Identification of honey bee (Apis mellifera) subspecies is important for their protection. It is also used by queen breeders to maintain some breeding lines. In this study, we compared three methods of subspecies identification based on the following: 17 microsatellite loci, COI-COII mitotypes and geometric morphometrics of forewing venation. The methods were used to classify colonies and workers from a mixed population of A. m. mellifera and A. m. carnica. There was highly significant correlation between results obtained using the three methods. More than three quarters of colonies were classified to the same subspecies by all three methods. The agreement was highest between microsatellites and morphometrics. More than 90 % of colonies were classified to the same subspecies by the two methods. There was also relatively high agreement (75 %) between microsatellites and morphometrics when workers were classified as pure subspecies or hybrids. In particular, one pure subspecies was never misclassified as other pure subspecies. The results presented here show that morphometrics can be used for detection of hybrids between A. m. mellifera and A. m. carnica.

Several honey bee (Apis mellifera) subspecies are in danger of local extinction because their feral population have almost completely disappeared. An important threat to the feral populations of bees is loss of habitat and loss of woodlands. In many places the only habitat suitable for honey bee nesting are rows of trees along roadsides. We studied a feral population of honey bees inhabiting avenues in northern Poland. We inspected 142 km of avenues and found 45 feral colonies. The estimated density of feral population inhabiting the avenues was 0.10 nest km⁻². Honey bees preferred to build their nests in trees with a thick trunk and a somewhat weak state of health. There was no strong preference of bees to any species of trees. We stress the importance of protection of existing avenues and creating new ones. This can provide suitable habitat not only for honey bees but also for other endangered species.

The genetic diversity of the north and western European subspecies of honey bee, Apis mellifera mellifera (the “dark bee”) is severely endangered due to hybridization with introduced bees of evolutionary branch C.

Species confined to temporally stable habitats are usually susceptible to habitat fragmentation, as living in long-lasting habitats is predicted to constrain evolution of dispersal ability. In Europe, saproxylic invertebrates associated with tree hollows are currently threatened due to the severe fragmentation of their habitat, but data on the population genetic consequences of such habitat decline are still scarce. By employing AFLP markers, we compared the spatial genetic structure of two ecologically and taxonomically related beetle species, Osmoderma barnabita and Protaetia marmorata (Cetoniidae). Both species are exclusively associated with tree hollows, but O. barnabita has a more restricted host preferences compared to P. marmorata. Analyses of spatial autocorrelation showed, in line with the predicted low dispersal potential of these saproxylic beetles, that both species are characterized by a strong kinship structure, which was more pronounced in the specialist O. barnabita than in the generalist P. marmorata. Individuals of both species sampled within single trees showed high relatedness (≈0.50 in O. barnabita and ≈0.15 in P. marmorata). Interestingly, groups of pheromone-emitting O. barnabita males sampled on the same tree trunk were found to be full brothers. Whether this result can be explained by kin selection to increase attraction of conspecific females for mating or by severe inbreeding of beetles within individual tree hollows needs further study. Although our studied populations were significantly inbred, our results suggest that the dispersal ability of Osmoderma beetles may be one order of magnitude greater than suggested by previous dispersal studies and acceptable levels of habitat fragmentation for metapopulation survival may be bigger than previously thought.

Carniolan honey bees (Apis mellifera carnica) are considered as an indigenous subspecies in Hungary adapted to most of the ecological and climatic conditions in this area. However, during the last decades Hungarian beekeepers have recognized morphological signs of the Italian honey bee (Apis mellifera ligustica). As the natural distribution of the honey bee subspecies can be affected by the importation of honey bee queens or by natural gene flow, we aimed at determining the genetic structure and characteristics of the local honey bee population using molecular markers. All together, 48 Hungarian and 84 foreign (Italian, Polish, Spanish, Liberian) pupae and/or workers were used for mitochondrial DNA analysis. Additionally, 53 sequences corresponding to 10 subspecies and the Buckfast hybrid were downloaded from GenBank. For the nuclear analysis, 236 Hungarian and 106 foreign honey bees were genotyped using nine microsatellites. Heterozygosity values, population‐specific alleles, FST values, principal coordinate analysis, assignment tests, structure analysis, and dendrograms were calculated. Haplotype and nucleotide diversity values showed moderate values. We found that one haplotype (H9) was dominant in Hungary. The presence of the black honey bee (Apis mellifera mellifera) was negligible, but a few individuals resembling other subspecies were identified. We proved that the Hungarian honey bee population is nearly homogeneous but also demonstrated introgression from the foreign subspecies. Both mitochondrial DNA and microsatellite analyses corroborated the observations of the beekeepers. Molecular analyses suggested that Carniolan honey bee in Hungary is slightly affected by Italian and black honey bee introgression. Genetic differences were detected between Polish and Hungarian Carniolan honey bee populations, suggesting the existence of at least two different gene pools within A. m. carnica.