We compared both the time course of neuromuscular blockade and the cardiovascular side-effects of suxamethonium and mivacurium during halothane and nitrous oxide anaesthesia in infants 2-12 months and children 1-12 years of age. Equipotent doses of mivacurium and suxamethonium were studied; 2.2 X ED(95) was used in four groups of infants and children, while 3.4 X ED(95) was used in two groups of children. Onset of neuromuscular block in infants was not significantly faster with suxamethonium than with mivacurium (P = 0.2). In all infants given suxamethonium, intubating conditions were excellent, while, in 6/10 infants given mivacurium, intubating conditions were excellent. Onset of complete neuromuscular block in children was significantly faster with suxamethonium, 0.9 min compared with mivacurium, 1.4 min (P less than or equal to 0.05). Increasing the dose of suxamethonium or mivacurium in children to 3.4 X ED(95) did not change the onset of neuromuscular block. Recovery of neuromuscular transmission to 25% of initial twitch height (T-25) in infants and children was significantly faster after suxamethonium than after mivacurium, at 2.5 and 6 min, respectively (P less than or equal to 0.05). In children given 3.4 X ED(95) of suxamethonium or mivacurium, recovery from neuromuscular block was almost identical with the dose of 2.2 X ED(95), with spontaneous recovery to T-25 prolonged by only 0.5 min. No infant or child had hypotension after the mivacurium bolus dose.