Skip to main content
NCBI home page
Ecology and Evolution logoLink to Ecology and Evolution
. 2018 Jun 25;8(14):7131–7142. doi: 10.1002/ece3.4237

Gene flow and Andean uplift shape the diversification of Gasteracantha cancriformis (Araneae: Araneidae) in Northern South America

Fabian C Salgado‐Roa 1,2, Carolina Pardo‐Diaz 1, Eloisa Lasso 2,3, Carlos F Arias 3, Vera Nisaka Solferini 4, Camilo Salazar 1,✉
PMCID: PMC6065347  PMID: 30073072

Abstract

The Andean uplift has played a major role in shaping the current Neotropical biodiversity. However, in arthropods other than butterflies, little is known about how this geographic barrier has impacted species historical diversification. Here, we examined the phylogeography of the widespread color polymorphic spider Gasteracantha cancriformis to evaluate the effect of the northern Andean uplift on its divergence and assess whether its diversification occurred in the presence of gene flow. We inferred phylogenetic relationships and divergence times in G. cancriformis using mitochondrial and nuclear data from 105 individuals in northern South America. Genetic diversity, divergence, and population structure were quantified. We also compared multiple demographic scenarios for this species using a model‐based approach (phrapl) to determine divergence with or without gene flow. At last, we evaluated the association between genetic variation and color polymorphism. Both nuclear and mitochondrial data supported two well‐differentiated clades, which correspond to populations occurring on opposite sides of the Eastern cordillera of the Colombian Andes. The final uplift of this cordillera was identified as the most likely force that shaped the diversification of G. cancriformis in northern South America, resulting in a cis‐ and trans‐Andean phylogeographic structure for the species. We also found shared genetic variation between the cis‐ and trans‐Andean clades, which is better explained by a scenario of historical divergence in the face of gene flow. This has been likely facilitated by the presence of low‐elevation passes across the Eastern Colombian cordillera. Our work constitutes the first example in which the Andean uplift coupled with gene flow influenced the evolutionary history of an arachnid lineage.

Keywords: Andean passes, Andean uplift, Colombian Andes, Gasteracantha cancriformis, gene flow

1. INTRODUCTION

The northern Andes in South America is one of the most biodiverse regions on the planet, and the origins of this rich diversity have been linked to past geological and climatic events such as the uplift of the Andes and quaternary climatic oscillations (Kattan, Franco, & Rojas, 2004). The effect of these geoclimatic events in promoting divergence between Neotropical populations and species can be elucidated using genetic data, especially by detecting deviations from the expected coalescent patterns in neutral loci (Rull, 2011). Most studies addressing this question have identified the uplift of the northern Andes as a major driver of Neotropical diversification in a scenario consistent with allopatric differentiation, wherein the complex topography of the Andes isolated populations on both sides of this barrier thus restricting gene flow (Antonelli et al., 2018; Brumfield & Capparella, 1996; Hoorn et al., 2010). In contrast with this predominant view, a recent comparative phylogeographic study found discordant divergence times for multiple avian lineages with cross‐Andean distribution, a result that is better explained by dispersal ability across the Andes rather than a single vicariant event (Smith, Harvey, Faircloth, Glenn, & Brumfield, 2014; Smith, McCormack et al., 2014). In line with this finding, new evidence supports the notion that common diversification modes in Neotropical birds include secondary contact between cross‐Andean populations or between divergence in the presence of gene flow, facilitated by low‐elevation corridors along the Andes (Cadena, Pedraza, & Brumfield, 2016; Oswald, Overcast, Mauck, Andersen, & Smith, 2017).

However, our current knowledge on the modes of animal diversification in the northern Andes is mostly based on vertebrates, and despite arthropods being the most diverse group of animals, analyses of their diversification in this region remain scarce (De‐silva et al., 2017; Turchetto‐Zolet, Pinheiro, Salgueiro, & Palma‐Silva, 2013). Some studies limited to insects, especially butterflies, show that the Andean mountains played an important role in their diversification, where speciation with and without gene flow across the Andes has occurred (Arias et al., 2012; Chazot et al., 2016, 2017; De‐silva et al., 2017; Díaz et al., 2014; Dick, Roubik, Gruber, & Bermingham, 2004; Elias et al., 2009). Despite this, a comprehensive understanding on how the Andean orogeny has promoted Neotropical animal diversification requires the inclusion of additional arthropod taxa, such as arachnids.

Gasteracantha cancriformis (Linnaeus, 1758) is a common Neotropical orb‐web spider that exhibits sexual dimorphism, with females being up to four times larger than males (5 to 13 mm vs 2 to 3 mm, respectively; Muma, 1971). Furthermore, while males apparently spin no definite web, females build a stronger net with reinforced guylines, distinct tufts of silk on the foundation lines and a definitive central disk (Muma, 1971). This orb‐web varies in size and position, and is used to capture prey such as whiteflies, flies, moths, and beetles (Muma, 1971). The species is also characterized by its color polymorphism, with at least eight known morphs (Gawryszewski, 2007), but the causes maintaining this variation remain unknown (Gawryszewski & Motta, 2012).

Gasteracantha cancriformis has a wide distribution in the Neotropical and subtropical region (from the south of the United States to northern Argentina; Levi, 1978) and consequently occurs on both sides of the Andes and in the Colombian inter‐Andean valleys. This distribution makes it an excellent model to test whether the uplift of the Andes has influenced its diversification at the population level. For example, the Andean uplift may have fragmented populations at opposite sides, thus limiting genetic connectivity; this scenario predicts reciprocally monophyletic clades at each flank of the Andes. Otherwise, long‐distance migration of individuals across dispersal corridors in the Andes could favor gene flow among areas; this scenario of divergence with gene flow predicts monophyletic groups at opposite sides of the Andes sharing genetic variation between them. At last, rampant migration of individuals across the Andes could promote genetic homogenization of populations in spite of the geographic barrier; in this scenario, it is expected to find population clustering not explained by geography.

Here, we implemented a multilocus approach to study the genetic connectivity between polymorphic populations of G. cancriformis across the northern Andes (Figure 1, Supporting information Table S1) and tested scenarios of strict vicariance versus divergence with gene flow. We also evaluated whether lineage clustering in this spider is explained by geography or color pattern. Overall, this work contributes to deepen our understanding of how Andean orogeny has shaped processes of biotic diversification and biogeography in the Neotropics.

Figure 1.

Figure 1

Open in a new tab

Map of the Neotropics showing the 18 sampling localities in Colombia and Brazil. White squares correspond to each locality sampled. Sampled localities are (a) Acre, (b) Praia do forte, (c) Lencois, (d) Villavicencio, (e) Buenavista, (f) Ibague, (g) Palomino, (h) Cali, (i) Boquia, (j) Bahia Malaga, (k) Bucaramanga, (l) Palmira, (m) Tolu, (n) Armero, (o) Leticia, (p) Cocalito, (q) Medellin, and (r) Campinas

2. MATERIALS AND METHODS

2.1. Sample collection

We used standard aerial searching and beating methods to sample a total of 105 individuals of G. cancriformis from 17 localities distributed from the north of Colombia to southeastern Brazil (Figure 1, Supporting information Table S1). Specimens were color coded following the categorization established by Gawryszewski (2007), preserved in a 20% dimethylsulfoxide (DMSO) solution saturated with NaCl, and stored at −80°C. Colombian samples were deposited in the “Colección de Artrópodos de la Universidad del Rosario” (CAUR#229), and Brazilian specimens were deposited in the “Coleção Científica de Aracnídeos e Miriápodes of the Instituto Butantan” (São Paulo, Brazil).

2.2. DNA extraction, amplification, and sequencing

DNA was extracted from legs using the Wizard Genomic DNA purification kit (Promega), following the manufacturer's protocol. We amplified two mitochondrial genes, the cytochrome oxidase I (587 pb; COI; Folmer, Black, Hoeh, Lutz, & Vrijenhoek, 1994) and the 16S ribosomal DNA gene (397 pb; 16s; Simon et al., 1994), using the conditions reported in previous studies (Blackledge et al., 2009; McHugh, Yablonsky, Binford, & Agnarsson, 2014; Peres et al., 2015). We also sequenced two nuclear loci, the internal transcribed spacer 2 (296 pb; ITS; White, Bruns, Lee, & Taylor, 1990) and 28S ribosomal DNA gene (681 pb; 28s; Hedin & Maddison, 2001), using primers previously developed for spiders (Moradmand, Schönhofer, & Jäger, 2014; Peres et al., 2015). At last, we used the published transcriptome of G. cancriformis (Prosdocimi et al., 2011) and designed the forward primer (5′‐ CAATTACACCTGGGAATCTTCTG‐3′) and reverse primer (5′‐CCCTGACAAAATTCAAATAGTCG‐3′) to amplify a 560‐bp fragment of the heat shock protein 90 (HSP90), a gene that was used previously in phylogenetic studies in Heliconius (Salazar et al., 2010). PCR reactions for this marker were performed in a 10 μl reaction volume containing 1× PCR buffer, 2.5 mM MgCl2, 500 μM dNTPs, 0.5 μM each primer, 0.5 U Platinum Taq (INVITROGEN), and 30–40 ng of DNA. The PCR cycling profile was 94°C for 5 min, 40 cycles at 94°C for 30 s, Tm 54°C for 30 s, and 72°C for 1 min and final extension at 72°C for 10 min. For all loci, we visualized 1 μL of the PCR reaction in a 1.5% agarose gel to verify the success of PCR. Positive amplicons were cleaned by ExoSAP‐IT (USB Corp., Cleveland, OH), and their sequencing was carried out by ELIM Biopharmaceuticals Inc. (Hayward, CA).

Gene sequences were read, edited, and assembled with CLC main workbench (QIAGEN) to obtain a consensus sequence per individual. Haplotype inference for heterozygous calls was conducted for nuclear loci with the PHASE algorithm implemented in dnaSP v5.10 (Librado & Rozas, 2009) with 5000 iterations per simulation and accepting inferred haplotypes with a confidence higher than 90%. Then, we used mega 6.0 (Tamura, Stecher, Peterson, Filipski, & Kumar, 2013) using the clustal W algorithm (Thompson, Higgins, & Gibson, 1994) to create alignments for each locus; these alignments were visually inspected and corrected accordingly. Alignments were translated to protein and verified for stop codons using mesquite v3.04 (Maddison & Maddison, 2015).

2.3. Molecular phylogenetics and divergence times

The nucleotide substitution model for each mitochondrial gene was selected using the BIC criterion in jmodeltest 0.1.1 (Posada, 2008). The most suitable models were HKY+I for COI and TIM+I for 16S.

Tree topologies were estimated with Bayesian inference (BI) using beast 1.7.4 (Drummond, Suchard, Xie, & Rambaut, 2012) and including two Micrathena vigorsi individuals as outgroups (Supporting information Table S1). We unlinked the substitution model for each gene and linked the clock model and the tree model. We applied a lognormal relaxed clock to estimate divergence times using a mutation rate of 0.0112 (SD = 0.001) substitution/site/million years previously reported for node dating and calibration in spiders (Bartoleti et al., 2017; Bidegaray‐Batista & Arnedo, 2011; Kuntner, Arnedo, Trontelj, Lokovšek, & Agnarsson, 2013). We ran two runs of 100 million generations sampling every 1000 generations. The initial 10,000 trees were discarded as burn‐in using treeannotator (Drummond & Bouckaert, 2015). We examined the output in tracer 1.7 (Rambaut, Drummond, Xie, Baele & Suchard, 2018) to confirm that all effective sample size (ESS) values were >200 and to confirm the convergence of the chains to the stationary distribution. The maximum credibility tree that best represented the posterior distribution was visualized and edited with figtree 1.4 (Rambaut, 2012).

Phylogenetic reconstruction was also performed with maximum likelihood (ML) in IQ‐tree (Nguyen, Schmidt, Von Haeseler, & Minh, 2015) using the same substitution models described before and applying the edge‐proportional partition. Node support was assessed with 1000 bootstrap replicates. We also constructed haplotype median‐joining networks per locus with popart (Leigh, Bryant, & Nakagawa, 2015).

2.4. Population genetics

We calculated haplotype diversity (h), nuclear diversity (π), number of segregating sites (ss), and Tajima's D with dnaSP v5.10 (Librado & Rozas, 2009). Genetic structure was evaluated using F ST at two different levels: among phylogenetic clades (i.e., populations occurring in opposite sides of the Eastern cordillera of the Colombian Andes) and among populations. Significance of deviations from panmixia was assessed with the Hudson permutation test (Hudson, Boos, & Kaplan, 1992). An analysis of molecular variance (AMOVA) was also calculated for the same levels of differentiation with arlequin 3.5 (Excoffier & Lischer, 2010) using 10,000 permutations.

Using the nuclear dataset, we identified the number of population genetic clusters (K) with the Bayesian clustering approach implemented in structure 2.3.4 (Pritchard, Stephens, & Donnelly, 2000). We ran the analysis under the admixture model with a 50,000 burn‐in, 100,000 MCMC sampling generations for K ranging from 1 to 13 (localities with only one individual were removed from this analysis), and 20 iterations for each value of K. We determined the best K by applying three complementary approaches as recommended by Janes et al. (2017): (i) according to the Δ K method of Evanno (Evanno, Regnaut, & Goudet, 2005), (ii) plotting the likelihood of K for each value of K (Earl & vonHoldt, 2012), and (iii) reporting multiple barplots for K values between 2 and 5. All these tests were implemented in clumpak (Kopelman, Mayzel, Jakobsson, Rosenberg, & Mayrose, 2015). We ran a multivariate analysis in order to make an additional validation of the genetic clusters for each locus. We did this by transforming fasta sequences into a genind object and loading it into the adegenet R package (Jombart & Ahmed, 2011), in which we performed a principal component analysis (PCA). We retained the first two components for a subsequent canonical discriminant analysis using the R package candisc (Friendly & Fox, 2017).

As isolation by distance (IBD) can obscure population structure signals, we investigated the presence of IBD for each locus using the Mantel (Mantel, 1967) with the R package vegan (Dixon, 2003). In order to do this, pairwise geographic distances among localities were calculated with the function distm from the R package geosphere (Hijmans, 2016), while genetic distances were estimated by linearizing the F ST values previously obtained. We also implemented a partial Mantel test (Smouse, Long, & Sokal, 1986) to separate the effect of geographic distance from the population assignments based on structure results.

Considering the recently highlighted limitations of the Mantel test (Legendre & Fortin, 2010; Legendre, Fortin, & Borcard, 2015; Meirmans, 2012), we also tested linear correlations between the logarithm of the geographic distances and genetic distances as recommended by Legendre and Fortin (2010) and Diniz‐filho et al. (2013).

2.5. Demographic model testing and parameter estimation

We used phrapl (Jackson, Morales, Carstens, & O'Meara, 2017) to choose a demographic model that fits our data. phrapl compares the topologies obtained from empirical data with those simulated under multiple demographic models and then, by calculating the proportion of times that simulated topologies match the empirical ones, it approximates the log‐likelihood of the data under a given model. phrapl calculates Akaike information criterion (AIC) as a measure of lack of model fit while the associated AIC weights (wAIC) can be interpreted as model probabilities.

phrapl can compare various demographic models, which can be broadly categorized as (a) isolation only (IO), (b) migration only (MO), (c) isolation with migration (IM), and (d) mixed models (ME). We tested six models that fall into the IO and IM categories. The first is an IO model with a single coalescent event and no migration, while the other five are IM models that assume constant gene flow along the branch length and differ in the direction and strength of migration.

As input for phrapl, we assigned each individual to its collection site (east or west of the Eastern cordillera) and loaded five midpoint rooted trees estimated with IQ‐tree (one per locus; Nguyen et al., 2015). We used jmodeltest (Posada, 2008) to select the most plausible pattern of sequence evolution for each gene (COI: HKY+I, 16S: TIM+I, ITS: K80 + I, 28S: F81 + I, HSP90: GTR+I).

We subsampled four tips per group and 100 subsamples per gene, giving a total of 500 observed trees. Then, following previous studies (Denton, Morales, & Gibbs, 2018; Hime et al., 2016; Morales, Jackson, Dewey, Meara, & Bryan, 2017), we conducted a simulation of 100,000 gene trees using a grid of parameter values for divergence time (t = 0.30, 0.58, 1.40, 2.54, 4.1) and migration (m = 0.10, 0.22, 0.46, 1.00, 2.15), in units of 4 N and 4 Nm, respectively. This array was designed to encompass the full range of potential values in each dataset as such information is not available yet for G. cancriformis. When we detected gene flow in our dataset, we tested two additional models that correspond to recent (τ‐9τ/10) and ancient secondary contact (τ‐τ/5), starting from the tips.

Using the model selected by phrapl, we estimated population sizes (θ), divergence time (τ), and migration rates (m) with the Bayesian coalescent program G‐phoCS (Gronau, Hubisz, Gulko, Danko, & Siepel, 2011). We ran independent analyses for mtDNA and nDNA, each with 50,000 burn‐in iterations and 500,000 additional sampling iterations; runs showed adequate mixing and convergence under these conditions. As G‐phoCS uses gamma distributions (α, β) to specify the prior distributions for the parameters, we tested deep (α = 1 and β = 30) and shallow (α = 1 and β = 300) divergence, both with migration rate priors set to α = 1 and β = 10. As results for both were very similar, we are presenting only those of β = 30. For the mitochondrial analysis, we converted raw parameter estimates using the mitochondrial rate of 0.0112 substitutions/site/million years and the nuclear rate of 0.0084 substitutions/site/million years (Bidegaray‐Batista & Arnedo, 2011).

2.6. Phenotype by genotype association

To test whether there is an association between the coloration of individuals and their genetic variation, we ran a chi‐square Monte Carlo test under the null hypothesis of independence between coloration and genetic haplotypes. This analysis was run for each locus and establishing color morphs categories following the categories of Gawryszewski (2007). The analysis included only Colombian specimens, due to the fact that Brazilian samples do not have color records.

3. RESULTS

3.1. Phylogenetic relationships and divergence time

Both BI and ML showed the same phylogenetic pattern which revealed two highly supported clades, but with their internal relationships unresolved (Figure 2). These major clades correspond to populations at the eastern (eEC) and western (wEC) sides of the Eastern Cordillera of the Colombian Andes (EC), with some individuals from the foothills of the EC showing shared haplotypes between both clades (Figure 2). Other physiographic features of the Neotropics, such as the Western and Central Cordilleras of the Colombian Andes or the Brazilian Dry Diagonal, do not cause population structure in G. cancriformis. Therefore, Brazilian samples were monophyletic within the eEC clade (Figure 2). beast estimated a mitochondrial divergence time for the two main clades at ca. 2.13 Ma (95% HPD = 0.98–3.93 Ma; Supporting information Figure S1). This estimate largely coincides to that of G‐phoCS using mitochondrial data, where divergence time was 1.68 Ma (95% HPD = 0.66–2.53 Ma; Supporting information Figure S1 and Table S2). Both of these dates are very close to the Pliocene/Pleistocene boundary and concordant with the final EC uplift (Gregory‐Wodzicki, 2000). When using nuclear data, G‐phoCS estimated a divergence time of 0.66 Ma (95% HPD = 0.37–0.96 Ma; Supporting information Figure S1).

Figure 2.

Figure 2

Open in a new tab

Mitochondrial phylogeny. Best recovered tree with mtDNA where node supports are represented by circles divided into two: The upper half corresponds to posterior probabilities obtained by Bayesian inference, and the lower half to the maximum‐likelihood bootstrap values after 1,000 bootstrap pseudoreplicates. Colored circles at the tips represent the color phenotype in the opisthosoma of each individual. Green and red squares highlight the eastern (eEC) and western (wEC) sides of the Eastern Cordillera of the Colombian Andes. Green branches highlight individuals sampled in the eEC that cluster into the wEC clade

3.2. Population genetics

Mitochondrial and nuclear median‐joining networks revealed two groups with shared haplotypes, corresponding to the eEC and wEC clades (Supporting information Figure S2). These clades are separated by 25 mutational steps in the mitochondrial marker COI and by 10 mutational steps in the mitochondrial marker 16S. In nuclear loci, the groups are differentiated by one mutation (Supporting information Figure S2). These groups are genetically differentiated as shown by our F ST analysis (F ST_mtDNA = 0.60; F ST_ITS = 0.24; F ST_28S = 0.25; F ST_HSP90 = 0.20; for all loci p < 0.05 in the Hudson permutation test).

Mitochondrial nucleotide diversity was higher in the eEC clade than in the wEC clade (Table 1); however, this did not hold true for nuclear loci, which may be due to differences in effective size and other causes of mito‐nuclear discordance (Toews & Brelsford, 2012). None of the loci showed significant Tajimas’ D suggesting neutral evolution (Table 1). For all loci, genetic structure was more pronounced among populations sampled at different sides of the EC than among populations at the same side (Supporting information Figure S3). This pattern was also reflected in the AMOVA analysis, where most of the mitochondrial variation is explained by differences among regions (eEC and wEC clades; Supporting information Table S3). However, for nuclear loci, most of the variance is due to differences within populations.

Table 1.

Population genetic summary statistics for the eastern (eEC) and western (wEC) sides of the Eastern Colombian cordillera for each locus

Pop Mitochondrial ITS 28S HSP90
N S π H d D N S π H d D N S π H d D N S π H d D
eEC 49 67 0.022 0.896 1.648 46 12 0.008 0.835 −0.021 49 7 0.002 0.703 −0.342 40 10 0.002 0.729 −1.435
wEC 56 45 0.007 0.927 −0.936 53 11 0.007 0.871 −0.606 57 10 0.002 0.72 −0.459 52 23 0.005 0.855 −1.078
Open in a new tab

Pop: Population; N: number of sequences; S: number of polymorphic sites; π: nucleotide diversity; H d: haplotype diversity; D: Tajima's D.

None of the loci showed Tajima's D values that departed from neutral expectations.

All methods applied to select the optimal value of K consistently revealed two groups (K = 2; Figure 3, Supporting information Figures S4 and S5), which is consistent with the phylogenetic analyses and the population pairwise F ST values (Figure 2, Supporting information Figure S3). In agreement with those results, the canonical discriminant analyses also identified two geographic clusters (Supporting information Figure S6). Individuals from both groups share variation between them, for example, most individuals from Villavicencio (eastern foothills of the EC) had either wEC or eEC mtDNA, and up to 30% of their nDNA was of wEC origin. Furthermore, there were two individuals from this locality with wEC mtDNA, and almost ~80% of their nDNA was of wEC origin (Figures 2 and 3, Supporting information Figure S7). Likewise, two individuals from Boquia and Bucaramanga (West and Central Cordillera, respectively) presented wEC mtDNA, but their nDNA showed a shared ancestry of almost 50% with the eEC populations (Figures 2 and 3; Supporting information Figure S7). We ruled out any effect of isolation by distance rather than Andean divergence causing the geographic population structure observed here (Supporting information Figure S8 and Table S4).

Figure 3.

Figure 3

Open in a new tab

Bayesian population assignment test based on nDNA. A population assignment test with the software structure based on three nuclear loci identified two distinct populations (K = 2). Bar plots show Bayesian assignment probabilities for individuals where each color represents the most likely ancestry from which the genotype was derived (green: eEC and red: wEC). Bars on the bottom indicate the geographic region that each population belongs to. Populations are coded as described in Figure 1. In population d, individuals 67 and 68 (light gray arrow) have almost 80% of their nDNA from wEC. Individual 78 in population i (mid gray arrow) and individual 95 in population m (black arrow) have wEC mtDNA, but their nDNA showed almost 50% of shared ancestry with the eEC populations

3.3. Demographic models

phrapl revealed different wAIC values and high ∆AIC between models with gene flow (IM) and isolation only (IO; Table 2 and Figure 4; Supporting information Table S5). The model with no migration had the lowest wAIC indicating that a single vicariant event with no genetic exchange is not plausible. However, it is difficult to differentiate between symmetrical and asymmetrical gene flow (Table 2; Supporting information Table S5). When we tested recent versus ancient secondary contact, the latter model was better supported, ruling out recent secondary contact but suggesting at least some isolation caused by the vicariant event (Table 2). These results imply that gene flow is responsible for the shared ancestry of G. cancriformis between eEC and wEC geographic regions. Likewise, G‐phoCS also detected reciprocal gene flow between individuals in these two regions (Supporting information Table S2).

Table 2.

Measures of fit of alternative models, labeled as in Figure 4

Model AIC lnL K ∆AIC wAIC
a 114.20 −56.10 1 15.66 1.50E−04
b 101.65 −48.82 2 3.10 0.08
c 102.26 −49.13 2 3.72 0.06
d 99.70 −47.85 2 1.16 0.21
e 101.72 −47.86 3 3.18 0.08
f 101.78 −47.89 3 3.24 0.07
g 100.88 −47.85 2 2.34 0.18
h 98.54 −47.27 2 0.00 0.38
Open in a new tab

Isolation only (a), isolation with migration (b to f), and secondary contact (g and h).

Figure 4.

Figure 4

Open in a new tab

Demographic scenarios tested for the evolution of G. cancriformis in the northern Andes. (a) Divergence with no migration, (b) divergence with unidirectional migration from wEC to eEC, (c) divergence with unidirectional migration from eEC to wEC, (d) divergence with bidirectional symmetrical migration, (e) divergence with bidirectional asymmetrical migration from eEC to wEC, and (f) divergence with bidirectional asymmetrical migration from wEC to eEC

3.4. Phenotype by genotype association

We found the white phenotype to be the most frequent morph among all populations studied, while the black–white morph was only present in the eEC populations. Nonetheless, in the Colombian Cauca valley (wEC), we collected a black morph that had not been previously reported (Supporting information Figure S9). Although our molecular sampling (mtDNA and nDNA) revealed a statistical association between genetic variation and geography, such association was not found for color polymorphism (Supporting information Table S6). This lack of association is also evident in the mtDNA phylogeny, as individuals of different colors cluster within same clade (Figure 2).

4. DISCUSSION

Our mitochondrial and nuclear data showed two well‐supported genetic clusters separated by the EC of the Colombian Andes, although this pattern was more evident in mtDNA than in nDNA. This discordance is likely due to the differences in effective population sizes between nDNA and mtDNA, the latter having four times less effective population size causing it to complete the process of lineage sorting faster and, as a consequence, estimate older coalescent times (Toews & Brelsford, 2012). In contrast, the larger effective population size of nDNA causes not only younger coalescent times but also the co‐existence of multiple haplotypes with minor differences among them, which increases the “within population” variance, as reflected in our AMOVA analysis. This mito‐nuclear discordance is not exclusive to our case but has also been observed in other studies which found the Andes mountains to play a role in splitting populations (Smith, Harvey et al., 2014; Toews & Brelsford, 2012).

Our findings suggest the divergence of these G. cancriformis groups likely occurred sometime around the Pliocene/Pleistocene boundary, which coincides with the final uplift of this part of the Andes (Gregory‐Wodzicki, 2000). However, we are unable to pinpoint an exact population divergence date. This is because differences in effective population sizes can also affect time estimates, either overestimating them (with mtDNA) or underestimating them (with nDNA). In addition, the use of standard substitution rates provides precise but inaccurate time estimates (although this effect is reduced when analyzing closely related species or population data) (dos Reis, Donoghue, & Yang, 2015).

In contrast, the Western and Central Colombian Cordilleras, as well as the Dry Diagonal in Brazil, do not seem to contribute to the diversification of these spider populations. This contrasts with previous reports where the Brazilian Dry Diagonal has been found as a natural barrier to gene flow in taxa such as frogs, birds, and lizards (Fouquet, Santana Cassini, Fernando Baptista Haddad, Pech, & Trefaut Rodrigues, 2014; Harvey & Brumfield, 2015; Werneck, Gamble, Colli, Rodrigues, & Sites, 2012). The genetic homogeneity found within the wEC clade may also be explained by the topography of the Western and Central Colombian Cordilleras, which are considerably narrower than the Eastern one (Guarnizo et al., 2015), and which might have facilitated the dispersal of individuals across this region.

Species diversification driven by the uplift of the Andes has been documented in several organisms including birds (Chaves, Pollinger, Smith, & LeBuhn, 2007; Fernandes, Wink, Sardelli, & Aleixo, 2014; Ribas, Moyle, Miyaki, & Cracraft, 2007), reptiles (Teixeira et al., 2016), amphibians (García‐R et al., 2012; Guarnizo, Amézquita, & Bermingham, 2009; Guarnizo et al., 2015), bees (Dick et al., 2004), and butterflies (Elias et al., 2009); however, the persistence of gene flow between populations separated by the Andes is far less known (Hoffmann & Baker, 2003; Miller et al., 2008; Oswald et al., 2017). Here, despite the vicariance associated with the Andean uplift that resulted in eEC and wEC Andean clades for G. cancriformis, we found individuals with shared ancestry between the main two geographic groups. The approximate likelihood demographic model implemented, identified gene flow as the most likely explanation for this. Furthermore, the model with the best support implies divergence in the face of gene flow after τ/5 generations forward in time, which suggests a short allopatric period.

Altitudinal depressions across the Andes can contribute to the dispersal of individuals, thus allowing admixture between populations that occur at opposite sides of this barrier. In fact, the EC of the Colombian Andes is not a uniform barrier along its length and has at least two depressions, the Andalucía pass and the Suaza‐Pescado valleys, which may be acting as dispersal corridors (Cadena et al., 2016). To our knowledge, scenarios consistent with Andean altitudinal depressions facilitating dispersion and gene flow have only been reported in the Peruvian Porculla pass, where six codistributed bird taxon pairs showed asynchronous divergence times, likely due to independent dispersal events coupled with gene flow (Oswald et al., 2017). In arthropods, there is evidence for dispersal across the Andes (Dick et al., 2004), but the persistence of gene flow across this barrier has not been demonstrated. We hypothesize that eEC and wEC populations of G. cancriformis have used such type of passes to cross the Eastern Cordillera and reproduce with populations on opposite sides, even after they achieved some degree of divergence. This could be facilitated by aerial dispersal mechanisms such as ballooning, where the friction between the air and the spider's silk can make an individual move up to 3,200 Km (Gressitt, 1965). Although this displacement strategy has not been observed in G. cancriformis, it is used by its sister taxa (Bell, Bohan, Shaw, & Weyman, 2005).

Color polymorphism in the opisthosoma of G. cancriformis did not explain the structure found in this species. However, the lack of association of mtDNA or nDNA haplotypes with coloration may be due to the nature of the loci studied, as they evolve neutrally and are not members of any known pigmentation pathway in arthropods (Wittkopp & Beldade, 2009). It is also important to bear in mind that information of color phenotype is not available for all samples (mainly for those from Brazil), which may be hindering a phenotype by genotype association. This seems, however, unlikely as most of the samples within the wEC clade have phenotype information and still do not cluster by phenotype. Even so, the mtDNA phylogenetic pattern suggests, to some extent, that this polymorphism pre‐dates the geographic split. Otherwise, the genetic connectivity between the populations at both sides of the Andes may be favoring the flow of color alleles thus maintaining phenotypic variation. This remains to be tested.

The scenario of vicariance coupled with gene flow found here supports the original ideas of Chapman (1917, 1926) and Haffer (1967a,b), who claimed that the similarities in the composition of the flora and fauna at both sides of the Andes might be due to dispersals through altitudinal depressions in the cordilleras (Chapman, 1917, 1926; Haffer, 1967a,b). However, because we are studying populations of a single species, the signatures of allele sharing we obtained could also be the result of incomplete lineage sorting (ILS), especially affecting nuclear loci. By bad luck, determining the extent to which gene flow and ILS contribute to such shared genetic variation is difficult (Holder, Anderson, & Holloway, 2001; Kubatko, 2009; Kubatko & Degnan, 2007). A more exhaustive evaluation of the role of the Andes and gene flow in the diversification of G. cancriformis would require the inclusion of more populations at the lowlands and at both sides of the Andes (especially those closer to altitudinal depressions in the cordilleras). This sampling would also allow further investigation into questions such as the origin and maintenance of color polymorphism in this spider.

This case constitutes one of the few phylogeographic studies in arthropods, and the first arachnids, showing that the Andes acts as a permeable barrier rather than an absolute one for the free movement of alleles. However, our findings are based on a limited sample of the genome. Therefore, further work on this topic implementing genomic approaches is needed, especially in invertebrates. This will allow the implementation of comprehensive analyses of comparative phylogeography to understand how geographic barriers and genetic connectivity across them shape genetic diversity and impact genetic structure.

CONFLICT OF INTEREST

All authors declare no conflict of interest.

AUTHOR CONTRIBUTIONS

F.C.S‐R., E.L.P., C.S., and V.N.S. conceived the idea and designed the experiments. F.C.S.‐R. collected the individuals. F.C.S.‐R. and C.P‐D. processed the samples in the laboratory and got the sequences. V.N.S. C.S. and C.F.A. contributed with material, tools, and reagents. F.C.S‐R. and C.S. analyzed the data. F.C.S.‐R, C.S., and C.P‐D wrote the manuscript. All authors approved the final version submitted.

DATA AVAILABILITY

Sequences were deposited in GenBank under accession numbers MH258252‐MH258747.

Supporting information

 

Click here for additional data file. (48.2KB, docx)

 

Click here for additional data file. (6.4MB, docx)

ACKNOWLEDGMENTS

For collecting permits, we thank Ministerio de Ambiente y Desarrollo Sostenible and ANLA in Colombia (Permiso Marco # 0530) and ICMBio in Brazil (27127 and 38889). We also thank multiple volunteers and friends for their logistic support and help with fieldwork in Colombia and Brazil. We are very grateful to Ariadna Morales for helping with the implementation of PHRAPL. FCS was funded by COLCIENCIAS (Joven Investigador Program Call 761‐2016, contract FCN1705‐CE101). CS and CPD were funded by COLCIENCIAS Grant FP44842‐005‐2017. CFA was supported by Convocatoria “Es tiempo de volver”‐COLCIENCIAS‐2014, contract No. 656–2014. NVS was funded by FAPESP (grants #2012/02526‐7, #2013/50491‐0, and #2013/08293‐7).

Salgado‐Roa FC, Pardo‐Diaz C, Lasso De Paulis E, Arias CF, Solferini VN, Salazar C. Gene flow and Andean uplift shape the diversification of Gasteracantha cancriformis (Araneae: Araneidae) in Northern South America. Ecol Evol. 2018;8:7131–7142. 10.1002/ece3.4237

REFERENCES

  1. Antonelli, A. , Quijada‐Mascareñas, A. , Crawford, A. J. , Bates, J. M. , Velazco, P. M. , & Wüster, W. (2009). Molecular studies and phylogeography of Amazonian Tetrapods and their relation to geological and climatic models In Hoorn C. & Wesselingh F. P. (Eds.), Amazonia: Landscape and species evolution (pp. 386–404). London, UK: Wiley‐Blackwell Publishing Ltd; 10.1002/9781444306408.ch24 [DOI] [Google Scholar]
  2. Arias, C. F. , Salazar, C. , Rosales, C. , Kronforst, M. R. , Linares, M. , Bermingham, E. , & McMillan, W. O. (2014). Phylogeography of Heliconius cydno and its closest relatives: Disentangling their origin and diversification. Molecular Ecology, 23(16), 4137–4152. 10.1111/mec.12844 [DOI] [PubMed] [Google Scholar]
  3. Bartoleti, L. F. M. , Peres, E. A. , Sobral‐Souza, T. , Fontes, F. H. M. , da Silva, M. J. , & Solferini, V. N. (2017). Phylogeography of the dry vegetation endemic species Nephila sexpunctata (Araneae: Araneidae) suggests recent expansion of the Neotropical Dry Diagonal. Journal of Biogeography, 44(9), 2007–2020. 10.1111/jbi.12998 [DOI] [Google Scholar]
  4. Bell, J. R. , Bohan, D. A. , Shaw, E. M. , & Weyman, G. S. (2005). Ballooning dispersal using silk: world fauna, phylogenies, genetics and models. Bulletin of Entomological Research, 95(02), 69–114. 10.1079/BER2004350 [DOI] [PubMed] [Google Scholar]
  5. Bidegaray‐Batista, L. , & Arnedo, M. A. (2011). Gone with the plate: The opening of the Western Mediterranean basin drove the diversification of ground‐dweller spiders. BMC Evolutionary Biology, 11(1), 317 10.1186/1471-2148-11-317 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Black, M. , Hoeh, W. , Lutz, R. , & Vrijenhoek, R. (1994). DNA primers for amplification of mitochondrial cytochrome c oxidase subunit I from diverse metazoan invertebrates. Molecular Marine Biology and Biotechnology, 3, 294–299. [PubMed] [Google Scholar]
  7. Blackledge, T. A. , Scharff, N. , Coddington, J. A. , Szüts, T. , Wenzel, J. W. , Hayashi, C. Y. , & Agnarsson, I. (2009). Reconstructing web evolution and spider diversification in the molecular era. Proceedings of the National Academy of Sciences of the United States of America, 106(13), 5229–5234. 10.1073/pnas.0901377106 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Brumfield, R. T. , & Capparella, A. P. (1996). Historical diversification of birds in northwestern South America: A molecular perspective on the role of vicariant events. Evolution, 50(4), 1607 10.2307/2410897 [DOI] [PubMed] [Google Scholar]
  9. Cadena, C. D. , Pedraza, C. A. , & Brumfield, R. T. (2016). Climate, habitat associations and the potential distributions of neotropical birds : Implications for diversification across the andes. Revista de la Academia Colombiana de Ciencias Exactas, Físicas y Naturales, 40(155), 275–287. 10.18257/raccefyn.280 [DOI] [Google Scholar]
  10. Chapman, F. M. (1917). The distribution of bird‐life in colombia: A contribution to a biological survey of South America.
  11. Chapman, F. M. (1926). The distribution of bird‐life in Ecuador: A contribution to a study of the origin of Andean bird‐life. Bulletin of the American Museum of Natural History, 55, 784. [Google Scholar]
  12. Chaves, J. A. , Pollinger, J. P. , Smith, T. B. , & LeBuhn, G. (2007). The role of geography and ecology in shaping the phylogeography of the speckled hummingbird (Adelomyia melanogenys) in Ecuador. Molecular Phylogenetics and Evolution, 43(3), 795–807. 10.1016/j.ympev.2006.11.006 [DOI] [PubMed] [Google Scholar]
  13. Chazot, N. , Willmott, K. R. , Condamine, F. L. , De‐Silva, D. L. , Freitas, A. V. L. , Lamas, G. , … Elias, M. (2016). Into the Andes: Multiple independent colonizations drive montane diversity in the Neotropical clearwing butterflies Godyridina. Molecular Ecology, 25(22), 5765–5784. 10.1111/mec.13773 [DOI] [PubMed] [Google Scholar]
  14. Chazot, N. , Willmott, K. R. , Lamas, G. , Freitas, A. V. L. , Piron‐prunier, F. , Arias, C. F. , … Elias, M. (2017). Renewed diversification following Miocene landscape turnover in a Neotropical butterfly radiation. bioRxiv. 10.1101/148189. [DOI] [Google Scholar]
  15. Denton, R. , Morales, A. , & Gibbs, H. L. (2018). Genome‐specific histories of divergence and introgression between an allopolyploid unisexual salamander lineage and two sexual species. bioRxiv. 10.1101/284950 [DOI] [PubMed] [Google Scholar]
  16. De‐silva, D. L. , Chazot, N. , Silva‐, K. L. , Lamas, G. , Mallet, J. , Willmott, K. R. , & Elias, M. (2017). North Andean origin and diversification of the largest ithomiine butterfly genus. Scientific Reports, 7, 45966 10.1038/srep45966 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Díaz, S. , Panzera, F. , Jaramillo‐O, N. , Pérez, R. , Fernández, R. , Vallejo, G. , … Gómez‐Palacio, A. (2014). Genetic, cytogenetic and morphological trends in the evolution of the rhodnius (Triatominae: Rhodniini) trans‐andean group. PLoS One, 9(2), e87493 10.1371/journal.pone.0087493 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Dick, C. W. , Roubik, D. W. , Gruber, K. F. , & Bermingham, E. (2004). Long‐distance gene flow and cross‐Andean dispersal of lowland rainforest bees (Apidae: Euglossini) revealed by comparative mitochondrial DNA phylogeography. Molecular Ecology, 13(12), 3775–3785. 10.1111/j.1365-294X.2004.02374.x [DOI] [PubMed] [Google Scholar]
  19. Diniz‐filho, J. A. F. , Soares, T. N. , Lima, J. S. , Dobrovolski, R. , Landeiro, V. L. , Pires, M. , … Bini, L. M. (2013). Mantel test in population genetics. Genetics and Molecular Biology, 485, 475–485. 10.1590/S1415-47572013000400002 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Dixon, P. (2003). VEGAN, a package of R functions for community ecology. Journal of Vegetation Science, 14(6), 927–930. 10.1111/j.1654-1103.2003.tb02228.x [DOI] [Google Scholar]
  21. Drummond, A. J. , & Bouckaert, R. R. (2015). Bayesian evolutionary analysis with BEAST. Cambridge, UK: Cambridge University Press; 10.1017/CBO9781139095112 [DOI] [Google Scholar]
  22. Drummond, A. J. , Suchard, M. A. , Xie, D. , & Rambaut, A. (2012). Bayesian phylogenetics with BEAUti and the BEAST 1.7. Molecular Biology and Evolution, 29(8), 1969–1973. 10.1093/molbev/mss075 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Earl, D. A. , & vonHoldt, B. M. (2012). STRUCTURE HARVESTER: A website and program for visualizing STRUCTURE output and implementing the Evanno method. Conservation Genetics Resources, 4(2), 359–361. 10.1007/s12686-011-9548-7 [DOI] [Google Scholar]
  24. Elias, M. , Joron, M. , Willmott, K. , Silva‐BrandÃo, K. L. , Kaiser, V. , Arias, C. F. , … Jiggins, C. D. (2009). Out of the Andes: Patterns of diversification in clearwing butterflies. Molecular Ecology, 18(8), 1716–1729. 10.1111/j.1365-294X.2009.04149.x [DOI] [PubMed] [Google Scholar]
  25. Evanno, G. , Regnaut, S. , & Goudet, J. (2005). Detecting the number of clusters of individuals using the software structure: A simulation study. Molecular Ecology, 14(8), 2611–2620. 10.1111/j.1365-294X.2005.02553.x [DOI] [PubMed] [Google Scholar]
  26. Excoffier, L. , & 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. 10.1111/j.1755-0998.2010.02847.x [DOI] [PubMed] [Google Scholar]
  27. Fernandes, A. M. , Wink, M. , Sardelli, C. H. , & Aleixo, A. (2014). Multiple speciation across the Andes and throughout Amazonia : the case of the spot‐backed antbird species complex (Hylophylax naevius/Hylophylax naevioides). Journal of Biogeography, 41(6), 1094–1104. 10.1111/jbi.12277 [DOI] [Google Scholar]
  28. Fouquet, A. , Santana Cassini, C. , Fernando Baptista Haddad, C. , Pech, N. , & Trefaut Rodrigues, M. (2014). Species delimitation, patterns of diversification and historical biogeography of the Neotropical frog genus Adenomera (Anura, Leptodactylidae). Journal of Biogeography, 41(5), 855–870. 10.1111/jbi.12250 [DOI] [Google Scholar]
  29. Friendly, M. , & Fox, J. (2017). Candisc: visualizing generalized canonical discriminant and canonical correlation analysis (version 1.0). Retrieved from http://CRAN.R-project.org/package=candisc, [Google Scholar]
  30. García‐R, J. C. , Crawford, A. J. , Mendoza, Á. M. , Ospina, O. , Cardenas, H. , & Castro, F. (2012). Comparative phylogeography of direct‐developing frogs (Anura: Craugastoridae: Pristimantis) in the Southern Andes of Colombia. PLoS One, 7(9), e46077 10.1371/journal.pone.0046077 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Gawryszewski, F. M. (2007). Policromatismo e stabilimentum em Gasteracantha cancriformis (Araneae, Araneidae): Caracterização e as hipóteses da atração de presas e da proteção da teia. Brasilia, Brazil: Universidade de Brasilia. [Google Scholar]
  32. Gawryszewski, F. M. , & Motta, P. C. (2012). Colouration of the orb‐web spider Gasteracantha cancriformis does not increase its foraging success. Ethology Ecology & Evolution, 24(1), 23–38. 10.1080/03949370.2011.582044 [DOI] [Google Scholar]
  33. Gregory‐Wodzicki, K. M. (2000). Uplift history of the Central and Northern Andes: A review. Geological Society of America Bulletin, 112(7), 1091–1105. 10.1130/0016-7606 [DOI] [Google Scholar]
  34. Gressitt, J. L. (1965). Biogeography and ecology of land arthropods of Antarctica In Van Miegham J. & Van Oye P. (Eds.), Biogeography and ecology of Antarctica. Monographiae Biologicae, vol. 15 (pp. 431–490). Dordrecht, the Netherlands: Springer; 10.1007/978-94-015-7204-0 [DOI] [Google Scholar]
  35. Gronau, I. , Hubisz, M. J. , Gulko, B. , Danko, C. G. , & Siepel, A. (2011). Bayesian inference of ancient human demography from individual genome sequences. Nature Genetics, 43, 1031 10.1038/ng.937 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Guarnizo, C. E. , Amézquita, A. , & Bermingham, E. (2009). The relative roles of vicariance versus elevational gradients in the genetic differentiation of the high Andean tree frog, Dendropsophus labialis . Molecular Phylogenetics and Evolution, 50(1), 84–92. 10.1016/j.ympev.2008.10.005 [DOI] [PubMed] [Google Scholar]
  37. Guarnizo, C. E. , Paz, A. , Muñoz‐Ortiz, A. , Flechas, S. V. , Méndez‐Narváez, J. , & Crawford, A. J. (2015). DNA barcoding survey of anurans across the eastern cordillera of Colombia and the impact of the andes on cryptic diversity. PLoS One, 10(5), e0127312 10.1371/journal.pone.0127312 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Haffer, J. (1967a). Speciation in Colombian forest birds west of Andes. Novitates Zoologicae, 2294(2294), 1–57. [Google Scholar]
  39. Haffer, J. (1967b). Zoogeographical notes on the “Non‐Forest” lowland bird faunas of northwestern South America. El Hornero, 10(4), 315–333. [Google Scholar]
  40. Harvey, M. G. , & Brumfield, R. T. (2015). Genomic variation in a widespread Neotropical bird (Xenops minutus) reveals divergence, population expansion, and gene flow. Molecular Phylogenetics and Evolution, 83, 305–316. 10.1016/j.ympev.2014.10.023 [DOI] [PubMed] [Google Scholar]
  41. Hedin, M. C. , & Maddison, W. P. (2001). A combined molecular approach to phylogeny of the jumping spider subfamily Dendryphantinae (Araneae: Salticidae). Molecular Phylogenetics and Evolution, 18(3), 386–403. 10.1006/mpev.2000.0883 [DOI] [PubMed] [Google Scholar]
  42. Hijmans, R. J. (2016). geosphere: Spherical Trigonometry.
  43. Hime, P. M. , Hotaling, S. , Grewelle, R. E. , O'Neill, E. M. , Voss, S. R. , Shaffer, H. B. , & Weisrock, D. W. (2016). The influence of locus number and information content on species delimitation: An empirical test case in an endangered Mexican salamander. Molecular Ecology, 25(23), 5959–5974. 10.1111/mec.13883 [DOI] [PubMed] [Google Scholar]
  44. Hoffmann, F. G. , & Baker, R. J. (2003). Comparative phylogeography of short‐tailed bats (Carollia: Phyllostomidae). Molecular Ecology, 12(12), 3403–3414. 10.1046/j.1365-294X.2003.02009.x [DOI] [PubMed] [Google Scholar]
  45. Holder, M. T. , Anderson, J. A. , & Holloway, A. K. (2001). Difficulties in detecting hybridization. Systematic Biology, 50(6), 978–982. http://www.jstor.org/stable/3070875 [DOI] [PubMed] [Google Scholar]
  46. Hoorn, C. , Wesselingh, F. P. , ter Steege, H. , Bermudez, M. A. , Mora, A. , Sevink, J. , … Antonelli, A. (2010). Amazonia through time: Andean uplift, climate change, landscape evolution, and biodiversity. Science (New York, N.Y.), 330(6006), 927–931. 10.1126/science.1194585 [DOI] [PubMed] [Google Scholar]
  47. Hudson, R. R. , Boos, D. D. , & Kaplan, N. L. (1992). A statistical test for detecting geographic subdivision. Molecular Biology and Evolution, 9(1), 138–151. [DOI] [PubMed] [Google Scholar]
  48. Jackson, N. D. , Morales, A. E. , Carstens, B. C. , & O'Meara, B. C. (2017). PHRAPL: Phylogeographic inference using approximate likelihoods. Systematic Biology, 19, 431–435. 10.1093/sysbio/syx001 [DOI] [PubMed] [Google Scholar]
  49. Janes, J. K. , Miller, J. M. , Dupuis, J. R. , Malenfant, R. M. , Gorrell, J. C. , Cullingham, C. I. , & Andrew, R. L. (2017). The K = 2 conundrum. Molecular Ecology, 26(14), 3594–3602. 10.1111/mec.14187 [DOI] [PubMed] [Google Scholar]
  50. Jombart, T. , & Ahmed, I. (2011). adegenet 1. 3‐1 : New tools for the analysis of genome‐wide SNP data. Bioinformatics, 27(21), 3070–3071. 10.1093/bioinformatics/btr521 [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Kattan, G. H. , Franco, P. , & Rojas, V. (2004). Biological diversification in a complex region : A spatial analysis of faunistic diversity and biogeography of the Andes of Colombia, 1829–1839. Journal of Biogeography, 31, 1829–1839. 10.1111/j.1365-2699.2004.01109.x [DOI] [Google Scholar]
  52. Kopelman, N. M. , Mayzel, J. , Jakobsson, M. , Rosenberg, N. A. , & Mayrose, I. (2015). CLUMPAK: A program for identifying clustering modes and packaging population structure inferences across K. Molecular Ecology Resources, 15, 1179–1191. 10.1111/1755-0998.12387 [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Kubatko, L. S. (2009). Identifying hybridization events in the presence of coalescence via model selection. Systematic Biology, 58(5), 478–488. 10.1093/sysbio/syp055 [DOI] [PubMed] [Google Scholar]
  54. Kubatko, L. S. , & Degnan, J. H. (2007). Inconsistency of phylogenetic estimates from concatenated data under coalescence. Systematic Biology, 56(1), 17–24. 10.1080/10635150601146041 [DOI] [PubMed] [Google Scholar]
  55. Kuntner, M. , Arnedo, M. A. , Trontelj, P. , Lokovšek, T. , & Agnarsson, I. (2013). A molecular phylogeny of nephilid spiders: Evolutionary history of a model lineage. Molecular Phylogenetics and Evolution, 69(3), 961–979. 10.1016/j.ympev.2013.06.008 [DOI] [PubMed] [Google Scholar]
  56. Legendre, P. , & Fortin, M.‐J. (2010). Comparison of the Mantel test and alternative approaches for detecting complex multivariate relationships in the spatial analysis of genetic data. Molecular Ecology Resources, 10(5), 831–844. 10.1111/j.1755-0998.2010.02866.x [DOI] [PubMed] [Google Scholar]
  57. Legendre, P. , Fortin, M.‐J. , & Borcard, D. (2015). Should the Mantel test be used in spatial analysis? Methods in Ecology and Evolution, 6(11), 1239–1247. 10.1111/2041-210X.12425 [DOI] [Google Scholar]
  58. Leigh, J. W. , Bryant, D. , & Nakagawa, S. (2015). popart: Full‐feature software for haplotype network construction. Methods in Ecology and Evolution, 6(9), 1110–1116. 10.1111/2041-210X.12410 [DOI] [Google Scholar]
  59. Levi, H. W. (1978). The American orb‐weaver genera Colphepeira, Micrathena and Gasteracantha North of Mexico (Araneae, Araneidae). Bulletin of the Museum of Comparative Zoology, 148, 417–442. [Google Scholar]
  60. Librado, P. , & Rozas, J. (2009). DnaSP v5 : A software for comprehensive analysis of DNA polymorphism data. Bioinformatics, 25(11), 1451–1452. 10.1093/bioinformatics/btp187 [DOI] [PubMed] [Google Scholar]
  61. Maddison, W. , & Maddison, D. (2015). Mesquite: A modular system for evolutionary analysis, version 3.04. Retrieved from http://mesquiteproject. org/mesquite/download/download.html.
  62. Mantel, N. (1967). The detection of disease clustering and a generalized regression approach. Cancer Research, 27(2 Part 1), 209 LP–220. [PubMed] [Google Scholar]
  63. McHugh, A. , Yablonsky, C. , Binford, G. J. , & Agnarsson, I. (2014). Molecular phylogenetics of Caribbean Micrathena (Araneae: Araneidae) suggests multiple colonization events and single island endemism. Invertebrate Systematics, Accepted F, 2012, 337–349. 10.1071/IS13051 [DOI] [Google Scholar]
  64. Meirmans, P. G. (2012). The trouble with isolation by distance. Molecular Ecology, 21(12), 2839–2846. 10.1111/j.1365-294X.2012.05578.x [DOI] [PubMed] [Google Scholar]
  65. Miller, M. J. , Bermingham, E. , Klicka, J. , Escalante, P. , do Amaral, F. S. R. , Weir, J. T. , & Winker, K. (2008). Out of Amazonia again and again: Episodic crossing of the Andes promotes diversification in a lowland forest flycatcher. Proceedings of the Royal Society of London B: Biological Sciences, 275(1639), 1133–1142. 10.1098/rspb.2008.0015 [DOI] [PMC free article] [PubMed] [Google Scholar]
  66. Moradmand, M. , Schönhofer, A. L. , & Jäger, P. (2014). Molecular phylogeny of the spider family Sparassidae with focus on the genus Eusparassus and notes on the RTA‐clade and “Laterigradae”. Molecular Phylogenetics and Evolution, 74, 48–65. 10.1016/j.ympev.2014.01.021 [DOI] [PubMed] [Google Scholar]
  67. Morales, A. E. , Jackson, N. D. , Dewey, T. A. , Meara, B. C. O. , & Bryan, C. (2017). Speciation with gene flow in North American Myotis bats. Systematic Biology, 66(3), 440–452. 10.1093/sysbio/syw100 [DOI] [PubMed] [Google Scholar]
  68. Muma, M. H. (1971). Biological and behavioral notes on Gasteracantha cancriformis (Arachnida: Araneidae). Florida Entomologist Society, 54(4), 345–351. 10.2307/3493600 [DOI] [Google Scholar]
  69. Nguyen, L. T. , Schmidt, H. A. , Von Haeseler, A. , & Minh, B. Q. (2015). IQ‐TREE: A fast and effective stochastic algorithm for estimating maximum‐likelihood phylogenies. Molecular Biology and Evolution, 32(1), 268–274. 10.1093/molbev/msu300 [DOI] [PMC free article] [PubMed] [Google Scholar]
  70. Oswald, J. A. , Overcast, I. , Mauck, W. M. , Andersen, M. J. , & Smith, B. T. (2017). Isolation with asymmetric gene flow during the nonsynchronous divergence of dry forest birds. Molecular Ecology, 26(5), 1386–1400. 10.1111/mec.14013 [DOI] [PubMed] [Google Scholar]
  71. Peres, E. A. , Sobral‐Souza, T. , Perez, M. F. , Bonatelli, I. A. S. , Silva, D. P. , Silva, M. J. , & Solferini, V. N. (2015). Pleistocene niche stability and lineage diversification in the subtropical spider Araneus omnicolor (Araneidae). PLoS One, 10(4), e0121543 10.1371/journal.pone.0121543 [DOI] [PMC free article] [PubMed] [Google Scholar]
  72. Posada, D. (2008). jModelTest: Phylogenetic model averaging. Molecular Biology and Evolution, 25(7), 1253–1256. 10.1093/molbev/msn083 [DOI] [PubMed] [Google Scholar]
  73. Pritchard, J. K. , Stephens, M. , & Donnelly, P. (2000). Inference of population structure using multilocus genotype data. Genetics, 155(2), 945–959. [DOI] [PMC free article] [PubMed] [Google Scholar]
  74. Prosdocimi, F. , Bittencourt, D. , da Silva, F. R. , Kirst, M. , Motta, P. C. , & Rech, E. L. (2011). Spinning gland transcriptomics from two main clades of spiders (Order: Araneae) ‐ Insights on their molecular, anatomical and behavioral evolution. PLoS One, 6(6), e21634 10.1371/journal.pone.0021634 [DOI] [PMC free article] [PubMed] [Google Scholar]
  75. Rambaut, A. , Drummond, A. J. , Xie, D. , Baele, G. , & Suchard, M. A. (2018). Posterior summarisation in Bayesian phylogenetics using Tracer 1.7. Systematic Biology, syy032. 10.1093/sysbio/syy032 [DOI] [PMC free article] [PubMed] [Google Scholar]
  76. Rambaut, A. (2012). FigTree, version 1.4, Retrieved from http://tree.bio.ed.ac.uk/software/figtree
  77. dos Reis, M. , Donoghue, P. C. J. , & Yang, Z. (2015). Bayesian molecular clock dating of species divergences in the genomics era. Nature Reviews Genetics, 17, 71 10.1038/nrg.2015.8 [DOI] [PubMed] [Google Scholar]
  78. Ribas, C. C. , Moyle, R. G. , Miyaki, C. Y. , & Cracraft, J. (2007). The assembly of montane biotas: Linking Andean tectonics and climatic oscillations to independent regimes of diversification in Pionus parrots. Proceedings of the Royal Society of London B: Biological Sciences, 274(1624), 2399–2408. 10.1098/rspb.2007.0613 [DOI] [PMC free article] [PubMed] [Google Scholar]
  79. Rull, V. (2011). Neotropical biodiversity: Timing and potential drivers. Trends in Ecology and Evolution, 26(10), 508–513. 10.1016/j.tree.2011.05.011 [DOI] [PubMed] [Google Scholar]
  80. Salazar, C. , Baxter, S. W. , Pardo‐Diaz, C. , Wu, G. , Surridge, A. , Linares, M. , … Jiggins, C. D. (2010). Genetic evidence for hybrid trait speciation in heliconius butterflies. PLOS Genetics, 6(4), e1000930 10.1371/journal.pgen.1000930 [DOI] [PMC free article] [PubMed] [Google Scholar]
  81. Simon, C. , Frati, F. , Beckenbach, A. , Crespi, B. , Liu, H. , & Flook, P. (1994). Evolution, weighting, and phylogenetic utility of mitochondrial gene sequences and a compilation of conserved polymerase chain reaction primers. Annals of the Entomological Society of America, 87(6), 651–701. 10.1093/aesa/87.6.651 [DOI] [Google Scholar]
  82. Smith, B. T. , Harvey, M. G. , Faircloth, B. C. , Glenn, T. C. , & Brumfield, R. T. (2014). Target capture and massively parallel sequencing of ultraconserved elements for comparative studies at shallow evolutionary time scales. Systematic Biology, 63(1), 83–95. 10.1093/sysbio/syt061 [DOI] [PubMed] [Google Scholar]
  83. Smith, B. T. , McCormack, J. E. , Cuervo, A. M. , Hickerson, M. J. , Aleixo, A. , Cadena, C. D. , … Brumfield, R. T. (2014). The drivers of tropical speciation. Nature, 515(7527), 406–409. 10.1038/nature13687 [DOI] [PubMed] [Google Scholar]
  84. Smouse, P. E. , Long, J. C. , & Sokal, R. R. (1986). Multiple regression and correlation extensions of the mantel test of matrix correspondence. Systematic Zoology, 35(4), 627 10.2307/2413122 [DOI] [Google Scholar]
  85. Tamura, K. , Stecher, G. , Peterson, D. , Filipski, A. , & Kumar, S. (2013). MEGA6: Molecular evolutionary genetics analysis version 6.0. Molecular Biology and Evolution, 30(12), 2725–2729. 10.1093/molbev/mst197 [DOI] [PMC free article] [PubMed] [Google Scholar]
  86. Teixeira, M. , Prates, I. , Nisa, C. , Silva‐Martins, N. S. C. , Strüssmann, C. , & Rodrigues, M. T. (2016). Molecular data reveal spatial and temporal patterns of diversification and a cryptic new species of lowland Stenocercus Duméril & Bibron, 1837 (Squamata: Tropiduridae). Molecular Phylogenetics and Evolution, 94(Part A), 410–423. 10.1016/j.ympev.2015.09.010 [DOI] [PubMed] [Google Scholar]
  87. Thompson, J. D. , Higgins, D. G. , & Gibson, T. J. (1994). CLUSTAL W: Improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position‐specific gap penalties and weight matrix choice. Nucleic Acids Research, 22(22), 4673–4680. 10.1093/nar/22.22.4673 [DOI] [PMC free article] [PubMed] [Google Scholar]
  88. Toews, D. P. L. , & Brelsford, A. (2012). The biogeography of mitochondrial and nuclear discordance in animals. Molecular Ecology, 21(16), 3907–3930. 10.1111/j.1365-294X.2012.05664.x [DOI] [PubMed] [Google Scholar]
  89. Turchetto‐Zolet, A. C. , Pinheiro, F. , Salgueiro, F. , & Palma‐Silva, C. (2013). Phylogeographical patterns shed light on evolutionary process in South America. Molecular Ecology, 22(5), 1193–1213. 10.1111/mec.12164 [DOI] [PubMed] [Google Scholar]
  90. Werneck, F. P. , Gamble, T. , Colli, G. R. , Rodrigues, M. T. , & Sites, J. W. Jr (2012). Deep diversification and long‐term persistence in the south american “dry diagonal”: Integrating continent‐wide phylogeography and distribution modeling of geckos. Evolution, 66(10), 3014–3034. 10.1111/j.1558-5646.2012.01682.x [DOI] [PubMed] [Google Scholar]
  91. White, T. J. , Bruns, T. , Lee, S. , & Taylor, J. (1990). Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics In Innis M. A., Gelfand D. H., Sninsky J. J. & White T. J. (Eds.), PCR protocols: A guide to methods and applications (pp. 315–322). Academic Press; NY: 10.1016/B978-0-12-372180-8.50042-1 [DOI] [Google Scholar]
  92. Wittkopp, P. J. , & Beldade, P. (2009). Development and evolution of insect pigmentation: Genetic mechanisms and the potential consequences of pleiotropy. Seminars in Cell & Developmental Biology, 20(1), 65–71. 10.1016/j.semcdb.2008.10.002 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

 

Click here for additional data file. (48.2KB, docx)

 

Click here for additional data file. (6.4MB, docx)

Data Availability Statement

Sequences were deposited in GenBank under accession numbers MH258252‐MH258747.

Articles from Ecology and Evolution are provided here courtesy of Wiley

RESOURCES