Prostaglandin analogues, used in the treatment of duodenal and benign gastric ulcer and in the prevention of gastric ulceration caused by non-steroidal anti-inflammatory drugs, are frequently associated with gastrointestinal side effects, particularly diarrhoea and abdominal cramps. We investigated the effects of misoprostol, a prostaglandin E1 derivative, on bowel motility and faecal loss of fat, water and bile acids in relation to its postprandial vs. preprandial administration. Twelve healthy subjects participated in a double-blind crossover study comparing three 5-day courses of therapy with a washout period of 1-2 weeks between courses. Following a Latin Square design, the dosing regimens were (a) 400 micrograms misoprostol b.d. after meals and placebo b.d. before meals; (b) 400 micrograms misoprostol b.d. before meals and placebo b.d. after meals; (c) placebo before and after meals. Orocaecal transit time measured by H2 breath tests following lactulose administration, was shortest during pre-prandial dosing but was also significantly decreased during post-prandial dosing. The overall treatment difference was highly significant (P less than 0.001), and the difference between each pair of treatments was also statistically significant. Whole bowel transit time studied by means of 3H-PEG 4000 determination in stools, was shorter for the two misoprostol regimens but statistical significance was borderline. The number of stools passed per day was similar in the three groups. During both misoprostol dosing periods, stools were less formed and their content of water, fat and bile acids was higher. There was also more urgency, flatulence, abdominal pain and nausea. It is concluded that the gastrointestinal side effects caused by misoprostol are mainly based on an increased orocaecal transit time. The effects are more important when the drug is administered before meals than after meals.