New molecular biologic techniques have opened a new era of pursuit of an infectious cause of Crohn's disease. Early data, however, are negative for the presence of DNA from mycobacteria and Chlamydia in Crohn's disease tissue. The role of the fecal stream in Crohn's disease has again been highlighted by excellent clinical observations in the pathogenesis of recurrent ileal disease following 'curative' resection and requires more detailed appraisal. The elegant demonstration of the intimate involvement of vascular endothelium in granulomas in Crohn's disease emphasizes the much-ignored role of endothelium in inflammatory events. More information on abnormalities of some key cytokines in the intestinal mucosa still leaves the question of 'cause versus effect' unanswered. A role for the 'impotent' neutrophil in Crohn's disease has again been supported by studies of postphagocytic activities. For ulcerative colitis, a study of monozygotic twins has confirmed that abnormalities of mucin glycoproteins in large bowel epithelium found in affected patients is unlikely to be a consequence of mucosal inflammation but, by itself, is not responsible for the development of such inflammation. The impressive disease specificity of perinuclear antineutrophil cytoplasmic antibodies to ulcerative colitis and to primary sclerosing cholangitis offers the exciting prospect of defining a putative autoantigen that would considerably expand our investigational capabilities, and provides a link between the two diseases. Pouchitis, a condition of ileoanal reservoirs that appears specific for patients with ulcerative colitis, may also offer clues to the cause of ulcerative colitis and recent studies have begun to examine roles of the lumen, the epithelium, and immunoinflammatory events in its pathogenesis. The void in our understanding of the pathogenesis of diarrhea in colitis has been narrowed by studies in experimental animals and humans on muscle, nerve and epithelial function, and on factors influencing them.