We conducted a double-blind, placebo-controlled, within-subject crossover study to investigate the effects of covert substitution of olestra, a non-energy-containing fat replacer, for conventional fat on food selection and energy intake in lean and obese men and women. Fifty-one subjects [BMI (kg/m2): 19-36; age: 25-63 y] were studied during two 14-d treatment periods (olestra and placebo), with a 7-d washout between feeding periods. During the intervention periods all foods were provided to the subjects. The aim was to produce a 10% dilution of total energy intake by replacing conventional triacylglycerol with olestra. To accomplish this, subjects were required to consume core foods providing 20-35 g olestra (depending on estimated energy needs) or the same foods containing placebo triacylglycerol. Additional items could be selected from foods that varied in macronutrient composition. When the two treatment periods were compared, total energy intake was 8% lower and fat intake 11% lower during the olestra period than during the placebo treatment period (P < 0.0001). Overall, subjects compensated for 15% of the fat and 20% of the total energy replaced by olestra. In absolute terms, subjects consumed 32% of total energy from fat during the placebo period and 27% of total energy from fat during the olestra period. Neither carbohydrate nor protein intake (g/d) differed between periods. The results did not differ as a function of BMI (lean compared with obese) or sex. Over a 2-wk period, covert substitution of olestra for conventional fat led to reductions in dietary fat intake and total energy intake in all subjects.