Interactions between stressors are involved in the decline of wild species and losses of managed ones. Those interactions are often assumed to be synergistic, and per se of the same nature, even though susceptibility can vary within a single species. However, empirical measures of interaction effects across levels of susceptibility remain scarce. Here, we show clear evidence for extreme differences in stressor interactions ranging from antagonism to synergism within honeybees, Apis mellifera. While female honeybee workers exposed to both malnutrition and the pathogen Nosema ceranae showed synergistic interactions and increased stress, male drones showed antagonistic interactions and decreased stress. Most likely sex and division of labour in the social insects underlie these findings. It appears inevitable to empirically test the actual nature of stressor interactions across a range of susceptibility factors within a single species, before drawing general conclusions.

The ubiquitous use of pesticides is one major driver for the current loss of biodiversity, and the common practice of simultaneously applying multiple agrochemicals may further contribute. Insect toxicology currently has a strong focus on survival to determine the potential hazards of a chemical routinely used in risk evaluations. However, studies revealing no effect on survival or even indicating enhanced survival are likely to be misleading, if potential trade-offs between survival and other physiological factors are overlooked. Here, we used standard laboratory experiments to investigate the sublethal (i.e., food consumption) and lethal (i.e., survival) effects of two common agricultural pesticides (Roundup^® and clothianidin) on adult female solitary bees, Osmia bicornis. The data showed no significant effect of the treatment on cumulative survival; however, a significant positive correlation between herbicide and insecticide exposure and age was revealed, i.e., bees exposed to higher dosages lived longer. As no significant differences in daily food consumption were observed across treatment groups, increased food intake can be excluded as a factor leading to the prolonged survival. While this study does not provide data on fitness effects, two previous studies using solitary bees observed significant negative effects of neonicotinoid insecticides on fitness, yet not on survival. Thus, we conjecture that the observed non-significant effects on longevity may result from a trade-off between survival and reproduction. The data suggest that a focus on survival can lead to false-negative results and it appears inevitable to include fitness or at least tokens of fitness at the earliest stage in future risk assessments.