Objective: To evaluate the impact of a dietary counseling in reducing the intake of energy-dense foods by infants. Design: A randomized controlled trial. Setting and Participants: Sao Leopoldo, Brazil. Mothers and infants of a low-income-group population were randomized into intervention (n = 163) and received dietary counseling during 10 home visits, or control (n = 234) groups. Main Outcomes Measures: Child consumption of sugar-dense (SD) and lipid-dense (LD) foods at 12 to 16 months. Analysis: The effect of the intervention was expressed by relative risks and 95% confidence intervals. Poisson regression analysis was used to determine the association between exclusive breastfeeding and the energy-dense foods intake. Results: A smaller proportion of infants from the intervention group consumed candy, soft drinks, honey, cookies, chocolate, and salty snacks. In the intervention group, there was a reduction of 40% and 50% in the proportion of infants who consumed LD and SD foods, respectively. Being breastfed up to 6 months reduced the risk for consumption of LD and SD foods by 58% and 67%, respectively. Conclusions and Implications: Dietary counseling to mothers may be effective in reducing the consumption of energy-dense foods among infants, and it is helpful in improving early dietary habits. (Contains 3 tables and 1 figure.)