There is conflicting evidence about the influence of fibrinogen genotype on plasma fibrinogen concentrations, and the relation between genotype and atherosclerotic disease has not been studied. In a population-based case-control study we aimed to find out whether certain fibrinogen genotypes are associated with an increased risk of peripheral atherosclerosis. 121 subjects with peripheral arterial disease and 126 healthy controls matched for age and sex were selected from a random population sample aged 55-74 years in the Edinburgh Artery Study. Mean fibrinogen concentrations were higher in cases than in controls (3.12 [95% confidence interval 2.99-3.26] vs 2.75 [2.64-2.85], p less than 0.001). A greater proportion of cases than controls were homozygous or heterozygous for an allele at the beta fibrinogen locus (4.2 kb allele, Bcl I digestion); the allele frequency was 0.197 in cases and 0.097 in controls (p less than 0.005). Extended haplotypes for 4.2 kb heterozygotes were also associated with an increased risk of peripheral arterial disease. However, haplotype had only a small effect on the association of plasma fibrinogen concentration with disease, and the relation of haplotype with disease was independent of age, sex, social class, smoking status, plasma fibrinogen, alcohol consumption, body mass index, and diabetes mellitus. We conclude that variation at the beta fibrinogen locus is associated with an increased risk of peripheral atherosclerosis. The influence is not mediated simply by way of increased fibrinogen concentrations but could be due to a structurally variant fibrinogen or linkage disequilibrium with a neighbouring gene.