Summary Increased fibre intake has been shown to reduce serum oestrogen concentrations. We hypothesized that fibre exerts this effect by decreasing the time available for reabsorption of oestrogens in the colon. We tested this in volunteers by measuring changes in serum oestrogen levels in response to manipulation of intestinal transit times with senna and loperamide, then comparing the results with changes caused by wheat bran. Forty healthy premenopausal volunteers were placed at random into one of three groups. The first group took senna for two menstrual cycles then, after a washout period, took wheat bran, again for two menstrual cycles. The second group did the reverse. The third group took loperamide for two menstrual cycles. At the beginning and end of each intervention a 4-day dietary record was kept and whole-gut transit time was measured; stools were taken for measurement of pH and j-glucuronidase activity and blood for measurement of oestrone and oestradiol and their non-protein-bound fractions and of oestrone sulphate. Senna and loperamide caused the intended alterations in intestinal transit, whereas on wheat bran supplements there was a trend towards faster transit. Serum oestrone sulphate fell with wheat bran (mean intake 19.8 g day-') and with senna; total-and non-protein-bound oestrone fell with senna. No significant changes in serum oestrogens were seen with loperamide. No significant changes were seen in faecal 3-glucuronidase activity. Stool pH changed only with senna, in which case it fell. In conclusion, speeding up intestinal transit can lower serum oestrogen concentrations.Keywords: intestinal transit; fibre; oestrogen; breast cancer There is substantial experimental, epidemiological and clinical evidence that breast cancer risk is influenced by endogenous hormones.Breast cancer is less common in rural Third World communities than in developed countries (Lea, 1966;Drasar and Irving, 1973;Armstrong and Doll, 1975;Miller, 1977) and becomes more common on migration from low-to high-risk areas, even within one generation (Staszewski and Haenszel, 1965;Buell, 1973), implying that environmental factors are important. Case-control studies have shown low fibre intake, with or without high fat intake, to be associated with increased risk of breast cancer (Katsouyanni et al, 1986;Lubin et al, 1986;Howe et al, 1990;Zaridze et al, 1991). How such diets exert this influence has not been established, but one possibility is by an effect on oestrogen metabolism.A high-fibre diet (Feng et al, 1993), or addition of wheat bran to the diet of healthy women (Rose et al, 1991), has been reported to reduce serum oestrogen levels. In rats given bran there was an increase in stool oestrogen (Neale, 1983). Vegetarians have higher faecal excretion and lower urinary excretion of oestrogens than omnivores (Armstrong et al, 1981;Goldin et al, 1982;Gorbach and Goldin, 1987). Plasma oestrogens have been found to be lower in vegetarians than omnivores in some (Shultz and Leklem, 1983) but not all studies (Goldin et al, ...