We evaluated the activity of teicoplanin against a type-III group B streptococcal strain in vitro and in vivo and compared the results with those of penicillin G. In vitro, the minimal inhibitory and minimal bactericidal concentrations of teicoplanin were 2- to 4-fold greater than those of penicillin G. In vivo studies were carried out with an experimental bacteremia and meningitis model in newborn rats. Eighty-one infected animals were randomized to receive teicoplanin 5, 10 or 20 mg/kg, twice daily, or penicillin G 50 or 200 mg/kg, twice daily, or saline (0.05 ml), twice daily. The mean serum levels of teicoplanin were maintained above 100 X the minimal bactericidal concentration for 7-8 h even with a dose of 5 mg/kg. The mean penetration of teicoplanin into the cerebrospinal fluid was estimated as 2.4-8.2% of those of concomitant levels in serum. The overall efficacy of teicoplanin was similar to that of penicillin G as judged by mortality rates. However, two bacteremic animals which were free of meningitis at the beginning of therapy developed this complication during 4 days of teicoplanin therapy, in contrast with none in the penicillin group. Further studies are needed to understand the reason(s) for these failures with teicoplanin therapy.