Queen Genotypes Research Articles

Estimation of the number of matings inApis mellifera: extensions of the model and comparison of different estimates

The multiple mating behaviour of Apis mellifera queens is an important reproductive characteristic of this species. Estimates of the number of drones that mate with a queen can be obtained through segregation analysis of genetic markers in samples of workers from several colonies. Statistical methods have been developed that allow estimation of this parameter using segregations of worker genotypes at one locus for which the genotypes of queens are known. Here we present simple modifications of previous models that allow estimation of the number of matings in Apis mellifera from multiple loci and genotypic segregations of workers and drones. Estimates of the mean and the variance are obtained by maximum likelihood. A sample consisting of 200 colonies of Africanized honeybees from Sâo Paulo State, Brazil, gives an estimate of 16.1 matings per queen, very close to the previous estimate (17.2) obtained 15 years before with a different gene marker and a different population of Africanized honeybees.

Maximum likelihood estimates of gene frequencies and racial admixture in Apis mellifera L. (Africanized honeybees)

In order to obtain estimates of gene frequencies and other genetic parameters in Apis mellifera, samples are usually obtained from groups of haploid male (drones) and sterile female (workers) offspring, each group being descended from a multiply-mated single queen. A probabilistic model of segregation is presented which allows for maximum-likelihood estimates of genetic parameters in the reproductive individuals using the genotypic segregations of workers and drones from different colonies. Marginal probabilities of these segregations are calculated for each queen genotype by assuming a Dirichlet distribution for the fertilization frequencies of the different male alleles stored in the spermathecae. Estimates of gene frequencies and racial proportions in Africanized honey bees from different localities in Brazil and Central America reveal geographical heterogeneity in these parameters, possibly as a result of different levels of gene flow from European races.

Conditions for genetic polymorphism in social Hymentopera under selection at the colony level

Selection at the colony level in social Hymenoptera with colonies containing single, once-mated queens is examined under a simple two-allele model. The condition for balanced polymorphism is X> 2V 2 (V + 1) , where V is the fitness of colonies with all workers homozygous and X that of colonies with both heterozygous and homozygous workers, relative to the fitness of colonies with all workers heterozygous. For certain fitness combinations satisfying the above relationship and characterized by values of V and X much lower than one, iteration reveals the development of stable limit cycles of allele frequencies rather than convergence to an equilibrium point. Addition of a third allele, or overlap between generations, eliminates these cycles. Queen-level overdominance is sufficient but not necessary for balanced polymorphism when V < 1, is both sufficient and necessary when V = 1, and is necessary but not sufficient when V > 1. Colony-level selection is a potentially powerful force maintaining genetic variation in populations of social insects, but does not imply correspondence between queen and worker genotype frequencies.