The results are reported of a multicenter randomized study of the effectiveness of maternal administration of betamethasone versus ambroxol, a substance of the group of the benzylamines, for prevention of RDS in preterm infants. Women of 27 to 34 weeks gestation with threatened premature delivery or planned premature delivery were admitted to the trial. Between September 1981 and November 1984 a total of 288 randomized patients delivered 315 neonates. The incidence of RDS was assessed in 169 viable neonates born before the 37th week. Of these 86 were born of 76 mothers treated with beta-methasone and 83 of 76 mothers treated with ambroxol. The overall incidence of RDS was significantly (P less than 0.05) higher in the betamethasone group (31%) than the ambroxol group (13%). Ambroxol was significantly more effective than betamethasone in twin births, in infants born before the 31st week, when ROM to delivery time was more than 48 hours, when treatment to delivery time was between 2 and 7 days and in female infants. The neonatal infection rate was significantly higher (P less than 0.05) in the group of betamethasone treated infants (18% with four fatalities) than in the group of ambroxol treated infants (9% with one fatality). These results suggest that ambroxol may be a valid alternative to steroids for prevention of RDS.