Style polymorphism in Linum (Linaceae): a case of Mediterranean parallel evolution?

1 Style polymorphism in Linum (Linaceae): a case of Mediterranean parallel 1 evolution? 2 3 Ruiz-Martín, J.1; Santos-Gally, R.1,2; Escudero, M1., Midgley, J.J.3; Pérez-Barrales, R1, 4.; 4 Arroyo, J.1 5 6 Addresses 7 1. Departamento de Biología Vegetal y Ecología, Universidad de Sevilla, Spain; 2. current8 address: CONACYT-Departamento de Ecología Evolutiva, Instituto de Ecología, UNAM, 9 México; 3. Department of Biological Sciences, University of Cape Town, South Africa; 4. 10 current address: School of Biological Sciences, University of Portsmouth, UK. 11 12 Running title: convergence of heterostyly in Linum 13 14 Author for correspondence: 15 J. Arroyo16 [email protected] 17 tel +34 954 557 058 fax +34 954 557 059 18 19 Keywords: Cape Floristic Region, divergence time, heterostyly, life-history, phylogeny, 20 pollination, polyploidy. 21 22 23 24 25 26 Page 1 of 54 • • • • Heterostyly is a sex polymorphism which has challenged evolutionary biologists ever since Darwin. One of the lineages where heterostyly, and related stylar conditions, appears more frequently is the family Linaceae and its most diverse and widespread genus, Linum. Thus, this group is particularly suitable for testing competing hypotheses about ancestral and transitional stages on the evolutionary building up of heterostyly. We generated a well-resolved phylogeny of Linum based on extensive sampling and plastid and nuclear DNA sequences, and used it to trace the evolution of character states of style polymorphism and its association with traits related to pollination and breeding systems, obtained from our samples and the literature. Our results supported former phylogenetic hypotheses: the paraphyly of Linum and the non-monophyly of current taxonomic sections. Heterostyly was common in the genus, but appeared concentrated in the Mediterranean basin and, to a lesser extent, in the South African Cape. Ancestral character state reconstruction failed to determine a unique state as the most probable condition for style polymorphism in the genus. In contrast, approach herkogamy was resolved as ancestral state in some clades, in agreement with recent hypotheses on the evolution of heterostyly. Some traits putatively related with heterostyly, such as life-history and polyploidy, did show marginal or non significant phylogenetic correlation respectively. Although pollinator data are limited, the available evidence suggests that beeflies are associated with specific cases of heterostyly. The consistent association between style polymorphism and heteromorphic incompatibility points out to ecological factors as drivers of the multiple evolution of style-polymorphism in Linum. Albeit based on limited evidence, we hypothesized that specialized pollinators and lack of mating opportunities drive evolution of style polymorphism and loss of the polymorphism, respectively. 2 27 Abstract 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 Introduction 56 Page 2 of 54 3 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 Page 3 of 54 The great variation of flowers across lineages has inspired modern plant classification since Linneaus (1735), as well as the formulation of hypotheses about the causes of extreme angiosperm diversification, otherwise known as the Darwin’s abominable mystery (Grant & Grant 1965; Stebbins 1970, 1974; see Friedman 2009 and references therein for an historical account of Darwin’s views). This floral variation also occurs within species and populations, can be continuous or discontinuous, and often appears associated with geographical variation, which has been important to bring insights on the biotic and abiotic causes of such variation (Herrera et al., 2006; Strauss & Whittall 2006, Gómez et al. 2009). Discontinuous variation at the population level, that is, presence of discrete and modal phenotypes, has been interpreted in the context of population divergence through disruptive selection (Ortiz et al. 2015). However, discontinuous variation sometimes results from negative frequency dependent selection, as the fitness of one phenotype strongly depends on the abundance of alternative phenotypes. At equilibrium, it is expected to find all phenotypes at the same proportion in the population. Discontinuous variation is better understood when accompanied by gender differentiation. With negative frequency selection, the success of the uncommon gender is larger than the common gender, as mate availability for the latter is lower (McCauley & Taylor 1997; Dufay et al. 2009). A similar situation can be achieved without gender differentiation (Pannell et al. 2005). Such is the case of reciprocal style polymorphisms, present in some hermaphroditic plants, where floral morphs display styles and stamens in a reciprocal position (Fig. 1), in a way that pollination and mating occurs more often between morphs rather than within morphs, maintaining the frequency of morphs at balance (Barrett 2002). The most common style polymorphism is heterostyly (Barrett & Shore 2008), for which flowers in populations present two (distyly) or three (tristyly) morphs. This polymorphism called the attention of evolutionists ever since Darwin (1877), and early geneticists, who soon discovered its apparently simple genetic basis (Bateson & Gregory 1905). Yet, in those early times, it was recognized that most heterostylous species showed the so-called heteromorphic incompatibility system (only crosses between different morphs are compatible, whereas self-fertilization and within-morph cross- 4 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 Page 4 of 54 fertilization is impeded, Darwin 1877, Dulberger 1992). During most of 20th century, heterostyly was used as model system to study the evolution of inbreeding avoidance. Specifically, most of the studies interpreted the evolutionary pathways of heterostyly following the proposals of Mather & de Winton (1941), with important modifications by Baker (1966), ultimately leading to the quantitative model of Charlesworth & Charlesworth (1979). In short, these models predict that reciprocal style polymorphism evolved after the appearance of the incompatibility system, with an ancestral state of non-herkogamous (homostylous) flowers showing high selfing rates and inbreeding depression. These models were challenged by that of Lloyd & Webb (1992 a, b), who suggested that the main driving force for the establishment of the polymorphism was the promotion of compatible cross pollination and the decrease pollen discount (enhanced male fitness, as Darwin himself proposed in 1877). The latter model presumed (i) an independent evolution of sex organ reciprocity and an heteromorphic incompatibility system, and (ii) an ancestral condition of an outcrosser with approach herkogamous flowers (i.e., with the stigma protruding the anthers). This model strongly emphasized the ecological context of pollination: specialized pollinators select for and maintain the style morphs if they are able to place pollen grains on different parts of the body, and legitimately deliver pollen to the opposite stigmas, with minimal pollen loss. The model of Lloyd & Webb (1992 a, b) has progressively gained more support from both microand macroevolutionary studies. Microevolutionary analyses have mostly examined the relative rates of pollination and mating between and within morphs in populations (Lau & Bosque 2003, reviewed in Costa 2017). In contrast, macroevolutionary models to study how the heterostylous floral syndrome evolved have been relatively scarce compared to population level studies. To this respect, macroevolutionary studies in some plant groups, such as Narcissus, Lithodora and related genera, Pontederiaceae, Exochaenium, Amsinckia, or Primula (Kohn et al. 1996; Schoen et al. 1997; Guggisberg et al. 2006; Pérez-Barrales et al. 2006; Ferrero et al. 2009; Kissling & Barrett 2013; Santos-Gally et al. 2013) have provided strong support to Lloyd & Webb’s (1992 a, b) ideas. Given that heterostyly is well represented both among lineages of Angiosperms (28 families across many orders in both monocots and dicots; Barrett & Shore 2008) and biomes, these studies offer good opportunities to explore the 5 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 Page 5 of 54 ecological and biogeographical correlates of heterostyly in order to infer the conditions that favour this polymorphism to arise and be maintained. For example, heterostyly should be common in plants with specialised pollination, or should be disadvantageous where outcrossing is at risk, as expected when pollinators are scarce, or in highly disturbed environments (Piper et al. 1986). Likewise, it would be unlikely to find heterostyly associated with hybridization and polyploidy (both associated with selffertilization as by-product, Ramsey & Schemske 1998), or with short-lived plants, particularly in annuals, as these typically present higher selfing rates and occur more frequently in disturbed places compared to perennial plants (Barrett 2002). Heterostyly in Linaceae was first reported in the seminal works of Darwin (1864, 1877) and Hildebrand (1864). In particular, Darwin’s experimental and observational work on Linum grandiflorum and L. perenne was influential in determining the function of the polymorphism. Later, it was suggested that other genera in the family could include distylous and tristylous species (Lloyd et al. 1990; Thompson et al. 1996). After Darwin´s work, geneticists used species of Linum to study the inheritance of heterostyly, and showed that style polymorphism and heteromorphic incompatibility appear linked (Lewis 1943; Dulberger 1992; Lewis & Jones 1992; Ushijima et al. 2012). Furthermore, the stability of heterostyly as a trait has been valuable for taxonomists, who used it as a binary character ("heterostylous" vs "homostylous") in identification keys and diagnoses (e.g., Ockendon & Walters 1968; Ockendon 1971; Martínez-Labarga & MuñozGarmendia 2015; Ruiz-Martín et al. 2015). Thus, taxonomic descriptions have been valuable to characterize species and conduct evolutionary reconstructions of the trait (McDill et al. 2009). However, Linum is a highly diverse genus with a wide geographic distribution, in which the diversity of stylar conditions is much greater than previously reported (Ruiz-Martín, unpublished data; Darwin 1877; Heitz 1980; Armbruster et al. 2006). Most of the taxonomic diversity appears in the Mediterranean and, surprisingly, the morphological variation on the types of polymorphism and other associated traits remains to be explored. Thus, Linum represents an excellent study system for testing macroevolutionary hypotheses and correlates with heterostyly. The specific aims of our study were: (1) to generate an updated phylogeny of Linum, including lineages and infrageneric taxa recognized in taxonomic studies, (2) to 6 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 Page 6 of 54 estimate divergence times in order to date events of evolutionary significance for the polymorphism, (3) to reconstruct ancestral states for stylar condition and other related traits, (4) to estimate the significance of correlated evolution between style polymorphism and those other traits across the phylogeny, and (5) to integrate all these results in a geographical and ecological context , in order to infer the conditions under which heterostyly most likely evolved. Ultimately, we wished to validate current evolutionary models of heterostyly. Material and methods Floral measurements and categorization Previous work reported that style polymorphism in Linum concentrates mostly in the Mediterranean basin and South Africa (McDill et al. 2009). Thus, we concentrated our field sampling efforts in these regions (although other regions were also explored), and also extracted information from published sources. We collected up to 100 flowers from 50 populations from 50 taxa of Linum (Table S1), and preserved flowers in 70% ethanol for morphological measurement in the laboratory. Linum flowers have five styles and five stamens, reaching each of five similar heights (we conducted a pilot study to assess within flower variation in the position of anthers and stigmas, and found that variation within flower was nearly negligible, results not shown). Anther and stigma heights were measured as the distance from base of the ovary to the top of the organ. All measurements were taken from digital images of the lateral view of flowers with petals removed, using ImageJ (Rasband 2008). Images were previously taken using a stereomicroscope (Zeiss Stemi-2000) with attached digital camera (Zeiss Axiocam). Data for the remaining Linum species and outgroups were collected from the literature (see Table S1 in Suppl. Material). We classified flowers of style polymorphic species as L-morph when the stigmas were positioned above the anther whorl, and S-morph when the stigmas were below the anther whorl. Style polymorphism includes two morphs (distyly and stigma height dimorphism) or three morphs (tristyly and stigma height trimorphism); and here we 7 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 Page 7 of 54 refer to stigma height polymorphism as the discrete variation in stigma height but not in anther height, a condition related with heterostyly (Barrett et al. 2000). Species with populations with only one floral morph were named monomorphic and classified as follows: homostylous (no apparent separation between sexual organs), approach or reverse herkogamous (stigmas placed above or below the anther whorl respectively), and horizontal herkogamous (anther-stigma separation along the horizontal plane of the flower). This classification was based on extensive flower measurements and the frequency distribution of sex organ heights among population (Ruiz-Martín, unpublished data). It is important to highlight that most of taxonomic references classify style polymorphism as heterostylous (sometimes discriminating distyly from tristyly) or homostylous; the latter referring to any style monomorphic condition, regardless the relative position of anthers and stigmas (see description above). This distinction is critical for testing models of evolution of heterostyly in relation to the ancestral stylar condition (true non-herkogamous homostyly in Charlesworth & Charlesworth 1979 vs. approach herkogamy in Lloyd & Webb 1992a). Hence, the species that could not be sampled in the field were we characterised using the quantitative information provided in taxonomic descriptions (e.g. approach or reverse herkogamous when no overlap was reported between stamen and style length, otherwise homostylous). We included other biological traits of species putatively related with style polymorphism, and gathered information from the literature on life-history, chromosome number, breeding system, pollinators, ancillary traits (polymorphism in size and form of pollen grains and/or stigma papillae) and genetic control of polymorphism (see Table S1 Suppl. Material, for references). Given the lack of a comprehensive monograph for species identification on Linum, we followed the most recent and comprehensive taxonomic treatment for regions with high species diversity in the genus: Yusepchuk (1949), Davis (1967), Ockendon & Walters (1968), Rogers (1981), Greuter et al. (1984), Yilmaz & Kaynak (2008) and McDill et al. (2009). Phylogeny and divergence times 8 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 Page 8 of 54 Sampling. 103 samples from 93 species or subspecies of Linum were included as ingroup, representing the five taxonomic sections. Two or three samples from different localities were included for nine Linum species with taxonomical doubts to test for monophyly. In addition, samples from eight species representing closely related genera (Anisadenia, Cliococca, Hesperolinon, Hugonia, Radiola, Reinwardtia, Sclerolinon and Tirpitzia, McDill et al. 2009) were included to evaluate if Linum is a monophyletic genus. Three species from closely related families (Hypericum perforatum from Hypericaceae, Viola pubescens from Violaceae, and Humiria balsamifera from Humiriaceae) were also included as outgroup (Table S1). Fifty-five leave samples from 48 species or subespecies of Linum were collected in field trips (vouchers stored at SEV herbarium; Table S1), whereas leaves from additional 18 taxa were obtained from herbaria collections (SEV, MA and E, Table S1). The DNA sequences from the remaining 29 species of Linum, eight of Linaceae and three from other families were directly downloaded from GenBank data base and previously published (see Table S1 for species and references). Two taxa were sampled in the field and obtained from herbaria. DNA extraction, PCR and sequencing. Total genomic DNA was extracted using DNEasy Plant Minikit (QIAGEN Inc., BIO Laboratories Inc., Carlsbad, CA, USA). One nuclear DNA region, ITS (internal transcribed spacer), and three plastid DNA regions, NADH dehydrogenase subunit F (ndhF) gene, maturase K (matK) gene and trnL-F spacer were amplified, purified and sequenced. PCR amplification was performed following McDill et al. (2009), with minor modifications. Products were purified using ExoSAP-IT (USB, Cleveland, Ohio, USA). Sequencing reactions were performed using the ABI BigDye® Terminator v3.1 Cycle Sequencing Kit (Thermo Fisher Scientific Inc., Massachusetts, U.S.A.) in Macrogene Europe Laboratory (Amsterdam, The Netherlands). Phylogenetic analyses. Sequences from the four DNA regions were aligned separately using MaffT 6.0 FFT-NS-I (Katoh & Toh 2008) as implemented in Geneious Pro™ 5.3 (Kearse et al. 2012). The resulting alignments were manually revised. Putative homoplasic regions were detected and removed from the alignments using GBlock v0.91b (Castresana 2000). Incongruence between DNA regions was discarded and the four DNA regions were combined in a single matrix (2,900 bp). 9 Bayesian inference analysis was performed using Markov chain Monte Carlo 241 (MCMC) as implemented in MrBayes3.0b4 (Huelsenbeck & Ronquist 2001). The best242 fitted model of DNA evolution for each DNA region was selected from the analysis in 243 ModelTest 3.06 (Posada & Crandall 1998). GTR +G +I was selected for ndhF and matK 244 regions and GTR +G for trnL-F and ITS regions. To avoid overparameterization, we 245 combined the three plastid regions in a matrix and analyzed it together using GTR +G +I 246 model. Two independent analyses of four Metropolis-coupled Markov chains were run 247 for 10 million generations. After a burn-in of 25%, the remaining trees (15,000) were 248 used to construct a majority-rule consensus tree using posterior probability values as a 249 measure of clade support. Phylogenetic analyses were performed using CIPRES Science 250 Gateway V. 3.3 portal (Miller et al. 2010). 251 Analyses of divergence times. The four DNA regions were combined in a single 252 partition (using GTR+G+I as DNA model of evolution). Analyses were conducted using 253 three independent MCMC runs of 120 million generations each, using Yule process as 254 tree model and relaxed clock log normal as clock model, as implemented in BEAST 255 v1.4.8 (Drummond & Rambaut 2007). Run convergence and burn-in were assessed in 256 Tracer 1.6 (Rambaut & Drummond 2007). Trees from the three independent runs were 257 combined using LogCombiner 1.4.8 (10% of burn-in). Maximum clade credibility trees 258 were calculated with TreeAnnotator 2.3.2 using a posterior probability limit of 0.95, 259 maximum clade credibility tree and the mean heights options. 260 Two calibration points were used: 1) a secondary calibration base on the age of 261 the stem node of Linaceae which is the Malpighiales crown node (Bell et al. 2010). 262 Specifically, a normal distribution with a mean of 93.5 Ma (95% CI 88-97 Ma) was used 263 as recommended for secondary calibrations. And, 2) a log-normal distribution with 264 mean = 0, standard deviation = 1.0 and zero offset = 33.9 for the crown node of genus 265 Linum (which includes genera Cliococca, Hesperolinon, Radiola and Sclerolinon). This last 266 calibration point accounts for the oldest Linum fossil. This is a pollen grain from Ebro 267 River Basin (33.9-37.2 Ma, Late Eocene, Cavagnetto & Anadón 1996). Analyses of times 268 of divergence were performed using CIPRES Science Gateway V. 3.3 portal (Miller et al. 269 2010) and the cluster located in Andalusian Scientific Information Technology Center 270 (CICA, Seville, Spain). 271 Page 9 of 54 16 trigynum clade; only marginally significant). Also within Clade B2 a transition from 456 homostyly or from polymorphic state to approach herkogamy was inferred (see L. 457 volkensii). 458 459 Trait correlations. There was marginal support for the correlation between presence 460 of stylar polymorphism and perennial life-history of species. Our results indicated that a 461 dependent model of evolution between life history and stylar polymorphism provided a 462 marginally significant better fit to the data than an independent model (difference 463 between likelihood–ratio = 9.136, p=0.057). For the set of 50 species where we were 464 able to obtain data on chromosome number, there was no significant correlation 465 between presence of stylar polymorphism and polyploidy (difference between 466 likelihood-ratio= 3.646, p= 0.456). 467 468 Discussion 469 470 Linaceae is a family that includes some of the largest morphological diversity of style 471 polymorphisms, with homostyly and different types of herkogamy, stigma-height 472 dimorphism and trimorphism, distyly, and tristyly, and Linum seems to display most of 473 this diversity. This allows testing evolutionary models for those traits where specific 474 transitions are predicted, as proposed by Charlesworth & Charlesworth (1979) and 475 Lloyd & Webb (1992a). Particularly, Lloyd & Webb’s (1992a) model challenged the 476 formerly prevalent ideas represented by Charlesworth & Charlesworth (1979), and 477 proposed an alternative ancestral condition (approach herkogamy, instead of 478 homostyly) to heterostyly. Interestingly, Hugonia within Linaceae was one of the study 479 cases that inspired the new model (Lloyd et al. 1990), which was later confirmed as 480 tristylous (Thompson et al. 1996; Meeus et al. 2011). Although the variation in Linum 481 inspired Darwin to interpret the adaptive significance of heterostyly (Darwin 1877), it is 482 surprising that the variation of stylar conditions in the genus has rarely been explored 483 (but see Armbruster et al. 2006 and McDill et al. 2009). In our study, we wished to 484 validate current evolutionary models, for which we generated an updated phylogeny, 485 Page 16 of 54 17 incorporated the wide variety of stylar conditions, and explored trait correlates to throw 486 light on the plausibility of the alternative models. As discussed below, our results failed 487 to ascertain clearly the ancestral condition in the genus, which precluded supporting 488 any of the competing models, with the exception perhaps of the South African clade, 489 which supported the Darwinian model of Lloyd & Webb (1992a). The information that 490 we gathered in addition to the stylar condition was limited, and precluded statistical 491 analyses to incorporate the evolutionary significance of breeding systems, pollination 492 biology and biogeography of species for this purpose. However, life-history and 493 polyploidy provided plausible explanations for the presence of style polymorphism. Our 494 main result is that, with the data available, both models could explain parts of the 495 evolution of heterostyly in Linum. 496 497 Phylogeny, divergence times and geographic ranges. We confirmed taxonomic 498 aspects that deserve further work (e.g., the inclusion of four Linaceae genera resulted in 499 the paraphyly of Linum, and the non-monophyly of some sections, see McDill et al. 2009 500 and McDill & Simpson 2011). Despite our sampling efforts almost duplicated sampling in 501 previous systematic work (McDill et al. 2009) and included a larger proportion of Linum 502 species, and that some of the DNA regions used were different, we obtained similar 503 results to those previously reported by McDill et al. (2009) and McDill & Simpson 504 (2011), making the phylogeny reported here more plausible and valuable for testing 505 evolutionary hypotheses. 506 In our study we found that, unlike species from other geographic regions, the 507 South African species, which all belong to the sect. Linopsis, formed a well-supported 508 monophyletic clade. In addition, the South African clade turned to be closely related to 509 the American clades, rather than the Euroasiatic clades from the same section. This 510 result has important implications for evolutionary interpretations because none of the 511 surveyed American Linum species present stylar polymorphisms, while species in sect. 512 Linopsis in Eurasia do. In our analyses, we were interested to estimate the sequence of 513 divergence dates leading to clades present in the Mediterranean Basin and South Africa, 514 the latter being the only region with style polymorphic Linum species outside the 515 Mediterranean basin. Thus, it is remarkable that the South African clade separated from 516 Page 17 of 54 18 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 Page 18 of 54 its monomorphic sister American clade in the late Miocene, about 9 MYA. In contrast, its closest Mediterranean clade, which includes members of sect. Linopsis and sect. Syllinum (with mostly western and eastern Mediterranean species respectively), diverged much earlier (in middle Miocene, more than 14 MYA). Unlike the American clade, Mediterranean clades include many style polymorphic species. By the time the clades split, continents were already separated, particularly Africa and the Americas. Thus, episodes of long distance dispersal should be invoked or, alternatively, massive extinctions of connecting clades in Africa, which would not have left a living or fossil trace. These episodes are coincident with last Antarctic glaciation and sharp decrease in temperature in southern Africa (Linder 2005). Regardless the specific events, it is remarkable that the American clades did not include any style polymorphic lineage. A proper biogeographical analysis incorporating explicit palaeogeographic settings would be necessary to ascertain the most likely scenario. Evolution of style polymorphism in Linum (models test). Previous work in Linum (McDill et al. 2009) provided a plausible reconstruction of pathways of heterostyly and “homostyly” (including all types of monomorphic conditions). Despite differences in sampling and molecular markers, our findings were similar to those previously reported (Fig. 3). Specifically, we were unable to determine the most likely ancestral stylar condition in the genus, which could be either style polymorphic and monomorphic (our terms). The variability of stylar conditions in Linaceae and in Linum (Ganders 1979; Lloyd et al. 1990; Thompson et al. 1996; Suksathan & Larsen 2006; McDill & Simpson 2011) combined with the inferred high transition rates among character states, and long-branches arising from the root of the phylogeny may explain this lack of resolution. An analysis at the family level would probably throw more light and allow better resolution of the ancestral condition. Despite lack of resolution at the basal stage, we detected several events of independent evolution of the polymorphism along the evolutionary history of Linum. Although some clades are integrated by mostly monomorphic or polymorphic species, any of these conditions appears secondarily lost, even in pairs of sister species. For example, loss of polymorphism was detected in L. seljukorum-L. pubescens, L. leoni-L. punctatum, L. lewisii-L pallescens, L. tenuifolum-L. 19 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 Page 19 of 54 suffruticossm, L. corymbulosum-L. trigynum. In addition, polymorphic species evolved in most of monomorphic clades, as shown by the species pairs L. grandiflorum-L. decumbens, L. comptoni-L. pungens; L. heterostylum-L. esterhuysenae. Particularly dynamic in evolutionary grounds was clade B2 (Fig. 3), especially most of the Western Mediterranean subclade, including species from L. virgatum to L. setaceum. This clade includes L. suffruticosum s.l., (López-González 1979; Martínez-Labarga & MuñozGarmendia 2015) with a special case of three-dimensional reciprocity (Armbruster et al. 2006), L. tenue, a polyphyletic species with substantial morphological variation in NW Africa (J. Arroyo and J. Ruiz-Martín, pers. observ.), as well as a recently named new distylous species, L. flos-carmini (Ruiz-Martín et al. 2015), different from its sister species, the homostylous L. setaceum. All this variation clearly reflects that further work is required in these taxa and geographic range. Perhaps one of the most remarkable outcomes is the independent evolution of heterostyly in two South African species within a clade integrated by 14 species. In his taxonomic review, Rogers (1981) suggested that heterostyly appeared in South Africa independently from its occurrence in the Mediterranean basin and nearby regions, which was later supported by McDill et al. (2009), and here we confirmed. Although limited, our population sampling allowed us to confirm the presence of distyly in L. comptonii and L. heterostylum. Because the South African Linum clade is monophyletic and closely related to the monomorphic clade of American Linum species, the independent evolution of the polymorphism is thus fully supported. Unlike American species, all South African Linum species, except L. thurnbergi, are restricted to Mediterranean type climate of the Cape Floristic Region (Rogers 1981). Thus, the presence of style polymorphism restricted to Mediterranean climates (the Cape and the Mediterranean basin) points out to an apparent case of parallel evolution linked directly or indirectly to climate. In other Mediterranean climate regions of the world the number of Linum species is much lower. The characterisation of monomorphism as homostyly and different types of herkogamy (Fig. 4) depicted a complex picture with regards the evolutionary reconstruction of pathways, but allowed us to explicitly test competing hypotheses of ancestral stylar state. Whereas the ancestral state at the genus level was unresolved, the 20 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 Page 20 of 54 only clade within Linum with certainty in the ancestral condition was the South African clade. Here, the Lloyd & Webb (1992a) model was fully supported, with approach herkogamy as ancestral condition. Interestingly, approach herkogamy is widespread in this clade. In contrast, approach herkogamy is uncommon in other clades (e.g. L. hologynum, L. lewisii, L. and volkensii) whereas homostyly appears frequently. This homostyly is secondary, derived from a polymorphic condition, and probably associated with shifts towards selfing to increase reproductive assurance (see for instance L. corymbulosum and L. trigynum, or L. leonii). Such shifts have been reported in other style polymorphic groups (Schoen et al. 1997; Guggisberg et al. 2006; Mast et al. 2006; PérezBarrales et al. 2006; Kissling & Barrett 2013; Santos-Gally et al. 2013). More detailed information on the breeding system of the species would confirm this hypothesis. Other stylar conditions are scarcer. Reverse herkogamy, a necessary phenotype in an intermediate step for the establishment of style polymorphism in any model, was detected in the Mediterranean L. nodiflorum and the two South American sister species L. littorale and L. prostratum. Surprisingly, reverse herkogamy appeared in these species as derived monomorphic condition. This transition has been reported in Exochaenium in the Gentianaceae (Kissling & Barrett 2013), although it remains unclear the mechanisms that favours the selection of monomorphic reverse herkogamy. Horizontal monomorphic herkogamy was detected in two Linum species, L. kingii and L. tenuifolium, and in two closely related genera, Hesperolinum and Radiola, which are placed within Linum. This condition might result from selection to avoid selfpollination, as in the self-compatible L. tenuifolium (Nicholls 1986) (see Fig. 1). Finally, it was not possible to include an evolutionary reconstruction of stigma height dimorphism, as it is an unusual condition in Linum, only present in L. grandiflorum and perhaps L. perenne (Heitz 1980). This condition has been reported as an intermediate and unstable state towards heterostyly (Lloyd & Webb 1992b, but see Barrett & Harder 2005), which is consistent with its unclear ancestral/derived condition. This evolutionary lability has been reported for stigma-height dimorphism in some Boraginaceae (Ferrero et al. 2009). 21 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 Page 21 of 54 Correlated evolution and trait associations. Few studies have attempted to investigate correlations between style polymorphisms and other traits in an explicit phylogenetic context, and these have focused on associations with other floral traits (e.g. corolla size and form: Santos-Gally et al. 2013; Kissling & Barrett 2013). In our study, we were interested to investigate the association between style polymorphism and life history (annual vs perennial). This association is expected (Dulberger 1992) because pollination of style polymorphic plants is often specialized (Darwin 1877; Lloyd & Webb 1992a; Lau & Bosque 2003), and short-lived plants, especially annuals, are more sensitive to loss of these pollinators or pollinator uncertainty, and shifts to selfing are more likely to occur. Our results showed that style polymorphism occurs more frequently among perennial than annual species, although the association was only marginally significant. However, we only gathered data for a subset of species, and data on breeding systems from more species would be particularly valuable here. Despite the limitations, this result suggests that reproductive assurance is probably important in annual species, and most likely plays a role against maintaining style polymorphism. An important trait associated with breeding system and thus with style polymorphism is polyploidy. The available evidence shows variation in the correlation between heterostyly and polyploidy, ranging from lack of association to heterostyly being frequent among diploids (Naiki 2012). Across families, a phylogenetic account of these studies suggests that this may stand only for Rubiaceae and Primulaceae (Naiki 2012). At least for Primula, it has been demonstrated that heterostyly is not present among allopolyploid taxa (Guggisberg et al. 2006), which has been also suggested for Turnera (Shore et al. 2006). This is in agreement with the mechanism of breakdown of heterostylous supergenes by recombination linked to hybridization (Lewis & Jones 1992). Although hybridization between some Linum species has been reported, the species involved displayed similar chromosome numbers (Seetharam 1972; Muravenko et al. 2003; Yurkevich et al. 2013), which does not promote breakdown of heterostyly. We were unable to detect a significant correlation between polyploidy and heterostyly in our data set of 50 species of Linum. It could be possible that our data includes mostly polyploidy series of autopolyploids. This is well illustrated by the closely related L. tenuifolium and L. suffruticosum. Linum tenuifolium is monomorphic, self-compatible and 22 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 Page 22 of 54 diploid across its wide range in Europe and western Asia (Nicholls 1986). In contrast, L. suffruticosum, with three-dimensional reciprocity (Fig. 1, Armbruster et al. 2006), displays a polyploid series from diploidy to decaploidy (Nicholls 1986; Ana Afonso, unpublished data) across its western Mediterranean range whilst maintaining the style polymorphism and heteromorphic incompatibility (Ruiz-Martín, unpublished). Despite the information on incompatibility systems in Linum is limited to only few species, all self-incompatible species display heteromorphic incompatibility, whereas selfcompatible species are monomorphic, with no intermediate cases being reported. Thus, the independent evolution of presence and type of self-incompatibility and style polymorphism proposed by Lloyd & Webb (1992a) is not supported. Interestingly, in eight style-polymorphic ancillary traits (dimorphism on pollen grains and stigmas) seemed to be linked to specific floral morphs, reinforcing the cohesiveness of the heterostylous syndrome in Linum. A possible role of pollinators in the evolution of style plymorphisms in Linum? One of the most insightful predictions made by Lloyd & Webb (1992a) stated that pollinators are critical for the selection of style polymorphisms. Pollinators need to fit tightly with flowers and contact anthers and stigmas in specific body parts to legitimately transfer pollen between morphs. This involves precise shape of flowers and behaviour of pollinators. At present, the scarcity of pollinator data on Linum precludes explicitly testing this hypothesis across the genus. However, studies on the pollination ecology of some species offer interesting insights. Specifically, flower morphology in Linum is relatively consistent in shape across species (funnel-like corolla of limited variation in tube width and length, Fig. 1), thus pollinator behaviour becomes crucial. This has been studied in L. pubescens (eastern Mediterranean range, sect. Dasylinum, clade A1 in Fig. 4; Johnson & Dafni 1998) and L. suffruticosum (western Mediterranean, sect. Linopsis, B2 in Fig. 4; Armbruster et al. 2006), both almost exclusively pollinated by Usia beeflies (Bombyliidae), with U. bicolor in L. pubescens and two species of different size in L. suffruticosum. In these two Linum species, the behaviour of Usia was similar and typical of these beeflies (Orueta 2002): they land on flowers and crawl to the bottom of the flower tube searching for nectar. Armbruster et al. (2006) described that the three 23 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 Page 23 of 54 dimensional reciprocity in L. suffruticosum allows separation of the placement of pollen from L and S flowers on the ventral and dorsal parts of the Usia body respectively. Those authors interpreted that the combination of the Usia behaviour with the three dimensional reciprocity probably increased legitimate pollinations between stylemorphs (Fig. 1). Usia species seem to commonly visit other Mediterranean distylous Linum species (Du Merle & Mazet 1978; and personal observations). Interestingly, Usia is a truly Mediterranean genus, with its highest species diversity in southern Iberian Peninsula, northwestern Africa, and Anatolia (Gibbs 2011; 2014), also with the highest diversity in Linum species. Whether heterostyly in Linum is restricted in the Northern Hemisphere to the Mediterranean basin due to its tight association with Usia flies, is a challenging hypothesis that deserves further insight. The examples of specialized pollination by Usia provide some support to the Darwinian model of Lloyd & Webb (1992a), particularly in L. suffruticosum. This species possess a heteromorphic incompatibility system, which prevents all illegitimate crosses bewteenand within morphs (Nicholls 1986; Ruiz-Martín, unpublished data). Why then has the sophisticated three-dimensional reciprocal distyly, including reciprocal torsion of stamens and styles, evolved apart from increasing efficiency of between-morph pollination and thus avoiding pollen discounting? Torsion of sex organs was first observed by Darwin in L. grandiflorum (Darwin 1877), and latter reported in the monomorphic L. usitatissimum (Schewe et al. 2011). Unfortunately, we lack information on the pollination ecology of heterostylous Linum species in the Cape Floristic Region (CFR) of South Africa , which prevents us to make strong inferences about the causes of the independent evolution of heterostyly there. Although Usia is not present in the CFR, fly pollination in South Africa is common (Johnson 2010), and it would not be surprising that other Bombyliidae or other fly families behave similarly to Usia. Interestingly, the recent description of three-dimensional reciprocity in a group of tristylous CFR Oxalis species (Oxalidaceae) (Turketti et al. 2012), with similar arrangement of stamens and styles to that described in L. suffruticosum and similar flower morphology (i.e. funnellike corollas) confirms the suggestion of Armbruster et al. (2006) that perhaps this kind of polymorphism is not so unusual, and closer examinations of sexual whorl 24 arrangement and pollinator fit can help identifying new examples, providing additional 701 support to the Darwinian view on the function and evolution of heterostyly. 702 703 Conclusions 704 Linum is a good model system for studying the evolution of heterostyly, both at macro 705 and microevolutionary levels. Our data revealed that Linum includes a wide range of 706 morphological variation related to the heterostylous floral syndrome. In contrast, 707 genetic systems linked to heterostyly seems to be rather invariant, for which it could be 708 assumed, as working hypothesis, that pollinators have moulded current floral 709 morphological variation on sex organs. Phylogenetic relationships have been reasonably 710 well resolved, allowing testing specific hypotheses about the evolutionary pathway that 711 allow the acquisition of the style polymorphism. While our analyses precluded inferring 712 the ancestral condition to style polymorphisms in the genus, some of its clades showed 713 that approach herkogamy appears to be the most likely ancestral condition, as Lloyd & 714 Webb (1992a) proposed. Interestingly, species with similar floral trait assemblages in 715 independent clades and in different areas of the Mediterranean basin and South Africa 716 are found. This suggests that ecological adaptations, perhaps mediated by pollinators, 717 rather than phylogenetic conservatism is probably the main driver for the evolution of 718 the stylar polymorphism. Future research to underpin the function of pollinators in the 719 promotion of disassortative pollen transfer in different conditions and regions is 720 necessary to provide further support to the Darwinian pollinator hypothesis for the 721 evolution of heterostyly. 722 723 Acknowledgements 724 This study forms part of a PhD project of JRM, who received a fellowship from MINECO 725 (FPI: BES-2008-003946). This study was funded by MINECO grants (CGL2013-45037-P, 726 CGL2010-11379-E, CGL2009-12565, CGL2006-13847-CO2-01). RSG was recipient of a 727 postdoctoral contract from the Andalusian regional government (excellence grant P09728 RNM-5280) and from the University of Seville. RPB had a postdoctoral contract of the 729 “Juan de la Cierva” program, and ME had a postdoctoral contract of MINECO. Many 730 Page 24 of 54 25 people helped in collecting or locating populations, particularly: J.J. Aldasoro, M. 731 Benavent, Y. Bouchenak-Khelladi, A. de Castro, S. Gómez-González, J. A. Mejías, P. 732 Peñalver, S. Moreno A. Pérez and Ross Turner. Blanca Arroyo, Yuval Sapir and Ross 733 Turner provided some photographs for Fig. 1 and Jordi Bosch identified bees on Linum 734 tenuifolium flowers. Ana Afonso, Silvia Castro and Joao Loureiro provided valuable 735 information on Linum chromosome numbers. We thank Andalusian Scientific 736 Information Technology Center (CICA, Seville, Spain) for providing computational 737 resources. 738 739 740 References 741 Armbruster W.S., Pérez-Barrales R., Arroyo J., Edwards M.E., Vargas P. (2006) Three-742 dimensional reciprocity of floral morphs in wild flax (Linum suffruticosum): a new 743 twist on heterostyly. New Phytologist 171, 581-590. 744 Baker H.G. (1966) The evolution, functioning and breakdown of heteromorphic 745 incompatibility systems. I. The Plumbaginaceae . Evolution 20, 349-368 746 Barrett S.C.H. (2002) The evolution of plant sexual diversity. Nature Reviews Genetics 3, 747 274-284.748 Barrett S.C.H., Jesson L.K., Baker A.M. (2000) The evolution and function of stylar 749 polymorphisms in flowering plants. Annals of Botany 85 (Supplement A): 253-265. 750 Barrett S.C.H., Harder L.D. (2005) The evolution of polymorphic sexual systems in 751 daffodils (Narcissus). New Phytologist 165, 45-53. 752 Barrett S.C.H., Shore J.S. (2008). New insights on heterostyly: comparative biology, 753 ecology and genetics. In: V.E. Franklin-Tong (Ed.) Self-incompatibility in flowering 754 plants (pp. 3-32). Springer, Berlin Heidelberg. 755 Bateson W., Gregory R.P. (1905) On the inheritance of heterostylism in Primula. 756 Proceedings of the Royal Society Series B 76, 581-586. 757 Bell C.D., Soltis D.E., Soltis P.S. (2010) The age and diversification of the angiosperms re-758 revisited. American Journal of Botany 97, 1296-1303 759 Page 25 of 54 32 Santos-Gally R., Gonzalez-Voyer A., Arroyo, J. (2013) Deconstructing heterostyly: the 932 evolutionary role of incompatibility system, pollinators, and floral architecture. 933 Evolution 67, 2072-2082. 934 Schewe L.C., Sawhney V.K., Davis A.R. (2011) Ontogeny of floral organs in flax (Linum 935 usitatissimum; Linaceae). American Journal of Botany 98, 1077-1085. 936 Schoen D.J., Johnston M.O., L'Heureux A.M., Marsolais J.V. (1997) Evolutionary history of 937 the mating system in Amsinckia (Boraginaceae). Evolution 51, 1090-1099. 938 Seetharam A. (1972) Interspecific hybridization in Linum. Euphytica 21, 489-495. 939 Shore J.S., Arbo M.M., Fernández A. (2006) Breeding system variation, genetics and 940 evolution in the Turneraceae. New Phytologist 171, 539-551. 941 Ssymank A., Hamm A., Vischer-Leopold M. (2009). Caring for pollinators safeguarding 942 agro-biodiversity and wild plant diversity. Federal Agency for Nature Conservation 943 (BfN) & Universität Bonn 944 Stebbins G.L. (1970) Adaptive radiation of reproductive characteristics in angiosperms. 945 I. Pollination mechanisms. Annual Review of Ecology and Systematics 1, 307–326.946 Stebbins G.L. (1974) Flowering plants: evolution above the species level. Belknap, 947 Cambridge, MA. USA. 948 Strauss S.Y., Whittall J.B. (2006) Non-pollinator agents of selection on floral traits. In: 949 Harder L.D., Barrett S.C.H. (Eds.) Ecology and evolution of flowers. Oxford University,. 950 Oxford, UK, pp. 120–138. 951 Suksathan P., K. Larsen (2006) A new species of Tirpitzia (Linaceae) from Thailand . Thai 952 Forest Bulletin 34, 201 – 205. 953 The Plant List (2013). Version 1. Published on the Internet; 954 http://www.theplantlist.org/ (accessed 10th March 2017). 955 Thompson J.D., Pailler T., Strasberg D., Manicacci D. (1996). Tristyly in the endangered 956 Mascarene Island endemic Hugonia serrata (Linaceae). American Journal of Botany 957 83, 1160-1167. 958 Page 32 of 54 33 Turketti S.S., Esler K.J., Dreyer L.L. (2012) Three-dimensional reciprocity: A new form of 959 tristyly in South African Oxalis (Oxalidaceae) species and its implications for 960 reproduction. South African Journal of Botany 78, 195-202. 961 Ushijima K., Nakano R., Bando M., Shigezane Y., Ikeda K., Namba Y., Kume S., Kitabata T., 962 Mori H., Kubo Y. (2012) Isolation of the floral morph-related genes in heterostylous 963 flax (Linum grandiflorum): the genetic polymorphism and the transcriptional and 964 post-transcriptional regulations of the S locus. The Plant Journal 69, 317-331. 965 Yang Z., Kumar S., Nei M. (1995) A new method of inference of ancestral nucleotide and 966 amino acid sequences. Genetics 141, 1641-1650. 967 Yılmaz Ö., Kaynak G. (2008) The check-list and chorology of the Linum L.(Linaceae) taxa 968 in the flora of Turkey. Journal of Biological and Environmental Sciences 2, 5-43. 969 Yurkevich O.Y., Naumenko-Svetlova A.A., Bolsheva N.L., Samatadze T.E., Rachinskaya 970 O.A., Kudryavtseva A.V., Zelenina D.A., Volkow A.V., Zelenin A.V., Muravenko O.V. 971 (2013) Investigation of genome polymorphism and seed coat anatomy of species of 972 section Adenolinum from the genus Linum. Genetic Resources and Crop Evolution 60, 973 661-676974 Yusepchuk S.V. (1949) Linaceae In: B.K. Shishkin (Ed) Flora of the USSR, vol. XIV. 975 Akademii Nauk SSSR, Moscow, Leningrad. 976 977 978 Page 33 of 54 34 Figure legends 979 980 Fig. 1. Floral variation and pollinators in Mediterranean Linum species: a) L. viscosum 981 visited by an Halictidae bee, b) L. narbonense with Usia pubera beefly collecting nectar, 982 c) and d) L. tenuifolium visited by Ceratina cucurbitina and Lasioglossum malachurum 983 bees, respectively, e) three dimensional reciprocity in L. suffruticosum, f) Usia sp. 984 collecting nectar in L. suffruticosum, g) conventional distyly in L. tenue. h) L. tenue with 985 nectar collecting U. pusilla, i) L. comptonii visited by pollen collecting Amegilla in South 986 Africa, and j) L. pubescens with Usia bicolor in Israel. Photographs by Blanca Arroyo (c, 987 d), Ross Turner (i) and Yuval Sapir (j). 988 989 Fig. 2. Phylogenetic tree of Linaceae based on BEAST analysis of combined nuclear (ITS) 990 and plastid (trnL-F, matK and ndhF) DNA regions. Numbers above each branch indicate 991 posterior probability support. Bars in each node indicate 95% CI of the age of each node. 992 Time scale on the horizontal axis is in millions of years. 993 994 Fig. 3. Maximum likelihood ancestral state reconstruction of stylar polymorphism in 995 Linum. Two ancestral states (blue = monomorphic, red = polymorphic) are considered 996 as the simplest way to understand the evolution of heterostyly (see Material and 997 methods for details). Letters above branches are referenced in main text. 998 999 Fig. 4. Maximum likelihood ancestral state reconstruction of stylar polymorphism in 1000 Linum. Five relevant states to the two competing hypotheses of the evolution of 1001 heterostyly are considered (Charlesworth & Charlesworth 1979, Lloyd & Webb, 1992a; 1002 see Materials and Methods for details). Colours represent the different stylar conditions: 1003 blue = homostyly, red = style polymorphism, green = approach herkogamy, yellow = 1004 horizontal herkogamy, and orange = reverse herkogamy. Letters above branches are 1005 referenced in main text. 1006 1007 Page 34 of 54 35 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 Page 35 of 54 Fig. S1. Phylogenetic tree of Linaceae based on BEAST analysis of combined nuclear (ITS) and plastid (trnL-F, matK and ndhF) DNA regions. Numbers above each branch indicate posterior probability support. Bars in each node indicate 95% CI of the age of each node. Time scale on horizontal axis is in millions of years. Tip labels include species name, section and distribution. Table S1. Sources of plant material and traits considered in the study. Taxa are arranged alphabetically by section and family. Sampled populations refer to GenBank accession numbers when obtained from published references, to samples supplied by herbaria (E: Royal Botanic Garden Edinburgh, MA: Royal Botanic Garden Madrid, SEV: University of Seville; codes for specimens are given). NA, not available. For Peer Review Fig. 1. Floral variation and pollinators in Mediterranean Linum species: a) L. viscosum visited by an Halictidae bee, b) L. narbonense with Usia pubera beefly collecting nectar, c) and d) L. tenuifolium visited by Ceratina cucurbitina and Lasioglossum malachurum bees, respectively, e) three dimensional reciprocity in L. suffruticosum, f) Usia sp. collecting nectar in L. suffruticosum, g) conventional distyly in L. tenue. h) L. tenue with nectar collecting U. pusilla, i) L. comptonii visited by pollen collecting Amegilla in South Africa, and j) L. pubescens with Usia bicolor in Israel. Photographs by Blanca Arroyo (c, d), Ross Turner (i) and Yuval Sapir (j). Page 36 of 54 For Peer Review Fig. 2. Phylogenetic tree of Linaceae based on BEAST analysis of combined nuclear (ITS) and plastid (trnLF, matK and ndhF) DNA regions. Numbers above each branch indicate posterior probability support. Bars in each node indicate 95% CI of the age of each node. Time scale on the horizontal axis is in millions of years. 405x958mm (200 x 200 DPI) Page 37 of 54 For Peer Review Fig. 3. Maximum likelihood ancestral state reconstruction of stylar polymorphism in Linum. Two ancestral states (blue = monomorphic, red = polymorphic) are considered as the simplest way to understand the evolution of heterostyly (see Material and methods for details). Letters above branches are referenced in main text. 210x297mm (200 x 200 DPI) Page 38 of 54 For Peer Review Fig. 4. Maximum likelihood ancestral state reconstruction of stylar polymorphism in Linum. Five relevant states to the two competing hypotheses of the evolution of heterostyly are considered (Charlesworth & Charlesworth 1979, Lloyd & Webb, 1992a; see Materials and Methods for details). Colours represent the different stylar conditions: blue = homostyly, red = style polymorphism, green = approach herkogamy, yellow = horizontal herkogamy, and orange = reverse herkogamy. Letters above branches are referenced in main text. 210x297mm (200 x 200 DPI) Page 39 of 54 For Peer Review Fig. S1. Phylogenetic tree of Linaceae based on BEAST analysis of combined nuclear (ITS) and plastid (trnLF, matK and ndhF) DNA regions. Numbers above each branch indicate posterior probability support. Bars in each node indicate 95% CI o f the age of each node. Time scale on horizontal axis is in millions of years. Tip labels include species name, section and distribution. 464x574mm (200 x 200 DPI) Page 40 of 54 1 Table S1. Source of plant material and traits considered in the study. Taxa are arranged alphabetically by section and family. Sampled populations refer to GenBank accession numbers when obtained from published references, to samples supplied by herbaria (E: Royal Botanic Garden Edinburgh, MA: Royal Botanic Garden at Madrid, SEV: University of Seville; codes for specimens are given). NA, not available. 1. Taxon 2. Section or family 3. Distribution 4. Sampled populations 5. Coordinates 6. GenBank Accession no. ITS 7. GenBank Accession no. ndhF5-8 8. GenBank Accession no. trnL-F 9. GenBank Accession no. matK 10. Stylar condition (binary) 11. Stylar condition (five states) 12. Lifeform 13. References for columns 10, 11, 12 14. Chromosome Number 15. References for column 14 16. Breeding system 17. References for column 16 Linum catharticum Cathartolinum N Medit. Ref. 1 NA FJ169533 FJ160796 FJ160880 HM544103 Monomorphic Homostylous Annual 2 n=8/2n=16 26 -- -- L. densiflorum A Dasylinum Azerbaijan This study E00450740 NA Forthcoming Forthcoming Forthcoming Forthcoming Polymorphic Polymorphic Perennial 3 -- -- -- -- L. densiflorum B Dasylinum Turkey This study 40º06'57.8''N 32º36'17.8''E Forthcoming Forthcoming Forthcoming Forthcoming Polymorphic Polymorphic Perennial 3 -- -- -- -- L. hirsutum Dasylinum Turkey Ref. 1 NA FJ169520 FJ160788 FJ160872 HM544106 Polymorphic Polymorphic Perennial 3 n=8; 2n=16/n=16; 2n=32 27, 28, 29 SI 52 L. hypericifolium Dasylinum Turkey Ref. 1 NA FJ169519 FJ160789 FJ160873 HM544107 Polymorphic Polymorphic perennial 4 -- -- -- -- L. olympicum Dasylinum Turkey, Greece This study E00450745 NA Forthcoming Forthcoming Forthcoming Forthcoming NA NA Perennial 5 -- -- -- -- L. pubescens Dasylinum Syria Ref. 1 NA FJ169518 FJ160790 FJ160874 NA Polymorphic Polymorphic Annual 6 2n=18 ; 2n=16 30, 31 SI 53 L. seljukorum Dasylinum Turkey This study E00450754 NA Forthcoming Forthcoming Forthcoming Forthcoming Monomorphic Homostylous Annual 3 n=8; 2n=16 -- -- -- L. spathulatum Dasylinum Greece This study 40º04.9'N 22º22.7'E Forthcoming Forthcoming Forthcoming Forthcoming Polymorphic Polymorphic Perennial 5 2n=16/2n=36 32;,28 -- -- L. unguiculatum Dasylinum Turkey This study E00450741 NA Forthcoming Forthcoming Forthcoming Forthcoming NA NA Perennial 3 -- -- -- -- L. viscosum Dasylinum Spain This study NA FJ169517 FJ160791 FJ160875 NA Polymorphic Polymorphic Perennial 2 n=8; 2n=16 27 -- -- L. acuticarpum Linopsis South Africa This study 33º59'55.6"S 20º26'33.7"E Forthcoming Forthcoming Forthcoming Forthcoming Monomorphic Approach herkogamous Perennial 7 n=15 33 -- -- L. adustum Linopsis South Africa This study 34º33'02.1"S 19º25'37.8"E Forthcoming Forthcoming Forthcoming Forthcoming Monomorphic Approach herkogamous Perennial 7 -- -- -- -- L. aethiopicum Linopsis South Africa This study 34º25'48.8"S 20º39'50.5"E Forthcoming Forthcoming Forthcoming Forthcoming Monomorphic Approach herkogamous Perennial 7 -- -- -- -- L. africanum Linopsis South Africa This study 34º09'39.4"S 18º52'16.0"E Forthcoming Forthcoming Forthcoming Forthcoming Monomorphic Approach herkogamous Perennial 7 n=16/2n=30; 2n=29 30, 33 SC 54 Page 41 of 54 8 80. Rogers C.M. (1980). In: Löve (Ed.), Chromosome number reports LXVII. Taxon 29, 347-367. 81. Rogers C.M. (1982) The systematics of Linum sect. Linopsis (Linaceae). Plant Systematics and Evolution 140, 225. 82. Gill S. (1987). Linseed. Indian Council of Agricultural Research Krishi Anusandhan Bhavan Pusa, New Delhi. 83. Devesa J.A., Talavera S., Galiano E.F. (1984) In A. Löve (Ed.). Chromosome Number Reports LXXXII. Taxon, 33, 126-134. Page 48 of 54