Replicate hybrid zones suggest a limited role of plumage in reproductive isolation among subspecies of the Variable Seedeater (Sporophila corvina)
Running title: Limited role of plumage in hybridization
Diego Ocampo1,2,3*, Kevin Winker4, Matthew J. Miller3,4,5, Luis Sandoval6, J. Albert C. Uy2
  1. Department of Biology, University of Miami, Coral Gables FL 33146, USA
  2. Department of Biology, University of Rochester, Rochester, NY 14627, USA
  3. Smithsonian Tropical Research Institute, 0843-03092 Balboa, Ancon, Panama
  1. University of Alaska Museum, 907 Yukon Drive, Fairbanks, AK 99775, USA
  2. Reneco International Wildlife Consultants, Abu Dhabi, UAE
  3. Laboratorio de Ecología Urbana y Comunicación Animal, Escuela de Biología, Universidad de Costa Rica, 11502 San José, Costa Rica.
* Corresponding author: ocampov.diego@gmail.com
KEYWORDS: genetic divergence, hybridization, plumage coloration, replicated contact zones, secondary contact.
Contact zones between hybridizing taxa offer a window into the speciation process (Barton & Hewitt, 1985; Harrison, 1990; Price, 2008). For instance, contact zones shed light on which traits are important for assortative mating and reproductive isolation, as well as the genomic basis of such traits (Barton & Gale, 1993; Nadeau et al., 2014; Semenov et al., 2017). Hybridization between differentiated taxa may have different outcomes (Abbott et al., 2013), such as the merger of distinct lineages (Todesco et al., 2016), or even the rise of a new and independent lineage through homoploid hybrid speciation (Mavárez & Linares, 2008). However, the degree of the homogenization is dependent on the biology of the taxa involved (Gompert et al., 2017; Irwin & Schluter, 2021). Identifying the factors that promote or limit gene flow among divergent populations is key to uncovering the origin of reproductive barriers, and thus the drivers of speciation (Coyne & Orr, 2004; Jiggins & Mallet, 2000).
Characterizations of hybrid individuals and hybrid zones offer further insights into the dynamics shaping interactions among lineages. For instance, comparisons of hybrid indices, which estimate the proportion of an individual’s ancestry from different lineages, and the level of heterozygosity at well-differentiated loci can classify individuals as first-generation hybrids, later-generation hybrids, or backcrosses (Fitzpatrick, 2012). Moreover, geographic clines provide insight into the role of deterministic processes in maintaining species boundaries, by fitting clinal models of how traits and genetic markers transition between distinctive populations (Endler, 1977). Here, the width of the cline provides a measure of the permeability of the reproductive barrier, and the concordance between cline widths of different traits can inform about the relative strength of selection on a given trait (Barton & Gale, 1993). Further, comparing cline centers may identify traits and genes that have undergone differential introgression (e.g., Baldassarre et al., 2014; Brumfield et al., 2001). Overall, these characterizations offer insights into which traits are important in causing reproductive isolation, the strength of selection on different traits, and even the relative fitness of individuals with mixed ancestry (e.g., Coster et al., 2018; Walsh et al., 2016).
Interpreting contact zone dynamics is often challenging, as the same pattern of phenotypic and genetic variation can result from different processes (Harrison, 1990). For example, an intergradation between two divergent populations might have occurred from ongoing parapatric divergence (i.e., primary contact) or from hybridization after differentiation in geographic isolation (i.e., secondary contact; Morales-Rozo et al., 2017). Comparing model support of alternative demographic scenarios based on coalescent theory may differentiate primary versus secondary contact, thus providing better estimates of important demographic parameters, such as rates and timing of gene flow, that have created the observed patterns of genomic variation (Marchi et al., 2021). Characterizing the situation at replicated hybrid zones increases the power and resolution to infer the factors driving divergence and introgression, because parallel results increase the support and likelihood for a given interpretation (Nadeau et al., 2014) and different results can help to distinguish independent effects of different factors shaping hybrid zone dynamics (Scordato et al., 2020).
The Variable Seedeater (Sporophila corvina ) is a small tanager species which consists of four subspecies demonstrating relatively low genetic differentiation yet substantial phenotypic divergence (Ocampo et al., 2022a). Three subspecies occur in relatively close proximity across Costa Rica and Panama. The nominate subspecies S. c. corvina has almost completely black plumage and is distributed along the Caribbean slope from northeastern Costa Rica through central Panama. The other two subspecies, S. c. hoffmanni and S. c. hicksii, have white collars, bellies, and rumps (a pied plumage), differing only on the extent of the white patch of the throat (Olson, 1981), and range along the Pacific slope from central Costa Rica to eastern Panama. The species is absent in the high-elevation mountain ranges that separate the Caribbean and Pacific lowlands of Costa Rica and western Panama, but the three subspecies have contact where their distributions overlap. The three contact zones are as follows. First, previous morphological studies addressing taxonomic limits and distributions in this species have characterized a hybrid swarm occurring between S. c. corvinaand S. c. hicksii in central Panama (Hellmayr, 1938; Olson, 1981; Stiles, 1996; Figure 1A, no. 1), while the contact zone between S. c. hoffmanni and S. c. hicksii is less clearly understood as an intergradation of these subspecies at the Pacific slope in Veraguas province, based on a few specimen records of intermediate phenotypes (Olson, 1981; Figure 1A, no. 2). Finally, recent observations in the Central Valley of Costa Rica suggest the existence of a third contact zone between populations of S. c. hoffmanni and S. c. corvina (examples of observations available at eBird: www.ebird.org; Figure 1A, no. 3). This final contact zone might have been mediated by deforestation and urbanization in Costa Rica over the last few decades (Joyce, 2006), creating artificial corridors that could have facilitated interbreeding between previously isolated clades (Carantón‑Ayala et al., 2018; Moulton et al., 2017). However, these contact zones and interactions between subspecies are poorly studied, and to date, little is known about the genetic consequences of these three independent contact zones.
Here, we take advantage of the replicate contact zones between subspecies pairs of S. corvina occurring in Costa Rica and Panama (Figure 1A), looking at the fine-scale genetic variation and structure between populations and subspecies, to evaluate the role of phenotypic divergence and ecological barriers in mediating gene flow. We evaluate how differences in morphology and plumage pattern covary with genomic divergence to test the hypothesis that differentiation in plumage color is an important signal for conspecific recognition and thus acts as a reproductive barrier. We expect lower genomic differentiation and higher rates of hybridization between phenotypically similar subspecies (i.e.,S. c. hoffmanni and S. c. hicksii ) than more phenotypically divergent subspecies (e.g., S. c. corvina andS. c. hicksii ). Moreover, we characterize genomic hybrids and hybrid zones, and use model-based demographic inference to infer the most likely evolutionary scenarios shaping current patterns of genomic admixture and differentiation between subspecies.
METHODS