Range-wide population genetics of Linum Suffruticosum S. L., an heterostylous polyploid and taxonomic complex

RANGE-WIDE POPULATION GENETICS OF LINUM SUFFRUTICOSUM S.L., AN HETEROSTYLOUS POLYPLOID AND TAXONOMIC COMPLEX Maria Antònia Vanrell Bosch Tutores: Juan Arroyo Marín y Violeta Simón-Porcar Máster Universitario en Biología Avanzada Facultad de Biología Noviembre de 2022 Abstract Linum suffruticosum L. s.l. (Linaceae) is a taxonomic complex widespread in the western Mediterranean basin. It is characterized by its high cytogenetic diversity and its 3D heterostyly system, which makes it an obligate outcrosser. In this work we studied the patterns of genetic diversity and population structure throughout its distribution using neutral genetic data obtained from microsatellite markers. It is shown that there is no relationship between taxonomic treatment and genetic variability, so a revision of the taxonomy would be useful. L. suffruticosum s.l. has a high level of genetic diversity, promoted by polyploidization processes, and maintained by the existence of abundant and reproductively connected populations. The populations are genetically structured following a pattern of isolation by distance, with the Rif mountains as the most important genetic barrier. However, genetic diversity does not covary at the latitudinal or altitudinal level, nor is it related to population size or the proportion of stylar morphs. This suggests that variations in genetic diversity are not mostly determined by selective forces, but by stochastic processes. Resumen Linum suffruticosum L. s.l. (Linaceae) es un complejo taxonómico ampliamente extendido en la cuenca mediterránea occidental. Se caracteriza por su alta diversidad citogenética y su sistema de heterostília en 3D, lo que lo convierte en un outcrosser obligado. En este trabajo se estudian los patrones de diversidad genética y estructura poblacional a lo largo de su distribución utilizando datos genéticos neutros obtenidos a partir de marcadores de microsatélites. Se ha demostrado que no existe relación entre el tratamiento taxonómico y la variabilidad genética, por lo que sería útil una revisión de la taxonomía. L. suffruticosum s.l. presenta una elevada diversidad genética, promovida por procesos de poliploidización, y mantenida por la existencia de poblaciones abundantes y conectadas reproductivamente. Las poblaciones están estructuradas genéticamente siguiendo un patrón de aislamiento por distancia, con las montañas del Rif como la barrera genética más importante. Sin embargo, la diversidad genética no covaría con la latitud o la altitud, ni está relacionada con el tamaño de la población o la proporción de morfos estilares. Esto sugiere que las variaciones en la diversidad genética no están determinadas principalmente por fuerzas selectivas, sino por procesos estocásticos. RANGE-WIDE POPULATION GENETICS OF LINUM SUFFRUTICOSUM S. Máster Universitario en Biología Avanzada 2022 Maria Antònia Vanrell 2 Introduction The Mediterranean basin is a biodiversity hotspot with an exceptional concentration of plant species (Heywood, 1995; Médail & Quezel, 1997). There are approximately 25 000 vascular plant taxa, thus representing 4.3% of the total plant species on the planet (Myers et al., 2000). Remarkably, about half of them (about 13 000 taxa) are endemic to the area. This exceptional species richness and high percentage of endemic taxa has been favored by the exceptional habitat heterogeneity, the climatic history strongly influenced by Quaternary glaciations, and the intense habitat fragmentation in this area (Greuter, 2001). Many widely distributed plant species present high variation in the morphological, demographic, and biological traits of populations, which are determined by geographic, ecological, and historical factors (López-Goldar et al., 2020). This diversity and the underlying microevolutionary processes are reflected in the patterns of genetic diversity and structure of the populations and are key for the destiny of the species. Linum suffruticosum L. s.l. (Linaceae) is a taxonomic complex of perennial woody plants with distinctive heterostylous white flowers widely distributed throughout the western Mediterranean basin (Martínez-Labarga & Muñoz -Garmendia, 2015). This species group has a particularly complex taxonomy due to its enormous morphological variability, which is probably increased by processes of hybridization, introgression between populations and polyploidy (Martínez-Labarga & Muñoz -Garmendia, 2015). In the absence of clear diagnostic characters, the definition of taxonomic entities has been made through morpho-geographical divisions. Historically, at least three taxa have been consensually accepted as distinct species: L. salsoloides Lam., L. appressum Caball. and L. suffruticosum L., (Jahandiez & Maire, 1932; López González, 1979; Ockendon & Walters, 1968), in addition to its different varieties. Nevertheless, the latest taxonomic treatment comprised more than 20 taxa for the Iberian Peninsula alone (Martínez- Labarga & Muñoz -Garmendia, 2015). Linum suffruticosum s.l. also presents a remarkable cytogenetic diversity. The group has been described as a polyploid complex with a basic chromosome number of 8 or 9, displaying a polyploid series from diploidy to decaploidy. This variation in the basic number of chromosomes and in the level of ploidy suggests multiple past events of chromosomal rearrangements, whole genome duplications and hybridization events (Afonso et al., 2021). The various cytotypes of L. suffruticosum s.l. are spatially segregated across its distribution and their geographic patterns are complex. The different ploidy levels are distributed parapatrically, with several contact zones between cytotypes that are not reproductively isolated, although mixed-ploidy populations are rarely found. This is expected to influence the genetic diversity and structure of these populations (Afonso et al., 2021). Polyploidization events can have a major effect on the genetic diversity of a lineage. In polyploidization, alteration of the genome triggers genetic and epigenetic responses, mostly of a stochastic nature (Madlung & Wendel, 2013). In the specific case of allopolyploidy, in which hybridization events have participated, complex gene expression combinations tend to be created, since new polyploid lineages are often formed from the hybridization of different populations, resulting in arrangements that incorporate greater genetic diversity (D. E. Soltis et al., 2009; P. S. Soltis & Soltis, 2000; van de Peer et al., 2021). Thus, polyploidy can promote diversity, but the formation of a new cytotype can cause a reproductive barrier to be created with the rest of the population. As a consequence, a bottleneck would be caused, and the isolation could be linked to a shift towards autogamy that prevents the exclusion of the minority cytotype (Levin, 1975; van Drunen & Husband, 2019). In that case there would be a reduction in genetic diversity. That said, polyploidy is considered a driver of speciation and rapid RANGE-WIDE POPULATION GENETICS OF LINUM SUFFRUTICOSUM S. Máster Universitario en Biología Avanzada 2022 Maria Antònia Vanrell 3 diversification (van de Peer et al., 2017, 2021), if populations are isolated but viable, they can promote genetic differentiation (Chester et al., 2012; Levin, 2013). On the contrary, if gene flow is maintained, it may happen that the populations increase their genetic variability and success or that the recombination results in the homogenization of the populations, thus maintaining diversity but not differentiation (Meimberg et al., 2009; Modliszewski & Willis, 2012). The effects of polyploidization are complex and can have a variety of evolutionary consequences in lineages. Linum suffruticosum s.l. presents an interesting reproductive strategy which includes floral polymorphism. Heterostyly - defined by reciprocal position of anthers and stigmas of opposite morphs within a population which improves pollen transfer between morphs (Barrett et al., 2000; Darwin, 1863, 1877) - is one of the most distinctive characteristics of L. suffruticosum (Armbruster et al., 2006; Martínez-Labarga & Muñoz -Garmendia, 2015). Among the flowers of L. suffruticosum, two different morphs can be found, but beyond the common distyly, in this case the stigmas and anthers of complementary morphs show full reciprocity in three dimensions: the anthers and stigmas are not reciprocal in height, but rather in how the stamens and styles twirl. In long-styled morph flowers (L-morph), styles form a column in the center of the flower with stigmas facing outward, and stamens spread from near the base, appressed to the corolla wall, with inwardly dehisced anthers. In short-styled morph flowers (S-morph), styles spread outward with the stigmas facing inwards, and the stamens are erect, forming a column in the center of the flower with their anthers rotated such that their dehiscence is outwards (Figure 1). The result of this polymorphism is that, depending on the morph, the pollen would be placed on the dorsal or ventral part of the pollinator; and, in the same way, the stigmas would come into contact only with the upper or underside of the pollinator. This mechanism ensures crosspollination in a more effective way than with the common distyly, in which the pollen lands in a more frontal or more distal part of the body of the pollinator. To date, Linum suffruticosum is the only known species with this type of 3D heterostyly (Armbruster et al., 2006). Figure 1. Flowers of Linum suffruticosum with petals removed, note the difference in the two morphs: The shortstyled morph (left) and the long-styled morph (right). Font: Armbruster et al. (2006). The morphological polymorphism in L. suffruticosum is associated to a heteromorphic selfincompatibility system that only allows fertilization between morphs, favoring disassortative mating (Afonso, 2022; Nicholls, 1985). Evolutionary models state that, in a population with disassortative mating in equilibrium, natural selection yields equal proportions of style morphs (Lloyd & Webb, 1992). Stochastic factors (e.g. habitat fragmentation) can deviate the morph ratio of populations. Unbalanced morph ratios reduce the abundance of compatible mates (i.e., the effective population size) (van Rossum et al., 2006a), and this can result in a loss of genetic diversity (Meeus et al., 2012). According to the observations of Afonso et al. (2021), most wild flax populations present an approximate 1:1 ratio of style morphs irrespective of ploidy and RANGE-WIDE POPULATION GENETICS OF LINUM SUFFRUTICOSUM S. Máster Universitario en Biología Avanzada 2022 Maria Antònia Vanrell 4 population size, suggesting that this complex reproductive system maintains selfincompatibility. Even that, it is unknown yet to what extent the heteromorphic selfincompatibility system and the proportion of morphs may vary in populations throughout the distribution range. Analyzing the genetic diversity and differentiation of L. suffruticosum s.l. populations appears as a key to understanding the evolutionary history and ecological forces driving the intra-specific variation in this taxonomic complex. The selective pressures faced by populations throughout the whole distribution, in addition to the stochastic processes that affect the differentiation between populations will be reflected in their neutral genetic variation (Barrett, 2010). Reproductive biology is also a relevant factor in explaining genetic diversity and population structure. The complex heterostylous system of the wild flax is expected to have important effects on the level of genetic differentiation of populations and will have different consequences than those found among self-fertilizing species (Crawford & Whitney, 2010). The main aim of this study was to determine the patterns of genetic diversity and population structure of L. suffruticosum s.l. in populations of contrasted environmental conditions throughout the distribution range of the species, through the use of nuclear microsatellite markers. In addition, we aimed to determine the correlates of the genetic variation with various biological and ecological (morph-ratio, population size, latitude, and elevation, ploidy level and taxonomic entity as surrogate of morphological variation) traits of populations. Apart from possible taxonomic implications, these results will shed light on the microevolution of this group and, in general, on the genetic structuring patterns of Mediterranean outcrossing polyploid species complexes. Materials and methods Sample collection We sampled a total of 32 L. suffruticosum s.l. populations throughout the entire species’ range, from Morocco to the southern East coast of France, through the East half of the Iberian Peninsula, during the 2018 to 2021 spring flowering seasons. For each population, we estimated population size by counting plants across haphazardly chosen areas and extrapolating the number for the complete population. We estimated the morph ratio by visual determination of the floral morph of 60 randomly chosen flowers from different individuals (one flower per plant, total N = 1427 flowers). In populations smaller than 60, all plants were checked. We individually collected leaves from 12 individuals of known morph and conserved them in silica gel. For 21 (65%) of these sampled populations, we compiled available information on ploidy level (diploid to octoploid populations) and entities of putative taxonomic value (determined as L. suffruticosum, Moroccan L. suffruticosum, or L. appressum-salsoloides) from Afonso et al. (2021). We determined the taxonomic entity of the remaining populations following the same criteria as in Afonso et al. (2021) (Table S1, Supplementary material). DNA isolation, SSR amplification and genotyping Genomic DNA was extracted from the dried leaf samples (total N=382 individuals) in 2022 using the Invisorb® Spin Plant Mini Kit (Invitek Molecular GmbH, Germany). The concentration and quality of the extracted DNA were assessed with a Nanodrop DeNovix DS-11 Spectophotometer (DeNovix Inc., USA). RANGE-WIDE POPULATION GENETICS OF LINUM SUFFRUTICOSUM S. Máster Universitario en Biología Avanzada 2022 Maria Antònia Vanrell 5 We used the specific L. suffruticosum SSR panel developed by Olmedo et al. (unpublished; Table S2, Supplementary material) to genotype the samples. Although this panel includes a total of 12 SSR markers, only six of these appeared suitable for this study after an exploratory probe in one individual from each of 16 populations. We performed simplex PCR for each of these six SSR in 20 μL of master mix including: 1x MyTaq Red Reaction Buffer (Bioline), 0.40 μM dye-labeled M13 primer, 0.40 μM PIG-tailed reverse primer, 0.04 μM M13-tailed forward primer, 0.01% bovine serum albumin (BSA, Promega), 0.5 μL MyTaqTM Red DNA Polymerase (Bioline), 50-70 ng gDNA and deionized water up to 20 μL. M13-tailed forward primers were labelled with either 6-FAM, VIC, NED, or PET fluorescent labels. The PCRs were performed in a Veriti™ 96- Well Thermal Cycler (Applied Biosystems™, USA) following the same touchdown procedure for all loci. This consisted of an initial denaturation for 2 min at 94°C; followed by 10 cycles of 92°C for 30 s, 30 s at 63°C with an increment of −1°C per cycle, and 30 s at 72°C; followed by 20 cycles of 94°C for 30 s, 30 s at 56°C, and 30 s at 72°C; and an extra extension of 5min at 72°C. Amplification products were genotyped by MACROGEN (Madrid, Spain) in an automatic ABI 3730 capillary DNA sequencer, using a GeneScan 500 LIZ internal size standard. The resulting electropherograms were analyzed for allele binning and calling using PeakScanner™ Software V2.0 (Applied Biosystems™, USA). Data analysis Genetic diversity The retrieved genotypes were used to characterize the genetic diversity of L. suffruticosum s.l. populations. Given the autopolyploid nature of the studied complex, GenoDive V3.06 (Meirmans, 2020) was used to calculate the standard metrics of genetic diversity per population and per locus: number of alleles (Na), effective number of alleles (Ne), expected heterozygosity (He) and total and corrected total heterozygosity (Ht, and H’t), the last two only on a per locus basis. The FIS statistic was used to test deviations from Hardy-Weinberg Equilibrium (HWE). Depending on the analysis, GenoDive corrects for the unknown dosage of alleles in polyploids in the estimates of genetic parameters by a maximum likelihood method based on random mating within populations or a method based on population allele frequencies. Since this correction is not possible in the calculation of observed heterozygosity (Ho) and inbreeding coefficient (GIS) statistics, these parameters were not estimated. Genetic structure We evaluated the genetic differentiation among populations, ploidy levels and taxonomic entities by analyses of molecular variance (AMOVA) run in GenoDive (Meirmans, 2020). The isolation by distance was assessed by regressing population pairwise genetic distances (based in FST) against the geographic Euclidean distances between populations using a Mantel test as implemented in the vegan 2.6-4 R package (Oksanen et al., 2022). Given the genetic differentiation found among ploidy levels, we carried out a second Mantel test to examine correlation between FST-based genetic distances and ploidy level-based distances (calculated as the pairwise difference between population ploidy levels; e.g. a value of six for a diploid - octoploid population pair). A Bayesian clustering of individuals was performed using the STRUCTURE V2.3.4 software (Pritchard et al., 2000), which has been shown to be the most suitable clustering method in mixed-ploidy systems (Stift et al., 2019). First, we estimated the population structure for the complete dataset using the admixture model and the locations as prior with a 2x104 burn-in period and subsequent 1x105 MCMC steps. We evaluated 1 ≤ K ≤ 10 potential genetic clusters (K) with 10 replicates for each K value. The optimal number of K was determined by the Evanno method (Evanno et al., 2005), checked with the aid of the Pophelper app (Francis, 2017). As in this first clustering the optimal K was found to be 2 (Figure S1 A, RANGE-WIDE POPULATION GENETICS OF LINUM SUFFRUTICOSUM S. Máster Universitario en Biología Avanzada 2022 Maria Antònia Vanrell 6 Supplementary material), we created a subsample of the dataset including the individuals of the populations assigned to cluster 1 and repeated the analysis using the same parameters. The graphical representation of population genetic structure was generated with the Pophelper app. Last, a principal component analysis (PCA) was carried out based on a matrix of covariance of allele frequencies among populations with GenoDive and plotted using the ggplot2 R package (Wickham, 2016). Ecological and biological correlates of genetic diversity We tested whether the genetic diversity of populations differed among ploidy levels, taxonomic entities and genetic clusters retrieved by STRUCTURE analyses (see below) using ANOAs and Tukey HSD post hoc tests, using the stats R package (R Core Team, 2021). Three populations had unassigned taxonomic entities and one population had a mixture of diploid and triploid individuals. The analyses were repeated by assigning these populations to each putative group. Similar results were retrieved in all cases, and hence we report the results for the analyses classifying such populations as L. suffruticosum and as diploid. We also analyzed the possible correlation between genetic diversity and the following continuous ecological variables of populations: percentage of L-morph, population size, latitude and elevation. Given the spatial nature of our data, we first calculated the Moran’s I statistic using the R package ape V5.6 (Paradis & Schliep, 2019) to evaluate the possible spatial autocorrelation of each variable. Given the lack of significant spatial autocorrelation for most of them (Table S3, Supplementary material), we performed linear regressions to analyze the association between each genetic diversity parameter and ecological variables. Results Characterization of SSR markers Among the 6 SSR markers analyzed, a total of 84 different alleles were found, with an average of 14 alleles per locus (range 10-30). X337 was the locus with the greatest genetic diversity: besides having the highest number of alleles (Na=30), it had the highest effective number of alleles (Ne = 3.883) and the highest expected, total and corrected total heterozygosity (He = 0.835, Ht = 0.924, H’t = 0.927) (Table 1). Genetic diversity of populations The average number of alleles in a population (Na) was 4.844, ranging from 2.167 (FLAX24) to 7.833 (FLAX78). Similarly, the average number of effective alleles in a population (Ne) was 3.05, ranging from 1.709 (FLAX24) to 4.86 (FLAX78). The expected heterozygosity (He) ranged from 0.381 (FLAX 61) to 0.744 (FLAX9), with an average of 0.584 (Table 2). Based on FIS statistics, all the populations deviated from HWE (P <0.001). Table 1. Indices of genetic diversity per locus. Abbreviations: Na, Number of alleles per locus; Ne Effective number of alleles per locus; He, Expected Heterozygosity; Ht, Total Heterozygosity; H't, Corrected Total Heterozygosity. Locus Na Ne He Ht H't X114 12 1.971 0.515 0.671 0.676 X337 30 3.883 0.835 0.924 0.927 X373 10 2.743 0.690 0.852 0.857 X395 10 1.243 0.209 0.713 0.729 X421 11 3.026 0.698 0.843 0.848 X943 11 2.044 0.549 0.766 0.773 Overall 14 ± 7.188 2.485 ± 0.848 0.583 ± 0.197 0.795 ± 0.087 0.802 ± 0.084 RANGE-WIDE POPULATION GENETICS OF LINUM SUFFRUTICOSUM S. Máster Universitario en Biología Avanzada 2022 Maria Antònia Vanrell 7 Genetic structure The analyses of molecular variance (AMOVA) revealed a significant differentiation among populations and among ploidy levels (Table 3 A), but not among taxonomic entities (Table 3 B). First, 45.9% of total genetic variation was found within populations, followed by a 42.2% found among populations, and an 11.9% found among ploidy levels. Nevertheless, there was no correlation between the genetic distance and ploidy differentiation of populations, according to the corresponding Mantel test (Table 4 A). However, another Mantel test showed a significant correlation between the genetic distance and the geographical distance of populations, fitting the isolation by distance model (Table 4 B). The Bayesian population analyses conducted with STRUCTURE found that the most likely ΔK was maximized at K=2 (Figure S1 A, Supplementary material), suggesting two distinct genetic clusters (Figure 2 A). Only three Moroccan populations were assigned to the cluster 2 (AA100, AA111 and AA112), and the other 29 populations were assigned to the cluster 1 (Figure 2 B). When removing the three populations of cluster 2 from the analysis, the ΔK was maximized in the value of K=7 (Figure S1 B, Supplementary material), which translates into seven differentiated genetic clusters (Figure 2). Table 2. Indices of genetic diversity in each population. Abbreviations: Na, Number of alleles per population; Ne, Effective number of alleles per population; He, Expected heterozygosity. Population Na Ne He FLAX2 4.333 2.709 0.640 FLAX5 4.000 2.494 0.520 FLAX9 3.833 2.883 0.744 FLAX13 4.167 2.413 0.602 FLAX24 2.167 1.709 0.398 FLAX29 5.000 2.926 0.546 FLAX30 4.833 3.371 0.601 FLAX33 7.167 4.533 0.736 FLAX36 2.833 2.026 0.459 FLAX37 5.167 3.282 0.531 FLAX39 4.833 3.075 0.518 FLAX45 5.500 2.841 0.581 FLAX46 3.667 2.622 0.539 FLAX53 4.500 3.050 0.583 FLAX56 5.167 3.373 0.646 FLAX59 5.333 3.590 0.603 FLAX61 2.333 1.840 0.381 FLAX62 5.167 3.332 0.567 FLAX69 4.000 2.797 0.661 FLAX71 7.333 4.200 0.680 FLAX73 3.833 2.934 0.508 FLAX75 7.333 4.266 0.689 FLAX77 4.500 2.518 0.560 FLAX78 7.833 4.860 0.717 FLAX79 6.500 3.675 0.544 FLAX80 6.833 3.448 0.611 FLAX81 5.333 2.814 0.497 AA100 5.000 2.988 0.669 AA105 3.667 2.550 0.647 AA107 3.167 2.117 0.515 AA111 6.000 3.419 0.690 AA112 3.667 2.955 0.510 Overall 4.844 ± 1.447 3.050 ± 0.731 0.584 ± 0.092 RANGE-WIDE POPULATION GENETICS OF LINUM SUFFRUTICOSUM S. Máster Universitario en Biología Avanzada 2022 Maria Antònia Vanrell 8 Table 3. Analysis of molecular variance (AMOVA). Test of the variations within populations, among populations and among ploidy levels (A) or among taxonomic entities (B). Source of variation P-value %Var (A) Within Population - 0.459 Among Population P < 0.001 * 0.422 Among ploidy levels P < 0.001 * 0.119 (B) Within Population - 0.482 Among Population P < 0.001 * 0.521 Among taxonomic entities 0.579 0 Table 4. Results of the Mantel's test showing the correlation between genetic differentiation (distance based on FST) and ploidy-based distance (A) or geographic Euclidean distance (B). Mantel statistic r Simulated P-value (A) FST distance – Ploidy-based distance -0.165 0.956 (B) FST distance – Geographic distance 0.281 0.001 * (A) (B) Figure 2. Genetic structure of the sampled populations of Linum suffruticosum . (A) Results of the Bayesian analysis of population structure (with software STRUCTURE, see Material and Methods). Plots showing two genetic clusters(K=2) above, and seven genetic clusters (K=7) after removing three Moroccan populations from the analyses (below). (B) Geographic distribution of the genetic clusters. Each population is represented by a pie chart that shows the percentage of membership in each genetic cluster for K=2 (left) and K=7 after removing three Moroccan populations (right). RANGE-WIDE POPULATION GENETICS OF LINUM SUFFRUTICOSUM S. Máster Universitario en Biología Avanzada 2022 Maria Antònia Vanrell 15 References Afonso, A. (2022). The role of polyploidy in shaping the diversity within Linum suffruticosum s.l.(Linaceae) [PhD Dissertation]. University of Coimbra. Afonso, A., Loureiro, J., Arroyo, J., Olmedo-Vicente, E., & Castro, S. (2021). Cytogenetic diversity in the polyploid complex Linum suffruticosum s.l. (Linaceae). Botanical Journal of the Linnean Society, 195(2), 216–232. https://doi.org/10.1093/BOTLINNEAN/BOAA060 Armbruster, W. S., Pérez-Barrales, R., Arroyo, J., Edwards, M. E., & Vargas, P. (2006). Three-dimensional reciprocity of floral morphs in wild flax (Linum suffruticosum): a new twist on heterostyly. New Phytologist, 171(3), 581–590. https://doi.org/10.1111/J.1469-8137.2006.01749.X Arroyo, J., Aparicio, A., Albaladejo, R. G., Muñoz, J., & Braza, R. (2008). Genetic structure and population differentiation of the Mediterranean pioneer spiny broom Calicotome villosa across the Strait of Gibraltar. Biological Journal of the Linnean Society, 93(1), 39–51. https://doi.org/10.1111/J.1095-8312.2007.00916.X Barrett, S. C. H. (2010). Darwin’s legacy: the forms, function and sexual diversity of flowers. Philosophical Transactions of the Royal Society B: Biological Sciences, 365(1539), 351–368. https://doi.org/10.1098/RSTB.2009.0212 Barrett, S. C. H., Jesson, L. K., & Baker, A. M. (2000). The Evolution and Function of Stylar Polymorphisms in Flowering Plants. Annals of Botany, 85(suppl_1), 253–265. https://doi.org/10.1006/ANBO.1999.1067 Barrett, S. C. H., Yakimowski, S. B., Field, D. L., & Pickup, M. (2010). Ecological genetics of sex ratios in plant populations. Philosophical Transactions of the Royal Society B: Biological Sciences, 365(1552), 2549–2557. https://doi.org/10.1098/RSTB.2010.0002 Chester, M., Gallagher, J. P., Symonds, V. V., da Silva, A. V. C., Mavrodiev, E. v., Leitch, A. R., Soltis, P. S., & Soltis, D. E. (2012). Extensive chromosomal variation in a recently formed natural allopolyploid species, Tragopogon miscellus (Asteraceae). Proceedings of the National Academy of Sciences of the United States of America, 109(4), 1176–1181. https://doi.org/10.1073/PNAS.1112041109/SUPPL_FILE/PNAS.201112041SI.PDF Crawford, K. M., & Whitney, K. D. (2010). Population genetic diversity influences colonization success. Molecular Ecology, 19(6), 1253–1263. https://doi.org/10.1111/J.1365-294X.2010.04550.X Darwin, C. (1863). On the existence of two forms, and on their reciprocal sexual relation, in several species of the genus Linum. Journal of the Proceedings of the Linnean Society of London. Botany, 7(26), 69–83. https://doi.org/10.1111/J.1095-8312.1863.TB01061A.X Darwin, C. (1877). The Different Forms of Flowers on Plants of the Same Species (J. Muray, Ed.). http://www.public-library.uk/pdfs/5/205.pdf Douaihy, B., Vendramin, G. G., Boratyński, A., Machon, N., & Dagher-Kharrat, M. B. (2011). High genetic diversity with moderate differentiation in Juniperus excelsa from Lebanon and the eastern Mediterranean region. AoB PLANTS, 2011(1). https://doi.org/10.1093/AOBPLA/PLR003 Eckert, C. G., Samis, K. E., & Lougheed, S. C. (2008). Genetic variation across species’ geographical ranges: the central–marginal hypothesis and beyond. Molecular Ecology, 17(5), 1170–1188. https://doi.org/10.1111/J.1365-294X.2007.03659.X Escudero, M., Vargas, P., Valcárcel, V., & Luceño, M. (2008). Strait of Gibraltar: an effective gene-flow barrier for wind-pollinated Carex helodes (Cyperaceae) as revealed by DNA sequences, AFLP, and cytogenetic variation. American Journal of Botany, 95(6), 745–755. https://doi.org/10.3732/AJB.2007342 RANGE-WIDE POPULATION GENETICS OF LINUM SUFFRUTICOSUM S. Máster Universitario en Biología Avanzada 2022 Maria Antònia Vanrell 16 Evanno, G., Regnaut, S., & Goudet, J. (2005). Detecting the number of clusters of individuals using the software structure: a simulation study. Molecular Ecology, 14(8), 2611–2620. https://doi.org/10.1111/J.1365- 294X.2005.02553.X Francis, R. M. (2017). pophelper: an R package and web app to analyse and visualize population structure. Molecular Ecology Resources, 17(1), 27–32. https://doi.org/10.1111/1755-0998.12509 Frankham, R., Ballou, J. D., Briscoe, D. A., & McInnes, K. H. (2002). Introduction to Conservation Genetics. Introduction to Conservation Genetics. https://doi.org/10.1017/CBO9780511808999 González-Martínez, S. C., Alía, R., & Gil, L. (2002). Population genetic structure in a Mediterranean pine (Pinus pinaster Ait.): a comparison of allozyme markers and quantitative traits. Heredity 2002 89:3, 89(3), 199– 206. https://doi.org/10.1038/sj.hdy.6800114 Greuter, W. (2001). Diversity of Mediterranean island floras. Bocconea, 13, 55–64. Heywood, V. H. (1995). The Mediterranean flora in the context of world biodiversity. Ecologia Mediterranea, 21(1–2), 11–18. Jahandiez, É., & Maire, R.-C.-J.-E. (1932). Catalogue des plantes du Maroc. Tome deuxiéme. Dicotylédones Archichalamydées. Alger : Imprimerie Minerva. En Vente à Paris, chez P. Lechevalier, libraire 12, Rue de Tournon (6o). https://doi.org/10.3/JQUERY-UI.JS Jump, A. S., Rico, L., Lloret, F., & Peñuelas, J. (2009). Microspatial population genetic structure of the Mediterranean shrub Fumana thymifolia. Plant Biology, 11(2), 152–160. https://doi.org/10.1111/J.1438- 8677.2008.00109.X Levin, D. A. (1975). Minority cytotipe exclusion in local plant populations. Taxon, 24(1), 35–43. https://doi.org/10.2307/1218997 Levin, D. A. (2013). The timetable for allopolyploidy in flowering plants. Annals of Botany, 112(7), 1201–1208. https://doi.org/10.1093/AOB/MCT194 Lloyd, D. G., & Webb, C. J. (1992). The Evolution of Heterostyly. In Evolution and Function of Heterostyly (pp. 151–178). Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-86656-2_6 López González, G. (1979). Algunas consideraciones sobre los linos del grupo Linum tenuifolium L. en España. In Memoires de la Société Botanique de Genéve (Vol. 1, pp. 99–109). https://doi.org/10.1016/J.TREE.2008.10.008 López-Goldar, X., Zas, R., & Sampedro, L. (2020). Resource availability drives microevolutionary patterns of plant defences. Functional Ecology, 34(8), 1640–1652. https://doi.org/10.1111/1365-2435.13610/SUPPINFO Loveless, M. D., & Hamrick, J. L. (1984). Ecological Determinants of Genetic Structure in Plant Populations. Annual Review of Ecology and Systematics, 15, 65–95. https://www.jstor.org/stable/2096943#metadata_info_tab_contents Madlung, A., & Wendel, J. F. (2013). Genetic and Epigenetic Aspects of Polyploid Evolution in Plants. Cytogenetic and Genome Research, 140(2–4), 270–285. https://doi.org/10.1159/000351430 Maguilla, E., Escudero, M., Jiménez-Lobato, V., Díaz-Lifante, Z., Andrés-Camacho, C., & Arroyo, J. (2021). Polyploidy Expands the Range of Centaurium (Gentianaceae). Frontiers in Plant Science, 12, 373. https://doi.org/10.3389/FPLS.2021.650551/BIBTEX Martínez-Labarga, J., & Muñoz -Garmendia, F. (2015). Linum L. In Castroviejo S, Muñoz Garmendia F, Navarro C, Quintanar A, & Buira A (Eds.), Flora Ibérica. Vol. 9 Rhamnaceae-Polygalaceae (Vol. 9, pp. 172–249). Real Jardín Botánico, CSIC. RANGE-WIDE POPULATION GENETICS OF LINUM SUFFRUTICOSUM S. Máster Universitario en Biología Avanzada 2022 Maria Antònia Vanrell 17 Martínez-López, V., García, C., Zapata, V., Robledano, F., & de la Rúa, P. (2020). Intercontinental long-distance seed dispersal across the Mediterranean Basin explains population genetic structure of a bird-dispersed shrub. Molecular Ecology, 29(8), 1408–1420. https://doi.org/10.1111/MEC.15413 Médail, F., & Quezel, P. (1997). Hot-spots analysis for conservation of plant biodiversity in the mediterranean basin. Annals of the Missouri Botanical Garden, 84(1), 112–127. Meeus, S., Honnay, O., Brys, R., & Jacquemyn, H. (2012). Biased morph ratios and skewed mating success contribute to loss of genetic diversity in the distylous Pulmonaria officinalis. Annals of Botany, 109(1), 227– 235. https://doi.org/10.1093/AOB/MCR272 Meimberg, H., Rice, K. J., Milan, N. F., Njoku, C. C., & McKay, J. K. (2009). Multiple origins promote the ecological amplitude of allopolyploid Aegilops (Poaceae). American Journal of Botany, 96(7), 1262–1273. https://doi.org/10.3732/AJB.0800345 Meirmans, P. G. (2020). genodive version 3.0: Easy-to-use software for the analysis of genetic data of diploids and polyploids. Molecular Ecology Resources, 20(4), 1126–1131. https://doi.org/10.1111/1755-0998.13145 Modliszewski, J. L., & Willis, J. H. (2012). Allotetraploid Mimulus sookensis are highly interfertile despite independent origins. Molecular Ecology, 21(21), 5280–5298. https://doi.org/10.1111/J.1365- 294X.2012.05706.X Molina-Venegas, R., Aparicio, A., Lavergne, S., & Arroyo, J. (2015). The building of a biodiversity hotspot across a land-bridge in the Mediterranean. Proceedings of the Royal Society B: Biological Sciences, 282(1813). https://doi.org/10.1098/RSPB.2015.1116 Myers, N., Mittermeler, R. A., Mittermeler, C. G., da Fonseca, G. A. B., & Kent, J. (2000). Biodiversity hotspots for conservation priorities. Nature, 403(6772), 853–858. Nicholls, M. S. (1985). Pollen flow, population composition, and the adaptive significance of distyly in Linum tenuifolium L. (Linaceae). Biological Journal of the Linnean Society, 25(3), 235–242. https://doi.org/10.1111/J.1095-8312.1985.TB00395.X Nicholls, M. S. (1986). Variation and evolution inLinum tenuifolium (Linaceae). Plant Systematics and Evolution 1986 153:3, 153(3), 243–258. https://doi.org/10.1007/BF00983691 Ockendon, D. J., & Walters, S. M. (1968). Linaceae. In Flora Europea (Vol. 2, pp. 584–603). Oksanen, J., Simpson, G. L., Blanchet, G., Kindt, R., Legendre, P., Minchin, P. R., & O’Hara, R. B. (2022). vegan: Community Ecology Package. R package (2.6-4). https://CRAN.R-project.org/package=vegan Paradis, E., & Schliep, K. (2019). ape 5.0: an environment for modern phylogenetics and evolutionary analyses in R. Bioinformatics, 35(3), 526–528. https://doi.org/10.1093/BIOINFORMATICS/BTY633 Pool, J. E., Hellmann, I., Jensen, J. D., & Nielsen, R. (2010). Population genetic inference from genomic sequence variation. Genome Research, 20(3), 291–300. https://doi.org/10.1101/GR.079509.108 Pritchard, J. K., Stephens, M., & Donnelly, P. (2000). Inference of population structure using multilocus genotype data. Genetics, 155(2), 945–959. https://doi.org/10.1093/GENETICS/155.2.945 R Core Team. (2021). R: A language and environment for statistical computing. R Foundation for Statistical Computing. In ci.nii.ac.jp. https://www.R-project.org/ Rice, A., Šmarda, P., Novosolov, M., Drori, M., Glick, L., Sabath, N., Meiri, S., Belmaker, J., & Mayrose, I. (2019). The global biogeography of polyploid plants. Nature Ecology & Evolution, 3(2), 265–273. https://doi.org/10.1038/s41559-018-0787-9 Ruiz-Martín, J., Santos-Gally, R., Escudero, M., Midgley, J. J., Pérez-Barrales, R., & Arroyo, J. (2018). Style polymorphism in Linum (Linaceae): a case of Mediterranean parallel evolution? Plant Biology, 20, 100–111. https://doi.org/10.1111/PLB.12670 RANGE-WIDE POPULATION GENETICS OF LINUM SUFFRUTICOSUM S. Máster Universitario en Biología Avanzada 2022 Maria Antònia Vanrell 18 Si, X., Gao, L., Song, Y., Khayatnezhad, M., & Minaeifar, A. A. (2020). Understanding population differentiation using geographical, morphological and genetic characterization in erodium cicunium. Indian Journal of Genetics and Plant Breeding, 80(4), 459–467. https://doi.org/10.31742/IJGPB.80.4.12 Soltis, D. E., Albert, V. A., Leebens-Mack, J., Bell, C. D., Paterson, A. H., Zheng, C., Sankoff, D., DePamphilis, C. W., Wall, P. K., & Soltis, P. S. (2009). Polyploidy and angiosperm diversification. American Journal of Botany, 96(1), 336–348. https://doi.org/10.3732/AJB.0800079 Soltis, P. S., & Soltis, D. E. (2000). The role of genetic and genomic attributes in the success of polyploids. Proceedings of the National Academy of Sciences, 97(13), 7051–7057. https://doi.org/10.1073/PNAS.97.13.7051 Stift, M., Kolář, F., & Meirmans, P. G. (2019). Structure is more robust than other clustering methods in simulated mixed-ploidy populations. Heredity 2019 123:4, 123(4), 429–441. https://doi.org/10.1038/s41437-019-0247-6 Stöcklin, J., Kuss, P., & Pluess, A. R. (2009). Genetic diversity, phenotypic variation and local adaptation in the alpine landscape: case studies with alpine plant species. Botanica Helvetica 2009 119:2, 119(2), 125–133. https://doi.org/10.1007/S00035-009-0065-1 Valdés, B. (1991). Andalucia and the Rif. Floristic links and common flora. Chronica Botanica , 10, 117–124. van de Peer, Y., Ashman, T. L., Soltis, P. S., & Soltis, D. E. (2021). Polyploidy: an evolutionary and ecological force in stressful times. The Plant Cell, 33(1), 11–26. https://doi.org/10.1093/PLCELL/KOAA015 van de Peer, Y., Mizrachi, E., & Marchal, K. (2017). The evolutionary significance of polyploidy. Nature Reviews Genetics 2017 18:7, 18(7), 411–424. https://doi.org/10.1038/nrg.2017.26 van Drunen, W. E., & Husband, B. C. (2019). Evolutionary associations between polyploidy, clonal reproduction, and perenniality in the angiosperms. New Phytologist, 224(3), 1266–1277. https://doi.org/10.1111/NPH.15999 van Rossum, F., de Sousa, S. C., & Triest, L. (2006a). Morph-specific differences in reproductive success in the distylous Primula veris in a context of habitat fragmentation. Acta Oecologica, 30(3), 426–433. https://doi.org/10.1016/J.ACTAO.2006.06.005 van Rossum, F., de Sousa, S. C., & Triest, L. (2006b). Morph-specific differences in reproductive success in the distylous Primula veris in a context of habitat fragmentation. Acta Oecologica, 30(3), 426–433. https://doi.org/10.1016/J.ACTAO.2006.06.005 van Rossum, F., & Triest, L. (2007). Fine-scale spatial genetic structure of the distylous Primula veris in fragmented habitats. Plant Biology, 9(3), 374–382. https://doi.org/10.1055/S-2006-924632/ID/60 Wedderburn, F., & Richards, A. J. (1990). Variation in within-morph incompatibility inhibition sites in heteromorphic Primula L. New Phytologist, 116(1), 149–162. https://doi.org/10.1111/J.1469- 8137.1990.TB00520.X Wickham, H. (2016). ggplot2. Elegant Graphics for Data Analysis. In ggplot2. Springer-Verlag. https://doi.org/10.1007/978-0-387-98141-3 Yuan, S., Barrett, S. C. H., Duan, T., Qian, X., Shi, M., & Zhang, D. (2017). Ecological correlates and genetic consequences of evolutionary transitions from distyly to homostyly. Annals of Botany, 120(5), 775–789. https://doi.org/10.1093/AOB/MCX098 RANGE-WIDE POPULATION GENETICS OF LINUM SUFFRUTICOSUM S. Máster Universitario en Biología Avanzada 2022 Maria Antònia Vanrell 19 Supplementary material Table S1. Ecological and biological variables corresponding to each of the analyzed populations. Abbreviations: POP, Population; LAT, Latitude; LONG, Longitude; 2x, diploid; 3x, triploid; 4x, tetraploid; 6x, hexaploid; 8x, octoploid; % L, Proportion of long-styled morph individuals in the population; Cluster (K2), assigned cluster by STRUCTURE assuming two populations; Cluster (K8), assigned cluster by STRUCTURE assuming eight populations. POP LAT LONG Location Ploidy Taxa % L Altitude (m.a.s.l) Population size Cluster (K2) Cluster (K8) FLAX2 36.516 -6.138 Cádiz, España 4x L. suffruticosum 0.5 33 50 1 1_1 FLAX5 37.194 -4.122 Granada, España 4x L. suffruticosum 0.492 765 200 1 1_1 FLAX9 37.198 -1.999 Almería, España 6x L. suffruticosum 0.444 597 500 1 1_2 FLAX13 36.846 -4.805 Málaga, España 4x L. suffruticosum 0.432 515 1000 1 1_2 FLAX24 37.127 -3.451 Granada, España 4x L. suffruticosum 0.574 1700 100 1 1_1 FLAX29 38.887 -3.053 Ciudad Real, España 4x L. suffruticosum 0.545 850 100 1 1_1 FLAX30 38.971 -2.896 Ciudad Real, España 4x L. suffruticosum 0.667 830 100 1 1_1 FLAX33 40.800 -3.613 Madrid, España 6x L. suffruticosum 0.473 805 300 1 1_6 FLAX36 38.349 -1.993 Albacete, España 4x L. suffruticosum 0.587 708 150 1 1_3 FLAX37 38.490 -2.412 Albacete, Spain 4x L. suffruticosum 0.438 956 500 1 1_6 FLAX39 42.381 0.791 Lérida, Spain 4x L. appressumsalsoloides 0.583 1078 100 1 1_2 FLAX45 42.353 1.842 Girona, Spain 2x-3x L. appressumsalsoloides 0.522 1259 5000 1 1_2 FLAX46 42.757 2.057 Aude, France 2x L. appressumsalsoloides 0.833 1020 20 1 1_2 FLAX53 44.523 3.307 Lozère, France 4x L. appressumsalsoloides 0.577 736 100 1 1_4 FLAX56 44.236 3.554 Lozère, France 2x L. appressumsalsoloides 0.44 989 25 1 1_4 FLAX59 43.992 5.242 Vaucluse, France 4x L. appressumsalsoloides 0.692 923 20 1 1_4 FLAX61 44.619 6.174 Hautes- France 2x L. appressumsalsoloides 0.538 1324 200 1 1_4 FLAX62 44.324 6.664 Alpes-de- Haute- Provence, France 4x L. appressumsalsoloides 0.617 1582 700 1 1_4 FLAX69 39.863 -2.135 Cuenca, Spain 4x NA 0.644 1090 500 1 1_1 FLAX71 40.573 -2.089 Cuenca, Spain 4x L. appressumsalsoloides 0.552 1132 100 1 1_6 FLAX73 41.263 -3.210 Guadalajara, Spain 2x L. appressumsalsoloides 0.524 1349 100 1 1_7 FLAX75 42.157 -2.109 La Rioja, Spain 8x Intermediate entity 0.517 842 100 1 1_6 FLAX77 42.504 -0.349 Huesca, Spain 2x L. suffruticosum 0.698 795 200 1 1_2 FLAX78 42.527 -0.556 Huesca, Spain 6x L. appressumsalsoloides 0.438 1044 20 1 1_6 FLAX79 41.038 -1.709 Zaragoza, Spain 6x NA 0.525 1025 1000 1 1_5 FLAX80 40.605 -0.327 Castellón, Spain 6x L. suffruticosum NA 850 100 1 1_5 FLAX81 41.787 -1.703 Zaragoza, Spain 6x L. appressumsalsoloides 0.541 730 300 1 1_5 AA100 34.785 -2.383 Berka, Morocco 4x L. Suffruticosum Morocco 0.514 723 3000 2 2 AA105 32.616 -4.799 Midelt, Morocco 4x L. Suffruticosum Morocco NA 1721 30 1 1_2 AA107 33.908 -4.031 Taza, Morocco 2x L. Suffruticosum Morocco 0.5 899 3000 1 1_2 AA111 34.876 -3.149 Nador, Morocco 6x L. Suffruticosum Morocco 0.629 471 80 2 2 AA112 34.901 -3.547 Driouch, Morocco 2x L. Suffruticosum Morocco 0.692 530 30 2 2 RANGE-WIDE POPULATION GENETICS OF LINUM SUFFRUTICOSUM S. Máster Universitario en Biología Avanzada 2022 Maria Antònia Vanrell 20 (A) (B) Figure S1. Delta-K plot, calculated based on Evanno et al. (2005) to estimate the optimal number of genetic clusters. (A) Delta-K was calculated from the output of the Bayesian analysis of the population structure (STRUCTURE software (Pritchard et al., 2000)), based on 6 SRRs and including the 32 populations involved in this study. (B) Delta-K was calculated from the output of the Bayesian analysis of the population structure (STRUCTURE software (Pritchard et al., 2000)), based on 6 SRRs and including 29 populations assignet to the Cluster 1 in the former analysis. Table S3. Moran's I spatial correlation test results. Abbreviations: Na, Number of alleles per population; Ne, Effective number of alleles per population; He, Expected heterozygosity; % L, Proportion of long-styled morph individuals in the population. Parameter Observed Expected Sd P-value Na 0.034 -0.032 0.054 0.219 Ne -0.01 -0.032 0.054 0.68 He -0.07 -0.032 0.054 0.481 Altitude (m.a.s.l.) 0.032 -0.032 0.053 0.22 Ploidy level -0.002 -0.032 0.054 0.58 Population size -0.061 -0.032 0.045 0.514 % L -0.083 -0.032 0.053 0.332 Table S2. Partial characterization of the 12 SSR loci identified in Linum suffruticosum by Olmedo et al. (unpublished), and their inclusion in this study based on their performance with M13 labelling in a preliminary test with 16 individuals from 16 populations. Primer sequences are not printed here for the sake of respecting publication authority. Locus Repeat motif Repeat length Amplicon size (bp) Included M13- labelling Ls_1145191 TGA 10 116 Yes 6-FAM Ls_1169143 TG 18 158 No Ls_1178187 TTC 14 189 No Ls_144692 TTC 23 243 No Ls_246481 CCA 11 103 No Ls_337128 AG 25 250 Yes PET Ls_37372 AC 16 195 Yes NED Ls_395648 AC 13 243 Yes 6-FAM Ls_421659 GGA 8 189 Yes PET Ls_807222 TTC 9 173 No Ls_889692 GAA 14 192 No Ls_9438 GAA 24 247 Yes VIC RANGE-WIDE POPULATION GENETICS OF LINUM SUFFRUTICOSUM S. Máster Universitario en Biología Avanzada 2022 Maria Antònia Vanrell 21 Table S4. Genetic diversity parameters’ means for each ploidy level (A), and for each genetic cluster (B). Abbreviations: Na, Number of alleles per population; Ne, Effective number of alleles per population; He, Expected heterozygosity. (A) Ploidy level Mean Na Mean Ne Mean He 2 3.979 2.65 0.53 4 5.521 2.907 0.577 6 6.214 3.662 0.648 8 7.333 4.266 0.689 (B) Genetic cluster 1_1 4.056 2.668 0.561 1_2 4.167 2.627 0.588 1_3 2.833 2.026 0.459 1_4 4.5 3.037 0.556 1_5 6.222 3.312 0.55 1_6 6.967 4.228 0.671 1_7 3.833 2.934 0.508 2 4.889 3.12 0.623 Table S5. Tukey HSD post hoc test results. Comparison of the differences in the Number of alleles (i), Effective number of alleles (ii) and Expected Heterozygosity (iii), between ploidy level pairs (A) and genetic cluster pairs (B). (A) Ploidy level pairs Difference Lower Upper P adjusted (i) Number of alleles 4 – 2 0.541 -0.834 1.917 0.707 6 – 2 2.235 0.590 3.880 0.004 * 8 – 2 3.354 -0.017 6.725 0.051 . 6 - 4 1.693 0.253 3.133 0.016 * 8 - 4 2.812 -0.463 6.088 0.112 8 - 6 1.119 -2.278 4.516 0.805 (ii) Effective number of alleles 4 – 2 0.256 -0.479 0.992 0.776 6 – 2 1.011 0.131 1.891 0.019 * 8 – 2 1.616 -0.186 3.419 0.091 6 - 4 0.754 -0.015 1.524 0.056 . 8 - 4 1.359 -0.392 3.111 0.171 8 - 6 0.604 -1.212 2.421 0.800 (iii) Expected heterozygosity 4 – 2 0.046 -0.051 0.144 0.570 6 – 2 0.118 0.001 0.235 0.046 . 8 – 2 0.159 -0.080 0.399 0.288 6 - 4 0.071 -0.030 0.174 0.246 8 - 4 0.112 -0.120 0.345 0.557 8 - 6 0.041 -0.201 0.282 0.966 RANGE-WIDE POPULATION GENETICS OF LINUM SUFFRUTICOSUM S. Máster Universitario en Biología Avanzada 2022 Maria Antònia Vanrell 22 (B) Genetic clusters pairs Difference Lower Upper P adjusted (i) Number of alleles 1_2-1_1 0.111 -1.665 1.887 1.000 1_3-1_1 -1.222 -4.773 2.329 0.941 1_4-1_1 0.444 -1.546 2.435 0.995 1_5-1_1 2.167 -0.158 4.491 0.080 1_6-1_1 2.911 0.920 4.902 0.001 * 1_7-1_1 -0.222 -3.773 3.329 1.000 2-1_1 0.833 -1.491 3.158 0.928 1_3-1_2 -1.333 -4.821 2.154 0.902 1_4-1_2 0.333 -1.541 2.208 0.999 1_5-1_2 2.056 -0.170 4.281 0.085 1_6-1_2 2.800 0.926 4.674 0.001 * 1_7-1_2 -0.333 -3.821 3.154 1.000 2-1_2 0.722 -1.504 2.948 0.956 1_4-1_3 1.667 -1.935 5.268 0.783 1_5-1_3 3.389 -0.408 7.185 0.105 1_6-1_3 4.133 0.532 7.735 0.017 * 1_7-1_3 1.000 -3.650 5.650 0.996 2-1_3 2.056 -1.741 5.852 0.630 1_5-1_4 1.722 -0.679 4.123 0.297 1_6-1_4 2.467 0.387 4.546 0.012 * 1_7-1_4 -0.667 -4.268 2.935 0.998 2-1_4 0.389 -2.012 2.790 0.999 1_6-1_5 0.744 -1.657 3.146 0.965 1_7-1_5 -2.389 -6.185 1.408 0.452 2-1_5 -1.333 -4.018 1.351 0.720 1_7-1_6 -3.133 -6.735 0.468 0.121 2-1_6 -2.078 -4.479 0.323 0.125 2-1_7 1.056 -2.741 4.852 0.981 (ii) Effective number of alleles 1_2-1_1 -0.040 -0.936 0.856 1.000 1_3-1_1 -0.642 -2.434 1.150 0.928 1_4-1_1 0.369 -0.635 1.374 0.919 1_5-1_1 0.644 -0.529 1.818 0.614 1_6-1_1 1.560 0.556 2.565 0.001 * 1_7-1_1 0.266 -1.526 2.058 1.000 2-1_1 0.453 -0.720 1.626 0.898 1_3-1_2 -0.602 -2.362 1.158 0.943 1_4-1_2 0.410 -0.536 1.356 0.832 1_5-1_2 0.685 -0.439 1.808 0.491 1_6-1_2 1.601 0.655 2.547 0.001 * 1_7-1_2 0.306 -1.454 2.066 0.999 2-1_2 0.493 -0.630 1.617 0.823 1_4-1_3 1.012 -0.806 2.829 0.599 1_5-1_3 1.287 -0.629 3.203 0.373 1_6-1_3 2.203 0.385 4.020 0.010 * 1_7-1_3 0.908 -1.438 3.255 0.897 2-1_3 1.095 -0.821 3.011 0.568 1_5-1_4 0.275 -0.937 1.487 0.994 1_6-1_4 1.191 0.142 2.241 0.018 * 1_7-1_4 -0.103 -1.921 1.714 1.000 2-1_4 0.084 -1.128 1.295 1.000 1_6-1_5 0.916 -0.296 2.128 0.241 1_7-1_5 -0.379 -2.294 1.537 0.997 2-1_5 -0.191 -1.546 1.163 1.000 1_7-1_6 -1.295 -3.112 0.523 0.305 2-1_6 -1.107 -2.319 0.104 0.091 2-1_7 0.187 -1.729 2.103 1.000 RANGE-WIDE POPULATION GENETICS OF LINUM SUFFRUTICOSUM S. Máster Universitario en Biología Avanzada 2022 Maria Antònia Vanrell 23 Table S6. Results of the linear regressions testing differences in genetic diversity parameters among ecological and biological variables: Proportion of long-styled morph individuals in the population (i), Population size (ii), Latitude (iii) and Altitude (m.a.s.l.) (iiii). Abbreviations: Na, Number of alleles per population; Ne, Effective number of alleles per population; He, Expected heterozygosity. Genetic diversity parameter Adjusted R-squared P-value (i) % L-morph individuals Na 0.019 0.212 Ne -0.002 0.337 He 0.023 0.195 (ii) Population size Na -0.033 0.939 Ne -0.006 0.377 He -0.033 0.956 (iii) Latitude Na 0.057 0.098 Ne 0.067 0.081 He 0.06 0.094 (iiii) Altitude (m.a.s.l.) Na -0.012 0.443 Ne -0.024 0.604 He -0.014 0.463