We examined relationships among alcohol intake, dietary fat composition, and total body fat (TBF) and central abdominal fat (CAF), independent of genetic confounders, and evaluated the modulating effect of genetic susceptibility. We studied 334 female twins (57.7 +/- 6.7 yr) after excluding dietary underreporters. Diet was assessed by Food-Frequency Questionnaire and body fat by dual-energy x-ray absorptiometry. Moderate alcohol consumers (12-17.9 g/d) had less TBF (20.6 +/- 5.6 vs. 24.8 +/- 8.4 kg, P = 0.03) and CAF (1.2 +/- 0.6 vs. 1.6 +/- 0.7 kg, P = 0.03) than abstainers. In multiple regression, alcohol consumption remained independently associated with body fat distribution. In cotwin case-control (monozygotic twin) analysis, moderate alcohol consumption accounted for 300 g less CAF, independent of genetic and other environmental factors. Gene-environment interaction analysis indicated that this association was limited to subjects at high genetic risk of abdominal obesity. There was no relationship between dietary fat composition and adiposity. However, in women at low genetic risk of abdominal obesity, subjects with polyunsaturated fat intakes in the highest tertile had about 50% less CAF than subjects with intakes in the lowest tertile (0.9 +/- 0.4 vs. 1.6 +/- 0.4 kg, P = 0.0007), an association absent in subjects with high genetic risk. In conclusion, genetic risk modulates relationships between dietary factors and adiposity. Lower abdominal fat may mediate associations between dietary intake and type 2 diabetes risk.