This is the first prospective study of urinary measures of the two major competing pathways of oestrogen metabolism, 16alpha-hydroxyoestrone (16alpha-OHE1) and 2-hydroxyoestrone (2-OHE1), in relation to incident breast cancer risk. Experimental and case-control study results suggest that metabolism favouring the more oestrogenic 16alpha-OHE1 pathway may be linked to higher breast cancer risk. Women aged 35 and older from Guernsey (n = 5104) were surveyed in 1977-85 and have been continuously monitored for breast cancer and mortality up to the present (Guernsey III, Imperial Cancer Research Fund). Incident cases of breast cancer were matched to three control subjects for comparison of urinary oestrogen metabolite levels measured by enzyme immunoassay (EIA) in spot urine samples collected at baseline and stored frozen for up to 19 years. Consistent with case-control study results, post-menopausal (but not premenopausal) women at baseline who went on to develop breast cancer showed about a 15% lower 2:16alpha-OHE1 ratio than matched control subjects. Further, subjects with metabolite ratios in the highest tertile of 2:16alpha-OHE1 had about a 30% lower risk than women with ratios in the lowest two-thirds, although results were not statistically significant (OR = 0.71, 95% CI = 0.29-1.75). It is of potential importance that, in contrast to most risk factors for breast cancer, such as late age at first birth, oestrogen metabolism appears to be modifiable via diet and exercise, offering women the possibility of lowering breast cancer risk through non-pharmacological measures, although this remains to be tested.