One hundred eleven newborn infants born of Spanish hepatitis B surface antigen (HBsAg) carrier mothers were consecutively assigned to one of three treatment groups. Group A was treated with three or four doses of hepatitis B immune globulin (HBIG) in one of three different schedules. Group B received one dose of hepatitis B vaccine (Hevac-B, Pasteur) at birth and at 1, 2 and 12 months. Group C was treated with the same vaccination schedule as group B and in addition received a single dose of HBIG at birth. Comparisons were made in the 85 babies who had strictly completed the immunization schedule and had been followed for at least 12 months. The three immunization protocols were equally effective, since none of the children became a chronic HBsAg carrier or developed acute symptomatic infection. There were five transient and subclinical infections among children who received only HBIG (group A), one transient infection in group B, and one in group C. There seems to be some correlation between anti-HBs levels and degree of protection, since all transient infections in group A occurred in the subgroups who did not maintain protective antibodies during the first 6 months. Although the percentage of responders in the two vaccinated groups did not differ significantly, children who received only vaccine reached higher antibody levels than those who also received HBIG. Our results suggest that any immunization schedule able to maintain anti-HBs levels during the first 6 months of life would be useful to prevent mother-to-infant transmission of the hepatitis B virus in areas where most of the carrier women are expected to be anti-HBe positive and hence relatively less infectious.