In the present study, ribostamycin concentrations in serum were measured by microbiological assay in 20 paediatric patients aged 3 months to 11 years after intramuscular ribostamycin 10, 15 and 20 mg/kg. All pharmacokinetic parameters and statistical analyses were calculated by computer. These results showed that the absorption rate constant (ka), elimination rate constant (ke), time to peak serum concentration (tmax), elimination half-life (t½), apparent volume of distribution (Vd/F), total body clearance (CL) and area under the serum concentration-time curve (AUC) were significantly different in infants under 6 months from those in children over 3 years (p < 0.05), except for the peak serum concentration (Cmax) and lag time from administration to the first appearance of drug in the serum (tlag) [p > 0.05]. The absorption of ribostamycin in infants was more rapid than that in children, but elimination was slower (p < 0.05). The Vd/F and CL in infants were also larger than in children (p < 0.01). There were significant positive correlations between Cmax, AUC and ribostamycin dosage (p < 0.01). Pharmacokinetic parameters for infants and children were compared with those observed in adults, and it was found that ribostamycin absorption and elimination were more rapid in the paediatric patients. The Cmaxin children and infants after intramuscular ribostamycin 10 mg/kg approached that in adults after an intramuscular dose of ribostamycin 500mg. Using a 1-compartment open pharmacokinetic model, the optimum intramuscular ribostamycin administration interval was estimated as 6.01 and 7.56h for children and infants, respectively, while the value was 8.5h in adults. When the drug was administered in multiple doses of 15 mg/kg intramuscularly every 8h, no accumulation occurred in children. It is suggested that ribostamycin be administered in intramuscular doses of 10 to 15 mg/kg twice daily in infants and 3 times daily in children, respectively.