Mosaic distribution of cytotypes in a mixed-ploidy plant species: nested environmental niches but low geographic overlap

PDF Proof 1Mosaic distribution of cytotypes in a mixed-ploidy plant species: nested 2environmental niches but low geographic overlap 3 4 5 M. Castro1, J. Loureiro1, M. Serrano2, D. Tavares1, B. C. Husband3, C. Siopa1, S. Castro1,4 61 CFE, Centre for Functional Ecology, Department of Life Sciences, University of Coimbra, Calçada Martim 7de Freitas, 3000-456 Coimbra, Portugal 82 Department of Botany, Faculty of Pharmacy, University of Santiago de Compostela, Galicia, Spain 93 Department of Integrative Biology, University of Guelph, Guelph, Ontario, N1G 2W1, Canada 10 4 Botanic Garden of the University of Coimbra, Calçada Martim de Freitas, 3000-456 Coimbra, Portugal 11 12 Short Running Title: Cytogeography in Jasione montana 13 14 For correspondence: 15 Mariana Castro 16 E-mail: [email protected] 17 Address: CFE, Centre for Functional Ecology, Department of Life Sciences, University of Coimbra, 18 Calçada Martim de Freitas, 3000-456 Coimbra, Portugal Page 1 of 60 Botanical Journal of the Linnean Society Botanical Journal of the Linnean Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 PDF Proof 19 ABSTRACT 20 Polyploids often have divergent geographic ranges compared to their diploid progenitors, but 21 the causes of such differentiation are poorly understood. The geographic ranges of cytotypes 22 within polyploid complexes may be caused by multiple factors, including historical events, 23 interactions among cytotypes, and divergent environmental tolerances, but the fine-scale 24 geographical arrangement of cytotypes is rarely known for most mixed-ploidy species. In this 25 study, we assessed cytotype diversity and distribution patterns in the Jasione montana polyploid 26 complex and examined whether environmental factors can explain the occurrence of 27 tetraploids. Specifically, we reviewed all chromosome counts available in the literature, 28 examined cytotype distributions in a large-scale population survey using flow cytometry (N = 29 278 populations, N = 3396 plants), and used niche modelling to compare cytotype 30 environmental associations. Two cytotypes were detected: diploids, which are widespread 31 across Europe, and tetraploids, restricted to the northwest quadrant of the Iberian Peninsula. 32 The two cytotypes were distributed in a mosaic with areas dominated by diploids intermixed 33 with those dominated by tetraploids, rarely forming mixed-ploidy populations (1.4%). Although 34 having low geographical overlap, the tetraploid niche is fully nested within the diploid niche 35 breadth and occupies only a subset of the environmental envelope of the diploid progenitor, 36 suggesting that polyploidization has not caused niche expansion due to novel environmental 37 preferences. The mosaic diploid-tetraploid contact zones and the lack of mixed-ploidy 38 populations suggest that frequency-dependent selection may play a role in excluding minority 39 cytotypes. Under this scenario, tetraploids would have to disperse to places unoccupied by 40 diploids to successfully establish. The aggregation of tetraploid populations in areas suitable for 41 diploids suggests that tetraploids may also outcompete diploids in certain areas. Collectively, 42 our results indicate that environmental sorting has played a role, at least on a broader scale, in 43 the successful establishment of polyploids in J. montana. 44 Page 2 of 60 Botanical Journal of the Linnean Society Botanical Journal of the Linnean Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 PDF Proof 45 ADDITIONAL KEYWORDS: diploids, flow cytometry, Jasione montana, minority cytotype 46 exclusion, niche modelling, polyploidy, tetraploids. Page 3 of 60 Botanical Journal of the Linnean Society Botanical Journal of the Linnean Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 PDF Proof 47 INTRODUCTION 48 The duplication of whole chromosome sets is a common event in nature (Wood et al., 2009; 49 Husband, Baldwin & Suda, 2013; Marques et al., 2017), giving rise to new polyploids that might 50 establish and spread within diploid/lower-ploidy progenitor populations. The recurrent 51 formation of polyploids has been documented multiples times throughout the evolutionary 52 history of particular plant groups (Wood et al., 2009; Otto & Whitton, 2000; Soltis et al., 2010), 53 but also in extant plant populations (e.g., Maceira et al., 1992; Burton & Husband, 2001; Ramsey, 54 2007; Castro et al., 2018). This is likely due to the frequent production of unreduced gametes in 55 nature (Bretagnolle & Thompson, 1995; Ramsey, 2007; Brownfield & Kohler, 2010), although 56 successful polyploid establishment depends on numerous factors (Levin, 1975; Rodriguez, 57 1996). Because polyploids arise frequently and can have significant ecological and evolutionary 58 effects, polyploidy is currently recognized as a major mechanism of sympatric speciation (Otto 59 & Whitton, 2000; Soltis et al., 2010), being considered an important mechanism of 60 diversification of flowering plants (Soltis & Soltis, 1999). Given its significant contribution, the 61 factors involved in the successful establishment of polyploid lineages have received increased 62 attention in the last decades. 63 Immediately after polyploid formation, the new cytotype is at a numerical disadvantage 64 within the population of its diploid/lower-ploidy progenitor. Theoretical models suggest that the 65 new polyploid can establish within the population of the progenitor only if it has the necessary 66 conditions to increase its numbers; otherwise, it will be excluded from the population due to 67 frequency-dependent selection (Levin, 1975; Rodriguez, 1996; Husband & Schemske, 2000). 68 Polyploid establishment will be favoured by features that increase the probability of successful 69 mating, such as recurring formation of polyploids, spatial clustering, perenniality, increased 70 selfing, and/or increased competitive ability (Fowler & Levin, 1984; Felber, 1991; Rodriguez, 71 1996; Husband & Schemske, 2000; Barringer, 2007). Assortative mating, enforced by various 72 reproductive barriers, may also promote the coexistence of polyploids and their progenitors Page 4 of 60 Botanical Journal of the Linnean Society Botanical Journal of the Linnean Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 PDF Proof 73 [e.g., Chamerion angustifolium (L.) Holub, Husband & Sabara 2004; Aster amellus L., Jersáková 74 et al., 2010; Castro et al., 2011; Gladiolus communis L., Castro et al., 2018]. Alternatively, 75 polyploids might disperse elsewhere, thereby escaping minority cytotype exclusion, and 76 establish new populations beyond the environmental limits of the parental individuals (niche 77 shift hypothesis; Levin, 1975, 2004; Husband & Schemske, 2000). 78 Different geographical patterns have been documented in diploid-polyploid complexes 79 (namely sympatric, parapatric or allopatric cytotype distributions; e.g., Petit, Bretagnolle & 80 Felber, 1999; Baack, 2004; Kolář et al., 2009; Castro et al., 2012). An association between ploidy 81 level and differences in environmental and spatial boundaries is frequently observed (Levin, 82 2002), with polyploids often having different geographic ranges from their diploid progenitors 83 (e.g., Balao et al., 2009; Kolář et al., 2009; Sonnleitner et al., 2010; Laport et al., 2013; Casazza 84 et al., 2017). The geographical range occupied by each cytotype is the result of several 85 potentially interacting forces, including historical patterns of origin or migration, interactions 86 among cytotypes, and divergence in environmental tolerances (Husband et al., 2013). Historical 87 processes, including the timing of polyploid emergence, dispersal and colonization patterns, are 88 among the factors determining geographical patterns (Lowry & Lester, 2006). Recurrent 89 polyploid formation generates primary contact zones and mosaic distributions, while secondary 90 contact zones originate after allopatric divergence and subsequent migration, bringing 91 cytotypes into contact and creating allopatric and parapatric distributions (Petit, Bretagnolle & 92 Felber, 1999). Additionally, changes in cell size and gene expression associated with whole 93 genome duplication may have broad-scale impacts on developmental processes and 94 environmental tolerances that might change the fitness of polyploids (Levin, 1983; Adams & 95 Wendel, 2005). These differences have been linked with increased ecological tolerances, niche 96 partitioning and/or wider geographic ranges (e.g., Levin, 1975; Husband & Schemske, 2000; 97 Buggs & Pannell, 2007; Ramsey, 2011; Hao et al., 2013) and may determine the likelihood of 98 polyploid establishment and subsequent spread. Consequently, knowing the geographic Page 5 of 60 Botanical Journal of the Linnean Society Botanical Journal of the Linnean Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 PDF Proof 99 arrangement of diploid-polyploid complexes in nature provides essential information for 100 inferring the processes involved in polyploid establishment, coexistence and divergence (e.g., 101 Levin, 2002; Petit et al., 1999; Lexer & van Loo, 2006; Castro et al., 2018). 102 Recently, ecological niche modelling (Warren, Glor & Turelli, 2008) and multivariate 103 analyses of niche variables (Broennimann et al., 2012) using cytotype occurrence data and 104 various abiotic factors (e.g., precipitation, temperature, soil characteristics and elevation) have 105 been used to characterize ecological niches and to compare environmental niches between 106 different taxa. This approach has been used to evaluate the niche shift hypothesis in related 107 diploid-polyploid species (e.g., Houstonia spp., Glennon, Rissler & Church, 2012; Leucanthemum 108 Iberian taxa, Oberprieler et al., 2012; Claytonia perfoliata Donn ex Willd. complex, McIntyre, 109 2012; Primula sect. Aleuritia complex, Theodoridis et al., 2013; Tolmeia spp., Visger et al., 2016), 110 or in the analysis of different cytotypes within a species (Houstonia purpurea L. and H. longifolia 111 Gaertn., Glennon et al., 2012; Heuchera cylindrica Douglas, Godsoe et al., 2013; Chamerion 112 angustifolium, Thompson, Husband & Maherali, 2014; Erysimum mediohispanicum Polatschek, 113 Muñoz-Pajares et al., 2018). The characterization of the ecological niches enables researchers, 114 not only to identify the potential environmental constraints on the distribution of different 115 cytotypes, but also to unmask other factors potentially affecting cytogeographic patterns, such 116 as inter-cytotype biotic interactions and historical processes (e.g., Godsoe et al., 2013; Laport et 117 al., 2013; Glennon, Ritchie & Segraves, 2014; Laport, Minckley & Ramsey, 2016; Castro, 2018; 118 Mairal et al., in press). Thus, modelling tools have enabled researchers to formulate hypotheses 119 about the successful establishment of certain polyploid complexes, and subsequently, to test 120 these hypotheses experimentally in the field or in controlled conditions, through reciprocal 121 transplants or competition experiments (e.g., Ranunculus adoneus A. Gray, Baack & Stanton, 122 2005; Chamerion angustifolium, Martin & Husband 2013; Jasione maritima (Duby) Merino, 123 Castro, 2018). Page 6 of 60 Botanical Journal of the Linnean Society Botanical Journal of the Linnean Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 PDF Proof 124 Jasione L. (Campanulaceae) is a small genus distributed in Europe, North Africa and 125 Southwest Asia, with its centre of diversity in the Iberian Peninsula and most of its species 126 showing restricted ranges (Tutin, 1973; Sales & Hedge, 2001a, 2001b; Pérez-Espona et al., 2005). 127 Phylogenetic analyses using ITS suggest a recent origin of the species within the genus, possibly 128 linked with the Pleistocene glaciations (Sales et al., 2004; Pérez-Espona et al., 2005), although 129 the evolutionary history of the genus is still unknown. Polyploidy is often expected to be 130 frequent during glaciation periods (Thompson, 2005; Marques et al., 2017), because of the 131 association between the production of unreduced gametes and temperature fluctuations 132 (Ramsey & Schemske, 1998; Mason et al., 2011). Jasione comprises several diploid taxa (e.g., J. 133 foliosa Cav., J. corymbosa Poir.; Silvestre, 1986; Parnell, 1987), but it is also rich in polyploid 134 complexes, including tetraploid species (e.g., J. sessiliflora Boiss. & Reut.; Favarger, Galland & 135 Kupfer, 1980) and species with several ploidy levels (e.g., J. montana, J. laevis Lam., J. maritima 136 (Duby) Merino, J. crispa (Pourr.) Samp.; Sales & Hedge, 2001a; Rubido-Bará, Horjales Luaces & 137 Villaverde, 2010). Among the latter is the widespread J. montana L., which has been formerly 138 described as diploid throughout its distribution range across Europe (e.g., Kovanda, 1968; 139 Bjorkqvist, 1969; Kliphuis & Wieffering 1972; Ubera, 1980), until Leitão & Paiva (1988) reported, 140 for the first-time, tetraploid plants in Central Portugal, and, more recently, Rubido-Bará et al., 141 (2010) described the occurrence of tetraploids in Galicia (Spain). The species exhibits high 142 morphological variability, with diversity of habits, growth form and organ’s size (Parnell, 1985, 143 1987; Bokhari & Sales, 2001; Sales et al., 2004). Although some minor morphological differences 144 in plant size, in root thickness and in leaf size have been reported between diploids and 145 tetraploids from Galicia (Rubido-Bará et al., 2010), the morphological traits reveal a high 146 variability and overlap between the two entities, hindering the identification of the cytotypes in 147 the field. This morphological similarity suggests an autopolyploid origin of the tetraploid J. 148 montana, although allopolyploidization cannot be excluded. While the cytogenetic diversity has Page 7 of 60 Botanical Journal of the Linnean Society Botanical Journal of the Linnean Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 PDF Proof 149 been characterized within some populations in Galicia (Spain), the geographical distribution of 150 the tetraploids and the environmental niche preferences of each cytotype are still unknown. 151 The main objective of this study was to explore the diversity and distribution of 152 cytotypes within the Jasione montana polyploid complex and identify possible factors involved 153 in the successful establishment and spread of tetraploids. In particular, the goals of the present 154 study were to: (1) delineate the geographical distribution of tetraploids in the Iberian Peninsula; 155 (2) identify minority cytotypes, mixed-ploidy populations and contact zones between cytotypes; 156 and (3) determine whether the cytotypes have different ecological associations that could 157 explain their observed geographical distributions. To accomplish this, we sampled populations 158 throughout the Iberian Peninsula, in particular at detected contact zones, to determine DNA 159 ploidy levels using flow cytometry and assess the distribution patterns of each cytotype. Niche 160 modelling tools were then used to explore the ecological associations of each cytotype in the 161 Iberian Peninsula. We hypothesize that polyploidization results in shifts in environmental 162 preference, which allows tetraploids to establish by colonizing different environmental niches 163 than diploids, thus exhibiting low geographic overlap. The information about cytotype diversity, 164 geographical patterns and environmental associations enabled us to explore the factors 165 involved in the establishment and spread of J. montana tetraploid individuals in nature. 166 167 MATERIALS AND METHODS 168 STUDY SYSTEM 169 Jasione montana L. (Campanulaceae) is distributed throughout Europe, from the Mediterranean 170 to approximately 62°N, and from western Asia to North Africa (Tutin, 1973). It grows on rocky 171 ground, heaths and grasslands with thin soil, and prefers acid soils rather than limestone regions 172 (Horwood, 1919). Individuals vary morphologically and may be annual, biennial or perennial; 173 still, no clear association between ploidy level and plant habit has been found. Jasione montana 174 plants form a rosette of leaves during the winter and produce erect or ascending stems in the Page 8 of 60 Botanical Journal of the Linnean Society Botanical Journal of the Linnean Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 PDF Proof 175 spring, each ending in a capituliform inflorescence of bluish flowers (Parnell, 1980; Sales & 176 Hedge, 2001a). The species comprises diploids (2n = 2x = 12 chromosomes) through most of its 177 distribution (e.g., Kovanda, 1968; Ubera, 1980; Luque & Mejas, 1986; Pastor, 1990; Rubido-Bará 178 et al., 2010), while tetraploids (2n = 4x = 24 chromosomes) have been reported in Coimbra 179 (Centre of Portugal; Leitão & Paiva, 1988) and in Galicia (Spain; Rubido-Bará et al., 2010). The 180 genome size of the two cytotypes has been estimated to be 2C = 3.24 pg for diploids, which is 181 roughly half that of tetraploids, 2C = 6.58 pg (Rubido-Bará et al., 2010). 182 183 CHROMOSOME COUNTS FROM THE LITERATURE 184 We reviewed the literature on the karyology of J. montana (available in Chromosome Counts 185 database, http://ccdb.tau.ac.il/, Rice et al., 2015; Floras and published manuscripts), compiled 186 all chromosome counts with reference to the locality of sampling and mapped the distribution 187 of cytotypes. A total of 40 references including 89 identified localities with karyological 188 information were compiled (Table S1) and mapped. 189 190 FIELD SAMPLING 191 Our sampling was mostly focused on the Iberian Peninsula, particularly the northwest quadrant, 192 where tetraploids have been previously reported (Leitão & Paiva, 1988; Rubido-Bará et al., 193 2010). We also sampled other Mediterranean areas to confirm the diploid dominance reported 194 in the literature. In total, 287 populations were sampled, including 279 in the Iberian Peninsula 195 and eight elsewhere (France, Ireland, Italy and Morocco; Table S2). During spring and summer 196 of 2013-2016, fresh leaves were collected from each population, placed in hermetic plastic bags 197 and stored at 4 °C for flow cytometric analyses. In each population, up to 38 individuals (mean 198 = 12 individuals/population) were randomly sampled, throughout the entire population (Table 199 S2). When the harvesting of fresh material was impractical, mature seeds were collected 200 (procedure followed in 14 populations). Geographic coordinates of each population were Page 9 of 60 Botanical Journal of the Linnean Society Botanical Journal of the Linnean Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 PDF Proof 355 In the northwest quadrant of the Iberian Peninsula, the diploid and tetraploid 356 populations appeared intermingled, although most areas are dominated by diploids (Figure 2B). 357 The tetraploid populations are mostly clustered in two regions, one in Central Portugal and 358 another in Galicia (Spain), creating several areas of contact between diploids and tetraploids, 359 including a few sympatric areas where cytotypes coexist and form mixed-ploidy populations in 360 Central Portugal (Figure 2B). The composition of the mixed-ploidy populations was variable: two 361 populations were dominated by tetraploids, with only one diploid individual being detected in 362 each population, one population had fairly similar cytotype proportions, and one small 363 population was dominated by diploids, bearing only one tetraploid individual (Table S2). Despite 364 the large sample size (n = 3396), especially in the contact zone, no other cytotypes were 365 detected. 366 367 CYTOTYPE ENVIRONMENTAL CHARACTERISTICS AND NICHE OVERLAP 368 Visual inspection of the distribution models reveals a high predicted suitability for diploids over 369 most of the Iberian Peninsula, except for the eastern calcareous areas, where the probability of 370 occurrence is very low (Figure 3A). In contrast, the predicted tetraploid distribution was 371 restricted to the northwest quadrant of the Iberian Peninsula, where this cytotype is currently 372 found, and to the eastern coast of Valencia (Figure 3B). 373 The first two axes of the PCA explained a high percentage (74.8%) of ecological (climatic 374 and edaphic) variance (51. 3% and 23.5% in Axis 1 and 2, respectively). Maximum temperature 375 of the warmest month (Bio_5) and precipitation in the driest month (Bio_14) had the highest 376 contributions to the first axis, followed by temperature and precipitation seasonality (Bio_4 and 377 Bio_15, respectively). Two soil variables (base saturation of the topsoil, bs_top, and dominant 378 surface textural class, txsrfdo) had lower contributions and mostly for the second axis (Figure 379 4A). Based on the first and second principal components, the environmental niche of the 380 tetraploid falls completely within the environmental range of the diploid (Figure 4B-C). However, Page 16 of 60 Botanical Journal of the Linnean Society Botanical Journal of the Linnean Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 PDF Proof 381 the reverse is not true; rather, the amplitude of the environmental niche of diploids was much 382 larger than that of the tetraploids (Figure 4B-C), and only 20.7% of the diploid environmental 383 niche overlaps with the tetraploid niche (green area in Figure 4B). 384 Even though the tetraploid niche is embedded in the diploid niche envelope (Figure 4B), 385 given the differences in occurrences density in the ecological space between cytotypes, the D 386 metric of niche overlap was low (D = 0.08). The niche equivalency test rejected the null 387 hypothesis (P < 0.05; Figure 4D) and indicates that the climatic niches of diploids and tetraploids 388 are not equivalent or identical (Glor & Warren, 2011). However, the niche similarity test revealed 389 that the observed D value falls within the 95th percentile of the simulated D values, thus failing 390 to reject the null hypothesis of niche similarity. Consequently, cytotype environmental niches 391 were not more similar (or different) from one another than the expected after random sampling 392 (P > 0.05, for both diploids growing in the tetraploid potential area, and tetraploids growing in 393 the diploid potential area; Figure 4E-F). 394 395 DISCUSSION 396 We found differences in both geographic distribution and environmental niche breadth between 397 diploid and tetraploid individuals of Jasione montana. Tetraploids are restricted to a relatively 398 small geographic area within the Iberian Peninsula, while diploids are widespread across Europe, 399 a pattern consistent with previous reports (compiled in Table S1 and Figure 2A; Leitão and Paiva, 400 1988; Rubido-Bará et al., 2010). Within the region of overlap, tetraploids and diploids mostly 401 occur as a mosaic of single-ploidy populations. Consistent with their smaller geographic range, 402 the environmental niche of tetraploids is more restricted than, and completely nested within, 403 that of diploids. Contrary to our hypothesis, these observations suggest that whole genome 404 duplication in this species is not associated with the origin of novel environmental requirements; 405 if anything, the tetraploid niche is a narrow subset of the extant diploid niche, a pattern seen in 406 other taxa (Glennon et al., 2014). Below, we discuss the mechanisms underlying these results Page 17 of 60 Botanical Journal of the Linnean Society Botanical Journal of the Linnean Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 PDF Proof 407 and their implications for our understanding of polyploid establishment and persistence in 408 plants. 409 410 PLOIDY LEVELS IN JASIONE MONTANA 411 Our study confirms the presence of two cytotypes in J. montana, diploids and tetraploids. 412 Polyploidization is common within the genus Jasione. Approximately half of the Jasione species 413 occurring on the Iberian Peninsula, the centre of diversity for the genus, are polyploid or harbour 414 multiple cytotypes (Sales & Hedge, 2001b; Rubido-Bará et al., 2010; Castro, 2018); J. montana is 415 one such species. The co-occurrence of diploids and tetraploids is commonly observed in mixed- 416 ploidy species, generally (Kolář et al., 2017), and is common among polyploid species in the 417 genus Jasione. For example, J. laevis and J. maritima (Sales & Hedge, 2001a; Rubido-Bará et al., 418 2010; Castro, 2018) both consist of diploids and tetraploids, whereas J. crispa has diploids, 419 tetraploids and hexaploids (Sales & Hedge, 2001a). More surprising is the absence of triploid 420 hybrids, especially in the contact zone of the Iberian Peninsula. This pattern may reflect the 421 spatial separation of cytotypes into single-ploidy populations or the presence of strong triploid 422 block (discussed below). The absence of triploids within the four mixed-ploidy populations 423 suggests that both factors may be operating in the diploid-tetraploid contact zones of Jasione 424 montana. 425 426 NESTED AND ASYMMETRICAL GEOGRAPHIC RANGES: HISTORICAL CAUSES 427 The nested and asymmetrical geographic ranges of diploids and tetraploids, and the restricted 428 geographic range of tetraploids in particular, may reflect the history of formation of tetraploids 429 from diploid ancestors in this species. Tetraploids may have arisen and successfully colonized 430 the northwest quadrant of the Iberian Peninsula relatively recently and, thus, have had limited 431 time to expand beyond the site of origin, as has been suggested for other mixed-ploidy species 432 (Godsoe et al., 2013; Laport et al., 2016). The limited geographic range also suggests that Page 18 of 60 Botanical Journal of the Linnean Society Botanical Journal of the Linnean Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 PDF Proof 433 tetraploids have arisen a limited number of times across the entire geographic range. While our 434 sampling was extensive, more intensive screening for rare tetraploids throughout the diploid 435 portion of the range is needed to determine whether tetraploids do in fact arise frequently but 436 are unlikely to establish. Multiple origins of polyploids have been documented in numerous taxa 437 and on a range of spatial scales (e.g., Soltis & Soltis, 1999; Segraves & Thompson, 1999; Sampson 438 & Byrne, 2011). Therefore, we cannot exclude the possibility that the two clusters of tetraploid 439 populations, one in Portugal and the other in Galicia (Spain), represent independent origins. To 440 date, we have little evidence with which to evaluate these historical scenarios in J. montana. 441 Phylogenetic analyses of the genus Jasione positioned J. montana in a clade with other species 442 currently occurring in the northwest quadrant of the Iberian Peninsula (Sales et al., 2004; Pérez- 443 Espona et al., 2005). However, the internal relationships of the clade remain unresolved and the 444 time of divergence of tetraploid J. montana relative to diploids is not known. Additional research 445 on genetic divergence and phylogeography of tetraploids and diploids may help to shed light on 446 the timing and frequency of whole genome duplication in this species. 447 448 ASYMMETRICAL RANGES: ENVIRONMENTAL NICHE SHIFTS? 449 Our results provide insight into the role of environmental differences between cytotypes in 450 determining establishment and geographic distribution of polyploids. Tetraploid plants of J. 451 montana occupy only a subset of the diploid environmental niche, as depicted in the PCA (Fig 452 4B, C), and the equivalency test indicates that climatic niches of diploids and tetraploids are not 453 equivalent; still, cytotype environmental niches were not more similar (or different) from one 454 another than expected by chance. Therefore, we conclude that there is no evidence that whole 455 genome duplication, leading to tetraploidy, resulted in a shift in environmental niche beyond 456 that of the diploid niche. Changes in environmental tolerance are often postulated as necessary 457 for polyploid establishment (Husband & Schemske, 2000; Baack & Stanton, 2005; Buggs & 458 Pannell, 2007; Ramsey, 2011) as they promote spatial segregation and within-cytotype mating, Page 19 of 60 Botanical Journal of the Linnean Society Botanical Journal of the Linnean Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 PDF Proof 459 allowing rare cytotypes to avoid frequency-dependent selection and exclusion (Levin, 1975; 460 Fowler & Levin, 1984; Felber, 1991; Hao et al., 2013). However, evidence in the literature for 461 niche shifts that occur in association with whole genome duplication is inconsistent. Examples 462 of environmental niche divergence between cytotypes have been reported in species such as 463 Chamerion angustifolium (Thompson et al., 2014), Centaurea stoebe (Glennon et al., 2014), 464 Tolmiea diplomenziesii / menziesii (Visger et al., 2016), and Erysimum mediohispanicum (Muñoz- 465 Pajares et al., 2018), although it is usually unclear whether this results directly from the 466 duplication or subsequent evolutionary divergence (see Maherali et al., 2009 for C. 467 angustifolium). In contrast, a test of the niche shift hypothesis using 20 different diploid- 468 polyploid species found that most pairs showed significant similarity or no difference in niche 469 between pairs (Glennon et al., 2014). Collectively, these results suggest that polyploids 470 frequently occupy similar environmental conditions to their diploid progenitors and that niche 471 differentiation, measured as climatic variables, may not be a necessary requirement for 472 polyploid establishment and persistence in nature (Glennon et al., 2014). 473 474 ASYMMETRICAL GEOGRAPHIC RANGES: ENVIRONMENTAL SORTING 475 While evidence for a significant environmental niche shift in tetraploids is lacking, environmental 476 sorting may still explain, in part, the differences in geographical range between cytotypes of J. 477 montana. The smaller geographical range of tetraploids is congruent with their narrower 478 environmental niche compared to diploids. This correlation suggests a possible causal 479 relationship; specifically, that the narrower niche of tetraploids is limiting the expansion and 480 establishment on a larger geographical scale. Additionally, polyploids with smaller niche ranges 481 than their diploid relatives may be well adapted, or even specialized, to a given set of 482 environmental conditions (Parisod & Broennimann, 2016). Thus, the restricted and spatially 483 clustered distribution of the tetraploids could reveal areas harbouring a combination of 484 environmental conditions where the tetraploid niche is at its optimum, allowing them to Page 20 of 60 Botanical Journal of the Linnean Society Botanical Journal of the Linnean Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 PDF Proof 485 outcompete the diploids. In the case of J. montana, the tetraploid distribution coincides with 486 the boundaries between four main biogeographic sectors (the Galicia and North Portugal sector, 487 the North Lusitania Sierra sector, the Montemuro and Estrela Sierras sector and the north part 488 of the Divisorio Portuguese sector; Rivas-Martínez et al., 2017) and is characterized by a 489 subatlantic-submediterranean climate and with similar floristic compositions (Rivas-Martinez et 490 al., 2017). This biogeographic region of contact between different sectors constitutes the 491 interface between the temperate and Mediterranean biomes and harbours several species 492 adapted to this transition (Amigo et al., 2017). Niche relationships between closely related 493 species or cytotypes in which the niche of one taxon/cytotype is much narrower than, and 494 nested within, the niche of the other, have been interpreted as indicative of environmental 495 specialization (Knouft et al., 2006; Vamosi et al., 2014; Parisod & Broennimann, 2016). To 496 confirm this hypothesis, reciprocal transplant experiments are necessary to establish that 497 tetraploids are, indeed, unable to thrive in diploid environments outside of their current 498 geographic range. 499 500 WHAT ALLOWS TETRAPLOIDS TO PERSIST WITH DIPLOIDS ON THE IBERIAN PENINSULA? 501 If tetraploid J. montana is restricted in distribution to a small portion of the diploid range, what 502 then allowed tetraploids to persist and coexist with diploids? Theoretical models predict that, in 503 randomly mating populations, rare cytotypes will experience a frequency-dependent mating 504 disadvantage, due to increased inter-cytotype mating, which is generally ineffectual. As a result, 505 dominant cytotypes will preclude establishment of newly arising cytotypes (Levin, 1975; 506 Rodriguez, 1996; Husband & Schemske, 2000) unless new cytotypes can disperse to unoccupied 507 locations (e.g., Godsoe et al., 2013; Thompson et al., 2014; Visger et al., 2016; Muñoz-Pajares et 508 al., 2018), resulting in a mosaic of single-ploidy populations on the landscape, each excluding 509 the other from their respective populations through minority cytotype exclusion (Petit et al., 510 1999; e.g., Baack, 2004; Kolář et al., 2009; Castro et al., 2012). In J. montana, the diploids and Page 21 of 60 Botanical Journal of the Linnean Society Botanical Journal of the Linnean Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 PDF Proof 511 tetraploids indeed form a complex mosaic of single-ploidy populations in the northwest 512 quadrant of the Iberian Peninsula, with areas dominated by diploids intermingled with areas 513 dominated by tetraploids. Furthermore, the rarity of mixed populations suggests that 514 reproductive barriers between the two cytotypes might be insufficient to prevent inter-cytotype 515 mating, resulting in minority cytotype disadvantage (triploid block; Levin, 1975; Husband, 2000). 516 Indeed, preliminary pollination experiments revealed that both cytotypes are unable to produce 517 viable offspring after inter-cytotype crosses (M. Castro, unpublished data). Minority cytotype 518 exclusion has been tested experimentally only by Husband (2000) in Chamerion angustifolium 519 and by Baack (2005) in Ranunculus adoneus, but it has been referred as an important mechanism 520 driving cytotype distribution patterns in numerous diploid-polyploid contact zones (e.g., Levin, 521 2002; Španiel et al., 2008; Castro et al., 2011). 522 Our results suggest that, to establish successfully, tetraploids of J. montana must 523 disperse and colonize unoccupied areas in the landscape and, thus, avoid the minority cytotype 524 exclusion (e.g., Godsoe et al., 2013; Thompson et al., 2014; Visger et al., 2016; Muñoz-Pajares 525 et al., 2018). Interestingly, there is a tendency for tetraploid J. montana to occupy ruderal and 526 more disturbed habitats than diploids, which can occupy both less disturbed areas as well as 527 some ruderal habitats (M. Castro & M. Serrano, Field observations). This pattern has been 528 observed in other polyploid complexes, in which tetraploids spread into ruderal habitats forming 529 mosaic contact zones even without differences in climatic niche (e.g., Španiel et al., 2008; Kolář 530 et al., 2016). Our observations remain to be confirmed but highlight the hypothesis that cytotype 531 coexistence in the contact zone can be enabled by microhabitat differences between cytotypes 532 within the existing climatic niche. Also, the aggregation of tetraploid populations in regions 533 suitable for diploids may indicate an ability to outcompete diploids in specific environments, 534 thereby excluding them from population. Fitness advantages such as increased competitive 535 ability (Maceira, Jacquard & Lumaret, 1993; Laport et al., 2013) and asymmetric assortative 536 mating (Husband & Sabara, 2004; Buggs & Pannell, 2006, 2007; Laport et al., 2016), among other Page 22 of 60 Botanical Journal of the Linnean Society Botanical Journal of the Linnean Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 PDF Proof 537 factors, play crucial roles in the success of polyploid lineages. In J. montana, some morphological 538 differences between cytotypes have been observed (Rubido-Bará et al., 2010) that might be 539 associated with plant fitness and competitive ability, such as increased plant size and tendency 540 for perenniality in the tetraploids (M. Serrano, field observations). Nevertheless, further studies 541 are needed to unravel possible fitness differences that could explain cytotype interactions at 542 contact zones. 543 544 CONCLUDING REMARKS 545 Our results indicate that polyploidization in J. montana has not caused expansion to novel 546 environmental conditions. Under this scenario, tetraploids will be subjected to strong 547 frequency-dependent selection and will be excluded by diploids unless they can disperse to sites 548 unoccupied by diploids, where they in turn may exclude diploid colonists. The mosaic of diploid 549 and tetraploid populations and a lack of mixed-ploidy populations is consistent with a frequency- 550 dependent selection influencing distributions within the mixed-ploidy zones, although the 551 influence of inter-cytotype competition cannot be ruled out. The interplay between both 552 mechanisms is unclear and future molecular and experimental studies will be important for fully 553 understanding the role of environmental sorting and evolutionary history in governing the 554 distribution of cytotypes within the contact zones of J. montana. 555 556 ACKNOWLEDGEMENTS 557 The authors are thankful to several researchers that assisted in the field collections, in particular 558 to Ana Afonso, Ana Martins, Joana Costa, José Cerca de Oliveira, José Miguel Costa, João Costa 559 and Lucie Mota. This research was supported by POPH/FSE funds by the Portuguese Foundation 560 for Science and Technology (FCT) with a doctoral grant to MC (SFRH/BD/89910/2012) and a 561 starting grant and exploratory project to SC (IF/01267/2013), and by Project RENATURE financed Page 23 of 60 Botanical Journal of the Linnean Society Botanical Journal of the Linnean Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 PDF Proof 562 by the “Programa Operacional Regional do Centro 2014-2020 (Centro2020) - CENTRO-01-0145- 563 FEDER-000007”. 564 565 SUPPORTING INFORMATION 566 The following additional information is available: 567 Table S1 – Chromosome counts of Jasione montana available in the bibliography. 568 Table S2 – Geographic information of Jasione montana populations sampled in this study. 569 Table S3 – Selected environmental variables for Jasione montana cytotype niche modelling. 570 Table S4 – Genome size estimates in Jasione montana. Page 24 of 60 Botanical Journal of the Linnean Society Botanical Journal of the Linnean Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 PDF Proof 571 REFERENCES 572 Adams SKL, Wendel JF. 2005. Novel patterns of gene expression in polyploidy. Trends in 573 Genetics 21(10): 539-543. 574 Araújo MB, New M. 2007. Ensemble forecasting of species distributions. Trends in Ecology & 575 Evolution 22(1): 42-47. 576 Baack, EJ. 2004. Cytotype segregation on regional and microgeographic scales in snow 577 buttercups (Ranunculus adoneus: Ranunculaceae). American Journal of Botany 91(11): 578 1783-1788. 579 Baack, EJ. 2005. Ecological factors influencing tetraploid establishment in snow buttercups 580 (Ranunculus adoneus, Ranunculaceae): minority cytotype exclusion and barriers to 581 triploid formation. American Journal of Botany 92(11): 1827-1835. 582 Baack EJ, Stanton ML. 2005. Ecological factors influencing tetraploid speciation in snow 583 buttercups (Ranunculus adoneus): niche differentiation and tetraploid 584 establishment. Evolution 59(9): 1936-1944. 585 Balao F, Casimiro-Soriguer R, Talavera M, Herrera J, Talavera S. 2009. Distribution and diversity 586 of cytotypes in Dianthus broteri as evidenced by genome size variations. Annals of 587 Botany 104(5): 965-973. 588 Barringer BC. 2007. Polyploidy and self-fertilization in flowering plants. American Journal of 589 Botany 94(9): 1527-1533. 590 Bates D, Maechler M, Bolker B, Walker S. 2014. lme4: linear mixed-effects models using Eigen 591 and S4. Available at http://CRAN.R-project.org/package=lme4 592 Bjorkqvist IR. 1969. Chromosome numbers in Iberian angiosperms. Nord. J. Bot. 122: 271-283. 593 Bokhari MH, Sales F. 2001. Jasione (Campanulaceae) anatomy in the Iberian Peninsula and its 594 taxonomic significance. Edinburgh Journal of Botany 58(3): 405-422. Page 25 of 60 Botanical Journal of the Linnean Society Botanical Journal of the Linnean Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 PDF Proof 745 Mairal M, Šurinová M, Castro S, Münzbergová Z. 2018. Multiple origin of autopolyploid 746 cytotypes followed by different demographic trajectories and niche differentiation in the 747 Aster amellus aggregate. Annals of Botany. 748 Marques I, Loureiro J, Draper D, Castro M, Castro S. 2017. How much do we know about the 749 frequency of hybridization and polyploidy in the Mediterranean region? Plant Biology 21- 750 37. 751 Martin SL, Husband BC. 2013. Adaptation of diploid and tetraploid Chamerion angustifolium to 752 elevation but not local environment. Evolution 67(6): 1780-1791. 753 Mason AS, Nelson MN, Yan G, Cowling WA. 2011. Production of viable male unreduced gametes 754 in Brassica interspecific hybrids is genotype specific and stimulated by cold 755 temperatures. BMC Plant Biology 11(1): 103. 756 McIntyre PJ. 2012. Polyploidy associated with altered and broader ecological niches in the 757 Claytonia perfoliata (Portulacaceae) species complex. American Journal of Botany 99(4): 758 655-662. 759 Muñoz-Pajares AJ, Perfectti F, Loureiro J, Abdelaziz M, Biella P, Castro M, Castro S, Gómez JM. 760 2018. Niche differences may explain the geographic distribution of cytotypes in Erysimum 761 mediohispanicum. Plant Biology 139-147. 762 Oberprieler C, Konowalik K, Altpeter S, Siegert E, Presti RML, Greiner R, Vogt R. 2012. Filling of 763 eco-climatological niches in a polyploid complex–A case study in the plant genus 764 Leucanthemum Mill. (Compositae, Anthemideae) from the Iberian Peninsula. Flora- 765 Morphology, Distribution, Functional Ecology of Plants 207(12): 862-867. 766 Otto SP, Whitton J. 2000. Polyploid incidence and evolution. Annual Review of Genetics 34(1): 767 401-437. 768 Panagos P, Van Liedekerke M, Jones A, Montanarella L. 2012. European Soil Data Centre: 769 Response to European policy support and public data requirements. Land use 770 policy 29(2): 329-338. Page 32 of 60 Botanical Journal of the Linnean Society Botanical Journal of the Linnean Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 PDF Proof 771 Parisod C, Broennimann O. 2016. Towards unified hypotheses of the impact of polyploidy on 772 ecological niches. New Phyologist 212: 540–542. 773 Parnell J. 1985. Jasione montana L. Journal of Ecology 73(1): 341-358. 774 Parnell J. 1987. Variation in Jasione montana L. (Campanulaceae) and related species in Europe 775 and North Africa. Watsonia 16(3): 249-267. 776 Pastor J. 1990. Números cromosómicos para la flora Española. 556-591. Lagascalia 15: 269-282. 777 Pérez-Espona S, Sales F, Hedge I, Möller M. 2005. Phylogeny and species relationships in Jasione 778 (Campanulaceae) with emphasis on the ‘Montana-complex’. Edinburgh Journal of 779 Botany 62(1-2): 29-51. 780 Petit C, Bretagnolle F, Felber F. 1999. Evolutionary consequences of diploid–polyploid hybrid 781 zones in wild species. Trends in Ecology & Evolution 14(8): 306-311. 782 Phillips SJ. 2008. Transferability, sample selection bias and background data in presence-only 783 modelling: a response to Peterson et al. (2007). Ecography 31(2): 272-278. 784 Phillips SJ, Anderson RP, Schapire RE. 2006. Maximum entropy modeling of species geographic 785 distributions. Ecological Modelling 190(3-4): 231-259. 786 R Core Development Team. 2016. A language and environment for statistical computing. 787 Vienna: R Foundation for Statistical Computing. Available at http://www.R-project.org/ 788 Ramsey J. 2007. Unreduced gametes and neopolyploids in natural populations of Achillea 789 borealis (Asteraceae). Heredity 98(3): 143. 790 Ramsey J. 2011. Polyploidy and ecological adaptation in wild yarrow. Proceedings of the 791 National Academy of Sciences 108(17): 7096-7101. 792 Ramsey J, Ramsey TS. 2014. Ecological studies of polyploidy in the 100 years following its 793 discovery. Phil. Trans. R. Soc. B 369(1648): 1-20. 794 Ramsey J, Schemske DW. 1998. Pathways, mechanisms, and rates of polyploid formation in 795 flowering plants. Annual Review of Ecology and Systematics 29(1): 467-501. Page 33 of 60 Botanical Journal of the Linnean Society Botanical Journal of the Linnean Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 PDF Proof 796 Rice A, Glick L, Abadi S, Einhorn M, Kopelman NM, Salman-Minkov A, Mayzel J, Chay O. 797 Mayrose I. 2015. The Chromosome Counts Database (CCDB) – a community resource of 798 plant chromosome numbers. New Phytologist 206(1): 19-26. 799 Rivas-Martínez S, Penas A, Díaz TE, Cantó P, del Río S, Costa JC, Herrero L, Molero J. 2017 800 Biogeographic units of the Iberian Peninsula and Islas Baleares to district level. A concise 801 synopsis. In: Loidi J, editor. The Vegetation of the Iberian Peninsula 1. Cham, Switzerland: 802 Springer Ed; p. 131–188. 803 Rodriguez DJ. 1996. A model for the establishment of polyploidy in plants. The American 804 Naturalist 147(1): 33-46. 805 Rubido-Bará M, Horjales Luaces M, Villaverde C. 2010. Dos nuevas subespecies del género 806 Jasione L. (Campanulaceae) en el Noroeste de la Península Ibérica. Nova Acta Científica 807 Compostela (Bioloxía) 19: 21-31. 808 Sales F, Hedge IC. 2001a. Myoporaceae-Campanulaceae. In: Castroviejo, S., Paiva, J., Sales, F., 809 Hedge, I.C., Aedo, C., Aldasoro, J.J., Herrero, A. & Velayos, M. eds., Flora Iberica. Plantas 810 Vasculares de la Península Ibérica e Islas Baleares. Vol. XIV. Real Jardin Botanico, C.S.I.C., 811 Madrid. 812 Sales F, Hedge IC. 2001b. Nomenclature and typification of western european Jasione 813 (Campanulaceae). Anales del Jardín Botánico de Madrid 59(1): 163-172. 814 Sales F, Hedge IC, Eddie W, Preston J, Moeller M. 2004. Jasione L. taxonomy and 815 phylogeny. Turkish Journal of Botany 28(1-2): 253-259. 816 Sampson JF, Byrne M. 2011. Genetic diversity and multiple origins of polyploid Atriplex 817 nummularia Lindl. (Chenopodiaceae). Biological Journal of the Linnean Society 105(1): 818 218-230. 819 Schoener TW. 1970. Nonsynchronous spatial overlap of lizards in patchy habitats. Ecology 51(3): 820 408-418. Page 34 of 60 Botanical Journal of the Linnean Society Botanical Journal of the Linnean Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 PDF Proof 821 Segraves KA, Thompson JN. 1999. Plant polyploidy and pollination: floral traits and insect visits 822 to diploid and tetraploid Heuchera grossulariifolia. Evolution 53(4): 1114-1127. 823 Silvestre S. 1986. Números cromosómicos para la flora española: 435-496. Números 435(455): 824 453. 825 Smith SA, Donoghue MJ. 2010. Combining historical biogeography with niche modeling in the 826 Caprifolium clade of Lonicera (Caprifoliaceae, Dipsacales). Systematic Biology 59(3): 322- 827 341. 828 Soltis DE, Soltis PS. 1999. Polyploidy: recurrent formation and genome evolution. Trends in 829 Ecology & Evolution 14(9): 348-352. 830 Soltis DE, Buggs RJ, Doyle JJ, Soltis PS. 2010. What we still don't know about 831 polyploidy. Taxon 59(5): 1387-1403. 832 Sonnleitner M, Flatscher R, Escobar García P, Rauchová J, Suda J, Schneeweiss GM, Hülber K, 833 Schönswetter P. 2010. Distribution and habitat segregation on different spatial scales 834 among diploid, tetraploid and hexaploid cytotypes of Senecio carniolicus (Asteraceae) in 835 the Eastern Alps. Annals of Botany 106(6): 967-977. 836 Španiel S, Marhold K, Hodálová I, Lihová J. 2008. Diploid and tetraploid cytotypes of Centaurea 837 stoebe (Asteraceae) in Central Europe: morphological differentiation and cytotype 838 distribution patterns. Folia Geobotanica 43(2): 131. 839 Suda J, Kron P, Husband BC, Trávníček P. 2007. Flow cytometry and ploidy: applications in plant 840 systematics, ecology and evolutionary biology. In Doležel J., Greilhuber J., Suda J., eds. 841 Flow cytometry with plant cells: analysis of genes, chromosomes and genomes. Weinheim: 842 Wiley-VCH, 103-130. 843 te Beest M, Le Roux JJ, Richardson DM, Brysting AK, Suda J, Kubešová M, Pyšek P. 2011. The 844 more the better? The role of polyploidy in facilitating plant invasions. Annals of 845 Botany 109(1): 19-45. Page 35 of 60 Botanical Journal of the Linnean Society Botanical Journal of the Linnean Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 PDF Proof 846 Theodoridis S, Randin C, Broennimann O, Patsiou T, Conti E. 2013. Divergent and narrower 847 climatic niches characterize polyploid species of European primroses in Primula sect. 848 Aleuritia. Journal of Biogeography 40(7): 1278-1289. 849 Thompson JD. 2005. Plant evolution in the Mediterranean. Oxford University Press on Demand. 850 Thompson KA, Husband BC, Maherali H. 2014. Climatic niche differences between diploid and 851 tetraploid cytotypes of Chamerion angustifolium (Onagraceae). American Journal of 852 Botany 101(11): 1868-1875. 853 Thuiller W, Georges D, Engler R, Breiner F, Georges MD, Thuiller CW. 2016. Package ‘biomod2’. 854 Available at https://cran.r-project.org/package=biomod2 855 Tutin TG. 1973. Flora Europaea: notulae systematicae ad floram Europaeam spectantes: no. 14. 856 Notes on Jasione L. Europe. Bot. L. Linn. Soc 67(3): 276-279. 857 Ubera JL. 1980. In Numeros cromosomicos para la flora Esparlola. 121-182. Lagascalia 9: 49 – 858 284. 859 Vamosi JC, Armbruster WS, Renner SS. 2014. Evolutionary ecology of specialization: insights 860 from phylogenetic analysis. Proceeding of the Royal Society of London B: Biological 861 Sciences 281: 2014-2004. 862 Visger CJ, Germain-Aubrey CC, Patel M, Sessa EB, Soltis PS, Soltis DE. 2016. Niche divergence 863 between diploid and autotetraploid Tolmiea. American Journal of Botany 103(8): 1396- 864 1406. 865 Warren DL, Glor RE, Turelli M. 2008. Environmental niche equivalency versus conservatism: 866 quantitative approaches to niche evolution. Evolution 62(11): 2868-2883. 867 Wood TE, Takebayashi N, Barker MS, Mayrose I, Greenspoon PB, Rieseberg LH. 2009. The 868 frequency of polyploid speciation in vascular plants. Proceedings of the national Academy 869 of sciences 106(33): 13875-13879. Page 36 of 60 Botanical Journal of the Linnean Society Botanical Journal of the Linnean Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 PDF Proof 870 FIGURE CAPTIONS 871 Figure 1 – Flow cytometric histogram of relative propidium iodide fluorescence intensity (PI 872 fluorescence) of nuclei isolated from fresh leaves of Solanum lycopersicum ‘Stupické’ (S.l.; 873 reference standard with 2C = 1.96 pg) and of Jasione montana diploid (2x) and tetraploid (4x) 874 plants. For each peak, the mean relative fluorescence (Mean FL), DNA index (DI, Mean FL of J. 875 montana peak / Mean FL of the reference standard) and coefficient of variation of each peak 876 (CV, in %) are provided. 877 878 Figure 2 – Jasione montana records. A) bibliographic information from chromosome counts and 879 ploidy levels in Europe (diamonds) and additional estimates provided in this study (circles); B) 880 detail of the Iberian Peninsula, including estimates from this study (big circles) and GBIF 881 occurrences (small circles). Diamonds – bibliographic records (BR); big circles – populations 882 screened in this study; small circles – GBIF occurrences; Ploidy levels: white – diploids (2x), grey 883 – tetraploids (4x), black – diploid-tetraploid populations. 884 885 Figure 3 – Predictive suitable niche for each cytotype: A) diploids and B) tetraploids of Jasione 886 montana in the Iberian Peninsula. With colors represent habitats with low suitability and black 887 colors habitats with high suitability. 888 889 Figure 4 – Ecological niche models for Jasione montana cytotypes at the Iberian Peninsula: A) 890 contribution of climatic and soil variables to the first two axes of the principal component 891 analyses (PCA) and the percentage of variance explained by each axis; B) environmental niche 892 of diploids and tetraploids, respectively, based on the PCA of selected variables; coloured areas 893 represent the following: dark grey – suitable habitats for diploids, green – denotes both the 894 suitable habitats for tetraploids and the overlapping areas between diploid and tetraploid 895 environmental niches; the continuous line corresponds to the whole climatic space, while the Page 37 of 60 Botanical Journal of the Linnean Society Botanical Journal of the Linnean Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 PDF Proof 896 dashed line indicates the 75th percentile; C) magnitude of environmental niches of each cytotype 897 (white and dark grey represent the occupation of niches in PCA1 and PCA2, respectively); D) 898 equivalency test results; E) similarity test results of diploids growing in the tetraploid 899 environmental niche; and F) similarity tests of tetraploids growing in the diploid environmental 900 niche. In D, E and F, dark grey diamonds represent the observed D values, while grey bars 901 correspond to the frequency of calculated D value resulting from a similarity test process. Page 38 of 60 Botanical Journal of the Linnean Society Botanical Journal of the Linnean Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 PDF Proof 902 TABLES 903 Table 1 – Genome size estimates in Jasione montana according with each cytotype. Ploidy level 904 and mean, standard deviation of the mean (SD), coefficient of variation (CV, in %), minimum and 905 maximum values of holoploid genome size (2C, in pg) are given. Mean and standard deviation 906 of the mean (SD) of estimated monoploid genome size (1Cx, in pg) and the total number of 907 populations and individuals analysed are also presented for each cytotype. Two ploidy levels 908 were observed: diploids (2x) and tetraploids (4x). Different letters correspond to statistically 909 significant differences at P < 0.05, n.s. denote non-significant differences at P > 0.05. Holoploid genome size (2C, pg) Monoploid genome size (1Cx, pg) Ploidy level Mean SD CV (%) Min Max Mean SD Populations (individuals) 2x 2.92a 0.07 2.9% 2.80 3.08 1.46n.s. 0.04 84 (205) 4x 5.86b 0.14 3.3% 5.63 6.06 1.46n.s. 0.04 24 (46) 910 Page 39 of 60 Botanical Journal of the Linnean Society Botanical Journal of the Linnean Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 PDF Proof Flow cytometric histogram of relative propidium iodide fluorescence intensity (PI fluorescence) of nuclei isolated from fresh leaves of Solanum lycopersicum ‘Stupické’ (S.l.; reference standard with 2C = 1.96 pg) and of Jasione montana diploid (2x) and tetraploid (4x) plants. For each peak, the mean relative fluorescence (Mean FL), DNA index (DI, Mean FL of J. montana peak / Mean FL of the reference standard) and coefficient of variation of each peak (CV, in %) are provided. 173x167mm (150 x 150 DPI) Page 40 of 60 Botanical Journal of the Linnean Society Botanical Journal of the Linnean Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 PDF Proof Jasione montana records. A) bibliographic information from chromosome counts and ploidy levels in Europe (diamonds) and additional estimates provided in this study (circles); B) detail of the Iberian Peninsula, including estimates from this study (big circles) and GBIF occurrences (small circles). Diamonds – bibliographic records (BR); big circles – populations screened in this study; small circles – GBIF occurrences; Ploidy levels: white – diploids (2x), grey – tetraploids (4x), black – diploid-tetraploid populations. 148x199mm (150 x 150 DPI) Page 41 of 60 Botanical Journal of the Linnean Society Botanical Journal of the Linnean Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 PDF Proof 89 Table S2 – Geographic information of the Jasione montana populations sampled in this study. 90 For each population, an ID code, estimated ploidy level, sample size (n) and information on the 91 location and GPS coordinates (angular) are presented. Two ploidy levels were observed: diploids 92 (2x) and tetraploids (4x). Populations were divided in two groups according with their ploidy 93 composition: single-ploidy populations and mixed-ploidy populations. The five locations marked 94 with an asterisk (*) constitute unpublished chromosome counts by Miguel Serrano. The 95 superscript notation (S) denote populations where ploidy level was assessed directly from seeds. Geographic coordinates ID code Ploidy level (n) Location Herbarium voucher Latitude Longitude Single-ploidy populations MS03-01 S 2x (15) Jbel Tazekka, Taza, Morocco SANT62794 34.08691 -4.18341 MS16-015 2x* Pazzano, Calabria, Italy SANT61450 38.46803 16.41768 AA002 2x (6) Pegões, Setúbal, Portugal 38.65242 -8.61802 MC357 2x (5) Arraiolos, Évora, Portugal 38.72755 -7.98722 MS05-01 S 2x (15) Aldea del Rey, Cuidad Real, Spain SANT60479 38.73722 -3.87583 MC356 2x (7) Coruche, Santarém, Portugal 38.94409 -8.49414 MC353 2x (17) Muge, Salvaterra de Magos, Portugal 39.06865 -8.66193 MC358 2x (3) Domingão, Ponte de Sor, Portugal 39.24543 -8.02592 SC122 2x (4) Jardim, Marvão, Portugal SANT68767 39.39334 -7.38971 SC121 2x (5) Quinta Mão do Novilheiro, Castelo de Vide, Portugal SANT68768 39.42844 -7.48734 DT017 2x (3) Chãos, Alcobertas, Portugal 39.42958 -8.92472 MC341 2x (19) Rio de Moinhos, Abrantes, Portugal 39.47490 -8.23640 MC342 2x (6) Constância, Santarém, Portugal 39.48597 -8.33026 MC340 2x (25) Penedo, Oleiros, Portugal 39.66357 -8.12899 MC221 2x (30) Peral, Proença-a-Nova, Portugal 39.69370 -7.79880 MC346 2x (26) Pias, Ferreira do Zêzere, Portugal 39.71898 -8.32942 SC191 2x (19) Vale de Urso, Proença-a-Nova, Portugal SANT70515 39.74483 -7.89155 MC309 2x (23) Albergaria, Marinha Grande, Portugal 39.74603 -8.88242 LM014 2x (6) Dornes, Ferreira do Zêzere, Portugal 39.76843 -8.26169 MC362 2x (13) Janardo, Leiria, Portugal 39.78695 -8.77387 MC350 2x (29) Vale da Meda, Ourém, Portugal 39.79381 -8.53367 SC192 2x (16) Sertã, Portugal SANT70473 39.80863 -8.11590 SC190 2x (3) Silveira dos Figos, Castelo Branco, Portugal SANT70477 39.83898 -7.67476 MC360 2x (1) Vermoil, Pombal, Portugal 39.84320 -8.66282 Page 48 of 60 Botanical Journal of the Linnean Society Botanical Journal of the Linnean Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 PDF Proof SC189 2x (3) Castelo Branco, Portugal SANT70476 39.84571 -7.47026 LM009 2x (25) Souto de Carpalhosa, Leiria, Portugal 39.84595 -8.83335 MC368 2x (16) Granja, Bajouca, Portugal 39.85191 -8.87618 MC361 2x (9) Valeirão, Pombal, Portugal 39.87102 -8.75234 EM001 2x (24) Pedrógão Pequeno, Sertã, Portugal 39.89622 -8.13783 SC193 2x (14) Poço Negro, Figueiró-dos-Vinhos, Portugal 39.91965 -8.25304 LM005 2x (6) Casal Fernão João, Pombal, Portugal SANT70512 39.91980 -8.64870 LM003 2x (5) Casal Fernão João, Pombal, Portugal SANT70513 39.92007 -8.65293 MC330 2x (20) Pedrogão Grande, Portugal 39.92688 -8.15120 MC336 2x (22) Cancinos, Oleiros, Portugal 39.92747 -7.92447 MC329 2x (18) Fato, Figueiró dos Vinho, Portugal 39.92843 -8.32886 MC371 2x (18) Matos do Carriço, Pombal, Portugal 39.98608 -8.79657 SC188 2x (9) São Miguel da Acha, Idanha-a-Nova, Portugal SANT70485 40.00324 -7.33108 MC354 2x (1) Redinha, Pombal, Portugal 40.00532 -8.58961 MC334 2x (12) Alvares, Góis, Portugal 40.00980 -8.10677 MC298 2x (11) Casal da Rola, Pombal, Portugal 40.02076 -8.71501 SC184 2x (11) Bogas de Baixo, Oleiros, Portugal SANT70531 40.03460 -7.77846 MC369 2x (5) Torneira, Pombal, Portugal 40.03966 -8.75807 SC182 2x (18) Vilarinho, Lousão, Portugal SANT70479 40.12720 -8.20791 AM002 2x (4) Vale, Lousã, Portugal 40.13018 -8.23468 AM001 2x (6) Vale, Lousã, Portugal 40.13082 -8.23518 SC187 2x (16) Penamacor, Portugal SANT70506 40.13903 -7.20258 SC186 2x (15) Fatela, Fundão, Portugal SANT70478 40.16101 -7.42409 MC271 2x (10) Caneiro, Miranda do Corvo, Portugal 40.17904 -8.31525 MC292 2x (12) Torre do Mondego, Coimbra, Portugal 40.19314 -8.38946 SC179 2x (2) Madorno, Montemor-o-Velho, Portugal SANT70480 40.19499 -8.65544 MS01-01 S 2x (15) San Martín de Trevejo, Cáceres, Spain SANT62914 40.20992 -6.78700 MC307 2x (9) Santo António dos Olivais, Coimbra, Portugal 40.21064 -8.40051 SC180 2x (10) Meães, Montemor-o-Velho, Portugal SANT70481 40.21384 -8.60850 AA001 2x (4) Salgueiro, Fundão, Portugal 40.22853 -7.33197 MC272 2x (6) Quinta Grande, Coimbra, Portugal 40.23437 -8.42012 MC306 2x (10) Secarias, Arganil, Portugal 40.24692 -8.03352 SC225 2x (10) Caria, Belmonte, Portugal 40.29917 -7.35661 SC230 2x (10) Malcata, Sabugal, Portugal 40.29997 -7.05845 Page 49 of 60 Botanical Journal of the Linnean Society Botanical Journal of the Linnean Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 PDF Proof SC231 2x (7) Quadrazais, Sabugal, Portugal 40.31300 -7.00049 SC229 2x (7) Sabugal, Portugal 40.33050 -7.09721 SC227 2x (4) Azenha, Sabugal, Portugal 40.33134 -7.22465 MS07-02 S 2x (15) Candelario, Salamanca, Spain SANT76423 40.33659 -5.76651 MC281 2x (9) Cácemes, Mortágua, Portugal 40.33928 -8.34688 MS15-062 2x (6) Montemayor del Río, Salamanca, Spain SANT71386 40.34574 -5.88048 MC287 2x (12) Almacinha, Mortágua, Portugal 40.35560 -8.19451 SC241 2x (3) Catraia do Buraco, Belmonte, Portugal 40.37214 -7.32858 SC232 2x (3) Alfaiates, sabugal, Portugal 40.37808 -6.94432 MC283 2x (17) Freixo, Mortágua, Portugal 40.38542 -8.19956 MC282 2x (14) Barracão, Mortágua, Portugal 40.38915 -8.28630 MC269 2x (9) Várzea, Mealhada, Portugal 40.39001 -8.38288 MC285 2x (12) Santa Comba Dão, Portugal 40.39751 -8.12613 SC127 2x (3) Senhora do Espinheiro, Seia, Portugal SANT68766 40.41127 -7.67038 SC169 2x (27) Póvoa da Forcada, Carregal do Sal, Portugal SANT70482 40.41677 -8.03776 JCO001 2x (20) Quintãzinha do Mouratão, Guarda, Portugal 40.51386 -7.21647 SC239 2x (3) Quinta do Souto, Belmonte, Portugal 40.42413 -7.25116 SC238 2x (4) A-de-Moura, Guarda, Portugal 40.45907 -7.2212 SC236 2x (7) Adão, Guarda, Portugal 40.46016 -7.14943 SC208 2x (11) Vila Verde, Seia, Portugal 40.46337 -7.77766 MC273 2x (23) Santo Amaro, Tondela, Portugal 40.50142 -8.07658 MS15-065 2x (6) Villanueva del Conde, Salamanca, Spain SANT71411 40.50603 -6.01425 SC176 2x (19) Palhaça, Oliveira do Bairro, Portugal SANT70475 40.51136 -8.59473 SC170 2x (21) Nelas, Portugal SANT70483 40.53123 -7.86441 MS15-069 2x (6) Peña de Francia, Salamanca, Spain SANT71409 40.53424 -6.14993 MG001 2x (11) Caramulo, Tondela, Portugal 40.57402 -8.17512 MC248 2x (17) Granja, Guarda, Portugal SANT70492 40.60071 -7.10506 SC210 2x (4) Lajeosa do Mondego, Guarda, Portugal 40.63036 -7.35552 MC249 2x (28) Germil, Penalva do Castelo, Portugal SANT70494 40.65076 -7.73303 SC171 2x (28) Abraveses, Viseu, Portugal SANT70484 40.68324 -7.92715 MC296 2x (2) Vale, Estarreja Portugal 40.70659 -8.55917 MS16-020 2x* Soto del Real, Madrid, Spain 40.75042 -3.79266 MC291 2x (14) Parada, Sever do Vouga, Portugal 40.77002 -8.29246 FM001 2x (10) Falachos, Trancoso, Portugal SANT70514 40.77475 -7.31173 MC247 2x (26) Palhais, Trancoso, Portugal SANT70491 40.81355 -7.43169 Page 50 of 60 Botanical Journal of the Linnean Society Botanical Journal of the Linnean Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 PDF Proof MC295 2x (6) Torre, Ovar, Portugal 40.82751 -8.57485 MS01-02 S 2x (15) Rascafria, Madrid, Spain SANT62736 40.84123 -3.89195 MC253 2x (25) Vila Nova de Paiva, Portugal 40.86050 -7.74368 EA001 2x (11) Folgosa, Castro Daire, Portugal SANT70496 40.89789 -7.89528 MC246A 2x (9) Devesas, Mêda, Portugal 40.90534 -7.22070 MS04-01 S 2x (15) Luras, Sardenha, Italy SANT61449 40.93919 9.19006 MC255 2x (23) Beselga, Penedono, Portugal 40.94033 -7.41715 MC294 2x (16) Gamoal, Santa Maria da Feira, Portugal 40.95436 -8.59781 MC275A 2x (1) Carvalhosa, Castro Daire, Portugal 40.95601 -7.97051 MC254 2x (22) Moimenta da Beira, Portugal 40.96929 -7.60611 MC265 2x (20) Bigorne, Lamego, Portugal SANT70501 41.00491 -7.88631 MS09-01 2x* Galllocanta, Zaragoza, Spain 41.01189 -1.48764 SC211 2x (16) Almendra, Vila Nova de foz Côa, Portugal 41.02128 -6.99439 MS04-02 2x* Li Cossi, Trinitá d'Agultu, Italy SANT61452 41.04651 8.93621 MC256 2x (20) Sebadelhe, Vila Nova de Foz Côa, Portugal 41.05878 -7.28996 MC276 2x (18) Fornelos, Resende, Portugal 41.10547 -7.97133 MC264 2x (15) Varais Lamego, Portugal SANT70974 41.14550 -7.77917 MC257 2x (14) Ervedosa do Douro, São João da Pesqueira, Portugal SANT70511 41.17167 -7.47849 MS01-03 S 2x (5) Cerezo de Arriba, Segóvia, Spain SANT62744 41.20521 -3.47356 SC220 2x (16) Cancelas, Paredes, Portugal 41.22045 -8.35026 SC221 2x (16) Gondalães, Paredes, Portugal 41.22737 -8.33442 SC222 2x (14) Vales, Vizela, Portugal 41.24451 -8.29639 SC219 2x (21) Grifão, Paredes, Portugal 41.24545 -8.34648 SC218 2x (15) Sobroso, Paredes, Portugal 41.25119 -8.32535 SC194 2x (9) Via Pouca, Vizela, Portugal 41.25188 -8.29536 MC258 2x (9) Alijo, Portugal SANT70504 41.26786 -7.45585 MC394 2x (8) Fontiela, Trofa, Portugal 41.28733 -8.56441 MC259 2x (9) Vila Flor, Portugal SANT70502 41.30274 -7.15472 MC263 2x (19) Bairro da Carvalha, Vil Real, Portugal SANT70933 41.31980 -7.73189 MC245 2x (31) Alfândega da Fé, Vinhais, Portugal SANT70488 41.36845 -6.95750 MC262 2x (15) Seixo, Murça, Portugal SANT70510 41.40044 -7.44749 MC388 2x (17) Infantas, Guimarães, Portugal 41.42072 -8.24202 MC260 2x (15) Golfeiras, Mirandela, Portugal SANT70503 41.48038 -7.20335 MC244 2x (30) Bairro de Santa Luzia, Miranda do Douro, Portugal SANT70497 41.49838 -6.29006 Page 51 of 60 Botanical Journal of the Linnean Society Botanical Journal of the Linnean Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 PDF Proof MC243 2x (29) Macedo de Cavaleiros, Portugal SANT70486 41.52176 -6.97683 MC385 2x (9) Vargens, Valpaços, Portugal 41.57491 -7.44362 MC239 2x (29) Rita, Povoa de Lenhoso, Portugal SANT70490 41.58439 -8.32088 MC393 2x (10) Rugem, Barcelos, Portugal 41.61484 -8.56782 MC384 2x (10) Torre de Dona Chama, Mirandela, Portugal 41.65511 -7.14501 SC099 2x (16) Outeiro, Montalegre, Portugal SANT76617 41.68602 -7.94060 MC391 2x (12) Bouças, Terras do Bouro, Portugal 41.72944 -8.30498 MC382 2x (19) Mosca, Baçal, Portugal 41.76454 -6.80383 MC381 2x (17) Bolideira, Chaves, Portugal 41.77525 -7.32473 SC120 2x (28) Praia de Afife, Viana do Castelo, Portugal SANT69671 41.78630 -8.87042 MC392 2x (12) Salvador, Ponte da Barca, Portugal 41.80221 -8.35737 SC104 2x (18) Torneiros, Ourense, Spain SANT69680 41.85503 -8.11290 SC100 2x (21) Pitões das Júnias, Montalegre, Portugal SANT69683 41.86905 -7.95285 MS08-01 S 2x (15) Zicavo, Córsega, France 41.87789 9.15295 MC237 2x (30) Coussourado, Paredes de Coura, Portugal SANT70489 41.92381 -8.63805 MS08-02 S 2x (1) Campo dell'Oro, Córsega, France 41.92438 8.782612 SC119 2x (13) As Eiras, Pontevedra, Spain SANT69674 41.92591 -8.78854 SC097 2x (9) Parâmio, Braçal, Portugal SANT68764 41.92869 -6.88478 MC380 2x (4) Pazos, Ourense, Spain 41.93630 -7.46737 SC151 2x (5) Porreiras, Paredes de Coura, Portugal 41.94870 -8.55439 SC105 2x (27) Guxinde, Ourense, Spain SANT68769 41.97209 -8.15188 SC098 2x (7) Castrelos, Zamora, Spain SANT69682 42.00266 -6.89702 MS15-049 2x (7) Mombuey, Zamora, Spain SANT71438 42.01321 -6.34481 MS15-051 2x (8) A, Gudiña, Ourense, Spain SANT71431 42.04722 -7.13633 MS15-046 2x (6) Riego de Loma, Zamora, Spain SANT71436 42.07390 -6.68260 SC107 2x (11) Fiães, Melgaço, Portugal SANT76434 42.09481 -8.19062 MC379 2x (13) Trandeiras, Ourense, Spain 42.10841 -7.66274 SC108 2x (22) San Bieito, Pontevedra, Spain SANT76619 42.11550 -8.31584 MS14-061 2x (2) Lagoa de Samabria, Zamora, Spain SANT70541 42.11619 -6.72132 MC378 2x (17) Mandrás, Ourense, Spain 42.14466 -7.94348 SC109 2x (12) Vixiáns, Pontevedra, Spain SANT69681 42.26487 -8.33807 SC110 2x (13) O Piñeiro da Igrexa, Pontevedra, Spain SANT69679 42.27738 -8.37261 SC111 2x (30) O Piñeiro da Igrexa, Pontevedra, Spain SANT76439 42.28235 -8.38272 MS11-01 S 2x (15) Carrión de los Condes, Paléncia, Spain 42.31130 -4.72234 MS14-126 2x (1) Saceda, León, Spain SANT70551 42.32858 -6.52247 Page 52 of 60 Botanical Journal of the Linnean Society Botanical Journal of the Linnean Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 PDF Proof MS02-01 S 2x (15) Queralbs, Girona, Spain SANT60186 42.35393 2.16396 MS14-052 2x (4) O Rego, Ourense, Spain SANT70534 42.38071 -8.00801 MS14-050 2x (2) Casanova, Ourense, Spain 42.41356 -8.04025 MS14-125 2x (1) Pombriego, León, Spain SANT70552 42.41944 -6.69567 SC162 2x (17) Montes de Valdueza, León, Spain SANT70517 42.44958 -6.59802 SC164 2x (6) San Clemente de Valdueza, León, Spain SANT70554 42.46647 -6.54864 SC283 2x (2) Pradorrey, León, Spain 42.48694 -6.11456 SC284 2x (3) Catro Camiños, Lugo, Spain 42.55973 -7.51190 MS14-110 2x (7) Alto do Cadám, A Noveliza, Pontevedra, Spain SANT76600 42.60561 -8.26591 MS15-120 2x (5) Moreda do Courel, Lugo, Spain SANT72398 42.62507 -7.11880 SC271 2x (11) Fresnedo, León, Spain SANT76432 42.65532 -6.57830 MS15-122 2x (5) Pedrachantada, Lugo, Spain SANT72404 42.75522 -7.38493 SC267 2x (4) Pedrachantada, Lugo, Spain 42.75539 -7.38570 MS14-129 2x (1) Pico Sacro, A Coruña, Spain 42.80715 -8.44611 SC272 2x (10) Palacios del Sil, León, Spain SANT76436 42.87388 -6.45031 SC273 2x (2) Caboalles de Abajo, León, Spain SANT76620 42.87388 -6.45031 MS14-121 2x (6) O leboreiro, A Coruña, Spain SANT76601 42.88406 -7.98270 SC075 2x (12) Gures, A Coruña, Spain SANT69213 42.91249 -9.14875 SC276 2x (16) Villasecino, León, Spain SANT76433 42.96125 -6.05127 SC274 2x (5) Brañas de Arriba, Asturias, Spain SANT76427 43.01310 -6.44518 SC275 2x (10) Pontarás, Asturias, Spain SANT76614 43.08285 -6.54761 MS13-004 2x (1) Cabo Vilán, LA Coruña, Spain SANT68615 43.15536 -9.20540 MS14-118 2x (1) Vegadécima, Asturias, Spain SANT76602 43.15547 -6.95791 SC265 2x (3) Toiral, Lugo, Spain 43.17893 -7.61298 MS14-048 2x (6) Soesto, La Coruña, Spain SANT70546 43.20609 -9.01620 MS14-117 2x (4) Lago, Asturias, Spain SANT76603 43.25826 -6.71970 MS04-03 S 2x (15) Oiartzun, Gipuzkoa, Spain SANT61509 43.27755 -1.85519 MS15-074 2x (3) Praia da Barda, A Coruña, Spain SANT76604 43.28271 -8.92655 SC261 2x (27) As Toxeiras de Riba, Lugo, Spain SANT76612 43.30417 -7.86336 SC245 2x (2) Beo, La Coruña, Spain 43.32059 -8.84373 MS14-045 2x (28) Faro de Punta Nariga, A Coruña, Spain SANT70544 43.32071 -8.91042 MS14-116 2x (4) La Estrella, Asturias, Spain SANT76605 43.32573 -6.46229 SC259 2x (19) O Reguengo, Lugo, Spain SANT76438 43.39240 -7.80027 MS14-112 2x (5) Vilariño, Lugo, Spain SANT76606 43.41257 -7.60451 MS14-115 2x (3) Castañedo, Asturias, Spain SANT76608 43.44945 -6.37391 Page 53 of 60 Botanical Journal of the Linnean Society Botanical Journal of the Linnean Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 PDF Proof MS14-113 2x (2) Serra de Xistral, Lugo, Spain SANT76607 43.46848 -7.54026 SC250 2x (3) Cariño, A Coruña, Spain SANT76428 43.47066 -8.31559 MS14-114 2x (5) Coaña, Asturias, Spain SANT76621 43.50813 -6.74835 SC257 2x (7) A Agraxoiba, A Coruña, Spain SANT76430 43.52127 -7.94350 SC253 2x (6) Cabo Prior, A Coruña, Spain SANT76422 43.55296 -8.31155 MS10-001 S 2x (15) Cabo Vidio, Asturias, Spain SANT64247 43.59356 -6.24420 SC256 2x (8) A Areosa, A Coruña, Spain SANT73435 43.68170 -8.04729 SC254 2x (14) Punta Candieiro, A Coruña, Spain SANT76437 43.70746 -8.05096 MS16-017 2x* Ambazac, Limoges, France 45.95951 1.39128 M. Sheeby n.n. 2x (15) Anascaul Lake, Kerry, Ireland ABD n.n. 52.17875 -10.06399 MC367 4x (6) Amieira, Marinha Grande, Portugal SANT 39.76900 -8.91144 SC038 4x (37) Amor, Leiria, Portugal SANT68779 39.79586 -8.85416 SC042 4x (34) Corucho, Bajouca, Portugal SANT68762 39.86877 -8.83948 SC183 4x (13) Moradias, Pampilhosa da Serra, Portugal SANT70526 40.09120 -7.99071 MC323 4x (8) Nossa Senhora da Piedade, Lousã, Portugal 40.09878 -8.23438 SC185 4x (13) São Martinho, Fundão, Portugal SANT70522 40.12071 -7.68483 JM002 4x (26) Parrozelos, Arganil, Portugal SANT70507 40.20324 -7.90134 JC002 4x (20) Mata da Margaraça, Arganil, Portugal 40.21075 -7.92777 JC001 4x (21) Relva Velha, Arganil, Portugal SANT70509 40.21613 -7.90580 JM001 4x (11) Mata da Margaraça, Arganil, Portugal SANT70505 40.21738 -7.92158 SC181 4x (5) São João do Campo, Coimbra, Portugal SANT70527 40.23508 -8.50015 MC290 4x (7) Caneiro, Penacova, Portugal 40.23770 -8.31345 SC168 4x (28) Ponte, Penacova, Portugal SANT70524 40.27446 -8.27590 MC268 4x (16) Mata, Montemor-o-Velho, Portugal 40.28449 -8.67892 SC299 4x (5) Gavinhos, Coimbra, Portugal 40.28935 -8.31675 MC289 4x (6) Miro, Penacova Portugal 40.29552 -8.24626 MC280 4x (26) Casalito, Penacova, Portugal 40.29764 -8.31378 AA004 4x (7) Terlamonte, Covilhã, Portugal 40.29845 -7.44582 SC285 4x (5) Casal do Céu, Figueira da Foz, Portugal 40.29999 -8.78180 MC267 4x (5) Casal do João, Cantanhede, Portugal 40.32617 -8.75311 SC178 4x (20) Lemede, Cantanhede, Portugal SANT70528 40.33496 -8.61014 MC286 4x (1) Almacinha, Mortágua, Portugal 40.36062 -8.19430 SC131 4x (3) Lagoa Seca, Serra da Estrela, Portugal SANT68773 40.37139 -7.63578 SC145 4x (5) Ponte Jugais, Seia, Portugal SANT68777 40.38427 -7.70510 JC011 4x (32) Seia, Portugal 40.41750 -7.70591 Page 54 of 60 Botanical Journal of the Linnean Society Botanical Journal of the Linnean Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 PDF Proof MC266 4x (6) Cabeços, Vagos, Portugal 40.42994 -8.65257 MC270 4x (14) S. Lourenço do Bairro, Anadia, Portugal 40.44003 -8.47327 MC297 4x (2) Bunheira, Oliveira do Bairro, Portugal 40.52468 -8.51850 SC175 4x (5) Bolfiar, Águeda, Portugal SANT70529 40.56382 -8.39382 SC174 4x (30) Arca, Oliveira de Frades, Portugal SANT70520 40.60769 -8.21392 SC173 4x (21) Agros, Vouzela, Portugal SANT70521 40.63570 -8.19801 MC141 4x (29) Ponte de S. Tiago, Sever do Vouga, Portugal 40.70100 -8.37497 SC172 4x (17) Vilar, São Pedro do Sul, Portugal SANT70525 40.77453 -8.07215 MC301 4x (14) Paredes, Vale de Cambra, Portugal 40.81652 -8.37489 DT001 4x (6) Paço de Mato, Vale de Cambra, Portugal SANT70530 40.85100 -8.31014 LM004 4x (7) Drave, Arouca, Portugal 40.85963 -8.11861 MC252 4x (13) Ribolhos, Castro Daire, Portugal 40.88682 -7.92844 MC275 4x (31) Carvalhosa, Castro Daire, Portugal 40.95807 -7.96297 MC279 4x (32) Canedo, Santa Maria da Feira, Portugal 41.01160 -8.45774 MC278 4x (32) Santa Cecília, Castelo de Paiva, Portugal 41.04545 -8.27445 MC277 4x (11) Cidadelhe, Cinfães, Portugal 41.07827 -8.08915 SC223 4x (19) Boialvo, Gondomar, Portugal 41.08075 -8.47271 SC091 4x (18) Fragas, Amarante, Portugal SANT68763 41.26528 -7.90186 SC152 4x (5) Cotorinho, Vila Real, Portugal 41.27313 -7.87249 SC089 4x (17) Candemil, Amarante, Portugal SANT76616 41.27966 -7.91785 MC387 4x (13) Veade, Celorico de Basto, Portugal 41.41861 -7.97825 MC241 4x (29) Portela de Santa Eulália, Ribeira de Pena, Portugal SANT70523 41.50069 -7.79296 MC389 4x (11) Figueiró do Monte, Fafe, Portugal 41.55806 -8.07242 MC390 4x (11) Pombal, Vieira do Minho, Portugal 41.57658 -8.11106 MC377 4x (16) Barracão, Montalegre, Portugal 41.76378 -7.70449 MS09-02 4x (1) Illa do Faro, Galiza, Spain SANT60536 42.21562 -8.90836 SC115 4x (10) Limens, Pontevedra, Spain SANT69676 42.25865 -8.81608 MS14-060 4x (1) As Ermidas, Ourense, Spain SANT70537 42.28400 -7.12801 MS14-053 4x (2) Vale de Marcelle, Ourense, Spain SANT70533 42.30219 -7.61265 MS15-113 4x (7) Soutomaior, Pontevedra, Spain SANT72405 42.33273 -8.56529 MS14-057 4x (1) Mendoia, Ourense, Spain SANT70532 42.33795 -7.22352 MS14-055 4x (3) Piñeira, Ourense, Spain SANT70536 42.36158 -7.40242 SC112 4x (19) Forzáns, Pontevedra, Spain SANT76440 42.36261 -8.45016 MS14-059 4x (2) O Barco, Ourense, Spain SANT70538 42.42409 -7.01482 MS15-043 4x (6) Fontaíñas, Ourense, Spain SANT71426 42.45200 -8.02619 Page 55 of 60 Botanical Journal of the Linnean Society Botanical Journal of the Linnean Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 PDF Proof 96 MS15-118 4x (5) A Frieira, Lugo, Spain SANT72397 42.52176 -7.36829 RC001 4x (5) Vilamor, Lugo, Spain SANT72406 42.56537 -7.23140 MS15-119 4x (4) Parada dos Montes, Lugo, Spain SANT72399 42.57203 -7.27245 MS14-049 4x (9) Taboada, Pontevedra, Spain SANT70535 42.68757 -8.22030 SC269 4x (7) Pedrafita do Cebreiro, Lugo, Spain SANT76431 42.71282 -7.00560 SC268 4x (6) Fillobal, Lugo, Spain SANT76615 42.73892 -7.18958 MS15-123 4x (4) Paradela, Lugo, Spain SANT72400 42.76807 -7.56978 MS15-124B 4x (3) Paradela, Lugo, Spain SANT72403 42.77391 -7.62728 MS15-124A 4x (4) Paradela, Lugo, Spain SANT72402 42.77436 -7.62736 SC072A 4x (7) Ventim, A Coruña, Spain SANT69208 42.80037 -9.02645 MS09-03 4x (2) Santiago de Compostela, A Coruña, Spain SANT60533 42.87518 -8.55576 SC266 4x (10) Sabarei de Abaixo, Lugo, Spain SANT76613 42.89582 -7.47154 SC195 4x (6) Parada, A Coruña, Spain 42.96313 -8.84410 MS14-119 4x (2) Gondar, Lugo, Spain SANT76609 43.02258 -7.41816 MS14-111 4x (5) O Porto Velho, Lugo, Spain SANT76610 43.18596 -7.95939 SC263 4x (7) San Salvador, Lugo, Spain SANT76611 43.20826 -7.83605 Mixed-ploidy populations SC203 2x (1), 4x (15) Vila Rosa, Seia, Portugal 40.50062 -7.81098 MC274 2x (1), 4x (25) Fareja, Castro Daire, Portugal 40.89768 -7.91486 MC235 2x (19), 4x (14) Sobroso, Paredes, Portugal SANT70500 41.25087 -8.33041 MC386 2x (8), 4x (1) Nozedo, Vila Pouca de Aguiar, Portugal 41.50717 -7.62201 Page 56 of 60 Botanical Journal of the Linnean Society Botanical Journal of the Linnean Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 PDF Proof 97 Table S3 – Selected environmental variables using the sampled diploid and tetraploid 98 populations of Jasione montana. For each cytotype, mean and standard error of the mean (Mean 99 ± SE) are given. Different letters represent statistically significant differences between cytotypes 100 at P < 0.05, n.s. denote non-significant differences at P > 0.05. Diploid Tetraploid Variables CODE Mean ± SE, n = 462 Mean ± SE, n = 53 Temperature seasonality Bio_4 5451.46 ± 41.39a 4530.78 ± 88.31 b Maximum temperature of warmest month Bio_5 278.04 ± 1.72 a 256.70 ± 2.51 b Precipitation of driest month Bio_14 19.56 ± 0.74 n.s. 18.82 ± 1.21 n.s. Precipitation seasonality Bio_15 41.57 ± 0.64 a 49.44 ± 0.82 b Base saturation of the tops bs_top 1.56 ± 0.03 a 1.90 ± 0.04 b Dominant surface textural class of the STU txsrfdo 1.89 ± 0.03 n.s. 1.85 ± 0.05 n.s. 101 Page 57 of 60 Botanical Journal of the Linnean Society Botanical Journal of the Linnean Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60