Much work on the relationship of exercise level to cardiovascular risk factors has been stimulated by the pioneering epidemiological studies of Dr. Ralph Paffenbarger. At Stanford we have conducted a series of studies since 1973, with the following major findings. Very active middle-aged men and women (compared with sedentary controls) have higher plasma lipoprotein concentrations of high-density lipoprotein (HDL) cholesterol, lower levels of very-low-density-lipoprotein (VLDL) cholesterol and triglyceride (TG), and often moderately lower levels of low-density lipoprotein (LDL) cholesterol. Lipoprotein subfractions are also different: higher levels of HDL2and lower levels of “small” LDL in the active groups. Very active people are leaner and smoke less. All of these characteristics predict lower risk of coronary heart disease (CHD). A 1-yr trial of jogging in sedentary men suggested that 8–10 miles-wk−1of running is required for significant HDL changes. Body fat loss correlated positively with distance run, with increase in HDL, and withincreasein caloric intake. A 1-yr trial in overweight men showed that fat loss by dieting alone or by exercising alone results in similar elevation of HDL cholesterol. A 1-yr trial in sedentary, overweight men and women losing weight on a hypocaloric low-fat diet showed thatadditionof regular exercise increases weight loss, improves lipoprotein pattern, and further decreases 12 yr CHD risk.