Molecular approaches reveal speciation between red- and blue-flowered plants in the Mediterranean Lysimachia arvensis and L. monelli (Primulaceae)

557 Botanical Journal of the Linnean Society, 2022, 199, 557–577. With 5 figures. © 2022 The Linnean Society of London, Botanical Journal of the Linnean Society, 2022, 199, 557–577 *Corresponding author. E-mail: [email protected] Molecular approaches reveal speciation between red- and blue-flowered plants in the Mediterranean Lysimachia arvensis and L. monelli (Primulaceae) FRANCISCO JAVIER JIMÉNEZ-LÓPEZ1,2,*, JUAN VIRUEL3,, MONTSERRAT ARISTA1, PEDRO L. ORTIZ1 and MARÍA TALAVERA1 1Departamento de Biología Vegetal y Ecología, Facultad de Biología, Universidad de Sevilla, Apdo. 1095, 41080 Sevilla, Spain 2Phenology Lab, Biosciences Institute, Department of Biodiversity, UNESP – São Paulo State University, São Paulo, Brazil 3Royal Botanic Gardens, Kew, TW9 3DS, Richmond, UK Received 16 April 2021; revised 6 September 2021; accepted for publication 4 October 2021 Flower colour has played a decisive role as an evolutionary force in many groups of plants by driving speciation. A well-known example of colour polymorphism is found across the Mediterranean populations of Lysimachia arvensis and L. monelli, in which blue- and red-flowered plants can be found. Previous studies recognized two lineages within L. arvensis differing in flower colour, but this variation has not yet been considered in a phylogenetic context. We have reconstructed the ancestral states of flower colour across Mediterranean Lysimachia spp. aiming at understanding its phylogenetic signal using the nuclear internal transcribed spacer (ITS) sequences and three plastid markers. All blue- and red-flowered specimens were nested in different clades in the ITS tree, thus supporting that L. arvensis and L. monelli are polyphyletic, whereas low phylogenetic resolution was found in plastid markers. Monophyly was reconstructed for blue-flowered L. arvensis and L. monelli samples, and similarly for red-flowered individuals of each species: (1) blue-flowered L. arvensis was reconstructed as sister to the strictly blue-flowered L. talaverae in a monophyletic clade sister to remaining Lysimachia; (2) red-flowered L. arvensis was resolved as sister to redflowered L. monelli in a monophyletic clade; and (3) clade 2 was sister to blue-flowered L. monelli and the strictly blue-flowered L. foemina. Our results suggest that colour lineages in L. arvensis and L. monelli constitute different species, but flower colour did not promote the separation of these lineages. We propose a new name for blue-flowered L. arvensis (L. loeflingii) and a new combination for red-flowered L. monelli (L. collina), maintaining L. arvensis for red-flowered plants and L. monelli for blue-flowered plants. ADDITIONAL KEYWORDS: Anagallis – colour polymorphism – molecular taxonomy – nuclear and plastid DNA regions – phylogeny. INTRODUCTION Corolla colour is one of the most important traits influencing pollinator attraction. Different pollinators show innate preferences for specific corolla colours (Lunau & Maier, 1995; Dyer et al., 2016) by increasing the frequency of visits to these flowers. Petal colour polymorphism is the presence of discrete flower phenotypes genetically determined within populations of a species (Huxley, 1955). Corolla colour polymorphism may promote speciation as it contributes to nonrandom mating, which can subsequently result in reproductive isolation (Sobel & Streisfeld, 2015). As a result of colour variation, species can be subject to divergent selection by pollinator preferences, and they can be subject to indirect selection resulting from genetically correlated traits (e.g. Levin & Brack, 1995; Armbruster et al., 1997; Armbruster, 2002; Frey, 2004). Indirect selection of corolla colour polymorphism, as a ‘magic trait’ (Servedio et al., 2011), may respond to both current and palaeoclimatic abiotic factors. Historical events involved in shaping of the Mediterranean climate, such as the Messinian Salinity Crisis applyparastyle “fig//caption/p[1]” parastyle “FigCapt” This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. Downloaded from https://academic.oup.com/botlinnean/article/199/2/557/6494517 by Universidad de Sevilla user on 04 November 2022 558 F. J. JIMÉNEZ-LÓPEZ ET AL. © 2022 The Linnean Society of London, Botanical Journal of the Linnean Society, 2022, 199, 557–577 (5.96–5.30 Mya; Krijgsman et al., 2010) and the onset of the Mediterranean climate (3.4–2.8 Mya; Suc, 1984), have played a key role in the diversification of species (e.g. Fiz-Palacios & Valcárcel, 2013). Lysimachia arvensis (L.) Manns & Anderb. (synonym: Anagallis arvensis L.) and L. monelli (L.) Manns & Anderb. (synonym: Anagallis monelli L.) are two Mediterranean species that show intraspecific flower colour polymorphism (Ferguson, 1972; Pujadas, 1997). Petal colours are genetically determined (Freyre & Griesbach, 2004; Sánchez-Cabrera et al., 2021), and in both species blue- and red-flowered individuals are found due to the presence of different anthocyanins in their petals (Ishikura, 1981; Quintana et al., 2008; Sánchez-Cabrera et al., 2021). The species differ in ploidy, L. arvensis being tetraploid (revised by Pastor, 1992) and L. monelli being diploid (Kress, 1969; García Pérez et al., 1997). In L. arvensis, colour morphs also differ in other traits such as flowering phenology or type of herkogamy (Arista et al., 2013; Jiménez-López et al., 2020c). The colour morphs show different geographical distributional patterns, blue-flowered plants appearing mainly in drier Mediterranean localities and red-flowered plants being predominant in more temperate areas (Arista et al., 2013). Blue- and red-flowered plants may appear in sympatric and allopatric populations, and pollinators show a preference for visiting blue-flowered plants in Mediterranean polymorphic populations (Ortiz et al., 2015) and high colour constancy patterns (Jiménez- López et al., 2020a). Hand-pollination between red and blue individuals gives rise to homogeneous F1 progeny with salmon-coloured flowers (Jiménez-López et al., 2020a), but these are infrequent in wild populations (Jiménez-López et al., 2020c). This ‘hybrid’ phenotype has been described as Anagallis × amoena Heldr. ex Halácsy (de Halácsy, 1904: 11). Nuclear microsatellite markers reconstructed two independent genetic groups for each colour morph, supporting this reproductive isolation between them (Jiménez-López et al., 2020b). All this ecological, morphological, reproductive and molecular evidence suggests that the two colour morphs of L. arvensis are independent lineages. In contrast, sympatric populations of blue- and red-flowered individuals of L. monelli have not been found. Blue-flowered plants are restricted to the Iberian Peninsula (Spain and Portugal), Morocco, Algeria, north-western Libya (Tripolitania) and Tunisia (Jahandiez & Maire, 1934), whereas redflowered plants grow mainly in Morocco, Sardinia and a restricted area of Catalonia (north-eastern Spain) where blue plants were not observed (Willkomm, 1870; Jahandiez & Maire, 1934). Apart from the selfincompatibility system exhibited by the blue-flowered plants (Talavera et al., 2001), little is known about reproduction of the two morphs of L. monelli. A study by Quintana et al. (2008) found low reproductive success after hand-pollination between blue- and red-flowered plants. Nevertheless, the strikingly different geographical ranges prevent any potential crosses between blue- and red-flowered individuals of L. monelli in nature. Previous phylogenetic studies have scarcely explored the potential implications of corolla colour polymorphism in taxon delimitation in L. arvensis and L. monelli because this trait was considered part of the infraspecific variation. Consequently, only one colour morph per species was sampled in most of the molecular analyses (Martins, Oberprieler & Hellwig., 2003; Manns & Anderberg, 2005, 2007a; Anderberg, Manns & Källersjö., 2007; Yan et al., 2018). Only in one study (Manns & Anderberg, 2007b) was a sample from each colour morph of L. arvensis from a Greek locality included. These two individuals were reconstructed as a monophyletic group based on plastid markers, but were nested in independent clades in the phylogenetic tree based on internal transcribed spacer (ITS) sequences. Larger sampling of plants of the two corolla colours for both L. arvensis and L. monelli is required in phylogenetic studies to clarify the implications of flower colour in taxon boundaries. In this context, the aims of our study are to (1) investigate their phylogenetic relationships, (2) reconstruct the ancestral state of flower colour evolution, (3) suggest possible species delimitation and (4) estimate the divergence time between petal colour morphs of L. arvensis and L. monelli. MATERIAL AND METHODS Study SpecieS Lysimachia arvensis is a Mediterranean species currently distributed across the world as an alien species. It is annual, self-compatible (Gibbs & Talavera, 2001) and tetraploid (2x = 40; revised by Pastor, 1992). Lysimachia monelli is distributed exclusively in the west of the Mediterranean Region. It is perennial, selfincompatible (Gibbs & Talavera, 2001; Talavera et al., 2001) and diploid (2x = 20; Kress, 1969; Valdés, 1970; Šveřepová, 1972; García Pérez et al., 1997). Plant material from the other two Lysimachia spp. present in the Mediterranean Basin was also included (Manns & Anderberg, 2009; Aymerich & Sáez, 2015). Lysimachia foemina (Mill.) Manns & Anderb. is an annual, self-compatible (Marsden-Jones, 1935; Marsden-Jones & Weiss, 1938) and tetraploid species (2x = 40; revised by Pastor, 1992). Lysimachia foemina, as L. arvensis, is currently distributed across the world as an alien species. Lysimachia talaverae L.Saéz & Aymerich (Aymerich & Sáez, 2015) is Downloaded from https://academic.oup.com/botlinnean/article/199/2/557/6494517 by Universidad de Sevilla user on 04 November 2022 PHYLOGENY OF MEDITERRANEAN LYSIMACHIA 559 © 2022 The Linnean Society of London, Botanical Journal of the Linnean Society, 2022, 199, 557–577 endemic to the western Mediterranean, and is annual, self-compatible (Gibbs & Talavera, 2001) and diploid (2x = 20; Šveřepová, 1972; García Pérez et al., 1997). Molecular SaMpling, dna extraction, pcr aMplification, Sequencing and alignMent Fresh leaf material from 28 populations of L. arvensis (LA), 12 of L. monelli (LM), three of L. foemina (LF) and four of L. talaverae (LT) (Fig. 1; Supporting Information, Table S1) was collected and dried in silica-gel (Chase & Hills, 1991). This sampling represents nine pure red and 12 pure blue populations and seven mixed (red and blue) populations of L. arvensis (the colour of each sample was coded as LA_R for red-flowered samples and LA_B for blue-flowered samples in L. arvensis). Sampling of L. monelli included four pure red (LM_R) and eight pure blue populations (LM_B). Also, one sample of L. azorica Hornem. ex Hook. (LZ) and one sample of two populations of L. linum-stellatum (L) Hoffmanns. & Link (LLS) were included (Table S1) in the phylogenetic analyses, and L. tenella L. (AY855150) and L. tyrrhenia U.Manns & Anderb. (AY855136) were used as outgroups. Total genomic DNA was isolated with Invisorb Vegetal DNA Kit HTS 96 (Invitek, Spain), with modifications following Jiménez-López et al. (2016). The quality of the extracted DNAs was checked in a 1% TAE-agarose gel, and DNA concentration was estimated using a NanoDrop DS-11 Spectrophotometer (DeNovix). The nuclear ribosomal DNA (nrDNA) ITS and three plastid markers were used for phylogenetic reconstructions. The partial 18S, ITS-1, 5.8S, ITS-2 and partial 28S regions of the rDNA were amplified using primers ITS5 and ITS4 (White et al., 1990). We tested 19 plastid markers (Supporting Information, Table S2) in eight samples of L. arvensis (four blue and four red specimens), but only three of them showed polymorphic sites between samples. PCR amplifications of ITS were performed following Jiménez-López et al. (2016) using 4 ng of DNA as input, and the polymorphic plastid markers rps16- trnK and rpl32-trnL were amplified following Shaw et al. (2007) and trnH-psbA was amplified following Hamilton (1999). After purification with ExoSAP-IT (Roche, Spain), fragments were sequenced in an ABI 3730 machine using the BigDye Terminator Cycle Sequencing Kit (Applied Biosystems, Foster City, CA, USA) at STAB Vida Lda. (Oeiras, Portugal). Both forward and reverse DNA sequences were obtained and assembled from double-stranded PCR products. The ITS sequences were carefully checked for double peaks. Twelve bases proved to have polymorphic states in some taxa, probably due to incomplete homogenization between different ITS paralogues. Eight of these bases, present in more than one sample, never in different taxa, were removed before analysis to avoid artificial results, following Manns & Anderberg (2007b). Sequences were edited using Geneious R10 (Biomatters Ltd, Auckland, New Zealand) and the ends of each sequence, of low quality, were trimmed. Alignments were conducted in MEGA v.X (Kumar et al., 2018; Stecher, Tamura & Kumar, 2020) using the algorithm Clustal W (Thompson, Gibson & Higgins, 2002), and visually inspected. Four insertions/deletions (indels) of single bases were identified for the ingroup in the ITS matrix, and seven indels separated ingroup Figure 1. Geographical distribution of the studied populations of Lysimachia arvensis (circles) and L. monelli (triangles). Colours correspond to flower colour of the studied populations: blue and red for pure blue and red populations, respectively, and pink for mixed populations. Map based on the projection generated by Google Earth for the coordinates of the studied populations (Table S1). Downloaded from https://academic.oup.com/botlinnean/article/199/2/557/6494517 by Universidad de Sevilla user on 04 November 2022 560 F. J. JIMÉNEZ-LÓPEZ ET AL. © 2022 The Linnean Society of London, Botanical Journal of the Linnean Society, 2022, 199, 557–577 taxa from outgroups; there were four indels (1–4 bp long) in ITS1 and three indels (4–7 bp long) in ITS2. Indels in the ITS matrix were coded as gaps and treated as partial deletions. In the plastid DNA regions, five indels were identified within the ingroup. Two indels (one of 4 bp and other of 1 bp) in the rps16-trnK matrix, two (1 bp) in the rpl32-trnL matrix and one (1 bp) in the trnH-psbA matrix were present. However, the outgroup showed 21 indels for the three plastid DNA regions studied: 11 in the rps16-trnK matrix (1–20 bp long), four in the rpl32-trnL matrix (5–9 bp long) and five in the trnH-psbA matrix (1–7 bp long). These indels were coded as gaps and treated as partial deletions. Finally, in the trnH-psbA matrix, one long insertion (69 bp) was observed in the outgroup with respect to the ingroup, which was excluded from the analysis. phylogenetic inference The Akaike information criterion (AIC) was used to estimate the evolution model that better fits each DNA matrix using jModeltest 2.1.4 (Darriba et al., 2012). Phylogenetic relationships were inferred based on maximum-parsimony (MP) inference using PAUP* (Swofford, 2003) and maximum-likelihood (ML) methods using RAxML v.8 (Stamatakis, 2014). Bootstrap support (BS) was estimated with 1000 bootstrap replicates following full heuristic searches. Bayesian inference (BI) was conducted using MrBayes (Ronquist et al., 2012) on XSEDE (3.2.7a) (CIPRES Science Gateway v,3.3; Miller, Pfeiffer & Schwartz, 2010). Phylogenetic reconstructions were calculated for each DNA region (ITS and each of the three plastid DNA markers), and for a concatenated matrix of the three plastid regions. Following Viruel et al. (2016), clades with BS of 75–100 or posterior probability (PPS) of 0.95–1.0 were considered moderately to strongly supported. Phylogenetic reconstruction with ITS was performed with and without the sample LA19_B according to the results of the recombination analysis (see below and Results). To test for incongruent topologies between the independent DNA data matrices, a partition homogeneity test was performed (Farris et al., 1994; Symonds & Lloyd, 2003). The incongruence length difference (ILD) test of Farris et al. (1994) implemented in PAUP* (Swofford, 2003) was performed using 1000 random-order-entry replicates to estimate if the nuclear and the three plastid data sets were significantly different from random partitions of the same size. recoMbination analySiS To test possible recombination events in the ITS sequences of all the samples, seven methods (RDP, GENECONV, BootScan, MaxChi, Chimaera, SiScan and 3SEQ) implemented in RDP4 v.484 were applied to infer recombination events (Martin et al., 2015; Viruel et al., 2018). anceStral State reconStruction To estimate the ancestral state of flower colour, we performed ancestral state reconstruction of colour classes using an ML approach with the function ‘ace’ and a Markov chain Monte Carlo (MCMC) approach called stochastic character mapping (SIMMAP; Huelsenbeck et al., 2003) included in the package ‘phytools’ (Revell, 2012) implemented in R. We performed the analyses on the RAxML phylogenetic tree obtained from the ITS dataset. We fitted three different models for both approaches (ML and SIMMAP): equal-rate model (ER), symmetric model (SYM) and all-rate-different model (ARD). The best-fitting model was selected according to the AIC. For SIMMAP analyses, we ran 1000 simulations per tree. Because we included more than one sample per species, we also pruned the additional samples for the same species in the case of monophyly. Ancestral state reconstruction was shown only on major clades. SpecieS deliMitation Putative species limits were explored with four methods: (1) a multi‐rate Poisson tree process (mPTP) provided at the mPTP webpage (https://mptp.h-its. org) (Kapli et al., 2017); (2) the Automatic Barcode Gap Discovery (ABGD) (Puillandre et al., 2012); (3) a ‘classical’ DNA barcoding gap analysis based on the Kimura two-parameter (K2P) method (Kimura, 1980) assessed with the APE R package (Paradis, Claude & Strimmer., 2004) and ggplot2 R package (Wickham, 2011); and (4) the species delimitation plugin (SDP) (Masters, Fan & Ross., 2011) implemented in Geneious. These methods were applied to the ITS sequence data, commonly used for inferring phylogenetic relationships at low taxonomic levels (Soltis & Soltis, 1998; Duminil & Di Michele, 2009). The mPTP inferred putative species boundaries on an ML phylogenetic input tree, assigning intra- or interspecific categories of each branch. The multi-rate implementation of the mPTP considers different evolutionary rates in the differentiation of each putative species and uses a single ML tree as input. To run the mPTP model on our dataset, we uploaded the previously estimated RAxML tree, following Hanusch et al. (2020). The K2P model estimated evolutionary divergence between sequences from the average number of base substitutions per site of all sequence pairs among groups, with a gamma distribution; and standard error estimates were obtained by a bootstrap procedure (100 replicates). The ABGD method statistically infers and finds the Downloaded from https://academic.oup.com/botlinnean/article/199/2/557/6494517 by Universidad de Sevilla user on 04 November 2022 PHYLOGENY OF MEDITERRANEAN LYSIMACHIA 561 © 2022 The Linnean Society of London, Botanical Journal of the Linnean Society, 2022, 199, 557–577 distance at which a barcode gap occurs and sorts the sequences into molecular operational taxonomic units (MOTUs) based on this distance (Puillandre et al., 2012). ABGD was run online (https://bioinfo.mnhn.fr/ abi/public/abgd/abgdweb.html) with default settings [Pmin = 0.001, Pmax = 0.1, Steps = 10, X (relative gap width):1.5 & Nb bins (for distance distribution): 20] and with Kimura (K80) TS/TV distances (Kimura, 1980). Finally, the SDP in the Geneious software evaluates the phylogenetic monophyly of each putative taxon considered as a clade by testing the probability that this monophyly occurred in a coalescent process, and it assesses the probability with which a putative taxon can be diagnosed successfully on a phylogenetic tree by comparing intra- and interspecific genetic distances (Masters et al. 2011). SDP was used to calculate intra/inter ratios (‘intra’ is the differentiation among members of a clade, and ‘inter’ is the genetic differentiation between the members of a clade and its closest neighbouring clade), values of P ID(Strict) (the mean probability that an unknown specimen of a given clade will effectively be placed within this clade), and two statistical tests: Rosenberg’s PAB (Rosenberg, 2007) and Rodrigo’s P (PD) (Rodrigo et al., 2008). Rosenberg’s PAB tests whether the monophyly of the selected group in the gene tree is due to an evolutionary process or if it is the result of random branching of the tree caused by an insufficient sample size (Rosenberg, 2007). Rodrigo’s P (Randomly Distinct) uses the ratio between the length of the branches within a selected group and the length between selected groups to estimate whether the distinctiveness of the groups is due to an evolutionary process or to random coalescent processes (Rodrigo et al., 2008). divergence-tiMe eStiMation We used molecular clock approaches to estimate divergence times among the main clades. We selected one random sequence for each colour and taxon studied, with sequences of other Lysimachia and related species from GenBank (Supporting Information, Tables S1 and S3). A phylogenetic reconstruction was inferred with ML methods with RAxML (Stamatakis, 2014) following the parameters described above. We used a Bayesian uncorrelated-lognormal relaxed-clock approach (Drummond et al., 2006) as implemented in BEAST v.2.5 (Bouckaert et al., 2019). Node ages and absolute branch substitution rates were estimated using two relaxed clock methods that represent two contrasting approaches to estimate divergence times (Magallón, Hilu & Quandt, 2013). We used both Yule speciation and birth–death tree priors (Heled & Drummond, 2012). We ran 30 independent analyses (50 million generations each, sampled every 5000 generations). Tracer 1.7.1 software (Rambaut et al., 2018) was used for convergence and satisfactory effective sample size (ESS) values (> 200) and verify that a burn-in of 10% was appropriate. Subsequent parameter distributions were obtained by combining the independent MCMCs with LogCombiner 1.10.4 and calculating the maximum credibility tree using TreeAnnotator 1.10.4 (Suchard et al., 2018). Fossil records are scarce because these plants lack woody parts, and both seed and pollen production are low. Maesa Forssk is a first-diverging taxon in Primulaceae (Anderberg et al., 2007; Yesson, Toomey & Culham, 2009) and its pollen is known from the Early Eocene (55.8 Mya; Huzioka & Takahasi, 1970) according to the Paleobiology Database (http:// paleodb.org/). Furthermore, fossil seeds similar to those of Lysimachia vulgaris L. and its close relatives (Oh et al., 2008) are found from the Mid-Miocene (12–16 Mya; Friis, 1985). We therefore used the pollen record to set the age prior for rooting and seed fossil to set the age prior for L. vulgaris. The most generally appropriate prior for fossil calibrations is the lognormal distribution (Ho & Phillips, 2009). In addition, a secondary calibration was applied to the stem age of core Lysimachia L. using an estimated mean age of 30 Mya from previous phylogenetic reconstructions (Yesson et al., 2009; Zanne et al., 2014). In this case, we used a standard normal prior distribution with mean of 30 Mya and standard deviation (SD) of 6 Mya, yielding a 95% highest posterior density (HPD) of 20–40 Mya, following Yan et al. (2018). To test the effect of different calibration points and tree prior approaches, we also performed analyses using only one fossil calibration, only secondary calibration and both calibration points. We then performed a regression between the node ages of the same clades from all analyses. RESULTS phylogenetic reconStructionS Fifty-two high-quality sequences were obtained for ITS, and 37 for the plastid regions. The ITS1 and ITS2 fragments in the ingroup alignment ranged from 230 to 231 and 205 to 207 bp, respectively. No significant variation among samples within each taxon was found, except six samples of LA_B (LA4_B, LA6_B, LA7_B, LA11_B, LA12_B & LA13_B), which showed three consistent transitions of 1 bp (two in ITS1 and one in ITS2) with respect to the other LA_B samples. A total of 625 aligned characters of ITS sequences were used for phylogenetic analyses. We observed 406 constant characters (65.0%), 27 variable parsimonyuninformative characters (4.3%) and 192 potentially parsimony-informative characters (30.7%). The Downloaded from https://academic.oup.com/botlinnean/article/199/2/557/6494517 by Universidad de Sevilla user on 04 November 2022 562 F. J. JIMÉNEZ-LÓPEZ ET AL. © 2022 The Linnean Society of London, Botanical Journal of the Linnean Society, 2022, 199, 557–577 heuristic search resulted in six MP trees, with a tree length (TL) of 304, a consistency index (CI) of 0.84 (excluding uninformative characters) and a retention index (RI) of 0.94. The ILD test showed significant differences between nuclear and plastid regions (P = 0.010), indicating inconsistency between these markers. Thus, phylogenetic analyses were performed for the ITS region separately from the concatenated dataset representing the three plastid markers. The MP, RAxML and Bayesian trees reconstructed the same topology for all analyses based on the ITS region (Fig. 2). Two main clades were observed: Clade I (C1, 62 BS, 0.9988 PPS) which contained LA_B samples (100 BS, 1.000 PPS) sister to LT (97 BS, 1.000 PPS), whereas the samples of L. arvensis from south-west Spain (LA4_B, LA6_B, LA7_B and LA11_B-LA13_B) formed a clade (82 BS, 0.975 PPS); and Clade II (C2; 80 BS, 0.850 PPS) comprising two subclades, named A and B. In subclade A (SCA; 87 BS, 0.968), LA_R samples (95 BS, 1.000 PPS) were sister to the LM_R samples (100 BS, 1.000 PPS). In subclade B (SCB; 94 BS, 0.9859), LM_B samples (99 BS, 0.999 PPS) were sister to LF (100 BS, 1.000 PPS). Therefore, blue and red individuals of both L. arvensis and L. monelli appeared in different groups. Blue- and Figure 2. Ancestral state reconstruction of flower colour of Lysimachia arvensis, L. monelli and related taxa (see Table 1). RAxML tree based on ITS sequences. The samples of L. arvensis from mixed populations are highlighted in bold. Bootstrap values > 50% are given above the branches. Posterior probabilities of the Bayesian approach were coded as: a (0.995–1.000 PPS), b (0.950–0.994 PPS) and c (0.850–0.949); these are indicated below branches. C1: Clade 1; C2: Clade 2; SCA: Subclade A and SCB: Subclade B. * (94 BS, 0.9859 PPS); † (82 BS, 0.975 PPS). Downloaded from https://academic.oup.com/botlinnean/article/199/2/557/6494517 by Universidad de Sevilla user on 04 November 2022 PHYLOGENY OF MEDITERRANEAN LYSIMACHIA 563 © 2022 The Linnean Society of London, Botanical Journal of the Linnean Society, 2022, 199, 557–577 red-flowered individuals from mixed populations of L. arvensis appeared in independent clades (Fig. 2; Supporting Information, Fig. S1). Each plastid DNA partition (rps16-trnK, rpl32-trnL and trnH-psbA) varied on aligned length, number of potentially informative sites and values of parsimony analyses (Table 1). Visual analysis of these sequences pointed to a split in two main haplotypes, particularly in the rps16-trnK and rpl32-trnL matrices. All LT and LA_R, and most LA_B samples share one haplotype (H1), whereas LM_B, LM_R, LF and some LA_B samples share another haplotype (H2). Both haplotypes are differentiated by the presence of indels and several single base mutations (Table 2), and H1 is more divergent from the outgroup than H2. For each plastid DNA region separately, the trees reconstructed showed a congruent topology in all analyses and markers (Supporting Information, Figs S2–S4). The concatenated plastid DNA dataset comprised 1695 aligned characters, of which 1459 were constant characters (86.1%), 159 variable parsimony-uninformative characters (9.4%) and 77 potentially parsimony-informative characters (4.5%). Heuristic search resulted in 100 MP trees, with a TL of 129, CI of 1.0 (excluding uninformative characters) and RI of 1.0. The phylogenetic tree reconstructed with the concatenated plastid DNA matrix showed two clades (Fig. 3; Fig. S5). The first clade (94 BS, 1.000 PPS) comprised all the red- and some blueflowered individuals of LA and LT. The second clade (86 BS, 0.890 PPS) included blue-flowered individuals of LA, LM (blue and red) and LF. Concordant results with the two haplotypes were observed visually. This topology was also observed for each plastid marker independently, although the bootstrap support was lower in all cases (Figs S2–S4). recoMbination analySeS No recombination events were invoked across samples, except for a blue-flowered L. arvensis (LA19_B), in which four of the seven recombination methods detected two recombinant regions of 18 nucleotides starting at position 16 from L. arvensis and of 25 nucleotides starting at position 133 from L. foemina. Therefore, this sample was excluded from the phylogenetic tree shown in Figure 2, but it is shown in Supporting Information, Figure S1. anceStral State reconStruction The ER model was the best-fitting model according to AIC both for ML (AIC ER: 0.99956394, SYM: 0.00043601 and ARD: 0.00000005) and for SIMMAP (AIC ER: 0.99956397, SYM: 0.00043601 and ARD: 0.00000002) analysis. Both approaches showed the same evolutionary transition of flower colour: a high probability of blue-flowered plants until the ancestor of the SCA clade, when red-flowered plants probably appeared (Fig. 2). SpecieS deliMitation The ITS phylogenetic reconstructions pointed to six MOTUs (LA_R, LM_R, LF, LM_B, LA_B and LT), which were consistent with the characterized indel positions (Table 2) and with the evolutionary divergence between samples estimated by the K2P model. Mean p-distances between the six MOTUs ranged from 0.0143 to 0.0622 and within each MOTU from 0.0000 to 0.0023 (Table 3). In addition, three of the four methods applied for species delimitation confirmed these clusters. Only the mPTP model differed by identifying an additional MOTU resulting from dividing LA_B in two: one with LA4_B, LA6_B, LA7_B, LA11_B, LA12_B and LA13_B (blue-flowered samples from populations of southwestern Spain), and one with the rest of the LA_B samples. The ABGD method was run with a prior maximum divergence of intraspecific diversity. The number of groups for the recursive partition was six with a prior of 0.008; seven with 0.005 and 0.003; and nine with 0.001. The primary partition was stable on the range of prior values, with six groups corresponding to the same groups obtained with a prior of 0.1. The same groups were supported by DNA barcoding gap analysis. Heat-map analysis showed that sequence dissimilarity was < 1 within each MOTU and 1–5 between samples of different MOTUs (Fig. 4). DNA barcoding gap analysis showed no overlap between intra- and interspecific distances among MOTUs (Fig. 4). Indeed, intraspecific genetic distances were close to 0 within each MOTU, whereas interspecific genetic distances ranged from 0.010 to 0.050 between MOTUs/ species (Supporting Information, Fig. S6). Table 1. Aligned information, parsimony analysis results and evolutionary model used for each plastid DNA partition rps16-trnK rpl32-trnL trnH-psbA Aligned base pairs 711–715 609–611 373 Potentially informative characters 18 14 9 MP trees/tree length 1/61 10/35 1/32 Consistency index 0.97 0.97 0.93 Retention index 0.96 0.98 0.71 Evolutionary model GTR TrN HKY Downloaded from https://academic.oup.com/botlinnean/article/199/2/557/6494517 by Universidad de Sevilla user on 04 November 2022 564 F. J. JIMÉNEZ-LÓPEZ ET AL. © 2022 The Linnean Society of London, Botanical Journal of the Linnean Society, 2022, 199, 557–577 Table 2. Summary of sequence variation in the nuclear ribosomal DNA regions (ITS 1 and ITS2) and plastid DNA partitions (rps16-trnK, rpl32-trnL and trnH-psbA) in the ingroup of the studied Lysimachia species ITS1 ITS2 Site 14 29 46 49 56 57 116 119 120 121 122 123 128 167 195 196 420 421 422 437 442 476 477 478 527 548 564 565 574 591 Type of change: Del Ins Del Consensus: T T C G C G G C G C A T C A T C C A - C G C T A G T C C C C LA_B A * * * * * * * * * * * * T * * * * * * T G A * * A * A * * LA_B (2)†A * * * T * * * * * * * T T * * * * * * T G A * * A T A * * LA_R * C T * * * A * C G T G * * * * - * * T * * * C * * * * T * LF * C * A * * T T A T C C * * * * * C A * * * * * - * * * * * LM_B * C * * * * C * A T C C * * A * * * * * * * * * * * * * * T LM_B (2)‡* C * * * * C * A T C A * * A * * * * * * * * C * * * * * T LM_R * C * * * * C * C G T G * * * G * * * * * * * * * * * * * A LT * * * * * A * * * * * C * T * * * * * * T G A * * A * * * * rps16-trnK rpl32-trnL trnH-psbA Site 113 424 467 468 471 472 473 474 475 498 589 59 173 176 188 333 342 584 40 360 365 Type of change del del del del del Ins Ins Del Consensus T A G A A – – – – G – G T T T – – G T G - H1 * * * * * * * * * * * * * * * * * * * * * H2 G G A */G1GAAAAATT*/C2C G T */T1T C A G Del: delection; Ins: insertion. †(LA4_B, LA6_B, LA7_B, LA11_B- LA13_B). ‡(LM3_B-LM5_B; see Supporting Information, Table S1). Haplotype 1 (H1)/Haplotype 2 (H2) comprise the clade 1/clade 2 samples of the concatenated plastid DNA tree (Fig. 3). 1, only on LF and LM_B samples of H2; 2, only on LA_B samples of H2; Fig. 3, Table S1). Downloaded from https://academic.oup.com/botlinnean/article/199/2/557/6494517 by Universidad de Sevilla user on 04 November 2022 PHYLOGENY OF MEDITERRANEAN LYSIMACHIA 565 © 2022 The Linnean Society of London, Botanical Journal of the Linnean Society, 2022, 199, 557–577 Finally, the results of the SDP also suggest six MOTUs/species for both BI and ML approaches, particularly with Rosenberg’s PAB statistics (Table 4). In general, the six initial MOTUs/species were monophyletic. For both BI and ML, most Intra/ Inter ratios were < 1, indicating that the divergence within each MOTU was low relative to the divergence with the closest MOTU. Values of P ID(Strict) in BI were higher for MOTUs LA_B and LA_R than for the remaining MOTUs, and hence were correctly identified. Rosenberg’s P(AB) values were significant (P < 0.05) for all MOTUs/species in the BI and ML trees. However, Rodrigo’s P(RD) values were significant (P < 0.05) for all MOTUs/species only in the BI tree. divergence tiMe eStiMation Similar time estimates and topologies were obtained using the three different calibration ages and different tree priors for the stem age of tribe Lysimachieae (Fig. 5; Supporting Information, Fig. S7). Divergence time results based on these different strategies were all similar to each other (Table S4). Our primary analyses (Fig. 5) used the full tree with the Yule prior and all Figure 3. Phylogenetic network (1) and maximum-likelihood (ML) (RAxML) (2) tree based on concatenated plastid sequence data (trnH-psbA, rps16-trnK and rpl32-trnL) of Lysimachia arvensis, L. monelli and related taxa (see Table 1). Main clades in trees from ML (RAxML) and Bayesian analysis (MrBayes) are identical. Bootstrap values > 50% and posterior probabilities of the Bayesian approach (BS/PPS) are given above the branches. Samples of L. arvensis from mixed populations are highlighted in bold. Downloaded from https://academic.oup.com/botlinnean/article/199/2/557/6494517 by Universidad de Sevilla user on 04 November 2022 572 F. J. JIMÉNEZ-LÓPEZ ET AL. © 2022 The Linnean Society of London, Botanical Journal of the Linnean Society, 2022, 199, 557–577 at the base, acute in the apex, superior in whorls of three leaves of 15–20 × 3–5 mm, elliptical, each one with a long pedicelled flower; pedicels of 25–35 mm, generally larger than the leaves. These branches come from the garden of George Clifford III (Hartekamp Garden, Holland). = Anagallis linifolia L. Sp. Pl. ed. 2, 1: 212 (1762) ≡ Anagallis monelli subsp. linifolia Maire in Jahandiez & Maire, Cat. Pl. Maroc.: 562 (1934) Ind. Loc.: ‘Habitat in Lusitania, Hispania. Claud Alstroemer’ Lectotype designated here: ‘linifolia [m. Linnaeus] A [Alstroemer]’: Herb. Linn. 208.4 (LINN-ML) The lectotype is formed by a single, apparently, annual plant 10 cm high, erect, with three short branches with fruits in the upper half, and other branches emerging near to the axonomorph root; stem with very short internodes, surpassed by the leaves; leaves of 6–10 × 0.1–0.2 mm, narrowly linear, truncated at the base, obtuse; recurved and rigid pedicels in well-developed fruits; fruits shorter than the calyx. The characters of this plant coincide with those indicated by Linnaeus when he described the species, so there is no doubt that this plant is the same as Linnaeus saw. The indication ‘Lusitania’ could have been made by Linnaeus from a plant from Portugal described by Tournefort and cited as a synonym. - Lysimachia collina (Schousb.) FJ.Jiménez- López, comb. nov. ≡ Anagallis collina Schousb., Iagttag. Vextrig. Marokko.: 78. (1800), basion. ≡ Anagallis monelli subsp. collina (Schousb.) H.Lindb. in Acta Soc. Sci. Fenn., ser. B, Opera Biol. 1(2): 115. (1932). ≡ Anagallis monelli var. collina (Schousb.) Pau in Mem. Real. Soc. Esp. Hist. Nat. 12 (5): 360 (1924) ≡ Anagallis linifolia var. collina (Schousb.) Ball in J. Linn. Soc. 16: 562 (1878). ≡ Anagallis linifolia f. rubriflora Batt. in Battandier & Trabut, Fl. Algérie, Dicot.: 723 (1890), [syn. subst.]. Ind. loc.: ‘Frequens ocurrit in collibus aridis provinciae Haha’ Red plants of L. monelli should be referred to L. collina because plants with this colour were described by Schousboe in 1800 as Anagallis collina. Due to Anagallis now being included in Lysimachia, we make a new combination. Lectotype designated here: Mogador [Haha] Schousboe [m. Schousboe]: C 10001180. The type material consists of a single woody stem of 15 cm length, without leaves in the lower half, deterMination key 1. Perennial plants, rarely annual herbs with a single stem; stem nodes with (two) three or four (five) verticillated leaves; fertile nodes with a single flower in the axil of each leaf; flowers with the corolla (14) 16–25 mm in diameter, styles 3–4 mm in length ............................................................................................2 - Annual plants; nodes generally with opposite leaves, rarely with three verticillated leaves; fertile nodes with two flowers, one in each leaf axil, rarely with a single flower; flowers with corolla 3–12 (14) mm in diameter; styles 1.0–2.5 mm in length ........................................................................................................... 3 2. Flowers with orange or red corolla................................................................................ Lysimachia collina - Flowers with blue corolla ............................................................................................. Lysimachia monelli 3. Plants generally compact; internodes generally shorter than the leaves; leaves, at least the upper ones, erect-patent, lanceolate or elliptic-lanceolate, acute; fruit pedicels generally shorter than internodes; flowers with blue corolla, with elliptical lobes, strongly serrated in the upper half of the margin, covered with hairs with (three) four (five) cells, the terminal elliptical or sub-cylindrical, about the size and shape of the adjacent cell, sometimes glabrous .................................................................... Lysimachia foemina - Plants generally graceful; internodes often longer than leaves; leaves patent, ovate or ovate-lanceolate, obtuse or subacute; fruit pedicels generally larger than internodes; flowers with blue or orange-red corolla, with lobes broadly ovate, denticulate, with the margin densely covered with hairs with three cells, the terminal globose, larger than the underlying cell ......................................................................................... 4 4. Flowers blue .................................................................................................................................................... 5 - Flowers orange or red .................................................................................................. Lysimachia arvensis 5. Small plants, generally < 10 cm in length; root neck generally covered by secondary roots; ovate leaves; flowers with blue or pale blue corolla, often surpassed by calyx; styles 1.0–1.5 mm in length; .................... .................................................................................................................................... Lysimachia talaverae - Generally large plants, up to 60 cm in length; bare root neck without secondary roots; ovate-lanceolate leaves, at least in the upper half of the stem; blue corolla, usually longer than the calyx; styles 2.0–2.5 mm in length .................................................................................................................... Lysimachia loeflingii Downloaded from https://academic.oup.com/botlinnean/article/199/2/557/6494517 by Universidad de Sevilla user on 04 November 2022 PHYLOGENY OF MEDITERRANEAN LYSIMACHIA 573 © 2022 The Linnean Society of London, Botanical Journal of the Linnean Society, 2022, 199, 557–577 branched in the upper half; each branch short, with some oval–lanceolate leaves, acute, shorter than the pedicels and the majority of the flowers in anthesis; isolectotype C 10001181. ACKNOWLEDGEMENTS We thank the following people for their contribution to sample collection and data analysis: F. Balao, R. Berjano, M. Buide, I. Casimiro-Soriguer, J. A. Mejías, M. A. Ortiz and S. Talavera. We thank S. Talavera for comments on the manuscript. This work was funded by FEDER funds from the Ministerio de Ciencia e Innovación and MINECO (CGL2008-02531-E & CGL2012-33270 to M.A. & CGL2015-63827 to M.A. & P.L.O.), the fellowship BES-2013-062859 and EEBB-I-17-12129 from MINECO and 88887.194785/2018-00 from CAPES_PRINT (Brazil) to F.J.J.-L. and the VPPI-US to M.T. Much of the analysis and partial evaluation of the results of the present study was carried out during the predoctoral stay (EEBB-I-17-12129) of the first author at the RBG Herbarium of Kew under Paul Wilkin’s supervision. We thank the Herbarium of the University of Sevilla (Her SEV, CITIUS) for their collaboration and the use of its new DNA Bank, and Royal Botanic Gardens, Kew. REFERENCES Anderberg AA, Manns U, Källersjö M. 2007. Phylogeny and floral evolution of the Lysimachieae (Ericales, Myrsinaceae). Evidence from ndhF sequence data. Willdenowia 37: 407–421. Arcangeli G. 1894. Compendio della Flora Italiana, Vol. 2. Turin: E. Loescher, 456. Arista M, Talavera M, Berjano R, Ortiz PL. 2013. Abiotic factors may explain the geographical distribution of flower colour morphs and the maintenance of colour polymorphism in the scarlet pimpernel. Journal of Ecology 101: 1613–1622. Armbruster WS. 1997. Exaptations link the evolution of plant–herbivore and plant–pollinator interactions: a phylogenetic inquiry. Ecology 78: 1661–1674. Armbruster WS. 2002. Can indirect selection and genetic context contribute to trait diversification? A transition‐probability study of blossom-colour evolution in two genera. Journal of Evolutionary Biology 15: 468–486. Aymerich P, Sáez Ll. 2015. Comentaris i precisions previs a la Checklist de la flora de Catalunya (Nord-Est de la península Ibèrica). Orsis 29: 23–90. Ball J. 1878. Introductory observations to the Spicilegium Florae Maroccanae. Journal of the Linnean Society 16: 377–772. Baum DA, Shaw KL, Hoch PC, Stephenson AG. 1995. Genealogical perspectives on the species problem, Experimental and molecular approaches to plant biosystematics. St. Louis Missouri Botanical Garden, 289–303. Barrelier J. 1714. Plantae per Galliam, Hispaniam et Italiam. Ed. Antoine de Joussieu. Parisii. Bouckaert R, Vaughan TG, Barido-Sottani J, Duchêne S, Fourment M, Gavryushkina A, Heled J, Jones G, Kühnert D, De Maio N, Matschiner M, Mendes FK, Müller NF, Ogilvie HA, du Plessis L, Popinga A, Rambaut A, Rasmussen D, Siveroni I, Suchard MA, Wu CH, Xie D, Zhang C, Stadler T, Drummond AJ. 2019. BEAST 2.5: an advanced software platform for Bayesian evolutionary analysis. PLoS Computational Biology 15: e1006650. Carlson JE, Holsinger KE. 2015. Extrapolating from local ecological processes to genus-wide patterns in colour polymorphism in South African Protea. Proceedings of the Royal Society B: Biological Sciences 282: 20150583. Chase MW, Hills HG. 1991. Silica gel: an ideal material for field preservation of leaf samples for DNA studies. Taxon 40: 215–220. Darriba D, Taboada GL, Doallo R, Posada D. 2012. jModelTest 2: more models, new heuristics and parallel computing. Nature Methods 9: 772. Degnan JH, Rosenberg NA. 2009. Gene tree discordance, phylogenetic inference and the multispecies coalescent. Trends in Ecology & Evolution 24: 332–340. Dörfler I. 1903. Herbarium Normale, Vol. 45. Vienna: O. Hensel, 143. Drummond AJ, Ho SY, Phillips MJ, Rambaut A. 2006. Relaxed phylogenetics and dating with confidence. PLoS Biology 4: e88. Duminil J, Di Michele M. 2009. Plant species delimitation: a comparison of morphological and molecular markers. Plant Biosystems 143: 528–542. Dyer RA. 1963. Primulaceae. In: Dyer RA, Codd IE, Rycroft HB. eds. Flora of southern Africa. Pretoria: Government Printer, 9–14. Dyer AG, Boyd-Gerny S, Shrestha M, Lunau K, Garcia JE, Koethe S, Wong BB. 2016. Innate colour preferences of the Australian native stingless bee Tetragonula carbonaria Sm. Journal of Comparative Physiology. A, Neuroethology, Sensory, Neural, and Behavioral Physiology 202: 603–613. Ellis TJ, Field DL. 2016. Repeated gains in yellow and anthocyanin pigmentation in flower colour transitions in the Antirrhineae. Annals of Botany 117: 1133–1140. Escudero M, Balao F, Martín-Bravo S, Valente L, Valcárcel V. 2018. Is the diversification of Mediterranean Basin plant lineages coupled to karyotypic changes? Plant Biology (Stuttgart, Germany) 20 (Suppl. 1): 166–175. Farris JS, Källersjö M, Kluge AG, Bult C. 1994. Testing significance of congruence. Cladistics 10: 315–320. Ferguson LF. 1972. Anagallis L. In: Tutin TG, Heywood VH, Burges NA, Moore DM, Valentine DH, Walters SM, Downloaded from https://academic.oup.com/botlinnean/article/199/2/557/6494517 by Universidad de Sevilla user on 04 November 2022 574 F. J. JIMÉNEZ-LÓPEZ ET AL. © 2022 The Linnean Society of London, Botanical Journal of the Linnean Society, 2022, 199, 557–577 Webb DA. eds. Flora Europea 3, Diapensiaceae-Myoporaceae. Cambridge: Cambridge University Press, 28–29. Fiz-Palacios O, Valcárcel V. 2013. From Messinian crisis to Mediterranean climate: a temporal gap of diversification recovered from multiple plant phylogenies. Perspectives in Plant Ecology, Evolution and Systematics 15: 130–137. Frey FM. 2004. Opposing natural selection from herbivores and pathogens may maintain floral-color variation in Claytonia virginica (Portulacaceae). Evolution; International Journal of Organic Evolution 58: 2426–2437. Freyre R, Griesbach RJ. 2004. Inheritance of flower color in Anagallis monelli L. HortScience 39: 1220–1223. Friis EM. 1985. Angiosperm fruits and seeds from the Middle Miocene of Jutland (Denmark). Det Kongelige Danske Videnskaberne Selskab 24: 1–165. García Pérez L, Talavera S, Ortiz PL, Arista M. 1997. Números cromosomáticos de plantas occidentales 734-737. Anales del Jardín Botánico de Madrid 55: 344–345. Gibbs PE, Talavera S. 2001. Breeding system studies in three species of Anagallis (Primulaceae): self-incompatibility and reduced female fertility in A. monelli L. Annals of Botany 88: 139–144. de Halácsy E. 1904. Conspectus Florae Graecae, Vol. 3. Leipzig: Engelmann. Hamilton MB. 1999. Four primer pairs for the amplification of chloroplast intergenic regions with intraspecific variation. Molecular Ecology 8: 521–523. Hanusch M, Ortiz EM, Patiño J, Schaefer H. 2020. Biogeography and integrative taxonomy of Epipterygium (Mniaceae, Bryophyta). Taxon 69: 1150–1171. Heled J, Drummond AJ. 2012. Calibrated tree priors for relaxed phylogenetics and divergence time estimation. Systematic Biology 61: 138–149. Ho SY, Phillips MJ. 2009. Accounting for calibration uncertainty in phylogenetic estimation of evolutionary divergence times. Systematic Biology 58: 367–380. Hoffmannsegg JCG, Link JHF. 1813-1820. Flore Portugaise, 1 (11). Berlin: Charles Fréderic Amelang, 325 Tab. 64. Hopkins R. 2013. Reinforcement in plants. The New Phytologist 197: 1095–1103. Hopkins R, Rausher MD. 2012. Pollinator-mediated selection on flower color allele drives reinforcement. Science (New York, N.Y.) 335: 1090–1092. Huelsenbeck JP, Nielsen R, Bollback JP. 2003. Stochastic mapping of morphological characters. Systematic Biology 52: 131–158. Huxley JS. 1955. Morphism and evolution. Heredity 9: 1–51. Huzioka K, Takahasi E. 1970. The Eocene flora of the Ube coal-field, southwest Honshu, Japan. Journal of the Mining College, Akita University, Series A: Mining Geology 4: 1–88. Ishikura N. 1981. Flavonoids in the petal cells of Anagallis arvensis f. coerulea containing a blue crystalline anthocyanin. Zeitschrift für Pflanzenphysiologie 103: 469–473. Jahandiez E, Maire R. 1934. Catalogue des plantes du Maroc, Vol 3. Algiers: Imprimerie Minerva, 563. Jiménez-López FJ, Matas L, Arista M, Ortiz PL. 2020a. Flower colour segregation and flower discrimination under the bee vision model in the polymorphic Lysimachia arvensis. Plant Biosystems 154: 535–543. Jiménez-López FJ, Ortiz MA, Berjano R, Talavera S, Terrab A. 2016. High population genetic substructure in Hypochaeris leontodontoides (Asteraceae), an endemic rupicolous species of the Atlas Mountains in NW Africa. Alpine Botany 126: 73–85. Jiménez-López FJ, Ortiz PL, Talavera M, Arista M. 2020b. Reproductive assurance maintains red-flowered plants of Lysimachia arvensis in Mediterranean populations despite inbreeding depression. Frontiers in Plant Science 11: 563110. Jiménez-López FJ, Ortiz PL, Talavera M, Pannell JR, Arista M. 2020c. The role of lateral and vertical herkogamy in the divergence of the blue- and red-flowered lineages of Lysimachia arvensis. Annals of Botany 125: 1127–1135. Jiménez-Moreno G, Fauquette S, Suc JP. 2010. Miocene to Pliocene vegetation reconstruction and climate estimates in the Iberian Peninsula from pollen data. Review of Palaeobotany and Palynology 162: 410–415. Kapli P, Lutteropp S, Zhang J, Kobert K, Pavlidis P, Stamatakis A, Flouri T. 2017. Multi-rate Poisson tree processes for single-locus species delimitation under maximum likelihood and Markov chain Monte Carlo. Bioinformatics (Oxford, England) 33: 1630–1638. Kay KM, Reeves PA, Olmstead RG, Schemske DW. 2005. Rapid speciation and the evolution of hummingbird pollination in neotropical Costus subgenus Costus (Costaceae): evidence from nrDNA ITS and ETS sequences. American Journal of Botany 92: 1899–1910. Kimura M. 1980. A simple method for estimating evolutionary rates of base substitutions through comparative studies of nucleotide sequences. Journal of Molecular Evolution 16: 111–120. Kirkpatrick M. 2000. Reinforcement and divergence under assortative mating. Proceedings of the Royal Society B: Biological Sciences 267: 1649–1655. Kollmann F, Feinbrun N. 1968. A cyto-taxonomic study in Palestinian Anagallis arvensis L. Notes of the Royal Botanic Garden Edinburgh 27: 173–186. Kress A. 1969. Cytotaxonomic studies on Primulaceae. Phyton 13: 211–225. Krijgsman W, Stoica M, Vasiliev I, Popov VV. 2010. Rise and fall of the Paratethys Sea during the Messinian Salinity Crisis. Earth and Planetary Science Letters 290: 183–191. Kumar S, Stecher G, Li M, Knyaz C, Tamura K. 2018. MEGA X: molecular evolutionary genetics analysis across computing platforms. Molecular Biology and Evolution 35: 1547–1549. Levin DA, Brack ET. 1995. Natural selection against white petals in Phlox. Evolution; International Journal of Organic Evolution 49: 1017–1022. Lihová J, Fuertes Aguilar J, Marhold K, Nieto Feliner G. 2004. Origin of the disjunct tetraploid Cardamine amporitana (Brassicaceae) assessed with nuclear and chloroplast DNA sequence data. American Journal of Botany 91: 1231–1242. Linnaeus C. 1753. Species Plantarum, Vol. 1. Stockholm: Impensis Laurentii Salvii, Holmiae, 148–149. Lunau K, Maier EJ. 1995. Innate colour preferences of flower visitors. Journal of Comparative Physiology A 177: 1–19. Downloaded from https://academic.oup.com/botlinnean/article/199/2/557/6494517 by Universidad de Sevilla user on 04 November 2022 PHYLOGENY OF MEDITERRANEAN LYSIMACHIA 575 © 2022 The Linnean Society of London, Botanical Journal of the Linnean Society, 2022, 199, 557–577 Magallón S, Hilu KW, Quandt D. 2013. Land plant evolutionary timeline: gene effects are secondary to fossil constraints in relaxed clock estimation of age and substitution rates. American Journal of Botany 100: 556–573. Manns U, Anderberg AA. 2005. Molecular phylogeny of Anagallis (Myrsinaceae) based on ITS, trnL-F, and ndhF sequence data. International Journal of Plant Sciences 166: 1019–1028. Manns U, Anderberg AA. 2007a. Character evolution in Anagallis (Myrsinaceae) inferred from morphological and molecular data. Systematic Botany 32: 166–179. Manns U, Anderberg AA. 2007b. Relationships of Anagallis foemina and A. arvensis (Myrsinaceae): new insights inferred from DNA sequence data. Molecular Phylogenetics and Evolution 45: 971–980. Manns U, Anderberg AA. 2009. New combinations and names in Lysimachia (Myrsinaceae) for species of Anagallis, Pelletiera and Trientalis. Willdenowia 39: 49–54. Marsden-Jones EM. 1935. The genetic of Anagallis arvensis Linn. and Anagallis foemina Mill. Proceedings of the Linnean Society of London 147: 105–106. Marsden-Jones EM, Weiss FE. 1938. The essential differences between Anagallis arvensis Linn. and Anagallis foemina Mill. Proceedings of the Linnean Society of London 150: 146–155. Martin DP, Murrell B, Golden M, Khoosal A, Muhire B. 2015. RDP4: detection and analysis of recombination patterns in virus genomes. Virus Evolution 1: vev003. Martins L, Oberprieler C, Hellwig FH. 2003. A phylogenetic analysis of Primulaceae s.l. based on internal transcribed spacer (ITS) DNA sequence data. Plant Systematics and Evolution 237: 75–85. Masters BC, Fan V, Ross HA. 2011. Species delimitation – a Geneious plugin for the exploration of species boundaries. Molecular Ecology Resources 11: 154–157. Miller MA, Pfeiffer W, Schwartz T. 2010. Creating the CIPRES Science Gateway for inference of large phylogenetic trees. In: Proceedings of the Gateway Computing Environments Workshop (GCE), New Orleans, ACM Journals, 1–8. Monein A, Atta II, Shehata AA. 2003. On the delimitation of Anagallis arvensis L. (Primulaceae) 1. Evidence based on micromorphological characters, palynological features and karyological studies. Pakistan Journal of Biological Sciences 6: 29–35. Narbona E, Wang H, Ortiz PL, Arista M, Imbert E. 2018. Flower colour polymorphism in the Mediterranean Basin: occurrence, maintenance and implications for speciation. Plant Biology 20: 8–20. Nordborg M. 2001. Coalescent theory. In: Balding DJ, Bishop M, Cannings C, eds. Handbook of statistical genetics. Chichester: John Wiley & Sons, 179–212. O’Kane SL, Schaal BA, Al-Shehbaz IA. 1996. The origins of Arabidopsis suecica (Brassicaceae) as indicated by nuclear rDNA sequences. Systematic Botany 21: 559–566. Oh IC, Anderberg AL, Schönenberger J, Anderberg AA. 2008. Comparative seed morphology and character evolution in the genus Lysimachia (Myrsinaceae) and related taxa. Plant Systematics and Evolution 271: 177–197. Ortiz PL, Berjano R, Talavera M, Rodríguez-Zayas L, Arista M. 2015. Flower colour polymorphism in Lysimachia arvensis: how is the red morph maintained in Mediterranean environments? Perspectives in Plant Ecology, Evolution and Systematics 17: 142–150. Otto SP, Whitton J. 2000. Polyploid incidence and evolution. Annual Review of Genetics 34: 401–437. Padilla-García N, Rojas-Andrés BM, López-González N, Castro M, Castro S, Loureiro J, Albache DC, Machon N, Martínez-Ortega MM. 2018. The challenge of species delimitation in the diploid–polyploid complex Veronica subsection Pentasepalae. Molecular Phylogenetics and Evolution 119: 196–209. Paradis E, Claude J, Strimmer K. 2004. APE: analyses of phylogenetics and evolution in R language. Bioinformatics (Oxford, England) 20: 289–290. Pastor J. 1992. Atlas cromosómico de la flora vascular de Andalucia Occidental. Seville: Universidad de Sevilla, 283–284. Pau C. 1915. About Anagallis monelli L. Boletín de la Sociedad Aragonesa de Ciencias Naturales 14: 100–107. Petit C, Bretagnolle F, Felber F. 1999. Evolutionary consequences of diploid–polyploid hybrid zones in wild species. Trends in Ecology & Evolution 14: 306–311. Postigo Mijarra JM, Barrón E, Gómez Manzaneque F, Morla C. 2009. Floristic changes in the Iberian Peninsula and Balearic Islands (south-west Europe) during the Cenozoic. Journal of Biogeography 36: 2025–2043. Puillandre N, Lambert A, Brouillet S, Achaz G. 2012. ABGD, automatic barcode gap discovery for primary species delimitation. Molecular Ecology 21: 1864–1877. Pujadas A. 1997. Anagallis L. In: Castroviejo S, Aedo C, Laínz M, Morales R, Muñoz Garmendia F, Nieto Feliner G, Paiva J. eds. Flora iberica, V. Ebenaceae-Saxifragaceae. Madrid: Real Jardín Botánico, CSIC, 57–62. Quintana A, Freyre R, Davis TM, Griesbach RJ. 2008. Genetic studies of flower color in Anagallis monelli L. HortScience 43: 1680–1685. Rambaut A, Drummond AJ, Xie D, Baele G, Suchard MA. 2018. Posterior summarization in Bayesian phylogenetics using Tracer 1.7. Systematic Biology 67: 901–904. Rausher MD. 2008. Evolutionary transitions in floral color. International Journal of Plant Sciences 169: 7–21. Revell LJ. 2012. Phytools: an R package for phylogenetic comparative biology (and other things). Methods in Ecology and Evolution 3: 217–223. Rieseberg LH, Raymond O, Rosenthal DM, Lai Z, Livingstone K, Nakazato T, Durphy JL, Schwarzbach AE, Donovan LA, Lexer C. 2003. Major ecological transitions in wild sunflowers facilitated by hybridization. Science 301: 1211–1216. Rodríguez-Sánchez F, Pérez-Barrales R, Ojeda F, Vargas P, Arroyo J. 2008. The Strait of Gibraltar as a melting pot for plant biodiversity. Quaternary Science Reviews 27: 2100–2117. Rodrigo A, Bertels F, Heled J, Noder R, Shearman H, Tsai P. 2008. The perils of plenty: what are we going to do with all these genes? Philosophical Transactions of the Royal Society B: Biological Sciences 363: 3893–3902. Downloaded from https://academic.oup.com/botlinnean/article/199/2/557/6494517 by Universidad de Sevilla user on 04 November 2022 576 F. J. JIMÉNEZ-LÓPEZ ET AL. © 2022 The Linnean Society of London, Botanical Journal of the Linnean Society, 2022, 199, 557–577 Ronquist F, Teslenko M, van der Mark P, Ayres DL, Darling A, Höhna S, Larget B, Liu L, Suchard MA, Huelsenbeck JP. 2012. MrBayes 3.2: efficient Bayesian phylogenetic inference and model choice across a large model space. Systematic Biology 61: 539–542. Rosenberg NA. 2007. Statistical tests for taxonomic distinctiveness from observations of monophyly. Evolution 61: 317–323. Sampaio G. 1900. Studies on the flora of Porto’s surroundings. I. Primulaceae (Vent.) Annales des Sciences Naturales 6: 51–62. Sánchez-Cabrera M, Jiménez López FJ, Narbona E, Arista M, Ortiz PL, Romero-Campero FJ, Ramanauskas K, Igic B, Fuller AA, Whittall JB. 2021. Changes at a critical branchpoint in the anthocyanin biosynthetic pathway underlie the blue to orange flower color transition in Lysimachia arvensis. Frontiers in Plant Science 12: 633979. Schousboe P. 1800. Iagttagelser over Vextriget i Marokko. [Observations on the Plant Kingdom in Morocco]. Copenhagen: Forfatteren hos K. H. Seidelin. Semerikova SA, Semerikov VL. 2016. Phylogeny of firs (genus Abies, Pinaceae) based on multilocus nuclear markers (AFLP). Russian Journal of Genetics 52: 1164–1175. Servedio MR, Van Doorn GS, Kopp M, Frame AM, Nosil P. 2011. Magic traits in speciation: ‘magic’ but not rare? Trends in Ecology & Evolution 26: 389. Shaw J, Lickey EB, Schilling EE, Small RL. 2007. Comparison of whole chloroplast genome sequences to choose noncoding regions for phylogenetic studies in angiosperms: the tortoise and the hare III. American Journal of Botany 94: 275–288. Sobel JM, Streisfeld MA. 2015. Strong premating reproductive isolation drives incipient speciation in Mimulus aurantiacus. Evolution; International Journal of Organic Evolution 69: 447–461. Soltis DE, Soltis PS. 1998. Choosing an approach and an appropriate gene for phylogenetic analysis. In Soltis DE, Soltis PS, Doyle JJ, eds. Molecular systematics of plants II. DNA sequencing. New York: Springer, 1–42. Soltis DE, Visger CJ, Soltis PS. 2014. The polyploidy revolution then…and now: Stebbins revisited. American Journal of Botany 101: 1057–1078. Soltis PS, Marchant DB, Van de Peer Y, Soltis DE. 2015. Polyploidy and genome evolution in plants. Current Opinion in Genetics & Development 35: 119–125. Souissi A, Gagnaire PA, Bonhomme F, Bahri-Sfar L. 2017. Introgressive hybridization and morphological transgression in the contact zone between two Mediterranean Solea species. Ecology and Evolution 7: 1394–1402. Stamatakis A. 2014. RAxML version 8: a tool for phylogenetic analysis and post-analysis of large phylogenies. Bioinformatics (Oxford, England) 30: 1312–1313. Stecher G, Tamura K, Kumar S. 2020. Molecular evolutionary genetics analysis (MEGA) for macOS. Molecular Biology and Evolution 37: 1237–1239. Strauss SY, Whittall JB. 2006. Non-pollinator agents of selection on floral traits. In: Harder LD, Barrett SCH, eds. Ecology and evolution of flowers. Oxford: Oxford University Press, 120–138. Suc JP. 1984. Origin and evolution of the Mediterranean vegetation and climate in Europe. Nature 307: 429–432. Suchard MA, Lemey P, Baele G, Ayres DL, Drummond AJ, Rambaut A. 2018. Bayesian phylogenetic and phylodynamic data integration using BEAST 1.10. Virus Evolution 4: vey016. Šveřepová G. 1972. On the cytotaxonomy of the genus Anagallis L. (K cytotaxonomi rodu Anagallis L.). Preslia 44: 219–226. Swofford DL. 2003. PAUP*. Phylogenetic analysis using parsimony (*and other methods), Version 4. Sunderland: Sinauer Associates. Symonds VV, Lloyd AM. 2003. An analysis of microsatellite loci in Arabidopsis thaliana: mutational dynamics and application. Genetics 165: 1475–1488. Takahashi Y, Takakura KI, Kawata M. 2016. Spatial distribution of flower color induced by interspecific sexual interaction. PLoS One 11: e0164381. Talavera S, Gibbs PE, Fernández-Piedra MP, Ortiz- Herrera MA. 2001. Genetic control of self-incompatibility in Anagallis monelli (Primulaceae: Myrsinaceae). Heredity 87: 589–597. Thompson JD, Gibson T, Higgins DG. 2002. Multiple sequence alignment using ClustalW and ClustalX. Current Protocols in Bioinformatics 1: 2–3. Tournefort JP. 1719. Instituciones rei herbariae, 3rd edn, Vol. 1. Paris: Typographia Regia. Valdés B. 1970. Números cromosómicos de algunas plantas españolas. Boletin de la Real Sociedad Española de Historia Natural (Biología) 68: 193–197. Vargas P. 2003. Molecular evidence for multiple diversification patterns of alpine plants in Mediterranean Europe. Taxon 52: 463–476. Viruel J, Forest F, Paun O, Chase MW, Devey D, Couto RS, Segarra-Moragues JG, Catalán P, Wilkin P. 2018. A nuclear Xdh phylogenetic analysis of yams (Dioscorea: Dioscoreaceae) congruent with plastid trees reveals a new Neotropical lineage. Botanical Journal of the Linnean Society 187: 232–246. Viruel J, Segarra-Moragues JG, Raz L, Forest F, Wilkin P, Sanmartín I, Catalán P. 2016. Late Cretaceous– Early Eocene origin of yams (Dioscorea, Dioscoreaceae) in the Laurasian Palaearctic and their subsequent Oligocene– Miocene diversification. Journal of Biogeography 43: 750–762. Volkov R, Borisjuk ANV, Panchuk II, Schweizer D, Hemleben V. 1999. Elimination and rearrangement of parental rDNA in the allotetraploid Nicotiana tabacum. Molecular Biology and Evolution 16: 311–320. Wakeley J. 2009. Coalescent theory: an introduction. Greenwood Village: Roberts and Company Publishers. Wang H, Conchou L, Bessière JM, Cazals G, Schatz B, Imbert E. 2013. Flower color polymorphism in Iris lutescens (Iridaceae): biochemical analyses in light of plant–insect interactions. Phytochemistry 94: 123–134. Downloaded from https://academic.oup.com/botlinnean/article/199/2/557/6494517 by Universidad de Sevilla user on 04 November 2022 PHYLOGENY OF MEDITERRANEAN LYSIMACHIA 577 © 2022 The Linnean Society of London, Botanical Journal of the Linnean Society, 2022, 199, 557–577 Wang J, Tian L, Lee HS, Chen ZJ. 2006. Nonadditive regulation of FRI and FLC loci mediates flowering-time variation in Arabidopsis allopolyploids. Genetics 173: 965–974. Wendel JF, Doyle JJ. 1998. Phylogenetic incongruence: window into genome history and molecular evolution. In: Soltis DE, Soltis PS, Doyle JJ, eds. Molecular systematics of plants II. DNA sequencing. New York: Springer, 265–296. Wessinger CA, Rausher MD. 2012. Lessons from flower colour evolution on targets of selection. Journal of Experimental Botany 63: 5741–5749. White TJ, Bruns T, Lee S, Taylor J. 1990. Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. In: Innis MA, Gelfand DH, Sninsky JJ, White TJ, eds. PCR protocols: a guide to methods and applications. New York: Academic Press, 315–322. Wickham H. 2011. ggplot2. Wiley Interdisciplinary Reviews: Computational Statistics 3: 180–85. Willkomm HM. 1870. Primulaceae. In: Willkomm HM, Lange JM, eds. Prodromus Florae Hispanicae, 2. Stuttgart: E. Schweizerbart (E. Koch), 635–650. Wilson P, Wolfe AD, Armbruster WS, Thomson JD. 2007. Constrained lability in floral evolution: counting convergent origins of hummingbird pollination in Penstemon and Keckiella. The New Phytologist 176: 883–890. Wood TE, Takebayashi N, Barker MS, Mayrose I, Greenspoon PB, Rieseberg LH. 2009. The frequency of polyploid speciation in vascular plants. Proceedings of the National Academy of Sciences USA 106: 13875–13879. Yan HF, Zhang CY, Anderberg AA, Hao G, Ge XJ, Wiens JJ. 2018. What explains high plant richness in East Asia? Time and diversification in the tribe Lysimachieae (Primulaceae). The New Phytologist 219: 436–448. Yesson C, Toomey NH, Culham A. 2009. Cyclamen: time, sea and speciation biogeography using a temporally calibrated phylogeny. Journal of Biogeography 36: 1234–1252. Zanne AE, Tank DC, Cornwell WK, Eastman JM, Smith SA, FitzJohn RG, McGlinn DJ, O’Meara BC, Moles AT, Reich PB, Royer DL, Soltis DE, Stevens PF, Westoby M, Wright IJ, Aarssen L, Bertin RI, Calaminus A, Govaerts R, Hemmings F, Leishman MR, Oleksyn J, Soltis PS, Swenson NG, Warman L, Beaulieu JM. 2014. Three keys to the radiation of angiosperms into freezing environments. Nature 506: 89–92. SUPPORTING INFORMATION Additional Supporting Information may be found in the online version of this article at the publisher’s web-site: Figure S1. Maximum clade credibility tree based on ITS sequences including a recombinant sample of Lysimachia arvensis (LA19_B). Figure S2. Phylogenetic network and maximum-likelihood tree of Lysimachia based on rps16-trnK plastid sequence data. Figure S3. Phylogenetic network and maximum-likelihood tree of Lysimachia based on rpl32-trnL plastid sequence data. Figure S4. Phylogenetic network and maximum-likelihood tree of Lysimachia based on psbA-trnH plastid sequence data. Figure S4. Maximum clade credibility tree based on three combined plastid sequences dataset. Figure S6. Frequency distributions of pairwise Kimura two-parameter distances. Figure S7. Maximum clade credibility tree of tribe Lysimachieae. Table S1. Population information of samples. Table S2. Condition and information of primer combinations tested. Table S3. GenBank accession numbers of sequences of tribe Lysimachieae used. Table S4. Divergence times of major clades based on different tree priors (Yule, birth–death) and calibration points. Table S5. Regression results between the node ages of all clades from the primary tree and the alternative trees. Downloaded from https://academic.oup.com/botlinnean/article/199/2/557/6494517 by Universidad de Sevilla user on 04 November 2022