inferred from ITS sequences and amplified fragment length polymorphism markers.11: 447–463. https://doi.org/10.1046/j.1365-294X.2002.01446.xKuklaG (2005) Saalian supercycle, Mindel/Riss interglacial and Milankivitch’s dating.24: 1573–1583. https://doi.org/10.1016/j.quascirev.2004.08.023KunešPPelánkováBChitrýMJanovskáVPokornýP (2008) Interpretation of the last-glacial vegetation of eastern-central Europe using modern analogues from southern Siberia.35: 2223–2236. https://doi.org/10.1111/j.1365-2699.2008.01974.xLangG (1994) Fischer, Jena, 462 pp.LisieckiLERaymoME (2005) A Pliocene-Pleistocene stack of 57 globally distributed benthic δ18O records.20: 1–17. https://doi.org/10.1029/2004PA001071LucekKButlinRKPatsiouT (2020) Secondary contact zones of closely-related Erebia butterflies overlap with narrow phenotypic and parasitic clines.33: 1152–1163. https://doi.org/10.1111/jeb.13669LucekKBouaouinaSJospinAGrillAde VosJ (2021) Prevalence and relationship of endosymbiontic Wolbachia in the genus Erebia. BMC Ecology and Evolution 21: e95. https://doi.org/10.1186/s12862-021-01822-9MarazziS (2005) Priuli & Verlucca, Torino, 416 pp.MartinJ-FGillesADescimonH (2000) Molecular phylogeny and evolutionary patterns of the european satyrids (Lepidoptera: Satyridae) as revealed by the mitochondrial gene sequence.15: 70–82. https://doi.org/10.1006/mpev.2000.0757MartinJ-FGillesALörtscherMDescimonH (2002) Phylogenetics and differentiation among the western taxa of the Erebiatyndarus group (Lepidoptera; Nymphalidae).75: 319–332. https://doi.org/10.1111/j.1095-8312.2002.tb02073.xMartinsonDGPisiasNGHaysJDImbrieJMooreTCShackletonNJ (1987) Age dating and the orbital theory of the ice ages: development of a high resolution 0–300,000 year chronostratigraphy.27: 1–29. https://doi.org/10.1016/0033-5894(87)90046-9MonegatoGRavazziCDoneganaMPiniRCalderoniGWickL (2007) Evidence of a two-fold glacial advance during the last glacial maximum in the Tagliamento end moraine system (eastern Alps).68: 284–302. https://doi.org/10.1016/j.yqres.2007.07.002MüllerCMProssJBibusE (2003) Vegetation response to rapid climate change in Central Europe during the past 140,000 yr based on evidence from the Füramoos pollen record.59: 235–245. https://doi.org/10.1016/S0033-5894(03)00005-XMusterCBerendonkTU (2006) Divergence and diversity: lessons from an arctic-alpine distribution (Pardosasaltuaria group, Lycosidae).15: 2921–2933. https://doi.org/10.1111/j.1365-294X.2006.02989.xNguyenL-TSchmidtHAvon HaeselerAMinhBQ (2015) IQ-TREE: A fast and effective stochastic algorithm for estimating maximum likelihood phylogenies.32: 268–274. https://doi.org/10.1093/molbev/msu300ParisodCBesnardG (2007) Glacial in situ survival in the Western Alps and polytopic autopolyploidy in Biscutellalaevigata L. (Brassicaceae).16: 2755–2767. https://doi.org/10.1111/j.1365-294X.2007.03315.xPaučulováLŠemelákováMMutanenMPristasPPanigajL (2017) Searching for the glacial refugia of Erebiaeuryale (Lepidoptera, Nymphalidae) – insights from mtDNA- and nDNA-based phylogeography in the Western Carpathians.55: 118–128. https://doi.org/10.1111/jzs.12156PeñaCMalmT (2012) VoSeq: a voucher and DNA sequence web application. PLoS ONE 7(6): e39071. https://doi.org/10.1371/journal.pone.0039071PeñaCWitthauerHKleckováIFricZWahlbergN (2015) Adaptive radiations in butterflies: evolutionary history of the genus Erebia (Nymphalidae: Satyrinae).116: 449–467. https://doi.org/10.1111/bij.12597PenckABrücknerE (1909) Tauchnitz, Leipzich, 1199 pp.PisaniDBentonMJWilkinsonM (2007) Congruence of morphological and molecular phylogenies.55: 269–281. https://doi.org/10.1007/s10441-007-9015-8RambautADrummondAJXieDBaeleGSuchardMA (2018) Posterior Summarization in Bayesian Phylogenetics Using Tracer 1.7.67: 901–904. https://doi.org/10.1093/sysbio/syy032RatnasinghamSHebertPDN (2007) BOLD: The Barcode of Life Data System (www.barcodinglife.org).7: 355–364. https://doi.org/10.1111/j.1471-8286.2007.01678.xRavazziCOrombelliGTanziG (2004) An outline of the flora and vegetation of Adriatic bassin (Northern Italy and eastern side of the Apennines) during the Last Glacial Maximum. In: AntonioliFVaiGB (Eds) Litho-paleoenvironmental maps of Italy during the Last Two Climatic Extremes; Explanatory Notes 32nd International Geological Congress., 15–20.ReinigWF (1938) Fischer, Jena, 146 pp.ReitnerJM (2011) Das Inngletschersystem während des Würmglazials. Arbeitstagung der Geologischen Bundesanstalt 2011 – Achenkirch Beiträge: 79–88.Rezbanyai-ReserL (1991) Die drei Zentralschweizer Kontaktstellen der Erebiaeuryale – Unterarten isarica Heyne und adyte Hbn. (Lep., Satyridae).25: 77–90.RitterSMichalskiSGSetteleJWiemersMFricZFSielezniewMŠašićMRozierYDurkaW (2013) Wolbachia infections mimic cryptic speciation in two parasitic butterfly species, Phengaristeleius and P.nausithous (Lepidoptera, Lycaenidae). PLoS ONE 8(11): e78107. https://doi.org/10.1371/journal.pone.0078107RogersARHarpendingH (1992) Population growth makes waves in the distribution of pairwise genetic differences.9: 552–569.RudnerZESümegiP (2001) Recurring Taiga forest-steppe habitats in the Carpathian Basin in the Upper Weichselian. Quaternary International 76/77: 177–189. https://doi.org/10.1016/S1040-6182(00)00101-4SchmittT (2007) Molecular biogeography of Europe: Pleistocene cycles and postglacial trends.4: 11–23. https://doi.org/10.1186/1742-9994-4-11SchmittT (2009) Biogeographical and evolutionary importance of the European high mountain systems. Frontiers in Zoology 6: e9. https://doi.org/10.1186/1742-9994-6-9SchmittTHabelJCRödderDLouyD (2014) Effects of recent and past climatic shifts on the genetic structure of the high mountain Yellow-spotted ringlet butterfly Erebiamanto (Lepidoptera, Satyrinae): a conservation problem.20: 2045–2061. https://doi.org/10.1111/gcb.12462SchmittTHaubrichK (2008) The genetic structure of the mountain forest butterfly Erebiaeuryale unravels the late Pleistocene and postglacial history of the mountain coniferous forest biome in Europe.17: 2194–2207. https://doi.org/10.1111/j.1365-294X.2007.03687.xSchmittTHewittGMMüllerP (2006) Disjunct distribution during glacial and interglacial periods in mountain butterflies: Erebiaepiphron as an example.19: 108–113. https://doi.org/10.1111/j.1420-9101.2005.00980.xSchmittTLouyDZimmermannEHabelJC (2016) Species radiation in the Alps: multiple range shifts caused diversification in Ringlet butterflies in the European high mountains.16: 791–808. https://doi.org/10.1007/s13127-016-0282-6SchmittTMüllerP (2007) Limited hybridization along a large contact zone between two genetic lineages of the butterfly Erebiamedusa (Satyrinae, lepidoptera) in Central Europe.45: 39–46. https://doi.org/10.1111/j.1439-0469.2006.00404.xSchönswetterPStehlikIHoldereggerRTribschA (2005) Molecular evidence for glacial refugia of mountain plants in the European Alps.14: 3547–3555. https://doi.org/10.1111/j.1365-294X.2005.02683.xSchönswetterPTribschASchneeweissGMNiklfeldH (2003) Disjunctions in relict alpine plants: phylogeography of Androsacebrevis and A.wulfeniana (Primulaceae).141: 437–446. https://doi.org/10.1046/j.0024-4074.2002.00134.xSchönswetterPTribschAStehlikINiklfeldH (2004) Glacial history of the high alpine Ranunculusglacialis (Ranunculaceae) in the European Alps in a comparative phylogeographical context.81: 183–195. https://doi.org/10.1111/j.1095-8312.2003.00289.xSimonsenTJWahlbergNBrowerAVZde JongR (2006) Morphology, molecules and fritillaries: approaching a stable phylogeny for Argynnini (Lepidoptrea; Nymphalidae).37: 405–418. https://doi.org/10.1163/187631206788831407SondereggerP (2005) Selbstverlag, Biel/Bienne, 712 pp.StehlikI (2000) Nunataks and peripheral refugia for alpine plants during quaternary glaciation in the middle parts of the Alps.110: 25–30.StehlikI (2003) Resistance or emigration? Response of alpine plants to the ice ages.52: 499–510. https://doi.org/10.2307/3647448StehlikISchnellerJJBachmannE (2002) Immigration and in situ glacial survival of the low-alpine Erinusalpinus (Scrophulariaceae).77: 87–103. https://doi.org/10.1046/j.1095-8312.2002.00094.xStephanovićSKosovacAKrstićOJovićJToševskiI (2016) Morphology versus DNA barcoding: two sides of the same coin. A case study of Ceutorhynchuserysimi and C.contractus identification.23: 638–648. https://doi.org/10.1111/1744-7917.12212StojakowitsPMayrCIvy-OchsSPreusserFReitnerJSpötlC (2020) Environments at the MIS 3/2 transitions in the northern Alps and their foreland. Quaternary International 581–582: 99–113. https://doi.org/10.1016/j.quaint.2020.08.003TribschASchönswetterP (2003) Patterns of endemism and comparative phylogeography confirm palaeoenvironmental evidence for Pleistocene refugia in the Eastern Alps.52: 477–497. https://doi.org/10.2307/3647447TzedakisPCLawsonITFrogleyMRHewittGMPreeceRC (2002) Buffered tree population changes in a quaternary refugium: evolutionary implications.297: 2044–2047. https://doi.org/10.1126/science.1073083van AndelTHTzedakisPC (1996) Palaeolithic landscapes of Europe and environs 150,000–25,000 years ago: an overview.15: 481–500. https://doi.org/10.1016/0277-3791(96)00028-5van HusenD (2004) Quaternary glaciations in Austria. In: EhlersJGibbardPL (Eds) Quaternary Glaciations, extent and chronology Part I: Europe., 1–13. https://doi.org/10.1016/S1571-0866(04)80051-4VargaZ (1996) Biogeography and evolution of oreal lepidoptera in the Palaearctic.42: 289–330.VargaZSchmittT (2008) Types of oreal and oreotundral disjunction in the western Palearctic.93: 415–430. https://doi.org/10.1111/j.1095-8312.2007.00934.xVilaMBjörklundM (2004) Testing biennialism in the butterfly Erebiapalarica (Nymphalidae: Satyrinae) by mtDNA sequencing.13: 213–217. https://doi.org/10.1111/j.0962-1075.2004.00472.xVilaMMarí-MenaNGuerreroASchmittT (2011) Some butterflies do not care about topography: a single genetic lineage of Erebiaeuryale (Nymphalidae) along the northern Iberian mountains.49: 119–132. https://doi.org/10.1111/j.1439-0469.2010.00587.xVilaMVidal-RomaníJRBjörklundM (2005) The importance of time scale and multiple refugia: incipient speciation and admixture of lineages in the butterfly Erebiatriaria (Nymphalidae).36: 249–260. https://doi.org/10.1016/j.ympev.2005.02.019WahlbergNNylinS (2005) Morphology versus molecules: resolution of the positions of Nymphalis, Polygonia and related genera (Lepidoptera: Nymphalidae).19: 213–223. https://doi.org/10.1111/j.1096-0031.2003.tb00364.xWiemersMChazotNWheatCWSchweigerOWahlbergN (2020) A complete time-calibrated multi-gene phylogeny of the European butterflies.938: 97–124. https://doi.org/10.3897/zookeys.938.50878WillienP (1985) Contribution lépidoptérique française à la Cartographie des Invertébrés Européens (C.I.E.) et travail préliminaire à l’établissement des Atlas nationaux du Secrétariat de la Faune et de la Flore (S.F.F.). Lépidoptères NymphalidaeSatyrinae: Erebia.14: 147–158.WillisKJvan AndelTH (2004) Trees or no trees? The environments of central and eastern Europe during the Last Glaciation.23: 2369–2387. https://doi.org/10.1016/j.quascirev.2004.06.002Appendix 110.3897/nl.45.68138.figurea13E59FDB7-7A42-5526-A9F7-1FB8B31B421F
Global temperature curve of the Quaternary period, based on the 16O/18O ratio in benthic Foraminifera (Martinson et al. 1987; Lisiecki and Raymo 2005). A. Epoch; Red – Holocene, yellow/orange – Pleistocene, green – Pliocene. B. Temperature curve. Flanking numbers indicate MIS stages: even numbers (left)=cold stages, odd numbers (right)=warm stages. From Cohen and Gibbard (2019), modified.
Explanation note: Table S1. Partial COI sequences, mined from GenBank. Morphological subspecies, accession numbers, geographical locations, and number of individuals per sequence.
https://binary.pensoft.net/file/635885This dataset is made available under the Open Database License (http://opendatacommons.org/licenses/odbl/1.0/). The Open Database License (ODbL) is a license agreement intended to allow users to freely share, modify, and use this Dataset while maintaining this same freedom for others, provided that the original source and author(s) are credited.Frans Cupedo, Camiel Doorenweerd