Clinical trials were conducted to measure the effect of dietary energy on total daily energy intakes by nine hospitalized young children recovering from protein-energy malnutrition. Semi-solid mixed diets containing either 50 or 100 kcal/100 g were offered at fixed intervals five times per day until the children refused further intake. Each diet was offered for seven consecutive days in alternate order. The mean +/- SD amount consumed with the low-density (LD) diet was significantly greater than with the high-density (HD) diet (220 +/- 35 vs 148 +/- 21 g/kg/d, P less than 0.001), but the amount of energy consumed with the HD diet was significantly greater than with the LD diet (148 +/- 21 vs 110 +/- 18 kcal/kg/d, P less than 0.001). There were no differences in total daily intake by day of diet period with the HD diet, but the intakes increased during the first 2-4 days with the LD diet. The maximum intakes at a single meal averaged between 40 and 66 g/kg/d for individual children receiving the LD diet, and were negatively related to the children's lengths and weight-for-length Z scores. We conclude that dietary energy density limited the amount of energy consumed from LD weaning diets offered at fixed frequencies, presumably because the volume of intake required to satisfy the children's energy needs exceeded their gastric capacity.