The health consequences of an adverse body-fat distribution (e.g., android, upper-body, visceral) have only recently concerned the medical community. Ninety years ago, however, actuarial study demonstrated the relationship of body-fat distribution to the mortality experience of insured, North American men. Thirty-four insurance companies pooled their data on males issued life policies between 1870 and 1899. Special classes of risk were defined by weight for height at baseline or by the observation that abdominal girth exceeded the girth of the expanded chest (abdominal obesity). The mortality experience of each risk class was compared to an age-stratified, actuarial table of the period. We present new analyses of these historical data relating specifically to the mortality impact of abdominal obesity. Among 163,567 overweight men, the prevalence of abdominal obesity increased with age and with degree of overweight. Among moderately overweight men, those with abdominal obesity experienced 133% of the expected mortality rate compared to 112% of the expected mortality for those who were not abdominally obese. Severely overweight men with abdominal obesity experienced 152% of the expected mortality compared to 135% of the expected mortality for severely overweight men who were not abdominally obese. We believe this nineteenth-century, acturial study of waist and chest girths was the first demonstration that body-fat distribution can influence longevity. These early actuarial findings, taken with more recent reports, establish that abdominal enlargement, but not necessarily an 'upper-body' fat distribution, constitutes a major health hazard. Future research must establish which abdominal-obesity index best predicts disease outcomes.