Serum bactericidal titers against Staphylococcus aureus were measured in 63 children who were receiving mafcillin or methicillin intravenously, or dicloxacillin, penicillin, or cephalexin orally. The SBTs obtained following unit does of 25 mg/kg of dicloxacillin, 35 mg/kg of penicillin, or 25 mg/kg of cephalexin with probenecid were comparable to those seen following intravenous doses of 40 mg/kg nafcillin or methicillin. Twenty-two children with acute hematogenous osteomyelitis proven or presumed to be due to S. aureus were treated intravenously until point tenderness and fever had resolved, and then with oral therapy. The mean duration of intravenous therapy was 14 days. Oral doses were adjusted so that a peak SBT of greater than or equal to 1:16 and a trough SBT of greater than or equal to 1:2 were obtained in most children. No recurrences occurred. The SBT proved to be a practical means of assessing the adequacy of oral therapy in children with infections due to S. aureus.