Dietary glycemic load, the mathematical product of the glycemic index (GI) of a food and its carbohydrate content, has been proposed as an indicator of the glucose response and insulin demand induced by a serving of food. To validate this concept in vivo, we tested the hypotheses that 1) portions of different foods with the same glycemic load produce similar glycemic responses; and 2) stepwise increases in glycemic load for a range of foods produce proportional increases in glycemia and insulinemia. In the first study, 10 healthy subjects consumed 10 different foods in random order in amounts calculated to have the same glycemic load as one slice of white bread. Capillary blood samples were taken at regular intervals over the next 2 h. The glycemic response as determined by area under the curve was not different from that of white bread for nine foods. However, lentils produced lower than predicted responses (P