Integrating top-down and bottom-up approaches to understand the genetic architecture of speciation across a monkeyflower hybrid zone.

Understanding the phenotypic and genetic architecture of reproductive isolation is a long-standing goal of speciation research. In several systems, large-effect loci contributing to barrier phenotypes have been characterized, but such causal connections are rarely known for more complex genetic architectures. In this study, we combine "top-down" and "bottom-up" approaches with demographic modelling toward an integrated understanding of speciation across a monkeyflower hybrid zone. Previous work suggests that pollinator visitation acts as a primary barrier to gene flow between two divergent red- and yellow-flowered ecotypes of Mimulus aurantiacus. Several candidate isolating traits and anonymous single nucleotide polymorphism loci under divergent selection have been identified, but their genomic positions remain unknown. Here, we report findings from demographic analyses that indicate this hybrid zone formed by secondary contact, but that subsequent gene flow was restricted by widespread barrier loci across the genome. Using a novel, geographic cline-based genome scan, we demonstrate that candidate barrier loci are broadly distributed across the genome, rather than mapping to one or a few "islands of speciation." Quantitative trait locus (QTL) mapping reveals that most floral traits are highly polygenic, with little evidence that QTL colocalize, indicating that most traits are genetically independent. Finally, we find little evidence that QTL and candidate barrier loci overlap, suggesting that some loci contribute to other forms of reproductive isolation. Our findings highlight the challenges of understanding the genetic architecture of reproductive isolation and reveal that barriers to gene flow other than pollinator isolation may play an important role in this system.

Widespread selection and gene flow shape the genomic landscape during a radiation of monkeyflowers.

Speciation genomic studies aim to interpret patterns of genome-wide variation in light of the processes that give rise to new species. However, interpreting the genomic "landscape" of speciation is difficult, because many evolutionary processes can impact levels of variation. Facilitated by the first chromosome-level assembly for the group, we use whole-genome sequencing and simulations to shed light on the processes that have shaped the genomic landscape during a radiation of monkeyflowers. After inferring the phylogenetic relationships among the 9 taxa in this radiation, we show that highly similar diversity (π) and differentiation (FST) landscapes have emerged across the group. Variation in these landscapes was strongly predicted by the local density of functional elements and the recombination rate, suggesting that the landscapes have been shaped by widespread natural selection. Using the varying divergence times between pairs of taxa, we show that the correlations between FST and genome features arose almost immediately after a population split and have become stronger over time. Simulations of genomic landscape evolution suggest that background selection (BGS; i.e., selection against deleterious mutations) alone is too subtle to generate the observed patterns, but scenarios that involve positive selection and genetic incompatibilities are plausible alternative explanations. Finally, tests for introgression among these taxa reveal widespread evidence of heterogeneous selection against gene flow during this radiation. Combined with previous evidence for adaptation in this system, we conclude that the correlation in FST among these taxa informs us about the processes contributing to adaptation and speciation during a rapid radiation.

Variation in ecophysiological traits might contribute to ecogeographic isolation and divergence between parapatric ecotypes of Mimulus aurantiacus.

Many forms of reproductive isolation contribute to speciation, and early-acting barriers may be especially important, because they have the first opportunity to limit gene flow. Ecogeographic isolation occurs when intrinsic traits of taxa contribute to disjunct geographic distributions, reducing the frequency of intertaxon mating. Characterizing this form of isolation requires knowledge of both the geographic arrangement of suitable habitats in nature and the identification of phenotypes involved in shaping geographic distributions. In Mimulus aurantiacus, red- and yellow-flowered ecotypes are incompletely isolated by divergent selection exerted by different pollinators. However, these emerging taxa are largely isolated spatially, with a hybrid zone occurring along a narrow region of contact. In order to assess whether responses to abiotic conditions contribute to the parapatric distribution of ecotypes, we measured a series of ecophysiological traits from populations along a transect, including drought sensitivity, leaf area and the concentrations of vegetative flavonoids. In contrast to the abrupt transitions in floral phenotypes, we found that ecophysiological traits exhibited a continuous geographic transition that largely mirrors variation in climatological variables. These traits may impede gene flow across a continuous environmental gradient, but they would be unlikely to result in ecotypic divergence alone. Nevertheless, we found a genetic correlation between vegetative and floral traits, providing a potential link between the two forms of isolation. Although neither barrier appears sufficient to cause divergence on its own, the combined impacts of local adaptation to abiotic conditions and regional adaptation to pollinators may interact to drive discontinuous variation in the face of gene flow in this system.

Geographic cline analysis as a tool for studying genome-wide variation: a case study of pollinator-mediated divergence in a monkeyflower.

A major goal of speciation research is to reveal the genomic signatures that accompany the speciation process. Genome scans are routinely used to explore genome-wide variation and identify highly differentiated loci that may contribute to ecological divergence, but they do not incorporate spatial, phenotypic or environmental data that might enhance outlier detection. Geographic cline analysis provides a potential framework for integrating diverse forms of data in a spatially explicit framework, but has not been used to study genome-wide patterns of divergence. Aided by a first-draft genome assembly, we combined an FCT scan and geographic cline analysis to characterize patterns of genome-wide divergence between divergent pollination ecotypes of Mimulus aurantiacus. FCT analysis of 58 872 SNPs generated via RAD-seq revealed little ecotypic differentiation (mean FCT  = 0.041), although a small number of loci were moderately-to-highly diverged. Consistent with our previous results from the gene MaMyb2, which contributes to differences in flower colour, 130 loci have cline shapes that recapitulate the spatial pattern of trait divergence, suggesting that they may reside in or near the genomic regions that contribute to pollinator isolation. In the narrow hybrid zone between the ecotypes, extensive admixture among individuals and low linkage disequilibrium between markers indicate that most outlier loci are scattered throughout the genome, rather than being restricted to one or a few divergent regions. In addition to revealing the genomic consequences of ecological divergence in this system, we discuss how geographic cline analysis is a powerful but under-utilized framework for studying genome-wide patterns of divergence.

Genomewide variation provides insight into evolutionary relationships in a monkeyflower species complex (Mimulus sect. Diplacus).

PREMISE OF THE STUDY: Evolutionary radiations provide excellent opportunities to study the origins of biodiversity, but rapid divergence and ongoing gene flow make inferring evolutionary relationships among taxa difficult. Consequently, combining morphological and genomic analyses will be necessary to clarify the evolutionary history of radiations. We used an integrative approach to shed light on relationships within a diverse radiation of monkeyflowers (Mimulus section Diplacus) with a controversial taxonomic history. METHODS: We used genomewide single nucleotide polymorphism data and a combination of phylogenetic and population genomic analyses to infer the evolutionary relationships within the group. Tests for hybridization were performed to reveal sources of shared variation, and multivariate analyses of floral trait data were conducted to examine the correspondence between phenotypic and phylogenetic data. KEY RESULTS: We identified four primary clades with evidence for some shared variation among them. We also detected evidence for recent gene flow between closely related subclades and populations. Strong discordance between floral trait and molecular data provides evidence for divergent and convergent phenotypic evolution. CONCLUSIONS: Mimulus section Diplacus has all the hallmarks of a rapid radiation, including diverse taxa that are at different stages of divergence, extensive shared variation among taxa, and complex patterns of phenotypic evolution. Our findings will direct future evolutionary research and have important taxonomic implications that highlight the need for a new revision of section Diplacus.

Population genomics of divergence within an obligate pollination mutualism: Selection maintains differences between Joshua tree species.

PREMISE OF THE STUDY: Speciation is a complex process that can be shaped by many factors, from geographic isolation to interspecific interactions. In Joshua trees, selection from pollinators on style length has been hypothesized to contribute to the maintenance of differentiation between two hybridizing sister species. We used population genomics approaches to measure the extent of genetic differentiation between these species, test whether selection maintains differences between them, and determine whether genetic variants associated with style length show signatures of selection. METHODS: Using restriction-site-associated DNA (RAD)-sequencing, we identified 9516 single nucleotide polymorphisms (SNPs) across the Joshua tree genome. We characterized the genomic composition of trees in a narrow hybrid zone and used genomic scans to search for signatures of selection acting on these SNPs. We used a genome-wide association study to identify SNPs associated with variation in phenotypic traits, including style length, and asked whether those SNPs were overrepresented among the group under selection. KEY RESULTS: The two species were highly genetically differentiated (FST = 0.25), and hybrids were relatively rare in the hybrid zone. Approximately 20% of SNPs showed evidence of selection maintaining divergence. While SNPs associated with style length were overrepresented among those under selection (P << 0.0001), the same was true for SNPs associated with highly differentiated vegetative traits. CONCLUSIONS: The two species of Joshua tree are clearly genetically distinct, and selection is maintaining differences between them. We found that loci associated with differentiated traits were likely to be under selection. However, many traits other than style length appeared to be under selection. Together with the dearth of intermediate hybrids, these findings reveal that these taxa are more strongly diverged than previously suspected and that selection, likely on many targets, is maintaining separation where the two species meet and hybridize.

Divergent selection drives genetic differentiation in an R2R3-MYB transcription factor that contributes to incipient speciation in Mimulus aurantiacus.

Identifying the molecular genetic basis of traits contributing to speciation is of crucial importance for understanding the ecological and evolutionary mechanisms that generate biodiversity. Despite several examples describing putative "speciation genes," it is often uncertain to what extent these genetic changes have contributed to gene flow reductions in nature. Therefore, considerable interest lies in characterizing the molecular basis of traits that actively confer reproductive isolation during the early stages of speciation, as these loci can be attributed directly to the process of divergence. In Southern California, two ecotypes of Mimulus aurantiacus are parapatric and differ primarily in flower color, with an anthocyanic, red-flowered morph in the west and an anthocyanin-lacking, yellow-flowered morph in the east. Evidence suggests that the genetic changes responsible for this shift in flower color have been essential for divergence and have become fixed in natural populations of each ecotype due to almost complete differences in pollinator preference. In this study, we demonstrate that a cis-regulatory mutation in an R2R3-MYB transcription factor results in differential regulation of enzymes in the anthocyanin biosynthetic pathway and is the major contributor to differences in floral pigmentation. In addition, molecular population genetic data show that, despite gene flow at neutral loci, divergent selection has driven the fixation of alternate alleles at this gene between ecotypes. Therefore, by identifying the genetic basis underlying ecologically based divergent selection in flower color between these ecotypes, we have revealed the ecological and functional mechanisms involved in the evolution of pre-mating isolation at the early stages of incipient speciation.

Introgressive hybridization facilitates adaptive divergence in a recent radiation of monkeyflowers.

A primary goal in evolutionary biology is to identify the historical events that have facilitated the origin and spread of adaptations. When these adaptations also lead to reproductive isolation, we can learn about the evolutionary mechanisms contributing to speciation. We reveal the complex history of the gene MaMyb2 in shaping flower colour divergence within a recent radiation of monkey flowers.In the Mimulus aurantiacus species complex, red-flowered M. a. ssp. puniceus and yellow-flowered M. a. ssp. australis are partially isolated because of differences in pollinator preferences. Phylogenetic analyses based on genome-wide variation across the complex suggest two origins of red flowers from a yellow-flowered ancestor: one in M. a. ssp. puniceus and one in M. a. ssp.flemingii. However, in both cases, red flowers are caused by cis-regulatory mutations in the gene MaMyb2. Although this could be due to distinct mutations in each lineage, we show that the red allele in M. a. ssp. puniceus did not evolve de novo or exist as standing variation in its yellow-flowered ancestor. Rather, our results suggest that a single red MaMyb2 allele evolved during the radiation of M. aurantiacus that was subsequently transferred to the yellow-flowered ancestor of M. a. ssp. puniceus via introgressive hybridization. Because gene flow is still possible among taxa, we conclude that introgressive hybridization can be a potent driver of adaptation at the early stages of divergence that can contribute to the origins of biodiversity.

Genomic studies on the nature of species: adaptation and speciation in Mimulus.

Evolutionary biology is in an exciting era, in which powerful genomic tools make the answers accessible to long-standing questions about variation, adaptation and speciation. The availability of a suite of genomic resources, a shared knowledge base and a long history of study have made the phenotypically diverse plant genus Mimulus an important system for understanding ecological and evolutionary processes. An international Mimulus Research Meeting was held at Duke University in June 2014 to discuss developments in ecological and evolutionary genetic studies in Mimulus. Here, we report major recent discoveries presented at the meeting that use genomic approaches to advance our understanding of three major themes: the parallel genetic basis of adaptation; the ecological genomics of speciation; and the evolutionary significance of structural genetic variation. We also suggest future research directions for studies of Mimulus and highlight challenges faced when developing new ecological and evolutionary model systems.

Genomic Approaches Are Improving Taxonomic Representation in Genetic Studies of Speciation.

Until recently, our understanding of the genetics of speciation was limited to a narrow group of model species with a specific set of characteristics that made genetic analysis feasible. Rapidly advancing genomic technologies are eliminating many of the distinctions between laboratory and natural systems. In light of these genomic developments, we review the history of speciation genetics, advances that have been gleaned from model and non-model organisms, the current state of the field, and prospects for broadening the diversity of taxa included in future studies. Responses to a survey of speciation scientists across the world reveal the ongoing division between the types of questions that are addressed in model and non-model organisms. To bridge this gap, we suggest integrating genetic studies from model systems that can be reared in the laboratory or greenhouse with genomic studies in related non-models where extensive ecological knowledge exists.

Genetics and the Evolution of Prezygotic Isolation.

The significance of prezygotic isolation for speciation has been recognized at least since the Modern Synthesis. However, fundamental questions remain. For example, how are genetic associations between traits that contribute to prezygotic isolation maintained? What is the source of genetic variation underlying the evolution of these traits? And how do prezygotic barriers affect patterns of gene flow? We address these questions by reviewing genetic features shared across plants and animals that influence prezygotic isolation. Emerging technologies increasingly enable the identification and functional characterization of the genes involved, allowing us to test established theoretical expectations. Embedding these genes in their developmental context will allow further predictions about what constrains the evolution of prezygotic isolation. Ongoing improvements in statistical and computational tools will reveal how pre- and postzygotic isolation may differ in how they influence gene flow across the genome. Finally, we highlight opportunities for progress by combining theory with appropriate data.

Ancient hybridization leads to the repeated evolution of red flowers across a monkeyflower radiation.

The reuse of old genetic variation can promote rapid diversification in evolutionary radiations, but in most cases, the historical events underlying this divergence are not known. For example, ancient hybridization can generate new combinations of alleles that sort into descendant lineages, potentially providing the raw material to initiate divergence. In the Mimulus aurantiacus species complex, there is evidence for widespread gene flow among members of this radiation. In addition, allelic variation in the MaMyb2 gene is responsible for differences in flower color between the closely related ecotypes of subspecies puniceus, contributing to reproductive isolation by pollinators. Previous work suggested that MaMyb2 was introgressed into the red-flowered ecotype of puniceus. However, additional taxa within the radiation have independently evolved red flowers from their yellow-flowered ancestors, raising the possibility that this introgression had a more ancient origin. In this study, we used repeated tests of admixture from whole-genome sequence data across this diverse radiation to demonstrate that there has been both ancient and recurrent hybridization in this group. However, most of the signal of this ancient introgression has been removed due to selection, suggesting that widespread barriers to gene flow are in place between taxa. Yet, a roughly 30 kb region that contains the MaMyb2 gene is currently shared only among the red-flowered taxa. Patterns of admixture, sequence divergence, and extended haplotype homozygosity across this region confirm a history of ancient hybridization, where functional variants have been preserved due to positive selection in red-flowered taxa but lost in their yellow-flowered counterparts. The results of this study reveal that selection against gene flow can reduce genomic signatures of ancient hybridization, but that historical introgression can provide essential genetic variation that facilitates the repeated evolution of phenotypic traits between lineages.

Multiple evolutionary origins of glyphosate resistance in Lolium multiflorum.

The multitude of herbicide resistance patterns that have evolved in different weed species is a remarkable example of the rapid adaptation to anthropogenic-driven disturbance. Recently, resistance to glyphosate was identified in multiple populations of Lolium multiflorum in Oregon. We used phenotypic approaches, as well as population genomic and gene expression analyses, to determine whether known mechanisms were responsible for glyphosate resistance and whether resistance phenotypes evolved independently in different populations, and to identify potential loci contributing to resistance. We found no evidence of genetic alterations or expression changes at known target and non-target-site resistance mechanisms of glyphosate. Population genomic analyses indicated that resistant populations tended to have largely distinct ancestry from one another, suggesting that glyphosate resistance did not spread among populations by gene flow. Rather, resistance appears to have evolved independently on different genetic backgrounds. We also detected potential loci associated with the resistance phenotype, some of which encode proteins with potential effects on herbicide metabolism. Our results suggest that Oregon populations of L. multiflorum evolved resistance to glyphosate due to a novel mechanism. Future studies that characterize the gene or genes involved in resistance will be necessary to confirm this conclusion.

Predictable patterns of constraint among anthocyanin-regulating transcription factors in Ipomoea.

• Transcription factors (TFs) may play a central role in plant morphological evolution. Variation in the nonsynonymous to synonymous nucleotide substitution rate (dN/dS) ratio among TFs can be attributed to either differences in constraint or the frequency of adaptive substitution. However, the relative contribution of these forces to the variation in dN/dS ratios is unknown. • We synthesize current and previous results comparing the variation in dN/dS ratios among members of the MYB-bHLH-WDR complex of TFs that regulates floral anthocyanin pigmentation in Ipomoea. • Low values of dN/dS in a WDR gene are the result of exceptionally strong purifying selection, with no evidence of positive selection. bHLH and MYB genes also fail to show evidence for positive selection, but have higher dN/dS ratios, indicating reduced selective constraint. • Differences in constraint are consistent with expectations based on the intrinsic features and regulatory network properties among these proteins. Significantly elevated dN/dS ratios in the MYB gene suggest that mutations experience reduced magnitudes of deleterious pleiotropy compared with the rest of the complex. Although reduced constraint may account for the observation that Myb mutations disproportionately contribute to differences in floral pigmentation, the lack of detectable positive selection in any of these TF proteins suggests that amino acid substitutions contribute little to flower colour evolution.

Population genomic and historical analysis suggests a global invasion by bridgehead processes in Mimulus guttatus.

Imperfect historical records and complex demographic histories present challenges for reconstructing the history of biological invasions. Here, we combine historical records, extensive worldwide and genome-wide sampling, and demographic analyses to investigate the global invasion of Mimulus guttatus from North America to Europe and the Southwest Pacific. By sampling 521 plants from 158 native and introduced populations genotyped at >44,000 loci, we determined that invasive M. guttatus was first likely introduced to the British Isles from the Aleutian Islands (Alaska), followed by admixture from multiple parts of the native range. We hypothesise that populations in the British Isles then served as a bridgehead for vanguard invasions worldwide. Our results emphasise the highly admixed nature of introduced M. guttatus and demonstrate the potential of introduced populations to serve as sources of secondary admixture, producing novel hybrids. Unravelling the history of biological invasions provides a starting point to understand how invasive populations adapt to novel environments.

Altered trans-regulatory control of gene expression in multiple anthocyanin genes contributes to adaptive flower color evolution in Mimulus aurantiacus.

A fundamental goal in evolutionary biology is to identify the molecular changes responsible for adaptive evolution. In this study, we describe a genetic analysis to determine whether the molecular changes contributing to adaptive flower color divergence in Mimulus aurantiacus affect gene expression or enzymatic activity. High performance liquid chromatography analysis confirms that flower color differences are caused by the presence versus absence of anthocyanin pigments. Cosegregation analysis and in vitro enzymatic assays rule out mutations that affect enzymatic function in the anthocyanin pathway genes. By contrast, cosegregation of gene expression with flower color suggests that tissue-specific differences in pigment production are caused by the coordinated regulatory control of three anthocyanin pathway genes. We provide evidence indicating that these expression differences are caused by a locus that acts in trans- and explains 45% of the phenotypic variance in flower color. A second locus with sequence similarity to the R2R3 MYB family of transcription factors explains 9% of the variation but does so in a complex fashion. These results demonstrate one of only two examples where we have clear evidence of both the adaptive nature of a flower color transition and evidence for its genetic basis. In both cases, mutations appear to affect expression of the anthocyanin structural genes. Future studies will allow us to determine whether these differences represent a real bias in favor of mutations that affect gene expression.

Genetic changes contributing to the parallel evolution of red floral pigmentation among Ipomoea species.

The repeated, independent evolution of phenotypic traits reflects adaptation to similar selective pressures. In some circumstances, parallel phenotypic evolution has a common genetic basis. Here, we investigate the types of genetic change responsible for the repeated evolution of red flowers among Ipomoea species. We identified three independent transitions from cyanidin- (blue/purple) to pelargonidin-type (red) anthocyanin pigments among Ipomoea species. The genetic basis for these transitions was examined using transgenics and gene expression assays. Using a literature survey to estimate the expected spectrum of mutation types capable of producing red flowers, we evaluated whether the observed distribution of mutation types differed from expectation. In these species, red floral pigmentation appears to be caused by the disruption of flux through the anthocyanin pathway at the same position. Results implicate tissue-specific regulatory changes in the same gene, which suggests the possibility that flower color evolved independently via the same genetic mechanism. Although multiple molecular mechanisms are capable of producing red flowers, we found a deviation between the distributions of observed and expected mutation types responsible for these evolutionary transitions. Regulatory mutations thus appear to be preferentially targeted during evolutionary change between species. We discuss possible explanations for this apparent bias.

Contrasting patterns of floral and molecular variation across a cline in Mimulus aurantiacus.

Steep clines in ecologically important traits may be caused by divergent natural selection. However, processes that do not necessarily invoke ongoing selection, such as secondary contact or restricted gene flow, can also cause patterns of phenotypic differentiation over short spatial scales. Distinguishing among all possible scenarios is difficult, but an attainable goal is to establish whether scenarios that imply selection need to be invoked. We compared the extent of morphological and genetic differentiation between geographically structured red and yellow floral races of Mimulus aurantiacus (bush monkeyflower; Phrymaceae). Flower color was assessed in a common garden as well as in the field to determine whether variation was genetic and to quantify the extent of geographical differentiation. Population genetic differentiation at marker loci was measured for both chloroplast and nuclear genomes, and the degree of population structure within and among the floral races was evaluated. Flower color shows both a strong genetic basis and a sharp geographic transition, with pure red-flowered populations in western San Diego County and pure yellow-flowered populations to the east. In the zone of contact, both pure and intermediate phenotypes occur. Patterns of genetic differentiation at marker loci are far less pronounced, as little of the variation is partitioned according to the differences in flower color. Phenotypic differentiation (Q(ST)) between populations with different flower colors is much greater than neutral genetic differentiation (F(ST)). When comparisons are made between populations of the same flower color, the opposite trend is evident. Limited neutral genetic structure between the floral races, combined with sharp differentiation in flower color, is consistent with the hypothesis that current or recent natural selection maintains the cline in flower color.

Strong premating reproductive isolation drives incipient speciation in Mimulus aurantiacus.

Determining which forms of reproductive isolation have the biggest impact on the process of divergence is a major goal of speciation research. These barriers are often divided into those that affect the potential for hybridization (premating isolation), and those that occur after mating (postmating isolation), and much debate has surrounded the relative importance of these categories. Within the species Mimulus aurantiacus, red- and yellow-flowered ecotypes occur in the southwest corner of California, and a hybrid zone occurs where their ranges overlap. We show that premating barriers are exclusively responsible for isolation in this system, with both ecogeographic and pollinator isolation contributing significantly to total isolation. Postmating forms of reproductive isolation have little or no impact on gene flow, indicating that hybrids likely contribute to introgression at neutral loci. Analysis of molecular variation across thousands of restriction-site associated DNA sequencing (RAD-seq) markers reveals that the genomes of these taxa are largely undifferentiated. However, structure analysis shows that these taxa are distinguishable genetically, likely due to the impact of loci underlying differentiated adaptive phenotypes. These data exhibit the power of divergent natural selection to maintain highly differentiated phenotypes in the face of gene flow during the early stages of speciation.

The geography of divergence with gene flow facilitates multitrait adaptation and the evolution of pollinator isolation in Mimulus aurantiacus.

Ecological adaptation is the driving force during divergence with gene flow and generates reproductive isolation early in speciation. Although gene flow opposes divergence, local adaptation can be facilitated by factors that prevent the breakup of favorable allelic combinations. We investigated how selection, genetic architecture, and geography have contributed to the maintenance of floral trait divergence and pollinator isolation between parapatric ecotypes of Mimulus aurantiacus. Combining greenhouse, field, and genomic studies, we show that sharp clines in floral traits are maintained by spatially varying selection. Although adaptation breaks down where the ecotypes co-occur, leading to the formation of a hybrid zone, the largely non-overlapping distributions of the ecotypes shield them from immigrant genes, facilitating divergence across most of the range. In contrast to the sharp genetic discontinuities observed across most hybrid zones, we observed a gradual cline in genome-wide divergence and a pattern of isolation by distance across the landscape. Thus, contrary to a long period of allopatry followed by recent re-contact, our data suggest that floral trait divergence in M. aurantiacus may have evolved with locally restricted, but ongoing gene flow. Therefore, our study reveals how the geographic distribution of an organism can contribute to the evolution of premating isolation in the early stages of divergence with gene flow.