There are important differences between on-target military attacks against relatively well protected Armed Forces and nerve agent attacks initiated by terrorists against a civilian population. In contrast to military personnel, civilians are unlikely to be pre-treated with pyridostigmine and protected by personal protective equipment. Furthermore, the time after exposure when specific therapy can first be administered to civilians is likely to be delayed. Even conservative estimates suggest a delay between exposure and the first administration of atropine/oxime of at least 30 minutes. The organophosphorus nerve agents are related chemically to organophosphorus insecticides and have a similar mechanism of toxicity, but a much higher mammalian acute toxicity, particularly via the dermal route. Nerve agents phosphonylate a serine hydroxyl group in the active site of the enzyme, acetylcholinesterase (AChE), which results in accumulation of acetylcholine and, in turn, causes enhancement and prolongation of cholinergic effects and depolarisation blockade. The rate of spontaneous reactivation of AChE is variable, which partly accounts for differences in acute toxicity between the nerve agents. With soman in particular, an additional reaction occurs known as 'aging'. This consists of monodealkylation of the dialkylphosphonyl enzyme, which is then resistant to spontaneous hydrolysis and reactivation by oximes. Monodealkylation occurs to some extent with all dialkylphosphonylated AChE complexes; however, in general, is only of clinical importance in relation to the treatment of soman poisoning, where it is a very serious problem. With soman, aging occurs so fast that no clinically relevant spontaneous reactivation of AChE occurs before aging has taken place. Hence, recovery of function depends on resynthesis of AChE. As a result, it is important that an oxime is administered as soon after soman exposure as possible so that some reactivation of AChE occurs before all the enzyme becomes aged. Even though aging occurs more slowly and reactivation occurs relatively rapidly in the case of nerve agents other than soman, early oxime administration is still clinically important in patients poisoned with these agents. Experimental studies on the treatment of nerve agent poisoning have to be interpreted with caution. Some studies have used prophylactic protocols, whereas the drugs concerned (atropine, oxime, diazepam) would only be given to a civilian population after exposure. The experimental use of pyridostigmine before nerve agent exposure, although rational, is not of relevance in the civilian context. With the possible exception of the treatment of cyclosarin (GF) and soman poisoning, when HI-6 might be preferred, a review of available experimental evidence suggests that there are no clinically important differences between pralidoxime, obidoxime and HI-6 in the treatment of nerve agent poisoning, if studies employing pre-treatment with pyridostigmine are excluded.