On request of the Dutch government a committee of the Health Council of the Netherlands has reviewed the role that results of field research in its broadest sense (i.e., including multi-species toxicity tests in the laboratory, research on model ecosystems et cetera) can play in ecotoxicological risk assessment for the authorisation of pesticides. The Committee believes that field research can provide valuable additional data about the exposure of non-target organisms and the resultant effects at population, community and ecosystem level. However, it frequently is unclear how these data might be used in reaching a decision about authorisation. To solve this problem, it is necessary to specify what is understood by "unacceptable damage". Both more clearly formulated protection goals of the government and a better understanding of the ecological significance of effects are needed to clarify this. Furthermore, the Committee points out that the statistical power of field trials must be sufficient to allow for the detection of changes that might be regarded as ecologically relevant. Finally, it recommends keeping a finger on the pulse in relation to authorised pesticides by monitoring their presence in environmental compartments and by investigating their role in suddenly occurring mortality among conspicuous animal species, such as birds, fish and honeybees. This kind of research forms a safety net for substances that have been wrongly authorised.