Nikola Kezić

Estimating the Density of Honeybee Colonies across Their Natural Range to Fill the Gap in Pollinator Decline Censuses

Although pollinator declines are a global biodiversity threat, the demography of the western honeybee (Apis mellifera) has not been considered by conservationists because it is biased by the activity of beekeepers. To fill this gap in pollinator decline censuses and to provide a broad picture of the current status of honeybees across their natural range, we used microsatellite genetic markers to estimate colony densities and genetic diversity at different locations in Europe, Africa, and central Asia that had different patterns of land use. Genetic diversity and colony densities were highest in South Africa and lowest in Northern Europe and were correlated with mean annual temperature. Confounding factors not related to climate, however, are also likely to influence genetic diversity and colony densities in honeybee populations. Land use showed a significantly negative influence over genetic diversity and the density of honeybee colonies over all sampling locations. In Europe honeybees sampled in nature reserves had genetic diversity and colony densities similar to those sampled in agricultural landscapes, which suggests that the former are not wild but may have come from managed hives. Other results also support this idea: putative wild bees were rare in our European samples, and the mean estimated density of honeybee colonies on the continent closely resembled the reported mean number of managed hives. Current densities of European honeybee populations are in the same range as those found in the adverse climatic conditions of the Kalahari and Saharan deserts, which suggests that beekeeping activities do not compensate for the loss of wild colonies. Our findings highlight the importance of reconsidering the conservation status of honeybees in Europe and of regarding beekeeping not only as a profitable business for producing honey, but also as an essential component of biodiversity conservation.

Standard methods for rearing and selection of Apis mellifera queens

Summary Here we cover a wide range of methods currently in use and recommended in modern queen rearing, selection and breeding. The recommendations are meant to equally serve as standards for both scientific and practical beekeeping purposes. The basic conditions and different management techniques for queen rearing are described, including recommendations for suitable technical equipment. As the success of breeding programmes strongly depends on the selective mating of queens, a subchapter is dedicated to the management and quality control of mating stations. Recommendations for the handling and quality control of queens complete the queen rearing section. The improvement of colony traits usually depends on a comparative testing of colonies. Standardized recommendations for the organization of performance tests and the measurement of the most common selection characters are presented. Statistical methods and data preconditions for the estimation of breeding values which integrate pedigree and performance data from as many colonies as possible are described as the most efficient selection method for large populations. Alternative breeding programmes for small populations or certain scientific questions are briefly mentioned, including also an overview of the young and fast developing field of molecular selection tools. Because the subject of queen rearing and selection is too large to be covered within this paper, plenty of references are given to facilitate comprehensive studies.

A review of methods for discrimination of honey bee populations as applied to European beekeeping

Summary Here, scientists from 19 European countries, most of them collaborating in Working Group 4: “Diversity and Vitality” of COST Action FA 0803 “Prevention of honey bee COlony LOSSes” (COLOSS), review the methodology applied in each country for discriminating between honey bee populations. Morphometric analyses (classical and geometric) and different molecular markers have been applied. Even if the approach has been similar, however, different methodologies regarding measurements, landmarks or molecular markers may have been used, as well as different statistical procedures. There is therefore the necessity to establish common methods in all countries in order to have results that can be directly compared. This is one of the goals of WG4 of the COLOSS project.

The influence of genetic origin and its interaction with environmental effects on the survival of Apis mellifera L. colonies in Europe

Summary The survival and performance of 597 honey bee colonies, representing five subspecies and 16 different genotypes, were comparatively studied in 20 apiaries across Europe. Started in October 2009, 15.7% of the colonies survived without any therapeutic treatment against diseases until spring 2012. The survival duration was strongly affected by environmental factors (apiary effects) and, to a lesser degree, by the genotypes and origin of queens. Varroa was identified as a main cause of losses (38.4%), followed by queen problems (16.9%) and Nosema infection (7.3%). On average, colonies with queens from local origin survived 83 days longer compared to non-local origins (p < 0.001). This result demonstrates strong genotype by environment interactions. Consequently, the conservation of bee diversity and the support of local breeding activities must be prioritised in order to prevent colony losses, to optimize a sustainable productivity and to enable a continuous adaptation to environmental changes.

Occurrence of parasites and pathogens in honey bee colonies used in a European genotype-environment interactions experiment

Summary Diseases are known to be one of the major contributors to colony losses. Within a Europe-wide experiment on genotype—environment interactions, an initial 621 colonies were set up and maintained from 2009 to 2012. The colonies were monitored to investigate the occurrence and levels of key pathogens. These included the mite Varroa destructor (mites per 10 g bees), Nosema spp. (spore loads and species determination), and viruses (presence/absence of acute bee paralysis virus (ABPV) and deformed wing virus (DWV)). Data from 2010 to the spring of 2011 are analysed in relation to the parameters: genotype, environment, and origin (local vs. non-local) of the colonies in the experiment. The relative importance of different pathogens as indicators of colony death within the experiment is compared. In addition, pathogen occurrence rates across the geographic locations are described.

Population dynamics of European honey bee genotypes under different environmental conditions

Summary Adaptation of honey bees to their environment is expressed by the annual development pattern of the colony, the balance with food sources and the host—parasite balance, all of which interact among each other with changes in the environment. In the present study, we analyse the development patterns over a period of two years in colonies belonging to 16 different genotypes and placed in areas grouped within six environmental clusters across Europe. The colonies were maintained with no chemical treatment against varroa mites. The aim of the study was to investigate the presence of genotype—environment interactions and their effects on colony development, which we use in this study as a measure of their vitality. We found that colonies placed in Southern Europe tend to have lower adult bee populations compared to colonies placed in colder conditions, while the brood population tends to be smaller in the North, thus reflecting the shorter longevity of bees in warmer climates and the shorter brood rearing period in the North. We found that both genotype and environment significantly affect colony development, and that specific adaptations exist, especially in terms of adult bee population and overwintering ability.

A review of methods used in some European countries for assessing the quality of honey bee queens through their physical characters and the performance of their colonies

Summary The term “quality” in relation to queens and drones refers to certain quantitative physical and/or behavioural characters. It is generally believed that a high quality queen should have the following physical characteristics: high live weight; high number of ovarioles; large size of spermatheca; high number of spermatozoa in spermatheca; and be free from diseases and pests. It is, however, also known that the performance of a honey bee colony is the result of its queens function as well as of that of the drones that mated with her. These two approaches are often considered together and give a general picture of the queen production technique and selection. Here we describe the most common and well known anatomical, physiological, behavioural and performance characters related to the queens, as measured in different European countries: the live weight of the virgin queen (Bulgaria); the live weight of the laying queen (Bulgaria, Italy); the diameter and volume of spermatheca (Bulgaria, Greece, Slovenia); the number of ovarioles (Greece, Italy, Slovenia); the weight of ovaries (Slovenia); the number of spermatozoa in spermatheca (Italy, Poland, Slovenia); the brood pattern (Bulgaria, Greece); the egg laying ability/fecundity (Bulgaria); the brood production (Croatia, Serbia); the colony population development (Croatia, Serbia, Slovakia); the honey production (Croatia, Denmark, Serbia, Slovakia); the hygienic behaviour (Croatia, Denmark, Serbia, Slovakia); the defence behaviour (Croatia); the calmness/sitting on the comb (Croatia, Denmark); and swarming (Croatia, Denmark). The data presented fit well with the findings of the same characters in the literature, and in general they support the argument for the term “quality characters”. Especially for the weight of the queen, the number of ovarioles, the volume of the spermatheca and the number of spermatozoa, data per country proved its own accuracy by repetition through the years. We also report that when instrumentally inseminated queens are kept under mass production conditions (in small cages in queen banks and with low number of attendants) they can transfer the semen to their spermatheca and clear their oviducts more efficiently when they have been inseminated with small amounts of semen in two or three sequences (but not four), compared to those inseminated with the same amount of semen at once (Poland). Furthermore, we had an inside view of the sanitary conditions of the colony: a. through the health status of the queen (nosema plus virus analysis) (Slovenia); and b. evaluating the nosema load of worker bees (Denmark) and of the queens (Greece). This is the first step to summarize this type of diverse data for such an important issue. The knowledge acquired can be used to fill in the existing gaps in the breeding or queen evaluation systems of each country in order to facilitate standardization of methodology for comparable results.

The genetic origin of honey bee colonies used in the COLOSS Genotype-Environment Interactions Experiment: a comparison of methods

Summary The COLOSS GEI (Genotype-Environment Interactions) Experiment was setup to further our understanding of recent honey bee colony losses. The main objective of the GEI experiment was to understand the effects of environmental factors on the vitality of European honey bee genotypes. This paper aims to describe the genetic background and population allocation of the bees used in this experiment. Two wing morphometric and two genetic methods were employed to discriminate bee populations. Classical morphometry of 11 angles on the wings were carried out on 350 bees. Geometric morphometry on 19 wing landmarks was carried out on 381 individuals. DNA microsatellite analysis was carried out on 315 individuals using 24 loci. Allozyme analysis was performed on 90 individuals using six enzyme systems. DNA microsatellite markers produced the best discrimination between the subspecies (Apis mellifera carnica, A. m. ligustica, A. m. macedonica, A. m. mellifera and A. m. siciliana) used in the experiment. Morphometric methods generally showed an intermediate level of discrimination, usually best separating A. m. siciliana and A. m. ligustica from the remaining populations. Allozyme markers lack power to discriminate at the level of individual bees, and given our sample size, also fail to differentiate subspecies. Based on DNA microsatellites, about 69% of the individuals were assigned to the same subspecies as originally declared, and 17% were found to belong to a different subspecies. Fourteen percent of the samples were found to be of mixed origin and could not be assigned to any subspecies with certainty. We further discuss the caveats of the methods and details of the sampled bees, their origins and breeding programmes in their respective locations.

Detecting population admixture in honey bees of Serbia

Summary Honey bee workers were sampled across the Serbian territory during 2009–2010 from mostly non-migratory apiaries to determine the population structure of these bees using morphometric, genetic, and spatial information. A total of 134 bees were sampled, of which 77 were analysed using classical wing morphometrics and 122 bees were successfully analysed using 24 DNA microsatellite markers. A combination of methods including multivariate statistics and assignment tests (frequency-based and Bayesian) revealed the honey bees of this region to resemble the subspecies Apis mellifera macedonica, Apis mellifera carnica or hybrids of these two subspecies. Based on Bayesian assignment (‘Structure’) and spatial PCA, honey bees within the Serbian territory were composed of 56%-58% A. m. carnica and 42%-44% A. m. macedonica. Spatial analysis showed the existence of a north-west to south-east cline in genetic differentiation. The bees in the north-west resemble A. m. carnica,while the bees in the south-east of the country are more similar to A. m. macedonica. Thus, the extent of A. m. macedonica within Serbia was greater than previously estimated. We define a line of hybridisation between A. m. carnica and A. m. macedonica within our study area. The cline of differentiation was still evident using a combination of genetic and spatial information, in spite of beekeeping activities including transhumance and breeding efforts.

Effect of genotype and environment on parasite and pathogen levels in one apiary - a case study

As part of the COLOSS GEI experiment, one apiary in Chalkidi, Greece was continuously monitored for various pests and pathogens including V. destructor mites, Nosema spp. spores, and quantitative titers of five viruses from late summer 2009 until March 2012. The apiary was established with 39 colonies which included ten A. m. carnica (CarV) colonies from Germany, ten A. m. ligustica (LigI) colonies from Italy, ten A. m. macedonica (MacB) colonies from Bulgaria and nine local A. m. macedonica (MacG) colonies from Greece. At the end of the monitoring period, eight colonies survived: six local MacG, one MacB and one LigI. The local MacG colonies consistently showed comparatively lower V. destructor infestation levels and, consequently, also low DWV titres. In contrast, LigI colonies had the highest Nosema spp. and BQCV titres. It is, however, difficult to attribute the low survival rate of non-local colonies to any single factor. Since all colonies were located in close proximity in a single apiary, and the local bees were in better health than non-local bees, we assume that the local bees were better adapted to the local environmental conditions, and handled environmental stressors better to avoid disease outbreak.