In vivo rectal dialysis was used to study rectal mucosal release of immunoreactive prostaglandin E2-like material and its relation to disease activity, rectal electrical potential difference (PD), and treatment in 24 patients with ulcerative colitis. In untreated colitics in remission and in relapse, median values for apparent mucosal prostaglandin E2 (PGE2) release were increased three-fold (P < 0.05) and 13-fold (P < 0.002) respectively over that found in control subjects. In patients in remission during treatment with sulphasalazine and/or corticosteroids, median apparent PGE2 release was similar to that of controls, but in colitics in relapse, despite treatment, it was greatly increased (P< 0.002). Ulcerative colitis in relapse was associated with a significant reduction in rectal PD(P < 0.002); in patients with quiescent ulcerative colitis, a smaller reduction was found (P < 0.05). In nine patients studied serially before and during treatment, there were associations between changes in disease activity assessed sigmoidoscopically, in PD and in apparent mucosal PGE2 release. Furthermore, rectal mucosal PGE2 release and PD were linearly correlated (P < 0.01). These findings indicate that mucosal PGE2 release is markedly enhanced in active ulcerative colitis, and they confirm the value of rectal PD as a guide to disease activity. In addition, they suggest that rectal dialysis may be a useful way of studying rectal prostaglandin metabolism in man.