To determine the energy and macronutrient consumption of elementary school children served modified lower fat and sodium or typical higher fat and sodium school lunches. The intervention school received lunches with < 30% of energy from fat and < 1000 mg sodium. The control school received typical lunches with 35% of energy from fat and > 1000 mg sodium. Served lunches were analyzed from menus and consumed lunches were analyzed using USDA plate waste methodology. Eighty intervention and 80 control subjects matched for ethnicity and economic status. Intervention school menus were significantly lower for energy, sodium, and fat compared to the control school. From analysis of plate waste, children from both schools consumed approximately 25% less energy than served (p < 0.05). Consumption of fat as a percentage of total energy and consumption of sodium was significantly less for the intervention compared to the control school. Compensation for the lower fat meals by the intervention school children (eating greater portions of high fat items) was not evident as the percentage of energy from fat consumed was only 1.3% greater than the percentage which was served. Lunch meals in this study were consumed as served. Intervention school children served lower fat and sodium meals consumed less fat and sodium than control school children and did not selectively eat only higher fat and sodium items. © 1996 American College of Nutrition.