Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2022 Nov 27;36(1):121–130. doi: 10.1111/jeb.14122

Intraspecific variation in reproductive barriers between two closely related Arabidopsis sister species

Antoine Perrier 1,2,✉, Yvonne Willi 1
PMCID: PMC10100320  PMID: 36436201

Abstract

Reproductive isolation (RI) is a critical component of speciation and varies strongly in timing and strength among different sister taxa, depending on, for example the geography of speciation and divergence time. However, these factors may also produce variation in timing and strength among populations within species. Here we tested for variation in the expression of RI among replicate population pairs between the sister taxa Arabidopsis lyrata subsp. lyrata and A. arenicola. While the former is predominantly outcrossing, the latter is predominantly selfing. We focused on intrinsic prezygotic and postzygotic RI as both species occur largely in allopatry. We assessed RI by performing within‐population crosses and interspecific between‐population crosses, and by raising offspring. RI was generally high between all interspecific population pairs, but it varied in timing and strength depending on population history. Prezygotic isolation was strongest between the closest‐related population pair, while early postzygotic isolation was high for all other population pairs. Furthermore, the timing and strength of RI depended strongly on cross direction. Our study provides empirical support that reproductive barriers between species are highly variable among population pairs and asymmetric within population pairs, and this variation seems to follow patterns typically described across species pairs.

Keywords: Arabidopsis, genetic distance, reproductive barrier, speciation, variation in reproductive isolation

Reproductive isolation (RI) is a critical component of speciation and varies strongly in timing and strength among different sister taxa, depending on, for example the geography of speciation and divergence time. However, these factors may also produce variation in timing and strength among populations within species. Here we tested for variation in postmating prezygotic and intrinsic postzygotic reproductive barriers among replicate population pairs of the recently diverged allopatric sister taxa Arabidopsis lyrata subsp. lyrata and A. arenicola. We found that RI among interspecific crosses was generally high, but varied strongly depending on population history and cross direction. As we discuss in the paper, this variation also seems to follow patterns typically described on the level of species.

graphic file with name JEB-36-121-g002.jpg

1. INTRODUCTION

The evolution of barriers to gene flow is essential for the speciation process and the maintenance of species (Coyne & Orr, 2004; Dobzhansky, 1937). The build‐up of reproductive isolation (RI) has, therefore, been studied intensively from diverse angles, including the type of barriers involved and the precise timing of expression during reproduction or the life cycle (e.g. Baack et al., 2015; Lowry et al., 2008; Matute & Cooper, 2021; Ramsey et al., 2003; Widmer et al., 2009). Such studies typically involve data collection on particular species pairs, by picking representatives of each species. However, estimates of RI may vary in timing of expression and strength within species depending on the populations studied (Corbett‐Detig et al., 2013; Cutter, 2012; Scopece et al., 2010; Widmer et al., 2009). This may especially apply to taxa that were in secondary contact in part of the range (e.g. Sweigart et al., 2007; Zuellig & Sweigart, 2018), or to sister taxa that do not show monophyletic splitting (e.g. Kozlowska et al., 2012). Despite the recurrent observation of within‐species variation in RI and its importance for understanding the processes leading to speciation, variation in the timing and strength of RI within taxa has rarely been addressed systematically (Cutter, 2012; Stankowski & Ravinet, 2021; The Marie Curie Speciation Network, 2012).

Reproductive barriers are usually distinguished based on the timing of expression relative to zygote formation. Premating barriers can include behavioural (Schaefer & Ruxton, 2015), ecological (Sobel et al., 2010) or mechanical (Grant, 1994) isolation. Another premating barrier is a selfing mating system (Hu, 2015; Wright et al., 2013). Postmating prezygotic barriers can be based on competitive mechanisms such as conspecific sperm or pollen precedence (Howard, 1999; Moore & Pannell, 2011) or on noncompetitive mechanisms such as signalling differences or gametophytic incompatibilities (Rieseberg & Willis, 2007; Swanson et al., 2004). Following fertilization, intrinsic postzygotic barriers can prevent the development, survival or fertility of hybrids (Coughlan & Matute, 2020; Coyne & Orr, 2004); such barriers are often linked to genetic incompatibilities (Dobzhansky, 1937; Muller, 1942). Finally, extrinsic postzygotic barriers can occur due to maladaptation of hybrids to either parental environment or due to mating preferences (Rundle & Nosil, 2005). A progression of speciation often requires the accumulation, interaction and/or coupling of two or more barriers to result in strong RI (Butlin & Smadja, 2018; Lowry et al., 2008).

The timing and strength of the expression of reproductive barriers may vary along the speciation continuum (Kulmuni et al., 2020). Whether pre‐ or postzygotic barriers contribute to speciation more has been an ongoing question in speciation research, with arguments for either scenario (Coughlan & Matute, 2020; Irwin, 2020; Lowry et al., 2008). Meanwhile, comparative analyses have suggested that which barrier evolves first depends on the taxon (Baack et al., 2015; Widmer et al., 2009). A more consistent finding has been that the strength of RI increases with increasing divergence time between species (Coyne & Orr, 1997; Matute & Cooper, 2021). This seems to apply particularly to genetically determined postzygotic barriers and species that evolve in allopatry (Ravinet et al., 2017), when closely related lineages become geographically isolated and diverge through genetic drift and selection, which result in the accumulation of RI (Orr, 1995). In contrast, increasing RI with increasing divergence time seems less consistent for prezygotic barriers (Christie & Strauss, 2018; Widmer et al., 2009; Willis & Donohue, 2017). Prezygotic barriers may evolve through different processes, such as reinforcement upon secondary contact, whereby selection leads to rapid accumulation of several such barriers (Butlin & Smadja, 2018).

Reproductive barriers are often asymmetric, that is their strength depends on the cross direction (Lewis & Crowe, 1958; Tiffin et al., 2001; Turelli & Moyle, 2007). Asymmetries in prezygotic barriers have been linked to divergence of genes necessary to discriminate against foreign gametes by the receiving sporophyte (Lewis & Crowe, 1958), as has been often found between species differing in mating system (Brys et al., 2014; Lewis & Crowe, 1958; Tiffin et al., 2001). In contrast, asymmetries in intrinsic postzygotic barriers have been linked to cytonuclear incompatibilities (Burton et al., 2013; Levin, 2003; Tiffin et al., 2001; Turelli & Moyle, 2007). In addition, mechanisms with strong parental control have been associated with so called parent‐of‐origin patterns in RI (Haig & Westboy, 1991; Lewis & Crowe, 1958; Tiffin et al., 2001), such as genomic imprinting (Wilkins & Haig, 2003) and male‐driven interlocus contest evolution (Rice & Holland, 1997).

A consequence of the diversity of processes and factors affecting the timing and strength of reproductive barriers is that RI may also vary depending on the population pairs used for interspecific contrasts. The patterns underlying variation in RI have been explored mostly between species without assessing intraspecific variation (Baack et al., 2015; Jewell et al., 2012; Lowry et al., 2008; Matute & Cooper, 2021; Moyle et al., 2004; Ramsey et al., 2003; Turissini et al., 2018; Widmer et al., 2009; Willis & Donohue, 2017). Only few studies compared the timing or strength of RI among replicate population pairs between two species, despite compelling evidence that RI may be heterogeneous depending on the populations under study (Christie & Strauss, 2019; Cutter, 2012; Zuellig & Sweigart, 2018). Assessing intraspecific variation in RI between two species may be crucial to uncover processes driving speciation at a finer scale than assessed in studies based on interspecific comparisons.

In two recently evolved sister species—the North American Arabidopsis lyrata subsp. lyrata (L.) and A. arenicola, we tested for differences in the timing and strength of postpollination RI among replicate population pairs differing in genetic distance. Both species occur nowadays largely in allopatry (Figure 1a), A. lyrata in the eastern and mid‐western US and mostly southern Canada and A. arenicola in subarctic to arctic eastern and mid‐western Canada and on Greenland (Warwick et al., 2006). A small contact zone in Saskatchewan, Canada was documented by herbaria specimen. Phylogeographic studies suggest that the predominantly outcrossing A. lyrata colonized North America via Beringia during the Quaternary and that A. arenicola is a young sister species of North American A. lyrata (Novikova et al., 2016; Schmickl et al., 2010). Further work on A. lyrata shows that the species expanded its range after the last glacial maximum from two refugia in the USA, one in the eastern and one in the mid‐western part of the current range (Griffin & Willi, 2014; Willi et al., 2018). During that last range expansion, A. arenicola likely emerged from the western lineage of A. lyrata (Hohmann et al., 2014; Novikova et al., 2016). Arabidopsis arenicola expanded its range towards the north, while the western lineage of A. lyrata continued its expansion laterally along the shores of the Great Lakes where range limits seem to coincide with niche limits (Lee‐Yaw et al., 2018; Willi et al., 2018). Because the two species are currently spatially well isolated, apart from the one small hybrid zone, we focused on testing for intraspecific variation in postpollination prezygotic and intrinsic postzygotic reproductive barriers, targeting fertilization, hybrid seed formation, survival, and fertility. We further assessed whether RI varied between population pairs depending on their genetic distance, and whether RI was asymmetric in strength within population pairs.

FIGURE 1.

FIGURE 1

Open in a new tab

Distribution map of Arabidopsis arenicola and A. lyrata with the six populations studied and genetic distances among them. (a) The blue shaded area represents the current range of A. arenicola, and the green shaded area represents the range of A. lyrata subsp. lyrata. Black triangles and dots represent the position of the studied populations of A. arenicola and A. lyrata, respectively. (b) Genetic distances among the six populations is represented by their relative position along the first two principal component axes of a principal component analysis on intergenic SNPs derived from pooled whole‐genome sequencing. A. arenicola populations are represented by blue triangles, A. lyrata populations by green dots.

2. MATERIAL AND METHODS

2.1. Plant material

Seeds of two populations of A. arenicola were collected in southern areas of its distribution, in northern Manitoba (AareMB) and eastern Quebec (AareQC), and seeds of four populations of A. lyrata were sampled in the western and eastern range of its distribution (Table S1, Figure 1a, labelled according to species and Canadian provinces or US states, respectively). The four populations of A. lyrata were chosen to represent variation in postglacial history across the species' range and in the role of genetic drift. The A. lyrata population from the north shore of Lake Superior, Ontario (AlyrON) represented the western genetic lineage that presumably gave rise to A. arenicola. This population was closest related to A. arenicola (see Results), and it had itself a history of postglacial colonization going back to the Driftless Area of Wisconsin (Willi et al., 2018), where the other western population (AlyrWI) came from. From the eastern genetic lineage, we chose a population from Upstate New York (AlyrNY) and one from Maryland (AlyrMD), both with a history of less genetic drift but longer divergence from A. arenicola. These four populations correspond to ON11, WI1, NY5 and MD2 in Willi et al. (2018). The two populations of A. arenicola were chosen because of their southern, but distant‐from‐each‐other location, assuming that they were generally more closely related to A. lyrata than more northern A. arenicola populations, with presumably a longer expansion history. Leaf material and seeds were collected from 30 to 50 plants per population in the field between 2007 and 2017 (Table S1). Leaf material was silica‐dried, and seeds were stored in separate bags for each sampled maternal plant at 4°C under dark and dry conditions.

Populations varied in the mating system. The two populations of A. arenicola as well as AlyrON, the population of A. lyrata closest‐related with A. arenicola, started setting seeds in the greenhouse before petals were fully open; they self‐fertilized autonomously. In contrast, the other three A. lyrata populations, AlyrWI, AlyrNY and AlyrMD, were mainly self‐incompatible and outcrossing, representing the mating system that is typical across the species' distribution, particularly in its core (Griffin & Willi, 2014). Selfing in A. arenicola and in A. lyrata in the area where the former likely evolved may be a powerful prezygotic barrier (Hu, 2015; Wright et al., 2013) and most likely contributed to the initiation of the speciation process by preventing cross‐pollination. In this study, the role of selfing as a reproductive barrier was not explicitly investigated, as we focused on postpollination barriers.

2.2. Genetic distance

Genetic distance between each population pair was calculated based on pooled, whole‐genome sequence information. DNA of 25 individuals per population was extracted, pooled in equimolar concentrations, pair‐end sequenced and sequences analysed following Willi et al. (2018) (data for A. lyrata was published under Bioproject PRJEB19338 (Willi et al., 2018); new data for A. arenicola was published under Bioproject PRJNA885817). We extracted single nucleotide polymorphisms (SNPs) using VCFtools 0.1.14 (Danecek et al., 2011). We kept only biallelic SNPs that had a coverage of 50–500, a minor population allele frequency of 0.03, and a maximum of 10% missing data across populations. Further filtering ensured that only SNPs of intergenic regions were retained, ending up with 77 859 SNPs. We estimated genetic distance as a proxy of population differentiation. We performed a principal component analysis (PCA) using the package pcadapt (Privé et al., 2020) in R version 4.1.2 (R Core Team, 2021). The genetic distance was calculated as the Euclidian distance between each population pair (Shirk et al., 2017), based on the first two principal components. These two PC axes explained 75.64% of variation.

2.3. Crossing experiment

To assess reproductive isolation between the two species, we produced within‐population, within‐species crosses (WSC) as well as between‐species crosses (BSC). In 2016 and 2017, we raised 26 individuals of different field‐collected maternal seed families per population in growth chambers and later transferred them to a greenhouse for crossing (see Table S2 for raising conditions). We performed non‐reciprocal WSC by crossing 12 randomly chosen “mother” plants (pollen recipients) with 12 other randomly chosen “father” plants (pollen donors) of the same population, resulting in 12 distinct WSC crosses per population. To generate F1 hybrids (BSC), we paired each A. arenicola population with two A. lyrata populations (detailed in Table S3). “Mother” individuals used to generate WSC were then used as both pollen recipients and pollen donors with randomly assigned plants of the partner population, resulting in 24 reciprocal BSC per population pair (12 per cross direction). Hand pollination was performed on buds that had been emasculated before the opening of anthers to prevent cross‐ and spontaneous self‐pollination. Each cross was attempted 3 times initially, to test for crossing success. Crossing continued until we had at least five siliques. Siliques were collected when ripe, dried for two weeks at ambient temperature in the dark and then stored as described before.

Crossing success was evaluated based on whether a pollinated pistil developed into a mature fruit. When the pistil contracted within a few days after pollination, the cross was considered unsuccessful. In the rare case of crossing success being unclear (e.g. short fruits, very thin fruits), fruits were opened, and seed set was assessed; crosses that had resulted in at least one fertilized ovule, even if aborted, were interpreted as a sign of at least partial fertilization. At least two mature fruits per cross were opened and the number of unfertilized ovules, early aborted embryos, late aborted embryos, and healthy‐looking seeds were counted.

2.4. Raising of offspring & postzygotic performance

To assess hybrid performance, we raised WSC and BSC offspring in a climate chamber. In total, we had seeds of 65 of the attempted 72 WSC (10 to 12 per population, Table S3). Due to high cross and seed‐set failure, we included only 36 BSC with sufficient seeds: 6 per cross direction of each population pair, randomly chosen within the available crosses. Crosses of AareQC(♀)xAlyrWI and AlyrON(♀)xAareQC systematically failed to produce fruits or healthy seeds and were excluded. For each cross, we only sowed seeds that were considered healthy (detailed in 2.3 Crossing experiment). For each cross, two seeds per pot (one if few seeds) were sown in three individual pots randomly assigned to one of three replicate blocks. Pots were filled with a standard substrate mixture of washed river sand and peat (1:1.5 sand:peat). Within blocks, pots were randomly positioned across two 54‐cell propagation trays (BK Qualipot, Otelfingen, Switzerland). Pots were saturated with water and then placed for 20 days at 4 °C in the dark in the climate chamber (Climecab 1400, Kälte 3000 AG, Landquart, Switzerland) for stratification. The climate chamber experiment simulated conditions close to the natural growth cycle of both species (growth conditions detailed in Table S2). Germination was initiated in a first phase of seven weeks simulating environmental conditions representative of fall. After 28 days, seedlings were randomly thinned to one per pot. The remaining seedlings were then vernalized (constant 4°C, low light) to simulate a winter phase of six weeks, followed by a spring and summer phase of 18 weeks to allow for reproduction. Trays were watered regularly, and fertilizer was added starting after vernalization, once every two weeks (2% v/v Wuxal universal fertilizer, Hauert Manna Düngerwerke GmbH, Nürnberg, Germany).

Hybrid performance was tracked at the level of the seedling until thinning, then at the level of the individual plant in a pot for the entire length of the experiment (recording details in Table S2). We recorded the day of germination, the day of death, and reproductive output four weeks after the first flower opened, by the total number of flowers produced until then and the remaining flower buds. Male fertility was estimated on two freshly opened flowers, collected 3 to 12 days after first flowering. All six anthers per flower were collected in 15 ml tubes and dried for 48 h at 60°C. Pollen counting was performed with a CASY TTC Cell Counter (Schärfe System GmbH, Reutlingen, Germany). Samples were first re‐hydrated with 12 ml of isotonic solution (CASYton, OMNI Life Science GmbH, Basel, Switzerland), and pollen was separated from the pollen bags in an ultrasound bath for 3 min. Pollen of three aliquots of 400 μl were counted per sample based on a particle‐size range of 8–40 μm. As we observed two distinct peaks in the size distribution of pollen grains in some samples, one at ~10 μm and one at 20 μm, we considered pollen with a size >15 μm, including the peak at 20 μm, to be viable (Willi, 2013). Pollen viability was calculated for each flower as the proportion of viable pollen grains relative to the total number of pollen grains.

2.5. Statistical analyses

We tested for variation in RI by focusing on intrinsic pre‐ and postzygotic barriers. Other prezygotic barriers (e.g. geographical isolation, difference in phenology, pollinators, or mating system) or environment‐driven postzygotic barriers (e.g. selection against hybrids in parental habitats) may also play a role in the system but were not the scope of this study. A first main analysis targeted the strength and reciprocity of total intrinsic postpollination RI. For this we calculated multiplicative performance (MP) on the level of a cross as the product of nine cross‐mean performance estimates describing successive life stages: crossing, fertilization of ovules, early and late seed development, germination, survival, bolting, reproductive output, and pollen viability (all described below and in Table S4). Then RI was calculated for each mother plant involved in both WSC and BSC using the formula derived from Sobel and Chen (2014): RI = 1–2 (MP BSC/[MP BSC + MP WSC]). An RI value of 1 represents complete reproductive isolation, one of 0 no isolation, and one of −1 complete facilitation of heterospecific pollination. Here, MP WSC was the average MP of the WSC performed on the two parental individuals used to generate each BSC. The independent variables were population pair, cross direction (A. arenicola as mother = 0, A. lyrata = 1) and their interaction analysed by an ANOVA in R. In‐depth analyses then tested whether RI MP of each cross direction within each population pair (hereafter population cross) was significantly different from 0 using the package emmeans (Lenth et al., 2022). Estimates of reproductive isolation were assumed to follow Gaussian distributions. This analysis focused on the absolute strength of each barrier (Coyne & Orr, 2004), that is only considering individuals that had successfully passed the previous life stage. For comparison, we also assessed the relative contribution of each barrier to RI.

A next analysis targeted the timing, strength and reciprocity of reproductive barriers. Prezygotic RI was calculated based on crossing success (fraction of successful crossing attempts) and fertilization rate of ovules. Postzygotic RI was based on the success of early seed development (fraction of late aborted and healthy seeds to fertilized ovules), success of late seed development (fraction of healthy seeds to late aborted and healthy seeds), germination rate, survival rate to flowering, bolting rate, reproductive output, and pollen viability (Table S4). Estimates of RI of life stages were split into separate entries (rows) such that life stage could be added in the model as an explanatory variable. The independent variables were life stage, population pair, cross direction (A. arenicola as mother = 0, A. lyrata = 1) and their interactions, analysed by ANOVA.

3. RESULTS

The six populations showed clear genetic structuring based on intergenic SNPs. The first two principal components, PC 1 (51.47%) and PC 2 (24.17%), suggested three clusters: one composed of the two A. arenicola populations and north‐western A. lyrata AlyrON, one with the two eastern A. lyrata populations AlyrNY and AlyrMD, and finally the population AlyrWI (Figure 1b). The genetic distance between populations estimated as pairwise Euclidian distance along the first two PC axes ranked the population pairs in the following order of increasing genetic distance (Table S5): AareQC – AlyrON (0.06) < AareMB – AlyrNY (1.11) < AareMB – AlyrMD (1.15) < AareQC – AlyrWI (1.44).

Between‐species crosses revealed significant reproductive isolation assessed on multiplicative performance RI MP , ranging from moderate isolation (0.37) to complete isolation (1) (Table S6, Figure 2a). Analyses supported that RI MP varied significantly among population pairs and cross direction, and the two interacted in their effect on RI (Table 1). RI MP seemed to increase with genetic distance between populations, but only for crosses with A. arenicola as mothers (Figure 2a, Table S7, Figure S8a). Crosses with A. lyrata as mothers showed less variation and seemed generally stronger than crosses with A. arenicola as mothers.

FIGURE 2.

FIGURE 2

Open in a new tab

Total reproductive isolation (RI) between population pairs of Arabidopsis arenicola and A. lyrata, and reproductive isolation split by timing of expression. Populations are sorted by increasing genetic distance (left to right). Reproductive isolation was estimated for (a) multiplicative performance (RI MP , dark grey for crosses with A. arenicola as mothers, light grey for reciprocal crosses) and (b) nine successive life stages, from the success of performing a cross to pollen viability of offspring. Bars represent population means (based on relative contributions in (b)), and numbers indicate average RI values significantly different from 0. Significance is indicated as: *p < 0.05, **p < 0.01, ***p < 0.001. Average RI and test statistics are reported in Table S6, relative contributions to RI in Table S9.

TABLE 1.

The effect of population pair (PP), cross direction (CD), life stage (LS, if applicable) and their interactions on reproductive isolation between Arabidopsis arenicola and A. lyrata

Fixed effects Reproductive isolation
Total RI (N = 88) Life stage‐specific RI (N = 409)
df F df F
Population pair 3 5.05** 3 6.63***
Cross direction 1 4.29* 1 6.05*
Life stage – 8 26.05***
PP × CD 3 6.09*** 3 8.15***
LS × PP – 20 12.38***
LS × CD – 4 74.91***
LS × PP × CD – 7 2.83**
Open in a new tab

Note: Population crosses with too low a replication number were excluded from the analysis (replication number detailed in Table S6). Variance explained (R 2) by the models on total RI and life stage‐specific RI was 0.33 and 0.66, respectively. F‐ratios with p‐values < 0.05 are written in bold; significance is indicated: (*) p < 0.1, *p < 0.05, **p < 0.01, ***p < 0.001.

When reproductive isolation was assessed for the nine pre‐ and postzygotic barriers separately, the effect of life stage and its interactions with population pair and with cross direction were significant (Table 1, Figure 2b). Reproductive isolation was considerable for prezygotic crossing success and for fertilization rate of ovules as well as during postzygotic late seed development and germination (Table S6, Figure 2b, relative RI values reported in Table S9 for comparison); all other life stages had generally lower RI. Population pairs varied in the timing of RI. The most obvious pattern was that RI in prezygotic crossing success was strongest in the population pair with the lowest genetic distance (Table S7, Figure S8b). In the other three population pairs, both prezygotic RI in crossing success and fertilization rate of ovules tended to increase with genetic distance, and RI in late seed development or germination were generally important (Table S7, Figure S8c). Furthermore, RI during particular life stages varied considerably with cross direction. Prezygotic RI and postzygotic RI in germination seemed strongest in crosses with A. lyrata as mothers, while postzygotic RI in late seed development tended to be very important when A. arenicola was the mother plant (Table S10, Figure S8c). Also, the three‐way interaction between life stage, population pair and cross direction was significant, indicating some deviations from these more general patterns. Finally, some population pairs also revealed evidence for partial hybrid advantage, for example for reproductive output if A. arenicola was the mother and pollen came from the closest‐related A. lyrata population, or for bolting when A. arenicola was the mother and pollen donor was the second‐closest A. lyrata population, overall contributing to lower total RI in these crosses.

4. DISCUSSION

The factors leading to variation in interspecific reproductive isolation (RI) have rarely been contrasted among multiple population pairs of two species, despite ample evidence for such variation (Cutter, 2012; The Marie Curie Speciation Network, 2012). In this study, we assessed the variation in timing and strength of postpollination prezygotic and intrinsic postzygotic reproductive barriers among replicate crosses between populations of the recently diverged Arabidopsis arenicola and A. lyrata. Total intrinsic postpollination RI among interspecific crosses was substantial, with considerable variation depending on population pair that could be associated with genetic distance and cross direction (Table 1, Figure 2a). When total RI was split into various pre‐ and postzygotic barriers, high variation in the timing and strength of these life stage‐specific barriers emerged, some of which was again linked to genetic distance between populations (Table 1, Figure 2b, Figure S9) and cross direction (Figure 2b).

We found strong total RI between A. arenicola and A. lyrata, even leading to complete isolation (RI = 1). While strong RI between sister species is not uncommon (Baack et al., 2015; Lowry et al., 2008), it seems remarkable for our study system considering the recent divergence, probably after the end of the last glacial maximum. Rapid accumulation of genetic incompatibilities in our system may have resulted from both selection and genetic drift. Divergent selection may have led to the increased fixation of different alleles in each of the species as they nowadays occur in largely distinct biogeographic areas with very different climate (Figure 1a). Such divergent selection increases the opportunity for genetic incompatibilities to arise (Rundle & Nosil, 2005). Furthermore, strong exposure to genetic drift has been documented in A. lyrata associated with rapid postglacial range expansion (Willi et al., 2018), and this must have occurred similarly in A. arenicola given that all area of occurrence had been covered by ice during the last glaciation (Schmickl et al., 2010). Genetic drift in A. arenicola must have been high also due to its selfing mating system (Nordborg & Donnelly, 1997; Pollak, 1987). Selfing further provides strong prepollination isolation (Hu, 2015), thereby facilitating the accumulation of genetic incompatibilities (Wright et al., 2013).

Apart from generally high RI, we also found significant variation in total RI among population pairs and between the two cross directions (Table 1, Figure 2). When A. arenicola was the mother, total RI increased with genetic distance between the population pairs (dark bars in Figure 2a). However, the pattern was mainly produced not by isolating mechanisms being weaker when the A. lyrata pollen‐donor population was closer related, but by an additional positive effect of hybridization, for example during F1 reproductive output or bolting in these population pairs (Figure 2b). In contrast, when A. lyrata was the mother of interspecific crosses, total RI was highest for the population pair with the lowest genetic distance (Figure 2a, first bright bar from the left) and quite lower in the population pair with highest distance. Here, positive effects of hybridization did not play a significant role. One potential explanation for partial hybrid superiority when A. arenicola was the mother plant and A. lyrata a more closely related pollen donor is heterosis due to mutation accumulation by selfing, combined with less time for accumulating genetic incompatibilities between these more related population pairs. However, we observed RI in pollen viability in the closest population pair when A. arenicola was the mother, indicating that some incompatibilities have still been able to accumulate.

Strong RI across population pairs was mainly expressed at the postpollination prezygotic and early postzygotic life stages (Figure 2b, Table S6). But the specifics of life stage‐based RI differed among population pairs and with cross direction (significant two‐way interactions). One interesting pattern was that the crosses between the closest population pair commonly failed to initiate fruit development, and more so when A. lyrata was the mother plant. Postpollination prezygotic RI (including ovule fertilization) sometimes also occurred in more distantly related population pairs when A. lyrata was the mother plant, but never to this extent. Stronger prezygotic isolation in the closest population pair could result from local idiosyncrasies; however, it may also be an indication that the evolution of isolating barriers happened in a different context compared to the other populations. High prezygotic RI may be due to reinforcement during early speciation (Butlin & Smadja, 2018). In this scenario, in areas where the two species were in close spatial proximity with higher opportunity for gene flow, prezygotic RI may have evolved in A. lyrata populations to limit the cost of pollen interferences and hybridization with a highly successful novel species.

For more distantly related population pairs, RI was, apart from the stages of crossing and fertilization, mainly expressed during late seed development or germination. RI by later hybrid failure was generally rare. An increase in postzygotic barriers has classically been attributed to the accumulation of genetic incompatibilities between diverging lineages (Orr, 1995). This pattern has been found particularly in allopatric species (Baack et al., 2015; Christie & Strauss, 2018; Coyne & Orr, 1997; Lowry et al., 2008; Matute & Cooper, 2021; Turissini et al., 2018; Willis & Donohue, 2017) and recently also between isolated lineages within a species (Barnard‐Kubow & Galloway, 2017). Beyond, our results reject the hypothesis that prezygotic barriers are generally stronger than postzygotic barriers (Lowry et al., 2008), but they are in line with postzygotic barriers being more likely to evolve in allopatry (Ravinet et al., 2017) and prezygotic barriers when species once were in contact and shared gene flow (Butlin & Smadja, 2018).

As indicated above, we found significant asymmetry in life stage‐specific RI. When selfing A. arenicola was the mother, RI in prezygotic crossing success and ovule fertilization was systematically lower (Figure S8b, Table S10). Such asymmetries have often been linked to lower capacity of self‐compatible species to discriminate against foreign pollen (Lewis & Crowe, 1958; Tiffin et al., 2001). However, the highest RI in crossing success was found when selfing A. lyrata from Ontario were mother plants, indicating that other processes may be at play in driving asymmetric prezygotic RI, such as the one discussed in the paragraph above. Asymmetry further occurred for postzygotic seed development and germination (Figure S8c, Table S10). In the three distantly related population pairs, RI in late seed development was important when A. arenicola was the mother plant. In contrast, in two of the three distantly related population pairs, RI in germination was strong when A. lyrata was the mother plant. Asymmetries in hybrid seed formation and germination may be imputed to genomic imprinting, such as imbalances in the parental dosage of the endosperm (Haig & Westboy, 1991; Lafon‐Placette et al., 2018; see also Supplementary S11, S12, and S13).

In summary, our results emphasize that there can be considerable variation in the timing and strength of reproductive isolation among populations of a species pair. Taking into account variation in RI between replicate population pairs is essential to assess the evolutionary processes driving speciation (Cutter, 2012; The Marie Curie Speciation Network, 2012). Our results indeed indicate that several hard‐to‐distinguish and partly overlapping processes may be involved. Here we discussed the potential implications of past contact and gene flow, the time of separate histories, genetic drift linked to past range dynamics and mating system shift, mutation accumulation, and divergent selection; that is the same processes leading to variation in RI between species pairs. To put results on RI in context, it is thus important to describe these parameters for the selected population pairs or target populations. Similarly, in the study of the speciation continuum, the selection of population pairs should be adjusted to the research questions. Finally, results have important implications for taxonomy and conservation. Young taxa can accumulate RI rapidly (Escudero et al., 2019), including both pre‐ and postzygotic barriers. This insight is especially important in the context of intrinsic postzygotic barriers, which have often been thought to require considerable time to evolve into strong barriers (Coughlan & Matute, 2020; Coyne & Orr, 1997; Matute & Cooper, 2021; Turissini et al., 2018). Recognizing such young species is thus important in the assessment and preservation of biodiversity.

AUTHOR CONTRIBUTIONS

All authors contributed to the study design. YW collected the seeds in the field, AP raised and crossed plants. AP analysed the data and wrote the manuscript, with inputs by YW.

FUNDING INFORMATION

This work was supported by the Swiss National Science Foundation (31003A_166322).

CONFLICT OF INTEREST

The authors declare that there is no conflict of interest.

PEER REVIEW

The peer review history for this article is available at https://publons.com/publon/10.1111/jeb.14122.

Supporting information

Appendix S1: Supporting Information

Click here for additional data file. (333KB, docx)

ACKNOWLEDGEMENTS

For assistance with crosses, seed counts and sowing, we thank Georg Armbruster, Olivier Bachmann, Markus Funk, Susanna Riedl and Darío Sánchez‐Castro. Pollen data produced and analysed in this paper were generated in collaboration with the Genetic Diversity Centre (GDC), ETH Zurich. For his advices on reproductive isolation and assistance in experimental design, we thank Kay Lucek. Collection permits were provided by the New York State Office of Parks, the Wisconsin Department of Natural Resources, and the United States National Park Service.

Perrier, A. , & Willi, Y. (2023). Intraspecific variation in reproductive barriers between two closely related Arabidopsis sister species. Journal of Evolutionary Biology, 36, 121–130. 10.1111/jeb.14122

DATA AVAILABILITY STATEMENT

Genetic data for A. lyrata were published under Bioproject PRJEB19338 (Willi et al., 2018; https://www.ncbi.nlm.nih.gov/bioproject/PRJEB19338/); new data for A. arenicola is stored under Bioproject PRJNA885817 (https://www.ncbi.nlm.nih.gov/bioproject/PRJNA885817/). The intergenic SNP table, raw phenotypic data and reproductive isolation estimates are stored in Zenodo (https://doi.org/10.5281/zenodo.7125244).

REFERENCES

  1. Baack, E. , Melo, M. C. , Rieseberg, L. H. , & Ortiz‐Barrientos, D. (2015). The origins of reproductive isolation in plants. The New Phytologist, 207, 968–984. 10.1111/nph.13424 [DOI] [PubMed] [Google Scholar]
  2. Barnard‐Kubow, K. B. , & Galloway, L. F. (2017). Variation in reproductive isolation across a species range. Ecology and Evolution, 7, 9347–9357. 10.1002/ece3.3400 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Brys, R. , Broeck, A. V. , Mergeay, J. , & Jacquemyn, H. (2014). The Contribution of mating system variation to reproductive isolation in two closely related Centaurium species (Gentianaceae) with a generalized flower morphology. Evolution, 68, 1281–1293. 10.1111/evo.12345 [DOI] [PubMed] [Google Scholar]
  4. Burton, R. S. , Pereira, R. J. , & Barreto, F. S. (2013). Cytonuclear genomic interactions and hybrid breakdown. Annual Review of Ecology, Evolution, and Systematics, 44, 281–302. 10.1146/annurev-ecolsys-110512-135758 [DOI] [Google Scholar]
  5. Butlin, R. K. , & Smadja, C. M. (2018). Coupling, reinforcement, and speciation. The American Naturalist, 191, 155–172. 10.1086/695136 [DOI] [PubMed] [Google Scholar]
  6. Christie, K. , & Strauss, S. Y. (2018). Along the speciation continuum: Quantifying intrinsic and extrinsic isolating barriers across five million years of evolutionary divergence in California jewelflowers. Evolution, 72, 1063–1079. 10.1111/evo.13477 [DOI] [PubMed] [Google Scholar]
  7. Christie, K. , & Strauss, S. Y. (2019). Reproductive isolation and the maintenance of species boundaries in two serpentine endemic jewelflowers. Evolution, 73, 1375–1391. 10.1111/evo.13767 [DOI] [PubMed] [Google Scholar]
  8. Corbett‐Detig, R. B. , Zhou, J. , Clark, A. G. , Hartl, D. L. , & Ayroles, J. F. (2013). Genetic incompatibilities are widespread within species. Nature, 504, 135–137. 10.1038/nature12678 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Coughlan, J. M. , & Matute, D. R. (2020). The importance of intrinsic postzygotic barriers throughout the speciation process. Philosophical Transactions of the Royal Society B: Biological Sciences, 375, 20190533. 10.1098/rstb.2019.0533 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Coyne, J. A. , & Orr, H. A. (1997). “Patterns of speciation in Drosophila” revisited. Evolution, 51, 295–303. 10.2307/2410984 [DOI] [PubMed] [Google Scholar]
  11. Coyne, J. A. , & Orr, H. A. (2004). Speciation. Sinauer Associates. [Google Scholar]
  12. Cutter, A. D. (2012). The polymorphic prelude to Bateson–Dobzhansky–Muller incompatibilities. Trends in Ecology & Evolution, 27, 209–218. 10.1016/j.tree.2011.11.004 [DOI] [PubMed] [Google Scholar]
  13. Danecek, P. , Auton, A. , Abecasis, G. , Albers, C. A. , Banks, E. , DePristo, M. A. , Handsaker, R. E. , Lunter, G. , Marth, G. T. , Sherry, S. T. , McVean, G. , Durbin, R. , & 1000 Genomes Project Analysis Group . (2011). The variant call format and VCFtools. Bioinformatics, 27, 2156–2158. 10.1093/bioinformatics/btr330 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Dobzhansky, T. (1937). Genetics and the origin of species. Columbia University Press. [Google Scholar]
  15. Escudero, M. , Lovit, M. , Brown, B. H. , & Hipp, A. L. (2019). Rapid plant speciation associated with the last glacial period: Reproductive isolation and genetic drift in sedges. Botanical Journal of the Linnean Society, 190, 303–314. 10.1093/botlinnean/boz016 [DOI] [Google Scholar]
  16. Grant, V. (1994). Modes and origins of mechanical and ethological isolation in angiosperms. Proceedings of the National Academy of Sciences, 91, 3–10. 10.1073/pnas.91.1.3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Griffin, P. C. , & Willi, Y. (2014). Evolutionary shifts to self‐fertilisation restricted to geographic range margins in North American Arabidopsis lyrata . Ecology Letters, 17, 484–490. 10.1111/ele.12248 [DOI] [PubMed] [Google Scholar]
  18. Haig, D. , & Westboy, M. (1991). Genomic imprinting in endosperm: its effect on seed development in crosses between species, and its implications for the evolution of apomixis. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences, 333, 1–13. 10.1098/rstb.1991.0057 [DOI] [Google Scholar]
  19. Hohmann, N. , Schmickl, R. , Chiang, T.‐Y. , Lučanová, M. , Kolář, F. , Marhold, K. , & Koch, M. A. (2014). Taming the wild: Resolving the gene pools of non‐model Arabidopsis lineages. BMC Evolutionary Biology, 14, 224. 10.1186/s12862-014-0224-x [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Howard, D. J. (1999). Conspecific sperm and pollen precedence and speciation. Annual Review of Ecology and Systematics, 30, 109–132. 10.1146/annurev.ecolsys.30.1.109 [DOI] [Google Scholar]
  21. Hu, X.‐S. (2015). Mating system as a barrier to gene flow. Evolution, 69, 1158–1177. 10.1111/evo.12660 [DOI] [PubMed] [Google Scholar]
  22. Irwin, D. E. (2020). Assortative mating in hybrid zones is remarkably ineffective in promoting speciation. The American Naturalist, 195, E150–E167. 10.1086/708529 [DOI] [PubMed] [Google Scholar]
  23. Jewell, C. , Papineau, A. D. , Freyre, R. , & Moyle, L. C. (2012). Patterns of reproductive isolation in Nolana (chilean Bellflower). Evolution, 66, 2628–2636. 10.1111/j.1558-5646.2012.01607.x [DOI] [PubMed] [Google Scholar]
  24. Kozlowska, J. L. , Ahmad, A. R. , Jahesh, E. , & Cutter, A. D. (2012). Genetic variation for postzygotic reproductive isolation between Caenorhabditis briggsae and Caenorhabditis sp. 9. Evolution, 66, 1180–1195. 10.1111/j.1558-5646.2011.01514.x [DOI] [PubMed] [Google Scholar]
  25. Kulmuni, J. , Butlin, R. K. , Lucek, K. , Savolainen, V. , & Westram, A. M. (2020). Towards the completion of speciation: the evolution of reproductive isolation beyond the first barriers. Philosophical Transactions of the Royal Society B: Biological Sciences, 375, 20190528. 10.1098/rstb.2019.0528 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Lafon‐Placette, C. , Hatorangan, M. R. , Steige, K. A. , Cornille, A. , Lascoux, M. , Slotte, T. , & Köhler, C. (2018). Paternally expressed imprinted genes associate with hybridization barriers in Capsella . Nature Plants, 4, 352–357. 10.1038/s41477-018-0161-6 [DOI] [PubMed] [Google Scholar]
  27. Lee‐Yaw, J. A. , Fracassetti, M. , & Willi, Y. (2018). Environmental marginality and geographic range limits: A case study with Arabidopsis lyrata ssp. lyrata . Ecography, 41, 622–634. 10.1111/ecog.02869 [DOI] [Google Scholar]
  28. Lenth, R. V. , Buerkner, P. , Herve, M. , Love, J. , Miguez, F. , Riebl, H. , & Singmann, H. (2022). emmeans: Estimated marginal means, aka least‐squares means. https://cran.r‐project.org/package=emmeans [Google Scholar]
  29. Levin, D. A. (2003). The cytoplasmic factor in plant speciation. Systematic Botany, 28, 5–11. [Google Scholar]
  30. Lewis, D. , & Crowe, L. K. (1958). Unilateral interspecific incompatibility in flowering plants. Heredity, 12, 233–256. 10.1038/hdy.1958.26 [DOI] [Google Scholar]
  31. Lowry, D. B. , Modliszewski, J. L. , Wright, K. M. , Wu, C. A. , & Willis, J. H. (2008). The strength and genetic basis of reproductive isolating barriers in flowering plants. Philosophical Transactions of the Royal Society B: Biological Sciences, 363, 3009–3021. 10.1098/rstb.2008.0064 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Matute, D. R. , & Cooper, B. S. (2021). Comparative studies on speciation: 30 years since Coyne and Orr. Evolution, 75, 764–778. 10.1111/evo.14181 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Moore, J. C. , & Pannell, J. R. (2011). Sexual selection in plants. Current Biology, 21, R176–R182. 10.1016/j.cub.2010.12.035 [DOI] [PubMed] [Google Scholar]
  34. Moyle, L. C. , Olson, M. S. , & Tiffin, P. (2004). Patterns of reproductive isolation in three Angiosperm genera. Evolution, 58, 1195–1208. 10.1111/j.0014-3820.2004.tb01700.x [DOI] [PubMed] [Google Scholar]
  35. Muller, H. J. (1942). Isolating mechanisms, evolution and temperature. Biology Symposium, 6, 71–125. [Google Scholar]
  36. Nordborg, M. , & Donnelly, P. (1997). The coalescent process with selfing. Genetics, 146, 1185–1195. 10.1093/genetics/146.3.1185 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Novikova, P. Y. , Hohmann, N. , Nizhynska, V. , Tsuchimatsu, T. , Ali, J. , Muir, G. , Guggisberg, A. , Paape, T. , Schmid, K. , Fedorenko, O. M. , Holm, S. , Säll, T. , Schlötterer, C. , Marhold, K. , Widmer, A. , Sese, J. , Shimizu, K. K. , Weigel, D. , Krämer, U. , … Nordborg, M. (2016). Sequencing of the genus Arabidopsis identifies a complex history of nonbifurcating speciation and abundant trans‐specific polymorphism. Nature Genetics, 48, 1077–1082. 10.1038/ng.3617 [DOI] [PubMed] [Google Scholar]
  38. Orr, H. A. (1995). The population genetics of speciation: The evolution of hybrid incompatibilities. Genetics, 139, 1805–1813. 10.1093/genetics/139.4.1805 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Pollak, E. (1987). On the theory of partially inbreeding finite populations. I. Partial selfing. Genetics, 117, 353–360. 10.1093/genetics/117.2.353 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Privé, F. , Luu, K. , Vilhjálmsson, B. J. , & Blum, M. G. B. (2020). Performing highly efficient genome scans for local adaptation with R package pcadapt version 4. Molecular Biology and Evolution, 37, 2153–2154. 10.1093/molbev/msaa053 [DOI] [PubMed] [Google Scholar]
  41. R Core Team . (2021). R: A language and environment for statistical computing. R Foundation for Statistical Computing. [Google Scholar]
  42. Ramsey, J. , Bradshaw, H. D., Jr. , & Schemske, D. W. (2003). Components of reproductive isolation between the monkeyflowers Mimulus lewisii and M. cardinalis (Phrymaceae). Evolution, 57, 1520–1534. 10.1111/j.0014-3820.2003.tb00360.x [DOI] [PubMed] [Google Scholar]
  43. Ravinet, M. , Faria, R. , Butlin, R. K. , Galindo, J. , Bierne, N. , Rafajlović, M. , Noor, M. A. F. , Mehlig, B. , & Westram, A. M. (2017). Interpreting the genomic landscape of speciation: A road map for finding barriers to gene flow. Journal of Evolutionary Biology, 30, 1450–1477. 10.1111/jeb.13047 [DOI] [PubMed] [Google Scholar]
  44. Rice, W. R. , & Holland, B. (1997). The enemies within: Intergenomic conflict, interlocus contest evolution (ICE), and the intraspecific Red Queen. Behavioral Ecology and Sociobiology, 41, 1–10. 10.1007/s002650050357 [DOI] [Google Scholar]
  45. Rieseberg, L. H. , & Willis, J. H. (2007). Plant speciation. Science, 317, 910–914. 10.1126/science.1137729 [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Rundle, H. D. , & Nosil, P. (2005). Ecological speciation. Ecology Letters, 8, 336–352. 10.1111/j.1461-0248.2004.00715.x [DOI] [Google Scholar]
  47. Schaefer, H. M. , & Ruxton, G. D. (2015). Signal diversity, sexual selection, and speciation. Annual Review of Ecology, Evolution, and Systematics, 46, 573–592. 10.1146/annurev-ecolsys-112414-054158 [DOI] [Google Scholar]
  48. Schmickl, R. , Jørgensen, M. H. , Brysting, A. K. , & Koch, M. A. (2010). The evolutionary history of the Arabidopsis lyrata complex: A hybrid in the amphi‐Beringian area closes a large distribution gap and builds up a genetic barrier. BMC Evolutionary Biology, 10, 98. 10.1186/1471-2148-10-98 [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Scopece, G. , Lexer, C. , Widmer, A. , & Cozzolino, S. (2010). Polymorphism of postmating reproductive isolation within plant species. Taxon, 59, 1367–1374. 10.1002/tax.595004 [DOI] [Google Scholar]
  50. Shirk, A. J. , Landguth, E. L. , & Cushman, S. A. (2017). A comparison of individual‐based genetic distance metrics for landscape genetics. Molecular Ecology Resources, 17, 1308–1317. 10.1111/1755-0998.12684 [DOI] [PubMed] [Google Scholar]
  51. Sobel, J. M. , & Chen, G. F. (2014). Unification of methods for estimating the strength of reproductive isolation. Evolution, 68, 1511–1522. 10.1111/evo.12362 [DOI] [PubMed] [Google Scholar]
  52. Sobel, J. M. , Chen, G. F. , Watt, L. R. , & Schemske, D. W. (2010). The biology of speciation. Evolution, 64, 295–315. 10.1111/j.1558-5646.2009.00877.x [DOI] [PubMed] [Google Scholar]
  53. Stankowski, S. , & Ravinet, M. (2021). Defining the speciation continuum. Evolution, 75, 1256–1273. 10.1111/evo.14215 [DOI] [PubMed] [Google Scholar]
  54. Swanson, R. , Edlund, A. F. , & Preuss, D. (2004). Species specificity in pollen‐pistil interactions. Annual Review of Genetics, 38, 793–818. 10.1146/annurev.genet.38.072902.092356 [DOI] [PubMed] [Google Scholar]
  55. Sweigart, A. L. , Mason, A. R. , & Willis, J. H. (2007). Natural variation for a hybrid incompatibility between two species of Mimulus . Evolution, 61, 141–151. 10.1111/j.1558-5646.2007.00011.x [DOI] [PubMed] [Google Scholar]
  56. The Marie Curie Speciation Network . (2012). What do we need to know about speciation? Trends in Ecology & Evolution, 27, 27–39. 10.1016/j.tree.2011.09.002 [DOI] [PubMed] [Google Scholar]
  57. Tiffin, P. , Olson, S. , & Moyle, L. C. (2001). Asymmetrical crossing barriers in angiosperms. Proceedings of the Royal Society of London ‐ Series B: Biological Sciences, 268, 861–867. 10.1098/rspb.2000.1578 [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Turelli, M. , & Moyle, L. C. (2007). Asymmetric postmating isolation: Darwin's corollary to Haldane's rule. Genetics, 176, 1059–1088. 10.1534/genetics.106.065979 [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Turissini, D. A. , McGirr, J. A. , Patel, S. S. , David, J. R. , & Matute, D. R. (2018). The rate of evolution of postmating‐prezygotic reproductive isolation in Drosophila . Molecular Biology and Evolution, 35, 312–334. 10.1093/molbev/msx271 [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Warwick, S. I. , Al‐Shehbaz, I. A. , & Sauder, C. A. (2006). Phylogenetic position of Arabis arenicola and generic limits of Aphragmus and Eutrema (Brassicaceae) based on sequences of nuclear ribosomal DNA. Canadian Journal of Botany, 84, 269–281. 10.1139/b05-161 [DOI] [Google Scholar]
  61. Widmer, A. , Lexer, C. , & Cozzolino, S. (2009). Evolution of reproductive isolation in plants. Heredity, 102, 31–38. 10.1038/hdy.2008.69 [DOI] [PubMed] [Google Scholar]
  62. Wilkins, J. F. , & Haig, D. (2003). What good is genomic imprinting: the function of parent‐specific gene expression. Nature Reviews Genetics, 4, 359–368. 10.1038/nrg1062 [DOI] [PubMed] [Google Scholar]
  63. Willi, Y. (2013). Mutational meltdown in selfing Arabidopsis lyrata . Evolution, 67, 806–815. 10.1111/j.1558-5646.2012.01818.x [DOI] [PubMed] [Google Scholar]
  64. Willi, Y. , Fracassetti, M. , Zoller, S. , & Van Buskirk, J. (2018). Accumulation of mutational load at the edges of a species range. Molecular Biology and Evolution, 35, 781–791. 10.1093/molbev/msy003 [DOI] [PubMed] [Google Scholar]
  65. Willis, C. G. , & Donohue, K. (2017). The evolution of intrinsic reproductive isolation in the genus Cakile (Brassicaceae). Journal of Evolutionary Biology, 30, 361–376. 10.1111/jeb.13011 [DOI] [PubMed] [Google Scholar]
  66. Wright, S. I. , Kalisz, S. , & Slotte, T. (2013). Evolutionary consequences of self‐fertilization in plants. Proceedings of the Royal Society B: Biological Sciences, 280, 20130133. 10.1098/rspb.2013.0133 [DOI] [PMC free article] [PubMed] [Google Scholar]
  67. Zuellig, M. P. , & Sweigart, A. L. (2018). A two‐locus hybrid incompatibility is widespread, polymorphic, and active in natural populations of Mimulus . Evolution, 72, 2394–2405. 10.1111/evo.13596 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Appendix S1: Supporting Information

Click here for additional data file. (333KB, docx)

Data Availability Statement

Genetic data for A. lyrata were published under Bioproject PRJEB19338 (Willi et al., 2018; https://www.ncbi.nlm.nih.gov/bioproject/PRJEB19338/); new data for A. arenicola is stored under Bioproject PRJNA885817 (https://www.ncbi.nlm.nih.gov/bioproject/PRJNA885817/). The intergenic SNP table, raw phenotypic data and reproductive isolation estimates are stored in Zenodo (https://doi.org/10.5281/zenodo.7125244).

Articles from Journal of Evolutionary Biology are provided here courtesy of Wiley

RESOURCES