Four patients, all possessing an atypical form of plasma cholinesterase, developed prolonged paralysis following succinylcholine administration. The clinical management of all four cases was facilitated by monitoring the train-of-four stimulus. All patients showed marked "fade" of the train-of-four ratio, the initial ratios of the fourth to the first twitches being 50 per cent or less, indicating variable degrees of nondepolarizing neuromuscular blockade. Reversal of paralysis with anticholinesterase agents was completely successful in three cases, but only partially effective in the fourth because of the probable presence of a mixture of both depolarizing block and nondepolarizing block. In such a situation, caution in the interpretation of the train-of-four ratio is necessary, since this test measures only the nondepolarizing component of the block. Whether or not reversal with anticholinesterase drugs is attempted, clinical estimates of neuromuscular function, such as head lift, vital capacity, and inspiratory force, must be carefully correlated with train-of-four values. If reversal is attempted, the brief action of edrophonium provides a useful clinical trial.