Introduction
Speciation is often defined as a process in which one species splits into two. However, new species can also form as a result of hybridization between different species (Abbott, 1992; Rieseberg, 1997; Mallet, 2007; Soltis & Soltis, 2010). This occurs most commonly when hybridization is accompanied by whole genome duplication (i.e., allopolyploid speciation). The genome duplication provides partial reproductive isolation between the new allopolyploid lineage and its parental species, which eases establishment (Baack et al., 2015). Less commonly, a new hybrid lineage establishes and becomes reproductively independent without a change in ploidy level (i.e., homoploid hybrid speciation). The ecological and evolutionary conditions necessary for homoploid hybrid speciation are more restrictive than for allopolyploidy because of the lack of a universal mechanism for creating reproductive isolation between the hybrid lineages and its parental species (Buerkle et al., 2000, 2003; Schumer et al., 2015; Comeault, 2018; Blanckaert et al., 2018, 2023).
Allopolyploidy is also easier to document than homoploid hybrid speciation because hybridization and whole genome duplication are readily detected from cytological and genomic data. Thus, the prevalence of allopolyploidy is fairly well established (Wood et al., 2009; Mayrose et al., 2011; Barker et al., 2016). In contrast, the frequency of homoploid hybrid speciation remains unclear, mainly because hybridization can lead to outcomes other than speciation, including maladaptive, neutral, or adaptive introgression, genetic swamping, outbreeding depression, stable hybrid zones, reinforcement, etc., and it can be difficult to distinguish between these different outcomes (Abbott et al., 2013; Taylor & Larson, 2019; Bock et al., 2023). Also, these outcomes are not mutually exclusive. For example, while some introgressions may be maladaptive and contribute to outbreeding depression, others from the same donor species could enhance local adaption and/or reproductive isolation (Schumer et al., 2018).
In an attempt to reduce this uncertainty, Schumer et al. (2014) put forward three criteria for establishing homoploid hybrid speciation: (1) reproductive isolation from the parental species; (2) evidence of previous admixture; and (3) demonstration that reproductive barriers are derived via hybridization. While criteria one and two are relatively straightforward to demonstrate, showing that hybridization led to the formation of reproductive isolation is more challenging. Indeed, Schumer et al. (2014) argued that only one butterfly species (Maverez et al. 2006) and three sunflower species (Rieseberg et al., 2003, but see Owens et al., 2023) satisfied all three criteria. Others have worried that the criteria may be too stringent and that a focus on reproductive isolation might shift attention away from other important aspects of the process such as ecological divergence and the generation of evolutionary novelty (Nieto Feliner et al., 2017; Wang, Xu et al., 2022). Also, in some instances biological mechanisms of reproductive isolation may not be critical to homoploid hybrid speciation, such as in the Oxford ragwort (Senecio Squalidus ), where the homoploid hybrid is allopatric with its parental species (James & Abbott, 2005; Nieto Feliner et al., 2017). These concerns have led to the suggestion that homoploid hybrid species be classified as type 1 when reproductive isolation is a direct result of hybridization and type 2 when reproductive isolation is a by-product of other processes (Ottenburghs, 2018).
An important consequence of this debate has been an increased focus on the potential role of hybridization in the evolution of reproductive isolation between hybrid lineages and their parental species (Schumer et al., 2018; Ottenburghs, 2018; Brennan et al., 2019). Partly as a result, there now are more than a dozen cases of homoploid hybrid speciation in which the evolution of reproductive barriers has been directly linked to hybridization (Salazar et al., 2010; Hermansen et al., 2014; Lukhtanov et al., 2015; Leducq et al., 2016; Lamichhaney et al., 2018; Brennan et al., 2019; Masello et al., 2019; Sun et al., 2020; Wang et al., 2021; Stevison et al., 2022; Wang, Kang et al., 2022; Zou et al., 2022; Wu et al., 2023; Zhang et al., 2023). More importantly, genomics approaches have been developed that offer a straightforward means for showing that the reproductive barriers isolating new homoploid hybrid species from their parents are indeed derived from hybridization (Sun et al., 2020; Wang et al., 2021). In this comment, we discuss a particularly effective strategy for making such connections that was developed by Jianquan Liu and his colleagues (Figure 1), the application of this method to several plant and animal species, and what we have learned about homoploid hybrid speciation as a consequence (Wang et al., 2021; Wang, Kang et al., 2022; Zou et al., 2022; Wu et al., 2023; Zhang et al., 2023).