The effects of socioeconomic disadvantage on behavioral, psychosocial, and physiological risk factors for cardiovascular disease were investigated in an economically depressed agricultural area of New York State. After adjustment for age and sex, at least one and typically two of the three dichotomized socioeconomic factors (manual labor, lack of high school graduation, and poverty) were associated with an increased prevalence of smoking, obesity, frequent salt use, cholesterol consumption, low levels of leisure activity, and social isolation. Education and occupation also made independent contributions to systolic blood pressure. After adjustment for age, sex, behavioral, and social isolation variables, those with both risk factors (less than 12 years of school, and manual labor) had a higher mean systolic blood pressure than those with neither risk factor (9.5 mm Hg; 95 percent confidence interval = 5.5, 13.5). In contrast, after adjustment for the other variables, those with the two risk factors, less than 12 years of school and income below the poverty level, had a lower mean serum cholesterol than those with neither risk factor (11 mg/100 ml; 95 percent confidence interval = 3.1, 18.9). The policy implications of these pervasive social gradients of cardiovascular risk and the paradoxical relationship with serum cholesterol are discussed.