In developing countries, the age at which breastfed infants are first given complementary foods is of public health importance because of the risk of diarrhoeal disease from contaminated weaning foods, and the potential risk of growth faltering if foods are inappropriately delayed. To evaluate whether there are any advantage of complementary feeding prior to 6 months, low-income primiparous mothers who had exclusively breastfed for 4 months were randomly assigned to one of 3 groups: continued exclusive breastfeeding to 6 months (EBF) (n = 50); introduction of complementary foods at 4 months with ad libitum nursing from 4-6 months (SF) (n = 47); and introduction of complementary foods at 4 months, with maintenance of baseline nursing frequency from 4-6 months (SF-M) (n = 44). Baby foods in jars were provided to the SF and SF-M groups from 4 to 6 months. Subjects were visited weekly and provided with lactation guidance; at 4, 5, and 6 months measurements were made of infant intake and breast milk composition. At 4 months, breast milk intake averaged 797 (139) g per day (no difference among groups). Between 4 and 6 months, breast milk intake was unchanged in EBF infants (+6) but decreased in the SF (-103), and SF-M (-62) groups (p < 0.001). Change in total energy intake (including solid foods) and infant weight and length gain did not differ significantly between groups. Weight and length gain from 4-6 months were comparable to those of breastfed infants in an affluent USA population. The results indicate that breastfed infants self-regulate their total energy intake when other foods are introduced. As a result, there is no advantage in introducing complementary foods before 6 months in this population, whereas there may be disadvantages if there is increased exposure to contaminated weaning foods.