A 25-year-old Chinese woman, was found HIV antibody positive on December 14, 1988. During our follow-up, we tested her American husband and found him to be seronegative. Unfortunately, her six-month-old infant was seropositive. The standard Western blot test was used in the first stage of analysis. The bands which appeared on the infant's strip were p15, p24, p31, p55 and gp120/gp160, but using the modified Western blot test the bands which appeared were p15, p24, p31, gp41, p53, p55, p64 and gp120/gp160. All the bands appearing on the infant's strips which used a modified Western blot test had higher intensities than those of a standard procedure. The mother was apparently infected with HIV through intercourse with her ex-boyfriend, who was a European. AZT was given to the mother because her T4 cell count was 338 per microliter and because of persistent cervical lymphadenopathy. The infant, which was bottle-fed and had been delivered by caesarean section, may have become HIV infected during the uterine stage.