Fluctuating reproductive isolation and stable ancestry structure in a fine-scaled mosaic of hybridizing Mimulus monkeyflowers

Matthew C. Farnitano; Keith Karoly; Andrea L. Sweigart

doi:10.1371/journal.pgen.1011624

Abstract

Hybridization among taxa impacts a variety of evolutionary processes from adaptation to extinction. We seek to understand both patterns of hybridization across taxa and the evolutionary and ecological forces driving those patterns. To this end, we use whole-genome low-coverage sequencing of 458 wild-grown and 1565 offspring individuals to characterize the structure, stability, and mating dynamics of admixed populations of Mimulus guttatus and Mimulus nasutus across a decade of sampling. In three streams, admixed genomes are common and a M. nasutus organellar haplotype is fixed in M. guttatus, but new hybridization events are rare. Admixture is strongly unidirectional, but each stream has a unique distribution of ancestry proportions. In one stream, three distinct cohorts of admixed ancestry are spatially structured at ~20-50m resolution and stable across years. Mating system provides almost complete isolation of M. nasutus from both M. guttatus and admixed cohorts, and is a partial barrier between admixed and M. guttatus cohorts. Isolation due to phenology is near-complete between M. guttatus and M. nasutus. Phenological isolation is a strong barrier in some years between admixed and M. guttatus cohorts, but a much weaker barrier in other years, providing a potential bridge for gene flow. These fluctuations are associated with differences in water availability across years, supporting a role for climate in mediating the strength of reproductive isolation. Together, mating system and phenology accurately predict fluctuations in assortative mating across years, which we estimate directly using paired maternal and offspring genotypes. Climate-driven fluctuations in reproductive isolation may promote the longer-term stability of a complex mosaic of hybrid ancestry, preventing either complete isolation or complete collapse of species barriers.

Author summary

Hybridization between species can create genetic novelty and promote adaptation, but can also erode species barriers and dilute genetic diversity. Climatic variation likely impacts the extent and eventual outcomes of hybridization, but these impacts are difficult to predict. We use population-scale genomic sequencing of hybridizing Mimulus monkeyflowers to better understand the influence of climatic variation on hybridization. We find evidence of hybridization in multiple populations, but the distribution of hybrid ancestry varies. In one population, groups of distinct hybrid ancestries are clustered in close proximity to each other. Isolation between these ancestry groups is driven in part by variation in self-fertilization and flowering time. However, some years have more overlap in flowering time than others, potentially driven by the timing of water availability. We see that hybrid ancestry is remarkably stable across a decade of measurements, despite fluctuations in the degree of isolation. Paradoxically, fluctuations in isolation may allow hybrid ancestry clusters to persist by preventing complete erosion of species barriers while still allowing some gene exchange. Climate change is expected to increase the variability of climatic factors such as precipitation and heat events; our study demonstrates one way these fluctuations could impact species.

Citation: Farnitano MC, Karoly K, Sweigart AL (2025) Fluctuating reproductive isolation and stable ancestry structure in a fine-scaled mosaic of hybridizing Mimulus monkeyflowers. PLoS Genet 21(3): e1011624. https://doi.org/10.1371/journal.pgen.1011624

Editor: Jesse Lasky, Pennsylvania State University, UNITED STATES OF AMERICA

Received: September 18, 2024; Accepted: February 16, 2025; Published: March 31, 2025

Copyright: © 2025 Farnitano et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All Illumina data generated for this project are archived at the Sequence Read Archive (https://www.ncbi.nlm.nih.gov/sra) under BioProject PRJNA1226178. Phenological data, sample information, and other metadata are archived at the DRYAD digital repository (DOI: https://doi.org/10.5061/dryad.kd51c5bhm). All analysis scripts are available on GitHub (github.com/mfarnitano/CAC_popgen).

Funding: ALS was supported by National Science Foundation award DEB1856180. MCF was supported by the National Institutes of Health Predoctoral Training Grant 5T32GM007103, the Society for the Study of Evolution GREG R.C. Lewontin Award, and a Doctoral Dissertation Improvement Grant (DDIG) from The Plant Center at the University of Georgia. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Hybridization can have a wide range of impacts on species, depending on both the success of hybrid individuals and their degree of reproductive isolation from progenitors. At one extreme, unfit hybrids can be an evolutionary dead-end [1–6]. Conversely, hybrids may freely and successfully mate with both progenitors, eroding differences between species until a single undifferentiated population remains [7–9]. Another possibility is that hybrids are successful but become strongly reproductively isolated from progenitors, forming a new, independent lineage [10–13]. Many cases fall between these extremes, with partial but incomplete reproductive barriers [14–17].

Partial reproductive isolation allows for ongoing gene flow between species (introgression) without collapse into a single lineage. Plant biologists have long understood the importance of hybridization among members of a ‘syngameon’, a group of species that exchange genes but maintain distinctiveness [18,19]. In the genomic era, introgression has been detected throughout the tree of life, suggesting that partial reproductive isolation is common [20,21]. Introgression plays an important role in numerous evolutionary processes such as adaptation [22,23], niche expansion [24,25], evolutionary rescue [26,27], extinction [28,29], and invasion [30,31]. But the dynamics of introgression are varied. Hybridization can be rare and transient, followed by backcrossing within a few generations to leave a signature of introgression at just a few genetic loci [32–34]. In other cases, hybridization produces a persistent swarm of intermating hybrids, which then have the potential to interact more extensively with progenitor species [35–38]. The coexistence of a partially isolated hybrid swarm alongside progenitor species provides an opportunity for significant adaptive introgression, since hybrids can act as a bridge for gene flow between progenitors [39,40]. However, hybridization can also introduce maladaptive alleles or allele combinations [1,41–43] that an admixed population must contend with [37,44–46]. Theory suggests that intermediate levels of reproductive isolation may, under certain circumstances, be an evolutionarily stable state rather than simply a transition state on the route to complete speciation [47]. But hybridization outcomes appear to be highly contingent on environmental and genetic conditions [35,48–50]. More empirical work is needed to understand what conditions lead to persistent partial reproductive isolation in hybrid populations.

One factor that could lead to partial reproductive isolation is spatial heterogeneity. When multiple distinct microhabitats are available, divergent directional selection can maintain multiple ecotypes in parallel niches [51–53]. Selection on locally favored alleles in these microhabitats can act as a premating reproductive barrier, reducing gene flow between ecotypes by eliminating migrants [54–57]. On the other hand, dispersal between these microhabitats could erode both local adaptation and reproductive isolation between ecotypes, leading to collapse into a single gene pool [7,58,59]. Within this pool, multiple microhabitats may instead promote a diversity of alleles through balancing selection [60,61]. The relative importance of divergent selection and balancing selection will depend on the scale and frequency of dispersal relative to the scale of environmental heterogeneity [62–65]. In some cases, dispersal may be sufficient to prevent complete isolation but not strong enough to erode differences completely. In fact, gene flow is often detected in sympatry even when premating reproductive barriers are strong [66–68]. Furthermore, hybrids may establish their own niche in intermediate or underutilized microhabitats, in which persistence is subject to the same balance of migration and selection [11,50,69–71].

Environmental heterogeneity over time, both within and across years, might also contribute to the maintenance of partial reproductive isolation. Differences in the timing of mating throughout the year can isolate groups [72–74]. If reproductive phenology is dependent on environmental cues, then variation in those cues across years can modulate the strength of temporal isolation [75–77]. Also, variation in selective pressures across years can lead to fluctuating selection that maintains a diversity of genotypes [78–81]. Within a hybridizing population, this can result in selection against hybridization in some years but not in others [82,83].

Understanding and predicting the effects of environmental variation on hybridization and reproductive isolation requires tracking hybrid populations across time. Multi-year studies have given us important insights into hybridization, demonstrating directional shifts in composition over time [84], shifts across space [85], stability over time [86,87], and fluctuations in the strength of species barriers [9,82,88]. Still, we know little about the environmental and genetic circumstances driving these different outcomes.

With this in mind, we turn to a previously identified hybridizing population of Mimulus guttatus and Mimulus nasutus monkeyflowers, at Catherine Creek (CAC) just north of the Columbia River Gorge in Washington, USA (Fig 1A). M. nasutus diverged from an M. guttatus progenitor ~200KYA, expanding to share much of the M. guttatus range across the western United States, where the two hybridize in multiple locations of secondary contact [89]. At Catherine Creek and the surrounding area, both species occupy a series of ephemeral seeps, where they co-occur at small spatial scales [90]. Previously, M. guttatus collections in 2012 from two parallel streams at Catherine Creek (CAC_Stream1 and CAC_Stream2, Fig 1A) showed levels of M. nasutus genomic ancestry ranging from near 0 to approximately 50%, indicating a history of hybridization followed by backcrossing to the M. guttatus parental population [89,90].

Download:

Fig 1. Sampling locations and local environmental variation.

A) Location of three sampling locations in the Columbia River Gorge area, Washington, USA. LM=Little Maui stream, CAC=Catherine Creek site with two parallel streams, S1=CAC_Stream1 and S2=CAC_Stream2. Approximate distance between CAC_Stream1 and CAC_Stream2 is ~120m, between CAC and LM is ~2.5 km. B-C) Interpolated precipitation totals (B) and average temperatures (C) in 14-day sliding windows for a 4km grid square covering the CAC and LM sites, downloaded from PRISM [147]. Colored lines indicate years where we sampled for this study; other intervening years are colored in grey. Dashed vertical lines indicate extent of typical flowering season for CAC and LM Mimulus, from April to June. D-E) Sample plots within each of three sampled streams: (D) LM, (E, left) CAC_Stream2, and (E, right) CAC_Stream1. A 0.5mx0.5m square was placed at each plot for flower counts and collections. Exact square placement varied slightly across years, and not all plot locations were sampled in each year. State and country outlines were obtained from gadm.org using the R/geodata package [178]. Aerial maps were obtained from the National Agricultural Imagery Program, courtesy of the U. S. Department of Agriculture, Farm Service Agency [179].

https://doi.org/10.1371/journal.pgen.1011624.g001

Despite this evidence of hybridization, several prezygotic reproductive barriers have been documented between M. guttatus and M. nasutus. One major source of isolation is mating system: M. guttatus is primarily outcrossing (though self-compatible), with large showy flowers visited by a variety of bee pollinators [91–93], while M. nasutus is predominantly selfing, with small and often cleistogamous flowers that self-pollinate prior to opening [94,95]. In addition, the species have phenological differences closely tied to water availability and drought escape: M. nasutus tends to be found on mossy rock outcrops that dry out more quickly, while the seepy microhabitats of M. guttatus stay wet later into the spring [57,90,96]. M. nasutus is better able to accelerate its life cycle to escape terminal drought, while the same drought stress causes a fitness loss in M. guttatus [57]. M. nasutus also flowers earlier in the season than M. guttatus, in part due to a shorter photoperiod requirement for flowering; this interspecific difference has been mapped to multiple large-effect genetic loci [97] and contributes to reproductive isolation in the wild by reducing phenological overlap [90]. But phenology may also be responsive to fluctuating environmental conditions, and the amount of phenological overlap could depend on a number of other factors besides a simple photoperiod cue, such as the duration of flowering and the quantity of floral displays. The Catherine Creek area has highly variable precipitation (both in quantity and seasonal timing, Fig 1B), so we predict that heterogeneous water availability plays an important role in the persistence and isolation of these species and their hybrids. Other environmental variables, such as temperature (Fig 1C) might also impact selection and isolation within these species and their hybrids.

Here, we use a low-coverage whole-genome approach to sequence wild-growing individuals from an additional nearby site 2.5 km away from Catherine Creek, named Little Maui (LM), in order to compare the structure of hybrid ancestry across multiple independent hybrid contact zones. In addition, we sequence three additional years of wild-growing individuals from one focal stream at Catherine Creek (labeled CAC_Stream1), and two years of samples from a second parallel stream ~150m away (labeled CAC_Stream2) along with offspring germinated from seeds of these wild individuals. This multi-year dataset, spanning a decade when combined with previous data, allows us to ask whether the hybrid population at Catherine Creek is stable across years. Next, we address multiple possible contributors to reproductive isolation among ancestry cohorts within Catherine Creek: spatial segregation, mating system variation, and phenological isolation across the flowering season. Using offspring genotypes, we document how these factors influence the actual mating dynamics of the admixed population, and how these dynamics change across years. We develop hypotheses about how fluctuating environmental conditions over time might interact with spatial heterogeneity to create contrasting ancestry patterns in different contact zones.

Results

Introgression from M. nasutus in replicated contact zones

To explore patterns of genomic variation within and between sympatric populations, we used a low-coverage whole-genome approach to sequence a total of 458 wild Mimulus samples collected from three streams in the Columbia River Gorge area (Table 1 and Fig 1A, 1D, and 1E): 265 from CAC_Stream1 collected in 2019, 2021, and 2022; 133 from CAC_Stream2 collected in 2021 and 2022; and 61 from Little Maui (LM, 2.5 km from CAC) collected only in 2021. To explore genomic variation within and between streams, we performed a genomic PCA on these samples (Fig 2A and 2B). PC axis 1 (17.91% of variation) separates species by ancestry, showing a clearly differentiated M. nasutus group and a cloud of M. guttatus-like individuals with varying levels of hybrid ancestry (Fig 2A). This primary axis of variation is highly correlated with hybrid index (proportion M. nasutus vs. M. guttatus ancestry) as determined by local ancestry inference (r²=0.978, Fig 2C). A separate NGSadmix structure analysis with K=2 is also highly correlated with hybrid index (r²=0.979, S1 Fig), further supporting hybrid ancestry proportion as the major axis of genomic differentiation in these populations.

Download:

Table 1. Maternal and offspring samples sequenced for this study from three streams.

https://doi.org/10.1371/journal.pgen.1011624.t001

Download:

Fig 2. Directional admixture shapes population structure in replicate streams.

A) Genomic PCA based on low-coverage genotype likelihoods at 19,633 variant sites for 264 individuals from CAC_Stream1 (CAC_S1, sampled in 2019, 2021, and 2022), 133 individuals from CAC_Stream2 (CAC_S2, sampled in 2021 and 2022), and 61 individuals from LM (sampled in 2021 only). PC1 separates M. nasutus samples (NAS) from M. guttatus and admixed samples, while PC2 separates Little Maui (LM) samples from Catherine Creek (CAC) samples. B) PC3 and PC4 from the same genomic PCA, with Little Maui (LM) and M. nasutus (NAS) samples removed from the visualization to more clearly show differentiation in the other samples. PC3 differentiates an admixed group found primarily in plots S1_1 and S1_2 (cluster CAC-A: note that these plots were not sampled in 2021), while PC4 separates a less-admixed group found primarily in plot S1_5 (cluster CAC-B). The remaining cluster (CAC-C) includes individuals from both CAC streams, which are not differentiated by these PC axes. Variation within Catherine Creek is not structured by year. C) PC axis 1 correlates strongly with hybrid index (the proportion of M. nasutus genomic ancestry, determined by local ancestry inference). NAS=M. nasutus individuals, SOOK=11 individuals of M. sookensis, an allopolyploid species between a M. guttatus-like progenitor and M. nasutus, which has an expected 50/50 ancestry composition between the two species. D) Hybrid index, the proportion of sites with M. nasutus ancestry across the genome for each individual, is distributed differently in each replicate stream, but this pattern is consistent across years. Hybrid index of 0.0 indicates M. guttatus, 1.0 indicates M. nasutus. Data from 2012 are from [90]; note that hybrid index for these samples was obtained using a different method.

https://doi.org/10.1371/journal.pgen.1011624.g002

Additional PC axes reveal population structure associated with geography, both between and within streams. PC2 (1.96% of variation) cleanly separates LM from CAC samples (true for both admixed individuals around PC1 ~ 0 and to a lesser extent for M. nasutus around PC1 ~ 0.1, Fig 2A). PC3 (1.74%) and PC4 (1.31%) further differentiate three clusters of admixed individuals within Catherine Creek (Fig 2B: LM and M. nasutus samples hidden for clarity). Two of these three clusters contain only samples from one stream (CAC_Stream1) while the third contains all remaining samples from both CAC streams. Within CAC_Stream1, each study plot is largely confined to one of these three clusters (S1 Table), suggesting that these axes of differentiation reflect spatial structure within CAC_Stream1 at a scale of <300m. We find similar clustering within CAC_Stream1 when running a separate PCA of only CAC individuals with hybrid index <0.8 (i.e., excluding LM and M. nasutus individuals, S2 Fig). Notably, despite differentiation within CAC_Stream1, samples from the parallel stream CAC_Stream2 (~150 m away) are not differentiated from CAC_Stream1 nor do they form multiple clusters, at least in these first four PC axes (main PCA, Fig 1A and 1B, and CAC-only PCA, S2 Fig). Each PCA cluster contains samples from multiple sampling years and therefore multiple sequencing batches (see Materials and Methods), except the LM cluster because LM was only sampled in one year, indicating that year and associated batch effects do not account for the clustering.

Consistent with previous population genomic analyses [89,90], the distribution of hybrid index (HI, the proportion of the genome with M. nasutus ancestry) suggests mostly unidirectional backcrossing of hybrids with M. guttatus (i.e., a majority of admixed individuals have HI<0.5, Fig 2D). In fact, virtually all CAC and LM M. guttatus-like individuals have at least some detectable M. nasutus ancestry: out of 374 majority-guttatus samples, only 12 have HI<0.05, while only two of those have HI<0.01. The M. nasutus samples, in contrast, have little to no introgression: out of 74 majority-nasutus samples, 61 have HI>0.95 and 57 of those have HI>0.99. The remaining 13 have HI between 0.5 and 0.8, indicating that backcrossing with M. nasutus does occasionally happen. No individuals were sampled with HI between 0.8 and 0.95, suggesting that backcrossing to M. nasutus does not typically continue for multiple generations as it does with M. guttatus. Despite pervasive admixture, we detect no first-generation (F1) hybrids (S3 Fig). We did discover a group of 11 individuals from CAC_Stream2 with HIs near 0.5, which were determined to be the allopolyploid species M. sookensis (Figs 2C, S3 and S4, and Materials and Methods). These provide a useful check of our ancestry calling method, since they have an expected HI of 0.5 due to apparent fixed heterozygosity caused by two homeologous copies of each locus [98]; once identified, they were removed from further analyses of admixture and reproductive isolation. Of the remaining 27 individuals with HIs between 0.4 and 0.6, all have ancestry heterozygosity values less than 0.62 (S3 Fig), indicating they are second- or later-generation hybrids as opposed to new F1s (which should have heterozygosity near 1.0). Taken together, our finding that directional introgression composed of later-generation and backcrossed hybrids is replicated in all three sampled streams indicates admixture in secondary contact is not only common but follows predictable patterns across the landscape.

Given the admixture we detect in the nuclear genome, we asked whether organellar genomes also show signatures of introgression. Remarkably, we found that all 325 CAC and LM samples included in our chloroplast network analysis – whether M. guttatus, admixed, or M. nasutus - have a single chloroplast haplotype (Fig 3). A similar pattern is seen in the mitochondria (S5 Fig). This same haplotype (or close derivatives) is carried by all sampled M. nasutus and M. sookensis (an allopolyploid with a maternal M. nasutus origin) lineages, including collections spanning most of the two species’ ranges [98]. In contrast, outside of CAC and LM, the haplotype is almost never found in M. guttatus, which has much higher organellar diversity across its geographic range (Figs 3 and S5): of 18 previously sampled non-CAC/LM M. guttatus accessions, only two carry this CAC/LM haplotype. One of these two accessions is the only other M. guttatus sample in our dataset from a known sympatric site: DPR, which is ~1000 km south of CAC/LM in the Sierras of California and shows a signature of M. nasutus introgression [46,89]. We therefore infer that this haplotype is derived from M. nasutus and has been completely captured by sympatric M. guttatus through introgression at both CAC and LM sites (and, potentially, at other sympatric populations elsewhere in the range, such as DPR). This haplotype structure suggests that not a single M. guttatus individual in our sample is free from the effects of introgression.

Download:

Fig 3. Complete chloroplast capture of a M. nasutus haplotype in sympatric M. guttatus.

NJ-Net haplotype network of chloroplast variation built from 328 CAC and LM maternal samples sequenced in this study (bolded captions) and 47 previously sequenced samples (non-bolded captions) from the M. guttatus species complex, including M. nasutus, M. guttatus, M. sookensis (a polyploid with M. nasutus as maternal parent), and M. decorus (another close relative) [98], using 102 total variant sites (43 parsimony-informative). Each circle represents a unique haplotype, with the size of the circle proportional to the number of samples matching that haplotype. Haplotypes are connected according to sequence similarity, with tick marks along the connections representing the number of nucleotide changes separating each sequence. A single chloroplast haplotype is present in all maternal samples from Catherine Creek, including M. guttatus, admixed, M. nasutus, and M. sookensis samples. All other M. nasutus samples from across the range share this same haplotype or a close derivative, as do samples from M. sookensis (a polyploid with M. nasutus as maternal parent). M. guttatus haplotypes are more variable, with only two samples not from CAC or LM sharing the M. nasutus haplotype, at least one of which (DPR84) is from another sympatric site with known introgression.

https://doi.org/10.1371/journal.pgen.1011624.g003

Patterns of ancestry at small spatial scales are persistent across years.

While we find substantial admixture in all three streams, ancestry proportions vary across the landscape (Fig 2D). In each stream, we see multiple distinct, though sometimes overlapping, peaks of ancestry, which we refer to as ancestry cohorts. All three streams have a clear M. nasutus cohort (HI>0.95). LM and CAC_Stream2 each have a second admixed cohort with majority-M. guttatus ancestry, but the two streams have different peak ancestries (HI=0.25-0.3 for LM and 0.1-0.2 for CAC_Stream2). Within CAC_Stream1, we see three cohorts: M. nasutus, a M. guttatus-like cohort centered around HI=0.05-0.1, and a more admixed cohort centered around HI=0.25-0.3. Where we have multiple years of sampling, the distribution of ancestry in each stream appears similar across years: in particular, all three cohorts (M. guttatus, admixed, and M. nasutus) in CAC_Stream1 are present in 2012, 2019, and 2022; note that M. nasutus at CAC were not sampled in 2021 because we arrived at the site later in the season that year. This year-to-year consistency both validates the reliability of our ancestry assignments and suggests that the three-cohort ancestry structure of CAC_Stream1 is stable across time, even though such structure is not found in other nearby streams.

We asked whether the multiple peaks of ancestry within CAC_Stream1 correspond to spatial structure at a smaller scale. The distribution of hybrid ancestry varies across plots, with clear spatial segregation between the three ancestry cohorts (Fig 4A). These differences are apparent on very fine scales: plots ~20m apart have distinct distributions of ancestry (i.e., plots S1_4, S1_5, and S1_6; Figs 1C and 4A). In contrast, LM and CAC_Stream2 appear to have less clear differentiation across space, though M. nasutus samples are clustered towards the lower plots in both (Fig 4B and 4C). The CAC_Stream1 spatial pattern of ancestry is remarkably consistent across years wherever we have multiple years of sampling (Fig 4A). This is true even comparing 2012 (which used a different sequencing methodology) to later years. There are a few exceptions: plot S1_7, for example, appears to have become more admixed between 2012 and 2019. But overall, differences in admixture proportion at small spatial scales are remarkably consistent across a decade of sampling in CAC_Stream1, contrasting with weaker spatial segregation in the other streams.

Download:

Fig 4. Fine-scaled spatial and phenological structure across years.

A) Histogram of maternal samples sequenced from each of eight plots within CAC_Stream1 (CAC_S1), binned by hybrid index, with sampling years in stacked bars. Hybrid index of 0.0 indicates M. guttatus, 1.0 indicates M. nasutus. Plots vary in the distribution of M. guttatus, admixed, and M. nasutus individuals, but most have a consistent distribution across years. B) Histograms of hybrid index within each of seven plots from CAC_Stream2 (CAC_S2), sampled in 2012 (previous data), 2021 (limited sampling), and 2022. C) Histograms of hybrid index within each of eight Little Maui (LM) plots, sampled in 2021 only. D) Median flower date and range of first to last open flower dates for each census plot in CAC_Stream1 and CAC_Stream2, compared to median hybrid index of sequenced samples from that plot. Median flower date is associated with median hybrid index. Flowering season duration was longer for most plots in 2022 compared to previous years. E) Date of marked open flowers is associated with hybrid index of the marked individual for CAC_Stream2, but not for CAC_Stream1. Random sets of flowers were marked when open throughout the season and those individuals were later sampled for sequencing; these are not necessarily the first open flowers for an individual. Colored lines and grey intervals represent model fits and 95% confidence intervals from a beta GAM regression, using year and stream as additional factor variables.

https://doi.org/10.1371/journal.pgen.1011624.g004

Partial reproductive isolation between admixture cohorts due to mating system

We next sought to investigate potential reasons for the persistence of spatial ancestry structure at such fine scales. One possible cause is premating isolation driven by self-fertilization, so we used paired maternal and offspring sequence data within the two CAC streams to infer selfing rates for our samples. After filtering for sequence quality and maternal match, we inferred selfing status for 1045 offspring from 189 individual fruits across 151 maternal families (Table 2). We found that, as expected, M. nasutus samples are highly selfing: 0 of 43 offspring from M. nasutus maternal plants were inferred to be outcrossed. M. guttatus and admixed cohorts had a mix of selfing and outcrossing: selfing rates for M. guttatus individuals with HI<0.15 ranged from 0 to 0.217 depending on year and stream (Table 2) while more admixed individuals (HI=0.15-0.8) had similar selfing rates ranging from 0.026-0.365, Table 2). Selfing rate significantly increased with increasing M. nasutus ancestry, though this pattern is primarily driven by M. nasutus individuals: when M. nasutus (HI>0.8) families are excluded, selfing increases with ancestry in 2021 and 2022 but decreases with ancestry in 2019 (S2 Table and S6 Fig). Selfing rate also significantly varied across years, with 2022 having the highest selfing rates across all ancestry cohorts (Tables 2 and S2 and S6 Fig). There was no significant effect of stream (CAC_Stream1 vs. CAC_Stream2) on selfing rates (Tables 2 and S2 and S6 Fig). Within the M. guttatus and admixed cohorts, most selfed offspring were found alongside outcrossed offspring in the same fruit: 13% of fruits with at least two called offspring had a mix of selfed and outcrossed offspring (S3 Table). Of these mixed fruits, most were majority outcrossing with a mean selfing proportion of 37% (S3 Table). Multiple paternity within a fruit was also common: out of 155 fruits with at least two outcrossed offspring, 119 (77%) had more than one inferred pollen donor, with an average probability of 52.9% that two outcrossed offspring in the same fruit were half-siblings rather than full-siblings (S3 Table).

Download:

Table 2. Mating system estimation.

https://doi.org/10.1371/journal.pgen.1011624.t002

We used selfing rates to estimate the strength of reproductive isolation between cohorts due to mating system. In CAC_Stream1, maternal families can be divided into three cohorts (M. guttatus, admixed, or M. nasutus) either according to plot identity or by maternal hybrid index; these classifications agree for 97% of maternal families (all but four: one M. guttatus and three M. nasutus maternal plants in plots classified as admixed). For simplicity, we chose to use maternal hybrid index as the classifier. For M. nasutus (HI>0.95) maternal plants, mating system is an almost complete barrier to reproduction (RI=1.0, Table 3). Between the M. guttatus (HI<0.15) and admixed (HI 0.15-0.8) cohorts in CAC_Stream1, mating system was a partial barrier, varying in strength by year and direction (RI=0.026-0.365, Table 3). Classification into cohorts in CAC_Stream2 is more ambiguous, because there is no plot-level division or clearly disjunct distribution between M. guttatus and admixed individuals, and because we did not sequence any M. nasutus offspring from this stream. If we arbitrarily use the same cutoff as CAC_Stream1 to divide M. guttatus and admixed individuals, we find that mating system would provide a similar weak but partial barrier to reproduction between these groups (0-0.113 depending on year and direction, Table 3). Note that all mating system estimates are based on the maternal plant and therefore do not account for paternal effects of selfing on reproductive isolation, such as lower contributions to the outcrossing pollen pool by self-fertilizing flowers.

Download:

Table 3. Measurements of premating reproductive isolation between cohorts.

https://doi.org/10.1371/journal.pgen.1011624.t003

Partial phenological isolation between ancestry cohorts

Another potential source of isolation between ancestry cohorts is flowering phenology, so we conducted a multi-year census of flowering phenology at CAC (with plot-level data from 2012, 2019, and 2022; and individual-level data from 2019 and 2022) to address phenological isolation. For a given plot, peak flowering times were typically consistent across years (S7 and S8 Figs), despite differences in overall flower abundance (S7 and S8 Figs). As previously shown for the 2012 growing season [90], we found that plot phenology at CAC consistently tracks hybrid ancestry: median flower date of plots is highly correlated with their median hybrid index (adjusted r² = 0.776, Fig 4D and S2 Table). The flowering time of individual plants is also associated with ancestry within CAC_Stream1 for both 2019 and 2022 (adjusted r² = 0.435, Fig 4E and S2 Table): M. nasutus flowers were typically marked early in the season, admixed individuals mid-season, and late-season M. guttatus individuals later in the spring. In contrast, individual flowering time was not associated with hybrid index for CAC_Stream2 (Fig 4E; see effect of stream and interaction effect of flower date by stream on hybrid index, S2 Table, p=0.0006 and 0.0040), though this may be due to a lack of variation in CAC_Stream2 hybrid index, since we did not mark any CAC_Stream2 M. nasutus individuals and the stream lacks a distinct M. guttatus cohort.

Using plot-level flowering census data from 2012, 2019, and 2022, we calculated the strength of phenological isolation among the three cohorts in CAC_Stream1: M. nasutus (plots S1_3 and S1_6), admixed (plots S1_2 and S1_4), and M. guttatus (plots S1_5 and S1_8); note as before that these plot assignments are generally predictive of individual-level ancestry, but not perfectly so (Fig 4A). Isolation between M. nasutus and M. guttatus was complete with zero flowering overlap in 2012 and 2019, and strong but incomplete in 2022 (RI=0.58 to 1.0, Table 3). This finding, along with the absence of any first-generation hybrids among 458 wild-collected CAC and LM samples (S3A Fig), suggests new interspecific crosses are rare. In contrast, phenological isolation between admixed individuals and M. guttatus was highly variable across the three study years: although the two groups were strongly isolated in 2012 (RI=0.88 and 0.99 depending on direction, Table 3) and in 2019 (RI=0.89 and 0.92, Table 3), they were much less so in 2022 (RI=0.26 and 0.47, Table 3).

What might explain the reduction in phenological isolation in 2022? Although median flowering time of each cohort was relatively stable across years, the duration of flowering was not: admixed plots in 2022 flowered later into the season than in 2019 or 2012 (Figs 4D and S7). As a result, there was greater phenological overlap between cohorts in 2022 compared to 2019 or 2012 (e.g., plots S1_1 and S1_2 overlapping with S1_5; Figs 4D and S7). We see this pattern in the individual-level data as well: in 2019, all marked flowers after May 15 were from the M. guttatus cohort, but in 2022, multiple flowers marked after May 15 were from substantially admixed individuals (Fig 4E). Differences in absolute abundance likely play an important role as well: 2019 had much lower flower counts throughout the season in all plots compared to both 2012 and 2022 (S7C and S8C Figs). We also counted relatively more flowers in admixed than in M. guttatus plots in both 2012 and 2022 (S7C and S8C Figs), explaining the asymmetry of reproductive isolation between the M. guttatus and admixed cohorts in those years (i.e., higher probability of pollen flow from admixed to M. guttatus plots: Table 3).

The strength of phenological isolation between admixed individuals and M. nasutus also varied across years, depending on the direction of pollen flow. In CAC_Stream1, admixed individuals were less likely to receive pollen from M. nasutus in 2022 (RI = 0.97) than in 2012 (RI = 0.83) or 2019 (RI = 0.68), in part due to a wave of late-flowering hybrids in 2022 (Table 3 and Figs 4B and S7B) that did not overlap with M. nasutus. For the reverse direction, in both 2012 and 2022, M. nasutus was poorly isolated from admixed individuals (RI = -0.51 and -0.267: Table 3), but these estimates are incomplete because our census of open flowers in these years began after M. nasutus had already begun flowering (S7B Fig). While CAC_Stream2 data are more limited, there was substantial phenological isolation preventing pollen flow from M. nasutus into admixed plots (RI=0.92, Table 3), but not in the reverse direction, although this asymmetry is due to a higher number of admixed than M. nasutus plots, which may not be representative of the stream demography as a whole.

Measured reproductive barriers accurately predict offspring ancestry shifts

Our paired sets of maternal-offspring genotypes allow us to directly test whether the strong premating barriers we detect at CAC actually translate to observed offspring identities. Any deviation in offspring hybrid indices from maternal values implies incomplete assortative mating (postmating barriers could shift allele frequencies at particular loci but are unlikely to cause systematic shifts in offspring HI). For M. nasutus, we observed nearly complete assortative mating due to self-fertilization: at CAC none of our 43 inferred offspring of M. nasutus plants across 11 fruits were inferred as outcrossed by BORICE (Table 2). We do find one example at LM where 2 of 7 offspring in a single M. nasutus fruit have admixed ancestry (HI=0.687 and 0.627), implying a partially-admixed M. guttatus-like pollen parent. In the reciprocal direction, one offspring from a 2022 CAC M. guttatus fruit (maternal HI=0.089) and one offspring from a 2021 LM admixed fruit (maternal HI=0.316) had hybrid indices consistent with a M. nasutus pollen parent. We can therefore estimate that the rate of M. nasutus mating outside its own ancestry cohort is 2/67 or 3.0% maternally, and 2/1501=0.1% paternally.

Next, we asked to what extent mating system differences and phenological isolation shape patterns of assortative mating between hybrids and M. guttatus at CAC. We reasoned that assortative mating should be maximized in selfed offspring and, indeed, we observe little to no deviation in HI for offspring inferred as selfed relative to their maternal HI values (Fig 5A and S2 Table). We do find a slight effect of year on this offspring deviation (which may be attributed to uncertainty in hybrid index estimation), but no effect of maternal HI or stream (S2 Table). For outcrossed offspring (which, by definition, have escaped the effects of mating system isolation), we might expect to observe a stronger deviation between maternal and offspring HIs if assortative mating due to phenological isolation is incomplete. This is precisely what we observe: hybrid indices of outcrossed offspring show a clear deviation from maternal values, with 45.9% of variance in offspring HI deviation attributed to maternal HI (Fig 5B and S2 Table). On average, M. guttatus mothers produced more-admixed offspring and admixed mothers produced less-admixed offspring than themselves.

Download:

Fig 5. Assortative mating agrees with patterns of reproductive isolation.

A) Comparison of offspring hybrid index to maternal hybrid index (HI, the proportion of M. nasutus ancestry across the genome) for selfed offspring from CAC_Stream1 (CAC_S1) and CAC_Stream2 (CAC_S2). Each point represents the mean of all offspring within a single fruit. Offspring HI deviation is equal to (Offspring HI – Maternal HI), with positive values indicating greater M. nasutus ancestry in offspring relative to maternal plants, negative values indicating less M. nasutus ancestry, and zero indicating that offspring HI exactly matches maternal HI. Lines indicate individual linear GAM model fits for each year, with shading around each line indicating 95% confidence intervals. There was no significant effect of stream on offspring deviation for selfed offspring (S2 Table), so the two streams are combined into a single linear fit for each year. The thin black dotted line shows the null expectation for selfed fruits, where offspring HI perfectly matches maternal HI. B) Equivalent plot for outcrossed offspring, showing a larger deviation from maternal ancestry, with the direction of deviation typically in the direction of the population mean. Each line indicates a linear GAM model fit for each year and stream, with shading indicating 95% confidence intervals.

https://doi.org/10.1371/journal.pgen.1011624.g005

Strikingly, we also find evidence that yearly variation in assortative mating mirrors patterns of phenological isolation at CAC. In 2019, when phenological isolation between M. guttatus and hybrids in CAC_Stream1 was nearly complete (RI: 0.92 and 0.88, Table 3), low-HI (M. guttatus-like) maternal plants almost never produced offspring with shifts in ancestry. In both 2019 and 2021, most large shifts in hybrid ancestry occurred in the highest-HI admixed individuals (i.e., HI <0.5), which are further from the population mean; they tended to produce offspring with lower hybrid indices, presumably by crossing with more abundant lower-HI hybrids (Fig 5B). In contrast, in 2022, when phenological isolation between these groups was at its lowest (RI: 0.26 to 0.47, Table 3), both low-HI M. guttatus and higher-HI admixed parents produced offspring with HIs shifted towards the population mean (Fig 5B), consistent with a breakdown in assortative mating allowing cross-cohort outcrossing. This difference is reflected in a significant effect of year (2022) on offspring HI deviation (p=0.01, S2 Table). High-HI individuals were also more likely to self in 2022 (Table 2). The net result of these changes is that, in 2022, outcrossed offspring overall experienced a slight shift to higher levels of admixture compared to maternal samples (mean shift in HI = +0.0172 ± 0.0002, Fig 5B), whereas in 2019 and 2021, there was a slight shift to lower levels of admixture (mean shift in HI = -0.0333 ± 0.0002 and -0.0155 ± 0.0004, Fig 5B).

Discussion

Here, we describe in detail the composition and mating dynamics within hybridizing populations of Mimulus guttatus and Mimulus nasutus during secondary contact. Admixture is prevalent in independent streams, with similar patterns of multi-generational, directional introgression from M. nasutus into M. guttatus. The distribution of hybrid ancestry across space is variable both at very fine (~50m) and coarser (~2.5 km) scales, but stable across a decade of sampling. We measure partial reproductive isolation between three ancestry cohorts by mating system and phenology, both of which likely contribute to the maintenance of ancestry structure across space. We demonstrate substantial year-to-year variation in phenological isolation, which may be associated with climatic variation. Using direct measurements of offspring ancestry composition, we confirm that variation in measured reproductive isolation predicts observed assortative mating. This is a rare direct confirmation of the effect of premating reproductive barriers on offspring outcomes. Our system demonstrates that the outcomes of hybridization can be dynamic and complex, highlighting the importance of sampling across both space and time.

Repeated cases of introgression across the landscape

While introgression between M. nasutus and M. guttatus has been detected previously at Catherine Creek and in a separate sympatric area in California [44,46,89,90,99], strong premating reproductive barriers between the two species imply that initial hybridization rates should be rare [66]. Our data suggest that ongoing admixture is common across the landscape, with independent admixed populations at Little Maui and Catherine Creek, despite no new hybridization events detected. This follows a trend in other systems where genomic signatures of introgression often co-occur with strong reproductive isolation [68,100], and a broader pattern of frequent signals of hybridization despite generally strong premating reproductive isolation across the tree of life [14,20,21]. Theory shows that even occasional migration events can strongly influence allele frequencies [68,101], and rare hybridization events can result in admixed populations if hybrids are persistent once formed. Our finding of weaker reproductive isolation between admixed groups compared to non-admixed progenitors means that, once a few admixed individuals are present, they may promote additional admixture, acting as a genetic ‘bridge’ between otherwise isolated populations and increasing the chance of adaptive introgression [39,40,102,103]. Admixed groups might also function as a genetic ‘sieve’: multiple generations of selection can purge incompatible allele combinations, while recombination breaks up linkage between incompatibilities and potentially adaptive alleles [7,46,104].

An open question is how repeatable introgression patterns will be when hybridization occurs multiple times [35,49,104–106]. The overall pattern of directional introgression, with pervasive M. nasutus ancestry in majority-M. guttatus genomes but very little signature of introgression into M. nasutus, is consistent across streams. The high selfing rate of M. nasutus explains this directionality, which is a common pattern in selfer-outcrosser pairs [77,99,107–109]. But despite similar asymmetries, each of our three streams have their own unique distribution of hybrid ancestry. This matches findings in other hybrid zones with a mosaic structure [37,104,106,110,111]. Differences in the timing and extent of water availability [77], the spatial distribution of microsites [111], the relative abundance of each species [110], or the presence of pollinators [112] could all influence these idiosyncratic patterns. We note that our data were not designed to capture the relative abundance of M. nasutus vs. M. guttatus (we likely undersampled M. nasutus), but more deliberate sampling in the future could test whether relative abundance impacts differences in the average ancestry of admixed individuals across streams. Expanding this work to additional streams and measuring ecological variables at higher resolution will help us understand the effects of these other factors. Overall, we see that a combination of consistent (i.e., selfing) and heterogeneous (i.e., flowering time) reproductive barriers can produce a patchwork of broadly similar but subtly different outcomes each time admixture occurs.

Complete organellar capture by hybridization

Hybridization often moves organellar genome haplotypes from one lineage into another, a phenomenon known as organellar capture [113–115]. Typically, organellar capture is assessed in just a small number of samples at phylogenetic resolution. Our population-scale sample provides a unique window into hybridization history – mainly, that an entire sympatric area has a single maternal origin. An M. nasutus maternal origin is consistent with the asymmetric nature of gene flow in our system and the observation, by us and others [66] that M. nasutus is more often the maternal parent when hybridizing. It also corroborates our finding that all our CAC and LM M. guttatus samples have at least small amounts of nuclear M. nasutus ancestry. Still, the extent of capture in even our most M. guttatus-like samples is striking and gives us insight into the formation of our hybridizing populations. Not only was the initial hybridization directional, but hybrids must have consistently remained as the seed parent across multiple generations of backcrossing, with nearby M. guttatus progenitors contributing primarily through pollen flow rather than seed dispersal. This pattern is consistent with pollen flow acting over longer distances than seed dispersal, so that most new seeds are from maternal plants from within a population, but pollen occasionally arrives from elsewhere. We might expect a similar pattern in other plant systems when pollen flow happens over longer distances than seed dispersal [116], or in animal systems for which males tend to disperse longer distances than females [117].

Another possibility is that the M. nasutus organellar haplotype might have some selective advantage within hybrid populations, perhaps due to segregating cytonuclear incompatibilities. Such interactions are common across eukaryotes [118–122], including between populations of M. guttatus and M. nasutus [123]. But we stress that asymmetries in reproductive isolation and dispersal are sufficient to explain these patterns without needing to invoke selection.

Persistent spatial structure with fluctuating reproductive isolation

Repeated sampling across years allows us to see that CAC populations have a stable distribution of hybrid ancestry, suggesting a lack of severe hybrid breakdown or maladaptation, though weak postzygotic barriers have been detected [44] and there is genomic evidence of selection against M. nasutus ancestry at Catherine Creek [44,90]. Ancestry levels are stably differentiated at ~20-50m scales within one stream (CAC_Stream1), much smaller than the scale of expected pollinator movement, suggesting that forces other than distance are maintaining isolation between cohorts. In contrast, this cohort structure is not apparent in the other streams we sampled (CAC_Stream2 and LM), despite similar distances between plots. The combination of factors that allow multiple ancestry cohorts to persist in one stream (CAC_Stream1) might be idiosyncratic to a particular heterogeneous landscape or population history, and are not necessarily replicable across different populations (e.g., CAC_Stream2 or LM); future work should examine what might be different about the landscape or history of these streams to drive differences in structure.

Population structure at fine scales has been described as a feature of M. guttatus populations [124,125], which often harbor substantial genetic diversity; but our study is, to our knowledge, the first to describe hybrid ancestry as a key component of this fine-scaled structure. Mosaic hybrid zones with stark fine-scaled structure have been described in other systems [126,127] and may be common in plants [128] but there are few studies of change over time in such systems. More broadly, hybrid zone studies have sometimes found stability across years [86,87], and other times found substantial shifts [9,84,85], but the reasons for these patterns are underexplored. We therefore set out to understand what components of reproductive isolation exist between cohorts within this fine-scaled mosaic, and how these components of reproductive isolation might change over time. Reproductive barriers between cohorts may serve as one force maintaining a stable mosaic distribution at such fine scales.

As expected, selfing is an important barrier isolating M. nasutus from both M. guttatus and admixed cohorts. Interestingly, it also provides a partial barrier between M. guttatus and admixed cohorts. For M. guttatus, our selfing rates generally agree with those of other studies, which range from about 25–50% [124,129–132]. Selfing rates in admixed Mimulus have not previously been documented; we find that they are generally closer to M. guttatus than M. nasutus, consistent with dominance of M. guttatus floral phenotypes [95]. We also see that selfing rates fluctuate slightly across years, possibly due to changes in pollinator abundance and timing [133], or in the size and number of flowers as a consequence of general plant health [134]. One intriguing possibility is that the milder conditions in 2022 allowed more plants to have multiple flowers open simultaneously, leading to an increase in geitonogamous (between-flower) selfing, an important selfing mode in M. guttatus [130]. But overall, selfing is probably a fairly consistent partial barrier across years.

Phenological isolation is also a strong barrier, confirming previous results [90], and is particularly important between the M. guttatus and admixed cohorts of CAC_Stream1, which are less isolated by mating system. However, phenological isolation is quite variable across years, a result confirmed by differential shifts in offspring ancestry across years. Phenology has a strong genetic component in these species: two major QTL contributing to differences in photoperiod response have been mapped to candidate genes [97]. Our results suggest that these genetic differences are only part of the picture, with isolation mediated by other factors. While based on a limited number of years, we note an intriguing trend of stronger isolation in drier (i.e., 2012, 2019) compared to wetter (i.e., 2022) seasons (Fig 1B), pointing to water availability as a potential factor influencing the strength of phenological isolation, acting through changes in flowering duration and abundance. 2022 was also an outlier for lower flowering-season temperatures (Fig 1C), which could play a role as well. Precipitation amount and variability are frequently associated with shifts in flowering phenology across plant taxa [135–138]. With ongoing sampling at Catherine Creek, we will be able to test whether the correlation between precipitation and phenological isolation holds across time. Direct measurements of water availability in microsites across the growing season will also help confirm this relationship in the future.

Impact of fluctuating environments on hybrid zones

Fluctuating environmental conditions can provide a form of balancing selection, maintaining allelic diversity by favoring different alleles in different years [60,78,80]. Similarly, environmental fluctuations may help maintain a variety of ancestry combinations after admixture: [77] found a correlation between interannual variance in precipitation and the extent of introgression across replicate contact zones. Environmental fluctuations could influence the distribution of hybrid ancestry in two main ways: varying the strength of selection against hybrids [83], or directly modulating the strength of premating reproductive isolation [75,76]. In our case, it appears the environment may be directly impacting the strength of premating reproductive isolation by varying the length of the flowering season and the consequent extent of flowering overlap; however, future work is needed to tie these fluctuations explicitly to particular environmental cues. Reproductive barriers that are sensitive to environmental cues, like phenology, may be important sources of fluctuating reproductive isolation across systems.

Climatic variability is projected to increase in many ecosystems with global climate change [139–141]. It is therefore imperative that we understand the effects of environmental variability on populations. In the context of climate change, hybridization is predicted to have both beneficial effects, such as increased genetic diversity and adaptive potential, and deleterious effects, such as swamping of rare taxa and homogenization of genetic diversity, each of which is likely to be context- and system-specific [75,142–146]. Our data raise the possibility that climatic variability itself can impact the extent of reproductive isolation in hybridizing populations. In a warmer, more unpredictable world, fluctuating reproductive isolation may become more common, leading to an increase in complex, dynamic scenarios of hybridization like this one. To manage these scenarios, we need a better understanding of which factors influence reproductive isolation and hybridization, how they change across space and time, and how they impact the structure and composition of admixed populations.

Materials and methods

Environmental data

We obtained daily climatic data for the years 2010–2022 from the PRISM online database [147], dataset ID AN91d, for the 4km grid square centered on (Latitude=45.7113, Longitude=-121.3637), which includes the Catherine Creek site. We calculated a rolling 14-day sum of total daily precipitation and a rolling 14-day mean of daily mean temperature, using the R package ‘zoo’ [148].

Field sampling and collections

We sampled tissue for DNA extraction and sequencing from wild-growing Mimulus individuals during the 2019, 2021, and 2022 growing seasons (April through June). In 2019, we sampled from a single stream at Catherine Creek, CAC_Stream1. In 2021 and 2022, we sampled from both CAC_Stream1 and an adjacent parallel stream ~120m to the west, CAC_Stream2 (Fig 1A). In 2021, we also sampled from a third more distant stream ~2.5 km away (Little Maui, or LM). In 2021, because we arrived at the field sites in mid-May (due to the ongoing COVID19 pandemic), CAC samples were limited to late-flowering individuals (flowering at LM is shifted to later in the season and thus was unaffected).

Within each stream, we set out 0.5m x 0.5m plots in sites along the stream where Mimulus individuals were growing; all samples were collected from within these plots. Plot locations were not always exactly the same across years; to simplify comparisons across years, we assigned plots within 10 m of each other to the same plot ID, resulting in a total of 8 plot IDs each for CAC_Stream1 and LM, plus 7 plot IDs for CAC_Stream2 (Fig 1C). Some plot IDs were not sampled in every year. A few plot IDs encompass samples from more than one adjacent plot from the same year; for these plots, we chose the one with the median number of flowers as a representative for phenological comparisons. A full accounting of plot locations across all years is provided in S4 Table. Approximately once per week throughout the flowering season in both 2019 and 2022, we counted the number of open Mimulus flowers within each plot. The total number of flowers across M. guttatus, M. nasutus, and potential hybrids was recorded for each plot at each time point; individual flowers were not field-identified to species as hybrids are difficult to distinguish in the field. Flowers are typically open for 1–3 days, and a single individual may have multiple flowers both simultaneously and in sequence (individual flower number ranges from 1 to >50). Phenology data were not collected in 2021.

During each visit to CAC in 2019 and 2022 and LM in 2021, we used acrylic paint to mark the calyx of three random open flowers per plot (if available). We used a different color of paint on each visit to indicate the date flowers were open. Later in the season, we attempted to relocate these same individuals and, if successful, collected fruits from the marked flowers, as well as leaf tissue into envelopes with silica for DNA. Seeds from one to three fruits per maternal plant were later germinated in the UGA Botany greenhouses; leaf/bud tissue was collected from the resulting offspring (1–16 offspring per wild-sampled maternal plant; final counts of families, fruits, and offspring are in S3 Table) and stored at -80°C for DNA extraction and sequencing. Sample metadata is archived at the Dryad Digital Repository [149].

DNA Extraction and Illumina sequencing

We extracted genomic DNA from both the dried wild-collected tissue samples and the greenhouse-grown offspring samples using a CTAB extraction protocol with phenol-chloroform extraction [150]. Dried tissue was flash-frozen in liquid nitrogen immediately before grinding; fresh tissue from greenhouse-grown offspring was kept at -80C until flash-freezing and grinding. DNA yield was quantified using a Quant-iT DNA quantification kit (Invitrogen P11496) and plate reader, then normalized to equal concentrations for library preparation.

To prepare libraries for Illumina sequencing, we used a low-coverage Tagmentation approach [151,152]; our protocol is available at [153]. Briefly, Tn5 enzyme was purified in bulk and pre-loaded with universal Illumina adapters following [154]. We then added the loaded Tn5 to approximately 1 ng of sample DNA and incubated to fragment DNA and add universal adapters. Next, we added OneTaq Hot Start polymerase (New England Biolabs M0488L) along with combinatorically barcoded forward and reverse primers and ran 18 cycles of PCR to amplify fragments. After PCR, samples were pooled into sets of 48 and cleaned using SPRI magnetic beads. These sets of 48 samples were quantified with a Qubit fluorometer and then merged in equimolar amounts for Illumina sequencing. Samples were sequenced at the Duke University Center for Genomic and Computational Biology using an Illumina NovaSeq 6000 machine to generate paired-end 150 bp reads with a targeted depth of ~1X coverage per sample. Reads were demultiplexed based on their combinatorial barcodes into individual samples.

Samples were prepared and sequenced across three separate runs: run 1 included a smaller test set of 31 maternal samples from 2019; run 2 included the remaining 2019 maternal and offspring samples (including re-sequencing the original 31) plus all 2021 maternal and offspring samples, and run 3 included all 2022 maternal and offspring samples. After a preliminary analysis on the 31 original samples, with data from runs 1 and 2 treated separately, one sample was excluded because hybrid index values (see Assigning local ancestry) obtained from the two runs did not agree; for the remaining 30 twice-sequenced samples, we concatenated their raw read files across both runs for the final analysis. From our full dataset, we also excluded 80 offspring samples whose maternal plant was not sequenced, and 1 maternal family (mother and offspring) with missing location data, leaving a total of 2946 samples. Across these samples, we sequenced ~4.8 billion read pairs, for an average of 1.64 million read pairs per sample (standard deviation 1.57 million read pairs). After filtering to remove samples with less than 25,000 called ancestry-informative sites (see Assigning local ancestry), our final dataset included 2706 samples with mean sequencing depth of 1.78 million read pairs per sample (standard deviation 1.56 million read pairs, range 45,098 read pairs to 15.72 million read pairs). This final dataset included 458 wild-collected maternal samples and 2248 greenhouse-grown offspring. Details about this final dataset are provided in Table 1.

Reference alignment, genotyping, and creation of SNP panels

For each Illumina sample, we used Trimmomatic v0.39 [155] to remove adapter sequences and low-quality ends. We aligned reads with bwa v0.7.17 ‘mem’ [156] to the Mimulus guttatus IM62 v3 reference genome (https://phytozome-next.jgi.doe.gov), removed duplicates with picard v2.21.6 ‘MarkDuplicates’ [157], and used samtools v1.10 [158] to keep only properly paired reads with map quality>=29. Coverage stats were obtained with qualimap v2.2.1 [159]. Across all 2946 samples, 83% of reads mapped successfully, for an average per-base genome coverage of 0.44 and an average of 14.0% of the genome covered by at least one read. For the 2706 samples retained after filtering (see below), average per-base genome coverage was 0.48 and an average of 15.3% of the genome was covered by at least one read.

We created panels of high-quality, informative SNPs using 36 previously sequenced lines (S5 Table), including one M. nasutus and 8 admixed M. guttatus lines from CAC, plus 3 additional M. nasutus and 24 allopatric M. guttatus from throughout the species’ ranges. M. nasutus has lower range-wide nucleotide diversity (π_syn~1.1%) than M. guttatus (π_syn~4.9%) so the smaller M. nasutus sample is informative [89]. We followed the same steps above to align these Illumina panel lines to the Mimulus guttatus IM62 v3 reference genome. The panel was genotyped using GATK 4.4.0.0 HaplotypeCaller and GenotypeGVCFs in all-sites mode [160]. Called sites were split into biallelic SNPs and invariant sites, with indels and multiallelic sites removed. SNPs were further filtered with GATK to remove sites that failed a set of standard quality filters (details in the associated code at github.com/mfarnitano/CAC_popgen). Sites were further filtered at the individual genotype level using vcftools v0.1.16 [161], setting genotypes to missing if individual sequencing depth was below 6 or above 100, or if genotype quality was less than 15 for that sample. Heterozygous calls were retained. From the resulting genotype file, we created a list of 3,493,514 SNPs called in at least 31 of the 36 reference individuals.

For a more targeted genotyping panel, we created a test set of 100 representative wild CAC+LM samples, then subset our variant list to 19,633 sites with nonzero read coverage in at least 60% of test samples and a minor allele frequency of at least 20%. This panel should include variants that are segregating in our focal populations, including both species-diagnostic and intraspecific segregating sites, and is tailored for relatively high coverage in our otherwise low-coverage dataset.

In addition, we created a panel of ancestry-informative sites distinguishing M. guttatus and M. nasutus, subsetting our full variant list to 208,560 SNPs with>=80% allele frequency difference between the 24 high-coverage allopatric M. guttatus lines and the four high-coverage M. nasutus lines. This panel should include only variants that are species-informative, but includes sites that may have low coverage across our full dataset. Given the high genetic diversity of M. guttatus relative to M. nasutus and the fact that M. nasutus variation is nested within range-wide M. guttatus variation [89], any one variant is not species-diagnostic and may be shared through ancestral polymorphism; however, the combination of multiple informative variants across a region, especially at high LD within M. nasutus ancestry blocks, can be diagnostic for local species ancestry [89,162].

Population structure analyses with ANGSD

We used ANGSD v0.940 [163] with the GATK likelihood model to obtain genotype likelihoods for all wild-collected samples at the targeted variant list of 19,633 SNPs. We then used these genotype likelihoods to run a genomic PCA analysis with PCAngsd [164]. The resulting covariance matrix was converted into principal components using the eigen function in R [165]. As a complementary approach to our ancestry HMM (below), we used NGSadmix [166] with K=2 groups to estimate admixture proportions using the ANGSD genotype likelihoods.

Assigning local ancestry

We used ancestry_HMM [162] to assign M. guttatus vs. M. nasutus ancestry across the genome for each maternal and offspring individual, following the ancestryinfer pipeline from [167] and using our 208,560 ancestry-informative sites (above). Reads for each maternal and offspring sample were aligned to both the Mimulus guttatus IM62 v3 reference and the Mimulus nasutus SF v2 reference (https://phytozome-next.jgi.doe.gov), keeping only reads that aligned exactly once to each genome. At each ancestry-informative site, we counted the number of reads supporting each allele. These read counts per allele were used as input for ancestry_HMM [162], along with allele frequencies in the 24 allopatric M. guttatus and 4 M. nasutus reference lines. Recombination rates between each SNP position were approximated using the bp distance between adjacent sites multiplied by the global recombination rate estimate of 3.9e-8 Morgans/bp, calculated using a total genetic map length of 14.7 Morgans [89] across a genome size of 375 Mb, approximately the mean genome size between M. guttatus (~430 Mb) and M. nasutus (~320 Mb). Ancestry_HMM was run with a simple model of two source populations and a single admixture pulse (specific parameters available within the code at github.com/mfarnitano/CAC_popgen) to estimate posterior probabilities of M. guttatus, M. nasutus, or heterozygous ancestry at each site for each sample. Posterior probabilities of at least 0.9 for any genotype were converted to hard genotype calls, with lower values set to missing. 240 samples with fewer than 25,000 called genotypes (out of 208,560 sites) were excluded from all analyses, leaving a final dataset of 2706 samples from an initial set of 2946 samples (Table 1). Hybrid index (HI) equals the number of sites with homozygous M. nasutus calls plus half the number of heterozygous calls, divided by the total number of called sites. Ancestry heterozygosity (AH) equals the number of heterozygous calls divided by the total number of called sites.

To test the reliability of our local ancestry measurements, we subset the raw reads from each allopatric M. guttatus, M. nasutus, and high-coverage CAC line (36 lines total) to 2 million read pairs per sample, and followed the above local ancestry pipeline to call ancestry across the genomes in these lines. All 4 M. nasutus lines had hybrid index >0.99; 18 of 24 allopatric M. guttatus lines had HI<0.01 and the remaining six had HI<0.05. The 8 high-coverage CAC M. guttatus were more admixed, with hybrid index ranging from 0.11 to 0.34, consistent with previous ancestry estimates for these lines [44].

Our data include 11 maternal samples from plots S2_1 and S2_2, all from 2022, that we determined to be the allopolyploid species M. sookensis. These individuals had hybrid indices near 0.5 and ancestry heterozygosity >0.85, which could indicate polyploidy or new F1 hybridization. Offspring of these individuals had very similar hybrid indices and similarly high ancestry heterozygosity, which is consistent with fixed heterozygosity in these highly selfing polyploids [98,168,169] but would not be true for the F2 progeny of F1 hybrids. In addition, these individuals formed a distinct cluster in PCA space: they diverged from the 1–1 correlation between hybrid index and PC1, and they were separated cleanly by PC5. Finally, offspring grown in the greenhouse had a small flower size and shape that matched M. sookensis and M. nasutus; F1 hybrids typically have larger flowers more similar to M. guttatus [95]. We removed all M. sookensis individuals from the analyses of reproductive isolation, individual phenology, mating system, and offspring ancestry deviations.

We combined our data with previously obtained hybrid indices from 75 individuals collected in 2012 from CAC_Stream1 and CAC_Stream2 [90]. These samples were genotyped using MSG sequencing [170], and hybrid indices were calculated from an HMM approach implemented in HapMix [171]; note that these are different data types and were processed differently than our 2019–2022 data. However, as both approaches use low-coverage whole-genome sequencing and HMM ancestry calling, we expect them to produce broadly comparable hybrid indices even if there are slight differences at individual loci.

Organellar haplotype networks

To investigate the haplotype structure of maternally transmitted organellar genomes in our population, we aligned each of our maternal samples, along with 8 high-coverage CAC lines (S5 Table), to the M. guttatus IM767 v1.1 chloroplast assembly (https://phytozome-next.jgi.doe.gov) using BWA v0.7.17 [156], and called genotypes using GATK v4.4.0.0 HaplotypeCaller and GenotypeGVCFs functions [160]. Read coverage aligned to the chloroplast genome was high compared to the nuclear genome, so we restricted each analysis to samples with mean coverage>=40, resulting in 328 of our samples and 6 of the previously sequenced CAC lines. We then subset each VCF to a set of SNPs from a recent analysis of organellar haplotypes across M. guttatus, M.nasutus, M. decorus, and M. sookensis [98], further removing any SNPs with missing data, heterozygous calls, or individual depth <4 for any sample using BCFtools v1.15.1 [158]. We merged this dataset with the genotype calls for the 41 samples from [98], output SNP calls as a fasta sequence for each sample using BCFtools consensus, and converted to nexus format using EMBOSS v6.6.0 [172]. We then used the Integer NJ Net function in popart [173] to generate a haplotype network from the resulting alignment. Our final chloroplast dataset included 102 segregating sites, of which 43 were parsimony-informative. We followed this same approach to generate a mitochondrial genome haplotype network, aligning sequences to an M. guttatus IM62 mitochondrial assembly [174] and excluding regions annotated as chloroplast-derived. After filtering for mean coverage>=40, we retained 150 of our samples and 8 previously sequenced CAC lines, which were merged with the 41 samples used in the recent analysis. Our final mitochondrial dataset included 120 segregating sites, of which 39 were parsimony-informative.

Estimating selfing rates

To estimate selfing and outcrossing rates, we started with 310 maternal families containing 1,993 offspring where both the maternal sample and at least one offspring sample had at least 25,000 sites with ancestry assignments (see Assigning local ancestry above). Then, we excluded maternal-offspring pairs with a large number of genotypes that contradict a maternal-offspring relationship, suggestive of high ancestry error rates in either the maternal or offspring sample, as follows. For each maternal-offspring pair, we used ancestry calls from ancestry_HMM (at 208,560 sites) to calculate the number of informative sites where genotypes were incompatible with maternal-offspring relationships (i.e., maternal and offspring samples were homozygous for opposite ancestries), divided by the total number of homozygous maternal calls with a called genotype in the offspring. Allowing for some uncertainty in ancestry calling, we excluded any offspring for which this incompatibility ratio was>=5%, as well as entire families if greater than half the offspring were excluded. After these filters, our final family dataset contained 185 maternal plants (60% of the original dataset) with offspring, plus 1565 offspring (79% of the original dataset). We note that this filter is designed to remove maternal or offspring samples with poor genotyping quality (high error rates), since selfing rate estimation is sensitive to such errors. The filter is purposefully strict: because our initial ancestry assignments use a posterior probability cutoff of 90%, for any pair of samples, a given site may have up to a 0.1+(0.9*0.1)=19% chance that at least one of the two samples has an incorrect ancestry assignment, prior to this filter. As such, failing this maternal-offspring filter does not necessarily imply that the offspring was incorrectly assigned to the mother, though the filter should remove any such cases as well.

We then randomly thinned our list of ancestry-informative sites to one per 10kb, leaving 17,514 sites, and calculated genotype likelihoods at each site for both maternal and offspring samples using ANGSD v0.940 with the GATK likelihood model [163]. Genotype likelihoods estimated in this way can be biased towards homozygosity when using low-coverage data, because only a single allele is typically sampled at each site. Furthermore, when short fragments are sequenced, forward and reverse paired-end reads can overlap, and ANGSD genotype likelihoods count this single fragment as two separate reads, artificially inflating the likelihood of a homozygous genotype. Since the estimation of selfing rates is sensitive to the likelihood of homozygous vs. heterozygous calls, we chose to mitigate this effect by using alignments of the forward reads only, discarding reverse reads, in our genotype likelihood estimation for selfing rate analyses. We estimated probabilities of selfing vs. outcrossing for each maternal-offspring pair, and probabilities of half-sibling vs. full-sibling relationships for each outcrossed sibling pair, using the Bayesian model implemented in BORICE.genomic v3 [124]. We tuned allele frequencies for 20 steps, then had 20 steps of burn-in and a total chain length of 100, with records kept every 2 steps. Offspring with a posterior probability>= 0.9 were called as selfed or outcrossed, and other offspring were excluded. We counted pairs as half-siblings if the posterior probability was >50%, full-siblings if it was <50%, and missing data if the posterior probability was exactly 50%.

Calculating the strength of reproductive isolation

We calculated standardized estimates [175] of reproductive isolation (RI) between three ancestry cohorts within CAC_Stream1, defined by hybrid index: an M. guttatus (lower admixture levels <0.15) cohort, an admixed (higher admixture levels >0.15), and an M. nasutus cohort. We calculated two components of premating reproductive isolation: mating system isolation, the degree to which self-fertilization reduces cross-cohort mating; and phenological isolation, the degree to which phenological non-overlap reduces cross-cohort mating; we also calculated their combined effects. Each measure ranges from -1 (complete disassortative mating) to 1 (complete assortative mating), with 0 indicating random mating. For CAC_Stream2, we calculated mating system isolation between M. guttatus and admixed cohorts only (due to a lack of M. nasutus sampling), using the same hybrid index cutoff as CAC_Stream1, though this cutoff may not be meaningful for this stream given its lack of cohort structure. Phenological RI was not calculated for LM due to a lack of the relevant cohort or plot structure; also, LM was only sampled in 2021, a year in which we did not collect phenological data. We did not calculate mating system isolation for LM since almost all (33/35 or 94%) available maternal families after filtering fell into the same admixed hybrid index cohort.

For mating system isolation, we used selfing rates estimated from BORICE, dividing individuals into three cohorts by hybrid index. We assumed for this calculation that outcrossing events contribute no isolation to the maternal cohort (RI=0), while selfing events contribute perfect isolation to their maternal cohort (RI=1). The degree of mating system isolation is then equal to the rate of selfing of maternal plants in each cohort. While selfing flowers may also contribute less to the pollen pool, causing increased mating system isolation from the paternal direction, this indirect effect is not included in our calculation.

To calculate phenological isolation between cohorts within streams CAC_Stream1 and CAC_Stream2, we utilized plot-level phenological data, assigning each plot to an ancestry cohort according to median hybrid index. For CAC_Stream1, we excluded two plots (S1_1 and S1_7) with apparently bimodal hybrid indices indicating the presence of multiple cohorts. CAC_Stream1 plots were assigned as follows: S1_5 and S1_8 representing M. guttatus, S1_2 and S1_4 representing admixed, and S1_3 and S1_6 representing M. nasutus. For CAC_Stream2, we assigned plots to two cohorts, M. nasutus (plots S2_1 and S2_2) and admixed (plots S2_3 to S2_7). For CAC_Stream2, we did not calculate phenology in 2012 because the only sampled M. nasutus plot in that year (S2_2) also had a substantial number of admixed individuals. For each pair of cohorts in a stream, we calculated the probability of within-cohort vs. between-cohort matings based on the number of open flowers in each plot group on each day: for each open flower, between-cohort mating probability is equal to B/(A+B) and within-plot mating probability is equal to A/(A+B), where A and B are the number of open flowers on the same day as the focal flower from the same and opposing plots, respectively. This assumes that each open flower contributes equally to the pollen pool and ignores the effects of physical distance. These probabilities were then summed across all open flowers across the season to obtain an estimated total number of between-cohort and within-cohort potential matings. Reproductive isolation was then calculated as RI=1-2P, where P is the relative probability of these between-cohort vs. within-cohort potential matings.

Following [175], we calculated the combined effect of mating system (RI_M) and phenological (RI_P) isolation using Equation 1 below, for any comparison where we had both values. Note that the order of isolating barriers does not matter for this calculation.

Equation 1. combined reproductive isolation

Using offspring to assess assortative mating

If assortative mating by ancestry is strong, we expect offspring hybrid index to closely match maternal hybrid index; in contrast, deviations between offspring and maternal hybrid index may indicate mating across different ancestry cohorts. To test for deviations from maternal hybrid index within the offspring, we first separated each fruit into selfed and outcrossed offspring based on BORICE. We then calculated the mean for each fruit of the difference between Offspring HI and Maternal HI (Offspring HI deviation), averaging selfed and outcrossed offspring separately. We expect a deviation near 0 for selfed offspring, or under strong assortative mating by ancestry; positive values indicate that the offspring have more M. nasutus ancestry than their maternal parent, while negative values indicate they have more M. guttatus ancestry than their maternal parent.

Statistical models

All statistical test were carried out using generalized additive models (GAMs) implemented with the R packages ‘gam’ [176] and ‘mgcv’ [177]. For each model below, we sequentially added independent variables in the order listed, starting with an intercept-only model and allowing for interactions, and retained a variable only if the variable had a significant main or interaction effect and it significantly improved the model fit according to a likelihood ratio test. All model details are provided in S2 Table. GAMs were used because they could handle a variety of model types (beta, binomial, and linear) and allow for incomplete factor interactions (e.g., not all combinations of stream and year were present in the data).

To test for predictors of selfing rates, we ran binomial GAMs (family = binomial, link = logit) with selfing rate as the dependent variable and the following independent variables: maternal hybrid index, stream (CAC_Stream1 or CAC_Stream2), and year (2019, 2021, or 2022). We ran these models for all families, then for all families except M. nasutus (HI>0.95) and for only families with maternal HI<0.5 (i.e., excluding the six majority-M. nasutus admixed individuals).

To test whether phenology was associated with hybrid index at the plot level, we paired flower censuses throughout the growing season from 2012, 2019, and 2022 with hybrid indices from our sequenced maternal plants from those plots. For each plot in each year, we calculated the date that the median censused flower was open in that plot. We then took the median hybrid index of all sequenced individuals from that plot. We ran beta regression GAMs (family = betar, link = logit) to test for an effect of median flower date, stream, and year on median hybrid index.

We also tested whether flower date was associated with hybrid index at the individual level. For sequenced individuals in 2019 and 2022, we have dates that marked flowers were open – note that these are not necessarily the first flowers on a given individual but were instead randomly chosen flowers from the plot on the day of marking. We ran beta regression GAMs (family = betar, link=logit) to test for an effect of individual marked flower date, stream, and year on maternal hybrid index.

We ran linear GAMs (family = gaussian, link = identity) to test for an effect of maternal hybrid index, stream, and year on offspring HI deviation (the difference between offspring hybrid index and maternal hybrid index, averaged per fruit). We ran separate linear models for the selfed offspring and for the outcrossed offspring, as well as a combined model with ‘mate type’ (selfed or outcrossed) as an additional predictor, added before stream and year. We also ran an ANOVA on each linear model using the ‘anova’ function in the R package ‘stats’ [165] to assess the relative effects of each predictor and interaction variable.

Supporting information

S1 Fig. Correlation between hybrid index and NGSAdmix structure results.

NGSAdmix value is the proportion assignment to one of two clusters by NGSAdmix with K=2; hybrid index is the proportion of sites with M. nasutus ancestry from local ancestry inference using Ancestry_HMM. The circled ‘SOOK’ cluster that deviates from the 1–1 line is composed of M. sookensis polyploid individuals (see Materials and Methods). ‘NAS’ = M. nasutus.

https://doi.org/10.1371/journal.pgen.1011624.s001

(TIFF)

S2 Fig. Ancestry heterozygosity vs. hybrid index for maternal and offspring plants.

Ancestry_HMM hybrid index and ancestry heterozygosity outputs for A) maternal and B) offspring individuals. Hybrid index or HI is the proportion of ancestry-informative sites called with M. nasutus ancestry from Ancestry_HMM. Ancestry heterozygosity or AH is the proportion of ancestry-informative sites called has heterozygous (out of the total number of called ancestry-informative sites). First-generation hybrids between 100% M. guttatus and 100% M. nasutus would have an expected HI=0.5 and AH=1.0; their offspring would be expected to have lower ancestry heterozygosity (~0.5 if selfed). The circled ‘SOOK’ cluster highlights a group of polyploid M. sookensis maternal plants; their offspring have similarly high AH values, which indicates fixed heterozygosity (polyploidy) rather than diploid F1 status.

https://doi.org/10.1371/journal.pgen.1011624.s002

(TIFF)

S3 Fig. PC5 identification of M. sookensis.

Principal components 1 and 5 from PCAngsd analysis (from the same analysis as PCs 1–4 shown in Fig 2A and 2B). PC1 correlates strongly with M. guttatus vs. M. nasutus ancestry (Fig 2B). PC5 clearly delineates a group of individuals identified as the polyploid species M. sookensis.

https://doi.org/10.1371/journal.pgen.1011624.s003

(TIFF)

S4 Fig. PCA analysis of only M. guttatus and admixed individuals from CAC_Stream1 and CAC_Stream2.

A) PC axis 1 vs. hybrid index, demonstrating that ancestry still drives the major axis of genomic variation in the dataset even without M. nasutus samples. B) PC axes 1 and 2, demonstrating some separation of certain CAC_Stream1 plots (particularly S1_1 and S1_2) from the rest of the data. C) PC axes 3 and 4, showing further variation within CAC_Stream1 that is partitioned by plot identity. Note that CAC_Stream2 is not differentiated from CAC_Stream1 in these first four axes, but the major axes of variation instead separate variation within CAC_Stream1. LM samples, M. nasutus samples, and M. sookensis samples were excluded prior to running thi PCA.

https://doi.org/10.1371/journal.pgen.1011624.s004

(TIFF)

S5 Fig. Complete mitochondrial capture of a M. nasutus haplotype in sympatric M. guttatus

. NJ-Net haplotype network of mitochondrial variation built from 154 CAC and LM maternal samples sequenced in this study (bolded captions) and 49 previously sequenced samples (non-bolded captions) from the M. guttatus species complex [98], using 120 total variant sites (39 parsimony-informative). The mitochondrial network largely agrees with the chloroplast network (Fig 3), with a single haplotype present in all maternal samples from Catherine Creek and Little Maui, including M. guttatus, admixed, M. nasutus, and M. sookensis samples. All other M. nasutus samples from across the range share this haplotype or a close derivative, as do samples from M. sookensis (a polyploid with M. nasutus as maternal parent). M. guttatus haplotypes are more variable, with only one sample not from CAC or LM sharing the M. nasutus haplotype. M. decorus is another member of the M. guttatus species complex with variable mitochondrial haplotypes.

https://doi.org/10.1371/journal.pgen.1011624.s005

(TIFF)

S6 Fig. Selfing rate vs. hybrid index.

Proportion of selfed vs. outcrossed offspring per maternal family vs. maternal hybrid index (proportion M. nasutus ancestry of the maternal plant), showing (A) a binomial model fit including M. nasutus maternal families, and (B) a binomial model fit excluding M. nasutus families. Self vs. outcross determined by the BORICE Bayesian model. Offspring without>=90% posterior probability of either state were removed. Lines and shaded regions indicate model fits and 95% confidence intervals from a binomial GAM regression using stream and year as additional factor variables (S2 Table).

https://doi.org/10.1371/journal.pgen.1011624.s006

(TIFF)

S7 Fig. Comparison of hybrid index, phenology, and maximum floral count for each plot within CAC_Stream1.

A) Histograms of binned hybrid index for sampled maternal plants in each CAC_Stream1 plot, with years as stacked colors. 0=M. guttatus, 1=M. nasutus. B) Phenology (counted open flowers) throughout the flowering season in seven plots within CAC_Stream1 during the 2012, 2019, and 2022 seasons, scaled to the maximum number of open flowers counted within that plot in that year. Plots with primarily M. nasutus individuals (plots S1_1,S1_3, and S1_6) had early flowers, plots with more admixed individuals (plots S1_2, S1_4, and S1_7) had intermediate peak flowering, and plots with primarily M. guttatus (plots S1_5 and S1_8) had later peak flowering. Peak flowering was similar but not identical across years. Tails of open flowers indicate that in 2022, plots with admixed individuals (plots S1_1,S1_2, and S1_4) continued flowering for much longer than in 2019, resulting in more overlap with M. guttatus plots (plots S1_5 and S1_8). C) Maximum number of open flowers on any one day for each plot. In 2019, there were far fewer open flowers throughout the growing season than in 2021 or 2022.

https://doi.org/10.1371/journal.pgen.1011624.s007

(TIFF)

S8 Fig. Comparison of hybrid index, phenology, and maximum floral count for each plot within CAC_Stream2.

D) Histograms of binned hybrid index for CAC_Stream2 plots. E) Phenology (counted open flowers) for CAC_Stream2 plots, scaled to the maximum number of open flowers counted within that plot in that year. F) Maximum number of open flowers for CAC_Stream2 plots.

https://doi.org/10.1371/journal.pgen.1011624.s008

(TIFF)

S1 Table. Distribution of samples from each plot in each PCA cluster.

https://doi.org/10.1371/journal.pgen.1011624.s009

(XLSX)

S2 Table. Statistical model output summaries.

https://doi.org/10.1371/journal.pgen.1011624.s010

(XLSX)

S3 Table. Summary of selfing and sibship estimation from BORICE.

https://doi.org/10.1371/journal.pgen.1011624.s011

(XLSX)

S4 Table. Plot names, locations and ID assignments for each year of sampling.

https://doi.org/10.1371/journal.pgen.1011624.s012

(XLSX)

S5 Table. Reference panel lines used for SNP panel creation.

https://doi.org/10.1371/journal.pgen.1011624.s013

(XLSX)

Acknowledgments

We thank Alex Sotola, Sara Hill, Autumn Knight, Liza Lesley, and Sherwin Shirazi for help processing samples, and Bob Schmitz for providing tagmentation enzyme for library preps. Thank you to Alex Sotola, Jill Anderson, Molly Schumer, Casey Bergman, and Kelly Dyer for advice during project development. We also thank Eleanore Ritter, Natalie Gonzalez, Logan Scott, and Sam Mantel for comments on an earlier draft of the manuscript.

References

1. Dobzhansky Th. Studies on hybrid sterility. Z Für Zellforsch Mikrosk Anat. 1934;21(2):169–223.
- View Article
- Google Scholar
2. Farnitano M, Sweigart A. Strong postmating reproductive isolation in Mimulus section Eunanus. Journal of Evolutionary Biology. 2023;36(10):1393–410.
- View Article
- Google Scholar
3. Richards TJ, Ortiz-Barrientos D. Immigrant inviability produces a strong barrier to gene flow between parapatric ecotypes of Senecio lautus. Evolution. 2016;70(6):1239–48.
- View Article
- Google Scholar
4. Scopece G, Widmer A, Cozzolino S. Evolution of postzygotic reproductive isolation in a guild of deceptive orchids. Am Nat. 2008;171(3):315–26. pmid:18198999
- View Article
- PubMed/NCBI
- Google Scholar
5. Stebbins GL. The role of hybridization in evolution. Proc Am Philos Soc. 1959;103(2):231–51.
- View Article
- Google Scholar
6. Wodsedalek JE. Causes of sterility in the mule. Biol Bull. 1916;30(1):1–56.
- View Article
- Google Scholar
7. Xiong T, Mallet J. On the impermanence of species: the collapse of genetic incompatibilities in hybridizing populations. Evolution. 2022;76(11):2498–512. pmid:36097352
- View Article
- PubMed/NCBI
- Google Scholar
8. Behm JE, Ives AR, Boughman JW. Breakdown in postmating isolation and the collapse of a species pair through hybridization. Am Nat. 2010;175(1):11–26. pmid:19916869
- View Article
- PubMed/NCBI
- Google Scholar
9. Kleindorfer S, O’Connor JA, Dudaniec RY, Myers SA, Robertson J, Sulloway FJ. Species collapse via hybridization in Darwin’s tree finches. Am Nat. 2014;183(3):325–41. pmid:24561597
- View Article
- PubMed/NCBI
- Google Scholar
10. Schumer M, Rosenthal G, Andolfatto P. How common is homoploid hybrid speciation?. Evolution. 2014;68(6):1553–60.
- View Article
- Google Scholar
11. Ungerer MC, Baird SJ, Pan J, Rieseberg LH. Rapid hybrid speciation in wild sunflowers. Proc Natl Acad Sci U S A. 1998;95(20):11757–62. pmid:9751738
- View Article
- PubMed/NCBI
- Google Scholar
12. Nice CC, Gompert Z, Fordyce JA, Forister ML, Lucas LK, Buerkle CA. Hybrid speciation and independent evolution in lineages of alpine butterflies. Evolution. 2013;67(4):1055–68. pmid:23550755
- View Article
- PubMed/NCBI
- Google Scholar
13. Rosser N, Seixas F, Queste LM, Cama B, Mori-Pezo R, Kryvokhyzha D, et al. Hybrid speciation driven by multilocus introgression of ecological traits. Nature. 2024;628(8009):811–7. pmid:38632397
- View Article
- PubMed/NCBI
- Google Scholar
14. Christie K, Fraser L, Lowry D. The strength of reproductive isolating barriers in seed plants: Insights from studies quantifying premating and postmating reproductive barriers over the past 15 years. Evolution. 2022;
- View Article
- Google Scholar
15. Coyne JA, Orr HA. Patterns of speciation in drosophila revisited. Evolution. 1997;51(1):295–303. pmid:28568795
- View Article
- PubMed/NCBI
- Google Scholar
16. Malone JH, Fontenot BE. Patterns of reproductive isolation in toads. PLoS One. 2008;3(12):e3900. pmid:19065271
- View Article
- PubMed/NCBI
- Google Scholar
17. Moyle LC, Olson MS, Tiffin P. Patterns of reproductive isolation in three angiosperm genera. Evolution. 2004;58(6):1195–208. pmid:15266970
- View Article
- PubMed/NCBI
- Google Scholar
18. Lotsy JP. On the species of the taxonomist in its relation to evolution. Genetica. 1931;13(1–2):1–16.
- View Article
- Google Scholar
19. Suarez-Gonzalez A, Lexer C, Cronk QCB. Adaptive introgression: a plant perspective. Biol Lett. 2018 Mar 14;14(3):20170688.
- View Article
- Google Scholar
20. Goulet BE, Roda F, Hopkins R. Hybridization in plants: old ideas, new techniques. Plant Physiol. 2017;173(1):65–78. pmid:27895205
- View Article
- PubMed/NCBI
- Google Scholar
21. Mallet J, Besansky N, Hahn MW. How reticulated are species?. Bioessays. 2016;38(2):140–9. pmid:26709836
- View Article
- PubMed/NCBI
- Google Scholar
22. Arnold ML, Kunte K. Adaptive genetic exchange: a tangled history of admixture and evolutionary innovation. Trends Ecol Evol. 2017;32(8):601–11. pmid:28645486
- View Article
- PubMed/NCBI
- Google Scholar
23. Chhatre VE, Evans LM, DiFazio SP, Keller SR. Adaptive introgression and maintenance of a trispecies hybrid complex in range-edge populations of Populus. Molecular Ecology. 2018;27(23):4820–38.
- View Article
- Google Scholar
24. Manzoor SA, Griffiths G, Obiakara MC, Esparza-Estrada CE, Lukac M. Evidence of ecological niche shift in Rhododendron ponticum (L.) in Britain: hybridization as a possible cause of rapid niche expansion. Ecol Evol. 2020;10(4):2040–50. pmid:32128136
- View Article
- PubMed/NCBI
- Google Scholar
25. Pfennig KS, Kelly AL, Pierce AA. Hybridization as a facilitator of species range expansion. Proc Royal Soc B: Biol Sci. 2016;283(1839):20161329.
- View Article
- Google Scholar
26. Baskett ML, Gomulkiewicz R. Introgressive hybridization as a mechanism for species rescue. Theoretical Ecol. 2011;4(2):223–39.
- View Article
- Google Scholar
27. Oziolor EM, Reid NM, Yair S, Lee KM, Guberman VerPloeg S, Bruns PC, et al. Adaptive introgression enables evolutionary rescue from extreme environmental pollution. Science. 2019;364(6439):455–7. pmid:31048485
- View Article
- PubMed/NCBI
- Google Scholar
28. Ayres DR, Zaremba K, Strong DR. Extinction of a common native species by hybridization with an invasive congener. Weed Technol. 2004;18(sp1):1288–91.
- View Article
- Google Scholar
29. Rhymer JM, Simberloff D. Extinction by hybridization and introgression. Ann Rev Ecol Evol Syst. 1996;27(1):83–109.
- View Article
- Google Scholar
30. Quilodrán CS, Tsoupas A, Currat M. The spatial signature of introgression after a biological invasion with hybridization. Front Ecol Evol. 2020;8.
- View Article
- Google Scholar
31. Wolfe L, Blair A, Penna B. Does intraspecific hybridization contribute to the evolution of invasiveness?: an experimental test. Biol Invasions. 2007;9(5):515–21.
- View Article
- Google Scholar
32. Goodman SJ, Barton NH, Swanson G, Abernethy K, Pemberton JM. Introgression through rare hybridization: a genetic study of a hybrid zone between red and sika deer (genus Cervus) in Argyll, Scotland. Genetics. 1999;152(1):355–71. pmid:10224266
- View Article
- PubMed/NCBI
- Google Scholar
33. Green R, Krause J, Briggs A, Maricic T, Stenzel U, Kircher M. A draft sequence of the Neandertal genome. Science. 2010;328(5979):710–22.
- View Article
- Google Scholar
34. Randi E, Lucchini V. Detecting rare introgression of domestic dog genes into wild wolf (Canis lupus) populations by Bayesian admixture analyses of microsatellite variation. Conserv Genet. 2002 Mar 1;3(1):29–43.
- View Article
- Google Scholar
35. Harrison R, Larson E. Heterogeneous genome divergence, differential introgression, and the origin and structure of hybrid zones. Mol Ecol. 2016;25(11):2454–66.
- View Article
- Google Scholar
36. Hasselman DJ, Argo EE, McBride MC, Bentzen P, Schultz TF, Perez-Umphrey AA, et al. Human disturbance causes the formation of a hybrid swarm between two naturally sympatric fish species. Mol Ecol. 2014;23(5):1137–52. pmid:24450302
- View Article
- PubMed/NCBI
- Google Scholar
37. Schumer M, Xu C, Powell DL, Durvasula A, Skov L, Holland C, et al. Natural selection interacts with recombination to shape the evolution of hybrid genomes. Science. 2018;360(6389):656–60. pmid:29674434
- View Article
- PubMed/NCBI
- Google Scholar
38. Yan L-J, Burgess KS, Zheng W, Tao Z-B, Li D-Z, Gao L-M. Incomplete reproductive isolation between Rhododendron taxa enables hybrid formation and persistence. J Integr Plant Biol. 2019;61(4):433–48. pmid:30192058
- View Article
- PubMed/NCBI
- Google Scholar
39. Larson EL, Andrés JA, Bogdanowicz SM, Harrison RG. Differential introgression in a mosaic hybrid zone reveals candidate barrier genes. Evolution. 2013;67(12):3653–61.
- View Article
- Google Scholar
40. Martinsen G, Whitham T, Turek R, Keim P. Hybrid populations selectively filter gene introgression between species. Evolution. 2001;55(7):1325–35.
- View Article
- Google Scholar
41. Fishman L, Sweigart AL. When Two Rights Make a Wrong: The Evolutionary Genetics of Plant Hybrid Incompatibilities. Annu Rev Plant Biol. 2018;69:707–31. pmid:29505737
- View Article
- PubMed/NCBI
- Google Scholar
42. Muller H. Isolating mechanisms, evolution, and temperature. Biological Symposium. 1942;6:71–125.
- View Article
- Google Scholar
43. Orr HA, Turelli M. The evolution of postzygotic isolation: accumulating Dobzhansky-Muller incompatibilities. Evolution. 2001;55(6):1085–94. pmid:11475044
- View Article
- PubMed/NCBI
- Google Scholar
44. Mantel SJ, Sweigart AL. Postzygotic barriers persist despite ongoing introgression in hybridizing Mimulus species. Mol Ecol. 2024;33(4):e17261. pmid:38174628
- View Article
- PubMed/NCBI
- Google Scholar
45. Shuker DM, Underwood K, King TM, Butlin RK. Patterns of male sterility in a grasshopper hybrid zone imply accumulation of hybrid incompatibilities without selection. Proc Biol Sci. 2005;272(1580):2491–7. pmid:16271974
- View Article
- PubMed/NCBI
- Google Scholar
46. Zuellig MP, Sweigart AL. A two-locus hybrid incompatibility is widespread, polymorphic, and active in natural populations of Mimulus. Evolution. 2018;72(11):2394–405. pmid:30194757
- View Article
- PubMed/NCBI
- Google Scholar
47. Servedio MR, Hermisson J. The evolution of partial reproductive isolation as an adaptive optimum. Evolution. 2020;74(1):4–14. pmid:31721186
- View Article
- PubMed/NCBI
- Google Scholar
48. Borge T, Lindroos K, Nádvorník P, Syvänen A-C, Saetre G-P. Amount of introgression in flycatcher hybrid zones reflects regional differences in pre and post-zygotic barriers to gene exchange. J Evol Biol. 2005;18(6):1416–24. pmid:16313454
- View Article
- PubMed/NCBI
- Google Scholar
49. Brice C, Zhang Z, Bendixsen D, Stelkens R. Hybridization outcomes have strong genomic and environmental contingencies. Am Nat. 2021;198(3):E53–67. pmid:34403309
- View Article
- PubMed/NCBI
- Google Scholar
50. Otis JA, Thornton D, Rutledge L, Murray DL. Ecological niche differentiation across a wolf-coyote hybrid zone in eastern North America. Divers Distrib. 2017;23(5):529–39.
- View Article
- Google Scholar
51. Butlin RK, Saura M, Charrier G, Jackson B, André C, Caballero A, et al. Parallel evolution of local adaptation and reproductive isolation in the face of gene flow. Evolution. 2014 Apr 1;68(4):935–49.
- View Article
- Google Scholar
52. Lowry DB, Willis JH. A widespread chromosomal inversion polymorphism contributes to a major life-history transition, local adaptation, and reproductive isolation. PLoS Biol. 2010;8(9):e1000500. pmid:20927411
- View Article
- PubMed/NCBI
- Google Scholar
53. Melo MC, Grealy A, Brittain B, Walter GM, Ortiz-Barrientos D. Strong extrinsic reproductive isolation between parapatric populations of an Australian groundsel. New Phytol. 2014;203(1):323–34. pmid:24684207
- View Article
- PubMed/NCBI
- Google Scholar
54. Borzée A, Kim JY, DA Cunha MAM, Lee D, Sin E, Oh S, et al. Temporal and spatial differentiation in microhabitat use: implications for reproductive isolation and ecological niche specification. Integr Zool. 2016;11(5):375–87. pmid:27059098
- View Article
- PubMed/NCBI
- Google Scholar
55. Ebersole JP. Niche separation of two damselfish species by aggression and differential microhabitat utilization. Ecology. 1985;66(1):14–20.
- View Article
- Google Scholar
56. Ling TC, Phokasem P, Sinpoo C, Yang Y-P, Disayathanoowat T. Microhabitat and pollinator differentiation drive reproductive isolation between two sympatric salvia species (Lamiaceae). Plants (Basel). 2022;11(18):2423. pmid:36145824
- View Article
- PubMed/NCBI
- Google Scholar
57. Mantel S, Sweigart A. Divergence in drought‐response traits between sympatric species of Mimulus. Ecol Evol. 2019;
- View Article
- Google Scholar
58. Bank C, Bürger R, Hermisson J. The limits to parapatric speciation: Dobzhansky-Muller incompatibilities in a continent-island model. Genetics. 2012;191(3):845–63. pmid:22542972
- View Article
- PubMed/NCBI
- Google Scholar
59. Lenormand T. Gene flow and the limits to natural selection. Trends in Ecol Evol. 2002;17(4):183–9.
- View Article
- Google Scholar
60. Delph LF, Kelly JK. On the importance of balancing selection in plants. New Phytol. 2014;201(1):45–56. pmid:23952298
- View Article
- PubMed/NCBI
- Google Scholar
61. Levene H. Genetic equilibrium when more than one ecological niche is available. The American Naturalist. 1953;87(836):331–3.
- View Article
- Google Scholar
62. Kisel Y, Barraclough TG. Speciation has a spatial scale that depends on levels of gene flow. Am Nat. 2010;175(3):316–34. pmid:20100106
- View Article
- PubMed/NCBI
- Google Scholar
63. Klein EK, Lagache-Navarro L, Petit RJ. Demographic and spatial determinants of hybridization rate. J Ecol. 2017;105(1):29–38.
- View Article
- Google Scholar
64. Musker SD, Ellis AG, Schlebusch SA, Verboom GA. Niche specificity influences gene flow across fine-scale habitat mosaics in Succulent Karoo plants. Mol Ecol. 2021;30(1):175–92. pmid:33152114
- View Article
- PubMed/NCBI
- Google Scholar
65. Slatkin M. Gene flow and the geographic structure of natural populations. Science. 1987;236(4803):787–92. pmid:3576198
- View Article
- PubMed/NCBI
- Google Scholar
66. Martin NH, Willis JH. Ecological divergence associated with mating system causes nearly complete reproductive isolation between sympatric Mimulus species. Evolution. 2007;61(1):68–82. pmid:17300428
- View Article
- PubMed/NCBI
- Google Scholar
67. Roda F, Mendes F, Hahn M, Hopkins R. Genomic evidence of gene flow during reinforcement in Texas Phlox. Mol Ecol. 2017;26(8):2317–30.
- View Article
- Google Scholar
68. Sambatti JBM, Strasburg JL, Ortiz-Barrientos D, Baack EJ, Rieseberg LH. Reconciling extremely strong barriers with high levels of gene exchange in annual sunflowers. Evolution. 2012;66(5):1459–73. pmid:22519784
- View Article
- PubMed/NCBI
- Google Scholar
69. Abbott RJ. Plant invasions, interspecific hybridization and the evolution of new plant taxa. Trends Ecol Evol. 1992;7(12):401–5.
- View Article
- Google Scholar
70. Schwarzbach AE, Donovan LA, Rieseberg LH. Transgressive character expression in a hybrid sunflower species. Am J Bot. 2001;88(2):270–7. pmid:11222249
- View Article
- PubMed/NCBI
- Google Scholar
71. Selz OM, Seehausen O. Interspecific hybridization can generate functional novelty in cichlid fish. Proc R Soc B Biol Sci. 2019;286(1913):20191621.
- View Article
- Google Scholar
72. Hendry AP, Berg OK, Quinn TP. Condition dependence and adaptation-by-time: breeding date, life history, and energy allocation within a population of salmon. Oikos. 1999;85(3):499.
- View Article
- Google Scholar
73. Hood G, Zhang L, Hu E, Ott J, Egan S. Cascading reproductive isolation: Plant phenology drives temporal isolation among populations of a host-specific herbivore. Evolution. 2019;73(3):554–68.
- View Article
- Google Scholar
74. Osmolovsky I, Shifrin M, Gamliel I, Belmaker J, Sapir Y. Eco-Geography and Phenology Are the Major Drivers of Reproductive Isolation in the Royal Irises, a Species Complex in the Course of Speciation. Plants (Basel). 2022;11(23):3306. pmid:36501345
- View Article
- PubMed/NCBI
- Google Scholar
75. Franks SJ, Weis AE. Climate change alters reproductive isolation and potential gene flow in an annual plant. Evol Appl. 2009;2(4):481–8. pmid:25567893
- View Article
- PubMed/NCBI
- Google Scholar
76. Seehausen O, Takimoto G, Roy D, Jokela J. Speciation reversal and biodiversity dynamics with hybridization in changing environments. Mol Ecol. 2008;17(1):30–44. pmid:18034800
- View Article
- PubMed/NCBI
- Google Scholar
77. Sianta SA, Moeller DA, Brandvain Y. The extent of introgression between incipient Clarkia species is determined by temporal environmental variation and mating system. Proc Natl Acad Sci U S A. 2024;121(12):e2316008121. pmid:38466849
- View Article
- PubMed/NCBI
- Google Scholar
78. Abdul-Rahman F, Tranchina D, Gresham D. Fluctuating environments maintain genetic diversity through neutral fitness effects and balancing selection. Mole Biol Evol. 2021;38(10):4362–75.
- View Article
- Google Scholar
79. Bergland AO, Behrman EL, O’Brien KR, Schmidt PS, Petrov DA. Genomic evidence of rapid and stable adaptive oscillations over seasonal time scales in Drosophila. PLoS Genet. 2014;10(11):e1004775. pmid:25375361
- View Article
- PubMed/NCBI
- Google Scholar
80. Han G, Wang W, Dong Y. Effects of balancing selection and microhabitat temperature variations on heat tolerance of the intertidal black mussel Septifer virgatus. Integr Zool. 2020;15(5):416–27. pmid:32297470
- View Article
- PubMed/NCBI
- Google Scholar
81. Troth A, Puzey JR, Kim RS, Willis JH, Kelly JK. Selective trade-offs maintain alleles underpinning complex trait variation in plants. Science. 2018;361(6401):475–8. pmid:30072534
- View Article
- PubMed/NCBI
- Google Scholar
82. Grant BR, Grant PR. High survival of darwin’s finch hybrids: effects of beak morphology and diets. Ecology. 1996;77(2):500–9.
- View Article
- Google Scholar
83. Tataru D, Wheeler E, Ferris K. Spatially and temporally varying selection influence species boundaries in two sympatric Mimulus. Proc R Soc B: Biol Sci. 2023;290(1992):20222279.
- View Article
- Google Scholar
84. Carney SE, Gardner KA, Rieseberg LH. Evolutionary changes over the fifty-year history of a hybrid population of sunflowers (helianthus). Evol. 2000;54(2):462.
- View Article
- Google Scholar
85. Taylor S, Curry R, White T, Ferretti V, Lovette I. Spatiotemporally consistent genomic signatures of reproductive isolation in a moving hybrid zone. Evolution. 2014;68(11):3066–81.
- View Article
- Google Scholar
86. Moore WS, Buchanan EB. Stability of the northern flicker hybrid zone in historical times: implications for adaptive speciation theory. Evolution. 1985;39(1):135–51. pmid:28563641
- View Article
- PubMed/NCBI
- Google Scholar
87. Sullivan B. Temporal stability in hybridization between Bufo microscaphus and Bufo woodhousii (Anura: Bufonidae): Behavior and morphology. J Evol Biology. 1995;8(2):233–47.
- View Article
- Google Scholar
88. Hendry AP, Wenburg JK, Bentzen P, Volk EC, Quinn TP. Rapid evolution of reproductive isolation in the wild: evidence from introduced salmon. Science. 2000;290(5491):516–9. pmid:11039932
- View Article
- PubMed/NCBI
- Google Scholar
89. Brandvain Y, Kenney A, Flagel L, Coop G, Sweigart A. Speciation and introgression between Mimulus nasutus and Mimulus guttatus. PLoS Genetics. 2014;10(6):e1004410.
- View Article
- Google Scholar
90. Kenney AM, Sweigart AL. Reproductive isolation and introgression between sympatric Mimulus species. Mol Ecol. 2016;25(11):2499–517. pmid:27038381
- View Article
- PubMed/NCBI
- Google Scholar
91. Arathi HS, Kelly JK. Corolla Morphology Facilitates Both Autogamy and Bumblebee Pollination in Mimulus guttatus. Int J Plant Sci. 2004;165(6):1039–45.
- View Article
- Google Scholar
92. Kelly JK, Willis JH. A Manipulative Experiment to Estimate Biparental Inbreeding in Monkeyflowers. Int J Plant Sci. 2002;163(4):575–9.
- View Article
- Google Scholar
93. Macnair MR, Macnair VE, Martin BE. Adaptive speciation in Mimulus: an ecological comparison of M. cupriphilus with its presumed progenitor, M. guttatus. New Phytologist. 1989;112(2):269–79.
- View Article
- Google Scholar
94. Dole JA. Reproductive assurance mechanisms in three taxa of the Mimulus guttatus complex (Scrophulariaceae). Am J Botany. 1992;79(6):650–9.
- View Article
- Google Scholar
95. Fishman L, Kelly AJ, Willis JH. Minor quantitative trait loci underlie floral traits associated with mating system divergence in mimulus. Evol. 2002;56(11):2138.
- View Article
- Google Scholar
96. Kiang YT, Hamrick JL. Reproductive Isolation in the Mimulus guttatus M. nasutus Complex. Am Midland Naturalist. 1978;100(2):269.
- View Article
- Google Scholar
97. Fishman L, Sweigart AL, Kenney AM, Campbell S. Major quantitative trait loci control divergence in critical photoperiod for flowering between selfing and outcrossing species of monkeyflower (Mimulus). New Phytol. 2014;201(4):1498–507. pmid:24304557
- View Article
- PubMed/NCBI
- Google Scholar
98. Whitener MR, Mangelson H, Sweigart AL. Patterns of genomic variation reveal a single evolutionary origin of the wild allotetraploid Mimulus sookensis. Friedman J, Connallon T, (Editors). Evolution. 2024 May 20;qpae079.
- View Article
- Google Scholar
99. Sweigart A, Willis J. Patterns of nucleotide diversity in two species of mimulus are affected by mating system and asymmetric introgression. Evolution. 2003;57(11):2490–506.
- View Article
- Google Scholar
100. Machado CA, Haselkorn TS, Noor MAF. Evaluation of the genomic extent of effects of fixed inversion differences on intraspecific variation and interspecific gene flow in Drosophila pseudoobscura and D. persimilis. Genetics. 2007;175(3):1289–306. pmid:17179068
- View Article
- PubMed/NCBI
- Google Scholar
101. Clarke B, Johnson M, Murray J, Hewitt G, Wragg G. Clines in the genetic distance between two species of island land snails: how ‘molecular leakage’ can mislead us about speciation. Philosophical Transactions of the Royal Society B: Biological Sciences. 1997;351(1341):773–84.
- View Article
- Google Scholar
102. Bettles CM, Docker MF, Dufour B, Heath DD. Hybridization dynamics between sympatric species of trout: loss of reproductive isolation. J Evol Biol. 2005;18(5):1220–33. pmid:16135118
- View Article
- PubMed/NCBI
- Google Scholar
103. Gilman RT, Behm JE. Hybridization, species collapse, and species reemergence after disturbance to premating mechanisms of reproductive isolation. Evolution. 2011;65(9):2592–605. pmid:21884058
- View Article
- PubMed/NCBI
- Google Scholar
104. Simon A, Fraïsse C, El Ayari T, Liautard-Haag C, Strelkov P, Welch JJ, et al. How do species barriers decay? Concordance and local introgression in mosaic hybrid zones of mussels. J Evol Biol. 2021;34(1):208–23. pmid:33045123
- View Article
- PubMed/NCBI
- Google Scholar
105. Langdon QK, Groh JS, Aguillon SM, Powell DL, Gunn T, Payne C, et al. Swordtail fish hybrids reveal that genome evolution is surprisingly predictable after initial hybridization. PLoS Biol. 2024;22(8):e3002742. pmid:39186811
- View Article
- PubMed/NCBI
- Google Scholar
106. Riquet F, Liautard‐Haag C, Woodall L, Bouza C, Louisy P, Hamer B. Parallel pattern of differentiation at a genomic island shared between clinal and mosaic hybrid zones in a complex of cryptic seahorse lineages. Evolution. 2019;73(4):817–35.
- View Article
- Google Scholar
107. Nelson TC, Stathos AM, Vanderpool DD, Finseth FR, Yuan Y-W, Fishman L. Ancient and recent introgression shape the evolutionary history of pollinator adaptation and speciation in a model monkeyflower radiation (Mimulus section Erythranthe). PLoS Genet. 2021;17(2):e1009095. pmid:33617525
- View Article
- PubMed/NCBI
- Google Scholar
108. Rifkin JL, Castillo AS, Liao IT, Rausher MD. Gene flow, divergent selection and resistance to introgression in two species of morning glories (Ipomoea). Mol Ecol. 2019;28(7):1709–29. pmid:30451335
- View Article
- PubMed/NCBI
- Google Scholar
109. Ruhsam M, Hollingsworth PM, Ennos RA. Early evolution in a hybrid swarm between outcrossing and selfing lineages in Geum. Heredity (Edinb). 2011;107(3):246–55. pmid:21448227
- View Article
- PubMed/NCBI
- Google Scholar
110. Lepais O, Petit RJ, Guichoux E, Lavabre JE, Alberto F, Kremer A, et al. Species relative abundance and direction of introgression in oaks. Mol Ecol. 2009;18(10):2228–42. pmid:19302359
- View Article
- PubMed/NCBI
- Google Scholar
111. Rand DM, Harrison RG. Ecological genetics of a mosaic hybrid zone: mitochondrial, nuclear, and reproductive differentiation of crickets by soil type. Evolution. 1989;43(2):432–49.
- View Article
- Google Scholar
112. Aldridge G. Variation in frequency of hybrids and spatial structure among Ipomopsis (Polemoniaceae) contact sites. New Phytol. 2005;167(1):279–88. pmid:15948849
- View Article
- PubMed/NCBI
- Google Scholar
113. Comes H, Abbott R. Molecular phylogeography, reticulation, and lineage sorting in mediterranean senecio sect. senecio (asteraceae). Evolution. 2001;55(10):1943–62.
- View Article
- Google Scholar
114. Folk RA, Mandel JR, Freudenstein JV. Ancestral gene flow and parallel organellar genome capture result in extreme phylogenomic discord in a lineage of angiosperms. Syst Biol. 2017;66(3):320–37.
- View Article
- Google Scholar
115. Tsitrone A, Kirkpatrick M, Levin D. A model for chloroplast capture. Evolution. 2003;57(8):1776–82.
- View Article
- Google Scholar
116. Webb CJ. The Selection of Pollen and Seed Dispersal in Plants. Plant Species Biol. 1998;13(1):57–67.
- View Article
- Google Scholar
117. Dobson SF. Competition for mates and predominant juvenile male dispersal in mammals. Anim Behav. 1982;30(4):1183–92.
- View Article
- Google Scholar
118. Barreto FS, Watson ET, Lima TG, Willett CS, Edmands S, Li W, et al. Genomic signatures of mitonuclear coevolution across populations of Tigriopus californicus. Nat Ecol Evol. 2018;2(8):1250–7. pmid:29988158
- View Article
- PubMed/NCBI
- Google Scholar
119. Bogdanova VS, Galieva ER, Kosterin OE. Genetic analysis of nuclear-cytoplasmic incompatibility in pea associated with cytoplasm of an accession of wild subspecies Pisum sativum subsp. elatius (Bieb.) Schmahl. Theor Appl Genet. 2009;118(4):801–9. pmid:19099285
- View Article
- PubMed/NCBI
- Google Scholar
120. Gobron N, Waszczak C, Simon M, Hiard S, Boivin S, Charif D, et al. A cryptic cytoplasmic male sterility unveils a possible gynodioecious past for Arabidopsis thaliana. PLoS One. 2013;8(4):e62450. pmid:23658632
- View Article
- PubMed/NCBI
- Google Scholar
121. Lee H-Y, Chou J-Y, Cheong L, Chang N-H, Yang S-Y, Leu J-Y. Incompatibility of nuclear and mitochondrial genomes causes hybrid sterility between two yeast species. Cell. 2008;135(6):1065–73. pmid:19070577
- View Article
- PubMed/NCBI
- Google Scholar
122. Meiklejohn CD, Holmbeck MA, Siddiq MA, Abt DN, Rand DM, Montooth KL. An Incompatibility between a mitochondrial tRNA and its nuclear-encoded tRNA synthetase compromises development and fitness in Drosophila. PLoS Genet. 2013;9(1):e1003238. pmid:23382693
- View Article
- PubMed/NCBI
- Google Scholar
123. Fishman L, Willis J. A cytonuclear incompatibility causes anther sterility in Mimulus hybrids. Evolution. 2006;60(7):1372–81.
- View Article
- Google Scholar
124. Colicchio J, Monnahan PJ, Wessinger CA, Brown K, Kern JR, Kelly JK. Individualized mating system estimation using genomic data. Mol Ecol Resour. 2020;20(1):333–47. pmid:31519042
- View Article
- PubMed/NCBI
- Google Scholar
125. Puzey JR, Willis JH, Kelly JK. Population structure and local selection yield high genomic variation in Mimulus guttatus. Mol Ecol. 2017;26(2):519–35. pmid:27859786
- View Article
- PubMed/NCBI
- Google Scholar
126. Ross CL, Harrison RG. A fine-scale spatial analysis of the mosaic hybrid zone between Gryllus firmus and Gryllus pennsylvanicus. Evolution. 2002;56(11):2296–312. pmid:12487359
- View Article
- PubMed/NCBI
- Google Scholar
127. Valbuena-Carabaña M, González-Martínez SC, Hardy OJ, Gil L. Fine-scale spatial genetic structure in mixed oak stands with different levels of hybridization. Mol Ecol. 2007;16(6):1207–19. pmid:17391407
- View Article
- PubMed/NCBI
- Google Scholar
128. Abbott RJ. Plant speciation across environmental gradients and the occurrence and nature of hybrid zones. J Syst Evol. 2017;55(4):238–58.
- View Article
- Google Scholar
129. Dudash MR, Ritland K. Multiple paternity and self-fertilization in relation to floral age in Mimulus guttatus (Scrophulariaceae). Am J Botany. 1991;78(12):1746–53.
- View Article
- Google Scholar
130. Leclerc-Potvin C, Ritland K. Modes of self-fertilization in Mimulus guttatus (Scrophulariaceae): a field experiment. Am J Botany. 1994;81(2):199–205.
- View Article
- Google Scholar
131. Ritland K. Correlated matings in the partial selfer mimulus guttatus. Evolution. 1989;43(4):848–59. pmid:28564194
- View Article
- PubMed/NCBI
- Google Scholar
132. Ritland K, Ganders FR. Covariation of selfing rates with parental gene fixation indices within populations of mimulus guttatus. Evolution. 1987;41(4):760–71. pmid:28564350
- View Article
- PubMed/NCBI
- Google Scholar
133. Karron JD, Holmquist KG, Flanagan RJ, Mitchell RJ. Pollinator visitation patterns strongly influence among-flower variation in selfing rate. Ann Bot. 2009;103(9):1379–83. pmid:19218584
- View Article
- PubMed/NCBI
- Google Scholar
134. Karron JD, Mitchell RJ, Holmquist KG, Bell JM, Funk B. The influence of floral display size on selfing rates in Mimulus ringens. Heredity (Edinb). 2004;92(3):242–8. pmid:14666135
- View Article
- PubMed/NCBI
- Google Scholar
135. Bartlett KB, Austin MW, Beck JB, Zanne AE, Smith AB. Beyond the usual climate? Factors determining flowering and fruiting phenology across a genus over 117 years. Am J Bot. 2023;110(7):e16188. pmid:37200535
- View Article
- PubMed/NCBI
- Google Scholar
136. Ganjurjav H, Gornish E, Hu G, Schwartz M, Wan Y, Li Y, et al. Warming and precipitation addition interact to affect plant spring phenology in alpine meadows on the central Qinghai-Tibetan Plateau. Agric Forest Meteorol. 2020;287:107943.
- View Article
- Google Scholar
137. Matthews ER, Mazer SJ. Historical changes in flowering phenology are governed by temperature × precipitation interactions in a widespread perennial herb in western North America. New Phytol. 2016;210(1):157–67. pmid:26595165
- View Article
- PubMed/NCBI
- Google Scholar
138. Sianta SA, Kay KM. Parallel evolution of phenological isolation across the speciation continuum in serpentine-adapted annual wildflowers. Proc Biol Sci. 2021;288(1948):20203076.
- View Article
- Google Scholar
139. Hunt B, Elliott T. Interaction of climatic variability with climatic change. Atmosphere-Ocean. 2004;42(3):145–72.
- View Article
- Google Scholar
140. Rind D, Goldberg R, Ruedy R. Change in climate variability in the 21st century. Clim Change. 1989;14(1):5–37.
- View Article
- Google Scholar
141. Salinger MJ. Climate variability and change: past, present and future – an overview. Clim Change. 2005;70(1):9–29.
- View Article
- Google Scholar
142. Brauer CJ, Sandoval-Castillo J, Gates K, Hammer MP, Unmack PJ, Bernatchez L, et al. Natural hybridization reduces vulnerability to climate change. Nat Clim Change. 2023 Mar;13(3):282–9.
- View Article
- Google Scholar
143. Chunco AJ. Hybridization in a warmer world. Ecol Evol. 2014;4(10):2019–31. pmid:24963394
- View Article
- PubMed/NCBI
- Google Scholar
144. Muhlfeld CC, Kovach RP, Jones LA, Al-Chokhachy R, Boyer MC, Leary RF, et al. Invasive hybridization in a threatened species is accelerated by climate change. Nature Clim Change. 2014;4(7):620–4.
- View Article
- Google Scholar
145. Savage J, Vellend M. Elevational shifts, biotic homogenization and time lags in vegetation change during 40 years of climate warming. Ecography. 2014;38(6):546–55.
- View Article
- Google Scholar
146. Vallejo-Marín M, Hiscock SJ. Hybridization and hybrid speciation under global change. New Phytol. 2016;211(4):1170–87. pmid:27214560
- View Article
- PubMed/NCBI
- Google Scholar
147. PRISM Climate Group. PRISM Time Series Data [Internet]. Oregon State University; 2024. Available from: https://prism.oregonstate.edu
148. Zeileis A, Grothendieck G. zoo: S3 Infrastructure for Regular and Irregular Time Series. J Statist Software. 2005;14(6):1–27.
- View Article
- Google Scholar
149. Farnitano M, Karoly K, Sweigart A. Data for “Fluctuating reproductive isolation and stable ancestry structure in a fine-scaled mosaic of hybridizing Mimulus monkeyflowers”. 2025.
- View Article
- Google Scholar
150. Fishman L. 96-well CTAB-chloroform DNA extraction v1. 2020.
- View Article
- Google Scholar
151. Adey A, Morrison HG, , Xun X, Kitzman JO, Turner EH, et al. Rapid, low-input, low-bias construction of shotgun fragment libraries by high-density in vitro transposition. Genome Biol. 2010;11(12):R119. pmid:21143862
- View Article
- PubMed/NCBI
- Google Scholar
152. Picelli S, Björklund AK, Reinius B, Sagasser S, Winberg G, Sandberg R. Tn5 transposase and tagmentation procedures for massively scaled sequencing projects. Genome Res. 2014;24(12):2033–40. pmid:25079858
- View Article
- PubMed/NCBI
- Google Scholar
153. Farnitano M, Sweigart A. Low-cost Tagmentation Library Prep for low-coverage Illumina sequencing v1. 2024.
- View Article
- Google Scholar
154. Lu Z, Hofmeister BT, Vollmers C, DuBois RM, Schmitz RJ. Combining ATAC-seq with nuclei sorting for discovery of cis-regulatory regions in plant genomes. Nucleic Acids Res. 2017;45(6):e41. pmid:27903897
- View Article
- PubMed/NCBI
- Google Scholar
155. Bolger AM, Lohse M, Usadel B. Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics. 2014;30(15):2114–20.
- View Article
- Google Scholar
156. Li H, Durbin R. Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinformatics. 2009;25(14):1754–60. pmid:19451168
- View Article
- PubMed/NCBI
- Google Scholar
157. Broad Institute. Picard Toolkit, GitHub repository [Internet]. 2019. Available from: https://broadinstitute.github.io/picard/
- View Article
- Google Scholar
158. Danecek P, Bonfield JK, Liddle J, Marshall J, Ohan V, Pollard MO, et al. Twelve years of SAMtools and BCFtools. Gigascience. 2021;10(2):giab008. pmid:33590861
- View Article
- PubMed/NCBI
- Google Scholar
159. Okonechnikov K, Conesa A, García-Alcalde F. Qualimap 2: advanced multi-sample quality control for high-throughput sequencing data. Bioinformatics. 2016;32(2):292–4.
- View Article
- Google Scholar
160. Van der Auwera G, O’Connor B. Genomics in the cloud: Using Docker, GATK, and WDL in Terra. O’Reilly Media 2020.
161. Danecek P, Auton A, Abecasis G, Albers C, Banks E, DePristo M, et al. The variant call format and VCFtools. Bioinformatics. 2011;27(15):2156.
- View Article
- Google Scholar
162. Corbett-Detig R, Nielsen R. A Hidden markov model approach for simultaneously estimating local ancestry and admixture time using next generation sequence data in samples of arbitrary ploidy. PLoS Genet. 2017;13(1):e1006529. pmid:28045893
- View Article
- PubMed/NCBI
- Google Scholar
163. Korneliussen TS, Albrechtsen A, Nielsen R. ANGSD: analysis of next generation sequencing data. BMC Bioinformatics. 2014;15(1):356. pmid:25420514
- View Article
- PubMed/NCBI
- Google Scholar
164. Meisner J, Albrechtsen A. Inferring population structure and admixture proportions in low-depth NGS Data. Genetics. 2018;210(2):719–31. pmid:30131346
- View Article
- PubMed/NCBI
- Google Scholar
165. R Core Team. R: A Language and Environment for Statistical Computing [Internet]. Vienna, Austria: R Foundation for Statistical Computing; 2023. Available from: https://www.R-project.org/
166. Skotte L, Korneliussen TS, Albrechtsen A. Estimating individual admixture proportions from next generation sequencing data. Genetics. 2013;195(3):693–702. pmid:24026093
- View Article
- PubMed/NCBI
- Google Scholar
167. Schumer M, Powell DL, Corbett-Detig R. Versatile simulations of admixture and accurate local ancestry inference with mixnmatch and ancestryinfer. Mol Ecol Resour. 2020;20(4):1141–51. pmid:32324964
- View Article
- PubMed/NCBI
- Google Scholar
168. Sweigart AL, Martin NH, Willis JH. Patterns of nucleotide variation and reproductive isolation between a Mimulus allotetraploid and its progenitor species. Mol Ecol. 2008;17(8):2089–100. pmid:18312552
- View Article
- PubMed/NCBI
- Google Scholar
169. Modliszewski JL, Willis JH. Allotetraploid Mimulus sookensis are highly interfertile despite independent origins. Mol Ecol. 2012;21(21):5280–98. pmid:22831280
- View Article
- PubMed/NCBI
- Google Scholar
170. Andolfatto P, Davison D, Erezyilmaz D, Hu TT, Mast J, Sunayama-Morita T, et al. Multiplexed shotgun genotyping for rapid and efficient genetic mapping. Genome Res. 2011;21(4):610–7. pmid:21233398
- View Article
- PubMed/NCBI
- Google Scholar
171. Price AL, Tandon A, Patterson N, Barnes KC, Rafaels N, Ruczinski I, et al. Sensitive detection of chromosomal segments of distinct ancestry in admixed populations. PLoS Genet. 2009;5(6):e1000519. pmid:19543370
- View Article
- PubMed/NCBI
- Google Scholar
172. Rice P, Longden I, Bleasby A. EMBOSS: the european molecular biology open software suite. Trends Genetics. 2000;16(6):276–7.
- View Article
- Google Scholar
173. Leigh JW, Bryant D. popart: full‐feature software for haplotype network construction. Methods Ecol Evol. 2015;6(9):1110–6.
- View Article
- Google Scholar
174. Mower J, Case A, Floro E, Willis J. Evidence against equimolarity of large repeat arrangements and a predominant master circle structure of the mitochondrial genome from a monkeyflower (Mimulus guttatus) lineage with cryptic CMS. Genome Biol Evol. 2012;4(5):670–86.
- View Article
- Google Scholar
175. Sobel JM, Chen GF. Unification of methods for estimating the strength of reproductive isolation. Evolution. 2014;68(5):1511–22. pmid:24450287
- View Article
- PubMed/NCBI
- Google Scholar
176. Hastie T. Generalized Additive Models. 2024. https://CRAN.R-project.org/package=gam
- View Article
- Google Scholar
177. Wood SN. Fast stable restricted maximum likelihood and marginal likelihood estimation of semiparametric generalized linear models. J R Statist Soc Series B (Statistical Methodology). 2011;73(1):3–36.
- View Article
- Google Scholar
178. Hijmans RJ, Barbosa M, Ghosh A, Mandel A. Geodata: download geographic data [Internet]. 2024 [cited 2024 Oct 10]. Available from: https://cran.r-project.org/web/packages/geodata/index.html
- View Article
- Google Scholar
179. USDA Farm Production and Conservation - Business Center, Geospatial Enterprise Operations. National Agriculture Imagery Program [Internet]. Google Earth Engine Data Catalog; [cited 2025 Jan 27]. Available from: https://developers.google.com/earth-engine/datasets/catalog/USDA_NAIP_DOQQ
- View Article
- Google Scholar

[ref1] 1. Dobzhansky Th. Studies on hybrid sterility. Z Für Zellforsch Mikrosk Anat. 1934;21(2):169–223.
View Article
Google Scholar

[2] View Article

[3] Google Scholar

[ref2] 2. Farnitano M, Sweigart A. Strong postmating reproductive isolation in Mimulus section Eunanus. Journal of Evolutionary Biology. 2023;36(10):1393–410.
View Article
Google Scholar

[5] View Article

[6] Google Scholar

[ref3] 3. Richards TJ, Ortiz-Barrientos D. Immigrant inviability produces a strong barrier to gene flow between parapatric ecotypes of Senecio lautus. Evolution. 2016;70(6):1239–48.
View Article
Google Scholar

[8] View Article

[9] Google Scholar

[ref4] 4. Scopece G, Widmer A, Cozzolino S. Evolution of postzygotic reproductive isolation in a guild of deceptive orchids. Am Nat. 2008;171(3):315–26. pmid:18198999
View Article
PubMed/NCBI
Google Scholar

[11] View Article

[12] PubMed/NCBI

[13] Google Scholar

[ref5] 5. Stebbins GL. The role of hybridization in evolution. Proc Am Philos Soc. 1959;103(2):231–51.
View Article
Google Scholar

[15] View Article

[16] Google Scholar

[ref6] 6. Wodsedalek JE. Causes of sterility in the mule. Biol Bull. 1916;30(1):1–56.
View Article
Google Scholar

[18] View Article

[19] Google Scholar

[ref7] 7. Xiong T, Mallet J. On the impermanence of species: the collapse of genetic incompatibilities in hybridizing populations. Evolution. 2022;76(11):2498–512. pmid:36097352
View Article
PubMed/NCBI
Google Scholar

[21] View Article

[22] PubMed/NCBI

[23] Google Scholar

[ref8] 8. Behm JE, Ives AR, Boughman JW. Breakdown in postmating isolation and the collapse of a species pair through hybridization. Am Nat. 2010;175(1):11–26. pmid:19916869
View Article
PubMed/NCBI
Google Scholar

[25] View Article

[26] PubMed/NCBI

[27] Google Scholar

[ref9] 9. Kleindorfer S, O’Connor JA, Dudaniec RY, Myers SA, Robertson J, Sulloway FJ. Species collapse via hybridization in Darwin’s tree finches. Am Nat. 2014;183(3):325–41. pmid:24561597
View Article
PubMed/NCBI
Google Scholar

[29] View Article

[30] PubMed/NCBI

[31] Google Scholar

[ref10] 10. Schumer M, Rosenthal G, Andolfatto P. How common is homoploid hybrid speciation?. Evolution. 2014;68(6):1553–60.
View Article
Google Scholar

[33] View Article

[34] Google Scholar

[ref11] 11. Ungerer MC, Baird SJ, Pan J, Rieseberg LH. Rapid hybrid speciation in wild sunflowers. Proc Natl Acad Sci U S A. 1998;95(20):11757–62. pmid:9751738
View Article
PubMed/NCBI
Google Scholar

[36] View Article

[37] PubMed/NCBI

[38] Google Scholar

[ref12] 12. Nice CC, Gompert Z, Fordyce JA, Forister ML, Lucas LK, Buerkle CA. Hybrid speciation and independent evolution in lineages of alpine butterflies. Evolution. 2013;67(4):1055–68. pmid:23550755
View Article
PubMed/NCBI
Google Scholar

[40] View Article

[41] PubMed/NCBI

[42] Google Scholar

[ref13] 13. Rosser N, Seixas F, Queste LM, Cama B, Mori-Pezo R, Kryvokhyzha D, et al. Hybrid speciation driven by multilocus introgression of ecological traits. Nature. 2024;628(8009):811–7. pmid:38632397
View Article
PubMed/NCBI
Google Scholar

[44] View Article

[45] PubMed/NCBI

[46] Google Scholar

[ref14] 14. Christie K, Fraser L, Lowry D. The strength of reproductive isolating barriers in seed plants: Insights from studies quantifying premating and postmating reproductive barriers over the past 15 years. Evolution. 2022;
View Article
Google Scholar

[48] View Article

[49] Google Scholar

[ref15] 15. Coyne JA, Orr HA. Patterns of speciation in drosophila revisited. Evolution. 1997;51(1):295–303. pmid:28568795
View Article
PubMed/NCBI
Google Scholar

[51] View Article

[52] PubMed/NCBI

[53] Google Scholar

[ref16] 16. Malone JH, Fontenot BE. Patterns of reproductive isolation in toads. PLoS One. 2008;3(12):e3900. pmid:19065271
View Article
PubMed/NCBI
Google Scholar

[55] View Article

[56] PubMed/NCBI

[57] Google Scholar

[ref17] 17. Moyle LC, Olson MS, Tiffin P. Patterns of reproductive isolation in three angiosperm genera. Evolution. 2004;58(6):1195–208. pmid:15266970
View Article
PubMed/NCBI
Google Scholar

[59] View Article

[60] PubMed/NCBI

[61] Google Scholar

[ref18] 18. Lotsy JP. On the species of the taxonomist in its relation to evolution. Genetica. 1931;13(1–2):1–16.
View Article
Google Scholar

[63] View Article

[64] Google Scholar

[ref19] 19. Suarez-Gonzalez A, Lexer C, Cronk QCB. Adaptive introgression: a plant perspective. Biol Lett. 2018 Mar 14;14(3):20170688.
View Article
Google Scholar

[66] View Article

[67] Google Scholar

[ref20] 20. Goulet BE, Roda F, Hopkins R. Hybridization in plants: old ideas, new techniques. Plant Physiol. 2017;173(1):65–78. pmid:27895205
View Article
PubMed/NCBI
Google Scholar

[69] View Article

[70] PubMed/NCBI

[71] Google Scholar

[ref21] 21. Mallet J, Besansky N, Hahn MW. How reticulated are species?. Bioessays. 2016;38(2):140–9. pmid:26709836
View Article
PubMed/NCBI
Google Scholar

[73] View Article

[74] PubMed/NCBI

[75] Google Scholar

[ref22] 22. Arnold ML, Kunte K. Adaptive genetic exchange: a tangled history of admixture and evolutionary innovation. Trends Ecol Evol. 2017;32(8):601–11. pmid:28645486
View Article
PubMed/NCBI
Google Scholar

[77] View Article

[78] PubMed/NCBI

[79] Google Scholar

[ref23] 23. Chhatre VE, Evans LM, DiFazio SP, Keller SR. Adaptive introgression and maintenance of a trispecies hybrid complex in range-edge populations of Populus. Molecular Ecology. 2018;27(23):4820–38.
View Article
Google Scholar

[81] View Article

[82] Google Scholar

[ref24] 24. Manzoor SA, Griffiths G, Obiakara MC, Esparza-Estrada CE, Lukac M. Evidence of ecological niche shift in Rhododendron ponticum (L.) in Britain: hybridization as a possible cause of rapid niche expansion. Ecol Evol. 2020;10(4):2040–50. pmid:32128136
View Article
PubMed/NCBI
Google Scholar

[84] View Article

[85] PubMed/NCBI

[86] Google Scholar

[ref25] 25. Pfennig KS, Kelly AL, Pierce AA. Hybridization as a facilitator of species range expansion. Proc Royal Soc B: Biol Sci. 2016;283(1839):20161329.
View Article
Google Scholar

[88] View Article

[89] Google Scholar

[ref26] 26. Baskett ML, Gomulkiewicz R. Introgressive hybridization as a mechanism for species rescue. Theoretical Ecol. 2011;4(2):223–39.
View Article
Google Scholar

[91] View Article

[92] Google Scholar

[ref27] 27. Oziolor EM, Reid NM, Yair S, Lee KM, Guberman VerPloeg S, Bruns PC, et al. Adaptive introgression enables evolutionary rescue from extreme environmental pollution. Science. 2019;364(6439):455–7. pmid:31048485
View Article
PubMed/NCBI
Google Scholar

[94] View Article

[95] PubMed/NCBI

[96] Google Scholar

[ref28] 28. Ayres DR, Zaremba K, Strong DR. Extinction of a common native species by hybridization with an invasive congener. Weed Technol. 2004;18(sp1):1288–91.
View Article
Google Scholar

[98] View Article

[99] Google Scholar

[ref29] 29. Rhymer JM, Simberloff D. Extinction by hybridization and introgression. Ann Rev Ecol Evol Syst. 1996;27(1):83–109.
View Article
Google Scholar

[101] View Article

[102] Google Scholar

[ref30] 30. Quilodrán CS, Tsoupas A, Currat M. The spatial signature of introgression after a biological invasion with hybridization. Front Ecol Evol. 2020;8.
View Article
Google Scholar

[104] View Article

[105] Google Scholar

[ref31] 31. Wolfe L, Blair A, Penna B. Does intraspecific hybridization contribute to the evolution of invasiveness?: an experimental test. Biol Invasions. 2007;9(5):515–21.
View Article
Google Scholar

[107] View Article

[108] Google Scholar

[ref32] 32. Goodman SJ, Barton NH, Swanson G, Abernethy K, Pemberton JM. Introgression through rare hybridization: a genetic study of a hybrid zone between red and sika deer (genus Cervus) in Argyll, Scotland. Genetics. 1999;152(1):355–71. pmid:10224266
View Article
PubMed/NCBI
Google Scholar

[110] View Article

[111] PubMed/NCBI

[112] Google Scholar

[ref33] 33. Green R, Krause J, Briggs A, Maricic T, Stenzel U, Kircher M. A draft sequence of the Neandertal genome. Science. 2010;328(5979):710–22.
View Article
Google Scholar

[114] View Article

[115] Google Scholar

[ref34] 34. Randi E, Lucchini V. Detecting rare introgression of domestic dog genes into wild wolf (Canis lupus) populations by Bayesian admixture analyses of microsatellite variation. Conserv Genet. 2002 Mar 1;3(1):29–43.
View Article
Google Scholar

[117] View Article

[118] Google Scholar

[ref35] 35. Harrison R, Larson E. Heterogeneous genome divergence, differential introgression, and the origin and structure of hybrid zones. Mol Ecol. 2016;25(11):2454–66.
View Article
Google Scholar

[120] View Article

[121] Google Scholar

[ref36] 36. Hasselman DJ, Argo EE, McBride MC, Bentzen P, Schultz TF, Perez-Umphrey AA, et al. Human disturbance causes the formation of a hybrid swarm between two naturally sympatric fish species. Mol Ecol. 2014;23(5):1137–52. pmid:24450302
View Article
PubMed/NCBI
Google Scholar

[123] View Article

[124] PubMed/NCBI

[125] Google Scholar

[ref37] 37. Schumer M, Xu C, Powell DL, Durvasula A, Skov L, Holland C, et al. Natural selection interacts with recombination to shape the evolution of hybrid genomes. Science. 2018;360(6389):656–60. pmid:29674434
View Article
PubMed/NCBI
Google Scholar

[127] View Article

[128] PubMed/NCBI

[129] Google Scholar

[ref38] 38. Yan L-J, Burgess KS, Zheng W, Tao Z-B, Li D-Z, Gao L-M. Incomplete reproductive isolation between Rhododendron taxa enables hybrid formation and persistence. J Integr Plant Biol. 2019;61(4):433–48. pmid:30192058
View Article
PubMed/NCBI
Google Scholar

[131] View Article

[132] PubMed/NCBI

[133] Google Scholar

[ref39] 39. Larson EL, Andrés JA, Bogdanowicz SM, Harrison RG. Differential introgression in a mosaic hybrid zone reveals candidate barrier genes. Evolution. 2013;67(12):3653–61.
View Article
Google Scholar

[135] View Article

[136] Google Scholar

[ref40] 40. Martinsen G, Whitham T, Turek R, Keim P. Hybrid populations selectively filter gene introgression between species. Evolution. 2001;55(7):1325–35.
View Article
Google Scholar

[138] View Article

[139] Google Scholar

[ref41] 41. Fishman L, Sweigart AL. When Two Rights Make a Wrong: The Evolutionary Genetics of Plant Hybrid Incompatibilities. Annu Rev Plant Biol. 2018;69:707–31. pmid:29505737
View Article
PubMed/NCBI
Google Scholar

[141] View Article

[142] PubMed/NCBI

[143] Google Scholar

[ref42] 42. Muller H. Isolating mechanisms, evolution, and temperature. Biological Symposium. 1942;6:71–125.
View Article
Google Scholar

[145] View Article

[146] Google Scholar

[ref43] 43. Orr HA, Turelli M. The evolution of postzygotic isolation: accumulating Dobzhansky-Muller incompatibilities. Evolution. 2001;55(6):1085–94. pmid:11475044
View Article
PubMed/NCBI
Google Scholar

[148] View Article

[149] PubMed/NCBI

[150] Google Scholar

[ref44] 44. Mantel SJ, Sweigart AL. Postzygotic barriers persist despite ongoing introgression in hybridizing Mimulus species. Mol Ecol. 2024;33(4):e17261. pmid:38174628
View Article
PubMed/NCBI
Google Scholar

[152] View Article

[153] PubMed/NCBI

[154] Google Scholar

[ref45] 45. Shuker DM, Underwood K, King TM, Butlin RK. Patterns of male sterility in a grasshopper hybrid zone imply accumulation of hybrid incompatibilities without selection. Proc Biol Sci. 2005;272(1580):2491–7. pmid:16271974
View Article
PubMed/NCBI
Google Scholar

[156] View Article

[157] PubMed/NCBI

[158] Google Scholar

[ref46] 46. Zuellig MP, Sweigart AL. A two-locus hybrid incompatibility is widespread, polymorphic, and active in natural populations of Mimulus. Evolution. 2018;72(11):2394–405. pmid:30194757
View Article
PubMed/NCBI
Google Scholar

[160] View Article

[161] PubMed/NCBI

[162] Google Scholar

[ref47] 47. Servedio MR, Hermisson J. The evolution of partial reproductive isolation as an adaptive optimum. Evolution. 2020;74(1):4–14. pmid:31721186
View Article
PubMed/NCBI
Google Scholar

[164] View Article

[165] PubMed/NCBI

[166] Google Scholar

[ref48] 48. Borge T, Lindroos K, Nádvorník P, Syvänen A-C, Saetre G-P. Amount of introgression in flycatcher hybrid zones reflects regional differences in pre and post-zygotic barriers to gene exchange. J Evol Biol. 2005;18(6):1416–24. pmid:16313454
View Article
PubMed/NCBI
Google Scholar

[168] View Article

[169] PubMed/NCBI

[170] Google Scholar

[ref49] 49. Brice C, Zhang Z, Bendixsen D, Stelkens R. Hybridization outcomes have strong genomic and environmental contingencies. Am Nat. 2021;198(3):E53–67. pmid:34403309
View Article
PubMed/NCBI
Google Scholar

[172] View Article

[173] PubMed/NCBI

[174] Google Scholar

[ref50] 50. Otis JA, Thornton D, Rutledge L, Murray DL. Ecological niche differentiation across a wolf-coyote hybrid zone in eastern North America. Divers Distrib. 2017;23(5):529–39.
View Article
Google Scholar

[176] View Article

[177] Google Scholar

[ref51] 51. Butlin RK, Saura M, Charrier G, Jackson B, André C, Caballero A, et al. Parallel evolution of local adaptation and reproductive isolation in the face of gene flow. Evolution. 2014 Apr 1;68(4):935–49.
View Article
Google Scholar

[179] View Article

[180] Google Scholar

[ref52] 52. Lowry DB, Willis JH. A widespread chromosomal inversion polymorphism contributes to a major life-history transition, local adaptation, and reproductive isolation. PLoS Biol. 2010;8(9):e1000500. pmid:20927411
View Article
PubMed/NCBI
Google Scholar

[182] View Article

[183] PubMed/NCBI

[184] Google Scholar

[ref53] 53. Melo MC, Grealy A, Brittain B, Walter GM, Ortiz-Barrientos D. Strong extrinsic reproductive isolation between parapatric populations of an Australian groundsel. New Phytol. 2014;203(1):323–34. pmid:24684207
View Article
PubMed/NCBI
Google Scholar

[186] View Article

[187] PubMed/NCBI

[188] Google Scholar

[ref54] 54. Borzée A, Kim JY, DA Cunha MAM, Lee D, Sin E, Oh S, et al. Temporal and spatial differentiation in microhabitat use: implications for reproductive isolation and ecological niche specification. Integr Zool. 2016;11(5):375–87. pmid:27059098
View Article
PubMed/NCBI
Google Scholar

[190] View Article

[191] PubMed/NCBI

[192] Google Scholar

[ref55] 55. Ebersole JP. Niche separation of two damselfish species by aggression and differential microhabitat utilization. Ecology. 1985;66(1):14–20.
View Article
Google Scholar

[194] View Article

[195] Google Scholar

[ref56] 56. Ling TC, Phokasem P, Sinpoo C, Yang Y-P, Disayathanoowat T. Microhabitat and pollinator differentiation drive reproductive isolation between two sympatric salvia species (Lamiaceae). Plants (Basel). 2022;11(18):2423. pmid:36145824
View Article
PubMed/NCBI
Google Scholar

[197] View Article

[198] PubMed/NCBI

[199] Google Scholar

[ref57] 57. Mantel S, Sweigart A. Divergence in drought‐response traits between sympatric species of Mimulus. Ecol Evol. 2019;
View Article
Google Scholar

[201] View Article

[202] Google Scholar

[ref58] 58. Bank C, Bürger R, Hermisson J. The limits to parapatric speciation: Dobzhansky-Muller incompatibilities in a continent-island model. Genetics. 2012;191(3):845–63. pmid:22542972
View Article
PubMed/NCBI
Google Scholar

[204] View Article

[205] PubMed/NCBI

[206] Google Scholar

[ref59] 59. Lenormand T. Gene flow and the limits to natural selection. Trends in Ecol Evol. 2002;17(4):183–9.
View Article
Google Scholar

[208] View Article

[209] Google Scholar

[ref60] 60. Delph LF, Kelly JK. On the importance of balancing selection in plants. New Phytol. 2014;201(1):45–56. pmid:23952298
View Article
PubMed/NCBI
Google Scholar

[211] View Article

[212] PubMed/NCBI

[213] Google Scholar

[ref61] 61. Levene H. Genetic equilibrium when more than one ecological niche is available. The American Naturalist. 1953;87(836):331–3.
View Article
Google Scholar

[215] View Article

[216] Google Scholar

[ref62] 62. Kisel Y, Barraclough TG. Speciation has a spatial scale that depends on levels of gene flow. Am Nat. 2010;175(3):316–34. pmid:20100106
View Article
PubMed/NCBI
Google Scholar

[218] View Article

[219] PubMed/NCBI

[220] Google Scholar

[ref63] 63. Klein EK, Lagache-Navarro L, Petit RJ. Demographic and spatial determinants of hybridization rate. J Ecol. 2017;105(1):29–38.
View Article
Google Scholar

[222] View Article

[223] Google Scholar

[ref64] 64. Musker SD, Ellis AG, Schlebusch SA, Verboom GA. Niche specificity influences gene flow across fine-scale habitat mosaics in Succulent Karoo plants. Mol Ecol. 2021;30(1):175–92. pmid:33152114
View Article
PubMed/NCBI
Google Scholar

[225] View Article

[226] PubMed/NCBI

[227] Google Scholar

[ref65] 65. Slatkin M. Gene flow and the geographic structure of natural populations. Science. 1987;236(4803):787–92. pmid:3576198
View Article
PubMed/NCBI
Google Scholar

[229] View Article

[230] PubMed/NCBI

[231] Google Scholar

[ref66] 66. Martin NH, Willis JH. Ecological divergence associated with mating system causes nearly complete reproductive isolation between sympatric Mimulus species. Evolution. 2007;61(1):68–82. pmid:17300428
View Article
PubMed/NCBI
Google Scholar

[233] View Article

[234] PubMed/NCBI

[235] Google Scholar

[ref67] 67. Roda F, Mendes F, Hahn M, Hopkins R. Genomic evidence of gene flow during reinforcement in Texas Phlox. Mol Ecol. 2017;26(8):2317–30.
View Article
Google Scholar

[237] View Article

[238] Google Scholar

[ref68] 68. Sambatti JBM, Strasburg JL, Ortiz-Barrientos D, Baack EJ, Rieseberg LH. Reconciling extremely strong barriers with high levels of gene exchange in annual sunflowers. Evolution. 2012;66(5):1459–73. pmid:22519784
View Article
PubMed/NCBI
Google Scholar

[240] View Article

[241] PubMed/NCBI

[242] Google Scholar

[ref69] 69. Abbott RJ. Plant invasions, interspecific hybridization and the evolution of new plant taxa. Trends Ecol Evol. 1992;7(12):401–5.
View Article
Google Scholar

[244] View Article

[245] Google Scholar

[ref70] 70. Schwarzbach AE, Donovan LA, Rieseberg LH. Transgressive character expression in a hybrid sunflower species. Am J Bot. 2001;88(2):270–7. pmid:11222249
View Article
PubMed/NCBI
Google Scholar

[247] View Article

[248] PubMed/NCBI

[249] Google Scholar

[ref71] 71. Selz OM, Seehausen O. Interspecific hybridization can generate functional novelty in cichlid fish. Proc R Soc B Biol Sci. 2019;286(1913):20191621.
View Article
Google Scholar

[251] View Article

[252] Google Scholar

[ref72] 72. Hendry AP, Berg OK, Quinn TP. Condition dependence and adaptation-by-time: breeding date, life history, and energy allocation within a population of salmon. Oikos. 1999;85(3):499.
View Article
Google Scholar

[254] View Article

[255] Google Scholar

[ref73] 73. Hood G, Zhang L, Hu E, Ott J, Egan S. Cascading reproductive isolation: Plant phenology drives temporal isolation among populations of a host-specific herbivore. Evolution. 2019;73(3):554–68.
View Article
Google Scholar

[257] View Article

[258] Google Scholar

[ref74] 74. Osmolovsky I, Shifrin M, Gamliel I, Belmaker J, Sapir Y. Eco-Geography and Phenology Are the Major Drivers of Reproductive Isolation in the Royal Irises, a Species Complex in the Course of Speciation. Plants (Basel). 2022;11(23):3306. pmid:36501345
View Article
PubMed/NCBI
Google Scholar

[260] View Article

[261] PubMed/NCBI

[262] Google Scholar

[ref75] 75. Franks SJ, Weis AE. Climate change alters reproductive isolation and potential gene flow in an annual plant. Evol Appl. 2009;2(4):481–8. pmid:25567893
View Article
PubMed/NCBI
Google Scholar

[264] View Article

[265] PubMed/NCBI

[266] Google Scholar

[ref76] 76. Seehausen O, Takimoto G, Roy D, Jokela J. Speciation reversal and biodiversity dynamics with hybridization in changing environments. Mol Ecol. 2008;17(1):30–44. pmid:18034800
View Article
PubMed/NCBI
Google Scholar

[268] View Article

[269] PubMed/NCBI

[270] Google Scholar

[ref77] 77. Sianta SA, Moeller DA, Brandvain Y. The extent of introgression between incipient Clarkia species is determined by temporal environmental variation and mating system. Proc Natl Acad Sci U S A. 2024;121(12):e2316008121. pmid:38466849
View Article
PubMed/NCBI
Google Scholar

[272] View Article

[273] PubMed/NCBI

[274] Google Scholar

[ref78] 78. Abdul-Rahman F, Tranchina D, Gresham D. Fluctuating environments maintain genetic diversity through neutral fitness effects and balancing selection. Mole Biol Evol. 2021;38(10):4362–75.
View Article
Google Scholar

[276] View Article

[277] Google Scholar

[ref79] 79. Bergland AO, Behrman EL, O’Brien KR, Schmidt PS, Petrov DA. Genomic evidence of rapid and stable adaptive oscillations over seasonal time scales in Drosophila. PLoS Genet. 2014;10(11):e1004775. pmid:25375361
View Article
PubMed/NCBI
Google Scholar

[279] View Article

[280] PubMed/NCBI

[281] Google Scholar

[ref80] 80. Han G, Wang W, Dong Y. Effects of balancing selection and microhabitat temperature variations on heat tolerance of the intertidal black mussel Septifer virgatus. Integr Zool. 2020;15(5):416–27. pmid:32297470
View Article
PubMed/NCBI
Google Scholar

[283] View Article

[284] PubMed/NCBI

[285] Google Scholar

[ref81] 81. Troth A, Puzey JR, Kim RS, Willis JH, Kelly JK. Selective trade-offs maintain alleles underpinning complex trait variation in plants. Science. 2018;361(6401):475–8. pmid:30072534
View Article
PubMed/NCBI
Google Scholar

[287] View Article

[288] PubMed/NCBI

[289] Google Scholar

[ref82] 82. Grant BR, Grant PR. High survival of darwin’s finch hybrids: effects of beak morphology and diets. Ecology. 1996;77(2):500–9.
View Article
Google Scholar

[291] View Article

[292] Google Scholar

[ref83] 83. Tataru D, Wheeler E, Ferris K. Spatially and temporally varying selection influence species boundaries in two sympatric Mimulus. Proc R Soc B: Biol Sci. 2023;290(1992):20222279.
View Article
Google Scholar

[294] View Article

[295] Google Scholar

[ref84] 84. Carney SE, Gardner KA, Rieseberg LH. Evolutionary changes over the fifty-year history of a hybrid population of sunflowers (helianthus). Evol. 2000;54(2):462.
View Article
Google Scholar

[297] View Article

[298] Google Scholar

[ref85] 85. Taylor S, Curry R, White T, Ferretti V, Lovette I. Spatiotemporally consistent genomic signatures of reproductive isolation in a moving hybrid zone. Evolution. 2014;68(11):3066–81.
View Article
Google Scholar

[300] View Article

[301] Google Scholar

[ref86] 86. Moore WS, Buchanan EB. Stability of the northern flicker hybrid zone in historical times: implications for adaptive speciation theory. Evolution. 1985;39(1):135–51. pmid:28563641
View Article
PubMed/NCBI
Google Scholar

[303] View Article

[304] PubMed/NCBI

[305] Google Scholar

[ref87] 87. Sullivan B. Temporal stability in hybridization between Bufo microscaphus and Bufo woodhousii (Anura: Bufonidae): Behavior and morphology. J Evol Biology. 1995;8(2):233–47.
View Article
Google Scholar

[307] View Article

[308] Google Scholar

[ref88] 88. Hendry AP, Wenburg JK, Bentzen P, Volk EC, Quinn TP. Rapid evolution of reproductive isolation in the wild: evidence from introduced salmon. Science. 2000;290(5491):516–9. pmid:11039932
View Article
PubMed/NCBI
Google Scholar

[310] View Article

[311] PubMed/NCBI

[312] Google Scholar

[ref89] 89. Brandvain Y, Kenney A, Flagel L, Coop G, Sweigart A. Speciation and introgression between Mimulus nasutus and Mimulus guttatus. PLoS Genetics. 2014;10(6):e1004410.
View Article
Google Scholar

[314] View Article

[315] Google Scholar

[ref90] 90. Kenney AM, Sweigart AL. Reproductive isolation and introgression between sympatric Mimulus species. Mol Ecol. 2016;25(11):2499–517. pmid:27038381
View Article
PubMed/NCBI
Google Scholar

[317] View Article

[318] PubMed/NCBI

[319] Google Scholar

[ref91] 91. Arathi HS, Kelly JK. Corolla Morphology Facilitates Both Autogamy and Bumblebee Pollination in Mimulus guttatus. Int J Plant Sci. 2004;165(6):1039–45.
View Article
Google Scholar

[321] View Article

[322] Google Scholar

[ref92] 92. Kelly JK, Willis JH. A Manipulative Experiment to Estimate Biparental Inbreeding in Monkeyflowers. Int J Plant Sci. 2002;163(4):575–9.
View Article
Google Scholar

[324] View Article

[325] Google Scholar

[ref93] 93. Macnair MR, Macnair VE, Martin BE. Adaptive speciation in Mimulus: an ecological comparison of M. cupriphilus with its presumed progenitor, M. guttatus. New Phytologist. 1989;112(2):269–79.
View Article
Google Scholar

[327] View Article

[328] Google Scholar

[ref94] 94. Dole JA. Reproductive assurance mechanisms in three taxa of the Mimulus guttatus complex (Scrophulariaceae). Am J Botany. 1992;79(6):650–9.
View Article
Google Scholar

[330] View Article

[331] Google Scholar

[ref95] 95. Fishman L, Kelly AJ, Willis JH. Minor quantitative trait loci underlie floral traits associated with mating system divergence in mimulus. Evol. 2002;56(11):2138.
View Article
Google Scholar

[333] View Article

[334] Google Scholar

[ref96] 96. Kiang YT, Hamrick JL. Reproductive Isolation in the Mimulus guttatus M. nasutus Complex. Am Midland Naturalist. 1978;100(2):269.
View Article
Google Scholar

[336] View Article

[337] Google Scholar

[ref97] 97. Fishman L, Sweigart AL, Kenney AM, Campbell S. Major quantitative trait loci control divergence in critical photoperiod for flowering between selfing and outcrossing species of monkeyflower (Mimulus). New Phytol. 2014;201(4):1498–507. pmid:24304557
View Article
PubMed/NCBI
Google Scholar

[339] View Article

[340] PubMed/NCBI

[341] Google Scholar

[ref98] 98. Whitener MR, Mangelson H, Sweigart AL. Patterns of genomic variation reveal a single evolutionary origin of the wild allotetraploid Mimulus sookensis. Friedman J, Connallon T, (Editors). Evolution. 2024 May 20;qpae079.
View Article
Google Scholar

[343] View Article

[344] Google Scholar

[ref99] 99. Sweigart A, Willis J. Patterns of nucleotide diversity in two species of mimulus are affected by mating system and asymmetric introgression. Evolution. 2003;57(11):2490–506.
View Article
Google Scholar

[346] View Article

[347] Google Scholar

[ref100] 100. Machado CA, Haselkorn TS, Noor MAF. Evaluation of the genomic extent of effects of fixed inversion differences on intraspecific variation and interspecific gene flow in Drosophila pseudoobscura and D. persimilis. Genetics. 2007;175(3):1289–306. pmid:17179068
View Article
PubMed/NCBI
Google Scholar

[349] View Article

[350] PubMed/NCBI

[351] Google Scholar

[ref101] 101. Clarke B, Johnson M, Murray J, Hewitt G, Wragg G. Clines in the genetic distance between two species of island land snails: how ‘molecular leakage’ can mislead us about speciation. Philosophical Transactions of the Royal Society B: Biological Sciences. 1997;351(1341):773–84.
View Article
Google Scholar

[353] View Article

[354] Google Scholar

[ref102] 102. Bettles CM, Docker MF, Dufour B, Heath DD. Hybridization dynamics between sympatric species of trout: loss of reproductive isolation. J Evol Biol. 2005;18(5):1220–33. pmid:16135118
View Article
PubMed/NCBI
Google Scholar

[356] View Article

[357] PubMed/NCBI

[358] Google Scholar

[ref103] 103. Gilman RT, Behm JE. Hybridization, species collapse, and species reemergence after disturbance to premating mechanisms of reproductive isolation. Evolution. 2011;65(9):2592–605. pmid:21884058
View Article
PubMed/NCBI
Google Scholar

[360] View Article

[361] PubMed/NCBI

[362] Google Scholar

[ref104] 104. Simon A, Fraïsse C, El Ayari T, Liautard-Haag C, Strelkov P, Welch JJ, et al. How do species barriers decay? Concordance and local introgression in mosaic hybrid zones of mussels. J Evol Biol. 2021;34(1):208–23. pmid:33045123
View Article
PubMed/NCBI
Google Scholar

[364] View Article

[365] PubMed/NCBI

[366] Google Scholar

[ref105] 105. Langdon QK, Groh JS, Aguillon SM, Powell DL, Gunn T, Payne C, et al. Swordtail fish hybrids reveal that genome evolution is surprisingly predictable after initial hybridization. PLoS Biol. 2024;22(8):e3002742. pmid:39186811
View Article
PubMed/NCBI
Google Scholar

[368] View Article

[369] PubMed/NCBI

[370] Google Scholar

[ref106] 106. Riquet F, Liautard‐Haag C, Woodall L, Bouza C, Louisy P, Hamer B. Parallel pattern of differentiation at a genomic island shared between clinal and mosaic hybrid zones in a complex of cryptic seahorse lineages. Evolution. 2019;73(4):817–35.
View Article
Google Scholar

[372] View Article

[373] Google Scholar

[ref107] 107. Nelson TC, Stathos AM, Vanderpool DD, Finseth FR, Yuan Y-W, Fishman L. Ancient and recent introgression shape the evolutionary history of pollinator adaptation and speciation in a model monkeyflower radiation (Mimulus section Erythranthe). PLoS Genet. 2021;17(2):e1009095. pmid:33617525
View Article
PubMed/NCBI
Google Scholar

[375] View Article

[376] PubMed/NCBI

[377] Google Scholar

[ref108] 108. Rifkin JL, Castillo AS, Liao IT, Rausher MD. Gene flow, divergent selection and resistance to introgression in two species of morning glories (Ipomoea). Mol Ecol. 2019;28(7):1709–29. pmid:30451335
View Article
PubMed/NCBI
Google Scholar

[379] View Article

[380] PubMed/NCBI

[381] Google Scholar

[ref109] 109. Ruhsam M, Hollingsworth PM, Ennos RA. Early evolution in a hybrid swarm between outcrossing and selfing lineages in Geum. Heredity (Edinb). 2011;107(3):246–55. pmid:21448227
View Article
PubMed/NCBI
Google Scholar

[383] View Article

[384] PubMed/NCBI

[385] Google Scholar

[ref110] 110. Lepais O, Petit RJ, Guichoux E, Lavabre JE, Alberto F, Kremer A, et al. Species relative abundance and direction of introgression in oaks. Mol Ecol. 2009;18(10):2228–42. pmid:19302359
View Article
PubMed/NCBI
Google Scholar

[387] View Article

[388] PubMed/NCBI

[389] Google Scholar

[ref111] 111. Rand DM, Harrison RG. Ecological genetics of a mosaic hybrid zone: mitochondrial, nuclear, and reproductive differentiation of crickets by soil type. Evolution. 1989;43(2):432–49.
View Article
Google Scholar

[391] View Article

[392] Google Scholar

[ref112] 112. Aldridge G. Variation in frequency of hybrids and spatial structure among Ipomopsis (Polemoniaceae) contact sites. New Phytol. 2005;167(1):279–88. pmid:15948849
View Article
PubMed/NCBI
Google Scholar

[394] View Article

[395] PubMed/NCBI

[396] Google Scholar

[ref113] 113. Comes H, Abbott R. Molecular phylogeography, reticulation, and lineage sorting in mediterranean senecio sect. senecio (asteraceae). Evolution. 2001;55(10):1943–62.
View Article
Google Scholar

[398] View Article

[399] Google Scholar

[ref114] 114. Folk RA, Mandel JR, Freudenstein JV. Ancestral gene flow and parallel organellar genome capture result in extreme phylogenomic discord in a lineage of angiosperms. Syst Biol. 2017;66(3):320–37.
View Article
Google Scholar

[401] View Article

[402] Google Scholar

[ref115] 115. Tsitrone A, Kirkpatrick M, Levin D. A model for chloroplast capture. Evolution. 2003;57(8):1776–82.
View Article
Google Scholar

[404] View Article

[405] Google Scholar

[ref116] 116. Webb CJ. The Selection of Pollen and Seed Dispersal in Plants. Plant Species Biol. 1998;13(1):57–67.
View Article
Google Scholar

[407] View Article

[408] Google Scholar

[ref117] 117. Dobson SF. Competition for mates and predominant juvenile male dispersal in mammals. Anim Behav. 1982;30(4):1183–92.
View Article
Google Scholar

[410] View Article

[411] Google Scholar

[ref118] 118. Barreto FS, Watson ET, Lima TG, Willett CS, Edmands S, Li W, et al. Genomic signatures of mitonuclear coevolution across populations of Tigriopus californicus. Nat Ecol Evol. 2018;2(8):1250–7. pmid:29988158
View Article
PubMed/NCBI
Google Scholar

[413] View Article

[414] PubMed/NCBI

[415] Google Scholar

[ref119] 119. Bogdanova VS, Galieva ER, Kosterin OE. Genetic analysis of nuclear-cytoplasmic incompatibility in pea associated with cytoplasm of an accession of wild subspecies Pisum sativum subsp. elatius (Bieb.) Schmahl. Theor Appl Genet. 2009;118(4):801–9. pmid:19099285
View Article
PubMed/NCBI
Google Scholar

[417] View Article

[418] PubMed/NCBI

[419] Google Scholar

[ref120] 120. Gobron N, Waszczak C, Simon M, Hiard S, Boivin S, Charif D, et al. A cryptic cytoplasmic male sterility unveils a possible gynodioecious past for Arabidopsis thaliana. PLoS One. 2013;8(4):e62450. pmid:23658632
View Article
PubMed/NCBI
Google Scholar

[421] View Article

[422] PubMed/NCBI

[423] Google Scholar

[ref121] 121. Lee H-Y, Chou J-Y, Cheong L, Chang N-H, Yang S-Y, Leu J-Y. Incompatibility of nuclear and mitochondrial genomes causes hybrid sterility between two yeast species. Cell. 2008;135(6):1065–73. pmid:19070577
View Article
PubMed/NCBI
Google Scholar

[425] View Article

[426] PubMed/NCBI

[427] Google Scholar

[ref122] 122. Meiklejohn CD, Holmbeck MA, Siddiq MA, Abt DN, Rand DM, Montooth KL. An Incompatibility between a mitochondrial tRNA and its nuclear-encoded tRNA synthetase compromises development and fitness in Drosophila. PLoS Genet. 2013;9(1):e1003238. pmid:23382693
View Article
PubMed/NCBI
Google Scholar

[429] View Article

[430] PubMed/NCBI

[431] Google Scholar

[ref123] 123. Fishman L, Willis J. A cytonuclear incompatibility causes anther sterility in Mimulus hybrids. Evolution. 2006;60(7):1372–81.
View Article
Google Scholar

[433] View Article

[434] Google Scholar

[ref124] 124. Colicchio J, Monnahan PJ, Wessinger CA, Brown K, Kern JR, Kelly JK. Individualized mating system estimation using genomic data. Mol Ecol Resour. 2020;20(1):333–47. pmid:31519042
View Article
PubMed/NCBI
Google Scholar

[436] View Article

[437] PubMed/NCBI

[438] Google Scholar

[ref125] 125. Puzey JR, Willis JH, Kelly JK. Population structure and local selection yield high genomic variation in Mimulus guttatus. Mol Ecol. 2017;26(2):519–35. pmid:27859786
View Article
PubMed/NCBI
Google Scholar

[440] View Article

[441] PubMed/NCBI

[442] Google Scholar

[ref126] 126. Ross CL, Harrison RG. A fine-scale spatial analysis of the mosaic hybrid zone between Gryllus firmus and Gryllus pennsylvanicus. Evolution. 2002;56(11):2296–312. pmid:12487359
View Article
PubMed/NCBI
Google Scholar

[444] View Article

[445] PubMed/NCBI

[446] Google Scholar

[ref127] 127. Valbuena-Carabaña M, González-Martínez SC, Hardy OJ, Gil L. Fine-scale spatial genetic structure in mixed oak stands with different levels of hybridization. Mol Ecol. 2007;16(6):1207–19. pmid:17391407
View Article
PubMed/NCBI
Google Scholar

[448] View Article

[449] PubMed/NCBI

[450] Google Scholar

[ref128] 128. Abbott RJ. Plant speciation across environmental gradients and the occurrence and nature of hybrid zones. J Syst Evol. 2017;55(4):238–58.
View Article
Google Scholar

[452] View Article

[453] Google Scholar

[ref129] 129. Dudash MR, Ritland K. Multiple paternity and self-fertilization in relation to floral age in Mimulus guttatus (Scrophulariaceae). Am J Botany. 1991;78(12):1746–53.
View Article
Google Scholar

[455] View Article

[456] Google Scholar

[ref130] 130. Leclerc-Potvin C, Ritland K. Modes of self-fertilization in Mimulus guttatus (Scrophulariaceae): a field experiment. Am J Botany. 1994;81(2):199–205.
View Article
Google Scholar

[458] View Article

[459] Google Scholar

[ref131] 131. Ritland K. Correlated matings in the partial selfer mimulus guttatus. Evolution. 1989;43(4):848–59. pmid:28564194
View Article
PubMed/NCBI
Google Scholar

[461] View Article

[462] PubMed/NCBI

[463] Google Scholar

[ref132] 132. Ritland K, Ganders FR. Covariation of selfing rates with parental gene fixation indices within populations of mimulus guttatus. Evolution. 1987;41(4):760–71. pmid:28564350
View Article
PubMed/NCBI
Google Scholar

[465] View Article

[466] PubMed/NCBI

[467] Google Scholar

[ref133] 133. Karron JD, Holmquist KG, Flanagan RJ, Mitchell RJ. Pollinator visitation patterns strongly influence among-flower variation in selfing rate. Ann Bot. 2009;103(9):1379–83. pmid:19218584
View Article
PubMed/NCBI
Google Scholar

[469] View Article

[470] PubMed/NCBI

[471] Google Scholar

[ref134] 134. Karron JD, Mitchell RJ, Holmquist KG, Bell JM, Funk B. The influence of floral display size on selfing rates in Mimulus ringens. Heredity (Edinb). 2004;92(3):242–8. pmid:14666135
View Article
PubMed/NCBI
Google Scholar

[473] View Article

[474] PubMed/NCBI

[475] Google Scholar

[ref135] 135. Bartlett KB, Austin MW, Beck JB, Zanne AE, Smith AB. Beyond the usual climate? Factors determining flowering and fruiting phenology across a genus over 117 years. Am J Bot. 2023;110(7):e16188. pmid:37200535
View Article
PubMed/NCBI
Google Scholar

[477] View Article

[478] PubMed/NCBI

[479] Google Scholar

[ref136] 136. Ganjurjav H, Gornish E, Hu G, Schwartz M, Wan Y, Li Y, et al. Warming and precipitation addition interact to affect plant spring phenology in alpine meadows on the central Qinghai-Tibetan Plateau. Agric Forest Meteorol. 2020;287:107943.
View Article
Google Scholar

[481] View Article

[482] Google Scholar

[ref137] 137. Matthews ER, Mazer SJ. Historical changes in flowering phenology are governed by temperature × precipitation interactions in a widespread perennial herb in western North America. New Phytol. 2016;210(1):157–67. pmid:26595165
View Article
PubMed/NCBI
Google Scholar

[484] View Article

[485] PubMed/NCBI

[486] Google Scholar

[ref138] 138. Sianta SA, Kay KM. Parallel evolution of phenological isolation across the speciation continuum in serpentine-adapted annual wildflowers. Proc Biol Sci. 2021;288(1948):20203076.
View Article
Google Scholar

[488] View Article

[489] Google Scholar

[ref139] 139. Hunt B, Elliott T. Interaction of climatic variability with climatic change. Atmosphere-Ocean. 2004;42(3):145–72.
View Article
Google Scholar

[491] View Article

[492] Google Scholar

[ref140] 140. Rind D, Goldberg R, Ruedy R. Change in climate variability in the 21st century. Clim Change. 1989;14(1):5–37.
View Article
Google Scholar

[494] View Article

[495] Google Scholar

[ref141] 141. Salinger MJ. Climate variability and change: past, present and future – an overview. Clim Change. 2005;70(1):9–29.
View Article
Google Scholar

[497] View Article

[498] Google Scholar

[ref142] 142. Brauer CJ, Sandoval-Castillo J, Gates K, Hammer MP, Unmack PJ, Bernatchez L, et al. Natural hybridization reduces vulnerability to climate change. Nat Clim Change. 2023 Mar;13(3):282–9.
View Article
Google Scholar

[500] View Article

[501] Google Scholar

Figures

Abstract

Author summary

Introduction

Results

Introgression from M. nasutus in replicated contact zones

Patterns of ancestry at small spatial scales are persistent across years.

Partial reproductive isolation between admixture cohorts due to mating system

Partial phenological isolation between ancestry cohorts

Measured reproductive barriers accurately predict offspring ancestry shifts

Discussion

Repeated cases of introgression across the landscape

Complete organellar capture by hybridization

Persistent spatial structure with fluctuating reproductive isolation

Impact of fluctuating environments on hybrid zones

Materials and methods

Environmental data

Field sampling and collections

DNA Extraction and Illumina sequencing

Reference alignment, genotyping, and creation of SNP panels

Population structure analyses with ANGSD

Assigning local ancestry

Organellar haplotype networks

Estimating selfing rates

Calculating the strength of reproductive isolation

Using offspring to assess assortative mating

Statistical models

Supporting information

S1 Fig. Correlation between hybrid index and NGSAdmix structure results.

S2 Fig. Ancestry heterozygosity vs. hybrid index for maternal and offspring plants.

S3 Fig. PC5 identification of M. sookensis.

S4 Fig. PCA analysis of only M. guttatus and admixed individuals from CAC_Stream1 and CAC_Stream2.

S5 Fig. Complete mitochondrial capture of a M. nasutus haplotype in sympatric M. guttatus

S6 Fig. Selfing rate vs. hybrid index.

S7 Fig. Comparison of hybrid index, phenology, and maximum floral count for each plot within CAC_Stream1.

S8 Fig. Comparison of hybrid index, phenology, and maximum floral count for each plot within CAC_Stream2.

S1 Table. Distribution of samples from each plot in each PCA cluster.

S2 Table. Statistical model output summaries.

S3 Table. Summary of selfing and sibship estimation from BORICE.

S4 Table. Plot names, locations and ID assignments for each year of sampling.

S5 Table. Reference panel lines used for SNP panel creation.

Acknowledgments

References