The Evolution of Extreme Polyandry in Social Insects: Insights from Army Ants

Matthias Benjamin Barth; Robin Frederik Alexander Moritz; Frank Bernhard Kraus

doi:10.1371/journal.pone.0105621

Abstract

The unique nomadic life-history pattern of army ants (army ant adaptive syndrome), including obligate colony fission and strongly male-biased sex-ratios, makes army ants prone to heavily reduced effective population sizes (N_e). Excessive multiple mating by queens (polyandry) has been suggested to compensate these negative effects by increasing genetic variance in colonies and populations. However, the combined effects and evolutionary consequences of polyandry and army ant life history on genetic colony and population structure have only been studied in a few selected species. Here we provide new genetic data on paternity frequencies, colony structure and paternity skew for the five Neotropical army ants Eciton mexicanum, E. vagans, Labidus coecus, L. praedator and Nomamyrmex esenbeckii; and compare those data among a total of nine army ant species (including literature data). The number of effective matings per queen ranged from about 6 up to 25 in our tested species, and we show that such extreme polyandry is in two ways highly adaptive. First, given the detected low intracolonial relatedness and population differentiation extreme polyandry may counteract inbreeding and low N_e. Second, as indicated by a negative correlation of paternity frequency and paternity skew, queens maximize intracolonial genotypic variance by increasingly equalizing paternity shares with higher numbers of sires. Thus, extreme polyandry is not only an integral part of the army ant syndrome, but generally adaptive in social insects by improving genetic variance, even at the high end spectrum of mating frequencies.

Citation: Barth MB, Moritz RFA, Kraus FB (2014) The Evolution of Extreme Polyandry in Social Insects: Insights from Army Ants. PLoS ONE 9(8): e105621. https://doi.org/10.1371/journal.pone.0105621

Editor: Fabio S. Nascimento, Universidade de São Paulo, Faculdade de Filosofia Ciências e Letras de Ribeirão Preto, Brazil

Received: September 13, 2013; Accepted: July 25, 2014; Published: August 21, 2014

Copyright: © 2014 Barth et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Funding: This work was supported by the Graduate Scholarship of Saxony Anhalt (http://www.uni-halle.de/forschung/service/wissenschaftlicher_nachwuchs/gradfoerd_lsa/#anchor1071166), Mexican-European grant MUTUAL (FONCICYT 94293). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Army ants are major arthropod predators and important keystone organisms in tropical and subtropical ecosystems, with a strong impact on the population dynamics of their prey and species community structures [1]–[5]. This is facilitated by a highly specialized combination of traits, the so called army ant adaptive syndrome, which includes group raiding and nomadism [6]–[8]. Several species can establish huge colonies with over a million workers sweeping the forest floor in fan-shaped swarm raids, thereby exhibiting impressive organizational complexity and efficient division of labor [1], [5], [6], [9], [10].

Nonetheless, as typical for the eusocial Hymenoptera (ants, social bees and social wasps), army ants may suffer from constrained effective population sizes (N_e) due to the limited number of reproductive queens and the haploidy of males [11]–[13]. This effect is enhanced by the army ant syndrome, because queens are wingless and can only disperse via colony fission, leading to strongly male-biased sex-ratios [4], [8], [14], which further reduces N_e and makes army ants susceptible to random drift and deleterious effects of inbreeding [15]. In the vast majority of army ant species, the colonies are headed by a single queen, which is highly polyandrous, mating with about 20 males [16]. Several hypotheses have been put forward to explain the adaptive value of such high degrees of polyandry in social insects, mainly invoking the enhancement of genotypic variance within colonies, but also the potential to increase overall N_e and thus counteract the risk of inbreeding.

The gain in intracolonial genetic variance (“genetic variance” hypothesis) is one of the most prominent hypotheses for the evolution of polyandry in social insects. Genetic variance may support overall colony fitness in various aspects like increasing colony productivity, tolerance to variable environments and pathogen resistance [17]–[22]. Moreover, the presence of many patrilines within a colony may improve worker caste specialization and thus division of labor [23]–[26]. A potential problem for the genetic variance hypothesis at first glance is to explain the evolution of extreme polyandry. This is due to the fact that the genotypic diversity gained with each additional mating drops sharply beyond six to ten effective matings as the average intracolonial relatedness approaches the 0.25 asymptote [27]–[29]. However, at colony level, the synergistic fitness gain of mutualistic interactions between genotypically diverse individuals, known as social heterosis [30], may well increase beyond this threshold and also explain extreme polyandry in concordance with the genetic variance hypothesis. Other hypotheses emphasize the importance of polyandry to overcome limited sperm supply (“sperm limitation” hypothesis) to facilitate huge colony sizes [31]–[34]. Finally, mating with males from many different colonies may increase N_e and counteract population fragmentation and inbreeding [35]–[37].

However, data on genetic colony structure and paternity frequencies in army ants are so far almost only available for the few intensively studied and easily accessible swarm-raiders of the genera Eciton and Dorylus (e.g., [16], [23], [35], [36], [38]). The study of Kronauer et al. [16] is so far the only one to compare four distantly related species, concluding that extreme polyandry is ancestral for army ants. Only a single reversion to near monandry (one to two matings) has recently been described for Neivamyrmex carolinensis, where colonies are polygynous instead [39], [40]. However, army ant communities in the tropics may comprise more than 20 sympatric species [41], all profoundly differing in their niches and life history [2]–[6], [10]. The majority of these species and their genetic colony structure are still unexplored and may offer important insights into the adaptive value of the army ant mating system.

Here we add paternity data on five Neotropical army ant species, namely Eciton mexicanum, E. vagans, Labidus coecus, L. praedator and Nomamyrmex esenbeckii (Ecitoninae), and compare them with respect to paternity frequency and genotypic colony composition. The two genera Labidus and Nomamyrmex were so far neglected in genetic army ant studies, despite their ubiquity and ecological significance [2], [3], [5], [42], [43]. E. mexicanum and E. vagans resemble their well-studied sister species, the swarm-raider E. burchellii, but have considerably smaller and more cryptic colonies [4]–[6], [9], [10]. By deducing queen and male genotypes from worker samples in classical microsatellite DNA analyses, we determine paternity to estimate queen mating frequencies, intracolonial relatedness and paternity skew (distribution of proportional paternity among siring males). Additionally we extract data of four more army ant species from the literature, for comparisons with our own empirical results.

With our data set we test critical predictions made by the various hypotheses for the evolution of extreme polyandry in social insects. First, relatedness analyses within colonies may reveal its potential to counteract inbreeding [36]. Second, by correlating paternity data with colony size we test for the sperm limitation hypothesis. Third, we use paternity skew analyses to test for the genetic variance hypothesis (i.e., for social heterosis). As a high skew can diminish intracolonial genotypic variance gained from polyandry by shifting the average intracolonial relatedness towards that of effective monandry [19], [34], [44], [45], queens are expected to equalize the paternity distribution among sires increasingly with higher paternity frequencies to maximize variance gains. In the absence of such a selection pressure male (sperm) competition should lead to monopolization of paternity [22], [34], [46], [47]. Consequently, Jaffé et al. [47] found a negative association of paternity skew and paternity frequency over a large paternity gradient across polyandrous Hymenoptera. However, they did not focus on extremely polyandrous species, while Kronauer et al. [16] hypothesized selective neutrality for the exact number of mates in this case. The present study is, to our knowledge, the first attempt to test this prediction and to show whether an association of paternity skew and frequency is maintained (i.e., genetic variance remains important) at the high end spectrum of polyandry within the clade of ancestrally extremely polyandrous army ants.

Materials and Methods

Sampling and genotyping

Sampling was carried out in Chiapas, Mexico, near the city of Tapachula at three different locations with a maximal distance of 18 km between them (Table 1): Cacahoatán (15°0′19 N, 92°10′17 W), Tapachula (14°55′58 N, 92°17′01 W) and Tuxtla Chico (14°58′25 N, 92° 9′34 W). Sampling permits were authorized by ECOSUR (El Colegio de la Frontera Sur, Unidad Tapachula) and the Agricultural Research Centre Rosario Izapa in Tuxtla Chico. The sampling did not involve endangered or protected species. Workers of the five study species were collected from columns and immediately stored in 95% ethanol and kept at –20°C until further laboratory processing: E. mexicanum s. str. Roger (3 colonies, n = 400 workers), E. vagans angustatum Roger (1 colony, n = 94), L. coecus Latreille (1 colony, n = 94), L. praedator s. str. Smith (3 colonies, n = 294) and N. esenbeckii wilsoni Santschi (2 colonies, n = 94). Species were identified using the key of Watkins [43].

Download:

Table 1. Sampling locations, paternity frequencies, genetic heterozygosities and paternity skew of up to three colonies of the five Neotropical army ants E. mexicanum (Em), E. vagans (Ev), L. coecus (Lc), L. praedator (Lp) and N. esenbeckii (Ne).

https://doi.org/10.1371/journal.pone.0105621.t001

We screened two existing primer sets (Eb and Dmo), originally developed for the army ants E. burchellii and Dorylus molestus [48], [49] for cross-amplification in all our five study species. Since only up to four loci were polymorphic in Labidus and Nomamyrmex, we developed five additional microsatellite primers (Lp2, Lp4, Lp14a, Lp30 and Lp38; GenBank accession numbers KF969232 to KF969236; Table S1) for L. praedator, following the microsatellite isolation protocol of Glenn and Schable [50]. These newly developed primers were also tested for cross-amplification resulting in a total of 14 microsatellite loci (Table S2) that individually all were polymorphic for at least one of the five tested species. Worker DNA was extracted from ant legs, following a 5% Chelex protocol [51] and amplified in standard PCR cocktails at an annealing temperature of T_a = 54°C (51°C for DmoD). Allele calling and scoring was performed using the Fragment Profiler software (Amersham Biosciences), and double checked by eye. DNA amplification and genotyping was repeated for 10% of the samples confirming the initial allele scoring. See Protocol S1, Tables S1 and S2 for further details of the microsatellite development and cross-species testing.

Paternity frequency analysis

Queen and male genotypes were deduced from the worker genotypes by Mendelian inference [52]. If a worker shared the same alleles as its heterozygous queen at a given locus, so that two patrilines were possible at that locus, workers were assigned in a way to minimize the number of patrilines (parsimonious). Workers with less than three unambiguously amplified loci were excluded (Tables 1 and S3). Taking advantage of the polyandrous mating system in army ants we used the inferred males and queens as a population sample from which we estimated the population wide allele frequencies with the program FSTAT 2.9.3 [53]. To obtain a diploid input file, males were entered as homozygote diploids and queens were duplicated. These allele frequencies were also used to confirm our manual queen and male deduction in the program MATESOFT version 1.0 [54]. The non-detection error for patrilines, i.e., the probability to obtain two males with the same allele combination at all loci [34], [45] was low for all species (2.2×10^–9 to 2.3×10^–2; Table S3). Therefore, the number of deduced patrilines was taken as the observed number of sires a queen mated with, i.e., the observed paternity frequency (k_obs). Correcting for the sample size dependent number of potentially unsampled patrilines (non-sampling error, [45]) we added the expected zero frequencies from fitted Poisson distributions to k_obs to achieve an estimate of the actual paternity frequencies of the queens (k_est). Additionally, we estimated the sample size corrected genetically effective paternity frequency (m_e), which accounts for potential paternity skew [55].

Population structure and relatedness

Nei’s observed and expected heterozygosities (H_O and H_S [56]) were calculated in FSTAT for all species. For H_O we used the worker genotypes of one randomly chosen worker per patriline to reduce an allele frequency bias from related workers (though some bias is still expected from the shared queen alleles of workers of the same colony). The unbiased deduced queen and male genotypes were used for H_S. Likewise, we used the worker genotypes to test for Hardy-Weinberg-Equilibrium, and the queen and male genotypes to test for linkage disequilibrium of loci in GENEPOP on the Web, using standard settings [57]. For the two species for which more than two colonies were available (E. mexicanum and L. praedator), queen and male genotypes were also used in FSTAT to estimate F_ST values for population sub-structuring [58], treating colonies as sub-populations. Moreover, the inbreeding coefficient was estimated from Nei’s heterozygosities: F_IS = 1– H_O/H_S [56].

To assess the significance of inbreeding from the within colony relatedness, E. mexicanum and L. praedator were also analyzed with the program RELATEDNESS 5.0.8 [59] in a similar way as described for other army ants [16], [35], [36]. The average pairwise regression genetic relatedness was estimated between queens and their mates (r_qm), among the males of each queen (r_mm) and among the workers (r_ww). In the absence of inbreeding the latter should not differ significantly from the expected pedigree relatedness, g_ww = 0.25+0.5/m_e [60], where m_e is the uncorrected effective paternity frequency from Starr [61]. Single-sample t-tests were used to test this relationship per colony (Bonferroni adjusted for multiple comparisons) and over all colonies. Likewise, r_qm and r_mm were tested against zero, to check whether queens mated with unrelated males and males were unrelated among themselves, respectively.

Paternity skew analyses within species

Two different indices were used with two different data sets to estimate intracolonial paternity skew. The first data set consisted of our five study species plus data of six E. burchellii colonies from Jaffé et al. [23]. For these six species raw data on per-queen paternity distribution among patrilines were available and were used to quantify and test paternity skew with the binomial B-index. This is considered the most robust paternity skew index [62], [63] and was calculated using the program SKEW CALCULATOR 2003 (https://www.eeb.ucla.edu/Faculty/Nonacs/PI.html, [63]). The B-index ranges from –1 (even paternity among all males) to 1 (monopoly by one male), with zero indicating a random paternity skew. SKEW CALCULATOR tests for deviation from random skew per species and colony based on one-tailed p-values from 1,000 simulations.

For the second data set we added paternity data of three more army ant species from the literature (D. molestus, Neivamyrmex nigrescens and N. carolinensis) to our first data set (Figure 1). As for the latter three species no raw paternity data was available to determine the B-index, we instead used the published k_obs and m_e values to calculate the effective-number-index of paternity skew (S-index), with S = (k_obs–m_e)/(k_obs–1) [64], for all nine species in the second data set, similar to Jaffé et al. [47]. The S-index ranges from 0 to 1, representing even or monopolized paternity, respectively. This index, however, does not account for random paternity skew which is always expected from stochastic effects, such as sampling biases. Therefore, we determined case-specific confidence intervals of randomly expected S-values for each species in standard spreadsheet based Monte Carlo simulations [63], [64]. To do so, we calculated S-values in such a way that each worker of a given colony had the same probability of belonging to any of the respective observed patrilines, and iterated this protocol 3,000 times, as described in detail in Jaffé et al. [47]. Observed S-values were considered significant if outside of the randomly expected confidence limits.

Download:

Figure 1. Phylogenetic tree (following Brady [7]) of all 13 army ant species for which paternity data is available so far.

Data on colony size, observed and effective paternity frequency (k_obs and m_e), effective-number-index of paternity skew (S) and inbreeding coefficient (F_IS) are given next to the tree. Colony size is represented by the average approximation of the number of workers in normal mature colonies as given in the literature. k_obs and m_e are the means given in this study or in the literature, from which S was calculated [47], [64]. Mean F_IS is also taken from this study or the literature (asterisks mark significance). Species abbreviations, as used in the following figure legends, are given in brackets behind species names. n/a, data not available.¹ Data on these species are listed for the sake of completeness, but have not been used in this study because m_e estimates were either not published, higher than the respective k_obs, or assumed to be underestimated [16], so that in these cases the S-index could not be calculated or would have been biased. ² Paternity and inbreeding data on these species are the result of this study.

https://doi.org/10.1371/journal.pone.0105621.g001

Paternity skew analyses among species

Both data sets were tested for an association of paternity skew and paternity frequency by applying a phylogenetically corrected generalized least-squared (GLS) regression, based on a maximum likelihood approach [65], [66], with k_obs as predictor of B and S skew indices over all species of the respective data set. To do so, the phylogenetic information from Brady [7] was used with our data in the program BayesTraits (http://www.evolution.reading.ac.uk/SoftwareMain.html [65]). According to the genetic variance hypothesis, this association should be negative if queens equalize paternity among sires to maximize colony level fitness gained from mating multiply, so that paternity skew decreases with increasing k_obs [47]. Furthermore, we also used colony size (Figure 1) as predictor of k_obs and the S-index, respectively, in GLS regressions over all nine species to test the sperm limitation hypothesis. Regression slopes were tested for deviation from zero using likelihood ratio tests.

As paternity skew (S-index) estimations in the second data set depend on k_obs, an association between these two variables is expected a priori. To test for a deviation from this null expectation, we determined case-specific confidence limits of slopes from Monte Carlo simulations by running the GLS model 100 times with 100 different sets of randomly chosen S-values from the first simulations of all nine species (see Jaffé et al. [47]). Negative S-values, which are not defined but may occur during the simulations if m_e exceeds k_obs, were conservatively treated as zero skew. Prior to GLS analyses the not normally distributed variables, paternity skew and colony size, were normalized using log (x+1) (accounting for zero skews) and simple log (x) transformations, respectively.

Differences in the degree of paternity skew among species can also be analyzed by testing paternity distributions among patrilines for deviation from homogeneity in contingency tables. Therefore, we used our first data set in a correspondence analysis in STATISTICA 8.0 [67], similar to Schlüns et al. [68]. Absolute patriline frequencies and the species were entered as categorical variables in rows and columns, respectively. The proportional contribution of each species to the deviation from homogeneity of the paternity skew is represented by relative inertia resulting from χ²-statistics.

Results

Genotypes and paternity frequencies

The primer characteristics of the five new L. praedator microsatellites, as well as all inferred queen and male genotypes are given in Tables S1 and S3, respectively. Locus Lp14a revealed homozygote excess at several alleles in E. mexicanum, suggesting the occurrence of one or more null alleles, and was therefore excluded from the analyses for this species. In the end, we used ten loci for E. mexicanum, eleven for E. vagans, three for L. coecus, nine for L. praedator, and six for N. esenbeckii (Table S2).

In all colonies, except for the E. mexicanum colony Em1, a single queen genotype explaining all worker genotypes could be inferred with 100% support in MATESOFT. For colony Em1 all but two workers could be explained by one queen, for which two different genotypes were possible. We thus used the queen genotype with the higher probability in MATESOFT (78%) and a more parsimonious number of patrilines. The remaining two unassigned workers were excluded for having possibly originated from a second queen as it might occur after recent colony fission.

All here analyzed queens were highly polyandrous with observed paternity frequencies (k_obs) ranging from 5 to 26 siring males (Table 1). Sample sizes were sufficient for all colonies since k_obs did not differ notably from the sample size corrected estimated paternity frequencies (k_est, 6.01 to 27.25), except for Ne1, which was excluded from the following paternity skew analyses because of potential underestimation of the paternity frequency (see Table 1). The genetically effective mating frequencies (m_e) ranged from 5.76 to 25.51 per colony and were lower than k_est throughout, indicating potential paternity skews [34], [55].

Population structure and relatedness

Observed heterozygosities (H_O) were high for all species (0.67 to 0.90) and within the range of expected heterozygosities (H_S, 0.68 to 0.86) over all loci (Table 1), suggesting no further homozygote excess after removal of null alleles. All species were in Hardy-Weinberg-Equilibrium over all loci and colonies (Fisher’s exact tests; E. mexicanum: χ² = 73.57, df = 60, p = 0.11; E. vagans: χ² = 16.32, df = 22, p = 0.80; L. coecus: χ² = 1.61, df = 6, p = 0.95; L. praedator: χ² = 47.04, df = 54, p = 0.74; N. esenbeckii: χ² = 4.65, df = 24, p>0.99). Linkage was only detected in L. praedator between two to four loci pairs per colony (Bonferroni adjusted p-values). However, as different loci pairs were linked in different colonies and no linkage was detected in the other species, the apparent linkage may actually reflect some degree of relatedness among queen’s mates in L. praedator, and we assume no physical linkage between loci.

For E. mexicanum and L. praedator genetic population sub-structuring in the sampling area was weak but significantly larger than zero (one-tailed t-tests; E. mexicanum: F_ST = 0.019±0.009 SE from jackknifing over loci, n = 56 chromosomal sets, t = 2.11, df = 9, p = 0.03; L. praedator: F_ST = 0.069±0.023, n = 76, t = 2.96, df = 8, p = 0.01). Likewise, the mean population inbreeding coefficient was also slightly positive for E. mexicanum (F_IS = 0.039±0.006 SE, t = 6.59, df = 9, p<0.001), but slightly negative in L. praedator (F_IS = –0.016±0.003, t = 5.71, df = 8, p<0.001).

In E. mexicanum the average regression genetic relatedness among workers (r_ww = 0.251±0.030 SE from jackknifing over loci) did not significantly differ from the pedigree relatedness (g_ww = 0.290) over all colonies (t = 1.31, p = 0.11). The overall average queen to mate relatedness was slightly negative, though not significantly different from zero (r_qm = –0.059±0.035, t = 1.7, p = 0.06). Also the relatedness among males did not significantly differ from zero (r_mm = 0.004±0.012, t = 0.37, p = 0.36). In colony Em2 r_ww = 0.199±0.034 was significantly lower than g_ww = 0.302, (t = 3.00, p<0.01), suggesting outbreeding. Also in L. praedator r_ww (0.326±0.036) and g_ww (0.280) were not significantly different (t = 1.29, p = 0.12) and the queen to mate relatedness did not differ from zero (r_qm = 0.017±0.026, t = 1.7, p = 0.27). The overall among male relatedness (r_mm = 0.041±0.018) was close to zero though significantly positive (t = 2.23, p = 0.03). See Table 2 for relatedness data.

Download:

Table 2. Relatedness estimates for three colonies of E. mexicanum and L. praedator from Chiapas, Mexico.

https://doi.org/10.1371/journal.pone.0105621.t002

Paternity skew analysis

Paternity skew was significant in both data sets, with mean B-values higher than zero (Table 1) and effective number indices S above randomly expected 95% confidence limits (Figure 1) for all tested species. Only for the two L. praedator colonies Lp1 and Lp3 paternity sharing of patrilines did not differ from a random skew (Bonferroni adjusted p-values). However, mean B-values over all L. praedator colonies indicate significant paternity skew, which is also confirmed by the S-index.

Among all species of the first data set the GLS model, corrected for phylogeny [7], [65], [66], revealed a significantly negative slope (b) for a simple regression of paternity skew B on k_obs (likelihood ratio test with df = 1; b = –0.001, R² = 0.74, χ² = 7.12, p = 0.008; Figure 2). This also held true when using the sample size corrected k_est as predictor of B (b = –0.001, R² = 0.73, χ² = 7.10, p = 0.008). Likewise, for the second data set the observed slope of S on k_obs was significantly negative (b = –0.005, R² = 0.63, χ² = 8.92, p = 0.003; Figure 3) and steeper than randomly expected slopes with a 95% confidence interval ranging from –0.003 to 0.002. Simple GLS regressions on colony size were not significant, neither of paternity frequency, nor of paternity skew among all nine species (k_obs: b = 0.712, R² = 0.00003, χ² = 0.01, p = 0.90; S-index: b = –0.015, R² = 0.03, χ² = 0.27, p = 0.61; Figures S1 and S2).

Download:

Figure 2. Association of mean paternity skew (B-index) and mean observed paternity frequencies (k_obs) across six Neotropical army ant species (represented by different symbols).

The solid line indicates the slope of a phylogenetically corrected GLS regression (b = –0.001, R² = 0.74, p = 0.008) and dashed lines the 95% confidence interval.

https://doi.org/10.1371/journal.pone.0105621.g002

Download:

Figure 3. Association of the effective-number-index of paternity skew (S-index) and mean observed paternity frequencies (k_obs) across the nine army ant species (represented by different symbols) for which S was estimated.

The solid line indicates the slope of a phylogenetically corrected GLS regression (b = –0.005, R² = 0.63, p = 0.003) and dashed lines the 95% confidence interval of expected slopes with the same intercept under random paternity distribution. These were estimated from 100 Monte Carlo iterations of the GLS regression with randomly generated S-values (see text and Jaffé et al. [47] for details). An observed slope below the confidence limits is significantly steeper than expected from random.

https://doi.org/10.1371/journal.pone.0105621.g003

The correspondence analysis showed significant deviation from homogeneity in the paternity share of patrilines among the six species of the first data set (overall χ² = 198.67, df = 125, p<0.001; Figure 4). Those species with the highest and lowest paternity frequencies (L. praedator and N. esenbeckii) contributed most to this differentiation in paternity skew, indicated by the highest relative inertia (0.395 and 0.219; Figure 4). Absolute patriline frequencies were summed over colonies per species for this analysis. Within species differences in paternity skew among colonies were not significant (Figure S3).

Download:

Figure 4. Paternity distribution of six Neotropical army ants (1–6 colonies) ranked according to mean observed paternity frequencies (k_obs).

Alternately shaded bars show the proportional paternity of all patrilines (summed over colonies for each species). Mean k_obs, total number of assigned workers (n), paternity skew (B-index with asterisks marking significance per species) and relative inertia of the correspondence analysis are shown above the bars. The overall χ²-statistics of the correspondence analysis indicates significant deviation from homogeneity of the paternity distribution among species (χ² = 198.67, df = 125, p<0.001). Relative inertia give the proportional contribution of each species to this deviation [67], showing the highest values for the species with the highest and lowest k_obs (L. praedator and N. esenbeckii, respectively).

https://doi.org/10.1371/journal.pone.0105621.g004

Discussion

Paternity frequencies

Effective paternity frequencies were high (up to 25.51) in all five army ant species analyzed in this study (Table 1), supporting the idea of ancestral evolution of extreme polyandry in army ants in close association with the army ant adaptive syndrome [7], [16]. Thus, all so far analyzed species exceed (most of them by far) the threshold above which a reduction of intracolonial genetic relatedness with further matings becomes negligible [27]–[29]. At this level of extreme polyandry fitness benefits resulting from genotypic variance can nevertheless be gained at colony level from mutualistic benefits of diverse, interacting genotypes that are more than the sum of their parts (social heterosis, [30]). Likewise, the maintenance of rare specialized genotypes could be advantageous at colony level [29]. For instance L. praedator, with the highest paternity frequencies and up to nine patrilines that are represented by only one worker (Table S3), is supportive to the latter hypothesis.

The only known not highly polyandrous army ant N. carolinensis does not contradict these hypotheses because in this species polyandry is secondarily replaced by polygyny with up to 19 queens [39], [40]. It has been argued that polyandry and polygyny can be alternative, negatively correlated strategies to increase intracolonial variance in social insects [17]–[19], [69].

Population structure and relatedness

Colony fission in combination with male-biased sex ratios [4], [6], [9] predisposes army ants for small effective population sizes (N_e), which severely increases the risk of genetic depletion and inbreeding [11], [13], [15]. Polyandry has been suggested as an escape route from these negative effects [11], [36] and indeed evidence for inbreeding is weak in army ants [16], [35], [36]. Male dispersal markedly facilitates gene flow [35], [37], and queens of E. burchellii can boost N_e by preferentially mating with unrelated males [36]. Also in our study we found only weak population sub-structuring (F_ST 0.019 and 0.069 for E. mexicanum and L. praedator) with high levels of heterozygosity (H_S 0.70 and 0.80, respectively), indicating high genetic variability and gene flow in both species. Some slight sub-structuring can nevertheless be expected because in army ants not all colonies of a population produce males in a given season [4], so that queens may have mated with different male population in different seasons.

Slightly positive and negative mean inbreeding coefficients (F_IS 0.039 and –0.016) in E. mexicanum and L. praedator, respectively, ranged well within previously observed values in other army ants (Figure 1). Negative F_IS as a result of heterozygote excess has also been found in E. burchellii [35] and, in combination with small N_e, can be a signal of outbreeding [70]–[73], e.g., if queens preferentially mate with males from distinct sub-populations. Slightly positive F_IS, as previously also found in N. nigrescens [16], however, not necessarily results from sib-mating, but may reflect slight heterozygote deficiency due to the observed population sub-structuring or due to the relatedness among some workers, from which H_O was estimated [15].

The average regression genetic relatedness among workers in E. mexicanum and L. praedator did not differ from the expected pedigree relatedness over all colonies (Table 2), confirming the absence of inbreeding, which was further supported by the lack of any relatedness between queens and their mates (in E. mexicanum even with a negative tendency). This and the significantly lower relatedness among workers compared to pedigree relatedness in colony Em2 (Table 2) again suggests outbreeding, as previously observed in E. burchellii [36]. In L. praedator a slightly positive average relatedness among males may reflect an uneven contribution of the surrounding colonies to the queens’ sires [35], [36]. This bias can explain the observed linkage between some loci pairs in L. praedator, but seems too small to significantly increase relatedness among workers.

Thus, our data illustrates the capability of army ant queens to avoid inbreeding by multiple mating with males to which they are unrelated to. At population level this seems to efficiently overcome the N_e constraints of the army ant syndrome by maintaining gene flow and genetic diversity. E.g., Chapman and Bourke [12] showed that the estimated N_e of an island population of E. burchellii increased from 75, assuming monandry, to 112 if assuming extreme polyandry. In contrast, in monandrous social insects small N_e often leads to high population sub-structuring and low heterozygosity ([74] for bumblebees, [75] for ants, and references therein).

Paternity skew and colony size

A strong paternity skew can diminish the effect of multiple mating by shifting the average intracolonial relatedness towards that of effective monandry [19], [34], [44]–[47]. According to the genetic variance hypothesis for the evolution of polyandry [17]–[22] a high intracolonial genotypic variance is an important parameter for colony fitness. Hence, it would be beneficial for queens to equalize the paternity share among siring males. Indeed, our data show that paternity skews decrease with an increasing mean number of sires of a queen (observed and sample size corrected paternity frequencies). In fact, the two colonies Lp1 and Lp3 (L. praedator) with the highest observed paternity frequencies (k_obs = 26) were also the ones with B-indices indifferent from a random paternity skew in our first data set (Table 1). Thus, this study provides evidence that intracolonial genotypic variance is an important factor for the evolution of polyandry, even in its most extreme form. Moreover, it is to our knowledge the first to show a negative association between paternity skew and paternity frequency across species within a clade of extremely polyandrous social insects, confirming earlier findings across social Hymenoptera in general [47].

Such a negative association can be interpreted as a trade-off across species between high genotypic variance and high relatedness within colonies. According to Hamilton’s inclusive fitness concept [76], the latter is an important prerequisite for social cohesion [19], [22], [28], [34], [47], [69]. In the case of extreme polyandry natural selection may favor queens to maximize intracolonial genotypic variance by paternity equalization, since the loss of relatedness with additional matings becomes negligible anyway. In contrast, in polyandrous insects with relatively low paternity frequencies, competition among males often leads to precedence of the first or last sire and a strong paternity skew [46], [77]–[80].

According to our phylogenetically corrected data, males appear to have little power to influence paternity skew in army ants, whereas queens seem to be adapted to equalize paternity. The correspondence analysis confirms significant paternity skew differences among the six species of the first data set, with peaks at the most extreme paternity frequencies (Figure 4), but not among colonies within species (Figure S3). Our findings fit the data on honeybees (genus Apis), the only other social Hymenoptera matching army ants in their high degree of polyandry [24], [27], [32], [52]. Honeybees also exhibit similar degrees of paternity skew and provide evidence that queens indeed control paternity equalization [68], [77]. Further evidence for queen control over sperm use by means of cryptic female choice [81] comes from polyandrous leaf-cutter ants [82], [83].

If queens mate multiply for sperm limitation reasons [31]–[34] using all available sperm, large colonies should exhibit skewed paternities, since males differ naturally in sperm number and contribution [27], [32], [84]. Moreover, under the sperm limitation hypothesis paternity frequencies are expected to increase with colony size. However, colony size was not significantly associated with k_obs (Figure S1), and showed a trend of a negative association (though not significant) with the S-index of paternity skew over nine army ant species (Figure S2). Thus, for army ants, this study neither supports nor rejects the sperm limitation hypothesis as an additional factor for the evolution of extreme polyandry.

Conclusion

Based on our results obtained from the analysis of paternity skew and genetic colony structure of nine army ants species, we suggest two main drivers for the evolution of extreme polyandry in army ants and in social insects in general. First, by mating with many unrelated males, queens seem to counteract problems connected with limited N_e [11]–[13], [36], facilitating gene flow and reducing the risk of inbreeding. This is supported by the fact that high polyandry in ants and honeybees is often associated with colony fission [16], [28], [84], which particularly reduces N_e (but see [85], [86]).

Second, in concordance with the genetic variance hypothesis polyandry may increase colony efficiency and parasite resistance [17]–[26], even beyond the expected threshold of six to ten matings [27]–[29], which can plausibly be explained by the principle of social heterosis [30]. Thus, an adaptive value of extreme polyandry, as suggested by Kraus et al. [87], seems more likely than selective neutrality, as proposed by Kronauer et al. [16], especially because the costs of mating are probably never zero. Even for army ant queens, which mate within the protection of the colony [4], [6] (reducing the risk of predation), potential costs such as energetic loss or disease/parasite transmission [17], [29] remain. The sperm limitation hypothesis [31]–[34] cannot be excluded here as an additional driver of extreme polyandry, but seems to play a minor role in army ants, fitting recent findings in stingless bees which can have enormously large colonies even in spite of monandrous queens [85].

Given our results, a possible scenario for the evolution of extreme polyandry in army ants could be that increasingly efficient group predation [1], [4]–[6], [9], [10] made colony fission as mode of reproduction obligate (requirement of a minimal worker number), which increased N_e limitations and led selection to favor higher paternity frequencies. The resulting genotypic variance, in turn, might have further boosted colony efficiency finally leading to the highly specialized life history of the army ant adaptive syndrome [6]–[8].

Supporting Information

Figure S1.

Association of mean observed paternity frequencies (k_obs) and average approximate colony sizes across the nine army ant species (represented by different symbols) for which also paternity skew was analyzed. The solid line indicates the slope of a phylogenetically corrected GLS regression (b = 0.712, R² = 0.00003, p = 0.90) and dashed lines the 95% confidence interval.

https://doi.org/10.1371/journal.pone.0105621.s001

(TIF)

Figure S2.

Association of the effective-number-index of paternity skew (S-index) and average approximate colony sizes across nine army ant species. The solid line indicates the slope of a phylogenetically corrected GLS regression (b = –0.015, R² = 0.03, p = 0.61) and dashed lines the 95% confidence interval.

https://doi.org/10.1371/journal.pone.0105621.s002

(TIF)

Figure S3.

Paternity distribution among 3–6 colonies of the three Neotropical army ant species E. burchellii (Eb), E. mexicanum (Em) and L. praedator (Lp). Alternately shaded bars show the proportional paternity of all patrilines. k_obs, number of assigned workers (n), paternity skew (B-index with asterisks marking Bonferroni adjusted significance) and relative inertia of the correspondence analysis are shown above the bars for each colony. Relative inertia give the proportional contribution of each colony to deviation from homogeneity of the paternity distribution within a species. The χ²-statistics of the correspondence analysis, given per species above the diagram, indicates no significant deviation among colonies for neither species.

https://doi.org/10.1371/journal.pone.0105621.s003

(TIF)

Table S1.

Characteristics of five new microsatellite loci developed for the Neotropical army ant Labidus praedator.

https://doi.org/10.1371/journal.pone.0105621.s004

(DOC)

Table S2.

Cross-amplification data of 14 microsatellite loci among five Neotropical army ant species.

https://doi.org/10.1371/journal.pone.0105621.s005

(DOC)

Table S3.

Inferred queen and patriline genotypes of five Neotropical army ants.

https://doi.org/10.1371/journal.pone.0105621.s006

(DOC)

Protocol S1.

Microsatellite isolation protocol for Labidus praedator and cross-amplification tests.

https://doi.org/10.1371/journal.pone.0105621.s007

(DOC)

Acknowledgments

We are grateful to the ECOSUR research center, particularly to R. B. M. Vandame for delightful discussions and to D. Sanchez and O. Argüello for providing equipment and their kind help during field work. Mating frequency data of E. burchellii from Mexico and Venezuela were kindly provided by R. Jaffé. Moreover, we want to thank D. Kleber and P. Leibe for technical assistance with the laboratory work at the MLU Halle-Wittenberg. Anonymous reviewers provided highly valuable comments that significantly improved an earlier draft of this paper.

Author Contributions

Conceived and designed the experiments: MBB RFAM FBK. Performed the experiments: MBB. Analyzed the data: MBB. Contributed reagents/materials/analysis tools: RFAM. Wrote the paper: MBB RFAM FBK.

References

1. Franks NR, Bossert WH (1983) The influence of swarm raiding army ants on the patchiness and diversity of a tropical leaf litter ant community. In: Sutton S, Whitmore T, Chadwick A, editors. Tropical Rain Forest: Ecology and Management. Oxford: Blackwell Publishing. 151–163.
2. Kaspari M, O’Donnell S (2003) High rates of army ant raids in the Neotropics and implications for ant colony and community structure. Evol Ecol Res 5: 933–939.
- View Article
- Google Scholar
3. Kaspari M, Powell S, Lattke J, O’Donnell S (2011) Predation and patchiness in the tropical litter: do swarm-raiding army ants skim the cream or drain the bottle? J Anim Ecol 80: 818–823
- View Article
- Google Scholar
4. Schneirla TC (1971) Army ants: a study in social organization. San Francisco: W.H. Freeman. 369 p.
5. Rettenmeyer CW (1963) Behavioral studies of army ants. Univ Kansas Sci Bull 44: 281–465.
- View Article
- Google Scholar
6. Gotwald WH (1995) Army ants: the biology of social predation. Ithaca: Comstock. 302 p.
7. Brady SG (2003) Evolution of the army ant syndrome: the origin and long-term evolutionary stasis of a complex of behavioral and reproductive adaptations. Proc Natl Acad Sci USA 100: 6575–6579
- View Article
- Google Scholar
8. Kronauer DJC (2009) Recent advances in army ant biology (Hymenoptera: Formicidae). Myrmecol News 12: 51–65.
- View Article
- Google Scholar
9. Franks NR (1985) Reproduction, foraging efficiency and worker polymorphism in army ants. In: Hölldobler B, Lindauer M, editors. Experimental Behavioural Ecology. Stuttgart: G Fisher. 91–107.
10. Powell S, Franks NR (2006) Ecology and the evolution of worker morphological diversity: a comparative analysis with Eciton army ants. Funct Ecol 20: 1105–1114
- View Article
- Google Scholar
11. Packer L, Owen R (2001) Population genetic aspects of pollinator decline. Conserv Ecol 5: 4.
- View Article
- Google Scholar
12. Chapman RE, Bourke AFG (2001) The influence of sociality on the conservation biology of social insects. Ecol Lett 4: 650–662
- View Article
- Google Scholar
13. Zayed A (2004) Effective population size in Hymenoptera with complementary sex determination. Heredity 93: 627–630
- View Article
- Google Scholar
14. Bulmer MG (1983) Sex ratio theory in social insects with swarming. J Theor Biol 100: 329–339
- View Article
- Google Scholar
15. Hartl DL, Clark AG (2007) Principles of population genetics. Sunderland: Sinauer Associates. 652 p.
16. Kronauer DJC, Johnson RA, Boomsma JJ (2007) The evolution of multiple mating in army ants. Evolution 61: 413–422
- View Article
- Google Scholar
17. Keller L, Reeve HK (1994) Genetic variability, queen number, and polyandry in social Hymenoptera. Evolution 48: 694–704.
- View Article
- Google Scholar
18. Schmid-Hempel P (1998) Parasites in social insects. Princeton: Princeton University Press. 424 p.
19. Boomsma JJ, Kronauer DJC, Pedersen JS (2009) The evolution of social insect mating systems. In: Gadau J, Fewell JH, Wilson EO, editors. Organization of insect societies: from genome to sociocomplexity. Cambridge: Harvard University Press. 3–25.
20. Hughes WOH, Boomsma JJ (2004) Genetic diversity and disease resistance in leaf-cutting ant societies. Evolution 58: 1251–1260
- View Article
- Google Scholar
21. Oldroyd BP, Fewell JH (2007) Genetic diversity promotes homeostasis in insect colonies. Trends Ecol Evol 22: 408–413
- View Article
- Google Scholar
22. Crozier RH, Fjerdingstad EJ (2001) Polyandry in social Hymenoptera – disunity in diversity? Ann Zool Fennici 38: 267–285.
- View Article
- Google Scholar
23. Jaffé R, Kronauer DJC, Kraus FB, Boomsma JJ, Moritz RFA (2007) Worker caste determination in the army ant Eciton burchellii. Biol Lett 3: 513–516
- View Article
- Google Scholar
24. Page RE, Robinson GE, Fondrik MK (1989) Genetic specialists, kin recognition and nepotism in honey-bee colonies. Nature 338: 576–579.
- View Article
- Google Scholar
25. Evison SEF, Hughes WOH (2011) Genetic caste polymorphism and the evolution of polyandry in Atta leaf-cutting ants. Naturwissenschaften 98: 643–649
- View Article
- Google Scholar
26. Kraus FB, Gerecke E, Moritz RFA (2011) Shift work has a genetic basis in honeybee pollen foragers (Apis mellifera L.). Behav Genet 41: 323–328
- View Article
- Google Scholar
27. Palmer KA, Oldroyd BP (2000) Evolution of multiple mating in the genus Apis. Apidologie 31: 235–248.
- View Article
- Google Scholar
28. Page RE, Metcalf RA (1982) Multiple mating, sperm utilization, and social evolution. Am Nat 119: 263–281.
- View Article
- Google Scholar
29. Fuchs S, Moritz RFA (1998) Evolution of extreme polyandry in the honeybee Apis mellifera L. Behav Ecol Sociobiol. 9: 269–275.
- View Article
- Google Scholar
30. Nonacs P, Kapheim KM (2007) Social heterosis and the maintenance of genetic diversity. J Evol Biol 20: 2253–2265
- View Article
- Google Scholar
31. Cole BJ (1983) Multiple mating and the evolution of social behavior in the hymenoptera. Behav Ecol Sociobiol 12: 191–201
- View Article
- Google Scholar
32. Kraus FB, Neumann P, van Praagh J, Moritz RFA (2004) Sperm limitation and the evolution of extreme polyandry in honeybees (Apis mellifera L.). Behav Ecol Sociobiol 55: 494–501
- View Article
- Google Scholar
33. Kraus FB, Moritz RFA (2010) Extreme polyandry in social Hymenoptera: evolutionary causes and consequences for colony organisation. In: Kappeler P, editor. Animal Behaviour: Evolution and Mechanisms. Berlin, Heidelberg: Springer. 413–439.
34. Boomsma JJ, Ratnieks FLW (1996) Paternity in Eusocial Hymenoptera. Philos Trans R Soc Lond B-Biol Sci 351: 947–975
- View Article
- Google Scholar
35. Berghoff SM, Kronauer DJC, Edwards KJ, Franks NR (2008) Dispersal and population structure of a New World predator, the army ant Eciton burchellii. J Evol Biol 21: 1125–1132
- View Article
- Google Scholar
36. Jaffé R, Moritz RFA, Kraus FB (2009) Gene flow is maintained by polyandry and male dispersal in the army ant Eciton burchellii. Popul Ecol 51: 227–236
- View Article
- Google Scholar
37. Barth MB, Moritz RFA, Pirk CWW, Kraus FB (2013) Male-biased dispersal promotes large scale gene flow in a subterranean army ant, Dorylus (Typhlopone) fulvus. Popul Ecol 55: 523–533
- View Article
- Google Scholar
38. Kronauer DJC, Peters MK, Schöning C, Boomsma JJ (2011) Hybridization in East African swarm-raiding army ants. Front Zool 8.
- View Article
- Google Scholar
39. Kronauer DJC, Boomsma JJ (2007) Multiple queens means fewer mates. Curr Biol 17: R753–755
- View Article
- Google Scholar
40. Rettenmeyer CW, Watkins JF (1978) Polygyny and monogyny in army ants (Hymenoptera: Formicidae). J Kansas Entomol Soc 51: 581–591.
- View Article
- Google Scholar
41. Longino JT, Coddington J, Colwell RK (2002) The ant fauna of a tropical rain forest: estimating species richness three different ways. Ecology 83: 689–702.
- View Article
- Google Scholar
42. Powell S, Clark E (2004) Combat between large derived societies: a subterranean army ant established as a predator of mature leaf-cutting ant colonies. Insectes Soc 51: 342–351
- View Article
- Google Scholar
43. Watkins JF (1976) The identification and distribution of New World army ants (Dorylinae: Formicidae). Waco: Baylor University Press. 102 p.
44. Sundström L, Boomsma JJ (2000) Reproductive alliances and posthumous fitness enhancement in male ants. Proc R Soc Lond B-Biol Sci 267: 1439–1444
- View Article
- Google Scholar
45. Jaffé R (2014) An updated guide to study polyandry in social insects. Sociobiology 61: 1–8
- View Article
- Google Scholar
46. Boomsma JJ, Sundström L (1998) Patterns of paternity skew in Formica ants. Behav Ecol Sociobiol 42: 85–92.
- View Article
- Google Scholar
47. Jaffé R, Garcia-Gonzalez F, den Boer SPA, Simmons LW, Baer B (2012) Patterns of paternity skew among polyandrous social insects: what can they tell us about the potential for sexual selection? Evolution 66: 3778–3788
- View Article
- Google Scholar
48. Kronauer DJC, Boomsma JJ, Gadau J (2004) Microsatellite markers for the driver ant Dorylus (Anomma) molestus. Mol Ecol Notes 4: 289–290
- View Article
- Google Scholar
49. Denny AJ, Franks NR, Edwards KJ (2004) Eight highly polymorphic microsatellite markers for the army ant Eciton burchellii. Mol Ecol Notes 4: 234–236
- View Article
- Google Scholar
50. Glenn TC, Schable NA (2005) Isolating microsatellite DNA loci. Methods Enzymol 395: 202–222
- View Article
- Google Scholar
51. Walsh PS, Metzger DA, Higuchi R (1991) Chelex-100 as a medium for simple extraction of DNA for Pcr-based typing from forensic material. Biotechniques 10: 506–513.
- View Article
- Google Scholar
52. Moritz RFA, Kryger P, Koeniger G, Koeniger N, Estoup A, et al. (1995) High degree of polyandry in Apis dorsata queens detected by DNA microsatellite variability. Behav Ecol Sociobiol 37: 357–363
- View Article
- Google Scholar
53. Goudet J (1995) FSTAT (Version 1.2): a computer program to calculate F-statistics. J Hered 86: 485–486.
- View Article
- Google Scholar
54. Moilanen A, Sundström L, Pedersen JS (2004) MATESOFT: a program for deducing parental genotypes and estimating mating system statistics in haplodiploid species. Mol Ecol Notes 4: 795–797
- View Article
- Google Scholar
55. Nielsen R, Tarpy DR, Reeve HK (2003) Estimating effective paternity number in social insects and the effective number of alleles in a population. Mol Ecol 12: 3157–3164
- View Article
- Google Scholar
56. Nei M (1987) Molecular evolutionary genetics. New York: Columbia University Press. 512 p.
57. Raymond M, Rousset F (1995) Genepop (Version-1.2) – population-genetics software for exact tests and ecumenicism. J Hered 86: 248–249.
- View Article
- Google Scholar
58. Weir BS, Cockerham CC (1984) Estimating F-Statistics for the analysis of population-structure. Evolution 38: 1358–1370.
- View Article
- Google Scholar
59. Queller DC, Goodnight KF (1989) Estimating relatedness using genetic markers. Evolution 43: 258–275.
- View Article
- Google Scholar
60. Pamilo P (1993) Polyandry and allele frequency differences between the sexes in the ant Formica aquionia. Heredity 70: 472–480.
- View Article
- Google Scholar
61. Starr CK (1984) Sperm competition, kinship, and sociality in the aculeate Hymenoptera. In: Smith RL, editor. Sperm competition and the evolution of animal mating systems. New York: Academic Press. 427–467.
62. Nonacs P (2000) Measuring and using skew in the study of social behavior and evolution. Am Nat 156: 577–589.
- View Article
- Google Scholar
63. Nonacs P (2003) Measuring the reliability of skew indices: is there one best index? Anim Behav 65: 615–627
- View Article
- Google Scholar
64. Pamilo P, Crozier RH (1996) Reproductive skew simplified. Oikos 75: 533–535.
- View Article
- Google Scholar
65. Pagel M (1999) Inferring the historical patterns of biological evolution. Nature 401: 877–884
- View Article
- Google Scholar
66. Freckleton RP, Harvey PH, Pagel M (2002) Phylogenetic analysis and comparative data: a test and review of evidence. Am Nat 160: 712–726.
- View Article
- Google Scholar
67. StatSoft Inc. (2008) STATISTICA (data analysis software system), version 8.0. www.statsoft.com.
68. Schlüns H, Moritz RFA, Lattorff HMG, Koeniger G (2005) Paternity skew in seven species of honeybees (Hymenoptera: Apidae: Apis). Apidologie 36: 201–209
- View Article
- Google Scholar
69. Hughes WOH, Ratnieks FLW, Oldroyd BP (2008) Multiple paternity or multiple queens: two routes to greater intracolonial genetic diversity in the eusocial Hymenoptera. J Evol Biol 21: 1090–1095
- View Article
- Google Scholar
70. Keller LF, Waller DM (2002) Inbreeding effects in wild populations. Conserv Biol 17: 19–23.
- View Article
- Google Scholar
71. Ohta T (2009) Associative overdominance caused by linked detrimental mutations. Genet Res 18: 277–286
- View Article
- Google Scholar
72. Reed DH, Frankham R (2003) Correlation between fitness and genetic diversity. Conserv Biol 17: 230–237.
- View Article
- Google Scholar
73. Szulkin M, Bierne N, David P (2010) Heterozygosity-fitness correlations: a time for reappraisal. Evolution 64: 1202–1217
- View Article
- Google Scholar
74. Goulson D (2009) Conservation of bumblebees. Glas Nat 25: 31–34.
- View Article
- Google Scholar
75. Seppä P (2008) Do ants (Hymenoptera: Formicidae) need conservation and does ant conservation need genetics. Myrmecol News 11: 161–172.
- View Article
- Google Scholar
76. Hamilton WD (1964) The genetical evolution of social behaviour I and II. J Theor Biol 7: 1–52.
- View Article
- Google Scholar
77. Franck P, Solignac M, Vautrin D, Cornuet J-M, Koeniger G, et al. (2002) Sperm competition and last-male precedence in the honeybee. Anim Behav 64: 503–509
- View Article
- Google Scholar
78. Parker GA (1970) Sperm competition and its evolutionary consequences in the insects. Biol Rev 45: 525–567.
- View Article
- Google Scholar
79. Simmons LW (2001) Sperm competition and its evolutionary consequences in the insects. Princeton: Princeton University Press. 435 p.
80. Arnqvist G, Rowe L (2005) Sexual conflict. Princeton: Princeton University Press. 330 p.
81. Eberhard WG (1996) Female control: sexual selection by cryptic female choice. Princeton: Princeton University Press. 472 p.
82. Holman L, Stürup M, Trontti K, Boomsma JJ (2011) Random sperm use and genetic effects on worker caste fate in Atta colombica leaf-cutting ants. Mol Ecol 20: 5092–5102
- View Article
- Google Scholar
83. Den Boer SPA, Baer B, Boomsma JJ (2010) Seminal fluid mediates ejaculate competition in social insects. Science 327: 1506–1509
- View Article
- Google Scholar
84. Pearcy M, Timmermans I, Allard D, Aron S (2008) Multiple mating in the ant Cataglyphis cursor: testing the sperm limitation and the diploid male load hypotheses. Insectes Soc 56: 94–102
- View Article
- Google Scholar
85. Jaffé R, Pioker-Hara FC, Dos Santos CF, Santiago LR, Alves DA, et al. (2014) Monogamy in large bee societies: a stingless paradox. Naturwissenschaften 101: 261–264
- View Article
- Google Scholar
86. Kronauer DJC, O’Donnell S, Boomsma JJ, Pierce NE (2011) Strict monandry in the ponerine army ant genus Simopelta suggests that colony size and complexity drive mating system evolution in social insects. Mol Ecol 20: 420–428
- View Article
- Google Scholar
87. Kraus FB, Neumann P, Moritz RFA (2005) Genetic variance of mating frequency in the honeybee (Apis mellifera L.). Insectes Soc 52: 1–5
- View Article
- Google Scholar

[ref1] 1. Franks NR, Bossert WH (1983) The influence of swarm raiding army ants on the patchiness and diversity of a tropical leaf litter ant community. In: Sutton S, Whitmore T, Chadwick A, editors. Tropical Rain Forest: Ecology and Management. Oxford: Blackwell Publishing. 151–163.

[ref2] 2. Kaspari M, O’Donnell S (2003) High rates of army ant raids in the Neotropics and implications for ant colony and community structure. Evol Ecol Res 5: 933–939.
View Article
Google Scholar

[3] View Article

[4] Google Scholar

[ref3] 3. Kaspari M, Powell S, Lattke J, O’Donnell S (2011) Predation and patchiness in the tropical litter: do swarm-raiding army ants skim the cream or drain the bottle? J Anim Ecol 80: 818–823
View Article
Google Scholar

[6] View Article

[7] Google Scholar

[ref4] 4. Schneirla TC (1971) Army ants: a study in social organization. San Francisco: W.H. Freeman. 369 p.

[ref5] 5. Rettenmeyer CW (1963) Behavioral studies of army ants. Univ Kansas Sci Bull 44: 281–465.
View Article
Google Scholar

[10] View Article

[11] Google Scholar

[ref6] 6. Gotwald WH (1995) Army ants: the biology of social predation. Ithaca: Comstock. 302 p.

[ref7] 7. Brady SG (2003) Evolution of the army ant syndrome: the origin and long-term evolutionary stasis of a complex of behavioral and reproductive adaptations. Proc Natl Acad Sci USA 100: 6575–6579
View Article
Google Scholar

[14] View Article

[15] Google Scholar

[ref8] 8. Kronauer DJC (2009) Recent advances in army ant biology (Hymenoptera: Formicidae). Myrmecol News 12: 51–65.
View Article
Google Scholar

[17] View Article

[18] Google Scholar

[ref9] 9. Franks NR (1985) Reproduction, foraging efficiency and worker polymorphism in army ants. In: Hölldobler B, Lindauer M, editors. Experimental Behavioural Ecology. Stuttgart: G Fisher. 91–107.

[ref10] 10. Powell S, Franks NR (2006) Ecology and the evolution of worker morphological diversity: a comparative analysis with Eciton army ants. Funct Ecol 20: 1105–1114
View Article
Google Scholar

[21] View Article

[22] Google Scholar

[ref11] 11. Packer L, Owen R (2001) Population genetic aspects of pollinator decline. Conserv Ecol 5: 4.
View Article
Google Scholar

[24] View Article

[25] Google Scholar

[ref12] 12. Chapman RE, Bourke AFG (2001) The influence of sociality on the conservation biology of social insects. Ecol Lett 4: 650–662
View Article
Google Scholar

[27] View Article

[28] Google Scholar

[ref13] 13. Zayed A (2004) Effective population size in Hymenoptera with complementary sex determination. Heredity 93: 627–630
View Article
Google Scholar

[30] View Article

[31] Google Scholar

[ref14] 14. Bulmer MG (1983) Sex ratio theory in social insects with swarming. J Theor Biol 100: 329–339
View Article
Google Scholar

[33] View Article

[34] Google Scholar

[ref15] 15. Hartl DL, Clark AG (2007) Principles of population genetics. Sunderland: Sinauer Associates. 652 p.

[ref16] 16. Kronauer DJC, Johnson RA, Boomsma JJ (2007) The evolution of multiple mating in army ants. Evolution 61: 413–422
View Article
Google Scholar

[37] View Article

[38] Google Scholar

[ref17] 17. Keller L, Reeve HK (1994) Genetic variability, queen number, and polyandry in social Hymenoptera. Evolution 48: 694–704.
View Article
Google Scholar

[40] View Article

[41] Google Scholar

[ref18] 18. Schmid-Hempel P (1998) Parasites in social insects. Princeton: Princeton University Press. 424 p.

[ref19] 19. Boomsma JJ, Kronauer DJC, Pedersen JS (2009) The evolution of social insect mating systems. In: Gadau J, Fewell JH, Wilson EO, editors. Organization of insect societies: from genome to sociocomplexity. Cambridge: Harvard University Press. 3–25.

[ref20] 20. Hughes WOH, Boomsma JJ (2004) Genetic diversity and disease resistance in leaf-cutting ant societies. Evolution 58: 1251–1260
View Article
Google Scholar

[45] View Article

[46] Google Scholar

[ref21] 21. Oldroyd BP, Fewell JH (2007) Genetic diversity promotes homeostasis in insect colonies. Trends Ecol Evol 22: 408–413
View Article
Google Scholar

[48] View Article

[49] Google Scholar

[ref22] 22. Crozier RH, Fjerdingstad EJ (2001) Polyandry in social Hymenoptera – disunity in diversity? Ann Zool Fennici 38: 267–285.
View Article
Google Scholar

[51] View Article

[52] Google Scholar

[ref23] 23. Jaffé R, Kronauer DJC, Kraus FB, Boomsma JJ, Moritz RFA (2007) Worker caste determination in the army ant Eciton burchellii. Biol Lett 3: 513–516
View Article
Google Scholar

[54] View Article

[55] Google Scholar

[ref24] 24. Page RE, Robinson GE, Fondrik MK (1989) Genetic specialists, kin recognition and nepotism in honey-bee colonies. Nature 338: 576–579.
View Article
Google Scholar

[57] View Article

[58] Google Scholar

[ref25] 25. Evison SEF, Hughes WOH (2011) Genetic caste polymorphism and the evolution of polyandry in Atta leaf-cutting ants. Naturwissenschaften 98: 643–649
View Article
Google Scholar

[60] View Article

[61] Google Scholar

[ref26] 26. Kraus FB, Gerecke E, Moritz RFA (2011) Shift work has a genetic basis in honeybee pollen foragers (Apis mellifera L.). Behav Genet 41: 323–328
View Article
Google Scholar

[63] View Article

[64] Google Scholar

[ref27] 27. Palmer KA, Oldroyd BP (2000) Evolution of multiple mating in the genus Apis. Apidologie 31: 235–248.
View Article
Google Scholar

[66] View Article

[67] Google Scholar

[ref28] 28. Page RE, Metcalf RA (1982) Multiple mating, sperm utilization, and social evolution. Am Nat 119: 263–281.
View Article
Google Scholar

[69] View Article

[70] Google Scholar

[ref29] 29. Fuchs S, Moritz RFA (1998) Evolution of extreme polyandry in the honeybee Apis mellifera L. Behav Ecol Sociobiol. 9: 269–275.
View Article
Google Scholar

[72] View Article

[73] Google Scholar

[ref30] 30. Nonacs P, Kapheim KM (2007) Social heterosis and the maintenance of genetic diversity. J Evol Biol 20: 2253–2265
View Article
Google Scholar

[75] View Article

[76] Google Scholar

[ref31] 31. Cole BJ (1983) Multiple mating and the evolution of social behavior in the hymenoptera. Behav Ecol Sociobiol 12: 191–201
View Article
Google Scholar

[78] View Article

[79] Google Scholar

[ref32] 32. Kraus FB, Neumann P, van Praagh J, Moritz RFA (2004) Sperm limitation and the evolution of extreme polyandry in honeybees (Apis mellifera L.). Behav Ecol Sociobiol 55: 494–501
View Article
Google Scholar

[81] View Article

[82] Google Scholar

[ref33] 33. Kraus FB, Moritz RFA (2010) Extreme polyandry in social Hymenoptera: evolutionary causes and consequences for colony organisation. In: Kappeler P, editor. Animal Behaviour: Evolution and Mechanisms. Berlin, Heidelberg: Springer. 413–439.

[ref34] 34. Boomsma JJ, Ratnieks FLW (1996) Paternity in Eusocial Hymenoptera. Philos Trans R Soc Lond B-Biol Sci 351: 947–975
View Article
Google Scholar

[85] View Article

[86] Google Scholar

[ref35] 35. Berghoff SM, Kronauer DJC, Edwards KJ, Franks NR (2008) Dispersal and population structure of a New World predator, the army ant Eciton burchellii. J Evol Biol 21: 1125–1132
View Article
Google Scholar

[88] View Article

[89] Google Scholar

[ref36] 36. Jaffé R, Moritz RFA, Kraus FB (2009) Gene flow is maintained by polyandry and male dispersal in the army ant Eciton burchellii. Popul Ecol 51: 227–236
View Article
Google Scholar

[91] View Article

[92] Google Scholar

[ref37] 37. Barth MB, Moritz RFA, Pirk CWW, Kraus FB (2013) Male-biased dispersal promotes large scale gene flow in a subterranean army ant, Dorylus (Typhlopone) fulvus. Popul Ecol 55: 523–533
View Article
Google Scholar

[94] View Article

[95] Google Scholar

[ref38] 38. Kronauer DJC, Peters MK, Schöning C, Boomsma JJ (2011) Hybridization in East African swarm-raiding army ants. Front Zool 8.
View Article
Google Scholar

[97] View Article

[98] Google Scholar

[ref39] 39. Kronauer DJC, Boomsma JJ (2007) Multiple queens means fewer mates. Curr Biol 17: R753–755
View Article
Google Scholar

[100] View Article

[101] Google Scholar

[ref40] 40. Rettenmeyer CW, Watkins JF (1978) Polygyny and monogyny in army ants (Hymenoptera: Formicidae). J Kansas Entomol Soc 51: 581–591.
View Article
Google Scholar

[103] View Article

[104] Google Scholar

[ref41] 41. Longino JT, Coddington J, Colwell RK (2002) The ant fauna of a tropical rain forest: estimating species richness three different ways. Ecology 83: 689–702.
View Article
Google Scholar

[106] View Article

[107] Google Scholar

[ref42] 42. Powell S, Clark E (2004) Combat between large derived societies: a subterranean army ant established as a predator of mature leaf-cutting ant colonies. Insectes Soc 51: 342–351
View Article
Google Scholar

[109] View Article

[110] Google Scholar

[ref43] 43. Watkins JF (1976) The identification and distribution of New World army ants (Dorylinae: Formicidae). Waco: Baylor University Press. 102 p.

[ref44] 44. Sundström L, Boomsma JJ (2000) Reproductive alliances and posthumous fitness enhancement in male ants. Proc R Soc Lond B-Biol Sci 267: 1439–1444
View Article
Google Scholar

[113] View Article

[114] Google Scholar

[ref45] 45. Jaffé R (2014) An updated guide to study polyandry in social insects. Sociobiology 61: 1–8
View Article
Google Scholar

[116] View Article

[117] Google Scholar

[ref46] 46. Boomsma JJ, Sundström L (1998) Patterns of paternity skew in Formica ants. Behav Ecol Sociobiol 42: 85–92.
View Article
Google Scholar

[119] View Article

[120] Google Scholar

[ref47] 47. Jaffé R, Garcia-Gonzalez F, den Boer SPA, Simmons LW, Baer B (2012) Patterns of paternity skew among polyandrous social insects: what can they tell us about the potential for sexual selection? Evolution 66: 3778–3788
View Article
Google Scholar

[122] View Article

[123] Google Scholar

[ref48] 48. Kronauer DJC, Boomsma JJ, Gadau J (2004) Microsatellite markers for the driver ant Dorylus (Anomma) molestus. Mol Ecol Notes 4: 289–290
View Article
Google Scholar

[125] View Article

[126] Google Scholar

[ref49] 49. Denny AJ, Franks NR, Edwards KJ (2004) Eight highly polymorphic microsatellite markers for the army ant Eciton burchellii. Mol Ecol Notes 4: 234–236
View Article
Google Scholar

[128] View Article

[129] Google Scholar

[ref50] 50. Glenn TC, Schable NA (2005) Isolating microsatellite DNA loci. Methods Enzymol 395: 202–222
View Article
Google Scholar

[131] View Article

[132] Google Scholar

[ref51] 51. Walsh PS, Metzger DA, Higuchi R (1991) Chelex-100 as a medium for simple extraction of DNA for Pcr-based typing from forensic material. Biotechniques 10: 506–513.
View Article
Google Scholar

[134] View Article

[135] Google Scholar

[ref52] 52. Moritz RFA, Kryger P, Koeniger G, Koeniger N, Estoup A, et al. (1995) High degree of polyandry in Apis dorsata queens detected by DNA microsatellite variability. Behav Ecol Sociobiol 37: 357–363
View Article
Google Scholar

[137] View Article

[138] Google Scholar

[ref53] 53. Goudet J (1995) FSTAT (Version 1.2): a computer program to calculate F-statistics. J Hered 86: 485–486.
View Article
Google Scholar

[140] View Article

[141] Google Scholar

[ref54] 54. Moilanen A, Sundström L, Pedersen JS (2004) MATESOFT: a program for deducing parental genotypes and estimating mating system statistics in haplodiploid species. Mol Ecol Notes 4: 795–797
View Article
Google Scholar

[143] View Article

[144] Google Scholar

[ref55] 55. Nielsen R, Tarpy DR, Reeve HK (2003) Estimating effective paternity number in social insects and the effective number of alleles in a population. Mol Ecol 12: 3157–3164
View Article
Google Scholar

[146] View Article

[147] Google Scholar

[ref56] 56. Nei M (1987) Molecular evolutionary genetics. New York: Columbia University Press. 512 p.

[ref57] 57. Raymond M, Rousset F (1995) Genepop (Version-1.2) – population-genetics software for exact tests and ecumenicism. J Hered 86: 248–249.
View Article
Google Scholar

[150] View Article

[151] Google Scholar

[ref58] 58. Weir BS, Cockerham CC (1984) Estimating F-Statistics for the analysis of population-structure. Evolution 38: 1358–1370.
View Article
Google Scholar

[153] View Article

[154] Google Scholar

[ref59] 59. Queller DC, Goodnight KF (1989) Estimating relatedness using genetic markers. Evolution 43: 258–275.
View Article
Google Scholar

[156] View Article

[157] Google Scholar

[ref60] 60. Pamilo P (1993) Polyandry and allele frequency differences between the sexes in the ant Formica aquionia. Heredity 70: 472–480.
View Article
Google Scholar

[159] View Article

[160] Google Scholar

[ref61] 61. Starr CK (1984) Sperm competition, kinship, and sociality in the aculeate Hymenoptera. In: Smith RL, editor. Sperm competition and the evolution of animal mating systems. New York: Academic Press. 427–467.

[ref62] 62. Nonacs P (2000) Measuring and using skew in the study of social behavior and evolution. Am Nat 156: 577–589.
View Article
Google Scholar

[163] View Article

[164] Google Scholar

[ref63] 63. Nonacs P (2003) Measuring the reliability of skew indices: is there one best index? Anim Behav 65: 615–627
View Article
Google Scholar

[166] View Article

[167] Google Scholar

[ref64] 64. Pamilo P, Crozier RH (1996) Reproductive skew simplified. Oikos 75: 533–535.
View Article
Google Scholar

[169] View Article

[170] Google Scholar

[ref65] 65. Pagel M (1999) Inferring the historical patterns of biological evolution. Nature 401: 877–884
View Article
Google Scholar

[172] View Article

[173] Google Scholar

[ref66] 66. Freckleton RP, Harvey PH, Pagel M (2002) Phylogenetic analysis and comparative data: a test and review of evidence. Am Nat 160: 712–726.
View Article
Google Scholar

[175] View Article

[176] Google Scholar

[ref67] 67. StatSoft Inc. (2008) STATISTICA (data analysis software system), version 8.0. www.statsoft.com.

[ref68] 68. Schlüns H, Moritz RFA, Lattorff HMG, Koeniger G (2005) Paternity skew in seven species of honeybees (Hymenoptera: Apidae: Apis). Apidologie 36: 201–209
View Article
Google Scholar

[179] View Article

[180] Google Scholar

[ref69] 69. Hughes WOH, Ratnieks FLW, Oldroyd BP (2008) Multiple paternity or multiple queens: two routes to greater intracolonial genetic diversity in the eusocial Hymenoptera. J Evol Biol 21: 1090–1095
View Article
Google Scholar

[182] View Article

[183] Google Scholar

[ref70] 70. Keller LF, Waller DM (2002) Inbreeding effects in wild populations. Conserv Biol 17: 19–23.
View Article
Google Scholar

[185] View Article

[186] Google Scholar

[ref71] 71. Ohta T (2009) Associative overdominance caused by linked detrimental mutations. Genet Res 18: 277–286
View Article
Google Scholar

[188] View Article

[189] Google Scholar

[ref72] 72. Reed DH, Frankham R (2003) Correlation between fitness and genetic diversity. Conserv Biol 17: 230–237.
View Article
Google Scholar

[191] View Article

[192] Google Scholar

[ref73] 73. Szulkin M, Bierne N, David P (2010) Heterozygosity-fitness correlations: a time for reappraisal. Evolution 64: 1202–1217
View Article
Google Scholar

[194] View Article

[195] Google Scholar

[ref74] 74. Goulson D (2009) Conservation of bumblebees. Glas Nat 25: 31–34.
View Article
Google Scholar

[197] View Article

[198] Google Scholar

[ref75] 75. Seppä P (2008) Do ants (Hymenoptera: Formicidae) need conservation and does ant conservation need genetics. Myrmecol News 11: 161–172.
View Article
Google Scholar

[200] View Article

[201] Google Scholar

[ref76] 76. Hamilton WD (1964) The genetical evolution of social behaviour I and II. J Theor Biol 7: 1–52.
View Article
Google Scholar

[203] View Article

[204] Google Scholar

[ref77] 77. Franck P, Solignac M, Vautrin D, Cornuet J-M, Koeniger G, et al. (2002) Sperm competition and last-male precedence in the honeybee. Anim Behav 64: 503–509
View Article
Google Scholar

[206] View Article

[207] Google Scholar

[ref78] 78. Parker GA (1970) Sperm competition and its evolutionary consequences in the insects. Biol Rev 45: 525–567.
View Article
Google Scholar

[209] View Article

[210] Google Scholar

[ref79] 79. Simmons LW (2001) Sperm competition and its evolutionary consequences in the insects. Princeton: Princeton University Press. 435 p.

[ref80] 80. Arnqvist G, Rowe L (2005) Sexual conflict. Princeton: Princeton University Press. 330 p.

[ref81] 81. Eberhard WG (1996) Female control: sexual selection by cryptic female choice. Princeton: Princeton University Press. 472 p.

[ref82] 82. Holman L, Stürup M, Trontti K, Boomsma JJ (2011) Random sperm use and genetic effects on worker caste fate in Atta colombica leaf-cutting ants. Mol Ecol 20: 5092–5102
View Article
Google Scholar

[215] View Article

[216] Google Scholar

[ref83] 83. Den Boer SPA, Baer B, Boomsma JJ (2010) Seminal fluid mediates ejaculate competition in social insects. Science 327: 1506–1509
View Article
Google Scholar

[218] View Article

[219] Google Scholar

[ref84] 84. Pearcy M, Timmermans I, Allard D, Aron S (2008) Multiple mating in the ant Cataglyphis cursor: testing the sperm limitation and the diploid male load hypotheses. Insectes Soc 56: 94–102
View Article
Google Scholar

[221] View Article

[222] Google Scholar

[ref85] 85. Jaffé R, Pioker-Hara FC, Dos Santos CF, Santiago LR, Alves DA, et al. (2014) Monogamy in large bee societies: a stingless paradox. Naturwissenschaften 101: 261–264
View Article
Google Scholar

[224] View Article

[225] Google Scholar

[ref86] 86. Kronauer DJC, O’Donnell S, Boomsma JJ, Pierce NE (2011) Strict monandry in the ponerine army ant genus Simopelta suggests that colony size and complexity drive mating system evolution in social insects. Mol Ecol 20: 420–428
View Article
Google Scholar

[227] View Article

[228] Google Scholar

[ref87] 87. Kraus FB, Neumann P, Moritz RFA (2005) Genetic variance of mating frequency in the honeybee (Apis mellifera L.). Insectes Soc 52: 1–5
View Article
Google Scholar

[230] View Article

[231] Google Scholar