Abstract

The structure and genetic diversity of a widely distributed species in a recently colonized area is influenced by the colonizing lineages, life-history traits, and biotic and abiotic factors. The connection established during the Pliocene between North and South America allowed the nine-banded armadillo (Dasypus novemcinctus) to expand its distributional range northward. High levels of genetic diversity have been recorded in South America, whereas low levels have been detected in populations in the United States, perhaps due to a founder effect during colonization. By sampling animals from Mexico and a few other areas, we test the hypothesis that armadillos in North America were derived from a single founding lineage, and assess whether this newly colonized region shows demographic signatures of expansion. We sequenced the mitochondrial control region of 157 individuals and genotyped microsatellites of 116 individuals. Our mitochondrial results showed 2 divergent lineages with high genetic variation in Mexico when compared to United States populations, suggesting that this species has a higher effective population size in Mexico. Samples from Central and South America indicate that both lineages differentiated prior to their arrival in Mexico. Lineages showed a historical demographic expansion, due probably to the large area of colonization. Clear genetic structure was observed with mitochondrial DNA, whereas microsatellites showed low levels of genetic differentiation. This contrasting pattern can be caused by male-biased dispersal. We conclude that North American populations of D. novemcinctus are derived from 2 founding lineages and show the consequences of the Great American Biotic Interchange influencing genetic patterns in the nine-banded armadillo in Mexico.

Patterns of genetic structure and variation of widely distributed species in areas of recent colonization are influenced by many factors, including the number and genetic composition of colonizing lineages (Ficetola et al. 2008; Virgilio et al. 2009), the speed of advance (Excoffier et al. 2009), disturbances such as climate change (Hewitt 2004), human-mediated habitat perturbations (Hájková et al. 2007; Mondol et al. 2009), and life-history traits peculiar to each species (Braaker and Heckel 2009). During natural colonizations, species gradually expand their geographic range by migrating to adjacent regions (Wilson et al. 2009), facilitating the spread of genetic lineages to new regions (Ruedi et al. 2008; Teacher et al. 2009).

At the end of the Pliocene and the beginning of the Pleistocene, the connection established between North and South America through the Isthmus of Panama allowed many species to expand their range (Webb 1991), in a phenomenon known as the Great American Biotic Interchange (Marshall et al. 1982; Webb 1976). The Great American Biotic Interchange is recognized as a major biogeographic event (Marshall et al. 1982; Pascual and Ortiz-Jaureguizar 2007) and influenced patterns of genetic diversity in many taxa (Culver et al. 2000; Eizirik et al. 2001; Wüster et al. 2005). The number of lineages of some bird and reptile species that contributed to the colonization has been explored (Daza et al. 2009; Tilston and Klicka 2010; Weir et al. 2009), but similar studies and details remain unclear for mammals.

Cingulata was one of the most representative mammal orders that migrated into North America during the Great American Biotic Interchange. This order is currently composed of 21 548 species of armadillos, most of which are confined to various parts of South America (Gardner 2005). The nine-banded armadillo (Dasypus novemcinctus) has the broadest distribution. Currently, it ranges from Uruguay and northeastern Argentina to the southern United States (Abba and Superina 2010). Examination of molecular data suggests that D. novemcinctus originated about 7 million years ago (mya— Delsuc et al. 2004) and paleontological data indicate that this origin probably occurred in northern South America (Carlini et al. 1997). The earliest remains of D. novemcinctus were found in Uruguay and date back to the late Pleistocene (Sopas Formation—Vizcaino et al. 1995). The oldest record of Dasypus in Mexico corresponds to the Holocene (approximately 8,000 years ago—Álvarez 1974). D. novemcinctus was 1st recorded in the lower Rio Grande valley of Texas in 1849 (Audubon and Bachman 1854), but rapidly colonized much of the southern United States. Because of the wide distribution of D. novemcinctus and its clear South American origin (Eisenberg 1981; Simpson 1980), this species is a good model to examine genetic structure in newly colonized areas because many genetic lineages could exist across the original range, and one or many of them might have participated in the colonization of North America. Consistent with expectations of a single founding lineage, previous studies showed that South American populations of nine-banded armadillos have higher levels of mitochondrial and nuclear genetic diversity (Frutos and Van Den Bussche 2002; Huchon et al. 1999) than their United States counterparts (Huchon et al. 1999; Moncrief 1988). This has been interpreted as the result of a founder effect during the colonization of North America (Huchon et al. 1999). However, this conclusion is based solely on data from populations in the United States. It remains unknown whether similar patterns occur farther south in Mexico.

We used nuclear and mitochondrial markers to screen a large number of nine-banded armadillos from Mexico, and a few other areas, to further test the hypothesis that armadillos in North America are derived from a single founding lineage. Additionally, we assessed whether the newly colonized region of North America shows demographic signatures of expansion (as might be expected for a recent colonization), or whether colonizing populations have had time to differentiate in the topographically diverse setting of Mesoamerica.

Materials and Methods

Sample collection.—We collected 157 specimens of D. novemcinctus from 110 localities, including 138 from Mexico (92 localities in 18 states), 12 from Colombia, 2 from Costa Rica, 1 from Guatemala, 2 from Nicaragua, and 2 from the United States. Among them, 71 were collected in the field from wild populations and 86 samples were museum specimens (Appendix I). All samples were collected in accordance with guidelines of the American Society of Mammalogists (Sikes et al. 2011), and fresh samples were collected in accordance with Mexico's wildlife legislation under Wildlife Department permit FAUT-0001. We used phosphate buffered saline solution at 1% (8 g of NaCl, 0.2 g of KCl, 1.44 g of NaHPO, 0.24 g of KH2PO4, pH 7.4) to clean and to hydrate the tissues for 24-72 h at 56°C. Total genomic DNA was extracted from each sample using a DNeasy Kit (Qiagen, Valencia, California), and stored at 4°C.

Amplification and sequencing.—A segment of the mitochondrial control region (408 base pairs [bp]) from each sample was amplified via polymerase chain reaction using primers D2 (5 ′ -ATTTYGGCGCTATGTAATTCG-3 ′ ; F. Delsuc, Université Montpellier II) and N4 (5 ′ -GGC AT AAGTC- CATCGAGATGTC-3 ′ ; designed in our laboratory). Taq polymerase was added at 2.5 U per 25 ul of reaction volume. The final l× buffer had 0.4 μM of each primer, 0.15 mM of deoxynucleoside triphosphate, and 2 mM of MgCl2. The thermal profile for amplification consisted of an initial denaturation cycle at 95 °C for 3 min, followed by 30 cycles at 94°C for 30 s, 62°C for 45 s, and 72°C for 120 s, and a final extension at 72°C for 10 min. All amplifications were performed in a Perkin-Elmer GeneAmp PCR system 9600 (Applied Biosystems, Foster City, California).

Each product was examined in a 1.5% agarose gel stained with ethidium bromide. Sequencing was done in both directions by the Sanger method in the Finch Lab at the University of Washington, Seattle, using forward and reverse primers described above. Mitochondrial DNA (mtDNA) sequences were aligned using ClustalW (Thompson et al. 1997) in BioEdit 7.0.5 (Hall 1999). The Akaike information criterion in jModelTest 0.1.1 (Posada 2008) was used to determine the best-fit evolution model for the phylogenetic analysis.

Micro satellite genotyping.—Five nuclear DNA (nDNA) autosomal microsatellite loci previously reported for this species (Dnov1, Dnov6, Dnov7, Dnov16, and Dnov24Prodöhl et al. 1996) were analyzed in the 116 individuals from Mexico. Because of DNA degradation in some museum samples, not all individuals amplified for all microsatellite loci.

Individuals were genotyped in a 10-μl reaction containing l× buffer with 0.3 μM of each primer, 0.15 mM of deoxynucleoside triphosphate, 1–2 mM of MgCl2 and 1.5 U of Taq DNA polymerase. The thermal profile for amplification consisted of an initial denaturation at 94°C for 5 min, followed by 35 cycles of 15 s at 94°C, 30 s at 56–58°C, and 1 s at 72°C, with a final extension of 7 min at 72°C. All reaction products were run on an ABI PRISM genetic analyzer with a 4- capillary system (Applied Biosystems) and sized with an internal lane standard using GENEMAPPER 4.0. The presence of null alleles was determined using MICRO- CHECKER (Van Oosterhout et al. 2004). These alleles may affect estimations of population differentiation, by reducing genetic diversity within populations (Chapuis and Estoup 2007).

Structure and genetic diversity in mtDNA.—A phylogenetic tree was inferred from unique mtDNA haplotypes using a Bayesian analysis implemented in BEAST 1.5.4 (Drummond and Rambaut 2007). Because D. novemcinctus migrated from South to North America (Webb 1976), haplotypes from 549 Colombia, Central America, and the United States were added in the analysis to examine whether the structure detected in Mexico was exclusive to this region or also present elsewhere (samples from countries other than Mexico were only used for the phylogenetic tree and the haplotype network). The closely related congener D. kappleri was used as the outgroup to root the tree. Three sequences of this species were obtained from the National Center for Biotechnology Information (GenBank accession numbers AJ010382, AJ010383, and AJ010384). We used an uncorrelated lognormal distribution to describe the variation in the molecular clock rate, as recommended by Drummond et al. (2007). This implies that there is no a priori correlation between the molecular clock rate of a lineage and that of its ancestor. We used an unweighted pair-group method using arithmetic averages tree as the starting tree. The other priors were set to default, and the program was repeatedly run to optimize the scale factors of the a priori function. After optimization, 10 million generations were run, with trees sampled every 1,000 generations. The first 1,000 trees were discarded as burn-in. Convergence on the parameter estimates was checked using Tracer version 1.5 (Rambaut and Drummond 2007).

For more closely related alleles, genealogical relationships among haplotypes were estimated using the median-joining algorithm of NETWORK 4.516 (http://www.fluxus-engineering.com). The median-joining method uses a maximum-parsimony approach to search for the shortest and least complex haplotypic trees for a given data set (Bandelt et al. 1999). When internal node haplotypes are not sampled, the median-joining method provides the best estimate of the genealogy (Cassens et al. 2005).

The distribution of mitochondrial variation was analyzed for the major mtDNA lineages identified. An analysis of molecular variance (AMOVA—Excoffier et al. 1992) was performed with Arlequin 3.11 (Excoffier et al. 2005) by partitioning the total sum of the squares into components representing variation among the major lineages and within them. The significance of FST was evaluated by comparing the observed value with the distribution of the values obtained from 1,000 random permutations between lineages. Estimates of nucleotide diversity (π), haplotype diversity (h), number of haplotypes, and mean number of pair-wise differences (K) were calculated for the complete data set and for each of the major mitochondrial lineages identified with DnaSP 5.0 (Librado and Rozas 2009).

Structure, genetic diversity, and gene flow in nuclear micro- satellites.—To explore the genetic structure revealed by the 5 microsatellites, we used the software Geneland 3.2.2 (Guillot et al. 2005; Guillot and Santos 2009). This Bayesian approach groups individuals into clusters (K) by minimizing Hardy- Weinberg disequilibrium and incorporating spatial information concerning the origin of the samples (Coulon et al. 2006). We tested values of K that ranged from 1 to 20 with 5 replicates each of 750,000 Markov chain Monte Carlo iterations (with thinning = 750 and a burn-in of 1,000 for each). We set the model to codominant data, correlated allele frequencies, a spatial Dirichlet model, a maximum rate of Poisson process of 100, and a maximum number of nuclei of 300. The number of clusters (K) was inferred from the modal value of K for these 5 runs, using the highest posterior probability. The process follows Coulon et al. (2006), who suggested inferring K in the 1st run and then running the algorithm again with K fixed at the previously inferred value in order to estimate other parameters. Thus, after K was defined by the mode of the posterior distribution of the Markov chain Monte Carlo chains, this was fixed and we then ran 500,000 Markov chain Monte Carlo iterations, 100 times, with settings equal to the previous runs. The mean logarithm of posterior probability for each of the 100 runs was calculated, and the 10 runs with highest values were selected. The geographical distribution of the subpopulations was reconstructed from the plot of posterior probabilities of each individual and their assignment to 1 or more of the genetic clusters. Independence between mitochondrial and nuclear genetic structure was assessed with a contingency table and chi-square analyses.

Nuclear genetic diversity parameters were estimated for the total sample and for the major mtDNA lineages identified. Estimates of observed (Ho) and expected (HE) heterozygosities and tests for departure from the Hardy-Weinberg equilibrium were obtained with Arlequin 3.11 (Excoffier et al. 2005). In addition, inbreeding coefficients, defined as a measure of the deficit of the heterozygotes inside the population (Wright 1951), and allele richness were calculated in FSTAT 2.9.3 (Goudet 2001). Linkage disequilibrium tests between pairs of loci within each lineage and overall were calculated with GENEPOP 3.4 (Raymond and Rousset 1995) using 10,000 permutations.

We calculated FST, RSt, and Dest to identify the contribution of historical and more current factors in the structure of the major mtDNA lineages. FST (Wright 1951) evaluates the difference in allelic frequencies between lineages, RSt (Slatkin 1995) is the fraction of the total variance of allele size between lineages, and the Dest estimate (Jost 2008) considers the proportion of alleles that are unique in each lineage. Arlequin 3.11 (Excoffier et al. 2005), FSTAT 2.9.3 (Goudet 2001), and SMOGD 1.2.5 (Crawford 2010; http://www.ngcrawford.com/django/jost/) were used to calculate these parameters, respectively. To avoid potential bias in FST due to the presence of null alleles in 2 microsatellite loci, FSt estimation was performed following the “Excluding Null Alleles” correction proposed by Chapuis and Estoup (2007) and estimated with the software FreeNA (http://wwwl.montpellier.inra.fr/URLB/). The distribution of nuclear variation (AMOVA— Excoffier et al. 1992) was analyzed for the major lineages identified and the clusters detected with Geneland.

Nuclear gene flow between mtDNA lineages was estimated using the Bayesian inference available in MIGRATE-n 3.0 (Beerli 2008). The parameters estimated with this software were θ (which is xNeμ) and Mt (which is the ratio of immigration and mutation rates, mi/μ) and the migration estimate is expressed as 4μM (Beerli 2008). The mutation rate 550 was assumed to be constant for all microsatellite loci, and initial values for theta and migration were obtained from FSt estimates. The Bayesian run consisted of 1 long chain with 1 million recorded steps and sampling increments of 100 generations. A total of 10 million genealogies (recorded steps multiplied by the sampling increment) were sampled, and the first 10,000 were discarded (burn-in).

Demography.—The historical demography of the lineages was explored with mtDNA. The mutation rate of the mitochondrial control region of D. novemcinctus was calculated because it is important in the estimation of demographic processes. It was approximated using the formula d = (tv + tvR)lm, where d is the number of nucleotide substitutions per site, ty is the number of transversions between species, R is the transversion/transition ratio within the focal species, and m is the length of the sequence (Rooney et al. 2001). The closely related congener D. kappleri was used as the outgroup to estimate the d-value. The rate of nucleotide substitution per site per year is π = d/2T, where T is the divergence time between the ingroup and the outgroup species. This was calculated using the estimated divergence time of 7 my a calculated by Delsuc et al. (2004). The rate of nucleotide substitution per site and generation is μ = πg (Rooney et al. 2001), where g is the generation time, which in nine-banded armadillos has been estimated at approximately 5 years (Abba and Superina 2010). The nucleotide substitution rate per haplotype (u) was calculated by multiplying the length (m) of the sequence by μ (Rooney et al. 2001).

Historical demographic trends within the main mtDNA lineages were estimated with Fu's neutrality Fs test (Fu 1997) and mismatch distribution implemented in Arlequin 3.11 (Excoffier et al. 2005). Negative values of Fu's test are expected in lineages that have undergone recent expansion, whereas positive values are expected in lineages that have recently experienced bottlenecks. Values are expected to be near zero in stable lineage sizes. Using the mismatch distribution, the time of expansion (t) was estimated from t = τ/2u (Rogers and Harpending 1992), where τ measures time in generations and u is the mutational rate per generation for the DNA sequence. The statistical significance value assessing the neutral model fit was calculated (Navascués et al. 2006).

Results

Structure and genetic diversity in mtDNA.—The best-fit evolution model for the control region of D. novemcinctus was HKY + I + G. Fifty-nine polymorphic sites were identified within 408 bp of the control region, resulting in 70 unique haplotypes in the 138 Mexican samples (Appendix II). Nucleotide diversity (π) and haplotype diversity (h) were high, 0.029 ±0.002 and 0.977 ±0.005, respectively. Thirteen haplotypes were identified in 19 samples from other countries. Two of these haplotypes (from Colombia and the United States) were shared with the Mexican samples.

The phylogenetic tree (Fig. 1) and haplotype network (Fig. 2) separated the haplotypes from Mexico into 2 major lineages corresponding to samples from the central-western (hereafter called West) part of the country, and the southeast (hereafter called East). In the phylogenetic tree, both lineages were monophyletic and supported by high posterior probabilities. One haplotype from Costa Rica and 4 from Colombia were nested within the West lineage, whereas 1 haplotype from Guatemala, 1 from Costa Rica, 2 from Nicaragua, and 2 from Colombia nested within the East lineage.

Fig. 1

Bayesian phylogenetic tree of haplotypes of Dasypus novemcinctus based on 408 bp of the mitochondrial control region, showing the 2 major lineages in Mexico: West (gray) and East (black). Labels are haplotype identification numbers (see Appendix II). Numbers above the branches indicate that the Bayesian posterior probabilities for the key node are >0.8. Arrows indicate haplotypes from Guatemala, Nicaragua, Costa Rica, Colombia, and the United States. Arrows within circles indicate haplotypes shared between Mexico and other countries. The tree is rooted with 3 sequences of D. kappleri.

Fig. 1

Bayesian phylogenetic tree of haplotypes of Dasypus novemcinctus based on 408 bp of the mitochondrial control region, showing the 2 major lineages in Mexico: West (gray) and East (black). Labels are haplotype identification numbers (see Appendix II). Numbers above the branches indicate that the Bayesian posterior probabilities for the key node are >0.8. Arrows indicate haplotypes from Guatemala, Nicaragua, Costa Rica, Colombia, and the United States. Arrows within circles indicate haplotypes shared between Mexico and other countries. The tree is rooted with 3 sequences of D. kappleri.

Fig. 2

Median-joining network based on 408 bp of the mitochondrial control region from 157 samples of Dasypus novemcinctus, showing the 2 major lineages: West (gray circles) and East (white circles). Circle size is proportional to haplotype frequencies; line length is roughly proportional to estimated number of mutation steps between haplotypes. Black solid circles represent unsampled or extinct haplotypes. Arrows indicate haplotypes from Guatemala, Nicaragua, Costa Rica, Colombia, and the United States. Arrows within circles indicate haplotypes shared between Mexico and other countries.

Fig. 2

Median-joining network based on 408 bp of the mitochondrial control region from 157 samples of Dasypus novemcinctus, showing the 2 major lineages: West (gray circles) and East (white circles). Circle size is proportional to haplotype frequencies; line length is roughly proportional to estimated number of mutation steps between haplotypes. Black solid circles represent unsampled or extinct haplotypes. Arrows indicate haplotypes from Guatemala, Nicaragua, Costa Rica, Colombia, and the United States. Arrows within circles indicate haplotypes shared between Mexico and other countries.

In Mexico, the 2 lineages exhibited an almost allopatric distribution, apparently separated by major mountain chains (Fig. 3). Four haplotypes from the East lineage were recorded in the geographic area occupied by the West lineage, and 1 haplotype from West was recorded in the area occupied by East lineage. Mitochondrial genetic diversity within each lineage was high (Table 1), and the average p-distance between them was 4.70%. The AMOVA showed that most of the genetic variance was found between lineages (72.6%; Table 2a), whereas within-lineage variance was relatively low (27.4%; Table 2a).

Fig. 3

Spatial location of the 157 samples from Dasypus novemcinctus analyzed for the mitochondrial control region (408 bp). Gray squares: West lineage samples. Black dots: East lineage samples. The geographic position of the Sierra Madre Oriental in Mexico is suggested as a boundary between the lineages in this country.

Fig. 3

Spatial location of the 157 samples from Dasypus novemcinctus analyzed for the mitochondrial control region (408 bp). Gray squares: West lineage samples. Black dots: East lineage samples. The geographic position of the Sierra Madre Oriental in Mexico is suggested as a boundary between the lineages in this country.

Table 1

Genetic variation estimates, Fu's Fstest, and mismatch distribution in the 2 major lineages, West and East, based on 408 bp of the mitochondrial control region from 138 Mexican samples of Dasypus novemcinctus.Parameters of the sudden expansion are presented as well as goodness-of-fit tests of the model, sum of squared deviations (SSD),time in generations since the sudden expansion occurred (τ [tau]—Rogers and Harpending 1992), and their corresponding confidence intervals.

Control region mtDNA West lineage East lineage 
Sample size (n54 84 
Nucleotide diversity (π) 0.009 0.014 
Haplotypic diversity (h0.93 0.968 
No. haplotypes 23 47 
Mean no. pair-wise difference (K3.922 6.069 
FU's Fstest (P-value) −25.932 (0.0001) 25.202 (0.0001) 
τ (confidence intervals) 3.293 (1.654–10.634) 6.289 (4.420–9.877) 
SSD(P-value) 0.009 (0.900) 0.004 (1.000) 
Control region mtDNA West lineage East lineage 
Sample size (n54 84 
Nucleotide diversity (π) 0.009 0.014 
Haplotypic diversity (h0.93 0.968 
No. haplotypes 23 47 
Mean no. pair-wise difference (K3.922 6.069 
FU's Fstest (P-value) −25.932 (0.0001) 25.202 (0.0001) 
τ (confidence intervals) 3.293 (1.654–10.634) 6.289 (4.420–9.877) 
SSD(P-value) 0.009 (0.900) 0.004 (1.000) 
Table 2

Hierarchical AMOVA based on a) 408 bp of the mitochondrial control region and b) 5 nuclear microsatellite loci. Percentage of variation associated to each level and fixation index are given CP-values are given in parentheses).

a) Source of variation mtDNA  
Major lineages (mtDNA)   
Among lineages (West versus East) 72.62 (FsT−0.72, P < 0.0001) 
Within lineages 27.38  
b) Source of variation nDNA  
Major lineages (mtDNA)   
Among lineages (West versus East) 2.74 (FIS0.027, P < 0.0001) 
Among Individuals within lineages 24.17 (F/s0.248, P < 0.0001) 
Within individuals 73.09 (FIT= 0.269, P < 0.0001) 
Clusters defined by Geneland (nDNA)   
Among clusters (K = 10) 12.67 (FST0.153, P< 0.0001) 
Among individuals within clusters 13.36 (FIS0.126, P< 0.0001) 
Within individuals 73.97 (FIT0.260, P < 0.0001) 
a) Source of variation mtDNA  
Major lineages (mtDNA)   
Among lineages (West versus East) 72.62 (FsT−0.72, P < 0.0001) 
Within lineages 27.38  
b) Source of variation nDNA  
Major lineages (mtDNA)   
Among lineages (West versus East) 2.74 (FIS0.027, P < 0.0001) 
Among Individuals within lineages 24.17 (F/s0.248, P < 0.0001) 
Within individuals 73.09 (FIT= 0.269, P < 0.0001) 
Clusters defined by Geneland (nDNA)   
Among clusters (K = 10) 12.67 (FST0.153, P< 0.0001) 
Among individuals within clusters 13.36 (FIS0.126, P< 0.0001) 
Within individuals 73.97 (FIT0.260, P < 0.0001) 

Structure, genetic diversity, and gene flow in nuclear microsatellites.—Sixty-three alleles were found for the 5 nDNA microsatellites (12, 20, 9, 10, and 12 for Dnovl, Dnov6, Dnov7, Dnov16, and Dnov24, respectively). Overall allelic richness was 11.7. No evidence of scoring errors due to stuttering or large allele dropout was found (Van Oosterhout et al. 2004). Null alleles, however, were detected in 2 loci (Dnovl6 and Dnov24), and significant linkage disequilibrium (P < 0.05) was found in 53% of all possible combinations.

The posterior distribution of the estimated number of clusters in Geneland revealed 10 distinct clusters. Individuals assigned to the same genetic cluster were not always spatially contiguous; 7 clusters contained members that were isolated by large intervening areas. A significant difference in the proportion of individuals from each mitochondrial lineage in the 10 Geneland clusters also was found (x2 = 64.335, P < 0.0001).

Overall, the observed heterozygosity values were lower than expected, and high allelic richness was observed in the 2 lineages (Table 3). The inbreeding coefficient was positive and significant for each of the lineages, and for the total sample. The presence of null alleles within the sample did not solely generate positive inbreeding coefficients because FISestimates in loci that did not show null alleles were also positive (Dnov1: 0.006; Dnov6: 0.100; Dnov7: 0.232). The FSTbetween lineages was small and different from zero, even when correcting for bias (with Excluding Null Alleles correction: FST = 0.020, 95% confidence interval [95% CI] = 0.013–0.029; without correction: FST = 0.027, 95% CI = 0.014-0.043). Global RST between lineages was 0.066 and Dest was 0.100. For the 2 different division scenarios (2 mitochondrial lineages and the clusters defined by Geneland), AMOVA revealed that most of the genetic variance in the nuclear microsatellites was explained by differences among individuals within the total sample and among individuals within lineages or clusters, whereas the difference between mitochondrial lineages and among nuclear clusters was lower, but still significant (Table 2b).

Table 3

Nuclear DNA genetic variation estimates in the 2 major lineages based on 5 autosomal microsatellite loci from 116 Mexican samples of Dasypus novemcinctus.

Nuclear microsatellites West lineage East lineage Total 
Sample size 42 74 116 
Allelic richness 10.16 11.54 12.6 
Inbreeding coefficient 0.314** 0.264** 0.249** 
Mean observed heterozygosity 0.551** 0.619** 0.617** 
Mean expected heterozygosity 0.8 0.84 0.821 
 
Nuclear microsatellites West lineage East lineage Total 
Sample size 42 74 116 
Allelic richness 10.16 11.54 12.6 
Inbreeding coefficient 0.314** 0.264** 0.249** 
Mean observed heterozygosity 0.551** 0.619** 0.617** 
Mean expected heterozygosity 0.8 0.84 0.821 
 
**

P <0.005.

High rates of nuclear gene flow between the 2 mtDNA lineages were found. The estimated number of immigrants from East to West was 34.234 (95% CI = 5.395–59.341) individuals per generation, and from West to East was 20.01 (95% CI = 1.10–39.560) individuals per generation.

Demography.—The number of nucleotide substitutions per site (d) in the mtDNA control region was 0.264 (tv = 33, R = 1.86, m = 357 bp). Based on the average divergence time calculated by Delsuc et al. (2004), the rate of nucleotide substitutions per site per lineage per year (π) was 1.90 × 10−8 (interval 2.6 x 10−8-2.4 × 10−8). Considering a generation time of 5 years, the rate of nucleotide substitutions per site per generation (μ) was 9.5 × 10−8, and for the 408 bp used, the nucleotide substitutions per haplotype per generation (u) was 3.8 × 10∼5.

Mismatch distribution for each lineage was bell-shaped, as was expected under the sudden expansion model (Fig. 4). The observed distribution was not significantly different from the expected distribution (Table 1). Average expansion time was estimated at 43,328 years (95% CI = 21,763-139,921 years calculated with τ) for the West lineage, and 82,750 years (95% CI = 58,157–129,960 years) for the East lineage. The x values and their confidence intervals for each lineage are presented in Table 1. Fu's Fs values were negative and significantly different from zero (Table 1), further supporting a demographic expansion for both lineages.

Fig. 4

Mismatch distribution for the West and East lineages of Dasypus novemcinctus based on 408 bp of the mitochondrial control region from 138 Mexican samples. Bars: pair-wise differences; solid lines: expected values under the sudden expansion model.

Fig. 4

Mismatch distribution for the West and East lineages of Dasypus novemcinctus based on 408 bp of the mitochondrial control region from 138 Mexican samples. Bars: pair-wise differences; solid lines: expected values under the sudden expansion model.

Discussion

The genetic structure of a species in a newly colonized area is strongly influenced by the different lineages that arrived during the increase in the distributional range. Several introductions from multiple founding sources have been detected in a wide range of invasive animals, plants, and fungi (Dlugosch and Parker 2008), and for species dispersing naturally into new areas (Hewitt 2011). Our study shows that the genetic structure of the nine-banded armadillo in Mexico, corresponds to 2 divergent mitochondrial lineages that are reciprocally monophyletic. The presence of haplotypes from both lineages in Central America and Colombia reveals a wider distribution of these genetic groups. Given the clear South American origin of the nine-banded armadillo (Eisenberg 1981; Simpson 1980), such evidence implies that these lineages differentiated prior to their arrival to Mexico and the current allopatric distribution in Mexico suggests independent northward colonization of each lineage. Nevertheless, more samples from a wider area are needed to test hypotheses about migration patterns, assess the geographic distribution of these lineages, and estimate the divergence time between them.

The allopatric distribution of the lineages in Mexico corresponds with the geographic distribution of the 2 subspecies of nine-banded armadillo described in this country. D. novemcinctus mexicanus is found in eastern Mexico, whereas D. n. davisi is restricted to western Mexico, occurring from the Balsas Basin north to Morelos (McBee and Baker 1982; Wetzel et al. 2007). However, the presence of both lineages in Costa Rica and Colombia, where other subspecies have been recorded, indicates that genetic information does not confirm the subspecies traditionally recognized based in geographical distribution and morphological data (McBee and Baker 1982; Wetzel et al. 2007).

Founder events are a hallmark of colonization and are expected to result in reduced genetic variation of founding populations (Wilson et al. 2009). Conversely, the nucleotide and haplotype diversity of the mtDNA control region of Mexican armadillos was high. It also was much greater than that recorded in the United States population (Huchon et al. 1999). According to our phylogeny, United States populations appear to be derived from the East lineage of Mexico. Nevertheless, this assumption needs to be tested with larger sample sizes, because we only analyzed 2 individuals from the United States. The low diversity levels reported for populations in the United States could be explained by a recent founder event in which there has been little time for mutation to generate new diversity. In contrast, the lineages in Mexico probably had either an initially large population size, or they recovered from their own founder event a long time ago.

A weak genetic structure in Mexican armadillos was revealed by microsatellites. Recent migration between the 2 mitochondrial lineages was inferred from the few differences in allelic frequencies between them (FST). Having a long time to accumulate differences between lineages can generate many private alleles (Dest), and a large variance in allele size in each lineage (Rst)- The significant difference estimated between the assignments of individuals to mitochondrial and nuclear clusters suggests that the patterns of genetic structure we uncovered are an accurate record of the species' history and not a reflection of stochastic effects limited to 1 gene, although some marker-specific effects are probable.

The high nuclear variation found in Mexican armadillos supports the idea that there is a large effective population size in Mexico or that the lineages have already recovered from a founder effect, or both. Each major lineage occurs across a large geographic area in Mexico, and shows substructure within it. The mixing of individuals from populations with different allelic frequencies can lead to a Wahlund effect, generating Hardy-Weinberg disequilibrium and a positive inbreeding coefficient (Hedrick 2005). Positive inbreeding coefficients can also be generated by endogamy (Hedrick 2005). This possibility should be explored in future studies of these populations. Regarding the linkage disequilibrium detected, in an analysis of 7 populations of armadillos in the United States, Loughry et al. (2009) observed no evidence of genotypic disequilibrium in the 5 autosomal microsatellites used in this study. Thus, it seems plausible that disequilibrium found was due to demographic changes unique to Mexico (Rafalski and Morgante 2004).

Different patterns of genetic structure among different markers can be caused by numerous factors, including those specific to the marker (mutation rate and ploidy) and those related to the life-history of the organism. Mitochondrial DNA, because it is matrilinearly inherited, is expected to show different patterns than nuclear genes if females and males have different dispersal patterns or rates. In our study, we found less genetic structure in microsatellites compared to mtDNA. Although genetic structure between the 2 types of markers was correlated, there was evidence for more gene flow in nuclear microsatellites. There are at least 3 nonexclusive explanations for the discrepancy between nuclear and mitochondrial patterns. First, the larger effective population size of nDNA compared to mtDNA could lead to less gene fixation through drift, and therefore less divergence and genetic structure in nDNA compared to mtDNA (Hare 2001). Although this could partially explain the lack of clear structure in microsatellites, the coalescence-based program MIGRATE (which assesses lineage sorting issues versus migration) indicated high levels of gene flow, suggesting that differences in effective population size alone were not the only explanation. Second, a slower mutation rate in nDNA could lead to less genetic structure in this marker; but microsatellite mutation rates are higher than those of mtDNA (Hedrick 2005). However, the most likely explanation is the differential mode of inheritance of the markers, combined with male-biased dispersal. We recognize that the fast expansion of nine-banded armadillos in the United States should be the result of extensive dispersion by both sexes. Nevertheless, we suggest that females show more philopatry than males and that this behavior maintains a stronger structure in mtDNA. The same pattern has been found in other mammalian species (Engelhaupt et al. 2009; Gauffre et al. 2009; Turmelle et al. 2011). Specifically, it has been recorded for some carnivores that increased their distributional range after the emergence of the Isthmus of Panama, and covered extensive areas in the newly colonized continent (Eizirik et al. 2001; Tchaicka et al. 2007).

Contrary to our inference from genetic data, a mark- recapture study (Loughry and McDonough 2001) suggested that in armadillos, juvenile males were more likely to remain in their natal populations than were females, although differences in dispersal compared to philopatry were not specifically tested. Frutos and Van Den Bussche (2002) assessed genetic variation in armadillo populations from Paraguay and argued that examination of their data indicated,, a high level of female dispersal. However, they used cytochrome b (mtDNA), and they did not compare genetic structure between nDNA and mtDNA markers. The haplotype shared by different populations from Paraguay could be the result of incomplete lineage sorting. Alternatively, the populations could belong to a single genetic lineage, whereas the inference proposed by us is based on 2 divergent lineages 555 that underwent secondary contact. Future studies using markers associated with the Y chromosome along with detailed field data will provide additional insight into dispersal patterns.

The mountain chains of central Mexico have clearly played an important role in the maintenance of genetic structuring of armadillos, consistent with phylogeographic studies of other mammals and birds (Leon-Paniagua et al 2007; McCormack et al. 2008; Sullivan et al. 2000). Mammalian studies, however, have only used mitochondrial markers, making it difficult to assess to what degree gene flow in nuclear genes might occur. As shown by our study, an assessment that also includes nuclear markers is necessary to fully understand whether the observed pattern is a consequence of stochastic factors or life-history traits. Examination of our data suggests that the Sierra Madre Oriental affected the genetic structure of armadillos, and probably the Tehuantepec Isthmus may act as a corridor for gene flow between lineages (Arteaga et al. 2011).

Population dynamics during the increase of a species range involve a change in effective population size. When a species arrives in a new area, it usually goes through demographic expansion once individuals have overcome potentially stressful biotic and abiotic environmental conditions (Sexton et al. 2009). The high genetic diversity in the 2 Mexican lineages of armadillos, despite the demographic changes detected in them, suggests the presence of a moderately large ancestral population size. At least in the last 100,000 years, both lineages underwent a constant expansion. The nine-banded armadillo is an ecologically tolerant species that occurs in a variety of environmental conditions, which suggests that climate cycling in the Pleistocene did not strongly influence its demographic history.

Our results demonstrate that 2 different lineages of D. novemcinctus arrived in Mexico from the south and colonized disjunct parts of North America. We suggest that females are philopatric, whereas males maintain a connection among populations through individual dispersal. Further studies that include samples from South American regions would be useful to complete the continent-wide structuring and genetic diversity patterns of armadillos. Such studies could reveal important information about the time and place of their evolutionary origin. Such sampling might also help to determine the time when armadillos crossed the Isthmus of Panama, and when they arrived in Mexico. The Great American Biotic Interchange influenced community and genetic structure of many species in the Americas through extended bouts of migration in both directions (Marshall et al. 1982; Pascual and Ortíz-Jaureguizar 2007; Webb 1976). This study clearly documents the consequences of the Great American Biotic Interchange in influencing genetic patterns in the nine-banded armadillo in a colonized area.

Acknowledgments

We thank 2 anonymous reviewers for constructive suggestions that improved the manuscript. We thank all the persons and institutions that provided biological samples for this study (Appendix I). We thank the Direction General de Vida Silvestre for the permits to collect samples of armadillos. We are grateful to O. Gaona, E. Aguirre, and L. Espinosa for their help with laboratory analyses and technical support. We thank M. Superina, J. McCormack, and R. Bello for valuable comments and English editing. This work is part of MCA's PhD dissertation in biological sciences at the Universidad Nacional Autónoma de México. MCA is grateful to the graduate program Doctorado en Ciencias Biológicas, to Universidad Nacional Autónoma de México for the scholarship granted for doctoral studies, and to Sigma Xi and the American Society of Mammalogists for research grants. The manuscript was finalized when RAM was on sabbatical at the Arizona-Sonora Desert Museum and the University of Arizona and he thanks Dirección General de Asuntos del Personal Académico-Universidad Nacional Autónoma de México for support.

Appendix I

Samples of Dasypus novemcinctusobtained from different museums.

Collection names Collection numbers of specimensa Total samples 
Mammal collection of American Museum of Natural History, United States (AMNH) 26007, 171919, 171921, 207420, 24053, 24054, 176676, 176675, 7278, 182075, 14663, 33148, 15463, 139318 14 
Colección Zoológica Regional, Instituto de Historia Natural y Ecología, México (CZRMA) 1581, 2255, 43, 50 
Colección de la Reserva de la Biósfera Los Tuxtlas, México A, B 
Colección Mastozoológica del Sureste de México, México (ECOSUR-SC) 52, 575, 980, 1270, 1570, 1572 
Colección de Mamíferos del Museo de Zoología “Alfonso L. Herrera,” Mexico (MZFC) 4254, 5078, 5079, 5075, 4900, 5077 
Colección Mastozoológica Instituto Tecnológico Agropecuario de Hidalgo, México 540, 544, 536, 543, 547, 534, 541, 535, 545, 531, 539, 537, 546, 532, 548 15 
Colección Nacional de Mamíferos, Instituto de Biología, Universidad Nacional Autónoma de México, México (IBUNAM) 1153, 3592, 10069, 11535, 17037, 16559, 31600, 43121, 16520, 15583, 27275, 16558, 14528, 43122, 14527, 37069, 16551, 16496 18 
Colección de Mamíferos, Escuela Nacional de Ciencias Biológicas, Instituto Politécnico Nacional, México (ENCB) 40868, 16019, 21096, 39110, 35629, 15099, 26577, 21669, 26067, 36004, 34214 11 
Colección de Mamíferos de la Universidad Michoacana San Nicolás de Hidalgo, México (UMSNH) 1323, 394, 887, 1983, C 
Mammal collection of Smithsonian National Museum of Natural History, United States (USMN) 553928, 281290, 281285, 281291, 281288, 337563 
Total  86 
Collection names Collection numbers of specimensa Total samples 
Mammal collection of American Museum of Natural History, United States (AMNH) 26007, 171919, 171921, 207420, 24053, 24054, 176676, 176675, 7278, 182075, 14663, 33148, 15463, 139318 14 
Colección Zoológica Regional, Instituto de Historia Natural y Ecología, México (CZRMA) 1581, 2255, 43, 50 
Colección de la Reserva de la Biósfera Los Tuxtlas, México A, B 
Colección Mastozoológica del Sureste de México, México (ECOSUR-SC) 52, 575, 980, 1270, 1570, 1572 
Colección de Mamíferos del Museo de Zoología “Alfonso L. Herrera,” Mexico (MZFC) 4254, 5078, 5079, 5075, 4900, 5077 
Colección Mastozoológica Instituto Tecnológico Agropecuario de Hidalgo, México 540, 544, 536, 543, 547, 534, 541, 535, 545, 531, 539, 537, 546, 532, 548 15 
Colección Nacional de Mamíferos, Instituto de Biología, Universidad Nacional Autónoma de México, México (IBUNAM) 1153, 3592, 10069, 11535, 17037, 16559, 31600, 43121, 16520, 15583, 27275, 16558, 14528, 43122, 14527, 37069, 16551, 16496 18 
Colección de Mamíferos, Escuela Nacional de Ciencias Biológicas, Instituto Politécnico Nacional, México (ENCB) 40868, 16019, 21096, 39110, 35629, 15099, 26577, 21669, 26067, 36004, 34214 11 
Colección de Mamíferos de la Universidad Michoacana San Nicolás de Hidalgo, México (UMSNH) 1323, 394, 887, 1983, C 
Mammal collection of Smithsonian National Museum of Natural History, United States (USMN) 553928, 281290, 281285, 281291, 281288, 337563 
Total  86 
a

A, B, and C are specimens without assigned numbers.

Appendix II

Control region haplotypes identified in 157 samples of Dasypus novemcinctusfrom Mexico, Colombia, Costa Rica, Guatemala, Nicaragua, and the United States.

Haplotype GenBank accession no. No. samples Mexican states from sample's origin (other countries from sample's origin) 
H1 JN602096 Quintana Roo 
H2 JN602097 Jalisco 
H3 JN602098 Veracruz 
H4 JN602099 Jalisco 
H5 JN602100 México, Guerrero 
H6 JN602101 Guerrero 
H7 JN602102 12 Morelos, Hidalgo, Colima, México, 
   Guerrero, Querétaro 
H8 JN602169 Jalisco 
H9 JN602103 Oaxaca 
H10 JN602104 Sinaloa 
H11 JN602105 Tamaulipas, Hidalgo, México 
H12 JN602106 Colima, Chiapas 
H13 JN602107 Nayarit 
H14 JN602108 Quintana Roo 
H15 JN602109 11 Quintana Roo, Chiapas, Veracruz 
H16 JN602110 Quintana Roo 
H17 JN602111 Quintana Roo, Chiapas 
H18 JN602112 Quintana Roo 
H19 JN602113 Chiapas 
H20 JN602114 Chiapas 
H21 JN602115 Chiapas 
H22 JN602116 Nuevo León, Chiapas 
H23 JN602117 Chiapas 
H24 JN602118 Chiapas 
H25 JN602119 Chiapas 
H26 JN602120 Chiapas 
H27 JN602121 Chiapas 
H28 JN602122 Chiapas 
H29 JN602123 Chiapas 
H30 JN602124 Hidalgo 
H31 JN602125 Hidalgo 
H32 JN602170 Hidalgo 
H33 JN602126 Hidalgo 
H34 JN602127 Hidalgo 
H35 JN602128 Hidalgo 
H36 JN602129 Hidalgo, Michoacá;n 
H37 JN602130 Jalisco, Colima 
H38 JN602131 Mexico, Guerrero, Morelos 
H39 JN602132 Mexico 
H40 JN602133 Mexico, Michoacá Guerrero 
H41 JN602134 Oaxaca 
H42 JN602135 Oaxaca 
H43 JN602136 Oaxaca 
H44 JN602137 Oaxaca 
H45 JN602138 Oaxaca 
H46 JN602139 Nayarit 
H47 JN602140 Nayarit 
H48 JN602172 Guerrero, Michoacá 
H49 JN602141 Guerrero, Michoacá 
H50 JN602142 Michoacá 
H51 JN602143 Veracruz 
H52 JN602144 Veracruz 
H53 JN602145 Tamaulipas 
H54 JN602146 Puebla (Colombia) 
H55 JN602147 Puebla 
H56 JN602148 Guerrero 
H57 JN602149 Guerrero 
H58 JN602150 Guerrero 
H59 JN602151 Guerrero 
H60 JN602152 Guerrero 
H61 JN602153 Guerrero 
H62 JN602154 Morelos 
H63 JN602155 Sinaloa 
H64 JN602156 Yucatá;n 
H65 JN602157 Veracruz 
H66 JN602158 Jalisco 
H67 JN602159 Hidalgo 
H68 JN602160 Hidalgo 
H69 JN602161  Hidalgo (United States) 
H70 JN602162 Veracruz 
H71 JN602163 (Costa Rica) 
H72 JN602164 (Costa Rica) 
H73 JN602165 (Nicaragua) 
H74 JN602171 (Colombia) 
H75 JN602166 (Colombia) 
H76 JN602167 (Colombia) 
H77 JN602168  (Colombia) 
H78 JN602173 (Colombia) 
H79 JN602174 (Colombia) 
H80 JN602175 (Nicaragua) 
H81 JN602176 1 (Guatemala) 
Haplotype GenBank accession no. No. samples Mexican states from sample's origin (other countries from sample's origin) 
H1 JN602096 Quintana Roo 
H2 JN602097 Jalisco 
H3 JN602098 Veracruz 
H4 JN602099 Jalisco 
H5 JN602100 México, Guerrero 
H6 JN602101 Guerrero 
H7 JN602102 12 Morelos, Hidalgo, Colima, México, 
   Guerrero, Querétaro 
H8 JN602169 Jalisco 
H9 JN602103 Oaxaca 
H10 JN602104 Sinaloa 
H11 JN602105 Tamaulipas, Hidalgo, México 
H12 JN602106 Colima, Chiapas 
H13 JN602107 Nayarit 
H14 JN602108 Quintana Roo 
H15 JN602109 11 Quintana Roo, Chiapas, Veracruz 
H16 JN602110 Quintana Roo 
H17 JN602111 Quintana Roo, Chiapas 
H18 JN602112 Quintana Roo 
H19 JN602113 Chiapas 
H20 JN602114 Chiapas 
H21 JN602115 Chiapas 
H22 JN602116 Nuevo León, Chiapas 
H23 JN602117 Chiapas 
H24 JN602118 Chiapas 
H25 JN602119 Chiapas 
H26 JN602120 Chiapas 
H27 JN602121 Chiapas 
H28 JN602122 Chiapas 
H29 JN602123 Chiapas 
H30 JN602124 Hidalgo 
H31 JN602125 Hidalgo 
H32 JN602170 Hidalgo 
H33 JN602126 Hidalgo 
H34 JN602127 Hidalgo 
H35 JN602128 Hidalgo 
H36 JN602129 Hidalgo, Michoacá;n 
H37 JN602130 Jalisco, Colima 
H38 JN602131 Mexico, Guerrero, Morelos 
H39 JN602132 Mexico 
H40 JN602133 Mexico, Michoacá Guerrero 
H41 JN602134 Oaxaca 
H42 JN602135 Oaxaca 
H43 JN602136 Oaxaca 
H44 JN602137 Oaxaca 
H45 JN602138 Oaxaca 
H46 JN602139 Nayarit 
H47 JN602140 Nayarit 
H48 JN602172 Guerrero, Michoacá 
H49 JN602141 Guerrero, Michoacá 
H50 JN602142 Michoacá 
H51 JN602143 Veracruz 
H52 JN602144 Veracruz 
H53 JN602145 Tamaulipas 
H54 JN602146 Puebla (Colombia) 
H55 JN602147 Puebla 
H56 JN602148 Guerrero 
H57 JN602149 Guerrero 
H58 JN602150 Guerrero 
H59 JN602151 Guerrero 
H60 JN602152 Guerrero 
H61 JN602153 Guerrero 
H62 JN602154 Morelos 
H63 JN602155 Sinaloa 
H64 JN602156 Yucatá;n 
H65 JN602157 Veracruz 
H66 JN602158 Jalisco 
H67 JN602159 Hidalgo 
H68 JN602160 Hidalgo 
H69 JN602161  Hidalgo (United States) 
H70 JN602162 Veracruz 
H71 JN602163 (Costa Rica) 
H72 JN602164 (Costa Rica) 
H73 JN602165 (Nicaragua) 
H74 JN602171 (Colombia) 
H75 JN602166 (Colombia) 
H76 JN602167 (Colombia) 
H77 JN602168  (Colombia) 
H78 JN602173 (Colombia) 
H79 JN602174 (Colombia) 
H80 JN602175 (Nicaragua) 
H81 JN602176 1 (Guatemala) 

Literature Cited

Abba
A. M.
Superina
M.
.
2010
.
The 2009/2010 Armadillo red list assessment
.
Edentata
 
11
:
135
184
.
Álvarez
T
.
1974
.
Identificación de los huesos rescatados en la segunda temporada en Santa Martha Chiapas (Cala HH/KK)
. Pp.
24
29
in
Excavaciones en el abrigo de Santa Marta Chiapas
 .
Serie Informes, Departamento de Prehistoria, Instituto Nacional de Antropología e Historia
,
México, Distito Federal, México
.
Arteaga
M. C
McCormack
J.
Eguiarte
L. E.
Medellín
R.
.
2011
.
Genetic admixture in multidimensional environmental space: asymmetrical niche similarity promotes gene flow in armadillos (Dasypus novemcinctus)
.
Evolution
 
65
:
2470
2480
.
Audubon
J. J.
Bachman
J.
.
1854
.
Quadrupeds of North America
 .
Privately published by J. J. Audubon
,
New York
.
Bandelt
H. J.
Forster
P.
ROhl
A.
.
1999
.
Median-joining networks for inferring intraspecific phylogenies
.
Molecular Biology and Evolution
 
16
:
37
48
.
Beerli
P.
2008
.
Migrate version 3.0: a maximum likelihood and Bayesian estimator of gene flow using the coalescent
 . http://popgen.sc.fsu.edu/Migrate/Migrate-n.
Accessed 20 March 2011
.
Braaker
S.
Heckel
G.
.
2009
.
Transalpine colonisation and partial phylogeographic erosion by dispersal in the common vole (Microtus arvalis)
.
Molecular Ecology
 
18
:
2518
2531
.
Carlini
A. A.
Vizcaíno
S. F.
Scillato-Yané
G J.
.
1997
.
Armoured xenarthrans: a unique taxonomic and ecologie assemblage
. Pp.
213
226
in
Vertebrate paleontology of the Miocene Honda Group, Republic of Colombia
  (
Kay
R. F.
Cifelli
R. L.
Flynn
J. J.
Madden
R. H.
, eds.).
Smithsonian Institution Press
,
Washington, D.C
.
Cassens
I.
Mardulyn
P.
Milinkovitch
M. C.
.
2005
.
Evaluating intraspecific “network” construction methods using simulated sequence data: do existing algorithms outperform the global maximum parsimony approach?
Systematic Biology
 
54
:
363
372
.
Chapuis
M. P.
Estoup
A.
.
2007
.
Microsatellite null alleles and estimation of population differentiation
.
Molecular Biology and Evolution
 
24
:
621
631
.
Coulon
A.
et al
.
2006
.
Genetic structure is influenced by landscape features: empirical evidence from a roe deer population
.
Molecular Ecology
 
15
:
1669
1679
.
Crawford
N. G
2010
.
SMOGD: software for the measurement of genetic diversity
.
Molecular Ecology Resources
 
10
:
556
557
.
Culver
M.
Johnson
W. E.
Pecon-Slattery
J. S.
O'Brien
S. J.
.
2000
.
Genomic ancestry of the American puma (Puma concolor)
.
Journal of Heredity
 
91
:
186
197
.
J. M
Daza
Smith
E. N.
Páez
V. P.
Parkinson
C. L.
.
2009
.
Complex evolution in the Neotropics: the origin and diversification of the widespread genus Leptodeira (Serpentes: Colubridae)
.
Molecular Phylogenetics and Evolution
 
53
:
653
667
.
Delsuc
F.
Vizcaino
S. F.
Douzery
E. J. P.
.
2004
.
Influence of Tertiary paleoenvironmental changes on the diversification of South American mammals: a relaxed molecular clock study within xenarthrans
.
BMC Evolutionary Biology
 
4
:
11
24
.
Dlugosch
K. M.
Parker
I. M.
.
2008
.
Founding events in species invasions: genetic variation, adaptive evolution, and the role of multiple introductions
.
Molecular Ecology
 
17
:
431
449
.
Drummond
A. J.
Ho
S. Y. W.
Rawlence
N.
Rambaut
A.
.
2007
.
A rough guide to BEAST 1.4
 .
Department of Computer Science, University of Auckland
,
Auckland, New Zealand
.
Drummond
A. J.
Rambaut
A.
.
2007
.
BEAST: Bayesian evolutionary analysis by sampling trees
.
BMC Evolutionary Biology
 
7
:
214
.
Eisenberg
J. F.
1981
.
The mammalian radiations
 .
Chicago University Press
,
Chicago, Illinois
.
Eizirik
E.
Kim
J.
Menotti-Raymond
M.
Crawshaw
P. G.
O'Brien
S. J.
.
2001
.
Phylogeography, population history and conservation genetics of jaguars (Panthera onca, Mammalia, Felidae)
.
Molecular Ecology
 
10
:
65
79
.
Engelhaupt
D.
et al
.
2009
.
Female philopatry in coastal basins and male dispersion across the North Atlantic in a highly mobile marine species, the sperm whale (Physeter macrocephalus)
.
Molecular Ecology
 
18
:
4193
4205
.
Excoffier
L.
Foll
M.
Petit
R. J.
.
2009
.
Genetic consequences of range expansions
.
Annual Review of Ecology, Evolution, and Systematics
 
40
:
481
501
.
Excoffier
L.
Laval
G
Schneider
S.
.
2005
.
Arlequin ver. 3.0: an integrated software package for population genetics data analysis
.
Evolutionary Bioinformatics
 
1
:
47
50
.
Excoffier
L.
Smouse
P. E.
Quattro
J. M.
.
1992
.
Analysis of molecular variance inferred from metric distance among DNA haplotypes: application to human mitochondrial DNA restriction data
.
Genetics
 
131
:
479
491
.
Ficetola
G. F.
Bonin
A.
Miaud
C.
.
2008
.
Population genetics origin and number of founders in biological invasion
.
Molecular Ecology
 
17
:
773
782
.
Frutos
S. D.
Van Den Bussche
R. A.
.
2002
.
Genetic diversity and gene flow in nine-banded armadillos in Paraguay
.
Journal of Mammalogy
 
83
:
815
823
.
Fu
Y.
1997
.
Statistical tests of neutrality of mutations against population growth, hitchhiking and background selection
.
Genetics
 
147
:
915
925
.
Gardner
A. L.
2005
.
Order Cingulata
. Pp.
94
99
in
Mammal species of the world: a taxonomic and geographic reference
  (
Wilson
D. E.
Reeder
D. M.
, eds.).
3
rd ed.
Johns Hopkins University Press
,
Baltimore, Maryland
.
Gauffre
B.
Petit
E.
Brodier
S.
Bretagnolle
V.
Cosson
J. F.
.
2009
.
Sex-biased dispersal patterns depend on the spatial scale in a social rodent. Proceedings of the Royal Society of London, B
.
Biological Sciences
 
276
:
3487
3494
.
Goudet
J.
2001
.
FSTAT, a program to estimate and test gene diversities and fixation indices ver. 2.9.3
 . http://www2.unil.ch/popgen/softwares/fstat.htm.
Accessed 17 October 2010
.
G
Guillot
Mortier
F.
Estoup
A.
.
2005
.
Geneland: a computer package for landscape genetics
.
Molecular Ecology Notes
 
5
:
708
711
.
Guillot
G.
Santos
F.
.
2009
.
A computer program to simulate multilocus genotype data with spatially auto-correlated allele frequencies
.
Molecular Ecology Resources
 
9
:
1112
1120
.
Hàjková
P.
et al
.
2007
.
Genetic structure and evidence for recent population decline in Eurasian otter populations in the Czech and Slovak Republics: implications for conservation
.
Journal of Zoology (London)
 
272
:
1
9
.
Hall
T. A.
1999
.
BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT
.
Nucleic Acids Symposium Series
 
41
:
95
98
Hare
M. P.
2001
.
Prospects for nuclear gene phylogeography
.
Trends in Ecology & Evolution
 
16
:
700
706
.
Hedrick
P. W.
2005
.
Genetics of populations
 .
3
rd ed.
Jones and Bartlett Publishers
,
Sudbury, Massachusetts
.
Hewitt
G. M.
2004
.
Genetic consequence of climatic oscillation in the Quaternary. Philosophical Transactions of the Royal Society of London, B
.
Biological Sciences
 
359
:
183
195
.
Hewitt
G. M.
2011
.
Quaternary phylogeography: the roots of the hybrid zones
.
Genetica
 
139
:
617
638
.
Huchon
D.
Delsuc
F.
Catzeflis
F. M.
Douzery
E. J. P.
.
1999
.
Armadillos exhibit less genetic polymorphism in North America than in South America: nuclear and mitochondrial data confirm a founder effect in Dasypus novemcinctus (Xenarthra)
.
Molecular Ecology
 
8
:
1743
1748
.
Jost
L.
2008
.
GST and its relatives do not measure differentiation
.
Molecular Ecology
 
17
:
4015
4026
.
Leon-Paniagua
L.
Navarro-Sigüenza
A. G.
Hernandez-Banos
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
:
653
664
.
Librado
P.
Rozas
J.
.
2009
.
DnaSP ver. 5: a software for comprehensive analysis of DNA polymorphism data
.
Bioinfor-matics
 
25
:
1451
1452
.
Loughry
W. J.
McDonough
C. M.
.
2001
.
Natal recruitment and adult retention in a population of nine-banded armadillos
.
Acta Theriologica
 
46
:
393
406
.
Loughry
W. J.
Truman
R. W.
McDonough
C. M.
Tilak
M.
Garnier
S.
Delsuc
F.
.
2009
.
Is leprosy spreading among nine-banded armadillos in the southeast United States?
Journal of Wildlife Diseases
 
45
:
144
152
.
Marshall
L. G
Webb
S. D.
Sepkoski
J. J.
Raup
D. M.
.
1982
.
Mammalian evolution and the Great American Interchange
.
Science
 
215
:
1351
1357
.
McBee
K.
Baker
R. J.
.
1982
.
Dasypus novemcinctus
.
Mammalian Species
 
162
:
1
9
.
McCormack
J. E.
Peterson
A. T.
Bonaccorso
E.
Smith
T. B.
.
2008
.
Speciation in the highlands of Mexico: genetic and phenotypic divergence in the Mexican jay (Aphelocoma ultramarina)
.
Molecular Ecology
 
17
:
2505
2521
.
Moncrief
N. D.
1988
.
Absence of genetic variation in a natural population of nine-banded armadillos, Dasypus novemcinctus (Dasypodidae)
.
Southwestern Naturalist
 
33
:
229
231
.
Mondol
S.
Karanth
K. U.
Ramakrishnan
U.
.
2009
.
Why the Indian subcontinent holds the key to global tiger recovery
.
PLoS Genetics
 
5
:
el000585
.
Navascués
M.
Vaxenvanidou
Z.
GonzáEz-Martínez
S.
Climent
J.
Gil
L.
Emerson
B.
.
2006
.
Chloroplast microsatellites reveal colonization and metapopulation dynamics in the Canary Island pine
.
Molecular Ecology
 
15
:
2691
2698
.
Pascual
R.
Ortí-Jaureguizar
E.
.
2007
.
The Gondwanan and South American episodes: two major and unrelated moments in the history of the South American mammals
.
Journal of Mammalian Evolution
 
14
:
75
137
.
Posada
D.
2008
.
jModelTest: phylogenetic model averaging
.
Molecular Biology and Evolution
 
25
:
1253
1256
.
Prodöhl
P.
Loughry
W. J.
McDonough
C. M.
Nelson
W. S.
Avise
J.
.
1996
.
Molecular documentation of polyembryony and the micro-spatial dispersion of clonal sibships in the nine-banded armadillo, Dasypus novemcinctus. Philosophical Transactions of the Royal Society of London, B
.
Biological Sciences
 
263
:
1643
1649
.
Rafalski
A.
Morgante
M.
.
2004
.
Linkage disequilibrium in two genomes of similar size
.
Trends in Genetics
 
20
:
103
111
.
Rambaut
A.
Drummond
A. J.
.
2007
.
Tracer vl.4
 . http://beast.bio.ed.ac.uk/Tracer.
Accessed 13 July 2010
.
Raymond
M.
Rousset
F.
.
1995
.
GENEPOP ver. 1.2: population genetics software for exact tests and ecumenicism
.
Journal of Heredity
 
86
:
248
249
.
Rogers
A. R.
Harpending
H.
.
1992
.
Population growth makes waves in the distribution of pairwise genetic differences
.
Molecular Biology and Evolution
 
9
:
552
569
.
Rooney
A. P.
Honeycutt
R. L.
Derr
J. N.
.
2001
.
Historical population size change of bowhead whales inferred from DNA sequence polymorphism data
.
Evolution
 
55
:
1685
1687
.
Ruedi
M.
Walter
S.
Fischer
M. C.
Scaravelli
D.
Excoffier
L.
Heckel
G.
.
2008
.
Italy as a major ice age refuge area for the bat Myotis myotis (Chiroptera: Vespertilionidae) in Europe
.
Molecular Ecology
 
17
:
1801
1814
.
Sexton
J. P.
McIntyre
R. J.
Angert
A. L.
Rice
K. J.
.
2009
.
Evolution and ecology of species range limits
.
Annual Review of Ecology, Evolution, and Systematics
 
40
:
415
436
.
Sikes
R. S.
Gannon
W. L.
the Animal Care,Use Committee of the American Society of Mammalogists
.
2011
.
Guidelines of the American Society of Mammalogists for the use of wild mammals in research
.
Journal of Mammalogy
 
92
:
235
253
.
Simpson
G. G
1980
.
Splendid isolation
.
The curious history of South American mammals
 .
Yale University Press
,
New Haven, Connecticut
.
Slatkin
M.
1995
.
A measure of population subdivision based on microsatellite allele frequencies
.
Genetics
 
139
:
457
462
.
Sullivan
J.
Arellano
E.
Rogers
D. S.
.
2000
.
Comparative phylogeography of Mesoamerican highland rodents: concerted versus independent response to past climatic fluctuations
.
American Naturalist
 
155
:
755
768
.
Tchaicka
L.
Eizirik
E.
de Oliveira
T. G.
Candido
J. F.
Jr.
Freitas
T. R. O.
.
2007
.
Phylogeography and population history of the crab-eating fox (Cerdocyon thous)
.
Molecular Ecology
 
16
:
819
838
.
Teacher
A. G. F.
Garner
T. W. J.
Nichols
R. A.
.
2009
.
European phylogeography of the common frog (Rana temporaria): routes of postglacial colonization into the British Isles and evidence for an Irish glacial refugium
.
Heredity
 
102
:
490
496
.
Thompson
J. D.
Gibson
T. J.
Plewniak
F.
Jeanmougin
F.
Higgins
D. G.
.
1997
.
The Clustal X Windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools
.
Nucleic Acids Research
 
25
:
4876
4882
.
Tilston
B.
Klicka
J.
.
2010
.
The profound influence of the late Pliocene Panamanian uplift on the exchange, diversification, and distribution of New World birds
.
Ecography
 
33
:
333
342
.
Turmelle
A. S.
Kunz
T. H.
Sorenson
M. D.
.
2011
.
A tale of two genomes: contrasting patterns of phylogeographic structure in a widely distributed bat
.
Molecular Ecology
 
20
:
357
375
.
Van Oosterhout
C.
Hutchinson
W. E.
Wills
D. P. M.
Shipley
P.
.
2004
.
MICRO-CHECKER: software for identifying and correcting genotyping errors in microsatellite data
.
Molecular Ecology Notes
 
4
:
535
538
.
Virgilio
M.
Fauvelot
C.
Costantini
F.
Abbiati
M.
Backeljau
T.
.
2009
.
Phylogeography of the common rag worm Hediste diversicolor (Polychaeta: Nereididae) reveals cryptic diversity and multiple colonization events across its distribution
.
Molecular Ecology
 
18
:
1980
1994
.
Vizcaino
S. F.
Perea
D.
Ubilla
M.
.
1995
.
Presencia de Dasypus (Mammalia, Edentata, Dasypodidae) en la formation Sopas (Pleistoceno tardio) de Uruguay. Distributión cronológica del género
.
Revista Chilena de Historia Natural
 
68
:
95
99
.
Webb
S. D.
1976
.
Mammalian faunal dynamics of the Great American Interchange
.
Paleobiology
 
2
:
216
234
.
Webb
S. D.
1991
.
Ecogeography and the Great American Interchange
.
Paleobiology
 
17
:
266
280
.
Weir
J. T.
Bermingham
E.
Schluter
D.
.
2009
.
The Great American Biotic Interchange in birds
.
Proceedingsof the National Academy of Sciences
 
106
:
21737
21742
.
Wetzel
R. M.
Gardner
A. L.
Redford
K. H.
Eisenberg
J. F.
.
2007
.
Order Cingulata
. Pp.
128
156
in
Mammals of South America, volume 1. Marsupials, xenarthrans, shrews, and bats
  (
Gardner
A. L.
, ed.).
University of Chicago Press
,
Chicago, Illinois
.
Wilson
J. R. U.
Dormontt
E. E.
Prentis
P. J.
Lowe
A. J.
Richardson
D. M.
.
2009
.
Something in the way you move: dispersal pathways affect invasion success
.
Trends in Ecology & Evolution
 
24
:
136
144
.
Wright
S.
1951
.
The genetical structure of populations
.
Annals of Eugenics
 
15
:
323
354
.
Wüster
W.
Ferguson
J. E.
Quijada-Mascareñas
J. A.
Pook
C. E.
Salomão
M. G.
Thorpe
R. S.
.
2005
.
Tracing an invasion: landbridges, réfugia and the phylogeography of the Neotropical rattlesnake (Serpentes: Viperidae: Crotalus durissus)
.
Molecular Ecology
 
14
:
1095
1108
.

Author notes

Associate Editor was Burton K. Lim.