The White-naped Brushfinch (Atlapetes albinucha) comprises up to eight allopatric subspecies mainly identified by the color of the underparts (gray vs. yellow belly). Yellow and gray bellied forms were long considered two different species (A. albinucha and A. gutturalis), but they are presently considered as one polytypic species. Previous studies in the genus Atlapetes have shown that the phylogeny, based on molecular data, is not congruent with characters such as coloration, ecology, or distributional patterns. The phylogeography of A. albinucha was analyzed using two mitochondrial DNA regions from samples including 24 different localities throughout montane areas from eastern Mexico to Colombia. Phylogeographic analyses using Bayesian inference, maximum likelihood and haplotype network revealed incomplete geographic structure. The genetic diversity pattern is congruent with a recent process of expansion, which is also supported by Ecological Niche Models (ENM) constructed for the species and projected into three past scenarios. Overall, the results revealed an incomplete genetic divergence among populations of A. albinucha in spite of the species’ ample range, which contrasts with previous results of phylogeographic patterns in other Neotropical montane forest bird species, suggesting idiosyncratic evolutionary histories for different taxa throughout the region.

allotypy, coalescent, Last Glacial Maximum, mtDNA, montane forest, Pleistocene, phylogeography, plumage differentiation

Phylogeographic analyses of widespread Neotropical montane forest bird species have indicated different levels of geographic structure in the variation of mitochondrial DNA (mtDNA) among populations (Bonaccorso et al. 2008, Navarro-Sigüenza et al. 2008, Barrera-Guzmán et al. 2012, Ortíz-Ramírez et al. 2016). Deep genetic divergence may suggest a long history of geographic isolation, usually accompanied by phenotypic divergence, whereas a shallow genetic differentiation has been attributed to possible scenarios of either recent range expansions or recent divergence (Cortés-Rodríguez et al. 2008, Arbeláez-Cortés et al. 2010).

The genus Atlapetes comprises a group of Neotropical finches inhabiting mainly humid montane forests from Mexico to northern Argentina (Paynter 1972, 1978), and has been considered one of the most species-rich clades (nearly 30 recognized species) among the New World passerines (Paynter 1972, 1978, Sánchez-González et al. 2015). Diversification in this group started about 5.2–3.2 Mya, and has probably occurred mainly due to changes in the range of montane forests related to the Pleistocene glacial cycles (Ricklefs and Latham 1992, Hooghiemstra et al. 2006, Sánchez-González et al. 2015). Atlapetes also provides a complex and intriguing case of evolutionary differentiation in plumage color (Paynter 1972, 1978, Remsen and Graves 1995). Most Atlapetes species have either yellow or gray underparts, which has led to the distinction of two main plumage patterns distributed in a “leapfrog” fashion along the Andes (Remsen 1984). Paynter (1972, 1978) included yellow and gray taxa in two different phylogenetic groups; a third group included yellow-plumaged species with bicolored crowns (A. albinucha, A. gutturalis, and A. pallidinucha). This phenotypically-based arrangement suggested that similarly plumaged species may have shared a common ancestor. However, Remsen and Graves (1995) proposed that some yellow and gray taxa may be representatives of the same species, suggesting that plumage differentiation patterns may be adaptive, as species with pale gray colors are generally found in drier and higher elevation habitats, whereas yellow-colored taxa are generally found in humid and lower elevations (Remsen and Graves 1995, García-Moreno and Fjeldså 1999).

Atlapetes albinucha (White-naped Brushfinch) is a widely distributed species found in montane regions from eastern Mexico to Colombia (AOU 1998, del Hoyo et al. 2011, Gill and Donsker 2015, Remsen et al. 2015). This species inhabits mainly humid and temperate montane forests (900 to 3000 m), as well as upper tropical zones and edges of clearings of cloud forests (Dwight and Griscom 1921, Paynter 1978, Howell and Webb 1995, del Hoyo et al. 2011). Up to eight allopatric subspecies showing two well-differentiated plumage coloration patterns with clear geographic structure have been recognized (Figure 1, del Hoyo et al. 2011, Remsen et al. 2015, Gill and Donsker 2018). Phenotypic differentiation in this taxon led some researchers to consider such differentiated populations as two species (Dwight and Griscom 1921, Howell and Webb 1995, Navarro-Sigüenza and Peterson 2004) or as part of a polytypic species (Paynter 1978, Remsen and Graves 1995, AOU 1998, del Hoyo et al. 2011, Gill and Donsker 2015, Remsen et al. 2015). Differences between subspecies with pale-gray underparts are apparently largely clinal, and are based in subtle variations of size and color (Paynter 1978); while color morphs with yellow- and pale-gray underparts are allopatric (Figure 1), and their ranges are separated by the low valley of Río Grijalva (1000 m) in Chiapas (southeastern Mexico), where the two morphotypes are less than 100 km apart (Paynter 1978).

Figure 1. 

Atlapetes albinucha distribution shown in green stapled lines, based on Sánchez-González et al. (2015) and Natureserve (http://natureserve.org). Blue dots depict tissue samples used in the present study. Red dots depict records of the species used to construct the distribution model. Bird pictures depict the geographic regions where color morphs are found. Blue line depicts the location of the putative distribution barrier of the morphs in Chiapas.

Paynter (1972, 1978) reviewed extensively the taxonomy and geographic variation in A. albinucha complex and suggested that plumage differentiation in these nearly parapatric populations were not indicative of a high genetic differentiation nor the product of ecological exclusion, but the result of a low river barrier, therefore implying isolation due to environmental factors as the main cause for this phenotypic differentiation. Current taxonomic schemes have adopted the proposal that A. albinucha represents a single polytypic species (Paynter 1978, Remsen and Graves 1995, AOU 1998, del Hoyo et al. 2011, Gill and Donsker 2015, 2018, Remsen et al. 2015). A study using mtDNA genetic analyses (Sánchez-González et al. 2015) at the genus level showed a phylogenetic reconstruction in which both yellow- and gray-plumaged birds in the A. albinucha complex were recovered mixed in a monophyletic group, sister to A. pileatus and well separated from the rest of Atlapetes, partially supporting the conclusions of Paynter (1978).

Here, using an extensive sampling for A. albinucha (39 individuals from 24 localities in four countries), we tested: a) if yellow- and gray-plumaged groups are reciprocally monophyletic, b) if there is phylogeographic structure in this widespread taxon across their range, and c) if past reconstructions of the environmental conditions were A. albinucha ranges at present indicate distributional changes that may be related with their genetic-geographic variation. These questions were approached using a mtDNA assessment of populations included within this complex, as well as environmental niche modelling analyses based on records of voucher specimens in biological collections.

The major result of our analyses using mtDNA sequence data for individuals of A. albinucha is that this taxon exhibits an incomplete genetic differentiation along their range in the Neotropical Montane Forest. The lack of clear phylogeographic structure in this montane bird taxon is in sharp contrast with expectations based on plumage differentiation which has resulted in the recognition of up to eight subspecies (Gill and Donsker 2018) as well as to the genetic divergence found in other birds from other naturally fragmented environments throughout the Neotropics (García-Moreno et al. 2004, Cadena et al. 2007, Navarro-Sigüenza et al. 2008, Weir et al. 2008, Pérez-Emán et al. 2010, Arbeláez-Cortés and Navarro-Sigüenza 2013). Despite not recovering a clear phylogeographic structure, the geographic distribution of the genetic variation in the geography is not completely random (as indicated by AMOVA and F_ST values), suggesting a scenario in which two groups (South America and Mesoamerica) have been diverging in isolation followed by range expansion, allowing the mixture of the genetic variation in periods favoring habitat connectivity during the LIG. Signatures for this vicariant scenario of genetic differentiation may be found in the phylogeographic topology, in which two groups including mainly South American and mainly Mesoamerican individuals were recovered, and in the significant variation among geographic groups determined by the AMOVA analyses (Table 3), as well as in the gene flow values (Table 4). Moreover, the low nucleotide diversity, but high haplotype diversity we found for mtDNA of A. albinucha, is thought to be consistent with populations passing through genetic bottlenecks followed by rapid population growth (see Grant and Bowen 1998).

The phylogeographic pattern of A. albinucha is consistent with allotypy, a term used to denote a stage in intermediate polyphyly (Omland et al. 2006). Allotypy is a likely intermediate stage of divergence characterized by local fixation of haplotypes on the path to reciprocal monophyly (Hudson 1990). This genetic pattern has been found in other bird species such as ducks (Peters et al. 2005, Peters and Omland 2007) and ravens (Omland et al. 2006) in the Holarctic, a raptor species in Neotropical lowlands (Johnson et al. 2007), and in passerines from Australia (Joseph and Wilke 2007) and from the Neotropical montane forests (Arbeláez-Cortés et al. 2010). For most of these cases, present distribution of the genetic diversity may reflect the retention of the genetic diversity found in the ancestor for a long time after isolation, which may have had a larger population size, thus increasing the time for some polymorphisms to be retained (Joseph and Omland 2009). Similarly, population expansion derived from populations with high effective sizes may also explain the haplotype and nucleotide diversities observed (Harpending et al. 1998, Ray et al. 2003, Ng et al. 2013, Jezkova et al. 2015).

In the case of A. albinucha, BSP (Figure 4) suggests a long period of population stability, with a slight increase from 750,000 to 250,000 ya, after which a relatively slight decrease in population size occurred ca. 100,000 ya. This population decrease, followed by rapid population growth and range expansion is coincident with the Last Glacial Maximum during the Late Pleistocene (~21,000 ya), when colder conditions may have allowed the formation of corridors between previously isolated humid montane forest patches (Figure 4a), likely enhanced by the downward altitudinal range changes of the forest belt (Hooghiemstra et al. 2006, Gutiérrez-Rodríguez et al. 2011, Rojas-Soto et al. 2012, Ornelas et al. 2010, 2013, Ramírez-Barahona and Eguiarte 2013). Such a scenario, probably promoted gene flow between previously isolated populations (e. g., Hewitt 2000, Zink and Blackwell-Rago 2000, Weir 2006, Hooghiemstra et al. 2006, Barber and Klicka 2010, Pérez-Emán et al. 2010, Wachowiak et al. 2013, Bagley and Johnson 2014, Ornelas et al. 2016). In addition, an interesting result emerging from our ENM is that regions inhabited by yellow-plumaged and grey-plumaged populations in Chiapas-Guatemala have apparently never been isolated, which seems to support conclusions by Paynter (1972, 1978) regarding the weakness of a low river valley as an effective barrier in separating these populations.

Causes of differentiation in plumage coloration in A. albinucha remain elusive in our analysis, as both plumage coloration patterns appeared intermixed in the tree topology, which suggest different processes for the configuration of the genetic variation and the phenotypic plumage differentiation. Therefore, the clear phenotypic differentiation between yellow-colored birds in northern Chiapas and gray-colored birds in the rest of the distributional range suggests that plumage may be under natural selection. Similar results have been obtained for other groups of birds in different geographical and ecological settings, such as in the Tropical Pacific islands (Filardi and Smith 2005, Uy et al. 2009), shorebirds (Rheindt et al. 2011), and Australian woodswallows (Joseph et al. 2006). In the case of Atlapetes brushfinches, yellow- and gray-plumage patterns are apparently ecologically segregated from each other at different elevations. Gray-plumaged birds tend to occupy high elevation, whereas yellow-plumaged birds tend to occupy lower elevations (Remsen and Graves 1995), thus suggesting that gray-plumages have evolved to deal with conditions on high elevations, but also some dry low-elevation environments (Sánchez-González et al. 2015). Similar changes in plumage patterns like the one detected in our study, and their correlation with environmental variables, have been also documented for other bird groups throughout the World (see Bowers 1960, Hall et al. 1966, Wunderle 1981, Galeotti and Cesaris 1996, Grunst et al. 2014, Reudnik et al. 2015). At the molecular level, plumage color changes are apparently a result of the concentration of lutein in the feather (Johnson and Brush 1972, Brush and Johnson 1976, McGraw and Hill 2006), however the specific mechanism in Atlapetes is unknown, although some studies point to single point mutations at the MCR-1 (melanocortin-1 receptor gene) as responsible for similar plumage changes in birds (reviewed in Mundy 2005, but see Cadena et al. 2011). Consistent with previous molecular-based studies, this study also supported that mtDNA variation does not correspond to plumage pattern differentiation in this species (Sánchez-González et al. 2015), suggesting that plumage coloration pattern in A. albinucha may be taxonomically misleading because it doesn’t reflect population history. In addition, it has been shown that vocal repertories are very similar and calls between color morphs cannot be reliably differentiated (Sánchez-González et al. 2015, Boesman 2016).

Results in this paper are not conclusive in terms of the currently accepted taxonomy for A. albinucha. Genetic divergence as a result of allotypy is apparent, suggesting that these taxa are likely at allotypy (Omland et al. 2006). Results in other bird taxa where allotypy has been found, show support to maintain recognized species, as genetic divergence is accompanied by morphological divergence (e. g., Peters and Omland 2007, Johnson et al. 2007), whereas others advocate to a single widespread (albeit genetically differentiated) species (e. g., Peters et al. 2005, Omland et al. 2006), lending an ambiguous support for species recognition. The study of Johnson et al. (2007) offered however, a threshold for species and subspecies recognition for a Neotropical raptor. The application of such a threshold for A. albinucha would support a subspecific status for all populations analyzed, thus maintaining the current taxonomic treatment (Paynter 1978, AOU 1998, Gill and Donsker 2015). However, further studies should be extended to include southern Central American populations and other genetic markers. Finally, results presented here underline that a general pattern for the evolution of montane bird taxa in Mesoamerica and Northern South America should consider several exceptions like the one depicted here for A. albinucha, as well as emphasize the role of idiosyncratic events in the recent evolution of bird taxa in this region, as it has been suggested for lowland bird taxa (Smith et al. 2014).

Genetic patterns found in A. albinucha were unexpected given previous findings in birds and other taxa codistributed in montane forests throughout the region (see Ornelas et al. 2013), which in general have showed corresponding patterns of genetic and morphological divergence (e. g., García-Moreno et al. 2004, Pérez-Emán et al. 2010).

The phylogeography of A. albinucha is consistent with allotypy, which has been suggested to represent an intermediate stage in the path to reciprocal monophyly (Omland et al. 2006). Most cases of allotypy have been reported in temperate birds from Eurasia (Peters et al. 2005, Omland et al. 2006, Peters and Omland 2007) as well as in the Eremian birds from Australia (Joseph and Wilke 2007), as well as in the Neotropics (see also Arbeláez-Cortés et al. 2010).

Environmental factors may have played a major role in shaping the evolution of morphological traits by natural selection that have been considered taxonomically relevant (Ball and Avise 1992), such as coloration pattern seen across the entire lineage (Paynter 1972, 1978, Remsen and Graves 1995, Sánchez-González et al. 2015), but that are not congruent with the genetic divergence indicated by mtDNA.

This contribution is part of AR-M Undergraduate Thesis and was supported by a scholarship from PAPIIT. Financial support was obtained from CONACyT 152060 and PAPIIT 215515 to AGN-S, and from Vicerrectoría de Investigación y Extensión of Universidad Industrial de Santander, Colombia, under Grant Proyecto de investigación capital semilla 2300 to EA-C. We thank the University of Kansas (M. B. Robbins & A. T. Peterson), the Colección de Tejidos del Instituto de Investigación de Recursos Biológicos “Alexander von Humboldt”, and the Museo de Zoología de la Facultad de Ciencias “Alfonso L. Herrera” and their respective staff for kindly providing tissue samples used in this study; a special acknowledgment to all the scientific collectors that obtained and deposited samples and voucher specimens of A. albinucha in these collections. We thank Natural Earth (http://www.naturalearthdata.com) for providing the raster data that was used in this study to map our species distribution models. Logistic help was obtained from Jano Núñez, Fanny Rebón, and Alejandro Gordillo (at UNAM), and by Diana Espitia Reina and Claudia A. Medina Uribe (at IAvH). Comments to early versions of this paper were obtained from Laura Márquez, Leticia Ochoa, and Erick García-Trejo. We are deeply indebted to Jon Fjeldså and Knud Jønsson for their comments, which greatly improved our manuscript.