Journal of Heredity 2004:95(2):119-126
© 2004 The American Genetic Association
Genetic Structure of an Apis dorsata Population: The Significance of Migration and Colony Aggregation
From the School of Biological Sciences, A12, University of Sydney, 2006 NSW, Australia (Paar and Oldroyd); Institute of Zoology, Karl-Franzens University, 8010 Graz, Austria (Paar and Kastberger); and Institute of Apiculture, 3293 Lunz, Austria (Huettinger). We thank D. K. Sharma and Chetri Bopal at Gauhati University for their assistance in Assam, India. The study was supported by the Austrian Science Foundation and the Australian Research Council. We thank Pierre Franck for help with the statistical analyses, and Keith Goodnight for advice on the use of his program Relatedness.
Address correspondence to Jürgen Paar, School of Biological Sciences, A12, University of Sydney, 2006 NSW, Australia, or e-mail: juergen_paar{at}aon.at.
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
Eight microsatellite loci were used to investigate the genetic structure of the giant honeybee (Apis dorsata) population in northeast India. This species migrates seasonally between summer and winter nesting sites, and queens appear to return to their previously occupied site. Furthermore, there is a strong tendency for colonies of this species to aggregate at perennially utilized nesting sites that may be shared by more than 150 colonies. These behavioral features suggest that colonies within aggregations should be more related than random colonies, but that the long-distance migration could act to minimize genetic differentiation both between geographical areas and within aggregations. Our genetic study supports these conjectures arising from natural history. A. dorsata aggregations are comprised of colonies that share more alleles than expected by chance. Although queens heading neighboring colonies are not close relatives, fixation indices show significant genetic differentiation among aggregation sites. However, there appears to be sufficient gene flow among aggregations to prevent high degrees of relatedness developing between colonies within aggregations. The results also suggest that there is significant population structuring between geographical regions, although the level of structuring caused by aggregation exceeds the differentiation attributable to geographic region.
The range of the Asian giant honeybee (A. dorsata Fabricius) extends from India to Southeast Asia, including the Philippines (Ruttner 1988). Its massive colonies are characterized by a single comb that is usually attached to a tree branch or cliff overhang, but sometimes to the eves of buildings or other urban structures.
Colonies comprise up to 100,000 individuals (Morse and Laigo 1969). Large amounts of honey (up to 45 kg; Ruttner 1988) can be stored by a colony, and for this reason, wild A. dorsata nests are frequently harvested throughout its range. For many people, honey (and sometimes brood) harvested from A. dorsata nests provides an important source of income. However, unsustainable harvesting methods, deforestation, and destruction of suitable nest sites may threaten local populations. Knowledge of the population structure and ecology of A. dorsata is needed for effective management and conservation of this species (Padmanabhan 2000).
The giant honeybee differs substantially in behavior and ecology from other Apis species (Morse and Laigo 1969). First, colonies are frequently found in dense aggregations (Kastberger and Sharma 2000; Koeniger and Koeniger 1980; Lindauer 1956; Roepke 1930; Ruttner 1988). Up to 200 colonies may occur on a single tree (Oldroyd et al. 2000) and these are often separated by only a few centimeters. Second, colonies often undergo seasonal migration between alternate nesting sites. In these populations, nest sites tend to be occupied for 3–4 months. Toward the end of this period, brood rearing ceases and the honey and pollen stores are depleted (Ruttner 1988). Eventually the colonies abscond to alternative nest sites which may be 200 km distant (Crane et al. 1993; Koeniger and Koeniger 1980). The proximate cause of migration may be related to available forage. A. dorsata swarms have been observed to travel between habitats with different blooming seasons (Crane et al. 1993; Koeniger and Koeniger 1980; Mahindre 2000). Absconding may also help control levels of the parasitic mite Tropilaelaps clareae, which needs brood in order to reproduce. Thus a colony may reduce infestation by this parasite with a period of broodless migration (Rinderer et al. 1994).
Nesting sites are reoccupied year after year over periods of several decades or more (Oldroyd et al. 2000). Amazingly, some returning colonies find their way back to exactly the same nesting structure they occupied the previous season (Neumann et al. 2000; Paar et al. 2000). This is despite the fact that workers probably live for less than 2 months (Otis et al. 1990), so only the queen could have direct experience of the former nest site. One colony is documented to have returned after an absence of 2 years (Paar et al. 2000). Such precise homing is only known for migratory vertebrates, particularly salmon, toads, storks, and seals (reviewed in Papi [1992]) and has not been reported for any other insect.
As in other honeybees, mating of virgin queens and drones in A. dorsata takes place during flight at some distance from the colony. Queens are known to mate with up to 100 drones over two to three successive mating flights (Wattanachaiyingcharoen et al. 2003). Drones leave their colonies at dusk, returning 15–30 min after departure (Koeniger et al. 1994; Koeniger and Wijayagunesekera 1976; Rinderer et al. 1993). The drones fly to well-defined perennial drone congregation areas (DCAs) (Koeniger et al. 1994). Although Koeniger et al. (1994) were able to track drones from a colony aggregation directly to a drone congregation area, it is unknown if all drones from all colonies in an aggregation fly to a single DCA, to several, or if several aggregations service the same DCA. In Apis cerana and Apis mellifera, drones from many colonies within an area visit the same DCA (Baudry et al. 1998; Koeniger et al. 1994). Baudry et al. (1998) calculated that 240 colonies were represented by drones in one A. mellifera congregation. This indicated that drones from most of the colonies within the recruitment range of this DCA were represented. This would lead to panmixis.
During reproductive swarming there seems to be two ways in which A. dorsata colonies divide (Lindauer 1956). In the first, the queen leaves the nest followed by a swarm of workers, as is observed in A. mellifera (Seeley 1985). The average distance those swarms move away from the natal nest is unknown, but on at least one occasion such a swarm traveled more than 500 m (Lindauer 1956). In the second mechanism, the queen and a group of workers separate from the nest and start a new colony a short distance from the mother colony without forming an intermediate cluster, and without taking flight. This form of reproduction is called "budding" (Lindauer 1956).
These features of A. dorsata biology raise the following questions regarding the population structure of this species: First, do aggregations or aggregations and nests in close proximity to these aggregations comprise closed subpopulations? A. dorsata shows very strong nest site fidelity (Neumann et al. 2000; Paar et al. 2000). If aggregations move as a cohesive unit during seasonal migration to an alternative aggregation site and return to the original site in the following season, this would reduce gene flow among aggregations. This would be especially so if each aggregation has its own unique DCA.
Second, do aggregations consist of closely related colonies? If "budding" is a common means of reproductive swarming, then it would be predicted that some colonies within aggregations would be related as mother and daughter colonies. Oldroyd et al. (2000) demonstrated that, out of seven nests within an aggregation in Malaysia, not one pair was this closely related. This suggests that budding, as observed by Lindauer (1956), may not be a common mechanism of reproductive swarming in A. dorsata.
Third, in what ways do seasonal migration influence population structure in A. dorsata? Subdivision of populations in an established range is not expected in species with long-range seasonal migration, as migration enhances gene flow (Arguedas and Parker 2000) and tends to lead to single, panmictic populations (Barrowclough 1980). Studies in the honeybee (A. m. mellifera) showed population subdivision on the European continent (Cornuet et al. 1975, 1978; Estoup et al. 1995) where colonies do not migrate, but no substructure is found in populations of the migratory African subspecies adansonii and intermissa (Chandler 1976; Cornuet et al. 1988; Gadbin et al. 1979; Grissa et al. 1990; Ruttner 1988).
Fourth, do geographic obstacles direct migration? Swarms of Apis species are known to travel for more than 10 km over open water (Hannabus 1939; Roubik 1989; Roubik and Boreham 1990) and calculations suggest that fully laden flying bees can cover distances of up to 90 (Otis et al. 1981) or even 110 km (Seeley 1985). However, despite their physiological ability to cover large distances, it may be that scout bees of migrating swarms are disinclined to cross hostile habitat. Very broad rivers, for example, could lead to subdivision within one population.
To explore these questions, we examined the genetic structure of the A. dorsata population in Assam, India, using microsatellites as genetic markers. Colonies were sampled from aggregations and solitary nests north and south of the Brahmaputra River, which divides Assam as a 8–15 km band of water and vegetation-free sandbanks. Colonies are thought to migrate at the end of dry season from the lowlands into the forests of the foothills of the Himalayas in the north and the Megalaya in the south, to return to the lowlands at the Brahmaputra at the beginning of the dry season 7–8 months later (Paar and Kastberger 1998).
To examine the genetic structure of aggregations we compared the degrees of relatedness of colonies within aggregations and between colonies in different aggregations in their vicinity. We also examined the possibility of the development of subpopulations by comparing relatedness coefficients and fixation indices among different aggregations and geographical regions.
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Honeybee Samples
Adult workers were collected from each of 32 colonies in November 1998 in Assam, northeast India (Figure 1). Here we define the following putative subdivisions within the A. dorsata population that were tested in this study (Figure 1): (1) the total population includes all colonies within the study; (2) the three regions—southwest, northwest, northeast—are geographical areas postulated to represent subpopulations, based on distance and geographic barriers; (3) aggregations within regions represented by a variable number of colonies, assembled on a single tree or building, including solitary nests in close (<40 m) proximity to the main group; and (4) single colonies (nests), which nested at least 1 km from aggregations.
|
= single coloniesOur study comprised samples from six aggregations (Figure 1). The sample from aggregation A consisted of eight colonies from the 30 nests present; the sample from aggregation C comprised two nests of the three present. All colonies present were sampled at aggregations B (three colonies), E (eight colonies), F (four colonies), and G (two colonies). There were also five solitary nests (Figure 1).
Samples were taken carefully in the early morning or in the late evening, without causing workers to take flight from the nest. Workers were collected from the bee curtain on the lower edge of the comb. The individuals sampled were stored in 99% ethanol.
DNA Extraction and PCR Amplification
DNA was extracted from one hind leg of each worker bee using a 5% Chelex solution (Walsh et al. 1991). The Chelex extracts were then diluted 1:4 with distilled water. The microsatellite loci used were A14, A88, A107, A76, and A24 cloned from A. mellifera (Estoup et al. 1993), B124 from Bombus terrestris (Estoup et al. 1993), Tc3-302 from Trigona cabonaria (Green et al. 2001), and Ad3, specifically cloned for this study from A. dorsata according to the procedures outlined in Green et al. (2001). The primer sequences for this latter locus are:
- 5' CCGTAACTGGACTTCTTTCCCTCC 3' (forward)
- 5' GACAATGGCGTACTTTGTGG 3' (reverse)
- 5' GACAATGGCGTACTTTGTGG 3' (reverse)
Data Analysis
Workers (24 to 34 bees per colony) were analyzed to infer the genotypes of the queens and their drone mates (Oldroyd et al. 1996). Under the assumption of monogyny (Oldroyd et al. 1996), all workers from a specific colony will carry one of the two alleles carried by the queen at any given locus. Having inferred the queen alleles, we excluded any worker that did not carry a queen allele as a drifted individual. (The frequency of drifted workers was very low; 0–1.2%).
Exact tests for Hardy-Weinberg equilibrium at each locus and genotypic linkage disequilibrium were calculated using the software package GENEPOP 3.2 (Raymond and Rousset 1995). Analyses were carried out with the queen alleles only. We caution that this test is invalid for the northeast region, where the sample size is too low. However, it is necessary to avoid using the actual worker genotypes to test for linkage and Hardy-Weinberg equilibrium because colony structure could cause significant deviations where none exists on a population-wide basis. Allele frequency differences (genic differentiation) between regions were also tested with GENEPOP 3.2.
FST calculations were based on a derived dataset. For social insects, population allele frequencies cannot be estimated from worker genotypes because every worker in a colony shares the queen's genotype. We therefore used inferred paternal alleles as a basis for calculations. To do this we first determined the genotype of the queen heading each colony from the array of worker genotypes of each colony (Oldroyd et al. 1996). Then the genotype of each worker's father was determined by subtracting the queen allele of each worker (Oldroyd et al. 1996). The array of paternal alleles was then diploidized to obtain a dataset for analysis. Genetic differentiation between the hierarchical levels of the population (total population, regions, and aggregations) was then quantified using FST estimates (Weir and Cockerham 1984) computed with the software FSTAT v. 2.9.1 (Goudet 2000).
GENEPOP 3.2 was used to test for isolation by distance. The correlations between FST/(1 – FST) and geographical distance between aggregations were tested using a permutation procedure (Mantel 1967). The effective number of migrants (Nm) between aggregations and regions was calculated according to Slatkin (1985) based on the distribution of rare alleles in the population and corrected for sample size following the procedure of Barton and Slatkin (1986). The rare allele estimate of Nm is preferred for data that use highly polymorphic loci such as microsatellites (Hedrick 1999).
Degrees of genetic relatedness, R, among colonies and between aggregations were computed with the program Relatedness 5.08 (by K. F. Goodnight), which uses an algorithm described by Queller and Goodnight (1989). Calculations of Nm and analysis with Relatedness 5.08 were based on worker genotypes. The respective regions provided the allele frequencies needed as reference populations (southwest: 15 colonies, 435 individuals; northeast: 14 colonies, 422 individuals).
In addition, intracolony pedigree relatedness was calculated as
|
|
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Hardy-Weinberg Equilibrium and Linkage Disequilibrium
Tests for Hardy-Weinberg equilibrium were performed for each region and each locus. No population showed significant deviation from Hardy-Weinberg equilibrium (Fisher's exact test, northeast: P = 0.16; southwest: P = 0.17). There was also no significant deviation from the equilibrium over all loci and all populations (P = 0.12). Exact tests for linkage disequilibrium showed no significant deviations in any of the 28 comparisons. Furthermore, no individual region (southwest, northwest, northeast) separately showed significant linkage disequilibrium (P
0.1 after Bonferroni correction).
Relatedness within Aggregations
No pair of colonies had sufficiently high intercolony relatedness, based on alleles shared by the workers, to suggest a mother-daughter or a half-sister relationship between the queens heading the colony pairs. Sixty-seven comparisons of queen genotypes within the six aggregations showed only one pair of queens that carried at least one identical allele at every locus studied (Table 1: aggregation E, colonies 1 and 2). These two queens could therefore be related as mother and daughter. The calculated relatedness based on allele sharing, R, of 0.151 suggests that these two individuals could have been related as cousins (expected value 0.125), but were probably not mother and daughter. A further three pairs of queens (Table 1: aggregation A, colonies 1 and 4; aggregation C, colonies 1 and 2; aggregation G, colonies 1 and 2) showed a level of relatedness compatible with them being cousins.
|
Observed intracolony relatedness, R, varied between 0.2 and 0.5, which is unexpected because in colonies with highly polyandrous queens, relatedness should approach 0.25 (Laidlaw and Page 1984; Oldroyd and Moran 1983). We therefore performed alternative calculations of pedigree relatedness, r, based on the effective number of queen mates (Crozier and Pamilo 1996) rather than allele sharing (Queller and Goodnight 1989). These calculations showed intracolony relatedness for all colonies ranging between 0.26 and 0.28, as expected.
Population Subdivision Between Aggregations
Interaggregational R values were universally slightly negative, and not significantly different from zero, indicating no relatedness between the aggregations studied, and a possible tendency of queens and drones to avoid mates from other aggregations (Figure 1). Within aggregations, however, mean relatedness between colonies was always positive (Figure 1). These results indicate that in the population studied, colonies within aggregations are more highly related than random colonies drawn from the same population.
Fisher's exact tests for multilocus identical allelic distributions across tested populations showed significant separation of aggregation E from aggregation F and aggregation E from aggregation G in the southwest and aggregation A from aggregation B in the northeast (Figure 2). The F values were significantly different from zero for all pairwise comparisons among aggregations (P < 0.01) (Figure 2). All fixation indices FAR and FAT, measuring the proportional reduction in heterozygosity of the aggregations relative to the regional population and the aggregations relative to the total population, were significantly greater than zero (Table 2). Testing for the correlation between genetic differentiation and geographic distance showed that this differentiation was not based on geographic separation (r = 0.23, P = 0.23, df = 15). Estimates of Nm, the effective number of migrants that come into each aggregation in each generation, resulted in a mean value of 0.95 (SD = 0.31) in the southwest region and 0.82 (SD = 0.15) in the northeast region. Even aggregations separated by only 0.2 km (A and C) showed a number of migrants as low as 0.97 individuals per generation.
|
|
Population Subdivision Between Regions
FST values between the three regions (southeast, northeast, northwest) were significantly different from zero (Figure 3). Tests for allele frequency differences (genic differentiation) among regions showed that the southwest and the northeast subpopulations had significantly different allele frequencies (Figure 3).
|
FRT between regions and the total population also differed significantly from zero (Table 2). However, comparison of the magnitude of FRT estimates with those of the FAT and FAR estimates (Table 2) shows that regions are much less important contributors to population subdivision than aggregations (t tests, P(FRT – FAT) = 0.017; P(FRT – FAR) = 0.001; df = 7).
Analysis of Nm showed that at an average of 1.72 (SD = 0.25) queens or drones move between regions in each generation. This value is significantly higher (t test, P = 0.002; df = 7) than the migration rate between aggregations within regions.
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
A. dorsata, a honeybee species with long-range seasonal migration, is not panmictic. Our results suggest that aggregations represent distinct breeding units, but that there is also significant gene flow among aggregations.
We regard three models of migratory behavior as plausible for A. dorsata. In the first, aggregations migrate as a single unit to an alternate nesting site and return in the same grouping in the following season. In the second, each nest oscillates between different aggregation sites. In the third, aggregations migrate as a cohesive group between aggregation sites, but with some colonies changing to other aggregations.
Based on the relatively low degree of relatedness, R, among colonies within aggregations (Oldroyd et al. 2000) (Figure 2), we can reject the first model. However, we are unable to distinguish between the other two hypotheses. Even a small number of migrants between breeding units (aggregations in our case) greatly reduces the level of relatedness within each aggregation (Hamilton 1975).
If great numbers of different colonies contribute drones to A. dorsata DCAs (as is observed in A. mellifera and A. cerana), then this would tend to lower the relatedness among colonies within aggregations. Mating flight distances of less than 5 km (Koeniger et al. 1994) would reduce the likelihood of gene flow between different aggregations. The significantly higher relatedness within an aggregation compared to the relatedness between different aggregations, even with aggregations only separated by 0.2 km, implies that each aggregation has its own DCA. This may occur because A. dorsata mates at dusk and mating flights are extremely brief relative to other honeybee species (Koeniger et al. 1994; Rinderer et al. 1993). In addition, the distances between aggregations are often more than 5 km, thus reducing the possibility of shared DCAs.
Despite impediments to gene flow among aggregations, they do not consist of closely related colonies. Our results confirm the findings of Oldroyd et al. (2000), who suggested that nests within aggregations are not related as mother and daughter colonies formed by budding. Further, we can say that aggregations are not comprised solely of relatives like half sisters, and only in rare cases can relatives such as cousins be found. Therefore the hypothesis that aggregations might occur because of budding (as observed by Lindauer [1956]) or very short movement by reproductive swarms (as speculated by Oldroyd et al. [1995]) can be rejected. This is also supported by the finding that subdivision was not correlated with geographic distance, as both mechanisms of reproductive swarming would lead to genetic differentiation as a function of distance (Pamilo 1998). Rather, the results support the hypothesis that aggregations may arise as an adaptive consequence of the mating system. A. dorsata is highly polyandrous (Moritz et al. 1995; Oldroyd et al. 1996), with effective queen mating frequencies up to 88.5 (Wattanachaiyingcharoen et al., 2003). As mating is potentially hazardous for the queen (Moritz 1985), the nearby presence of unrelated colonies servicing the same DCA may enhance fitness (Oldroyd et al. 1995).
The results also support our original hypothesis of differentiation of the population in Assam, India, into the three regions—southeast, northeast, northwest (Figure 1). Slight, but significant structuring is found between the postulated regions, although the degree of differentiation between regions is not as pronounced as that between aggregations. The estimated effective number of migrants per generation (Nm) between regions and between aggregations strengthens this interpretation, as there are significantly more migrants between regions than between aggregations.
Comparing the degree of genetic differentiation between the two regions on the north bank of the Brahmaputra River and between regions south of the river showed no significant differences. This indicates that the Brahmaputra River is not an important barrier to gene flow, implying that geographical obstacles seem to have no role in directing migrating swarms and that scout bees are not disinclined to cross hostile habitat. Rather, it seems that nest site fidelity and mating at DCAs lead to elevated relatedness among colonies within aggregations, but long-distance migration tends to equalize allele frequencies across a broad area.
The surprisingly large intracolony R values based on the allele frequencies (Queller and Goodnight 1989) compared to the pedigree relatedness, r, based on the effective paternity frequency, as well as the slightly negative values of intercolony relatedness, are probably explained by the limited number of individuals and loci studied. However, this does not detract from our general conclusion of some population structure based on aggregations and regions.
The importance of aggregations to A. dorsata biology suggests that management and conservation strategies should focus not on whole populations, but on smaller geographical areas or single aggregation sites. A. dorsata is an important pollinator and a valuable source of income through honey and wax production for many rural people throughout Asia. Therefore considerate, nondestructive methods of harvesting and protection of aggregation sites will probably need to be introduced to prevent extinction of local populations.
|
Footnotes |
|---|
Corresponding Editor: Stephen Schaeffer
Received June 12, 2003
Accepted December 23, 2003
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
-
Arguedas N, Parker P, 2000. Seasonal migration and genetic population structure in house wrens. Condor. 102:517-528.[CrossRef]
Barrowclough GF, 1980. Gene flow, effective population sizes, and genetic variance components in birds. Evolution. 34:789-798.[CrossRef]
Barton NH, Slatkin M, 1986. A quasi-equilibrium theory of the distribution of rare alleles in a subdivided population. Heredity. 56:409-415.
Baudry E, Solignac M, Garnery L, Cornuet JM, Koeniger N, 1998. Relatedness among honeybees (Apis mellifera) of a drone congregation. Proc R Soc Lond B. 265:2009-2014.[CrossRef]
Chandler MT, 1976. The African honeybee—Apis mellifera adansonii: the biological basis of its management. In: Apiculture in tropical climates (Crane E, ed). London: International Bee Research Association; 61–68.
Cornuet J, Daoudi A, Mohssine EH, Fresnaye J, 1988. Étude biométrique de populations d'abeilles marocaines. Apidologie. 19:355-366.
Cornuet J, Fresnaye J, Lavie P, 1978. Étude biométrique de deux populations d'abeilles cévenoles. Apidologie. 9:41-55.
Cornuet J, Fresnaye J, Tassencourt L, 1975. Discrimination et classification de populations d'abeilles a patir de caractéres biométrique. Apidologie. 6:145-187.
Crane E, van Luyen V, Mulder V, Ta TC, 1993. Traditional management system for Apis dorsata in submerged forests in southern Vietnam and central Kalimantan. Bee World. 74:27-40.
Crozier R, 1970. Coefficients of relationship and the identity of genes by descent in the social Hymenoptera. Am Nat. 104:216-217.[CrossRef][ISI]
Crozier R, Pamilo P, 1996. Sex allocation and kin-selection in social insects. Oxford: Oxford University Press.
Estoup A, Garnery L, Solignac M, Cornuet J, 1995. Microsatellite variation in honey bee (Apis mellifera L.) populations: hierarchical genetic structure and test of the infinite allele and stepwise mutation model. Genetics. 140:679-695.[Abstract]
Estoup A, Solignac M, Harry M, Cornuet J, 1993. Characterization of (GT)n and (CT)n microsatellites in two insect species: Apis mellifera and Bombus terrestris. Nucleic Acids Res. 21:1427-1431.
Gadbin C, Cornuet J, Fresnaye J, 1979. Approche biométrique de la variété locale d'Apis mellifera L. dans le sud tchadien. Apidologie. 10:137-148.
Goudet J, 1995. FSTAT (version 1.2): a computer program to calculate F-statistics. J Hered. 86:485-486.
Goudet J, 2000. FSTAT, a program to estimate and test gene diversities and fixation indexes (updated from Goudet 1995) http://www.unil.ch/izea/softwares/fstat.html.
Green CL, Franck P, Oldroyd B, 2001. Characterization of microsatellites for Trigona carbonaria, a stingless bee endemic to Australia. Mol Ecol Notes. 1:89-92.
Grissa K, Cornuet J, Msadda K, Fresnaye J, 1990. Étude biométrique de populations d'abeilles tunisiennes. Apidologie. 21:303-310.
Hamilton WD, 1975. Innate social aptitudes of man: an approach from evolutionary genetics. In: Biosocial anthropology (Fox R, ed). New York: Wiley; 133–155.
Hannabus CH, 1939. Swarm flies seven miles. S Afr Bee J. 14:8.
Hedrick PW, 1999. Perspective: highly variable loci and their interpretation in evolution and conservation. Evolution. 53:313-318.[CrossRef][ISI]
Kastberger G, Sharma DK, 2000. The predator-prey interaction between blue-bearded bee eaters (Nyctyornis athertoni Jardine and Selby 1830) and giant honeybees (Apis dorsata Fabricius 1798). Apidologie. 31:727-736.[CrossRef]
Koeniger N, Koeniger G, 1980. Observations and experiments on migration and dance communication of Apis dorsata in Sri Lanka. J Apic Res. 19:21-34.
Koeniger N, Koeniger G, Tingek S, Kalitu A, Mardan M, 1994. Drones of Apis dorsata (Fabricius 1793) congregate under the canopy of tall emergent trees in Borneo. Apidologie. 25:249-264.
Koeniger N, Wijayagunasekera HNP, 1976. Time of drone flight in the three Asiatic honeybee species (Apis cerana, Apis florea, Apis dorsata). J Apic Res. 15:67-71.
Laidlaw HH, Page R, 1984. Polyandry in honey bees (Apis mellifera L.): sperm utilization and intracolony genetic relationships. Genetics. 108:985-997.
Lindauer M, 1956. Communication among the honeybees and stingless bees of India. Bee World. 38:34-39.
Mahindre DB, 2000. Developments in the management of Apis dorsata colonies. Bee World. 81:155-163.
Mantel N, 1967. The detection of disease clustering and generalized regression approach. Cancer Res. 27:209-220.[ISI][Medline]
Moritz R, 1985. The effects of multiple mating on the worker-queen conflict in Apis mellifera. Behav Ecol Sociobiol. 16:375-377.[CrossRef]
Moritz RFA, Kryger P, Koeniger G, Koeniger N, Estoup A, Tingek S, 1995. High degree of polyandry in Apis dorsata queens detected by DNA microsatellite variability. Behav Ecol Sociobiol. 37:357-363.[CrossRef]
Morse R, Laigo FM, 1969. Apis dorsata in the Philippines. Laguana: Philippine Association of Entomologists.
Neumann P, Koeniger N, Koeniger G, Tingek S, Kryger P, Moritz MFA, 2000. Home-site fidelity in migratory honeybees. Nature. 406:474-475.[CrossRef][Medline]
Oldroyd B, Moran C, 1983. Heritability of worker characters in the honeybee (Apis mellifera). Aust J Biol Sci. 36:323-332.
Oldroyd B, Osborne K, Mardan M, 2000. Colony relatedness in aggregations of Apis dorsata Fabricius (Hymenoptera, Apidae). Insectes Soc. 47:94-95.[CrossRef]
Oldroyd B, Smolenski AJ, Cornuet J, Wongsiri S, Estoup A, Rinderer TE, Crozier RH, 1996. Levels of polyandry and intracolonial genetic relationships in Apis dorsata (Hymenoptera: Apidae). Ann Entomol Soc Am. 89:276-283.
Oldroyd B, Smolenski A, Lawler S, Estoup A, Crozier R, 1995. Colony aggregations in Apis mellifera. L. Apidologie. 26:119-130.
Otis GW, Mardan M, McGee K, 1990. Age polytheism in Apis dorsata. In: Social insects and the environment. Proceedings of the 11th International Congress of IUSSI, 1990 (Veeresh GK, Mallick B, and Viraktamath CA, eds). New Delhi: Oxford & IBH Publishing; 378.
Otis GW, Winston ML, Taylor OR, 1981. Engorgement and dispersal of Africanized honeybee swarms. J Apic Res. 20:3-12.
Paar J, Kastberger G, 1998. Seasonal distribution of giant honeybees in northeast India. In: Proceedings of the XIII International Congress of IUSSI, Adelaide, 1998 (Schwarz MP and Hogendoorn K, eds). XIII International Congress of IUSSI, Inc.; 358.
Paar J, Oldroyd B, Kastberger G, 2000. Giant honeybees return to their nest sites. Nature. 406:475.[CrossRef][Medline]
Padmanabhan P, 2000. Problems of bee-keeping in Kerala, S. India. In: Seventh IBRA Conference on Tropical Bees: Management and Diversity, Chiang Mai, 2001 (Wongsiri S, ed). Cardiff, UK: International Bee Research Association; 191.
Pamilo P, 1998. Genetic structure of populations and sociobiology. In: Proceedings of the XIII International Congress of IUSSI, Adelaide, 1998 (Schwarz MP and Hogendoorn K, eds). XIII International Congress of IUSSI, Inc.; 11.
Papi F, 1992. Animal homing. London: Chapman and Hall.
Queller D, Goodnight KF, 1989. Estimating relatedness using genetic markers. Evolution. 43:258-275.[CrossRef][ISI]
Raymond M, Rousset F, 1995. GENEPOP (version 1.2): population genetics software for exact tests and ecumenicism. J Hered. 86:248-249.
Rinderer T, Oldroyd B, Wongsiri S, Sylvester HA, De Guzman LI, Potichot S, Sheppard WS, Buchmann SL, 1993. Time of drone flight in four honey bee species in south-eastern Thailand. J Apic Res. 32:27-33.
Rinderer T, Oldroyd B, Lekprayoon C, Wongsiri S, Boonthai C, Thapa R, 1994. Extended survival of the parasitic honey bee mite Tropilaelaps clareae on adult workers of Apis mellifera and Apis dorsata. J Apic Res. 33:171-173.
Roepke W, 1930. Beobachtungen an indischen Honigbienen, insbesondere an Apis dorsata. Meded Landbouwhoogeschool Wageningen. 34:1-28.
Roubik D, 1989. Ecology and natural history of tropical bees. Cambridge: Cambridge University Press.
Roubik D, Boreham MW, 1990. Learning to live with Africanized honey bees. Interciencia. 15:146-153.
Ruttner F, 1988. Biogeography and taxonomy of honeybees. Berlin: Springer-Verlag.
Seeley TD, 1985. Honeybee ecology: a study of adaptation in social life. Princeton, NJ: Princeton University Press.
Slatkin M, 1985. Rare alleles as indicators of gene flow. Evolution. 39:53-65.[CrossRef]
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. 10:507.
Weir BS, Cockerham CC, 1984. Estimating F-statistics for the analysis of population structure. Evolution. 38:1358-1370.[CrossRef][ISI]
Wattanachaiyingcharoen W, Oldroyd B, Wongsiri S, Palmer K, Paar J, 2003. A scientific note on the mating frequency of Apis dorsata. Apidologie. 34:85-86.[CrossRef]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||