Field bioassays were conducted to assess the toxicity of three insecticides, chlorpyrifos, cypermethrin and pirimicarb, to four species of springtails, Isotoma viridis, Isotomurus palustris, Folsomia candida (Collembola: Isotomidae) and Sminthurus viridis (Collembola: Sminthuridae). Spray residues on two soil types (a sandy clay loam and a sandy soil) were obtained in the field, in the presence and absence of a wheat crop canopy, after spray application by a commercial tractor-mounted sprayer. Collembola were then confined for 24-h periods on the sprayed soils in a constant laboratory environment at 1, 2, 3, 8 and 15 days after treatment. Residual insecticide toxicity was compared between species, insecticides, soils and exposure conditions (crop or no crop) using the age of residue at which median mortality occurred (DAT,,). Cypermethrin and pirimicarb residues were of low toxicity, causing less than 10% mortality, whereas residues of chloryprifos were toxic to all four species of Collembola on both soil types and in both exposure treatments. Interspecific differences in collembolan susceptibility to chlorpyrifos residues gave the ranking (from most to least susceptible) S. viridis > F. candida > Isotomurus palustris > Isotoma viridis. Residues on the sandy soil were more toxic than those on the sandy clay loam. These results are discussed in terms of how field bioassay approaches may be used to determine pesticide residual toxicity to microarthropods. We conclude that field bioassays offer a feasible method for evaluating the toxicity of pesticides and the persistence of toxic effects on Collembola. Advantages and disadvantages of this method are considered.