Understanding how evolutionary trends have changed through time is challenging, as it involves the interaction of several factors. The most important among these challenges are the ability to distinguish true evolutionary patterns (actual evolutionary events) from those influenced by sampling bias towards the present (i.e., only extant representatives of diverse clades) and the unequal extinction rates across different groups of plants.

The macroevolutionary study of chromosomal changes through evolutionary history utilizing comparative phylogenetic frameworks is no different. These studies focus on karyotypic changes that lead to changes in chromosome number, through poly- and dysploidy, as chromosome counts are widely available for many plants (see Figure 1 in Carta and Escudero, 2023; The Index to Plant Chromosome Numbers: Goldblatt and Johnson, 2011; Goldblatt and Lowry, 2011; The Chromosome Counts Database: Rice et al., 2015; Rice and Mayrose, 2023). In contrast, genomic or detailed karyotypic data needed for detecting other types of CRs, such as inversions or translocations, have only started to be generated in the last two to three decades.

The two major processes underpinning chromosome losses and gains (i.e. aneuploidy and dysploidy) have different impacts on lineage diversification. While aneuploidy leads to unstable lineages that persist for shorter periods over evolutionary time (under 1–5 million years; Yona et al., 2012), dysploidy has a less drastic impact, resulting in more stable lineages throughout time (Otto, 2007). Most, if not all, of the chromosome gains and losses that have been inferred in phylogenies of angiosperms are caused by dysploidy (Escudero et al., 2014; Carta et al., 2020). For instance, in the genus Carex L. (Cyperaceae), one of the most diverse genera of vascular plants in terms of species richness and chromosome number diversity, there is a tendency for dysploidy to be the main underlying mechanism responsible for chromosome evolution (Hipp et al., 2009).

Multiple studies have found that rates of polyploidy have increased in more recent evolutionary time in angiosperms (Carta et al., 2020; Escudero et al., 2014; see in Figure 2 data from Zhan et al., 2021). In the case of dysploidy, the rates are more constant and slightly accelerate towards the present (Escudero et al., 2014; see in Figure 2 ). Data suggest that there are no significant differences between descending and ascending dysploidy rates ( Figure 3 ). Here, we show the rate of dysploidy and polyploidy against time using the data from Zhan et al. (2021), who inferred rates of polyploidy and dysploidy using ChromEvol (Glick and Mayrose, 2014; Mayrose et al., 2010) from phylogenetic (Qian and Jin, 2016; Zanne et al., 2014) and chromosome count data (Rice et al., 2015) for 30,000 taxa representing 46 orders and 147 families of angiosperms. Their analyses show that both polyploidy and dysploidy increase exponentially with time, but the increase in rate is much greater for polyploidy ( Figure 2 ). The evolution of CRs through time is yet to be explored for gymnosperms and ferns, which could be especially interesting in the latter as they have an even greater chromosome number variation than angiosperms (Leitch and Leitch, 2012) and similar rates of paleopolyploidy (Li et al., 2024).

This acceleration towards the present in CRs is probably partially caused by a detection bias, as reconstructing genomes further back in time requires increasingly extensive genomic data to remain accurate. Chromosome rearrangement inference from chromosome counts is less reliable with increasing phylogenetic depth as homoplasies become more frequent, especially within groups with high karyological instability (Mayrose and Lysak, 2021; Weiss-Schneeweiss and Schneeweiss, 2013). Methods based on either genetic (gene duplicate analysis and paralogue divergence) or genomic data (synteny analysis), are more reliable at higher depths than those based on chromosome counts, even allowing detection of other CRs (Wang et al., 2021). As more genomic data become available for a wider range of taxa, studies of CR events in angiosperm history will become more accurate. One of the most recent studies addressing this issue was performed by McKibben et al. (2024). The authors studied synonymous divergences of paralogs (Ks) and orthologs, along with syntenic analysis on 462 genomes distributed throughout the angiosperm phylogenetic tree. Their analyses inferred multiple ancient WGD events and concluded that most extant species have experienced, on average, three WGD events, while also detected an increase in WGD towards the tips of the phylogenetic tree.

While a bias towards more recent events may contribute to the apparent acceleration in rates of CR, it does not account for the higher pattern of acceleration observed in polyploids compared to dysploids ( Figure 2 ). This bias would be expected to influence both types of CRs, and potentially even more strongly for polyploidy because it is generally easier to detect than dysploidy. Thus, the discrepancy in acceleration rates suggests additional factors are at play. One hypothesis shows that the detection of polyploidization in the past is more difficult because they stay “polyploid” for a shorter time due to the activity of diploidization processes leading to the rearrangement of multiple sets of parental chromosomes in their polyploid ancestors (Zhang et al., 2019). Considering the amount of WGD throughout the evolution of angiosperms (McKibben et al., 2024), extant species would be expected to have a much higher number of chromosomes and larger genomes than what is currently observed. This discrepancy underscores the impact of diploidization in shaping angiosperm genomes (Wang et al., 2021). Most species undergo a process of diploidization after WGD, which impacts at a genomic, epigenomic and proteomic level (Dodsworth et al., 2016; Wendel et al., 2016). This process often results in a reduction in genome size and a decrease in chromosome number through dysploidy, sometimes reaching a number of chromosomes equal, or even lower, than the original diploid number (Li et al., 2021; Mandáková and Lysak, 2018). Different factors have been proposed to explain the so-called “large genome constraint hypothesis”, which is primarily linked to genome size. These include limitations in phosphorus (P) and nitrogen (N) availability, constraints related to life cycle duration, narrow ecological tolerances, or small population sizes (Knight et al., 2005; Carta and Peruzzi, 2016; Guignard et al., 2016; Wang et al., 2021). Also, there is strong evidence that supports a lower diversification rate for neopolyploids in both ferns and angiosperms (Mayrose et al., 2010). Diploidization mechanisms remain to be studied in gymnosperms and ferns (Li et al., 2021) in order to understand how it impacts CR rates. A study in the hexaploid coast redwood (Sequoia sempervirens), one of the few polyploid species in conifers (see other examples in Farhat et al., 2019; 2022), showed very low diploidization rates (Scott et al., 2016). This finding may help to explain the evolutionary success of polyploid lineages and a limited chromosome number variation in gymnosperms. Ferns propose a more interesting, yet unexplored, study system for the impact of diploidization on CR rates. Unlike angiosperms, ferns undergo diploidization primarily through gene deletion and pseudogenization rather than gene loss (Haufler, 1987; Zheng et al., 2024), while still maintaining a high rate of diploidization (Haufler and Soltis, 1986; Haufler, 1987; Wolf et al., 1987).

Descending dysploidy is one of the most frequent routes of diploidization contributing to the reversion towards functional diploidy of polyploid angiosperms (Mandáková and Lysak, 2018). Therefore, descending dysploidy is expected to accelerate after WGD, while this acceleration is not present in species undergoing increasing dysploidy. The latter is often driven by chromosomal fragmentation but does not directly impact genome size significantly, unless accompanied by changes in DNA content. To differentiate the decreasing dysploidy associated with diploidization, we compared decreasing and increasing dysploidy rates to test if they were significantly different ( Figure 3 ). However, we did not observe a difference, suggesting that the decreasing dysploidy associated with subsequent diploidization is not as easily detected as ancient WGD ( Figure 2 ). This suggests that ancient WGD and the associated chromosome number reductions (decreasing dysploidy) during diploidization are difficult to detect, as polyploids may undergo rapid diploidization, which may limit their diversification potential in the polyploid state. Further support for this comes from models predicting 3–4 rounds of WGD in angiosperms, with high rates of diploidization alongside lower polyploid diversification rates compared to diploids (Barker et al., 2016; 2024). This instability is not observed following dysploidy, as its rate appears to remain more constant throughout angiosperm evolutionary time (Escudero et al., 2014; Figure 2 ). Such findings support the hypothesis that dysploidy, unlike polyploidy, is not as disadvantageous in generating long-term persisting lineages and does not entail significant changes in DNA content (Escudero et al., 2014). Overall, this highlights the importance of further exploring the impact of dysploidy in evolution, which has traditionally received less attention compared to polyploidy, despite its potential influence on chromosomal evolution and species diversification at both macro- and microevolutionary scales. Even less is known about the role of dysploidy in other plant lineages. Descending dysploidy associated with diploidization has been demonstrated to be lower in ferns than in angiosperms even for similar rates of paleoploidy (Haufler, 1987; Li et al., 2024; Zheng et al., 2024), leading to high chromosome numbers (Leitch and Leitch, 2012).

In conclusion, the increase in CR rates in more recent angiosperms is influenced by both biological processes and methodological biases. While sampling biases favor the detection of recent chromosomal changes, true evolutionary mechanisms, particularly polyploidy, also contribute to this tendency. Polyploidy rates increase more rapidly than dysploidy, where ancient WGD followed by diploidization are common. Dysploidy, while more constant over time, shows a slight acceleration toward the present. Models like ChromoSSE have been developed to reconstruct chromosomal evolution by integrating chromosomal changes, speciation, and extinction rates. These models offer valuable insights into how CRs, such as dysploidy and polyploidy, influence diversification across lineages. However, the detection of these patterns is greatly affected by the availability of data (phylogenetic and chromosome number data). Incorporating more comprehensive datasets in future studies, not only within angiosperms but expanding to the rest of plants, may provide a clearer understanding of the great differences between the main plant groups. Further studies should focus on chromosome evolution in ferns, in order to understand how the differences in diploidization mechanisms have shaped chromosome evolution in ferns and angiosperms, both groups with high polyploidy rates.

Chromosomal rearrangements have the potential to drive speciation by reducing gene flow between divergent populations (Berdan et al., 2024; Lucek et al., 2023). Specifically, chromosomal changes may occur at the speciation event, either by directly initiating a cladogenetic process or by reinforcing speciation that was already initiated or completed by other geographical or ecological drivers or other genomic mutations independent of CR. Allopolyploid speciation is an example of chromosomal changes and cladogenesis coinciding (see section related to allopolyploidy). Currently, two primary models explain how CRs may contribute to reproductive isolation and speciation (Rieseberg, 2001a): hybrid dysfunction and the suppression of recombination.

The classical models are based on hybrid dysfunction and hypothesize that hybrids resulting from crosses between two different chromosomal races have reduced fitness (Coyne and Orr, 2004). This reduced fitness can lead to strong selection against hybrids, primarily because newly arising CRs are often underdominant. While strongly underdominant rearrangements are unlikely to reach fixation, those with weaker underdominance may become fixed but usually create only shallow barriers to reproductive isolation, making them unlikely to drive speciation (Rieseberg, 2001a; Faria and Navarro, 2010). This model of chromosomal speciation suggests that chromosomal changes occur at, or just prior to a cladogenetic event, as such changes are expected to establish significant barriers to gene flow that may result in speciation. In plants, ploidy changes are generally expected to lead to hybrid dysfunction (Escudero et al., 2016b; 2024) because for example, a cross between a diploid and a tetraploid typically produces an inviable or sterile triploid (Köhler et al., 2010). However, there are also counter examples where gene flow occurs between different ploidy levels through at least partially fertile transitional cytotypes. This gene flow can facilitate heteroploid gene transfer, contribute to adaptation via adaptive introgression and even lead to the de novo formation of a new polyploid (Čertner et al., 2017; Kolář et al., 2017; Peskoller et al., 2021, reviewed by Brown et al., 2024; Bartolić et al., 2024). This might lead to a slowing down of the genomic separation of given cytotypes and hampering diversification and speciation as well. Besides, there are many cases of intraspecific variation in ploidy levels within the same species (e.g., Hieracium subgenus Pilosella, Suda et al., 2007; Elettaria cardamomum, Anjali et al., 2016; Phragmites australis, Wang et al., 2024), suggesting that auto-polyploidization does not necessarily drive rapid speciation. Regarding intraspecific genetic variation, ploidy level in populations may represent (i) different genetic groups (Balao et al., 2010), (ii) only a partial correspondence with the genetic clustering (Vanrell et al., 2024), or (iii) a complete mismatch indicating a lack of genetic differentiation among different ploidy levels within a species (Chumová et al., 2024; Kauai et al., 2024). Given the limited number of ancient polyploidization events inferred for plants, it seems that most of these frequent intraspecific polyploidy variations do not persist. Otherwise, we would observe many more ancient polyploidization events in extant species. This conclusion is congruent with recent polyploids showing often lower diversification rates than their diploid progenitors (Mayrose et al., 2010). This aligns with reports suggesting that polyploidization from a macroevolutionary viewpoint is an evolutionary “dead end” since polyploids exhibit higher rates of extinction than their diploid relatives (Arrigo and Barker, 2012; Mayrose et al., 2014; Shu et al., 2022). In contrast, there are also examples of autopolyploid cytotypes that have undergone speciation processes being ultimately recognized as independent species (Fernández et al., 2022). Other alternative views suggest that there is no significant association between shifts in diversification rates and ancient polyploidization (Landis et al., 2018).

The second line of theory emphasizes the role of CRs for recombination, whereby CRs become fixed through natural selection as they suppress recombination in locally advantageous groups of genes, known as supergenes (Ayala and Coluzzi, 2005). By acting as barrier loci, they create genomic regions that facilitate the maintenance of beneficial combinations of alleles within populations, even in the presence of gene flow and can ultimately promote reproductive isolation and speciation. The suppression-recombination model of chromosomal speciation predicts that chromosomal changes (inversions, dysploidy, deletions/insertions, duplications, and reciprocal translocations) that may affect recombination rates will occur before a cladogenetic event, where over time, locally adapted alleles accumulate, eventually resulting in a cladogenetic event. Under this model, we predict genome variation among populations of the same species, a variation that is indeed observed (Rieseberg, 2001a). Furthermore, the speciation process can occur concurrently with gene flow between different karyotypes, indicating that speciation is not an instantaneous process but often a rather gradual one influenced by ongoing evolutionary forces (Rieseberg, 2001a; Ravinet et al., 2017).

Both theoretical frameworks suggest that variation in CRs among populations facilitates reproductive isolation, thereby enhancing the potential for speciation (Lucek et al., 2023). In addition, these two models of chromosomal speciation are not mutually exclusive: some CRs may on one hand reduce gene flow between different karyotypes, resulting in partial hybrid dysfunction, and on the other hand suppress recombination. Together, these factors may eventually lead to cladogenesis. In this context, CRs that act as barriers to gene flow are predicted to occur either before the speciation process is complete or afterward, preventing interspecific gene flow, e.g. during secondary contact (Faria et al., 2011; Berdan et al., 2024). Finally, there are some CRs that are not or less important for speciation and are likely to become extinct over time (Lucek et al., 2023) or could be retained through a process similar to incomplete lineage sorting, where the origin of a CR predates the speciation event.

In a phylogenetic framework, the hybrid dysfunction type model is consistent with cladogenesis, whereas the suppression of recombination type model is more consistent with anagenesis (Lucek et al., 2022). However, only few phylogenetic approaches exist to model chromosomal evolution (Mayrose and Lysak, 2021). The joint modeling of chromosome evolution, speciation, and extinction is implemented in ChromoSSE (Freyman and Höhna, 2018). In this model, chromosomal changes may occur anagenetically (along the branches of the phylogeny) or cladogenetically (at the time of speciation). From the few examples for plants that implement this model (Freyman and Höhna, 2018; Valdés-Florido et al., 2023, 2024b; Tribble et al., 2025) a clear pattern emerges: the vast majority of chromosomal changes happen anagenetically, while only a small percentage occur cladogenetically. In summary, although chromosomal changes can occur around the speciation event, most changes occur between cladogenetic events. Additionally, it is thought that only a small percentage of chromosomal changes are able to survive through the filter of speciation, with most eventually being lost or becoming extinct.

Ultimately, chromosomal changes can indeed occur at the time of speciation, either initiating or reinforcing the process of cladogenesis (see more details in Box 2 ). In the hybrid dysfunction model, CRs such as polyploidy create reproductive barriers by reducing hybrid fitness. In contrast, the recombination suppression model proposes that chromosomal changes, like inversions or translocations, reduce recombination allowing locally adapted alleles to accumulate potentially driving speciation. In most cases, chromosomal changes do not persist long-term, as only a small percentage survive the speciation process, with most becoming extinct. Alternatively, CRs can also exist without affecting phenotypes or physiological functions, remaining neutral and simply persisting. Therefore, while CRs can contribute to speciation, the majority of these changes occur anagenetically, with relatively few occurring directly at the moment of cladogenesis.

Changes in physical traits, in the chromosomes, and speciation can occur simultaneously in plants, but their interplay is complex and influenced by multiple factors. Intraspecific phenotypic and allelic changes are driven by natural selection, genetic drift or adaptation to local biotic and abiotic factors. Duplications, inversions, or translocations can result in gene expression changes that lead to significant phenotypic effects (Wray et al., 2003). As a consequence, these rearrangements can also lead to new linkage relationships or the formation of new genes. It is important to highlight that few breakpoints have been characterized for inversions with clear phenotypic effects (Hoffmann and Rieseberg, 2008; Guan et al., 2021; Chen et al., 2024). Previous studies have demonstrated that chromosomal inversions have putatively evolved as a response to environmental conditions because they were associated with morphological traits and showed increased fitness in adapted environments (Lowry and Willis, 2010; Lee et al., 2016). For instance, strong karyotype differences between closely related Mediterranean orchid species that also share pollinators have shown that CRs play an important role in reducing hybrid fitness and maintaining reproductive isolation (Cozzolino and Scopece, 2008). Otherwise, it is widely known that chromosomal deletions can have significant phenotypic consequences, since dominant alleles can be deleted, exposing recessive alleles in heterozygosity (Huettel et al., 2008). The phenotypic effects derived from other SVs have been less studied in plants. However, comparative genomic mapping has begun to facilitate their identification and annotation techniques have allowed the identification of possible SVs (including gene presence/absence and copy number variations) responsible for phenotypic traits (Huang and Rieseberg, 2020; Zhang et al., 2020). These kinds of studies have mainly focused on crops, identifying how SV impacts in genes with agronomic value (Yuan et al., 2021). Since crop gene pools are often derived from multiple species, sequencing and assembly efforts are put into all the species within the genus of interest. This has led to the development of super-pangenomes, which enable the detection of conserved and diverged genomic regions, as well as their frequencies within populations (Zhao et al., 2020). The relevance of pangenomics has grown significantly with the availability of high-quality genomes assemblies from multiple cultivars, especially in agriculturally important crops (Zhao et al., 2018).

Otherwise, WGD are common in plants, with extensive impacts on gene expression, cellular function, and organism phenotype. Polyploids can display differences in floral traits (Balao et al., 2011; McCarthy et al., 2015), chemical scents (Vereecken et al., 2010; Jersáková et al., 2010), and flowering phenology (Schranz and Osborn, 2000; Pegoraro et al., 2019). Such phenotypic changes evolved immediately after polyploidization, and it may have served to establish and stabilize novel cytotypes (Oswald and Nuismer, 2011; Clo and Kolář, 2021). Thus, polyploidy often results in reproductive isolation, leading to rapid speciation because new polyploid individuals may not be able to reproduce with their diploid progenitors. These new species frequently exhibit novel phenotypic traits as a consequence of changes in gene expression caused by the increased chromosome number (Chen, 2007; Balao et al., 2011; Basit and Lim, 2024). Phenotypic and morphological changes known to be induced by polyploidy are those related to variation in flower number and flowering time (Schranz and Osborn, 2000), plant structure, or alterations in plant physiology under stress tolerance (Cohen et al., 2013; Van de Peer et al., 2021; Turcotte et al., 2024). Polyploidy may for instance contribute to higher tolerance to nutrient-poor soils and resistance to stressful environments such as drought, cold or pathogens (Levin, 2002; Sader et al., 2019). In fact, a common phenomenon in polyploid species is the “gigas effect”, which results in increased cell sizes and overall plant features in comparison with their diploid parents (Stebbins, 1971; Soltis et al., 2014). In addition, CRs can generate genetic diversity through evolutionary changes, since bursts of lineage splitting in plants often result in adaptive radiation (Parent et al., 2020). In these cases, the accumulation of genomic changes leads to rapid phenotypic evolution promoted by genetic variation and accelerated evolution, particularly under changing environmental conditions. These phenotypic changes are not gradual but instead occur in bursts, often linked to speciation, which frequently occurs simultaneously. Consequently, long periods of evolutionary stasis could be interrupted by short and rapid bursts of evolutionary change linked to chromosomal events (Stebbins, 1971; Levin, 2002; Lysak et al., 2006). However, although most of intraspecific polyploidy variations do not persist over time (see above section), some studies have revealed that species with recent polyploid origins may undergo rapid speciation and significant phenotypic divergence (Fehrer et al., 2022). Floral evolution in the genus Calochortus Pursh (Liliaceae) represents a case of radiation, where selection for adaptation to diverse local habitats drives the specialization of flowers to various pollinators. This contrasts with adaptive radiation, which typically involves selection for specific pollinators within a single habitat (Patterson and Givnish, 2004). The ability to reproduce can also be directly linked to polyploidization, and typical patterns of cytotype distribution have been found in different studies (Krak et al., 2013; Valdés-Florido et al., 2024a). On the other hand, in isolated environments like Hawaiian Islands, plant species often display bursts of phenotypic diversity and chromosomal evolution as a consequence of rapid adaptation resulting in speciation events (Barrier et al., 1999; Bellinger et al., 2022). Thus, polyploidization is a process that assists speciation and diversification into new areas, being able to entail evolution of reproductive strategies. However, sometimes polyploidization could be related to the loss of a trait, such as it occurs with the loss of heterostyly for the two major families that present this trait (Primulaceae and Rubiaceae; Guggisberg et al., 2006; Naiki, 2012).

Aneuploidy has a very drastic impact on genetic dosage (Birchler et al., 2007; Birchler and Veitia, 2007) and is most commonly deleterious, while dysploidy is much more widespread with a high impact on plant evolution (Escudero et al., 2012). While dysploidy does not involve significant changes in DNA content, it can also have an impact on phenotype through structural rearrangements. At macroevolutionary scales, some studies highlight that gain or loss in the number of chromosomes can be associated with novel morphological features and influence diversification processes (Farminhão et al., 2021; Sader et al., 2019). For instance, recent research has shown that ascending dysploidy together with genome size expansion correlates both with larger flowers and higher diversification rates in the subgenus Passiflora L., suggesting a positive selection towards bigger genome sizes through morphological/ecological changes (Sader et al., 2019). The recurrent and parallel evolution of the same dysploid cytotype in the genus Soldanella L. has consistently resulted in speciation events (Slovák et al., 2023; Rurik et al., 2024). Similarly, Farminhão et al. (2021) found a correlation between dysploidy events and the evolution of leaflessness in the Dendrophylax-Microcoelia clade of angraecoids (Orchidaceae) with an eventful karyotypic history dominated by descending dysploidy, although the underlying mechanisms remain unexplored. No increases in net diversification rates could be related to chromosome number changes with the predominance of karyotypic stasis. However, species experiencing shifts in chromosome number appear to show parallel evolution of some phenotypic structures, leaflessness, and changes in floral color (Farminhão et al., 2021).

In summary, bursts of phenotypic and chromosomal evolution can occur simultaneously with speciation, but their relationship is complex due to the timing and interplay between these processes being highly dependent on evolutionary forces and environmental factors (see Box 3 for a case study in the genus Linum). On the one hand, CRs, such as inversions or duplications, can lead to changes in traits like morphology, fitness, or reproductive isolation. On the other hand, polyploidy often leads to rapid phenotypic evolution and speciation due to changes in chromosome number, resulting in traits such as altered flowering times or increased environmental tolerance. Adaptive radiation often triggers rapid speciation and significant phenotypic divergence, but the persistence of these changes can vary since some chromosomal changes may become extinct over time, while others promote long-term diversification. Additionally, shifts in chromosome number, such as dysploidy or aneuploidy, can also contribute to phenotypic diversity and speciation, although their effects vary depending on the context.

One of the best-known drivers of speciation is geographic isolation, where populations of the same species become separated as a consequence of space in the presence or absence of geographical and ecological barriers (Dobzhansky, 1951; Coyne and Orr, 2004). Geographical isolation can provide the conditions for chromosome changes to accumulate either through drift or selection and without the diluting effects of gene flow from other populations. These chromosomal changes further enhance population differentiation in morphology, ecology, pre-and/or post-zygotic barriers, cumulatively or individually giving rise to reproductive isolation and eventually speciation (Levin, 2002; Scopece et al., 2010). This indicates that intrinsic postzygotic mechanisms may trigger polymorphism among allopatric conspecific plant populations. In contrast, a study found no links between chromosome transitions and the major diversification events associated with ecological events in the temperate grasses (Pooideae) (Pimentel et al., 2017). However, we should consider confounding factors in such macroevolutionary studies, such as the effect of undetected polyploidization followed by diploidization processes or genome sampling bias due to the use of few genetic markers, which does not negate the existence of chromosomal changes. In addition, chromosome evolution and cladogenetic processes were not modeled together which may lead to biased results if chromosomal changes are, in fact, affecting cladogenesis.

Changes in the environment (e.g., altitude, temperature) can also trigger adaptive responses through chromosome evolution (chromosomal inversion in adaptation; Huang and Rieseberg, 2020). Moreover, the occurrence of polyploidy in the tree of life also seems to correlate with periods of environmental change (Van de Peer et al., 2017). For instance, polyploidy can cause variation in plant functional traits and generate individuals that can adapt and exploit new environmental niches (Wan et al., 2020) and can facilitate adaptive response to harsh environmental conditions (Alix et al., 2017). Specifically, environmental stress has been proposed to foster the production of unreduced gametes, which are the main drivers of polyploidization in angiosperms (Bretagnolle and Thompson, 1995; Levin, 2002). The formation of diploid pollen grains has been promoted by low temperatures in the genera Solanum L. (Solanaceae), Datura L. (Solanaceae), Oenothera L. (Onagraceae), or Epilobium L. (Onagraceae) (e.g., McHale, 1983; Alsamir et al., 2021; Krakos et al., 2022). However, not only do low temperatures enhance the production of unreduced gametes, high temperature environments also have the potential to increase ploidy levels as seen in the genera Rosa L. (Rosaceae) (Pécrix et al., 2011; Crespel et al., 2015) and Populus L. (Salicaceae) (Wang et al., 2017). Other environmental factors, such as temperature fluctuations, low nutrient stress, or the presence of parasites and viruses have similarly been reported to promote the formation of unreduced gametes (Levin, 2002). These strategies are consistent with a broader adaptability and ecological tolerance and higher invasive potential of polyploids than their diploid relatives (Pandit et al., 2011; Te Beest et al., 2012; Marks et al., 2024).

Geographical and environmental pressures (or only one of them) may also occur simultaneously, driving chromosomal changes that indirectly promote speciation (Coyne and Orr, 2004). In addition, chromosome evolution has been strongly linked to biogeography in angiosperms (Rice et al., 2015), with polyploidization showing significant evolutionary implications, including the potential for range expansion (Te Beest et al., 2012; Soltis et al., 2015). For instance, in the genus Panax L. (Araliaceae) it has been demonstrated that the ancient and recent WGDs along with geographical and ecological isolations might have together contributed to the diversification of this genus, suggesting that distinct selection pressures appear to have acted during the genus’ evolutionary history (Shi et al., 2015). In the genus Centaurium (L.) Hill polyploid species may have an optimal climatic niche related to harsher environments (Valdés-Florido et al., 2024b; see Box 4 ). However, this cytotype adaptation is not always linked to speciation, as cytotypes can coexist within a single species as seen in the case of cryptic invasion of polyploid Centaurea stoebe L. expanding into the range of its diploid relative in Europe (Rosche et al., 2025).

Taking into consideration the above, geographic and environmental changes can trigger chromosome evolution, which can either coincide or accumulate post exposure together putatively driven by the severity of the changes experienced by plants. Geographic isolation is a key driver of this process, as it allows for chromosomal changes to undergo fixation in populations without gene flow. These chromosomal changes, including inversions and polyploidy, can contribute to further differentiation in traits, such as morphology and ecology (stress factors like temperature fluctuations or low nutrients availability), fostering reproductive isolation and speciation. In some cases, both geographic isolation and environmental pressures work together, while in other instances, environmental factors alone can influence chromosome evolution and diversification.

Ecological interactions describe the diverse mechanisms through which organisms influence each other’s survival, reproduction success, and distribution within ecosystems. These interactions can occur between individuals of the same species (intraspecific) or between different species (interspecific) (Schoener, 1990), and play an important role in evolution, as they can act as selective forces during speciation (Thompson, 2009). From a microevolutionary perspective, empirical evidence supports the idea that ecological interactions have acted as selective forces on certain CRs (Burak et al., 2018). However, from a macroevolutionary point of view, there are not many cases where such interactions have left detectable signature in the speciation patterns among plant groups.

Among the aforementioned CRs, polyploidy yet again, is the most widely studied in the context of ecological interactions (Thompson et al., 2004; Segraves and Anneberg, 2016). Polyploidization has the potential to create genetically isolated entities with divergent genetic and phenotypic traits that can shape the interactions of plants with other organisms, and likewise, these organisms can act as selective forces in stabilizing new polyploid races (Segraves and Anneberg, 2016). For species that rely on animal-mediated pollination, pollinators can contribute to the reproductive isolation of polyploids from their diploid relatives through assortative mating (Rezende et al., 2020). With the exception of the genus Nicotiana L. (Solanaceae), where pollinators have significantly influenced macroevolutionary patterns of speciation through floral color selection (McCarthy et al., 2015), clear examples of pollinators shaping diversification patterns in polyploid lineages are rare. In fact, the results of a current study, which examined whether neo-polyploidization in Arabidopsis arenosa (L.) Lawalrée led to changes in flower size that might influence pollinator behavior, found no evidence for assortative mating due to polyploidization. Instead, it was observed that polyploidization facilitated pollen exchange between different ploidy levels (Schmickl et al., 2024). Additionally, polyploids may evolve new defense mechanisms against herbivores, such as the production of new secondary chemical compounds (Orians, 2000; Edger et al., 2015; see more details in Box 5 ) or an expanded host range (Nuismer and Thompson, 2001; Arvanitis et al., 2010), which could also influence patterns of speciation in polyploid lineages and their diploid relatives. One of the most compelling examples of how key innovations responses to herbivory can shape macroevolutionary speciation patterns in plants through gene and WGD as described by Edger et al. (2015). In this scenario, the evolutionary arms race between members of the order Brassicales and pierid butterflies has played a significant role in driving the diversification rates of both groups ( Figure 9 ). The evolutionary interplay between the two organismal groups has led to co-evolutionary dynamics, where plants evolve new chemical defenses against herbivory, while the butterflies develop mechanisms to overcome these defenses. Changes in the secondary chemistry of polyploids can also influence their interactions with other organisms, offering potential protection from parasites and pathogens (Burdon and Marshall, 1981; Vleugels et al., 2013) or disrupting relationships with mutualistic fungi, possibly resulting in fewer associations (Gundel et al., 2014; Franco et al., 2015). However, the underlying mechanisms remain complex and no documented cases have demonstrated a lasting impact on the macroevolutionary patterns of speciation in polyploid races. Lastly, an increase in genetic variability provided by polyploidization can enhance the environmental tolerance of polyploids, making them more competitive and prone to invasions (Pandit et al., 2011; Thébault et al., 2011; Te Beest et al., 2012; Cheng et al., 2021; Moura et al., 2021). This phenomenon is exemplified by certain species of the genus Spartina Schreb. (Poaceae), where meso- and neo-polyploid events have enabled them to be more competitive in stressful habitats by introducing novel regulatory patterns in gene expression (Giraud et al., 2021).

However, not only polyploidy but also dysploidy events may also have an impact in ecological interactions. A higher chromosome number may lead to higher recombination rates (Nijalingappa, 1974; Bell, 1982; Nokkala et al., 2004; Escudero et al., 2012). Higher recombination rates provide more evolutionary potential, which is advantageous in highly competitive communities with temporarily predictable environments (Koella, 1993; Escudero et al., 2013a; Escudero and Hipp, 2013). Holocentric chromosomes distribute centromeric activity along their length, unlike the single centromere of monocentric chromosomes (Márquez-Corro et al., 2018). This structure enables fragmented chromosomes to retain kinetochore function, reducing segregation errors during meiosis and potentially facilitating chromosomal fissions and fusions without compromising genome stability. In a highly competitive environment, this genetic flexibility can offer an advantage. Conversely, low rates of recombination should be positively selected in unstable and low competition communities where a pioneering strategy could be successful (Stebbins, 1958; Grant, 1958; Bell, 1982; Koella, 1993). Accordingly, in high competition environments, higher numbers of chromosomes are expected in holocentric species. This hypothesis has been tested at the microevolutionary level within Carex laevigata group populations (sect. Spirostachyae, Escudero et al., 2013a). The results indicated that chromosome numbers are indeed higher in lowland ancestral areas where competition is more intense. However, when the same hypothesis was extended to the whole genus Carex, the relationship between chromosome number and competition was less clear (Bell, 1982; Escudero et al., 2012). This indicates that other evolutionary or ecological factors could have shaped chromosome number evolution at a broader scale within this group.

Chromosomal rearrangements that do not involve changes in chromosome number can also play a role in adaptation by reducing recombination between favorable combinations of alleles (Kirkpatrick and Barton, 2006; Lowry and Willis, 2010). This reduced recombination contributes to speciation in a similar way - by suppressing recombination between local adapted alleles and those causing assortative mating (Trickett and Butlin, 1994). One of the examples is the differentiation of monkeyflowers (Erythranthe guttata G.L.Nesom, Phrymaceae) populations in two ecotypes, where a large inversion has been shown to affect the growth form (Lowry and Willis, 2010), herbivore resistance through secondary compounds synthesis (Kollar et al., 2024), as well as also causing assortative mating by allochrony in flowering time (Lowry and Willis, 2010). At a broader evolutionary scale, in this same genus, another inversion has been associated with differences in corolla length and flower color contributing to both prezygotic and postzygotic isolation of a sister species pair: Erythranthe lewisii (Pursh) G.L.Nesom and N.S.Fraga and Erythranthe cardinalis (Douglas ex Benth.) Spach (Fishman et al., 2013). Besides, Johnson et al. (2009) described SVs as having an effect on ecological interactions in plants at the macroevolutionary level. Under this scenario, some species from the Onagraceae family display Permanent Translocation Heterozygosity (PTH), a characteristic that prevents pairing of homologous chromosomes in meiosis and thus recombination. Moreover, most of these species are self-fertilizing, resulting in offspring that are genetically identical to the parent. The study found that species with PTH are more susceptible to generalist herbivores, suggesting that these may have reduced defenses against herbivory, likely due to their limited genetic variability.

Overall, ecological interactions leave a detectable signal on chromosome macroevolution, although the relationship is complex and not always straightforward. Evidence shows that ecological interactions, such as pollinator preferences, herbivory, and competition, can act as selective forces on CRs and influence speciation patterns in a microevolutionary and macroevolutionary scale.

Hybridization serves as a critical evolutionary mechanism that plays a significant role in shaping diversification and speciation processes in plants (Soltis and Soltis, 2009; Taylor and Larson, 2019). It can play a creative role in species evolution, having both positive and negative effects, contributing to species evolution by triggering hybrid speciation (Abbott et al., 2013; Buerkle et al., 2000; Taylor and Larson, 2019), facilitating adaptive introgression (Suarez-Gonzalez et al., 2018), and even fueling adaptive radiations (Barrier et al., 1999; Mandáková et al., 2010b; Stankowski and Streisfeld, 2015). However, hybridization can also result in negative consequences, such as complete hybrid sterility, extensive introgression that merges previously separated gene pools, thereby hindering speciation and potentially leading to the extinction of parental taxa (Kearns et al., 2018; Slovák et al., 2023; Todesco et al., 2016). From a karyotypic perspective, the effects of hybridization depend on the compatibility of parental genomes, with ploidy level being a key factor. Hybridization can occur without changes in chromosome number (homoploid hybridization) or may involve WGD (allopolyploidization) (Ramsey and Schemske, 1998; Nieto Feliner et al., 2020; see a case study in Box 6 for details on both mechanisms).

Allopolyploidization promotes speciation by integrating distinct parental genomes and establishing thus immediate reproductive barriers (Ramsey and Schemske, 1998; Pikaard, 2001; Mallet, 2007; Qiu et al., 2020). A major evolutionary advantage of allopolyploids is their increased heterozygosity, which confers hybrid vigor while overcoming some challenges of homoploid hybridization such as (sub)genomic incompatibility (Comai, 2005; Qiu et al., 2020). This can drive phenotypic changes and adaptations, potentially leading to further lineage diversification (Tayalé and Parisod, 2013; Mutti et al., 2017). However, allopolyploidization also presents remarkable challenges that can affect the viability and establishment of new allopolyploids, with potential negative implications for speciation. Newly formed allopolyploids face extrinsic challenges, such as population bottlenecks (Novikova et al., 2017) and competition with parental species (Fowler and Levin, 1984), as well as intrinsic challenges, including complications in chromosome segregation (Bomblies et al., 2016) and changes in the genome structure (Leitch and Leitch, 2008). As WGD significantly reduces or eliminates homeologous recombination in the hybrid genome, potential incompatibilities between divergent parental subgenomes cannot be effectively purged, which may pose significant challenges for newly formed allopolyploids (Rieseberg, 2001b; Qiu et al., 2020). As a result, the genomes of newly evolved allopolyploids must undergo continuous alterations at the genetic, epigenetic, transcriptomic, and proteomic levels during the early stages of establishment to achieve genomic stabilization (Chelaifa et al., 2010; Chester et al., 2012; Edger et al., 2017; Qiu et al., 2020). These genomic changes are typically absent in older allopolyploid lineages (Burns et al., 2021), indicating that structural and expression plasticity of genomes in newly formed allopolyploids is crucial for their stabilization and integration, with the timing of these changes varying among species (Wendel et al., 2018; Burns et al., 2021).

The association between allopolyploidy and diversification processes is frequently linked to biological traits; however, even greater attention has been given to adaptations within ecological and biogeographical contexts. One key aspect of establishment and subsequent evolution of allopolyploids is their niche breadth and shifts relative to their diploid progenitors. In contrast to previous assumptions (Stebbins, 1984), recent comparative studies have shown that ecological niche shifts in allopolyploids, relative to their diploid progenitors, are highly variable, exhibiting patterns of expansion, contraction, intermediacy, and novelty (Blaine Marchant et al., 2016; Parisod and Broennimann, 2016). Mata et al. (2023) further found no consistent differences in the distribution ranges or habitat types of allopolyploids in relation to extreme conditions; allopolyploid species do not necessarily occupy more extreme environments or broader geographic ranges compared to their diploid progenitors. The significant overlap observed between the niches and distribution ranges of allopolyploids and their progenitors suggests that these niches are largely shaped by the climatic and geographical characteristics of the parental species. However, biotic and microclimatic factors likely play a significant role in the establishment of allopolyploids (Blaine Marchant et al., 2016; Griffiths et al., 2019; Akiyama et al., 2020; Mata et al., 2023). Reevaluating the classical view that allopolyploids predominantly thrive in deglaciated temperate habitats (Stebbins, 1984) reveals contradictions considering recent findings. Indeed, an increasing number of studies emphasize the evolution of allopolyploids in the Mediterranean region, where they often exhibit restricted distribution ranges and specialized ecological niches (López-González et al., 2021; Šlenker et al., 2021; Kantor et al., 2023). It seems that the ecological dynamics of allopolyploids may be more complex than previously recognized, as their successful establishment in these environments suggests adaptive strategies that enable them to thrive despite often limited geographical distributions.

The critical question of whether allopolyploidy facilitates species diversification, especially at the macroevolutionary level through mechanisms such as species radiation, remains inadequately understood. Nevertheless, studies on allopolyploids in the tribe Microlepidieae (Brassicaceae) Al‑Shehbaz and approximately 50 taxa of Nicotiana sect. Suaveolentes Goodsp. (Solanaceae) suggest that all taxa in both groups are derived from a single allopolyploid ancestor, from which they diversified and radiated across the Southern Hemisphere (Clarkson et al., 2004; Kelly et al., 2013; Mandáková et al., 2017; Chase et al., 2023). In both cases, diploidization has led to the evolution of diverse dysploid lineages, facilitating further diversification processes (Mandáková et al., 2017; Chase et al., 2023). Similarly, a study by Tomlin et al. (2024) uncovered the allopolyploid origin of monophyletic Hawaiian mints (Lamiaceae), which are derived from North American ancestors of the genus Stachys L While the authors did not directly test the impact of allopolyploidy on the diversification and radiation of Hawaiian mints, they proposed that this allopolyploid ancestry could provide a genomic substrate for morphological differentiation within the lineage and potentially foster evolutionary radiation in the rapidly evolving Hawaiian landscape (Tomlin et al., 2024). In contrast, Estep et al. (2014) found that while one-third of species in the grass tribe Andropogoneae Dumort. (Poaceae) are allopolyploids, diversification did not precede the allopolyploidization event and does not correlate with subsequent speciation bursts. The emergence of Andropogoneae species in the Late Miocene coincides with the expansion of major C4 grasslands, and although allopolyploidy remains a significant mode of speciation within this tribe, its role in diversification is less clear. Furthermore, additional studies indicate that allopolyploidy may catalyze diversification and even radiation, although these hypotheses still need to be tested (Barrier et al., 1999; Julca et al., 2018).

In conclusion, allopolyploidy plays a significant role in rapid speciation by generating new species that exhibit hybrid vigor and enhanced ecological potential (Mallet, 2007). While it facilitates evolutionary diversification, the relationship between chromosomal evolution and ecological adaptation is variable and influenced by the ecological requirements of the progenitors. Although some studies suggest that allopolyploidy can trigger lineage diversification or rapid radiations, further comprehensive empirical investigations are needed to support these findings and allow for broader generalizations.

Homoploid hybrid speciation (HHS), in contrast to allopolyploid speciation, occurs without WGD, resulting in hybrid species that retain the same chromosome number as their parental species (Stebbins, 1958; Mallet, 2007; Soltis and Soltis, 2009; Abbott et al., 2010; Schumer et al., 2014). Unlike allopolyploidy, HHS does not immediately generate reproductive barriers. Instead, hybrid sterility, which restricts gene flow between hybrids and progenitors, must evolve through genetic incompatibilities (genic sterility) or CRs (chromosomal sterility) (Stebbins, 1958; Abbott et al., 2010; Yakimowski and Rieseberg, 2014). In the absence of reproductive barriers, homoploid hybrids are susceptible to backcrossing with their parental species, which can lead to introgression, the formation of hybrid zones, reinforcement, or even genetic assimilation (Soltis and Soltis, 2009; Todesco et al., 2016; Aguillon et al., 2022). An alternative mechanism for the stabilization and establishment of homoploid hybrid species involves extrinsic factors, such as spatial isolation from parental taxa, often accompanied by ecological divergence (Grant, 1981; Gross and Rieseberg, 2005; Abbott et al., 2010; Yakimowski and Rieseberg, 2014). Chromosomal rearrangements are widely recognized as a key factor for establishing intrinsic reproductive barriers between homoploid hybrids and their parental species, a phenomenon explained by the recombinational model of HHS (Stebbins, 1957; Grant, 1958, 1981; Buerkle et al., 2000). According to this model, parental species exhibit at least two independent CRs, resulting in reduced gamete viability in F1 hybrids due to heterozygosity. Subsequently, homozygous recombinants in the F2 generation may restore self-fertility while remaining incompatible with the parental species, potentially enabling sympatric speciation. This recombinational model demonstrates how hybrid lineages can achieve reproductive isolation from their parental species in sympatry, suggesting it as a likely pathway for initiating HHS. However, extrinsic reproductive barriers, such as ecological divergence between hybrids and parental species, may further influence the success of HHS (Buerkle et al., 2000).

Demonstrating HHS driven by chromosomal or genetic mechanisms, however, is not straightforward and requires a comprehensive investigation that integrates (cyto)genomic, ecological, and experimental approaches (Seehausen, 2004; Mallet, 2007; Schumer et al., 2014). This evidence includes reproductive isolation from parental species, documentation of past hybridization events, and confirmation that isolating mechanisms have emerged as a result of hybridization (Schumer et al., 2014). Therefore, evidence of HHS driven by CRs has only been documented in a limited number of plant systems (Rieseberg et al., 1995; Archibald et al., 2005; Wu and Tanksley, 2010; Yakimowski and Rieseberg, 2014; Ostevik et al., 2020). Nonetheless, the genus Helianthus L. (sunflowers) serves as the most iconic model system for understanding HHS, where CRs play a crucial role in establishing chromosomal sterility and driving speciation (Rieseberg et al., 1995; Ostevik et al., 2020; Todesco et al., 2020). The dynamic chromosomal evolution in sunflowers, driven by rearrangements, has facilitated rapid diversification, primarily through intrachromosomal inversions and interchromosomal translocations (Rieseberg et al., 1995; Ostevik et al., 2020). Both types of rearrangements are common in plant genome and karyotype evolution (Weiss-Schneeweiss and Schneeweiss, 2013), although intrachromosomal rearrangements tend to occur more frequently than interchromosomal ones in plant evolution (Wu and Tanksley, 2010; Ostevik et al., 2020). In addition, CRs that induce hybrid sterility appear to be strongly linked to an annual life strategy. These rearrangements are more frequent in annuals, which undergo more meiotic events per generation, thereby accelerating chromosomal mutation rates (Archibald et al., 2005; Owens and Rieseberg, 2014; Yakimowski and Rieseberg, 2014). While there is growing evidence supporting interspecific hybridization prior to adaptive radiations in both plants and animals (Seehausen, 2004; Meier et al., 2017; Stankowski and Streisfeld, 2015; Svardal et al., 2020; Skopalíková et al., 2023), documented cases of ancient homoploid hybridization preceding lineage diversification and radiation in plants remain scarce (Pease et al., 2016; Liu et al., 2017). Nevertheless, HHS occurs through genetic or chromosomal sterility. If the latter is involved, the extent to which CRs contribute to the homoploid speciation in the ancestors of these radiations remains unclear. In conclusion, the role of CRs in establishing reproductive barriers against closely related or more distantly related species during HHS-induced diversifications, and particularly radiation, remains largely unexplored. We propose that investigating the role of HHS and particularly the influence of CRs as a mechanism of speciation in plant diversification, presents significant opportunities for future research.