Are honeybees suitable surrogates for use in pesticide risk assessment for non‐<i>Apis</i> bees?

Historically, bee regulatory risk assessment for pesticides has centred on the European honeybee (Apis mellifera), primarily due to its availability and adaptability to laboratory conditions. Recently, there have been efforts to develop a battery of laboratory toxicity tests for a range of non-Apis bee species to directly assess the risk to them. However, it is not clear whether the substantial investment associated with the development and implementation of such routine screening will actually improve the level of protection of non-Apis bees. We argue, using published acute toxicity data from a range of bee species and standard regulatory exposure scenarios, that current first-tier honeybee acute risk assessment schemes utilised by regulatory authorities are protective of other bee species and further tests should be conducted only in cases of concern. We propose similar analysis of alternative exposure scenarios (chronic and developmental) once reliable data for non-Apis bees are available to expand our approach to these scenarios. In addition, we propose that in silico (simulation) approaches can then be used to address population-level effects in more field-realistic scenarios. Such an approach could lead to a protective, but also workable, risk assessment for non-Apis species while contributing to pollination security in agricultural landscapes around the globe. © 2019 Society of Chemical Industry.