Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Pleistocene climate changes shaped the population structure of Partamona seridoensis (Apidae, Meliponini), an endemic stingless bee from the Neotropical dry forest

  • Elder Assis Miranda ,

    Affiliations Laboratório de Genética Evolutiva de Himenópteros, Departamento de Genética e Evolução, Universidade Federal de São Carlos, São Carlos, SP, Brazil, Departamento de Ciências Biológicas, Universidade Estadual de Santa Cruz, Ilhéus, BA, Brazil

  • Kátia Maria Ferreira,

    Affiliation Laboratório de Genética Evolutiva de Himenópteros, Departamento de Genética e Evolução, Universidade Federal de São Carlos, São Carlos, SP, Brazil

  • Airton Torres Carvalho,

    Affiliation Unidade Acadêmica de Serra Talhada, Universidade Federal Rural de Pernambuco, Serra Talhada, PE, Brazil

  • Celso Feitosa Martins,

    Affiliation Departamento de Sistemática e Ecologia, Universidade Federal da Paraíba, João Pessoa, PB, Brazil

  • Carlo Rivero Fernandes,

    Affiliation União de Ensino Superior de Campina Grande, UNESC, Campina Grande, PB, Brazil

  • Marco Antonio Del Lama

    Affiliation Laboratório de Genética Evolutiva de Himenópteros, Departamento de Genética e Evolução, Universidade Federal de São Carlos, São Carlos, SP, Brazil

Pleistocene climate changes shaped the population structure of Partamona seridoensis (Apidae, Meliponini), an endemic stingless bee from the Neotropical dry forest

  • Elder Assis Miranda, 
  • Kátia Maria Ferreira, 
  • Airton Torres Carvalho, 
  • Celso Feitosa Martins, 
  • Carlo Rivero Fernandes, 
  • Marco Antonio Del Lama


Partamona seridoensis is an endemic stingless bee from the Caatinga, a Neotropical dry forest in northeastern Brazil. Like other stingless bees, this species plays an important ecological role as a pollinator. The aim of the present study was to investigate the genetic structure and evolutionary history of P. seridoensis across its current geographic range. Workers from 84 nests from 17 localities were analyzed for COI and Cytb genic regions. The population structure tests (Bayesian phylogenetic inference, AMOVA and haplotype network) consistently characterized two haplogroups (northwestern and eastern), with little gene flow between them, generating a high differentiation between them as well as among the populations within each haplogroup. The Mantel test revealed no isolation by distance. No evidence of a potential geographic barrier in the present that could explain the diversification between the P. seridoensis haplogroups was found. However, Pleistocene climatic changes may explain this differentiation, since the initial time for the P. seridoensis lineages diversification took place during the mid-Pleistocene, specifically the interglacial period, when the biota is presumed to have been more associated with dry conditions and had more restricted, fragmented geographical distribution. This event may have driven diversification by isolating the two haplogroups. Otherwise, the climatic changes in the late Pleistocene must not have drastically affected the population dynamics of P. seridoensis, since the Bayesian Skyline Plot did not reveal any substantial fluctuation in effective population size in either haplogroup. Considering its importance and the fact that it is an endemic bee from a very threatened Neotropical dry forest, the results herein could be useful to the development of conservation strategies for P. seridoensis.


Social bees from the tribe Meliponini, known as stingless bees, are distributed in tropical and southern subtropical areas throughout the world, with greater diversity in Neotropical and Indo-Malayan regions [1,2]. So far, a total of 641 names for stingless bees are listed for the Neotropical region, 417 of which are considered valid [2]. Recent data indicate 244 valid species and another 89 undescribed forms [3] occurring in Brazil alone. Stingless bees play an important ecological role as pollinators of many wild plant species and are considered key elements for the maintenance and conservation of natural ecosystems as well as economically important agricultural systems [4]. However, recent studies have shown the decline of these pollinators and the consequent loss of pollination services in many areas of the world due mainly to human activities [5,6,7,8]. Despite the diversity, importance and threats, little is known regarding the population genetics and phylogeography of these bees.

The stingless bee Partamona seridoensis [9] occupies xeric regions, such as the Caatinga (phytogeographic domain), a Neotropical dry forest in the northeastern region of Brazil. This species nests only in active arboreal termite nests, mainly those of Constrictotermes cyphergaster. (Termitidae, Nasutitermitinae) and Microcerotermes spp [9,10,11]. Partamona seridoensis has been poorly studied and only one report is found in the literature on its population genetics, in which two populations were analyzed using allozymic and microsatellite markers [12]. There have been no studies on its conservation status, although this bee occupies the Caatinga, which is a threatened region. A recent study involving Partamona rustica has suggested the importance of concentrating efforts on the conservation of endemic stingless bees from Brazilian dry forests, due to bee hunters and deforestation, which have threatened these populations [13].

The Caatinga covers nearly 85,000 Km2 and is a highly heterogeneous phytogeographical domain, being the second largest center of endemism of xeric environments in South America [14,15]. Studies on animals and plants of the Caatinga have shown that this environment harbors high levels of species richness and specific interactions between insects and plants [13,14,16,17]. However, the Caatinga is one of the least protected areas by conservation policies in Brazil, with only 5.9% of its area included within conservation units [18], and one of the least studied areas of the Neotropical dry forest diagonal, constituted by the Chaco, Cerrado and Caatinga phytogeographic domains [14]. Moreover, deforestation combined with the semi-arid climate of the Caatinga has led to a process of desertification in many areas, making this phytogeographical domain severely threatened by climate changes [19].

Many studies have shown the effects of Pleistocene climatic events on the population structure and demography of bees around the world [20,21,22]. However, these studies are scarce for endemic bees in xeric habitats, such as the Neotropical dry forest diagonal [23], especially with regard to stingless bee populations of the Caatinga. A recent study analyzed the genetic structure and evolutionary processes involved in the current genetic variation of Melipona subnitida, an endemic species in northeastern Brazil, and suggested that the paleoclimatic changes of the Pleistocene could have played an important role in the diversification of this species [24]. Another recent study reconstructed the evolutionary history of Partamona rustica, an endemic stingless bee from the Neotropical dry forest in Brazil, and identified two groups of populations with no evidence of substantial changes in effective population size during the late Pleistocene [23].

Understanding the evolutionary history of a species and its genetic structure is essential to the development of effective conservation strategies [23,24], especially for species of the threatened Caatinga. Given the potentially serious consequences of inbreeding in stingless bees due the complementary sex determination mechanism [25,26], it is essential to understand the genetic population structure of these organisms to enable the implementation of management strategies and conservation policies. Thus, the aim of the present study was to investigate the genetic structure and evolutionary history of P. seridoensis across its current geographic range.

Materials and methods


Adult workers were collected from P. seridoensis nests from March 2008 to August 2014. Through active searches, 84 nests were sampled from 17 localities in the northeastern region of Brazil, with an average of five nests per locality (Table 1 and Fig 1A). The sampled localities were chosen through prior knowledge of occurrence records available on online database and others studies as well as through our active searches. In each locality where nests of P. seridoensis were sampled, we look for nests distant from each other at least 300 meters to avoid high kinship between them. All specimens were preserved in ethanol prior to the molecular analyses and vouchers were deposited in the Camargo Entomological Collection at the Universidade de São Paulo (FFCLRP—USP) and in the Laboratório de Genética Evolutiva de Himenópteros at the Universidade Federal de São Carlos (LGEH–UFSCar). All necessary research permits for fieldwork and collection of samples were issued by the Brazilian Institute for Biodiversity Conservation (Instituto Chico Mendes de Conservação da Biodiversidade—ICMBio) recorded by SISBio (permit number 31750). Field studies did not involve endangered or protected species.

Fig 1.

Geographical distribution of P. seridoensis populations (A). Samples are indicated by circle and number and listed according to Table 1. Bayesian inference using concatenated COI and Cytb mitochondrial gene regions in P. seridoensis and outgroups (B). Median-joining haplotype network in P. seridoensis (C). Caatinga area is represented by downward diagonal lines and the dotted line represents the geographical distribution of P. seridoensis [9] in A. Bayesian implementations of the Generalized Mixed Yule Coalescent model (bGMYC) for P. seridoensis, shown in B. Two lines represent two mutation steps between haplotypes and no line between haplotypes represents one mutation step in C. Gray color represents northwestern haplogroup and black color represents eastern haplogroup (in all figures).

Table 1. Geographical origin (with coordinates) of the P. seridoensis nests.

N = number of colonies; H = haplotypes found in each locality.

Amplification and sequencing

Total DNA was extracted from one worker per colony using a phenol-chloroform protocol [27]. Fragments of the mitochondrial genes cytochrome c oxidase I (COI) [28] and cytochrome b (Cytb) [29], commonly used for this type of study, were amplified. All polymerase chain reactions (PCR) contained template DNA (50 ng), 1X of Taq buffer (Invitrogen, USA), 250 μM of each dNTP, l.0 μM of each primer, 2.5 mM of MgCl2 and 1 U of Platinum Taq DNA polymerase (Invitrogen, USA) in a final volume of 25 μL. The PCR conditions for all gene regions were as follows: initial denaturation at 94°C for 5 minutes, followed by 40 cycles of denaturation at 94°C for 30 s, annealing at 50°C and extension at 72°C for 1 min, and completed with an additional extension step at 72°C for 10 min. The PCR products were electrophoresed in agarose gel stained with GelRedTM. The PCR product was purified using the Illustra ExoProStar 1-Step Kit (GE Healthcare, Buckinghamshire, UK). The mitochondrial genes were sequenced in both directions with the same amplification primers using the BigDye v 3.0 Dye Terminator Cycle Sequencing Kit (Applied Biosystems, Inc., Carlsbad, CA, USA) in an automated sequencer ABI 3730 XL (Applied Biosystems).

Genetic diversity

The electropherograms were edited using the Codon Code v3.7.1 program (CodonCode, Dedham, Massachusetts, USA). Sequences were aligned using the Clustal W algorithm with the aid of the BioEdit program [30]. All alignments were visually inspected and corrected. All haplotype sequences were deposited in the GenBank (accession numbers KX839192-KX839206). The number of variable sites (S), number of haplotypes (h), haplotype diversity (Hd) and nucleotide diversity (π) considering the concatenated gene regions were estimated using the DnaSP v5.10.01 program [31].

Phylogeographic structure tests

Bayesian inference was used to estimate the phylogeny of P. seridoensis lineages based on the concatenated mitochondrial haplotypes. The model with the best fit was selected with the aid of the MrModeltest v2.2 program [32] based on the Akaike information criterion (AIC) for two partitions (COI and Cytb). Two independent Bayesian runs of 20 million generations were performed, each with four Markov Chain Monte Carlo (MCMC) simulations. The first two thousand generations were discarded as burn-in, after which trees were sampled every 1000 generations. This analysis was performed using the MrBayes 3.1.2 program [33]. Chain convergence was checked using likelihood plots for each run with the aid of the Tracer 1.5 software [34] and results with ESS values > 200 were accepted. The consensus species tree for each inference was visualized and edited using the FigTree 1.4 software ( For this analysis, Partamona cupira and Partamona ailyae were used as the outgroup, since these species are part of the same group as P. seridoensis.

To check if the haplogroups observed are not cryptic species and assess uncertainty in phylogenetic tree estimation and model parameters, a Bayesian implementation of the generalized mixed Yule coalescent model (bGMYC) was used, which integrates over these potential sources of error via MCMC simulation [35]. The R package ‘bGMYC’ was used to calculate marginal posterior probabilities of species limits from the posterior distribution of ultrametric trees reconstructed by BEAST 1.8.0. A post-burn-in sample of 100 trees resampled from that posterior was used to calculate the posterior distribution of the bGMYC model. The bGMYC analysis was run for 100,000 generations, with a burn-in of 90,000 generations and a thinning interval of 100 samples.

Haplotype networks were constructed employing a median-joining network [36] in the NETWORK program ( Analysis of molecular variance (AMOVA) was performed to estimate the degree of population genetic differentiation among populations using three hierarchical levels. The ΦST and ΦCT indexes were also estimated. These analyses were implemented using the ARLEQUIN 3.11 program [37] and significance was determined with 1000 permutations. A Mantel test was carried out using the program IBDWS v.3.23 [38] to test the hypothesis of isolation by distance, considering each haplogroup separately and all populations together. Significance of this analysis was determined with 1000 permutations. All these analyses were implemented using the concatenated gene regions.

Population divergence time

A Bayesian approach implemented in BEAST 1.8.0 program [39] was used to estimate the time since the most recent common ancestor (TMRCA) of all samples and for each recovered lineage, based on the concatenated regions. Two independent runs were implemented for each estimate with the following parameters: coalescent constant size as a tree prior, UPGMA as an initial tree, a strict clock, 100 million generations, parameter sampling every 5000 generations and 10% of the MCMC simulations as burn-in. As there is currently no measure of mitochondrial mutation rate for corbiculate bees, to obtain the absolute times, a COI substitution rate of 1.9% per lineage per million years per generation (one year) (under a normal distributed prior) calibrated for other hymenopterans [40] was adopted. The substitution models for the northwestern and eastern haplogroups (see Results) were TrN and HKI+I, respectively, which were selected using the jModeltest v.2.1.5 program [41] and based on the AIC. Convergence between runs and analysis performance were checked using Tracer 1.5. Results with ESS values > 200 were accepted. Independent runs of each TMRCA estimate were combined in LogCombiner [39].

Demographic inference

A Bayesian Skyline Plot (BSP) reconstruction was implemented for possible groups identified in the phylogenetic reconstruction to evaluate population size dynamics over time using the BEAST 1.8.0 program [39]. All the concatenated sequences for each group were used. As no selection was observed for each gene, as well as for their respective codon positions (data not shown), we used all codon positions for each gene on this analysis. The evolutionary model, clock and genes regions were the same as those used in TMRCA inference. The following parameters were employed: 20 million generations for each haplogroup, parameter sampling every 1000 generations of the MCMC analysis and a 10% burn-in period. Convergence between runs and the performance of the analysis were checked using Tracer 1.5 and the plots were constructed with the aid of the same program [34].

Gene flow and migration

We evaluated the gene flow between haplogroups by the structure tests using Migrate-n v3.6 [42]. The migration parameters were estimated using concatenated gene regions. Migrate-n v3.6 estimates migration rate divided by the mutation rate M (m/μ). Four migration models were tested and compared by the bayes factor: (1) a full model with two population sizes and two migration rates; (2) a model with two population sizes and one migration rate (gene flow into the northwestern population); (3) a model with two population sizes and one migration rate (gene flow into the eastern population); (4) a model considering eastern and northwestern as a panmictic population. We used a Bayesian approach and thermodynamic integration of four chains with a static heating swap scheme (temperatures: 1.0, 1.5, 3.0, 106), with two independent runs, sampling at every 600th step for a total of 600,000 of recorded steps and a burn-in of 150,000 steps.


Genetic diversity

A fragment of 1,015 base pairs (Cytb– 433 bp; COI– 582 bp) was obtained for the concatenated genes of P. seridoensis and no indels occurred in these amplicons. We have no ambiguous bases or missing data in our dataset. Sixteen variable sites and 14 haplotypes were identified. The species exhibited low nucleotide diversity (π = 0.00266 ± 0.00019) and high haplotype diversity (Hd = 0.756 ± 0.044) (Table 2).

Table 2. Genetic diversity of P. seridoensis.

Number of samples (N), number of haplotypes (h), variables sites (S), nucleotide diversity (π) and haplotype diversity (Hd) for concatenated mitochondrial genes COI and Cytb in each haplogroup and all populations.

Phylogeographic structure

Based on the AIC, the evolutionary model selected for the Bayesian Inference, including the outgroup, was GTR + I. Fig 1B shows the topology based on this inference. Two haplogroups with a high value of branching support were identified. The first, eastern haplogroup, comprised samples from 11 locations (1 to 11 in Table 1), in which six haplotypes were identified. In this haplogroup, haplotype H1 was shared among six populations and all the other haplotypes were privative of the localities (black haplotypes in Fig 1C). The second, northwestern haplogroup, comprised samples from six locations (12 to 17 in Table 1), which exhibited eight haplotypes that were privative of the locations (grey haplotypes in Fig 1C). The bGMYC shows both haplogroups as a single evolutionary entity, with poor posterior probability values (PP< 0.9). AMOVA indicated high differentiation between haplogroups and among populations (ΦCT = 0.740, p < 0.0001; ΦST = 0.984; p < 0.0001, respectively).

The Mantel test demonstrated no correlation between genetic and geographic distances considering all populations (r = 0.171; p = 0.097) and populations in each haplogroup separately (reastern = -0.222, p = 0.812; rnortheastern = 0.245, p = 0.2468), indicating no isolation by distance among populations.

Divergence time and demographic inferences

The coalescent-based analysis considering all P. seridoensis populations indicated a TMRCA of 163 kya (highest posterior density [HPD 95%:]: 87.8–257.6 kya), suggesting that this could be the initial diversification time for the two P. seridoensis lineages analyzed herein. The TMRCA for each group indicated a recent age for these two lineages, with coalescence of northwestern and eastern groups respectively occurring 84.5 kya (HPD 95%: 35.7–149.6 kya) and 106.9 kya (HPD 95%: 49.3–180.5 kya). The BSP did not reveal any substantial fluctuations in the effective population size (Ne) in either haplogroup (Fig 2).

Fig 2. Coalescent Bayesian Skyline Plot (BSP) used to infer demographic history of P. seridoensis population groups.

Black horizontal line shows median BSP estimate and blue area shows upper and lower 95% highest posterior density limits.

Migration rates

We detected little gene flow between the eastern and northwestern populations in all models. Migration from the eastern to the northwestern populations was the most probable model (0.72) of gene flow (Table 3), with the number of migrants per generation (Nemeastern) of 0.63 (95% credibility interval: 0–2.87 migrants). All these parameters showed posterior probability curves with unique peaks and effective sample sizes (ESS) higher than 1,000, indicating convergence in the run.

Table 3. Bayes factors and log marginal likelihoods of the four evaluated migration models between northwestern and eastern haplogroups of P. seridoensis.


The population structure tests (phylogenetic inference, AMOVA and haplotype network) consistently characterized two haplogroups (northwestern and eastern). The results revealed a high number of privative haplotypes and a high level of differentiation between haplogroups as well as among all populations, with little gene flow mainly from the eastern to the northwestern haplogroup when the population genetic structure is assessed by mitochondrial genes. Moreover, the bGMYC analysis recovered a single evolutionary entity, suggesting that these haplogroups are not different species or even cryptic species. Previous study involving allozymic and microsatellite loci found a low population differentiation between two geographically close P. seridoensis populations in the state of Paraíba (Brazil) [12], suggesting restrictions to gene flow between the populations as the present study. Some biological traits of stingless bees limit dispersal. New queens of species of Partamona are philopatric, remaining in the place of origin and not flying more than 300 meters from the maternal nest during the swarming process [43,44,45]. Moreover, P. seridoensis builds its nests on specific substrates, nesting in arboreal termite nests [9,10,11]. Since stingless bees exhibit limited dispersal capacity, relatively weak geographical barriers may effectively restrict gene flow [46].

There is no evidence of potential geographic barriers (such as mountains, large rivers or valleys) in the present that could explain the diversification found between the two haplogroups of P. seridoensis. Moreover, no correlation was found between genetic and geographical distances, suggesting no isolation by distance among the populations analyzed. However, the differentiation that resulted in the two haplogroups may be explained by climatic changes in the mid-Pleistocene, since the initial diversification time for these lineages was estimated at approximately 163 kya, specifically in the interglacial period [47], a time when the biota is presumed to have been more associated with dry conditions and had more restricted and fragmented geographical distribution [24]. This event may have driven the diversification process, isolating northwestern and eastern haplogroups. Furthermore, like other stingless bees, P. seridoensis depends on plant resources, such as pollen, nectar and resins, as well the existence of suitable sites for nesting, such as termite nests. Such resources may also have had restricted geographical distribution due to climatic changes. These aspects and the presumed limited dispersal capacity of this stingless bee lend support to the hypothesis that mid-Pleistocene climatic changes may have driven the diversification of P. seridoensis into at least two haplogroups.

Similar results were observed to populations of Melipona subnitida, a stingless bee with a similar distribution area in the Caatinga [24]. The authors characterized two evolutionary lineages of the species, which also diverged in the mid-Pleistocene, and suggest that paleoclimatic changes in this period could have played an important role in this diversification. Other phylogeographic studies involving species associated with xeric vegetation in interglacial periods report data that are in agreement with the present findings [23,48,49,50]. Some researchers have also shown that climate oscillations in the Pleistocene may have promoted changes in the range of Neotropical forests in response to more warm and humid (interglacial periods) or more cold and dry (glacial periods) climatic conditions [51,52,53]. Moreover, climatic changes may have also influenced biogeographic patterns, genetic diversity, historical demography and species richness in the Neotropical biota [54,55] as well as the biota of other regions around the world [20,21,56].

Studies have suggested that forests expanded into the Caatinga, at least locally, during the wettest periods [57,58]. Population expansion for two evolutionary lineages of M. subnitida in the Caatinga during the Pleistocene period was reported [24]. However, the late Pleistocene climatic changes do not seem to have had a substantial influence on the demographic patterns of the P. seridoensis groups, since the BSP did not reveal any substantial fluctuations in the effective size of the populations in either haplogroup analyzed herein (Fig 2). As the two P. seridoensis haplogroups have a very recent origin, diverging after the Last Interglacial Maximum (approximately 120 kya), oscillations occurred in the late Pleistocene must not have drastically affected the population dynamics. A putative refuge area for Seasonally Dry Tropical Forests in the Caatinga between the Last Glacial Maximum and the present, which covers most of the current distribution area [59] was recently proposed. Such an area may have served as refuge for the species during this period, which allowed the demographic stability of the two lineages. This finding is in agreement with another study, which also found that climatic changes of the late Pleistocene did not seem to have a remarkable influence on the demographic patterns of P. rustica populations in Brazilian dry forests [23].

Unfortunately, there are no substitution rates yet available for the mtDNA (such as COI or Cytb genes) of bees. Therefore, recent studies involving bees [22,23,24] published in important journals have also used calibration described for other insects to estimate divergence time as well as historical demography. Studies have shown that the use of a “universal” mutation rate may overestimate the timing of population-level events [60,61,62]. It should be noted that the rate used in the present study agrees with others calibration rates estimated for the COI of other insects, such as 1.28 for beetles [63], 1.45 for ants [64] and 2.0 for Drosophila [65], as well as the COI and Cytb concatenated regions, such as 1.5–2.0 for Orthoptera [66] 0.8–1.3 for Diptera [67] and 2.1 for Coleoptera [68]. Therefore, the rate used in the present study does not differ much from those estimated for others insects and should not markedly compromise the results obtained herein.

The population differentiation of P. seridoensis found with mitochondrial genes is expected to be lower when assessed using nuclear genes, as observed in a previous study involving the population genetics of P. seridoensis using nuclear markers [12]. Like most Hymenopterans, stingless bees have a single multiallele locus that controls sex determination [26,69,70], in which hemizygous individuals are males, whereas diploid individuals that are heterozygous will develop into females and those homozygous at this sex locus will be diploid males [71]. Thus, this model predicts that inbreeding should produce diploids males that are highly harmful to colonies and populations. Due to the philopatric behavior of females, gene flow among populations must be accomplished by males, leading to asymmetrical sex dispersal, as also observed in other stingless bees [12,23,72,] and even bees from the tribe Euglossini in Neotropical regions [22,73,74,]. Hence, one would expect lower differentiation among populations based on nuclear genes due to the dispersal performed by males in these bee populations [23,72].

Recent study stressed the need for further, more detailed investigations of dry environments of the Neotropical region, since few phylogeographical studies have considered the biota of this area [23]. The present results contribute to a better understanding of evolutionary processes occurring in the Caatinga during the Pleistocene, as we have shown that the lineages of P. seridoensis diversified during the mid-Pleistocene likely due to climatic changes, resulting in two haplogroups. The conservation of different evolutionary lineages is important to preserving better adapted genotypes for greater survival during drought periods in the Caatinga [24]. Moreover, for species with low population gene flow and, consequently, high differentiation, as observed in the present study, it is important to preserve local populations in order to maintain the species throughout its range, which is an important source of information for planning conservation strategies [75,76]. Recent studies have shown that bee hunters, intense deforestation in Caatinga areas and livestock farming represent serious threat to stingless bees [13,23,77]. Studies involving organisms from dry forests as Caatinga are important since it has been excluded from discussions on conservation [78]. Given its importance as a pollinator and the fact that it is an endemic bee from a threatened dry forest, the present findings can be useful to the development of conservation strategies aimed at P. seridoensis populations, especially its two evolutionary lineages. Conservation policies directed at preserving pollinating species may be an effective mechanism for preserving the biodiversity of Neotropical biomes, which is an urgent task that must be pursued.


The authors are grateful to Dr. Silvia Regina de Menezes Pedro (FFCLRP—USP) for the bee identification, Carolina Machado, Carlos Congrains and Henrique Batalha Filho for help in some analysis, Michael Hrncir, Thiago Oliveira and Jânio Felix for help in the field and all beekeepers and locals who assisted in the fieldwork.

Author Contributions

  1. Conceptualization: EAM MADL.
  2. Formal analysis: EAM.
  3. Funding acquisition: EAM MADL.
  4. Investigation: EAM.
  5. Methodology: EAM KMF ATC.
  6. Resources: EAM MADL KMF CFM ATC CRF.
  7. Supervision: MADL.
  8. Writing – original draft: EAM MADL CFM.
  9. Writing – review & editing: EAM MADL.


  1. 1. Michener CD. The Bees of the World, 2nd ed. The Johns Hopkins University Press, Baltimore, Maryland; 2007.
  2. 2. Camargo JMF, & Pedro SEM. Meliponini Lepeletier, 1836. In Moure JS, Urban D, Melo GAR (Orgs). Catalogue of Bees (Hymenoptera, Apoidea) in the Neotropical Region—online version. 2013. Available at Accessed Feb/04/2017.
  3. 3. Pedro SRM. The stingless bee fauna in Brazil (Hymenoptera: Apidae). Sociobiology, 2014; 61: 348–354.
  4. 4. Slaa EJ, Chaves LAS, Malagodi-Braga KS, Hofsted FE. Stingless bees in applied pollination: practice and perspectives. Apidologie, 2006; 37: 293–315.
  5. 5. Klein AM, Vaissiere BE, Cane JH, Steffan-Dewenter I, Cunningham SA, Kremen C, et al. Importance of pollinators in changing landscapes for world crops. Proc R Soc Lond B Biol Sci. 2007; 274: 303–313.
  6. 6. Brown MJF, Paxton RJ. The conservation of bees: a global perspective. Apidologie. 2009; 40:410–416.
  7. 7. Basu P, Bhattacharya R, Ianetta PPM. A decline in pollinator dependent vegetable crop productivity in India indicates pollination limitation and consequent agro-economic crises. 2011. Available from: Nature Precedings,
  8. 8. Smith MR, Singh GM, Mozaffarian D, Myers SS. Effects of decreases of animal pollinators on human nutrition and global health: a modelling analysis. The Lancet. 2015; 386: 1964–1972.
  9. 9. Camargo JMF, Pedro SRM. Meliponini Neotropicais: o gênero Partamona Schwarz, 1939 (Hymenoptera: Apidae, Apinae)–Bionomia e Biogeografia. Rev Bras Entomol. 2003; 47: 31–372.
  10. 10. Fernandes CRM, Almeida AB, Del Lama MA; Martins CF. Nesting substrate characteristics of Partamona seridoensis Pedro & Camargo (Hymenoptera: Apidae) in areas of dry forest in Brazil. Sociobiology. 2017; 64(1): Forthcoming.
  11. 11. Oliveira TFFN, Silva LL, Hrncir M. Opportunistic occupation of nests of Microcerotermes spp. Silvestri (Termitidae: Termitinae) by Partamona seridoensis Camargo & Pedro (Apidae: Meliponini) in the Brazilian Tropical Dry Forest. Sociobiology. 2016.
  12. 12. Fernandes CR, Martins CF, Ferreira KM, Del Lama MA. Gene variation, population differentiation, and sociogenetic structure of nests of Partamona seridoensis (Hymenoptera: Apidae, Meliponini). Biochem Genet. 2011; 50: 325–335. pmid:21938561
  13. 13. Miranda EA, Carvalho AF, Andrade-Silva ACR, Silva CI, Del Lama MA. Natural history and biogeography of Partamona rustica, an endemic bee in dry forests of Brazil. Insect Soc. 2015; 62: 255–263.
  14. 14. Werneck FP. The diversification of eastern South American open vegetation biomes: historical biogeography and perspectives. Quat Sci Rev. 2011; 30: 1630–1648.
  15. 15. Ministério do Meio Ambiente (MMA) Caatinga. 2014. Available from:
  16. 16. Zanella FCV, Martins CF (2003). Abelhas da Caatinga: biogeografia, ecologia e conservação. In: Leal IR, Tabarelli M, Silva JMC (Org.). Ecologia e conservação da Caatinga. Recife: Ed. Universitária da UFPE, pp 75–134.
  17. 17. Giannini TC, Acosta AL, Silva CI, Oliveira PEAM, Imperatriz-Fonseca VL, Saraiva AM. Identifying the areas to preserve passion fruit pollination service in Brazilian Tropical Savannas under climate change. Agric Ecosyst Environ. 2013; 171: 39–46.
  18. 18. Serviço Florestal Brasileiro. Florestas do Brasil em resumo-2013: dados de 2007–2012. SFB, Brasília, Brazil. 2013. Available from:
  19. 19. Sánchez-Azofeifa GA, Quesada M, Rodríguez JP. Research priorities for neotropical dry forests. Biotropica. 2005; 37: 477–485.
  20. 20. Dellicour S, Kastally C, Varela S, Michez D, Rasmont P, Mardulyn P, Lecocq T. Ecological niche modelling and coalescent simulations to explore the recent geographical range history of five widespread bumblebee species in Europe. J. Biogeogr. 2016; 44: 39–50.
  21. 21. Groom S, Stevens M, Schwarz MP. Parallel responses of bees to Pleistocene climate change in three isolated archipelagos of the southwestern. Proc R Soc Lond B Biol Sci. 2014; 281: 20133293.
  22. 22. López-Uribe MM, Zamudio KR, Cardoso CF, Danforth BN. Climate, physiological tolerance and sex-biased dispersal shape genetic structure of Neotropical orchid bees. Mol Ecol. 2014; 23: 1874–1890. pmid:24641728
  23. 23. Miranda EA, Batalha-Filho H, Congrains C, Carvalho AF, Ferreira KM, Del Lama MA (2016) Phylogeography of Partamona rustica (Hymenoptera, Apidae), an endemic stingless bee from the Neotropical dry forest diagonal. PLoS ONE 11(10): e0164441. pmid:27723778
  24. 24. Bonatti V, Simões ZLP, Franco FF, Francoy TM. Evidence of at least two evolutionary lineages in Melipona subnitida (Apidae, Meliponini) suggested by mtDNA variability and geometric morphometrics of forewings. Naturwissenschaften. 2014; 101: 17–24. pmid:24384774
  25. 25. Cook JM, Crozier RH. Sex determination and population biology in the Hymenoptera. Trends Ecol Evol. 1995; 10: 281–286. pmid:21237037
  26. 26. Hasselmann M, Gempe T, Schiott M, Nunes-Silva CG, Otte M, Beye M. Evidence for the evolutionary nascence of a novel sex determination pathway in honeybees. Nature. 2008; 454: 519–523. pmid:18594516
  27. 27. Sheppard WS, McPheron BA. Ribosomal DNA diversity in Apidae. In: Smith DR (ed) Diversity in the genus Apis, Westview, Boulder (CO); 1991; pp. 89–102.
  28. 28. Dick CW, Roubik DW, Gruber KF, Bermingham E. Long-distance gene flow and cross-Andean dispersal of lowland rainforest bees (Apidae: Euglossini) revealed by comparative mitochondrial DNA phylogeography. Mol Ecol. 2004; 13: 3775–3785. pmid:15548290
  29. 29. Crozier RH, Crozier YC. The mitochondrial genome of the honeybee Apis mellifera: complete sequence and genome organization. Genetics. 1993; 133:97–117. pmid:8417993
  30. 30. Hall TA. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symp Ser. 1999; 41: 95–98.
  31. 31. Librado P, Rozas J. DnaSP v5: a software for comprehensive analysis of DNA polymorphism data. Bioinformatics. 2009; 25: 1451–1452. pmid:19346325
  32. 32. Nylander JAA. MrModeltest 2.2. Program Distributed by the Author. Evolutionary Biology Centre, Uppsala University; 2004.
  33. 33. Huelsenbeck JP, Ronquist F. MRBAYES: Bayesian inference of phylogeny. Bioinformatics. 2001; 17: 754–755. pmid:11524383
  34. 34. Rambaut A, Drummond AJ. Tracer, version 1.5. 2009. Available at:
  35. 35. Reid N, Carstens B. Phylogenetic estimation error can decrease the accuracy of species delimitation: a Bayesian implementation of the general mixed Yule-coalescent model. BMC Evol Biol. 2012; 12:196. pmid:23031350
  36. 36. Bandelt HJ, Forster P, Röhl A. Median-joining networks for inferring intraspecific phylogenies. Mol Biol Evol. 1999; 16: 37–48. pmid:10331250
  37. 37. Excoffier L, Lischer HEL. Arlequin version 3.5: a new series of programs to perform population genetics analyses under Linux and Windows. Mol Ecol Resour. 2010; 10: 564–567. pmid:21565059
  38. 38. Jensen JL, Bohonak AJ, Kelley ST. Isolation by distance web service v.3.23. BMC Genetics. 2005; 6: 13. pmid:15760479
  39. 39. Drummond AJ, Rambaut A, Suchard M. BEAST v1.8.0 2002–2013. Bayesian Evolutionary Analysis Sampling Trees. 2013. Available from:
  40. 40. Machado CA, Jousselin E, Kjellberg F, Compton SG and Herre EA. Phylogenetic relationships, historical biogeography and character evolution of fig-pollinating wasps. Proc R Soc Lond B Biol Sci. 2001; 268: 685–694.
  41. 41. Darriba D, Taboada GL, Doallo R, Posada D jModelTest 2: more models, new heuristics and parallel computing. Nat Methods. 2012; 9: 772.
  42. 42. Beerli P, Palczewski M. Unified framework to evaluate panmixia and migration direction among multiple sampling locations. Genetics. 2010; 185: 313–326. pmid:20176979
  43. 43. Engels W, Imperatriz-Fonseca VL. Caste development, reproductive strategies and control of fertility in honeybees and stingless bees. In: Social Insects: An Evolutionary Approach to Castes and Reproduction (Engels W, ed.), Springer-Verlag, Berlin; 1990. pp. 166–230.
  44. 44. Wille A, Orozco E. Observations on the founding of a new colony by Trigona cupira (Hymenoptera: Apidae) in Costa Rica. Rev Biol Trop. 1975; 22: 253–287.
  45. 45. Nogueira-Neto P. Notas bionômicas sobre Meliponíneos. III. Sobre a enxameagem. Arquivos do Museu Nacional do Rio de Janeiro. 1954; 42: 419–451.
  46. 46. Batalha-Filho H, Waldschmidt AM, Campos LAO, Tavares MG, Fernandes-Salomão TM. Phylogeography and historical demography of the Neotropical stingless bee Melipona quadrifasciata (Hymenoptera, Apidae): incongruence between morphology and mitochondrial DNA. Apidologie. 2010; 41: 534–547.
  47. 47. Kukla G. Saalian supercycle, Mindel/Riss interglacial and Milankovitch’s dating. Quat Sci Rev. 2005; 24: 1573–1583.
  48. 48. Wüster W, Ferguson JE, Quijada-Mascareñas JA, Pook CE, Salomão MG, Thorpe RS. Tracing an invasion: landbridges, refugia, and the phylogeography of the Neotropical rattlesnake (Serpentes: Viperidae: Crotalus durissus). Mol Ecol. 2005; 14: 1095–1108. pmid:15773938
  49. 49. Almeida FC, Bonvicino CR, Cordeiro-Estrela P. Phylogeny and temporal diversification of Calomys (Rodentia, Sigmodontinae): implications for the biogeography of an endemic genus of the open/dry biomes of South America. Mol Phylogenet Evol. 2007; 42: 449–466. pmid:16971145
  50. 50. Franco FF, Manfrin MH. (2013) Recent demographic history of cactophilic Drosophila species can be related to Quaternary palaeoclimatic changes in South America. J Biogeogr. 2013; 40: 142–154.
  51. 51. Ab’Saber . Espaços ocupados pela expansão dos climas secos da América do Sul, por ocasião dos períodos glaciais quaternários. Paleoclimas. 1977; 3: 1–19.
  52. 52. Pennington RT, Prado DE, Pendry CA. Neotropical seasonally dry forests and Quaternary vegetation changes. J Biogeogr. 2000; 27: 261–273.
  53. 53. Carnaval AC, Waltari E, Rodrigues MT, Rosauer D, VanDerWal J, Damasceno R, et al. Prediction of phylogeographic endemism in an environmentally complex biome. Proc R Soc Lond B Biol Sci. 2014; 281, 20141461.
  54. 54. Carnaval AC, Moritz C. Historical climate modelling predicts patterns of current biodiversity in the Brazilian Atlantic forest. J Biogeog. 2008; 35: 1187–1201.
  55. 55. Werneck FP, Leite RN, Geurgas SR, Rodrigues MT. Biogeographic history and cryptic diversity of saxicolous Tropiduridae lizards endemic to the semiarid Caatinga. BMC Evol Biol. 2015;
  56. 56. Xu X, Liu F, Chen J, Ono H, Agnarsson I, Li D, Kuntner M. Pre-Pleistocene geological events shaping diversification and distribution of primitively segmented spiders on East Asian margins. J Biogeog. 2016; 43: 1004–1019
  57. 57. Auler AS, Wang X, Edwards RL, Cheng H, Cristalli PS, Smart PL, et al. Quaternary ecological and geomorphic changes associated with rainfall events in presently semi-arid northeastern Brazil. J Quat Sci. 2004; 19: 693–701.
  58. 58. Carnaval AC, Bates JM. Amphibian DNA shows marked genetic structure and tracks Pleistocene climate change in Northeastern Brazil. Evolution. 2007; 61: 2942–2957. pmid:17941838
  59. 59. Werneck FP, Costa GC, Colli GR, Prado DE, Sites JW Jr. Revisiting the historical distribution of Seasonally Dry Tropical Forests: new insights based on palaeodistribution modelling and palynological evidence. Glob Ecol Biogeog. 2011; 20: 272–288.
  60. 60. Ho SYW, Lanfear R, Bromham L, Phillips MJ, Soubrier J, Rodrigo AG, Cooper A. Time-dependent rates of molecular evolution. Mol Ecol. 2011; 15: 3087–3101.
  61. 61. Papadopoulou A, Anastasiou I, Vogler AP. Revisiting the insect mitochondrial molecular clock: the mid-aegean trench Calibration. Mol Biol Evol. 2010; 27: 1659–1672. pmid:20167609
  62. 62. Grant WS. Problems and Cautions With Sequence Mismatch Analysis and Bayesian Skyline Plots to Infer Historical Demography. J Hered. 2015; 1–14.
  63. 63. Farrell BD. Evolutionary assembly of the milkweed fauna: cytochrome oxidase 1 and the age of Tetraopes beetles, Mol. Phylogenet Evol. 2001; 18: 469–478.
  64. 64. Resende HC, Yotoko KSC, Delabie JHC, Costa MA, Campiolo S, Tavares MG, Campos LAO, Fernandes-Salomão TM. Pliocene and Pleistocene events shaping the genetic diversity within the central corridor of the Brazilian Atlantic Forest. Biol J Linn Soc. 2010; 101:949–960.
  65. 65. Beckenbach A, Wang YW, Liu H. Relationships in the Drosophila obscura species group inferred from mitochondrial cytochrome oxidase II sequences. Mol Biol Evol. 1993; 10: 619–634. pmid:8393127
  66. 66. Shapiro LH, Strazanac JS, Roderick GK. Molecular phylogeny of Banza (Orthoptera: Tettigoniidae), the endemic katydids of the Hawaiian Archipelago. Mol Phylogen Evol. 2006; 41:53–63.
  67. 67. Esseghir S, Ready PD, Ben-Ismail R. Speciation of Phlebotomus sandflies of the subgenus Larroussius coincided with the late Miocene-Pliocene aridification of the Mediterranean subregion. Biol J Linn Soc. 2000; 70:189–219.
  68. 68. Pons J, Vogler AP. Complex pattern of coalescence and fast evolution of a mitochondrial rRNA pseudogene in a recent radiation of tiger beetles. Mol Biol Evol. 2005; 22: 991–1000. pmid:15647517
  69. 69. Hasselmann M, Gempe T, Schiott M, Nunes-Silva CG, Otte M, Beye M. Evidence for the evolutionary nascence of a novel sex determination pathway in honeybees. Nature. 2008; 454: 519–523. pmid:18594516
  70. 70. Brito DV, Nunes Silva CG, Hasselmann M, Viana LS, Astolfi-Filho S, Carvalho-Zilse GA. Molecular characterization of the gene feminizer in the stingless bee Melipona interrupta (Hymenoptera: Apidae) reveals association to sex and caste development. Insect Biochem Mol Biol. 2015; 66: 24–30. pmid:26393998
  71. 71. Tavares MG, Carvalho CR, Soares FAF, Fernandes A. Detection of diploid males in a natural colony of the cleptobiotic bee Lestrimelitta sp (Hymenoptera, Apidae). Genet Mol Biol. 2010: 33: 491–493. pmid:21637422
  72. 72. Cameron EC, Franck P, Oldroyd BP. Genetic structure of nest aggregations and drone congregations of the southeast Asian stingless bee Trigona collina. Mol Ecol. 2004; 13: 2357–2364. pmid:15245407
  73. 73. Cerântola NCM, Oi CA, Cervini M, Del Lama MA. Genetic differentiation of urban populations of Euglossa cordata Linnaeus 1758 from the State of São Paulo, Brazil. Apidologie. 2011; 42: 214–222.
  74. 74. Rocha-Filho LC, Cerântola NCM, Garófalo CA, Imperatriz-Fonseca VL, Del Lama MA. Genetic differentiation of the Euglossini (Hymenoptera, Apidae) populations on a mainland coastal plain and an island in southeastern Brazil. Genetica. 2013; 141: 65–74. pmid:23443762
  75. 75. Frankham R, Ballou JD, Briscoe DA. Introduction to conservation genetics, 2nd ed. Cambridge University Press; 2010.
  76. 76. Pullin AS, Báldi A, Can OEet al. Conservation focus on Europe: major conservation policy issues that need to be informed by conservation science. Conserv Biol. 2009; 23:818–824. pmid:19627313
  77. 77. Neves EL, Castro MS. Mandaçaia: uma abelha chave para a conservação da Caatinga. Candombá—Revista Virtual. Salvador. 2006; 2:1–3.
  78. 78. Klink CA, Machado RB. Conservation of the Brazilian Cerrado. Conserv Biol. 2005; 19: 707–713.