Fifty-two moderately obese adult women were stratified according to their baseline breakfast-eating habits and randomly assigned a weight-loss program. The no-breakfast group ate two meals per day and the breakfast group ate three meals per day. The energy content of the two weight-loss programs was identical. After the 12-wk treatment, baseline breakfast eaters lost 8.9 kg in the no-breakfast treatment and 6.2 kg in the breakfast treatment. Baseline breakfast skippers lost 7.7 kg in the breakfast treatment and 6.0 kg in the no-breakfast treatment. This treatment-by-strata-by-time interaction effect (P less than 0.06) suggests that those who had to make the most substantial changes in eating habits to comply with the program achieved better results. Analyses of behavioral data suggested that eating breakfast helped reduce dietary fat and minimize impulsive snacking and therefore may be an important part of a weight-reduction program.