Fipronil is a highly effective, broad-spectrum insecticide with potential value for the control of a wide range of crop, public hygiene, amenity, and veterinary pests. It can generally be applied at low to very low dose rates to achieve effective pest control. Application rates vary between 0.6 and 200 g a.i./ha, depending on the target pest and formulation. It belongs to the phenyl pyrazole or fiprole group of chemicals and is a potent disrupter of the insect central nervous system via interference with the gamma-aminobutyric acid (GABA-) regulated chloride channel. Fipronil degrades slowly on vegetation and relatively slowly in soil and in water, with a half-life ranging between 36 hr and 7.3 mon depending on substrate and conditions. It is relatively immobile in soil and has low potential to leach into groundwater. One of its main degradation products, fipronil desulfinyl, is generally more toxic than the parent compound and is very persistent. There is evidence that fipronil and some of its degradates may bioaccumulate, particularly in fish. Further investigation on bioaccumulation is warranted, especially for the desulfinyl degradate. The suitability of fipronil for use in IPM must be evaluated on a case-by-case basis. In certain situations, fipronil may disrupt natural enemy populations, depending on the groups and species involved and the timing of application. The indications are that fipronil may be incompatible with locust IPM; hence, this possibility requires further urgent investigation. It is very highly toxic to termites and has severe and long-lasting negative impacts on termite populations. It thus presents a long-term risk to nutrient cycling and soil fertility where termites are "beneficial" key species in these ecological processes. Its toxicity to termites also increases the risk to the ecology of habitats in which termites are a dominant group, due to their importance as a food source to many higher animals. This risk has been demonstrated in Madagascar, where two endemic species of lizard and an endemic mammal decline in abundance because of their food chain link to termites. Fipronil is highly toxic to bees (LD50 = 0.004 microgram/bee), lizards [LD50 for Acanthodactylus dumerili (Lacertidae) is 30 micrograms a.i./g bw], and gallinaceous birds (LD50 = 11.3 mg/kg for Northern bobwhite quail), but shows low toxicity to waterfowl (LD50 > 2150 mg/kg for mallard duck). It is moderately toxic to laboratory mammals by oral exposure (LD50 = 97 mg/kg for rats; LD50 = 91 mg/kg for mice). Technical fipronil is in toxicity categories II and III, depending on route of administration, and is classed as a nonsensitizer. There are indications of carcinogenic action in rats at 300 ppm, but it is not carcinogenic to female mice at doses of 30 ppm. The acute toxicity of fipronil varies widely even in animals within the same taxonomic groups. Thus, toxicological findings from results on standard test animals are not necessarily applicable to animals in the wild. Testing on local species seems particularly important in determining the suitability of fipronil-based products for registration in different countries or habitats and the potential associated risk to nontarget wildlife. Risk assessment predictions have shown that some fipronil formulations present a risk to endangered bird, fish, and aquatic and marine invertebrates. Great care should thus be taken in using these formulations where they may impact any of these endangered wildlife groups. Work in Madagascar has highlighted field evidence of this risk. The dose levels at which fipronil produces thyroid cancer in rats are very high and are unlikely to occur under normal conditions of use. There is also dispute as to whether this is relevant to human health risk. However, as fipronil is a relatively new insecticide that has not been in use for long enough to evaluate the risk it may pose to human health, from data on human exposure to the product, a precautionary approach may be warranted. The use of some fipronil-based products on domestic animals is not recommended where handlers spend significant amounts of time grooming or handling treated animals. In general, it would appear unwise to use fipronil-based insecticides without accompanying environmental and human health monitoring, in situations, regions, or countries where it has not been used before, and where its use may lead to its introduction into the wider environment or bring it into contact with people. Further work is needed on the impacts of fipronil on nontarget vertebrate fauna (amphibians, reptiles, birds, and mammals) in the field before the risk to wildlife from this insecticide can be adequately validated. Further field study of the effects of fipronil on the nutrient cycling and soil water-infiltration activities of beneficial termites is required to assess the ecological impacts of the known toxicity of fipronil to these insects.