Genetic and Ecological Divergence of Cinnamon Hummingbird Amazilia rutila (Aves: Trochilidae) Continental Populations Separated by Geographical and Environmental Barriers

Abstract

Background and Research Aims: Historical geological events and climatic changes have played important roles in shaping population differentiation and distribution within species. Amazilia rutila (Trochilidae) is a widespread hummingbird species in the tropical dry forest along the Pacific slope and the Yucatán Peninsula in Mexico. Methods: We used mitochondrial DNA sequence, ecological niche modelling and niche divergence tests to determine the effects of major geographic barriers and environmental variability on genetic and niche divergence of A. rutila continental populations. Results: Our results revealed three genetic groups without haplotype sharing corresponding to the distribution of individuals/populations from the Pacific slope W of the Isthmus of Tehuantepec (PAC), in Oaxaca and Chiapas E of the Isthmus of Tehuantepec (CHIS_OAX) and those from the Yucatán Peninsula and Guatemala (YUC). Values of neutrality tests suggest past demographic expansion without effective population size changes over time, and the time since the demographic expansion ranged between 39.4 and 84.45 ka BP. Each genetic group differed in their position in environmental space, with low-to-very limited overlap in the fundamental climatic niche dimensions of all groups analyzed, particularly between YUC and PAC. Analysis of climate differentiation and ecological niche comparisons showed that the environmental space occupied by these mtDNA groups is similar but not identical. Conclusion: We conclude that the genetic differentiation of A. rutila is consistent with a model of population isolation by geographical barriers and environmental differences. Inferences about the consequences of past demographic expansion and isolation underlying intraspecific evolutionary relationships await further study. Implications for Conservation: Our findings highlight the importance of preserving evolutionary significant units of this widespread hummingbird species. Conservation actions must consider intrinsic requirements of evolutionarily distinct populations and the environmental drivers that shape their distributions, maximizing preservation of intraspecific genetic variability and monitoring changes in genetic diversity.

Keywords

demographic expansion Mesoamerica Mexico niche divergence refugia Trochilidae tropical dry forest

Introduction

Phylogeography has proven successful for understanding how genetic variation across populations is geographically structured and shaped through time (Avise, 2000; Avise & Walker, 1998). Given the importance of genetic diversity to the maintenance of biological diversity, phylogeographic studies are pivotal in providing baseline data for cataloguing and mapping of intraspecific diversity and for the identification and maintenance of within-species evolutionary potential (Beheregaray, 2008; Macqueen, 2012). In the Mexican landscape, avian phylogeographic studies have revealed two general (opposing) patterns: (1) strong geographic structuring associated with the uplift of the mountains and ancient vicariance events (e.g., Arbeláez-Cortés et al., 2010; Barrera-Guzmán et al., 2012; Maldonado-Sánchez et al., 2016; McCormack et al., 2008; Ortiz-Ramírez et al., 2016); or (2) weak geographic structuring associated with high levels of gene flow through permeable geographical barriers during the Pleistocene and population connectivity during recent events of expansion and secondary contact (e.g., van Els et al., 2014).

The first pattern has been found in hummingbirds (Trochilidae), with strong geographic structuring (i) in species with populations separated by biogeographic barriers such as the Isthmus of Tehuantepec, the Trans-Mexican Volcanic Belt, and/or the Motagua-Polochic-Jocotán fault system (Hernández-Soto et al., 2018; Jiménez & Ornelas, 2016; Malpica & Ornelas, 2014; Rodríguez-Gómez et al., 2013, 2021, Zamudio-Beltrán et al., 2020, 2020b); (ii) species with disjunct distribution of populations (Arbeláez-Cortés & Navarro-Sigüenza, 2013; González et al., 2011; Licona-Vera & Ornelas, 2014); (iii) or in species inhabiting naturally fragmented habitats such as cloud forests (Cortés-Rodríguez et al., 2008a; Ornelas et al., 2016; Zamudio-Beltrán & Hernández-Baños, 2018). In contrast, the second pattern of weak geographic structuring and high levels of gene flow has been found in hummingbird species inhabiting lowland humid forest edges and coastal habitats, oases in tropical deciduous forests and arid montane scrub (González-Rubio et al., 2016; Licona-Vera et al., 2018a; Miller et al., 2011; Rodríguez-Gómez & Ornelas, 2015, 2018). Although studies with lowland hummingbird species have not shown a strong genetic structure, studies with other bird species distributed along the Mexican Pacific slope have shown stronger genetic structure, in which the genetic breaks were partially compatible with climatically stable areas (e.g., Arbeláez-Cortés et al., 2010). Nonetheless, the pervasive role of geo-climatic factors remains poorly understood for taxa at lower elevations, owing to a paucity of studies on widely distributed species in lowland tropical forests.

The tropical dry forest (hereafter TDF) is widespread and almost continuously distributed over extensive areas on the Pacific slope of Mexico, from central Sonora to southeastern Chihuahua to the southern state of Chiapas and continuing on to Central America with no apparent geographic barriers to dispersal, and on the Gulf of Mexico slope (central Veracruz) and the northwestern of the Yucatán Peninsula (Rzedowski, 1978). The TDF is characterized by a pronounced seasonality in precipitation with a marked dry period during several months (Portillo-Quintero & Sánchez-Azofeifa, 2010). In northern Mesoamerica, geographical and ecological barriers associated with the Isthmus of Tehuantepec, Motagua-Polochic-Jocotán fault system, and the Nicaraguan Depression circumscribe the distribution of TDF, in which the biotic diversification involved a mixture of vicariance and dispersal events (Prieto-Torres et al., 2019). Palynological information suggests that fragmentation and subsequent range expansion of forest fragments occurred in different areas within the continental distribution of the TDF associated with Pleistocene climatic fluctuations (Graham & Dilcher, 1995; Pennington et al., 2004; Toledo, 1981). However, there is also evidence of species divergence prior to the Pleistocene (Pennington et al., 2004; Weir, 2006), in which earlier historical and climatic events within the region contributed to deeper phylogeographical breaks within taxa (e.g., Becerra, 2005; Pennington et al., 2000; Zarza et al., 2018). In Mexico, the TDF is the predominant type of tropical vegetation, originally covering 15 % of the Mexican territory and over 60 % of the current total area of tropical vegetation in the country (Trejo & Dirzo, 2000). This type of vegetation has both a high diversity and endemism, with a considerably spatial variation in structure and species composition (Prieto-Torres et al., 2019; Rzedowski, 1978; Trejo & Dirzo, 2000). Despite the widespread distribution, the TDF is severely threatened in the country by deforestation at the local scale, with 60 % of the original vegetation lost and remnant forests (19 % in a forested condition) restricted to areas with steep slopes and by an increased frequency and intensity of droughts forecasted under future climate change scenarios predicting substantial changes in rainfall regimes in the TDF (Allen et al., 2017; Prieto-Torres et al., 2016; Trejo & Dirzo, 2000).

The Cinnamon Hummingbird (Amazilia rutila DeLattre, 1843) is a good model to understand the historical biogeography of the TDF in Mexico because its widespread distribution crosses the Isthmus of Tehuantepec and Motagua-Polochic-Jocotán fault system geographical barriers and allopatric distribution on the Yucatán Peninsula, which could have produced partially- or fully reproductively isolated populations. The inclusion of several Amazilia rutila samples in Ornelas et al. (2014) Amazilia’s phylogeny highlights within-species geographical structure of genetic variation. In a recent study, Vázquez-López et al. (2021) assessed genetic and morphometric variation in the Cinnamon Hummingbird Amazilia rutila complex. A phylogenetic analysis using DNA sequences of two mitochondrial (mtDNA) and two nuclear genes (nDNA) yielded three genetic groups: the Pacific slope, the Chiapas region, and the Yucatán Peninsula and northern Central America. Individuals from the Tres Marías Islands, much larger than those from continental populations, formed a monophyletic group but nested within the Mexican Pacific slope group in the concatenated dataset. The three genetic groups were recovered by individual phylogenies for mitochondrial loci; however, individual nuclear locus phylogenies did not recover any geographic structure (Vázquez-López et al., 2021). In the present study, we combined mitochondrial DNA sequences of three gene regions (ND2, ATPase 6, ATPase 8), ecological niche modelling and niche divergence tests to explore the historical demography and level of ecological divergence within A. rutila. First, we inferred genetic differentiation and genetic structure among continental populations of Cinnamon Hummingbird across northern Mesoamerica for a different set of individuals as compared to those used in Vázquez-López et al. (2021). With a set of abiotic variables (temperature and precipitation), the current distribution of suitable habitat and fundamental ecological niches were then estimated based on occurrence data of the species to assess whether geography or environmental variability played a role on genetic and niche divergence of this TDF specialist. Specifically, we asked: (1) whether the genetic variation of Amazilia rutila across its geographic range in northern Mesoamerica corresponds to existing geographic barriers; (2) whether divergence and demographic changes correspond temporally to the formation of geographic barriers; and (3) tested genetic groups of A. rutila for niche differences. Given that the environments of A. rutila in TDF are heterogeneous across its geographical range and that potential climatic and/or geographical barriers to gene flow or the permeability of the landscape for passage of A. rutila might have changed over time, we expect a weak pattern of genetic structuring and extensive gene flow and admixture across its distribution.

Material and Methods

Study Species

Cinnamon Hummingbird is distributed from Sinaloa south along the Pacific slope (including the Tres Marías Islands) to southern Chiapas, and along the coast of the Yucatán Peninsula in Mexico to Belize, with additional Central American populations extending to NW Costa Rica (Friedmann et al., 1950; Howell & Webb, 1995). Although sexes are similar (plumage), males have the bright red bill with black tip and females have it with mostly black above (Howell & Webb, 1995). In Mexico, the Cinnamon Hummingbird occupies the tropical dry forests, thorn forests and gallery forests from sea level up to 1600 m of altitude (Arizmendi et al., 2020; Howell & Webb, 1995). In the TDF of western Mexico and the Motagua Valley in Guatemala, Cinnamon Hummingbird is a resident and most abundant territorial species, and the core hummingbird species of the hummingbird-plant species network (Arizmendi & Ornelas, 1990; Bustamante Castillo et al., 2018, 2020; Díaz Infante et al., 2020). The clade composed of Amazilia rutila and sister species A. yucatanensis Cabot, 1845 and A. tzacatl De la Llave, 1833 is retrieved as monophyletic in molecular phylogenies (McGuire et al., 2014; Ornelas et al., 2014) and split from South American species c. 14–11 million years ago (Ornelas et al., 2014). The split rutila / tzacatl-yucatanensis is estimated to have occurred c. 12 million years ago, and the age of the A. rutila clade is estimated at 8 million years ago while the split A. tzacatl / A. yucatanensis occurring at c. 7–5 million years ago (Ornelas et al., 2014). These estimates coincide with high diversification rates of Bursera trees (Burseraceae, torchwood, copal) on the Pacific Coast and Balsas River Basin around 7.5 million years ago and the lower diversification in the Atlantic Coast and recent entry into Central American lowlands (Becerra, 2005), where A. yucatanensis and A. tzacatl are currently distributed.

Four subspecies are recognized on the basis of geographic variation in size and plumage colour (Dickinson & Remsen, 2013; Gill et al., 2020; Friedmann et al., 1950; Schuchmann, 1999; Weller, 1999): (1) A. r. diluta Van Rossem, 1938 from W Mexico (Sinaloa and Nayarit), sometimes considered a synonym of A. r. rutila (Friedmann et al., 1950; Van Rossem, 1938); (2) A. r. rutila (DeLattre, 1843) on the Pacific slope of Mexico, from Jalisco to Oaxaca, and south to Chiapas and the Yucatán Peninsula south to the arid valleys of interior Guatemala to NW Costa Rica; (3) A. r. graysoni Lawrence, 1866 off W Mexico on Tres Marías Islands; proposed to be elevated to species Amazilia graysoni (Gómez de Silva et al., 2020; Vázquez-López et al., 2021); and (4) A. r. corallirostris (Bourcier & Mulsant, 1846) SE Mexico on the Pacific slope in southern Chiapas, Mexico, south through western Guatemala to the Lempa River, El Salvador (Friedmann et al., 1950); or includes the populations from the Yucatán Peninsula south to Costa Rica, usually attributed to nominate rutila (Weller, 1999). Amazilia rutila graysoni of Tres Marías Islands are darker (more bronzy above and duskier below) and larger than mainland birds (Howell & Webb, 1995; Vázquez-López et al., 2021), A. r. diluta is similar to A. r. rutila but coloration below is paler and more pinkish cinnamon and upperparts slightly more golden bronze (less greenish), whereas A. r. corallirostris birds have darker rufous coloration underparts with chin more conspicuously spotted white (Howell & Webb, 1995; Schuchmann, 1999; Van Rossem, 1938). Populations on the Yucatán Peninsula are allopatrically distributed from those of A. r. rutila on the Pacific slope and from those of A. r. corallirostris from southern Chiapas, Mexico through western Guatemala to El Salvador and Costa Rica (Howell & Webb, 1995; Weller, 1999). However, there is no agreement with regard the phenotypic differences distinguishing these subspecies and their distribution limits and taxonomic status (Dickinson & Remsen, 2013; Friedmann et al., 1950; Gill et al., 2020; Schuchmann, 1999; Weller, 1999).

Sample Collection and DNA Sequencing

For molecular analysis, we sampled 71 individuals from 20 sites across the distribution of A. rutila in Mexico and Guatemala between almost sea level and 1095 m of altitude (Table 1, Supplementary Figure S1). Hummingbirds were captured using mist nets, and two rectrices were collected from each hummingbird as a source of DNA for subsequent genetic analysis before the bird was released. Samples were collected under the required permits and using approved animal welfare protocols. According to current known range of subspecies, we sampled four individuals of A. r. diluta, 30 individuals of A. r. rutila, and 37 of A. r. corallirostris (Table 1, Supplementary Figure S1). The sampling presented in this study practically covers the continental distribution limits of A. rutila in northern Mesoamerica; however, large extensions along the Pacific slope and Central America remain unsampled and the limits and precise distribution of subspecies are disputed (Weller, 1999).

Table 1.

Geographic Information and Sample Sizes of the 20 Amazilia rutila Localities Sampled in this Study.

Location number	Location	Population number	Grouping	n	Altitude (m)	Latitude (N)	Longitude (W)
1	Sinaloa, Culiacán	1	PAC	4	9	18° 10´ 43´´	102° 28´ 08´´
2	Jalisco, Chamela	2	PAC	3	21	18° 36´ 98´´	101° 26´ 83´´
3	Jalisco, Chamela, Camino Antiguo Sur	3	PAC	1	47	19° 29´ 48´´	105° 02´ 20´´
4	Jalisco, Playa Chalacatepec	2	PAC	2	15	17° 12´ 48´´	100° 43´ 25´´
5	Jalisco, La Manzanilla	3	PAC	1	41	19° 16´ 34´´	104° 47´ 20´´
6	Jalisco, Río Las Joyas, La Manzanilla	4	PAC	3	22	19° 35´ 13´´	104° 16´ 27´´
7	Jalisco, Careyitos	2	PAC	3	1	19° 26´ 18´´	105° 01´ 31´´
8	Jalisco, Barranca El Choncho	3	PAC	2	241	19° 20´ 25´´	104° 43´ 47´´
9	Guerrero, Río Santiago Atoyac	5	PAC	3	123	18° 36´ 19´´	95° 05´ 31´´
10	Guerrero, San Andrés Atoyac	6	PAC	3	18	18° 31´ 16´´	95° 02´ 26´´
11	Oaxaca, San Gabriel Mixtepec	7	PAC	3	700	16° 09´ 46´´	97° 08´ 16´´
12	Oaxaca, Santa Efigenia	8	CHIS_OAX	3	8	19° 13´ 21´´	96° 18´ 56´´
13	Chiapas, Tapachula, Finca Irlanda	9	CHIS_OAX	3	1095	15° 17´ 34´´	92° 33´ 59´´
14	Guatemala, Esquipulas, Finca El Rosario	10	YUC	4	990	14° 55´ 57´´	89° 34´ 42´´
15	Campeche, Escárcega	11	YUC	1	7	18° 36´ 16´´	90° 43´ 55´´
16	Quintana Roo, Sian Kaan	11	YUC	2	602	16° 45´ 05´´	93° 31´ 05´´
17	Quintana Roo, Laguna Guerrero	11	YUC	2	10	18° 42´ 42´´	88° 14´ 19´´
18	Yucatán, Ria Lagartos	12	YUC	9	3	18° 31´ 08´´	88° 16´ 11´´
19	Yucatán, Chicxulub	13	YUC	11	11	18° 54´ 19´´	91° 03´ 31´´
20	Yucatán, Puerto Progreso	14	YUC	8	7	21° 01´ 58´´	89° 45´ 12´´
	n			71

Abbreviations: PAC, Pacific slope west of the Isthmus of Tehuantepec; CHIS_OAX, Pacific slope east of the Isthmus of Tehuantepec; YUC, Yucatán Peninsula and Central America.

DNA was extracted from tail feathers with the DNeasy Blood and Tissue extraction kit (Qiagen, Valencia, CA, USA), following the protocol recommended by the manufacturer. We amplified and sequenced three gene regions: NADH nicotinamide dehydrogenase subunit 2 (ND2, 398 bp), and the mitochondrial adenosine triphosphatase synthase 6 and 8 genes (ATPase 6 and ATPase 8, hereafter ATPase, 682 bp). Amplification of ND2 was conducted with primers L5215–H5578 (Hackett, 1996), whereas for the ATPase we used primers L8929 (Sorenson et al., 1999) and H9855 (Eberhard & Bermingham, 2004). Protocols for DNA extraction, PCR, and for sequencing the PCR products are described elsewhere (Licona-Vera & Ornelas, 2014). All newly acquired sequences have been submitted to GenBank (Accession nos. ND2: OP837824–OP837894; ATPase 6–8: OP837895–OP837953).

Relationships among Haplotypes

Statistical parsimony networks for the single (ND2, ATPase 6–8) and combined (ND2+ATPase 6–8) mtDNA datasets were constructed to infer relationships among haplotypes using TCS v1.2.1 (Clement et al., 2000), with the 95 % probability connection limit and gaps treated as single evolutionary events, and visualized using POPART (Leigh & Bryant, 2015). Loops were resolved following the criteria given by Pfenninger & Posada (2002). The aligned sequences of ND2 and ATPase for haplotype networks (Figure 1(a) and (b)) were 398 (71 sequences) and 682 (59 sequences) base pairs (bp) long, respectively. The combined ND2+ATPase sequences (59 sequences) were 1080 bp. An additional statistical parsimony network was constructed as described using a composite ND2 dataset that included our newly generated sequences and those in Vázquez-López et al. (2021) downloaded from the GenBank (Accession nos. MZ998668–MZ998740). All populations are newly sampled in this study, particularly those from most of the locations from Jalisco (locations 3–8), Santa Efigenia, Oaxaca (location 12), and Esquipulas, Guatemala (location 14) were not sampled in the previous work (Supplementary Table S1; see also Table S1 in Vázquez-López et al., 2021). However, the resulting matrix (143 sequences, 1079 bp after alignment) has a lot of missing data after alignment. Our ND2 sequences had a length of 398 bp and those in Vázquez-López et al. study had 605 to 1041 bp. Haplotype networks can be misleading in the presence of missing data because produce biased network relationships (collapsing of different sequences into a single haplotype) and/or fail to indicate alternative positions for sequences (Joly et al., 2007). Thus, we cut the sequences at the same length (235 bp) to avoid bias. The composite ND2 dataset (143 sequences) retrieved a single network and yielded 15 haplotypes (Supplementary Figure S2); however, we observed that in some sequences from the Vázquez-López et al. study insertions were presumably introduced during alignment and some sequences with missing data were coded incorrectly. These problems can affect results of analyses that rely on network topology, such as haplotype diversity and population differentiation indices or past population history inferences, and could bias estimates of population structure and migration rates (Joly et al., 2007). For these reasons, we do not use the composite ND2 dataset in further analysis.

Figure 1.

Statistical parsimony networks of (a) single ND2 and (b) ATPase data and (c) combined ND2+ATPase data of Amazilia rutila overlaid on a map of North America. Numbers correspond to sampling localities. Refer to Table 1 and Supplementary Table S1 (Supporting information) for sampling locality information and haplotype distribution. Haplotypes are coded with a different colour according to geographic regions, Pacific slope west of the Isthmus of Tehuantepec (PAC, yellow), Pacific slope of Chiapas and Oaxaca east of the Isthmus of Tehuantepec (dark red), and Yucatán Peninsula (YUC, turquoise), and a number codes each of the haplotypes. The size of sections of the pie charts corresponds to the number of individuals with that haplotype. In the inset (d), a Bayesian analysis of population genetic structure (BAPS) is presented based on the ND2+ATPase sequences. Colours indicate different genetic clusters (K = 3), as explained above.

Then, to estimate the most likely number of genetically differentiated clusters without making a priori assumptions about the partitioning of genetic diversity, we performed a nonspatial genetic mixture analysis with a Bayesian model-based approach employing the program BAPS v5.3 (Corander et al., 2008) and the combined ND2+ATPase 6–8 dataset. BAPS assesses the most likely number of genetically different clusters using the module for linked molecular data. The codon linkage model, appropriate for our sequence data, was applied. We surveyed the probability of a different number of genetic clusters (PP) under the independent loci model in two independent runs with the number of proposed clusters (K) ranging from 2 to 15, with 10 runs for each K.

Population Indices and Phylogeographic Structuring

To describe intraspecific genetic variation of A. rutila, molecular diversity indices (h, gene diversity; π, nucleotide diversity) and pairwise comparisons of F_ST values between genetic groups were calculated using ARLEQUIN v3.5 (Excoffier & Lischer, 2010) with 20,000 permutations. Haplotype diversity indices for each population (h_S, v_S) and at the species level (h_T, v_T), and coefficients of population differentiation (G_ST, N_ST) were estimated using PERMUT v2.0 (Pons & Petit, 1996). We further compared the G_ST and N_ST values and tested for phylogeographic structure using PERMUT with 10,000 permutations and the U-statistic. An N_ST value significantly higher than the G_ST value provides evidence of phylogeographic structure (Pons & Petit, 1996). Note that ‘populations’ are sampling localities, whereas ‘groups’ are sets of pooled populations, as specified in Table 1. Locations with one or two samples were lumped with closest location (seven samples from locations 2, 4 and 7 into population 2, four samples from locations 3, 5 and 8 into population 3, and five samples from locations 15, 16 and 17 into population 11).

Geographic Structure of Populations

To test for the presence of hierarchical population structure, analyses of molecular variance (AMOVA) were run in ARLEQUIN v3.5. Populations were treated (a) as one group to determine how much variation is explained by differences between populations sampled, and grouped into (b) two groups of populations separated by the Isthmus of Tehuantepec, from Sinaloa and Durango to Oaxaca on the Pacific slope of Mexico (west of the Isthmus of Tehuantepec; locations 1–11, Table 1) and from southern Chiapas, Mexico through western Guatemala to El Salvador and the Yucatán Peninsula (east of the Isthmus of Tehuantepec; locations 12–20, Table 1), or (c) three groups (PAC = locations 1–11, CHIS_OAX = locations 12–13, YUC = locations 14–20; Table 1). Significance of AMOVAs was determined with 20,000 permutations each. These groupings allow us to test several models for the presence of hierarchical population structure and genetic variability found among geographic regions.

We also infer the optimal number of geographically homogeneous and maximally differentiated groups (K) using a spatial analysis of molecular variance (SAMOVA) implemented in SAMOVA v1.0 (Dupanloup et al., 2002) and testing values of K from 2 to 5, with 10 replicates per each K value.

Historical Demography

Signatures of demographic expansion in A. rutila were addressed by means of neutrality tests, Fu’s Fs (Fu, 1997), Tajima’s D (Tajima, 1989) and Ramos-Onsins and Rozas’ R2 (Ramos-Onsins & Rozas, 2002) statistics of neutrality. R2 is the most powerful tests used to detect population growth (Ramos-Onsins & Rozas, 2002). Significance was evaluated by comparing observed values with null distributions generated by 10,000 replicates, using the empirical population sample size and the observed number of segregating sites in the pegas package of R v4.1.1 (Paradis, 2010; R Core Team, 2020). Also, we compared the distributions of pairwise nucleotide differences among haplotypes (mismatch distribution) to expectations of a sudden-expansion model (Rogers, 1995) for the genetic groups. Mismatch distributions were generated with ARLEQUIN v3.5 using the sum of squared deviations test (SSD) and Harpending’s raggedness index (Harpending, 1994), both of which are higher in stable, nonexpanding populations (Rogers & Harpending, 1992). Deviations from the null model of population expansion were tested with 1,000 parametric bootstrapping (Schneider & Excoffier, 1999) in ARLEQUIN. Significant negative values of D and Fs and small positive values of R2 indicate an excess of low frequency mutations relative to expectations under the standard neutral model (i.e. strict selective neutrality of variants, constant population size, and lack of subdivision and gene flow). The mismatch distribution analysis was carried out using the sudden demographic expansion model of Schneider & Excoffier (1999) in unsubdivided populations and the spatial expansion model in a subdivided population (Excoffier, 2004). We employed 20,000 replicates to test the goodness-of-fit of the observed mismatch distribution to that expected under the spatial and sudden demographic expansion model using the sum of squares differences (SSD) and Harpending’s raggedness index (Hri index; Harpending, 1994) according to Rogers & Harpending (1992).

The time at which the spatial expansion event took place was dated following the expression, t = τ/2 μk, where τ is the estimated number of generations since the expansion, μ is the mutation rate per site per generation, and k is the sequence length (Schneider & Excoffier, 1999). The expansion parameter tau (τ) was estimated using ARLEQUIN in genetic lineages in which signs of sudden demographic expansion were evident. To convert the time since expansion (t), we used a 2.1-years generation time based on the observation that the age of maturity begins 1 year after hatching, and an assumed low annual adult survival rate of 0.52 reported for Bassilina (Hylocharis) leucotis (Ruiz-Gutiérrez et al., 2012). The approximate average generation time (T) is calculated according to T = a + [s / (1–s)] (Lande et al., 2003), where a is the time to maturity and s is the adult annual survival rate. Based on this, the estimate for T was 2.1 years.

Lastly, changes in effective population size (N_e) through time were estimated using Bayesian skyline plots (BSP) in BEAST v2.4.2 (Bouckaert et al., 2014). We chose the HKY+I (for A. rutila sensu lato) and HKY (for PAC and YUC groups) substitution models with empirical base frequencies, a strict clock model, and a piecewise-linear coalescent Bayesian skyline tree prior with five starting groups. One independent run of 30 million generations each was run, with trees and parameters sampled every 1000-iterations, with a burn-in of 10 %. Results of each run were visualized using TRACER to ensure that stationarity and convergence had been reached (ESS > 200). The time axis was scaled using the mtDNA geometric mean substitution rate of 0.001214 s/s/l/Myr (Lerner et al., 2011). The BSP for the CHIS_OAX group was not estimated due to small sample size.

Ecological Niche Modeling (ENM)

We constructed an ENM model in MAXENT v3.3.3k (Phillips et al., 2006) to predict the current distribution of suitable habitat occupied by Amazilia rutila. Coordinates of occurrence data obtained from the Global Biodiversity Information Facility (GBIF.org (13 October 2021) GBIF Occurrence Download, https://doi.org/10.15468/dl.ps6zn8) were assembled and supplemented with our geo-referenced records from field collection. We debugged these data by selecting the records from museum collections and iNaturalist research-grade observations but eliminating those outside the known distribution of the species. After careful verification of every data location, we excluded duplicate occurrence records or in close proximity to each other (c. 1 km²) to reduce the effects of spatial autocorrelation using GridSample package in R (Thomson et al., 2017). The dataset was restricted to 312 unique presence records for the analysis.

We used 19 climatic variables summarizing data of precipitation and temperature (BIO1–BIO19 variables) as climate layers from WorldClim v1.4 (Booth et al., 2014; Hijmans et al., 2005) at c. 1 km² (30 arc-sec) spatial resolution to predict the current distribution of suitable habitat occupied by A. rutila. An important step in ENM is to delineate a realistic calibration region (‘M’, BAM diagram; Soberón & Peterson, 2005); that is, the set of sites accessible to a species over which models are calibrated (Atauchi et al., 2020; Barve et al., 2011; Freeman et al., 2019; Soberón & Peterson, 2005). In this study, we calibrated the ENM model for A. rutila (in practice a mask or GIS polygon) using a geographical clipping to the species range based on the ecoregions of Mexico and Central America proposed by Olson et al. (2001) as the ‘M’ accessible areas considering potential boundaries on the landscape to dispersal and altitude range limits (Supplementary Table S2 and Figure S3; Barve et al., 2011). Climate layers were clipped with the ‘M’ region for use in MAXENT analysis. We used the variance inflation factor (VIF) in the usdm package (Naimi, 2015) from the R v4.1.1 (R Core Team, 2020) to exclude highly correlated variables and avoid multicollinearity. We retained bioclimatic variables with VIF values lower than 10 for all species records in each run, until all remaining VIF values were less than 10, as VIF values higher than 10 indicate strong collinearity (e.g., Ranjitkar et al., 2016). After removing the highly correlated variables, eight variables were used in the final analysis (BIO2 = Mean Diurnal Range, BIO3 = Isothermality, BIO8 = Mean Temperature of Wettest Quarter, BIO13 = Precipitation of Wettest Month, BIO14 = Precipitation of Driest Month, BIO15 = Precipitation Seasonality, BIO18 = Precipitation of Warmest Quarter, and BIO19 = Precipitation of Coldest Quarter). Final model was constructed in MAXENT with ten cross-validation replicates without extrapolation and considering the average output grids as the final predictive model. We applied a binary transformation (absence or presence, zero or one) using the 10th percentile training presence logistic threshold (T10LT). The geographic representations of the climatically suitable areas were constructed using a continuous representation of environmental suitability values, which were spatially projected using ArcMap. The area under the receiver operating characteristic (AUC) curve was used to evaluate the prediction performance of the model, in which values around 0.5 represent distribution models no better than random and those around 1 represent a perfect fit between the observed and the predicted species distribution; acceptable models are those with > 0.7 AUC values (Phillips et al., 2006). However, several criticisms have been associated with this approach (e.g., Cobos et al., 2019; Lobo et al., 2008; Merow et al., 2014). For this reason, we also evaluated model using the partial-ROC test (Peterson et al., 2008). Within a value range from 0 to 2, values over 1 suggest a better performance than chance, by analysing the presences versus the absence against the total area predicted by MaxEnt (Osorio-Olvera et al., 2020). Lastly, we also used the true skill statistic (TSS), which is a threshold-dependent measure of model performance, to evaluate the accuracy of predictive maps generated by presence-only data (Allouche et al., 2006; Liu et al., 2013), where TSS values ranging between 0.4 and 0.8 are considered useful (Fielding & Bell, 1997; Landis & Koch, 1977). For each replicate, TSS was calculated using the T10LT and then TSS values were averaged among replicates using the sp package (Pebesma & Bivand, 2005) in R v4.1.1.

Niche Divergence

We quantified the differentiation (or overlap) between climatic niches of the Amazilia rutila genetic groups (PAC, CHIS_OAX, YUC) based on their ENMs. The databases were worked individually to avoid identification errors at the limits of the distribution areas. Climate niche overlaps among groups were estimated using the PCA-env method proposed by Broennimann et al. (2012) with variables selected in the ENM by assessing their variance inflation factors (VIF) and then eliminating the multicollinear predictors. After that, we reduced the 19 variables to the same eight variables used in our ENM. Principal component analysis (PCA) was used to transform the environmental space of the investigated or selected environmental variables into a two-dimensional space defined by the first and second principal components (Strubbe et al., 2015). The PCA-env is carried out to transform the climate layers into a reduced number of linearly uncorrelated variables, i.e. principal components (Broennimann et al., 2012). The first component represents the largest possible amount of variability in the original variables, and each subsequent component represents the largest part of the remaining variability. This test compares the available environmental conditions for each species within a defined study extension (background) with their observed occurrences and calculates the available environmental space defined by the first two PCA axes. The differences in the position of the species along the principal components reflect their environmental differences.

Subsequently, the overlapping of the niches by pairs of the groups (PAC vs. YUC, PAC vs. CHIS_OAX, YUC vs. CHIS_OAX) was calculated using the Schoener’s D metric (Schoener, 1970). The values of this metric range from 0 (meaning that the niches are completely different) to 1 (meaning that the niches completely overlap) (Broennimann et al., 2012) and the graphs were made to observe the surfaces density of occurrences for each group. To facilitate the interpretation of results, the outputs were condensed into five classes as suggested by Rödder & Engler (2011): 0–0.2=no or very limited overlap, 0.2–0.4=low overlap, 0.4–0.6=moderate overlap, 0.6–0.8=high overlap and 0.8–1.0=very high overlap. Finally, two different randomization tests were used to test the niche evolution hypotheses. An equivalency test, which determines whether the niches of two entities in two geographic ranges are equivalent (that is, whether the niche overlap is constant by randomly reallocating the occurrences of both entities between the two ranges), and a similarity test which compares the niche overlap of one randomly distributed range on its background while keeping the other unchanged, and then performs reciprocal comparison. Each randomization process is repeated 100 times (to ensure that the null hypothesis can be rejected with a high level of confidence), producing a null distribution of overlapping values against which the observed score was compared. If the observed value of D is located outside 95 % of the density of the simulated values, the null hypothesis is rejected (H0= the niches are similar or equivalent), which implies that the groups occupy different segments of the environmental space. The stochastically simulated values were generated using the ENM of each Cinnamon Hummingbird genetic group and an ENM created with random points drawn from the minimum convex polygon surrounding the original occurrence records of the other genetic groups of Cinnamon Hummingbird. The geographic ranges of the genetic groups were used as backgrounds individually. For the background test, we used the minimum convex polygon surrounding the original occurrence records and the entire Mexico and Guatemala as the species range. All analyses were computed with the ecospat package (Di Cola et al., 2017) in R.

Results

Haplotype network

The combined ND2+ATPase of A. rutila from 20 localities yielded 23 haplotypes (Figure 1(c), Table 1, Supplementary Table S3). The statistical parsimony analysis retrieved a single network, in which three haplogroups (mtDNA groups) were revealed: populations 1–11 west of the Isthmus of Tehuantepec (PAC, A. r. diluta+A. r. rutila), populations 12–13 from Oaxaca and Chiapas east of the Isthmus of Tehuantepec (CHIS–OAX, A. r. rutila or A. r. corallirostris), and populations 14–20 from the Yucatán Peninsula and Guatemala (YUC, A. r. rutila or A. r. corallirostris) (Figure 1(c), Table 1, Supplementary Table S3).

The most widespread haplotype was H2, which forms the core of the first haplogroup composed of ten haplotypes (YUC), eight haplotypes exclusively found in populations from the Yucatán Peninsula and two haplotypes exclusively found in Guatemala (Figure 1(c), Supplementary Table S2). The second most frequent haplotype (H1) recovered for the second haplogroup composed of eight haplotypes was distributed exclusively in PAC populations west of the Isthmus of Tehuantepec (Figure 1(c), Supplementary Table S3). Forming the third haplogroup, haplotypes H7, H10 and H21–H23 were shared among individuals from populations east of the Isthmus of Tehuantepec, from Chiapas and Oaxaca (CHIS_OAX). No haplotypes separated by several mutational steps were shared between populations of the three-mtDNA groups (Figure 1(c), Supplementary Table S3).

BAPS analysis with mtDNA sequences and spatial clustering of groups of individuals resulted in three congruent clusters (K = 3) as the best partition (log marginal likelihood = –686.98, PP = 1.0; Figure 1(d)). In agreement with the haplotype network, populations west of the IT fell into a single cluster (A. r. diluta + A. r. rutila). Populations east of the IT from Oaxaca and Chiapas clustered together (A. r. rutila or A. r. corallirostris), and populations from the Yucatán Peninsula and Guatemala (A. r. rutila or A. r. corallirostris) formed the third cluster in the BAPS plot (Figure 1(d)).

Population Indices and Phylogeographic Structuring

Number of haplotypes varied among groups, from five in CHIS_OAX with six samples to ten in YUC with 31 samples (Table 2). Gene diversity and nucleotide diversity values were highest for CHIS_OAX (0.87, 0.002), respectively, followed by those for PAC (0.73, 0.001) and YUC (0.66, 0.001) (Table 2).

Table 2.

Results of Neutrality Tests and Mismatch Distribution of Amazilia rutila Samples (ND2+ATPase) by Genetic Group to Infer Demographic Expansion.

Parameter	PAC	CHIS_OAX	YUC	A. rutila
N	22	6	31	59
N _H	8	5	10	23
h	0.732 ± 0.090	0.867 ± 0.129	0.660 ± 0.032	0.868 ± 0.032
π	0.00108 ± 0.00018	0.00181 ± 0.00039	0.00128 ± 0.00046	0.03107 ± 0.00154
D _T	–0.996	0.355	–2.169***	–0.937
Fs	–4.247***	–2.263*	–5.168***	–3.893***
SSD	0.012	0.007	0.010	0.079
Hri	0.097	0.062	0.107	0.040
R2	0.032***	0.195	0.042***	0.070
t	52.375	84.448	39.401	58.461

N = number of individuals, N_H = number of haplotypes per population, h = gene diversity, π = nucleotide diversity, D_T = Tajima’s D, F_S = Fu’s Fs, SSD = differences in the sum of squares or mismatch distribution, Hri = Harpending’s raggedness index. R2 = Ramos-Onsins and Rozas statistic. t = time since expansion (kyr BP). Positive values for D_T and F_S are indicative of mutation-drift-equilibrium, which is typical of stable populations, whereas negative values that result from an excess of rare haplotypes, indicate that populations have undergone recent expansions, often preceded by a bottleneck. Significantly negative values (at the P = 0.05 level for Tajima’s D test and P < 0.02 for Fs test; Excoffier & Lischer, 2010) in both tests reveal historic demographic expansion events. Significant (P ≤ 0.05) SSD and Hri values indicate deviations from the sudden expansion model. Small positive values of R2 indicate an excess of low frequency mutations relative to expectations under the standard neutral model. *p < 0.05, **p < 0.01, *** p < 0.001. Group abbreviations are as follows: PAC: Pacific, CHIS_OAX: Chiapas-Oaxaca and YUC: Yucatán Peninsula.

Observed genetic differentiation among populations based on ND2+ATPase (G_ST = 0.437, SE = 0.0838) indicated that A. rutila is genetically subdivided. Genetic diversity across all populations (h_T = 0.993, SE = 0.0407; v_T = 0.950, SE = 0.0490) was higher than the average within-population value (h_S = 0.525, SE = 0.0947; v_S = 0.256, SE = 0.1093). PERMUT analysis showed that N_ST (0.730, SE = 0.1177) and G_ST values were statistically different (p < 0.05), indicating phylogeographical structuring. Pairwise comparisons of F_ST values were high and significant when groups of populations were compared (PAC vs. YUC, F_ST = 0.9777, p < 0.001; PAC vs. CHIS_OAX, F_ST = 0.9616, p < 0.001; YUC vs. CHIS_OAX, F_ST = 0.9779, p < 0.001).

Geographic Structure of Populations

The AMOVAs showed significant genetic differentiation at each hierarchical level (Table 3). When groups were not defined, the AMOVA showed that the highest percentage of variation (97%) was explained by differences among populations and only 2.2 % by differences within populations (Table 3). When grouping populations as separated by the Isthmus of Tehuantepec, the highest percentage of genetic variance (73.4 %) was explained by differences between groups, 1.5 % by differences within populations, and 25.3 % of the variance accounted for differences among populations within groups. The F_CT value was high and significant (0.73, p < 0.01), indicating genetic differentiation between populations separated by the Isthmus. When AMOVA was performed for three groups (PAC, CHIS_OAX, YUC), the genetic differentiation was higher and significant (F_CT = 0.97, p < 0.001, Table 3).

Table 3.

Results of AMOVA and SAMOVA Models on Amazilia rutila Populations for Mitochondrial DNA (ND2 and ATPase).

Source of variation	df	Sum of squares	Estimated variance	% variance	Fixation indices
(1) No groups defined
Among populations	12	964.45	18.00	97.7	F_ST = 0.97***
Within populations	46	19.16	0.41	2.2
Total	58	983.61	18.41
(2) Two groups (east and west of the Isthmus of Tehuantepec)
Among populations	1	613.96	20.93	73.41	F_CT = 0.73**
Between populations within groups	11	350.48	7.16	25.13	F_CS = 0.94***
Within populations	46	19.16	0.41	1.46	F_ST = 0.98***
Total	58	983.61	28.51
(3) Three groups (A. r. diluta+A. r. rutila, A. r. corallirostris from Oaxaca and Chiapas through western Guatemala, A. r. corallirostris from the Yucatán Peninsula through Central America)
Among populations	2	947.10	27.83	97.50	F_CT = 0.97***
Between populations within groups	10	17.35	0.29	1.04	F_CS = 0.41***
Within populations	46	19.16	0.41	1.46	F_ST = 0.98***
Total	58	983.61	28.55
(4) SAMOVA, K = 3
Among groups	2	893.42	26.25	97.38	F_CT = 0.97***
Among populations within groups	10	17.15	0.29	1.08	F_CS = 0.41***
Within populations	46	19.04	0.41	1.54	F_ST = 0.98***
Total	58	929.62	26.96

F-statistics were used to estimate the proportion of genetic variability found among populations (F_ST), among populations within groups (F_SC) and among groups (F_CT). ns, not significant (p > 0.05), * p < 0.05, ** p < 0.001, *** p < 0.0001.

The SAMOVA detected strong geographical structure with same three groups of populations inferred as the optimal number of geographical clusters (Table 3).

Historical Demography

For ND2+ATPase, Tajima’s D and Fu’s Fs values were negative and significantly different from zero, except Tajima’s D in PAC and CHIS_OAX (Table 2). Also, the low and non-significant values of SSD and Hri indicate a good fit to the demographic expansion model and consistent with a scenario of a sudden demographic expansion, so the expansion model was not rejected (Table 2). R2 statistic showed positive, small and highly significant values for PAC and YUC groups, indicating that these groups presented past demographic expansion (Table 2). Based on our estimated values of τ, the average time since the demographic expansion was 52.37 ka BP for PAC, 84.45 ka BP for CHIS_OAX, and 39.40 ka BP for YUC (Table 2).

The Bayesian skyline plots suggest that the effective population size was stable over time in YUC and PAC, except a marginal decrease in Amazilia rutila as a whole (100,000–50,000 years ago; Figure 2).

Figure 2.

Bayesian skyline plots showing historical demographic trends for YUC and PAC genetic groups and for Amazilia rutila as a whole using mitochondrial sequences (ND2 and ATPase data).

Ecological Niche Modelling

The current distribution of A. rutila was supported by high predictive power (AUC, mean ± SD, 0.796 ± 0.33) indicating adequate model performance. The partial ROC test (1.651 ± 0.03) showed that models were statistically significant (p < 0.01). Thus, performance values for the model assessment approach indicated that the distribution model was statistically accurate. Determination of the threshold probability for predicted presence using TSS resulted in a mean proportion of correctly classified training observations (TSS, mean ± SD, 0.476 ± 0.056). The projections of estimated current distribution suggest that areas of suitable habitat for A. rutila are continuous along the Pacific slope and that the predicted distribution on the Yucatán Peninsula is separated from the distribution predicted along the Pacific slope (Figure 3(a)).

Figure 3.

(a) Predicted distribution of Amazilia rutila at present. The output of MAXENT consists of grid maps with each cell having an index of suitability indicated between 0 and 1. Biogeographic barriers Isthmus of Tehuantepec (IT) and Motagua-Polochic-Jocotán (MPJ) fault system shown as dashed lines. The approximate geographic range separation between A. r. diluta and A. r. rutila subspecies is shown on the map by solid horizontal line. (b) PCA plot showing the first two components of PCA-env in A. rutila. (c) PAC, (d) CHIS_OAX and (e) YUC population niche displayed on the same multi-dimensional scale represented by the first two axes of a principal components analysis (PCA) summarizing the entire study area. Grey shadings show the density of the occurrences of the species by cell. The solid and dashed contour lines illustrate 100 % and 50 % of the available (background) environment, respectively. Abbreviations: PAC, populations west of the Isthmus of Tehuantepec (A. r. diluta + A. r. rutila); CHIS_OAX, populations from Oaxaca and Chiapas east of the Isthmus of Tehuantepec (Amazilia ‘saturata’ according to Vázquez-López et al., 2021); and YUC, populations from the Yucatán Peninsula and Guatemala (A. r. corallirostris).

Niche divergence

The PCA indicated three niche axes that together explain 64.38 % of the total environmental variation along climatic gradients in A. rutila (PC1 = 49.95 %, PC2 = 14.43 %; Figure 3(b)). PC1 is positively associated with mean diurnal range of temperature (BIO2) and precipitation seasonality (BIO15) and negatively associated with precipitation variables (BIO13, BIO14, BIO18, BIO19), and the second niche axis is positively associated with precipitation of warmest quarter (BIO18) and negatively associated with temperature variables (BIO3, BIO8) (Table 4). The occurrence density surfaces in environmental space, as determined by PCA-env, showed that the position in environmental space varied among genetic groups (Figure 3(c)–(e)). The contribution of the climatic variables on the two axes of the PCA-env and the percentage of inertia explained by the two axes is presented in Figure 4(a). Each genetic group differed in their position in environmental space. In general, the Schoener’s D niche overlap scores, which ranged from 0.17 to 0.39, indicate a low to very limited overlap in the fundamental climatic niche dimensions of all groups analyzed, particularly the very limited niche overlap between YUC and PAC (Table 5). The occupied niches by the genetic groups were not identical (p < 0.009); the null hypothesis of niche equivalency was rejected for all comparisons between genetic groups (Figure 4(b)–(d), Table 5). For niche similarity tests, however, all comparisons were rejected (Figure 4(e)–(j)), indicating that their environmental space is more similar to each other than expected by chance (Table 5).

Table 4.

Contributions of the First Two PCA-env Axes to Environmental Space.

Bioclimatic variable	PC1 (49.95%)	PC2 (14.43%)
BIO2 = Mean Diurnal Range	202.0032	85.4200
BIO3 = Isothermality	–122.5425	–142.1315
BIO8 = Mean Temperature of Wettest Quarter	–54.0571	–104.9207
BIO13 = Precipitation of Wettest Month	–193.0196	23.8774
BIO14 = Precipitation of Driest Month	–200.6128	69.1043
BIO15 = Precipitation Seasonality	190.2408	–5.8811
BIO18 = Precipitation of Warmest Quarter	–130.9561	148.2826
BIO19 = Precipitation of Coldest Quarter	–196.5559	6.7901

Figure 4.

(a) The contribution of the climatic variables on the two axes of the PCA-env and the percentage of inertia explained by the two axes (PC1 and PC2). Histograms show simulated niche overlaps (grey bars) and the observed niche overlap between groups (bars with a circle) on which tests of niche equivalency (b–d) and niche similarity (e–j) were calculated from 100 iterations. The significance of the tests is shown. Abbreviations: PAC, populations west of the Isthmus of Tehuantepec (A. r. diluta + A. r. rutila); CHIS_OAX, populations from Oaxaca and Chiapas east of the Isthmus of Tehuantepec (Amazilia ‘saturata’ according to Vázquez-López et al., 2021); and YUC, populations from the Yucatán Peninsula and Guatemala (A. r. corallirostris).

Table 5.

Ecological Niche Comparisons for the Amazilia rutila Genetic Groups.

Comparisons		Niche overlap (D)	Niche similarity		Niche equivalency
a	b		a → b	b → a
PAC	YUC	0.1784	Similar	Similar	Non-equivalent**
PAC	CHIS–OAX	0.2271	Similar	Similar	Non-equivalent**
YUC	CHIS–OAX	0.3997	Similar	Similar	Non-equivalent**

Pairwise niche overlap values are presented for the comparisons of niche similarity and equivalency of species a with species b in terms of Schoener’s D, niche similarity p-values and equivalency via randomization test. The ecological niches are significantly more similar than expected by chance (niche similarity test), or more equivalent than expected by chance (niche equivalency test). **p < 0.01.

Discussion

Genetic Differentiation among Amazilia rutila Populations

In this study, we elucidate the geographic structure of genetic variation of A. rutila populations in Mexico and Guatemala using mitochondrial DNA sequence data, and determine the effects of major geographic barriers, geographic distribution of suitable habitat and environmental variability on population divergence. The haplotype network, F_ST statistics, PERMUT, BAPS, AMOVA, and SAMOVA estimates revealed that individuals from the Yucatán Peninsula and Guatemala are genetically distinct from those genetically differentiated in the Pacific slope separated by the Isthmus of Tehuantepec (west of the Isthmus from Sinaloa to Oaxaca and east of the Isthmus in Oaxaca and Chiapas), supporting the hypothesis that geographical or ecological barriers have limited gene flow and promoted isolation between populations in the three geographic areas. In a recent study, Vázquez-López et al. (2021) evaluated the genetic differentiation of Amazilia rutila populations, recognizing the existence of three genetic groups: Mexican Pacific slope, Chiapas, and Yucatán Peninsula. In their phylogeny, individuals from the Tres Marías Islands were nested within the Mexican Pacific slope group. Despite this, they proposed elevating each group to species status, Amazilia rutila (Mexican Pacific slope), Amazilia graysoni (Tres Marías Islands), Amazilia saturata (Chiapas), and Amazilia corallirostris (Yucatán Peninsula and Central America). Using samples from a different set of individuals and the implementation of a new mitochondrial marker (ATPase 6–8), we recovered three genetic groups in continental populations and these results partially coincide with the groups that present Vázquez-López et al. (2021). In our study the CHIS_OAX genetic group (Chiapas lineage or A. saturata according to Vázquez-López et al., 2021) also includes populations of Oaxaca located east of the Isthmus of Tehuantepec (Figure 1(c)).

The geographical break between PAC and CHIS_OAX at the Isthmus of Tehuantepec favours the hypothesis of genetic divergence due to isolation between populations separated by this barrier (allopatric fragmentation). This break is spatially congruent with those of other hummingbird species (Cortés-Rodríguez et al., 2008a; González et al., 2011; Jiménez & Ornelas, 2016; Malpica & Ornelas, 2014; Ornelas et al., 2016; Rodríguez-Gómez et al., 2013; Rodríguez-Gómez & Ornelas, 2018), and those of other bird (e.g., Álvarez et al., 2016; Barrera-Guzman et al., 2012; Cortés-Rodríguez et al., 2013; Maldonado-Sánchez et al., 2016) and vertebrate taxa (e.g., León-Paniagua et al., 2007; Castoe et al., 2009; Mendoza et al., 2019; Mulcahy et al., 2006) inhabiting montane woodlands in Mesoamerica, including plants (e.g., Gutiérrez-Rodríguez et al., 2011; Martínez de León et al., 2022; Ornelas & Rodríguez-Gómez, 2015). However, these vicariant events seemly occurred at different times (Barber & Klicka, 2010; Ornelas et al., 2013, 2015), indicating that genetic divergence occurred variously during different temporal windows. In addition, phylogeographic studies have found low levels of genetic differentiation and gene flow between populations separated by the Isthmus, for both highland and lowland hummingbird species (Hernández-Soto et al., 2018; Rodríguez-Gómez et al., 2013, 2021, Zamudio-Beltrán et al., 2020a, 2020b) and other taxa (Arbeláez-Cortés et al., 2010; Cortés-Rodríguez et al., 2008b; Mendoza et al., 2019; Navarro-Sigüenza et al., 2008; Ornelas et al., 2010; Vázquez-Miranda et al., 2009), leading these authors to suggest that the Isthmus is a semipermeable barrier to gene flow. Divergence time estimates for the split between populations of A. rutila separated by the Isthmus (4.84–3.16 million years ago; Vázquez-López et al., 2021) support the hypothesis that the split between populations by the Isthmus pre-date the Pleistocene glacial periods. Although the existence of several mutational steps between populations separated by the Isthmus is indeed indicative of ancient allopatric fragmentation, it could also result from non-sampled haplotypes from intermediate populations for this lowland hummingbird species. Thus, increased population sampling along the coast of Oaxaca (between localities 11 and 12; Figure 1(c)) would be needed to further assess the hypothesis of allopatric fragmentation and the evolutionary distinctiveness of the PAC and CHIS_OAX mtDNA groups of A. rutila separated by the Isthmus.

Our survey of molecular sequence data in populations of A. rutila identified a third genetic group in the Yucatán Peninsula (YUC). According to Vázquez-López et al. (2021), the split between populations of A. rutila on the Yucatán Peninsula and those along the Pacific slope occurred 8.29–5.88 million years ago, with an age of 2.6–1.42 million years ago for the crown node of the YUC clade. The existence of unique haplotypes on the Yucatán Peninsula is consistent with a hypothesis of isolation (restricted northward gene flow). The ecological features and biogeography of the Yucatán Peninsula show the climatic effects of its 65-Myr history, submerging under warm tropical waters by numerous marine transgressions to have occurred since to the Pliocene-Miocene that resulted in basically a large limestone slab slowly emerging from south to north with a rain gradient, from a very humid in the south-southeast to dry in the north-northwest pattern (Espadas-Manrique et al., 2003; Licona-Vera et al., 2018b; Vázquez-Domínguez & Arita, 2010). The vegetation also follows the SE-NW rain gradient, from tropical rainforests in the southern Petén region to tropical scrubland in the extreme NW portion of the Yucatán peninsula, with extensive areas covered with deciduous or semideciduous tropical forests between these two extremes (Ramírez-Barahona et al., 2009; Vázquez-Domínguez & Arita, 2010). Because of this distinctiveness, the Yucatán Peninsula is biogeographically divided into two provinces, the Petén province in the south and the dry Yucatán province in the north (e.g., Espadas-Manrique et al., 2003; Ramírez-Barahona et al., 2009; Vázquez-Domínguez & Arita, 2010).

The biogeography of the region is not only influenced by the effects of the formation and emergence of the Yucatán Peninsula but also by those related to the geology of and tectonic activity and mountain ranges on the Maya block, between the biogeographic breaks Isthmus of Tehuantepec and Motagua-Polochic-Jocotán fault system (Gutiérrez-García & Vázquez-Domínguez, 2012). Although these geological and topographical features and isolation by mountain ranges in the Maya block surely influenced the evolutionary history of Mesoamerican species (e.g., González et al., 2011; Guevara-Chumacero et al., 2010; Gutiérrez-García & Vázquez-Domínguez, 2012; Jiménez & Ornelas, 2016; Rodríguez-Gómez & Ornelas, 2014; Williford et al., 2016), few studies have included phylogeographies of species with patterns of historical divergence and population genetic differentiation between groups of populations from Chiapas and from the Yucatán Peninsula (Gutiérrez-García & Vázquez-Domínguez, 2012; Licona-Vera et al., 2018b; Ortiz-Rodriguez et al., 2020; Oyama et al., 2016). In these studies, the observed patterns of genetic divergence are consistent with the hypothesis that isolation of the dry Yucatán province by semideciduous tropical rain forest along the Petén region and Chiapas restricted northward gene flow from locations of the Pacific slope in Mexico into the TDF in the extreme NW portion of the Yucatán peninsula (an “ecological barrier”; Licona-Vera et al., 2018b and references therein). Together, our results are consistent with both a model of allopatric fragmentation and reduced gene flow by both physical and environmental barriers. Although individuals from Guatemala had non-shared haplotypes with those in the Yucatán Peninsula, in support of the hypothesis that the Motagua-Polochic-Jocotán fault system is a geographic barrier, only one mutational step separated haplotype H9 from most frequent haplotype in the Yucatán Peninsula (H2). Increased molecular markers and population sampling along the coast of Chiapas (between localities 12 and 13; Figure 1(c)) and within Guatemala would be needed to further assess the hypothesis of genetic divergence by isolation and the evolutionary distinctiveness of A. rutila on the Yucatán Peninsula. Given that this region is poorly studied at the phylogeography level, the phylogeography of A. rutila in combination with a niche-modelling approach and nuclear DNA markers with higher resolution (microsatellites) is particularly important to explore the genetic structure and to understand the effects of the Motagua-Polochic-Jocotán fault system and evolution of the TDF biodiversity on the Yucatán Peninsula.

Historical Demography

Our results reveal a high level of gene diversity across the 20 populations (Table 2). The high levels of genetic differentiation (pairwise comparisons of F_ST values) and the significant phylogeographic structure may result from low levels of historical gene flow among the studied populations and implies that isolation is important in shaping the genetic differentiation. The genetic divergence between groups of populations highlights the importance of geographical barriers in driving isolation followed by intraspecific differentiation and genetic structuring of A. rutila populations (N_ST > G_ST). However, low frequency haplotypes within each of the mtDNA genetic groups (PAC, CHIS_OAX, YUC) differ from each other by only one mutation in most cases and from high frequency central haplotypes within each haplogroup, suggesting its recent formation. Also, the combination of high genetic diversity, low nucleotide diversity values, and the presence of many low frequency single haplotypes separated by few mutational steps, indicate rapid population growth from ancestral populations with small effective population size (Avise, 2000). Indeed, the negative and significant values of neutrality tests and BSPs suggest past demographic expansion without effective population size changes over time for the PAC and YUC genetic groups, and the average time since the demographic expansion was between 52.37 ka BP and 39.40 ka BP, respectively. Projections on the distribution of suitable habitat under past conditions in Vázquez-López et al. (2021) study revealed that suitable habitat for A. rutila was continuous along the Pacific slope but more restricted than its current distribution during the Last Inter Glacial (140–120 kyr BP), and isolated and restricted to the extreme NW portion of the Yucatán peninsula. They proposed that Pleistocene climatic changes were a crucial factor in the divergence and emergence of the current lineages within A. rutila. However, estimated divergence times between groups contradict this hypothesis. Although recent studies using genomic sequencing of hummingbirds have demonstrated inconsistencies of dated phylogenies analysed using nuclear and mitochondrial DNA (Andermann et al., 2019), time since the demographic expansion based on mtDNA sequences in our study suggest that population expansion within mtDNA lineages is more recent than the origin and formation of geographical barriers.

Our results highlight a pattern of low intraspecific genetic divergence in a hummingbird species with a continuous and widespread distribution in the TDF, which is unique when contrasted to higher genetic divergence observed in other co-distributed bird species along the Mexican Pacific slope. For instance, Arbeláez-Cortés et al. (2010) showed marked phylogeographic structure in three co-distributed bird species along the Mexican Pacific slope. All three species showed marked phylogeographic structure, with breaks found in roughly similar areas reported in other taxa, such as the border between the Mexican states of Nayarit and Jalisco, southern Jalisco and Michoacán, Guerrero and Oaxaca, and between Oaxaca and Chiapas, genetic breaks partially compatible with climatically stable areas.

Instead, our results are similar to the ones of other widespread hummingbird species and other bird taxa with ranges divided by geographic barriers such as the Isthmus of Tehuantepec and the Motagua-Polochic-Jocotán fault system. The climatic changes of the Pleistocene would explain the current distribution of the genetic groups, but would not provide an explanation for the marked genetic differentiation among mtDNA lineages without haplotype sharing and the distribution of the CHIS_OAX group between the eastern portion of the Isthmus of Tehuantepec and the Motagua-Polochic-Jocotán fault system, indicating that geographical barriers played a role on the differentiation of these mtDNA genetic groups. Based on past distribution models in Vázquez-López et al. (2021), the ancestral population of the CHIS_OAX and PAC groups probably presented a continuous distribution along the Pacific slope during the Pliocene. After the rise in sea level and the flooding of the Isthmus of Tehuantepec, this distribution was interrupted, causing populations to the east and west of the Isthmus to remain separated. Therefore, it is likely that the ancestral population that remained west to the Isthmus expanded its distribution along the Pacific slope and to the Tres Marías Islands (Vázquez-López et al., 2021).

Niche Divergence

The genetic structuring within A. rutila suggests isolation and low levels of gene flow between mtDNA groups of populations across their geographic ranges. This scenario is supported by niche overlap and niche equivalency tests, which indicated that the three-mtDNA groups have low-to-very limited niche’s overlap (D = 0.39–0.17) and similar but not identical niches (Table 5). The three groups of A. rutila hummingbirds occupy similar environmental space (fundamental niche), probably due to shared ancestral habitat preferences (niche conservatism; Soberón & Nakamura, 2009; Wiens & Graham, 2005). Subspecies differences in environmental space (niche divergence) have been shown in Buff-Bellied Hummingbird (Amazilia yucatanensis), sister to A. rutila (Vásquez-Aguilar et al., 2021), suggesting that the distribution of one subspecies or genetic group cannot be implied by the distribution of another one. If allopatric fragmentation occurred in A. rutila, one would expect genetically divergent groups of populations to retain certain aspects of their fundamental niche. However, significant differences between mtDNA groups in niche equivalency (niches spaces are not identical) suggest niche divergence; while allopatric fragmentation may have been important for population genetic differentiation of A. rutila, our data support an environmental differentiation scenario.

The environmental niches of the A. rutila mtDNA groups are similar in terms of temperature and precipitation based on the PCA-env results. The similarity of environmental conditions between regions might be directly related to the availability of floral resources, which in turn constitutes the limiting factor in the distribution of genetic groups (Abrahamczyk & Kessler, 2015). However, the environmental niches for the three genetic groups were not equivalent. For the PC1 (50 % of total variation), the PCA-env showed that the greater differences among the three mtDNA groups correspond to differences in temperature and precipitation variables: PC1 is positively associated with mean diurnal range in temperature (BIO2) and precipitation seasonality (BIO15), and negatively associated with precipitation of wettest and driest month (BIO13, BIO14) and precipitation of warmest and coldest quarter (BIO18, BIO19). These results indicate that precipitation variables may be more important than temperature in determining the limits in the distribution of all three mtDNA genetic groups. While these data alone do not represent a rigorous test of the causes of population divergence or adaptive evolution, the tests of niche equivalency and niche similarity show that the climatic niches of the mtDNA lineages of A. rutila are more different than expected based on random predictions, and that the differences among genetic groups exceed those predicted by the background environments of the regions they inhabit. Thus, our findings based on the non-equivalency of the fundamental niches support the hypothesis that the mtDNA lineages of A. rutila have undergone coarse-scale niche divergence and are constrained by a set of climatic and macro-environmental conditions that might determine the distribution and availability of floral resources to which A. rutila interact with within each of the regions.

Implications for Population Conservation

Significant genetic differentiation in mtDNA sequences and genetic structure among A. rutila populations suggest that isolation by geographical barriers played a role on population differentiation. We further provide evidence for rejection of the niche equivalency hypothesis that confirmed that the retrieved genetic groups exist in distinct environmental niche spaces. Given the high population genetic diversity and identification of possible processes influencing the genetic structure of the species, our study highlights recognition of evolutionary significant units for decision-making in preserving this widespread hummingbird species. Additional work is needed to understand factors other than simply variability in the background environment maintaining population separation and intraspecific divergence within A. rutila.

Supplemental Material

Supplemental Material - Genetic and Ecological Divergence of Cinnamon Hummingbird Amazilia rutila (Aves: Trochilidae) Continental Populations Separated by Geographical and Environmental Barriers

Supplemental Material for Genetic and Ecological Divergence of Cinnamon Hummingbird Amazilia rutila (Aves: Trochilidae) Continental Populations Separated by Geographical and Environmental Barriers by Evelyn González-Rodríguez, Antonio Acini Vásquez-Aguilar and Juan Francisco Ornelas in Tropical Conservation Science

Supplemental Material

Supplemental Material - Genetic and Ecological Divergence of Cinnamon Hummingbird Amazilia rutila (Aves: Trochilidae) Continental Populations Separated by Geographical and Environmental Barriers

Supplemental Material

Supplemental Material - Genetic and Ecological Divergence of Cinnamon Hummingbird Amazilia rutila (Aves: Trochilidae) Continental Populations Separated by Geographical and Environmental Barriers

Supplemental Material

Supplemental Material - Genetic and Ecological Divergence of Cinnamon Hummingbird Amazilia rutila (Aves: Trochilidae) Continental Populations Separated by Geographical and Environmental Barriers

Supplemental Material

Supplemental Material - Genetic and Ecological Divergence of Cinnamon Hummingbird Amazilia rutila (Aves: Trochilidae) Continental Populations Separated by Geographical and Environmental Barriers

Supplemental Material

Supplemental Material - Genetic and Ecological Divergence of Cinnamon Hummingbird Amazilia rutila (Aves: Trochilidae) Continental Populations Separated by Geographical and Environmental Barriers

Footnotes

Acknowledgements

We thank Cristina Bárcenas, Andreia Malpica, Clementina González, Rosa Alicia Jiménez, Michelle Bustamante-Castillo, Mariana Hernández-Soto, M. Cristina MacSwiney G., José Manuel García-Enríquez, Yuyini Licona-Vera, and Andrés E. Ortiz-Rodriguez for field and laboratory assistance; Diego F. Angulo for data analysis; and five anonymous for providing useful comments on the manuscript. The samples collected in Mexico were conducted with the permission of the Secretaría de Medio Ambiente y Recursos Naturales, Instituto de Ecología, Dirección General de Vida Silvestre (permit numbers: INE: SEMARNAP, D00-02/3269, INE SGPA/DGVS/02038/07, 01568/08, 02517/09, 07701/11, 13528/14, 02577/15, 06448/16, 5050/19).

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the research competitive grants 61710, 155686, A1-S-26134 (awarded to JFO) from the Consejo Nacional de Ciencia y Tecnología (CONACyT; http://www.conacyt.mx) and research funds (20030/10563) from the Departamento de Biología Evolutiva, Instituto de Ecología, AC (INECOL; ) to JFO. E.G-R. was supported by a research assistant scholarship (20140) from the Sistema Nacional de Investigadores (SNI) granted to JFO (16464).

ORCID iDs

Evelyn González-Rodríguez

Antonio Acini Vásquez-Aguilar

Juan Francisco Ornelas

Data Availability Statement

The data given in this article are available from the corresponding author upon reasonable request. All unique sequences used in this study have been deposited in Genbank under accession numbers: ND2 locus: OP837824–OP837894, ATPase 6–8: OP837895–OP837953 ().

Supplemental Material

Supplemental material for this article is available online.

References

Abrahamczyk

Kessler

(2015). Morphological and behavioural adaptations to feed on nectar: How feeding ecology determines the diversity and composition of hummingbird assemblages. Journal of Ornithology, 156, 333–347. https://doi.org/10.1007/s10336-014-1146-5

Álvarez

Salter

J. F.

McCormack

J. E.

Milá

(2016). Speciation in mountain refugia: Phylogeography and demographic history of the pine siskin and black-capped siskin complex. Journal of Avian Biology, 47(3), 335–345. https://doi.org/10.1111/jav.00814

Allen

Dupuy

J. M.

Gei

M. G.

Hulshof

Medvigy

Pizano

Salgado-Negret

Smith

C. M.

Trierweiler

Van Bloem

S. J.

Waring

B. G.

Powers

J. S.

(2017). Will seasonally dry tropical forests be sensitive or resistant to future changes in rainfall regimes? Environmental Research Letters, 12(2), 023001. https://doi.org/10.1088/1748-9326/aa5968

Allouche

Tsoar

Kadmon

(2006). Assessing the accuracy of species distribution models: Prevalence, kappa, and the true skill statistic (TSS). Journal of Applied Ecology, 43(6), 1223–1232. https://doi.org/10.1111/j.1365-2664.2006.01214.x

Andermann

Fernandes

A. M.

Olsson

Töpel

Pfeil

Oxelman

Aleixo

Faircloth

B. C.

Antonelli

(2019). Allele phasing greatly improves the phylogenetic utility of ultraconserved elements. Systematic Biology 68(1), 32–46. https://doi.org/10.1093/sysbio/syy039

Arbeláez-Cortés

Navarro-Sigüenza

A. G.

(2013). Molecular evidence of the taxonomic status of western Mexican populations of Phaethornis longirostris (Aves: Trochilidae). Zootaxa 3716(1), 81–97. https://doi.org/10.11646/zootaxa.3716.1.7

Arbeláez-Cortés

Nyári

A. S.

Navarro-Sigüenza

A. G.

(2010). The differential effect of lowlands on the phylogeographic pattern of a Mesoamerican montane species (Lepidocolaptes affinis, Aves: Furnariidae). Molecular Phylogenetics and Evolution, 57(2), 658–668. https://doi.org/10.1016/j.ympev.2010.06.013

Arizmendi

M. C.

Ornelas

J. F.

(1990). Hummingbirds and their floral resources in a tropical dry forest in Mexico. Biotropica, 22(2), 172–180. https://doi.org/10.2307/2388410

Arizmendi

M. C.

Rodríguez-Flores

C. I.

Soberanes-González

C. A.

Schulenberg

T. S.

(2020). Cinnamon Hummingbird (Amazilia rutila), version 1.0. In: Schulenberg TS, Birds of the World (Ed). Cornell Lab of Ornithology, Ithaca, NY, USA. https://doi.org/10.2173/bow.cinhum1.01

10.

Atauchi

P. J.

Aucca-Chutas

Ferro

Prieto-Torres

D. A.

(2020). Present and future potential distribution of the endangered Anairetes alpinus (Passeriformes: Tyrannidae) under global climate change scenarios. Journal of Ornithology, 161, 723–738. https://doi.org/10.1007/s10336-020-01762-z

11.

Avise

J. C.

(2000). Phylogeography: The history and formation of species. Harvard University Press, Cambridge

12.

Avise

J. C.

Walker

(1998). Pleistocene phylogeographic effects on avian populations and the speciation process. Proceedings of the Royal Society of London Series B, 265(1395), 457–463. https://doi.org/10.1098/rspb.1998.0317

13.

Barber

B. R.

Klicka

(2010). Two pulses of diversification across the Isthmus of Tehuantepec in a montane Mexican bird fauna. Proceedings of the Royal Society of London Series B, 277(1694), 2675–2681. https://doi.org/10.1098/rspb.2010.0343

14.

Barrera-Guzmán

A. O.

Milá

Sánchez-González

L. A.

Navarro-Sigüenza

A. G.

(2012). Speciation in an avian complex endemic to the mountains of Middle America (Ergaticus, Aves: Parulidae). Molecular Phylogenetics and Evolution, 62(3), 907–920. https://doi.org/10.1016/j.ympev.2011.11.020

15.

Barve

Jiménez-Valverde

Lira-Noriega

Maher

S. P.

Peterson

A. T.

Soberón

Villalobos

(2011). The crucial role of the accessible area in ecological niche modeling and species distribution modeling. Ecological Modelling, 222(11), 1810–1819. https://doi.org/10.1016/j.ecolmodel.2011.02.011

16.

Becerra

J. X.

(2005). Timing the origin and expansion of the Mexican tropical dry forest. Proceedings of the National Academy of Sciences USA, 102(31), 10919–10923. https://doi.org/10.1073/pnas.0409127102

17.

Beheregaray

L. B.

(2008). Twenty years of phylogeography: The state of the field and the challenges for the Southern Hemisphere. Molecular Ecology, 17(17), 3754–3774. https://doi.org/10.1111/j.1365-294X.2008.03857.x

18.

Booth

T. H.

Nix

H. A.

Busby

J. R.

Hutchinson

M. F.

(2014). BIOCLIM: The first species distribution modelling package, its early applications and relevance to most current MAXENT studies. Diversity and Distributions, 20(1), 1–9. https://doi.org/10.1111/ddi.12144

19.

Bouckaert

Heled

Kühnert

C. H.

Xie

Suchard

M. A.

Rambaut

Drummond

A. J.

(2014). BEAST 2: A software platform for Bayesian evolutionary analysis. PLoS Computational Biology, 10(4), e1003537. https://doi.org/10.1371/journal.pcbi.1003537

20.

Broennimann

Fitzpatrick

M. C.

Pearman

P. B.

Petitpierre

Pellissier

Yoccoz

N. G.

Thuiller

Fortin

M. J.

Randin

Zimmermann

N. E.

Graham

C. H.

Guisan

(2012). Measuring ecological niche overlap from occurrence and spatial environmental data. Global Ecology and Biogeography, 21(4), 481–497. https://doi.org/10.1111/j.1466-8238.2011.00698.x

21.

Bustamante-Castillo

Hernández-Baños

B. E.

Arizmendi

M. C.

(2018). Hummingbird diversity and assemblage composition in a disturbed tropical dry forest of Guatemala. Tropical Conservation Science, 11, 1–15. https://doi.org/10.1177/1940082918793303

22.

Bustamante-Castillo

Hernández-Baños

B. E.

Arizmendi

M. C.

(2020). Hummingbird-plant visitation networks in agricultural and forested areas in a tropical dry forest region of Guatemala. Journal of Ornithology, 161, 189–201. https://doi.org/10.1007/s10336-019-01712-4

23.

Castoe

T. A.

Daza

J. M.

Smith

E. N.

Sasa

M. M.

Kuch

Campbell

J. A.

Chippindale

P. T.

Parkinson

C. L.

(2009). Comparative phylogeography of pitvipers suggests a consensus of ancient Middle American highland biogeography. Journal of Biogeography, 36(1), 88–103. https://doi.org/10.1111/j.1365-2699.2008.01991.x

24.

Chesser

R. T.

Billerman

S. M.

Burns

K. J.

Cicero

Dunn

J. L.

Kratter

A. W.

Lovette

I. J.

Mason

N. A.

Rasmussen

P. C.

Remsen

J. V.

Jr. Stotz

D. F.

Winker

(2020). Sixty-first Supplement to the American Ornithological Society’s Check-list of North American Birds. Auk, 137 , ukaa030. https://doi.org/10.1093/auk/ukaa030

25.

Clement

Posada

Crandall

K. A.

(2000). TCS: A computer program to estimate genealogies. Molecular Ecology, 9(10), 1657–1659. https://doi.org/10.1046/j.1365-294x.2000.01020.x

26.

Cobos

M. E.

Peterson

A. T.

Barve

Osorio-Olvera

(2019). kuenm: An R package for detailed development of ecological niche models using Maxent. PeerJ, 7, e6281. https://doi.org/10.7717/peerj.6281

27.

Corander

Sirén

Arjas

(2008). Bayesian spatial modeling of genetic population structure. Computer Statistics, 23, 111–129. https://doi.org/10.1007/s00180-007-0072-x

28.

Cortés-Rodríguez

Hernández-Baños

B. E.

Navarro-Sigüenza

A. G.

Omland

K. E.

(2008b). Geographic variation and genetic structure in the Streak-backed Oriole: Low mitochondrial differentiation reveals recent divergence. Condor, 110(4), 729–739. https://doi.org/10.1525/cond.2008.8578

29.

Cortés-Rodríguez

Hernández-Baños

B. E.

Navarro-Sigüenza

A. G.

Peterson

A. T.

García-Moreno

(2008a). Phylogeography and population genetics of the Amethyst-throated Hummingbird (Lampornis amethystinus). Molecular Phylogenetics and Evolution, 48(1), 1–11. https://doi.org/10.1016/j.ympev.2008.02.005.

30.

Cortés-Rodríguez

Jacobsen

Hernández-Baños

B. E.

Navarro-Sigüenza

A. G.

Peters

J. L.

Omland

K. E.

(2013). Coalescent analyses show isolation without migration in two closely related tropical orioles: The case of Icterus graduacauda and Icterus chrysater. Ecology and Evolution, 3(13), 4377–4387. https://doi.org/10.1002/ece3.768

31.

Díaz Infante

Lara

Arizmendi

M. C.

(2020). Temporal dynamics of the hummingbird-plant interaction network of a dry forest in Chamela, Mexico: A 30-year follow-up after two hurricanes. PeerJ, 8, e8338. https://doi.org/10.7717/peerj.8338

32.

Dickinson

E. C.

Remsen

J. V.

(2013). The Howard and Moore Complete Checklist of the Birds of the World . Vol. 1.

Non-passerines Press

. Eastbourne, UK: Aves.

33.

Di Cola

Broennimann

Petitpierre

Breiner

F. T.

D’Amen

Randin

Engler

Pottier

Pio

Dubuis

Pellissier

Mateo

R. G.

Hordijk

Salamin

Guisan

(2017). ecospat: An R package to support spatial analyses and modeling of species niches and distributions. Ecography, 40(6), 774–787. https://doi.org/10.1111/ecog.02671

34.

Dupanloup

Schneider

Excoffier

(2002). A simulated annealing approach to define the genetic structure of populations. Molecular Ecology, 11(12), 2571–2581. https://doi.org/10.1046/j.1365-294X.2002.01650.x

35.

Eberhard

J. R.

Bermingham

(2004). Phylogeny and biogeography of the Amazona ochrocephala (Aves: Psittacidae) complex. Auk, 121(2), 318–332. https://doi.org/10.1093/auk/121.2.318

36.

Espadas-Manrique

Durán

Argáez

(2003). Phytogeographic analysis of taxa endemic to the Yucatan Peninsula using geographic information systems, the domain heuristic method and parsimony analysis of endemicity. Diversity and Distributions, 9(4), 313–330. https://doi.org/10.1046/j.1472-4642.2003.00034.x

37.

Excoffier

(2004). Patterns of DNA sequence diversity and genetic structure after a range expansion: Lessons from the infinite-island model. Molecular Ecology, 13(4), 853–864. https://doi.org/10.1046/j.1365-294X.2003.02004.x

38.

Excoffier

Lischer

H. E. L.

(2010). Arlequin suite ver 3.5: A new series of programs to perform population genetics analyses under Linux and Windows. Molecular Ecology Resources, 10(3), 564–567. https://doi.org/10.1111/j.1755-0998.2010.02847.x

39.

Fielding

A. H.

Bell

J. F.

(1997). A review of methods for the assessment of prediction errors in conservation presence/absence models. Environmental Conservation, 24(1), 38–49. https://doi.org/10.1017/S0376892997000088

40.

Freeman

Sunnarborg

Peterson

A. T.

(2019). Effects of climate change on the distributional potential of three range-restricted West African bird species. Condor, 121, duz012. https://doi.org/10.1093/condor/duz012.

41.

Friedmann

Griscom

Moore

R. T.

(1950). Distributional check-list of the birds of Mexico. Part 1. Pacific Coast Avifauna, 29, 1–202

42.

Y. X.

(1997). Statistical neutrality of mutations against population growth, hitchhiking and background selection. Genetics, 147(2), 915–925. https://doi.org/10.1093/genetics/147.2.915

43.

Gill

Donsker

Rasmussen

(Eds.) (2020). IOC World List (v10.2). Accessed at:https://doi.org/10.14344/IOC.ML.10.2

44.

Gómez de Silva

Pérez Villafaña

M. G.

Cruz-Nieto

A. A.

(2020). Are some of the birds endemic to the Tres Marías Islands (Mexico) species? Bulletin of the British Ornithologists’ Club, 140(1), 7–37. https://doi.org/10.25226/bboc.v140i1

45.

González

Ornelas

J. F.

Gutiérrez-Rodríguez

(2011). Selection and geographic isolation influence hummingbird speciation: Genetic, acoustic and morphological divergence in the wedge-tailed sabrewing (Campylopterus curvipennis). BMC Evolutionary Biology, 11, 38. https://doi.org/10.1186/1471-2148-11-38

46.

González-Rubio

García-De León

F. J.

Rodríguez-Estrella

(2016). Phylogeography of endemic Xantus’ hummingbird (Hylocharis xantusii) shows a different history of vicariance in the Baja California Peninsula. Molecular Phylogenetics and Evolution, 102, 265–277. https://doi.org/10.1016/j.ympev.2016.05.039

47.

Graham

Dilcher

D. L.

(1995). The Cenozoic record of tropical dry forest in northern Latin America and the southern United States. In: Bullock

Mooney

Medina

(Eds) Seasonally Dry Tropical Forests. Cambridge University Press, Cambridge, pp. 124–145.

48.

Guevara-Chumacero

L. M.

López-Wilchis

Pedroche

F. F.

Juste

Ibáñez

Barriga-Sosa

I. D. L. A.

(2010). Molecular phylogeography of Pteronotus davyi (Chiroptera: Mormoopidae) in Mexico. Journal of Mammalogy, 91(1), 220–232. https://doi.org/10.1644/08-MAMM-A-212R3.1

49.

Gutiérrez-García

T. A.

Vázquez-Domínguez

(2012). Biogeographically dynamic genetic structure bridging two continents in the monotypic Central American rodent Ototylomys phyllotis. Biological Journal of the Linnean Society, 107(3), 593–610. https://doi.org/10.1111/j.1095-8312.2012.01966.x

50.

Gutiérrez-Rodríguez

Ornelas

J. F.

Rodríguez-Gómez

(2011). Chloroplast DNA phylogeography of a distylous shrub (Palicourea padifolia, Rubiaceae) reveals past fragmentation and demographic expansion in Mexican cloud forests. Molecular Phylogenetics and Evolution, 61(3), 603–615. https://doi.org/10.1016/j.ympev.2011.08.023

51.

Harpending

R. C.

(1994). Signature of ancient population growth in a low-resolution mitochondrial DNA mismatch distribution. Human Biology, 66(4), 591–600.

52.

Hackett

S. J

. (1996). Molecular phylogenetics and biogeography of tanagers in the genus Ramphocelus (Aves). Molecular Phylogenetics and Evolution, 5(2), 368–382. https://doi.org/10.1006/mpev.1996.0032

53.

Hernández-Soto

Licona-Vera

Lara

Ornelas

J. F.

(2018). Molecular and climate data reveal expansion and genetic differentiation of Mexican Violet-ear Colibri thalassinus thalassinus (Aves: Trochilidae) populations separated by the Isthmus of Tehuantepec. Journal of Ornithology, 159, 687–702. https://doi.org/10.1007/s10336-018-1540-5

54.

Hijmans

R.J.

Cameron

S. E.

Parra

J. L.

Jones

P. G.

Jarvis

(2005). Very high resolution interpolated climate surfaces for global land areas. International Journal of Climatology, 25(15), 1965–1978. https://doi.org/10.1002/joc.1276

55.

Howell

S. N. G.

Webb

(1995). A guide to the birds of Mexico and northern Central America. Oxford University Press, New York, New York.

56.

Jiménez

R. A.

Ornelas

J. F.

(2016). Historical and current introgression in a Mesoamerican hummingbird species complex: A biogeographic perspective. PeerJ, 4, e1556. https://doi.org/10.7717/peerj.1556

57.

Joly

Stevens

M. I.

van Vuuren

B. J.

(2007). Haplotype networks can be misleading in the presence of missing data. Systematic Biology, 56(5), 857–862. https://doi.org/10.1080/10635150701633153

58.

Lande

Engen

Sæther

B. E.

(2003). Stochastic population dynamics in ecology and conservation. Oxford University Press, Oxford, UK.

59.

Landis

J. R.

Koch

G. G.

(1977). The measurement of observer agreement for categorical data. Biometrics, 33(1), 159–174.

60.

Leigh

J. W.

Bryant

(2015). POPART: Full-feature software for haplotype network construction. Methods in Ecology and Evolution, 6(9), 1110–1116. https://doi.org/10.1111/2041-210X.12410

61.

León-Paniagua

Navarro-Sigüenza

A. G.

Hernández-Baños

B. E.

Morales

J. C.

(2007). Diversification of the arboreal mice of the genus Habromys (Rodentia: Cricetidae: Neotominae) in the Mesoamerican highlands. Molecular Phylogenetics and Evolution, 42(3), 653–664. https://doi.org/10.1016/j.ympev.2006.08.019

62.

Lerner

H. R. L.

Meyer

James

H. F.

Hofreiter

Fleischer

R. C.

(2011). Multilocus resolution of phylogeny and timescale in the extant adaptive radiation of Hawaiian honeycreepers. Current Biology, 21(21), 1838–1844. https://doi.org/10.1016/j.cub.2011.09.039

63.

Licona-Vera

Ornelas

J. F.

(2014). Genetic, ecological and morphological divergence between populations of the endangered Mexican Sheartail Hummingbird (Doricha eliza). PLoS ONE, 9, e101870. https://doi.org/10.1371/journal.pone.0101870

64.

Licona-Vera

Ornelas

J. F.

Wethington

Bryan

K. B.

(2018a). Pleistocene range expansions promote divergence with gene flow between migratory and sedentary populations of Calothorax hummingbirds. Biological Journal of the Linnean Society, 124(4), 645–667. https://doi.org/10.1093/biolinnean/bly084.

65.

Licona-Vera

Ortiz-Rodríguez

A. E.

Vásquez-Aguilar

A. A.

Ornelas

J. F.

(2018b). Lay mistletoes on the Yucatán Peninsula: Post-glacial expansion and genetic differentiation of Psittacanthus mayanus (Loranthaceae). Botanical Journal of the Linnean Society, 186(3), 334–360. https://doi.org/10.1093/botlinnean/box098

66.

Liu

White

Newell

(2013). Selecting thresholds for the prediction of species occurrence with presence-only data. Journal of Biogeography, 40(4), 778–789. https://doi.org/10.1111/jbi.12058

67.

Lobo

J. M.

Jiménez-Valverde

Real

(2008). AUC: A misleading measure of the performance of predictive distribution models. Global Ecology and Biogeography, 17(2), 145–151. https://doi.org/10.1111/j.1466-8238.2007.00358.x

68.

Macqueen

(2012). Last chance to see: The role of phylogeography in the preservation of tropical biodiversity. Tropical Conservation Science, 5(4), 417–425. https://doi.org/10.1177/194008291200500401

69.

Maldonado-Sánchez

Gutiérrez-Rodríguez

Ornelas

J. F.

(2016). Genetic divergence in the Common Bush-Tanager Chlorospingus ophthalmicus (Aves: Emberizidae) throughout Mexican cloud forests: The role of geography, ecology and Pleistocene climatic fluctuations. Molecular Phylogenetics and Evolution, 99, 76–88. https://doi.org/10.1016/j.ympev.2016.03.014

70.

Malpica

Ornelas

J. F.

(2014). Postglacial northward expansion and genetic differentiation between migratory and sedentary populations of the broad-tailed hummingbird (Selasphorus platycercus). Molecular Ecology, 23(2), 435–452. https://doi.org/10.1111/mec.12614

71.

Martínez de León

Castellanos-Morales

Moreno-Letelier

(2022). Incipient speciation, high genetic diversity, and ecological divergence in the alligator bark juniper suggest complex demographic changes during the Pleistocene. PeerJ, 10, e13802. https://doi.org/10.7717/peerj.13802

72.

McCormack

J. E.

Peterson

A. T.

Bonaccorso

Smith

T. B.

(2008). Speciation in the highlands of Mexico: Genetic and phenotypic divergence in the Mexican jay (Aphelocoma ultramarina). Molecular Ecology, 17(10), 2505–2521. https://doi.org/10.1111/j.1365-294X.2008.03776.x

73.

McGuire

J. A.

Witt

C. C.

Remsen

J. V.

Jr Corl

Rabosky

D. L.

Altshuler

D. L.

Dudley

(2014). Molecular phylogenetics and the diversification of hummingbirds. Current Biology, 24(8), 910–916. https://doi.org/10.1016/j.cub.2014.03.016

74.

Mendoza

A. M.

Bolívar-García

Vázquez-Domínguez

Ibáñez

Parra Olea

(2019). The role of Central American barriers in shaping the evolutionary history of the northernmost glassfrog, Hyalinobatrachium fleischmanni (Anura: Centrolenidae). PeerJ, 7, e6115. https://doi.org/10.7717/peerj.6115

75.

Merow

Smith

M. J.

Edwards

T. C.

Jr. Guisan

McMahon

S. M.

Normand

Thuiller

Wüest

R. O.

Zimmermann

N. E.

Elith

(2014). What do we gain from simplicity versus complexity in species distribution models? Ecography, 37(12), 1267–1281. https://doi.org/10.1111/ecog.00845

76.

Miller

M. J.

Lelevier

M. J.

Bermingham

Klicka

J. T.

Escalante

Winker

(2011). Phylogeography of the Rufous-tailed Hummingbird (Amazilia tzacatl). Condor, 113(4), 806–816. https://doi.org/10.1525/cond.2011.100226

77.

Morrone

J. J.

(2014). Biogeographical regionalisation of the Neotropical region. Zootaxa, 3782(1), 1–110. https://dx.doi.org/10.11646/zootaxa.3782.1.1

78.

Mulcahy

D. G.

Morrill

B. H.

Mendelson III

J. R.

(2006). Historical biogeography of lowland species of toads (Bufo) across the Trans-Mexican Volcanic Belt and the Isthmus of Tehuantepec. Journal of Biogeography, 33(11), 1889–1904. https://doi.org/10.1111/j.1365-2699.2006.01546.x.

79.

Naimi

(2015). usdm: Uncertainty analysis for species distribution models. R package version, 1.1.18.

80.

Navarro-Sigüenza

A. G.

Peterson

A. T.

Nyári

García-Deras

G. M.

García-Moreno

(2008). Phylogeography of the Buarremon brush-finch complex (Aves, Emberizidae) in Mesoamerica. Molecular Phylogenetics and Evolution, 47(1), 21–35. https://doi.org/10.1016/j.ympev.2007.11.030

81.

Olson

D. M.

Dinerstein

Wikramanayake

E. D.

Burgess

N. D.

Powell

G. V. N.

Underwood

E. C.

D’amico

J. A.

Itoua

Strand

H. E.

Morrison

J. C.

Loucks

C. J.

Allnutt

T. F.

Ricketts

T. H.

Kura

Lamoreux

J. F.

Wettengel

W. W.

Hedao

Kassem

K. R.

(2001). Terrestrial ecoregions of the world: A new map of life on EarthA new global map of terrestrial ecoregions provides an innovative tool for conserving biodiversity. BioScience, 51(11), 933–938. https://doi.org/10.1641/0006-3568(2001)051[0933:TEOTWA]2.0.CO;2

82.

Ornelas

J. F.

Rodríguez-Gómez

(2015). Influence of Pleistocene glacial/interglacial cycles on the genetic structure of the mistletoe cactus Rhipsalis baccifera (Cactaceae) in Mesoamerica. Journal of Heredity, 106(2), 196–210. https://doi.org/10.1093/jhered/esu113

83.

Ornelas

J. F.

Ruiz-Sánchez

Sosa

(2010). Phylogeography of Podocarpus matudae (Podocarpaceae): Pre-Quaternary relicts in northern Mesoamerican cloud forests. Journal of Biogeography, 37(12), 2384–2396. https://doi.org/10.1111/j.1365-2699.2010.02372.x

84.

Ornelas

J. F.

González

Espinosa de los Monteros

Rodríguez-Gómez

García-Feria

L. M.

(2014). In and out of Mesoamerica: Temporal divergence of Amazilia hummingbirds pre-dates the orthodox account of the completion of the Isthmus of Panama. Journal of Biogeography, 41(1), 168–181. https://doi.org/10.1111/jbi.12184

85.

Ornelas

J. F.

González

Hernández-Baños

B. E.

García-Moreno

(2016). Molecular and iridescent feather reflectance data reveal recent genetic diversification and phenotypic differentiation in a cloud forest hummingbird. Ecology and Evolution, 6(4), 1104–1127. https://doi.org/10.1002/ece3.1950

86.

Ornelas

J. F.

González de León

González

Licona-Vera

Ortiz-Rodríguez

A. E.

Rodríguez-Gómez

(2015). Comparative palaeodistribution of eight hummingbird species reveal a link between genetic diversity and Quaternary habitat and climate stability in Mexico. Folia Zoologica, 64(3), 246–259. https://doi.org/10.25225/fozo.v64.i3.a6.2015.

87.

Ornelas

J. F.

Sosa

Soltis

D. E.

Daza

J. M.

González

Soltis

P. S.

Gutiérrez-Rodríguez

Espinosa de los Monteros

Castoe

T. A.

Bell

Ruiz-Sánchez

(2013). Comparative phylogeographic analyses illustrate the complex evolutionary history of threatened cloud forests of northern Mesoamerica. PLoS ONE, 8, e56283. https://doi.org/10.1371/journal.pone.0056283

88.

Ortiz-Ramírez

M. F.

Andersen

M. J.

Zaldívar-Riverón

Ornelas

J. F.

Navarro-Sigüenza

A. G.

(2016). Geographic isolation drives divergence of uncorrelated genetic and song variation in the Ruddy-capped Nightingale-Thrush (Catharus frantzii; Aves: Turdidae). Molecular Phylogenetics and Evolution, 94, 74–86. https://doi.org/10.1016/j.ympev.2015.08.017

89.

Ortiz-Rodriguez

A. E.

Licona-Vera

Vásquez-Aguilar

A. A.

Hernández-Soto

López-Huicochea

E. A.

Ornelas

J. F.

(2020). Genetic differentiation among Psittacanthus rhynchanthus (Loranthaceae) populations: Novel phylogeographic patterns in the Mesoamerican tropical lowlands. Plant Systematics and Evolution, 306, 10. https://doi.org/10.1007/s00606-020-01638-y

90.

Osorio-Olvera

Lira-Noriega

Soberón

Peterson

A. T.

Falconi

Contreras-Díaz

R. G.

Martínez-Meyer

Barve

(2020). ntbox: An r package with graphical user interface for modelling and evaluating multidimensional ecological niches. Methods in Ecology and Evolution, 11(10), 1199–1206. https://doi.org/10.1111/2041-210X.13452

91.

Oyama

Martínez-Ramos

Peñaloza-Ramírez

J. M.

Rocha-Ramírez

Armenta-Medina

E. G.

Hernández-Soto

(2016). Population genetic structure of an extremely logged tree species Guaiacum sanctum L. in the Yucatan Peninsula, Mexico. Botanical Sciences, 94(2), 345–356. https://doi.org/10.17129/botsci.278

92.

Paradis

(2010). Pegas: An R package for population genetics with an integrated-modular approach. Bioinformatics, 26(3), 419–420. https://doi.org/10.1093/bioinformatics/btp696

93.

Pebesma

Bivand

R. S.

(2005). S classes and methods for spatial data: The sp package. R news, 5(2), 9–13.

94.

Pennington

R. T.

Lavin

Prado

D. E.

Pendry

C. A.

Pell

S. K.

Butterworth

C. A.

(2004). Historical climate change and speciation: Neotropical seasonally dry forest plants show patterns of both tertiary and quaternary diversification. Philosophical Transactions of the Royal Society B, 359(1443), 515–537. https://doi.org/10.1098/rstb.2003.1435

95.

Pennington

R. T.

Klitgaard

B. B.

Ireland

Lavin

(2000). New insights into floral evolution of basal Papilionoideae from molecular phylogenies. In: Herendeen

Bruneau

(Eds) Advances in Legume Systematics, Part 9. Royal Botanic Gardens, Kew, UK, pp 233–248.

96.

Peterson

A. T.

Papeş

Soberón

(2008). Rethinking receiver operating characteristic analysis applications in ecological niche modeling. Ecological Modelling, 213(1), 63–72. https://doi.org/10.1016/j.ecolmodel.2007.11.008

97.

Pfenninger

Posada

(2002). Phylogeographic history of the land snail Candidula unifasciata (Helicellinae, Stylommatophora): Fragmentation, corridor migration, and secondary contact. Evolution, 56(9), 1776–1788. https://doi.org/10.1111/j.0014-3820.2002.tb00191.x

98.

Phillips

S. J.

Anderson

R. P.

Schapire

R. E.

(2006). Maximum entropy modeling of species geographic distributions. Ecological Modelling, 190(3–4):231–259. https://doi.org/10.1016/j.ecolmodel.2005.03.026

99.

Pons

Petit

R. J.

(1996). Measuring and testing genetic differentiation with ordered versus unordered alleles. Genetics, 144(3), 1237–1245. https://doi.org/10.1093/genetics/144.3.1237

100.

Portillo-Quintero

C. A.

Sánchez-Azofeifa

G. A.

(2010). Extent and conservation of tropical dry forests in the Americas. Biological Conservation, 143(1), 144–155. https://doi.org/10.1016/j.biocon.2009.09.020

101.

Prieto-Torres

D. A.

Navarro-Sigüenza

A. G.

Santiago-Alarcón

Rojas-Soto

O. R.

(2016). Response of the endangered tropical dry forests to climate change and the role of Mexican Protected Areas for their conservation. Global Change Biology, 22(1), 364–379. https://doi.org/10.1111/gcb.13090

102.

Prieto-Torres

D. A.

Rojas-Soto

O. R.

Bonaccorso

Santiago-Alarcón

Navarro-Sigüenza

A. G.

(2019). Distributional patterns of Neotropical seasonally dry forest birds: A biogeographical regionalization. Cladistics, 35(4), 446–460. https://doi.org/10.1111/cla.12366

103.

R Core Team . (2020). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Viena, Austria. https://www.R-project.org/

104.

Ramírez-Barahona

Torres-Miranda

Palacios-Ríos

Luna-Vega

(2009). Historical biogeography of the Yucatan Peninsula, Mexico: A perspective from ferns (Monilophyta) and lycopods (Lycophyta). Biological Journal of the Linnean Society, 98(4), 775–786. https://doi.org/10.1111/j.1095-8312.2009.01331.x

105.

Ramos-Onsins

S. E.

Rozas

(2002). Statistical properties of new neutrality tests against population growth. Molecular Biology and Evolution, 19(12), 2092–2100. https://doi.org/10.1093/molbev/msl052

106.

Ranjitkar

Sujakhu

N. M.

Wang

Mortimer

P. E.

Kindt

Zomer

R. J.

(2016). Climate modelling for agroforestry species selection in Yunnan Province, China. Environmental Modelling & Software, 75, 263–272. https://doi.org/10.1016/j.envsoft.2015.10.027

107.

Rödder

Engler

J. O.

(2011). Quantitative metrics of overlaps in Grinnellian niches, advances and possible drawbacks. Global Ecology and Biogeography, 20(6), 915–927. https://doi.org/10.1111/j.1466-8238.2011.00659.x

108.

Rodríguez-Gómez

Ornelas

J. F.

(2014). Genetic divergence of the Mesoamerican azure-crowned hummingbird (Amazilia cyanocephala) across the Motagua-Polochic-Jocotán fault system. Journal of Zoological Systematics and Evolutionary Research, 52(2), 142–153. https://doi.org/10.1111/jzs.12047

109.

Rodríguez-Gómez

Ornelas

J. F.

(2015). At the passing gate: Past introgression in the process of species formation between Amazilia violiceps and A. viridifrons hummingbirds along the Mexican Transition Zone. Journal of Biogeography, 42(7), 1305–1318. https://doi.org/10.1111/jbi.12506

110.

Rodríguez-Gómez

Ornelas

J. F.

(2018). Genetic structuring and secondary contact in the white-chested Amazilia hummingbird species complex. Journal of Avian Biology, 49, e01536. https://doi.org/10.1111/jav.01536

111.

Rodríguez-Gómez

Gutiérrez-Rodríguez

Ornelas

J. F.

(2013). Genetic, phenotypic and ecological divergence with gene flow at the Isthmus of Tehuantepec: The case of the azure-crowned hummingbird (Amazilia cyanocephala). Journal of Biogeography, 40(7), 1360–1373. https://doi.org/10.1111/jbi.12093

112.

Rodríguez-Gómez

Licona-Vera

Silva-Cárdenas

Ornelas

J. F.

(2021). Phylogeography, morphology and ecological niche modelling to explore the evolutionary history of Azure-crowned Hummingbird (Amazilia cyanocephala, Trochilidae) in Mesoamerica. Journal of Ornithology, 162, 529–547. https://doi.org/10.1007/s10336-020-01853-x

113.

Rogers

A. R.

(1995). Genetic evidence for a Pleistocene population explosion. Evolution, 49(4), 608–615. https://doi.org/10.1111/j.1558-5646.1995.tb02297.x

114.

Rogers

A. R.

Harpending

(1992). Population growth makes waves in the distribution of pairwise genetic differences. Molecular Biology and Evolution, 9(3), 552–569. https://doi.org/10.1093/oxfordjournals.molbev.a040727

115.

Ruiz-Gutiérrez

Doherty

P. F.

Santana

Contreras Martínez

Schondube

Verdugo Munguía

Iñigo-Elias

(2012). Survival of resident Neotropical birds: Considerations for sampling and analysis based on 20 years of bird-banding efforts in Mexico. Auk, 129(3), 500–509. https://doi.org/10.1525/auk.2012.11171

116.

Rzedowski

(1978). Vegetación de México. México: Limusa, Noriega Editores.

117.

Schneider

Excoffier

(1999). Estimation of demographic parameters from the distribution of pairwise differences when the mutation rates vary among sites: Application to human mitochondrial DNA. Genetics, 152(3), 1079–1089. https://doi.org/10.1093/genetics/152.3.1079

118.

Schoener

T. W.

(1970). Nonsynchronous spatial overlap of lizards in patchy habitats. Ecology, 51(3), 408–418. https://doi.org/10.2307/1935376

119.

Schuchmann

K-L.

(1999). Family Trochilidae, hummingbirds. In: del Hoyo

Elliott

Sargatal

(Eds) Handbook of the birds of the world. Volume 5. Lynx Edicions, Barcelona, pp. 468–680.

120.

Soberón

Nakamura

(2009). Niches and distributional areas: Concepts, methods, and assumptions. Proceedings of the National Academy of Sciences USA, 106(suppl. 2), 19644–19650. https://doi.org/10.1073/pnas.0901637106

121.

Soberón

Peterson

A. T.

(2005). Interpretation of models of fundamental ecological niches and species’ distributional areas. Biodiversity Informatics, 2, 1–10. https://doi.org/10.17161/bi.v2i0.4

122.

Sorenson

M. D.

Ast

J. C.

Dimcheff

D. E.

Yuri

Mindell

D. P.

(1999). Primers for a PCR-based approach to mitochondrial genome sequencing in birds and other vertebrates. Molecular Phylogenetics and Evolution 12(2), 105–114. https://doi.org/10.1006/mpev.1998.0602

123.

Strubbe

Beauchard

Matthysen

(2015). Niche conservatism among nonnative vertebrates in Europe and North America. Ecography, 38(3), 321–329. https://doi.org/10.1111/ecog.00632

124.

Tajima

(1989). Statistical-method for testing the neutral mutation hypothesis by DNA polymorphism. Genetics, 123(3), 585–595. https://doi.org/10.1093/genetics/123.3.585

125.

Thomson

D. R.

Stevens

F. R.

Ruktanonchai

N. W.

Tatem

A. J.

Castro

M. C.

(2017). GridSample: An R package to generate household survey primary sampling units (PSUs) from gridded population data. International Journal of Health Geographics, 16, 1–19. https://doi.org/10.1186/s12942-017-0098-4

126.

Toledo

V. M.

(1981). Pleistocene changes of vegetation in tropical México. In: Prance

(Ed) Biological diversification in the tropics. Columbia University Press, New York, NY, USA, pp. 93–111

127.

Trejo

Dirzo

(2000). Deforestation of seasonally dry tropical forest: A national and local analysis in Mexico. Biological Conservation, 94(2), 133–142. https://doi.org/10.1016/S0006-3207(99)00188-3

128.

Van Els

Spellman

G. M.

Smith

B. T.

Klicka

(2014). Extensive gene flow characterizes the phylogeography of a North American migrant bird: Black-headed Grosbeak (Pheucticus melanocephalus). Molecular Phylogenetics and Evolution, 78, 148–159. https://doi.org/10.1016/j.ympev.2014.04.028

129.

Van Rossem

A. J.

(1938). A northwest race of the Cinnamon Hummingbird. Condor, 40(5), 226–227. https://doi.org/10.1093/condor/40.5.226a

130.

Vásquez-Aguilar

A. A.

Ornelas

J. F.

Rodríguez-Gómez

MacSwiney

G. M. C.

(2021). Modeling future potential distribution of Buff-bellied Hummingbird (Amazilia yucatanensis) under climate change: species vs. subspecies. Tropical Conservation Science, 14, 1–18. https://doi.org/10.1177/19400829211030834

131.

Vázquez-Domínguez

Arita

H. T.

(2010). The Yucatan peninsula: Biogeographical history 65 million years in the making. Ecography, 33(2), 212–219. https://doi.org/10.1111/j.1600-0587.2009.06293.x

132.

Vázquez-López

Cortés-Rodríguez

Robles-Bello

S. M.

Bueno-Hernández

Zamudio-Beltrán

L. E.

Ruegg

Hernández-Baños

B. E.

(2021). Phylogeography and morphometric variation in the Cinnamon Hummingbird complex: Amazilia rutila (Aves: Trochilidae). Avian Research, 12, 61. https://doi.org/10.1186/s40657-021-00295-0.

133.

Vázquez-Miranda

Navarro-Sigüenza

A. G.

Omland

K. E.

(2009). Phylogeography of the Rufous-naped wren (Campylorhynchus rufinucha): Speciation and hybridization in Mesoamerica. Auk, 126(4), 765–778. https://doi.org/10.1525/auk.2009.07048

134.

Weir

J. T.

(2006). Divergent timing and patterns of species accumulation in lowland and highland Neotropical birds. Evolution, 60(4), 842–855. https://doi.org/10.1111/j.0014-3820.2006.tb01161.x

135.

Weller

(1999). Cinnamon Hummingbird. In: del Hoyo

Elliott

Sargatal

(Ed) Handbook of the birds of the world. Barn-Owls to hummingbirds. Volume 5. Lynx Edicions, Barcelona, Spain, pp. 596–597.

136.

Wiens

J. J.

Graham

C. H.

(2005). Niche conservatism: Integrating evolution, ecology, and conservation biology. Annual Review in Ecology, Evolution, and Systematics, 36, 519–539. https://doi.org/10.1146/annurev.ecolsys.36.102803.095431

137.

Williford

Deyoung

R. W.

Honeycutt

R. L.

Brennan

L. A.

Hernández

(2016). Phylogeography of the Bobwhite (Colinus) Quails. Wildlife Monographs, 193(1), 1–49. https://doi.org/10.1002/wmon.1017

138.

Zamudio-Beltrán

L. E.

Hernández-Baños

B. E.

(2018). Genetic and morphometric divergence in the Garnet-Throated Hummingbird Lamprolaima rhami (Aves: Trochilidae). PeerJ, 6, e5733. https://doi.org/10.7717/peerj.5733

139.

Zamudio-Beltrán

L. E.

Licona-Vera

Hernández-Baños

B. E.

Klicka

Ornelas

J. F.

(2020b). Phylogeography of the widespread white-eared hummingbird (Hylocharis leucotis): Preglacial expansion and genetic differentiation of populations separated by the Isthmus of Tehuantepec. Biological Journal of the Linnean Society, 130(2), 247–267. https://doi.org/10.1093/biolinnean/blaa043.

140.

Zamudio-Beltrán

L. E.

Ornelas

J. F.

Malpica

Hernández-Baños

B. E.

(2020a). Genetic and morphological differentiation among populations of the Rivoli’s Hummingbird (Eugenes fulgens) species complex (Aves: Trochilidae). Auk, 137 , ukaa032. https://doi.org/10.1093/auk/ukaa032

141.

Zarza

Connors

E. M.

Maley

J. M.

Tsai

W. L. E.

Heimes

Kaplan

McCormack

J. E.

(2018). Combining ultraconserved elements and mtDNA data to uncover lineage diversity in a Mexican highland frog (Sarcohyla; Hylidae). PeerJ, 6, e6045. https://doi.org/10.7717/peerj.6045

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.74 MB

0.31 MB

1.39 MB

0.07 MB

0.04 MB