Young adult male rhesus monkeys (Macaca mulatta) were fed an atherogenic diet for 38 months. After 38 months of atherosclerosis induction, a baseline group was selected and necropsied to determine the extent and severity of atherosclerosis before regression regimens were begun. The remaining animals were fed diets that varied in cholesterol concentration in order to maintain plasma cholesterol concentrations of approximately 200 or 300 mg/dl for either 24 or 48 months. The progression or regression of atherosclerosis in coronary arteries, abdominal aorta, and carotid arteries was determined by comparing them to the baseline group. Coronary artery atherosclerosis regressed in the majority of animals after 4 years but not after 2 years when plasma cholesterol concentrations were about 200 mg/dl. Among the animals maintained at plasma cholesterol concentrations of about 300 mg/dl, about half the animals progressed in the extent of coronary artery atherosclerosis while about half regressed. The majority of the animals that progressed in lesion extent were genetic hyperresponders to dietary cholesterol whereas those that regressed were predominantly hyporesponders, even though their plasma lipid concentrations were equivalent during the regression phase. The changes seen in atherosclerosis extent in the abdominal aorta were quite similar to the changes seen in coronary arteries. Changes at this site were not pronounced after 2 years, but after 4 years animals with plasma cholesterol concentrations of about 300 mg/dl progressed while the animals at 200 mg/dl were mostly unchanged. No evidence for atherosclerosis regression was found in the common carotid arteries or in the carotid bifurcations.