Between 1977 and 1980, 1442 pregnant women in Thies, Senegal, were tested for serologic markers of hepatitis B virus (HBV) infection. Of these, 9.8% were HBsAg(+), 59.9% were anti-HBs(+), and 15.6% had anti-HBc alone. Of 116 HBsAg(+) pregnant women, only 19.8% were HBeAg(+), a much lower proportion of infectious carriers than seen in Asian populations. Cord blood from 1353 babies was HBsAg(-), implying that the babies were not infected prior to birth. Four hundred sixty-two babies, including 88 born to HBsAg(+) mothers, were observed for 2 weeks to 38 months after birth. In contrast to observations in Asia, none of the babies became HBsAg(+) before 5 months of age, and only three of the 16 born to HBeAg(+) mothers became HBsAg(+) within the first year of life; all three developed chronic infections (i.e., HBsAg(+) for greater than or equal to 6 months. In the second year of life, six of 34 babies born to HBsAg(+), HBeAg(-)/anti-HBe(-) mothers became infected with HBV, and four of the six developed chronic infections. During the first 3 years of life, infections occurred at a higher rate in infants born to HBsAg(+) (17%) than to HBsAg(-) (4%) women. The latter group of infants included 4.0% of those born to anti-HBs(+) mothers, 4.6% born to anti-HBcAg(+), and 3.2% born to uninfected women. These observations indicate that HBV infections in Senegal usually do not occur perinatally, but do occur at high incidence later in infancy and childhood. Such infections can be prevented by the use of hepatitis B vaccine alone; administration of hepatitis B immune globulin should not be needed.