Guidelines for testing host specificity of agents for biological control of arthropod pests

Prior to the release of exotic organisms for biological control of pests, their safety and acceptability must first be demonstrated. For weeds, protocols ensure that potential agents are sufficiently host specific to have no detrimental impact on crops, ornamentals or native plants, but for arthropod pests, protocols to assess any development of agents on non-target taxa have not been so widely adopted. Agents for many arthropod pests have been safely released although known to be less specific than most agents for weeds. When testing potential agents, if any feeding or development occurs on beneficial or native organisms, such agents must be assessed to ensure that their establishment will not detrimentally affect abundance or survival of non-target taxa. When development of an agent occurs on non-target taxa, the benefits of biological control of a target pest must be weighed against those potential detrimental effects. The number and range of non-target arthropods to be tested with an exotic agent should be selected carefully since it is impractical to test an extensive list of taxa. The logistics of maintaining a range of arthropods in culture when the life histories of so many are unknown, may limit the number that can reasonably be tested. However, tests on a selection of taxa with relationships to the target species can provide meaningful information necessary to assess potential agents. Guidelines are proposed for assessing the acceptability of agents for biological control of arthropod pests.