Hosted by the College of Agricultural and Environmental Sciences at the University of Georgia
CAP banner

Goals & Objective: Goal 3
Objective 3.2

Obj 3.2: Elucidate lethal and sub-lethal effects of insecticides on non-Apis


Rationale and significance
Many insecticides are toxic to bumble bees, particularly after direct spray or through exposure to treated foliage (Thomson 2001). However, there is a void in toxicity tables for non-Apis, particularly for newer chemistries. Pesticide toxicities determined for honey bees are not always predictive of toxicities to other bees NRC (2007). There is debate on the extent to which neonicotinoids accumulate in pollen and nectar and impact bee health. In the case of non-Apis bees, the results are mixed, with some showing no effect (Franklin et al. 2004), and others showing sub-lethal impacts on Bombus activity (Gels et al. 2002), behavior, and learning (Morandin and Winston 2003). There is need to expand this knowledge base with the most commonly used insecticides and most commonly cultured non-Apis bees.

Expected outcomes:

To gain:

  1. new information on basic toxicology of some of our most common non-Apis managed pollinators and
  2. new information on sub-lethal effects of neonicotinoids on non-Apis bees

Summary Statement for Goal 3
The non-Apis CAP group has focused on the incidence of cross-over infections of Apis mellifera pathogens into neighboring populations of wild bees. At two farm sites stocked with either Apis or Bombus, the honey bee viruses DWV, BQCV and SBV detections were common in non-Apis bees. In contrast, target viruses were absent at an isolated site with no stocked Apidae. These data suggest that common and prevalent honey bee viruses are shared across multiple native bee genera. Whether these native bees serve as biological hosts or have an impact on populations is unknown. Concerning bumble bees, of 34 individuals caught at one site, the majority was positive for DWV and SBV and at a second site most were positive for BQCV. Because commercial bumble bees are reared on pollen collected from honey bee colonies, they could have picked up viruses from contaminated pollen. At the two non-farm, isolated sites, no BQCV or SBV were detected and only 6% of samples were positive for DWV. The data suggest that common and prevalent honey bee viruses may be shared across many Bombus species.

The group hypothesized that there would be different virus genotypes infecting different bee species and that these differences would be in the capsid genes (structural proteins). A phylogenetic analysis was conducted of DWV virus and representative bee taxa: 12 Apis
, 42 Bombus and 12 non Apis non Bombus bees. The sequences from these bees were all greater than 98% identical to DWV RdRp sequences in the data base indicating that all of these bees were infected with DWV. Some Bombus isolates and some non-Apis non Bombus isolates clustered with DWV from Apis. Some Apis isolates, some Bombus isolates and some non-Apis non-Bombus isolates clustered in a less defined group outside the Apis DWV cluster. Most interestingly, there was one genotype found only in Bombus. This isolate was found in slightly more than half of the Bombus samples that were taken on two different collection dates. These data suggest that one genotype is favored in Bombus and that as the virus becomes established in a bumble bee population, a certain genotype is favored and predominates.

Acute dermal LD50 tests done in year 2 showed unexplained variation in mortality among bees when they were exposed to topical applications of imidacloprid, even when body size was accounted for. We tested a hypothesis that varying Nosema load across the bees could render some more vulnerable to doses of the neonicotinoid. Our data showed an insufficient differences to explain observed variation in mortality.

We established in year 2 that the imidacloprid LD50 for B. impatiens (Koppert origin) was 13 ng/ bee. In a homing experiment, individuals were marked and treated with either the sublethal dose of 5 ng of technical grade imidacloprid in acetone or acetone alone and all were released 0.5 km from their hive’s location. Work to date has not resolved geographic differences between sites and will be repeated another year.



Methodology, data and analysis of results to date are shared in an annual report to USDA. Papers generated by team members during the time of the CAP are listed and periodically updated within each objective. Beyond the citation of published papers, the consensus of the group is that it would otherwise be unhelpful or possibly misleading to state preliminary results within the context on this web site.

Publications of objective 3.2 principal investigator (Averill) to date during the CAP

Morkeski, A. and A.L. Averill. 2010. “Wild bee status and evidence for pathogen spillover with honey bees.” American Bee Journal, 150(11):1049-1052

Further Background Information
Documentation of CAP progress in general, and of this objective in particular, is available through the following sources:

  1. Bee Health, an eXention initiative for peer-reviewed scientific recommendations
  2. Colony Collapse Disorder Progress Report for 2009
  3. When Varroacides Interact
  4. Honey Bee “Medical Records”: The Stationary Apiary Monitoring Project
  5. Pesticides Applied to Crops and Honey Bee Toxicity
  6. Sustainable Beekeeping
  7. Wild Bee Status and Evidence for Pathogen 'Spillover' with Honey Bees 
  8. Assessing the Risks of Honey Bee Exposure to Pesticides
  9. Pesticides and their involvement in Colony Collapse Disorder

Updated August 19, 2011.