The mucosal barrier has three major components, the mucus layer, the epithelial glycocalyx and the surface epithelium itself, whose integrity largely depends on tight junction function. In health, there is relatively little direct interaction between the luminal microbiota and the epithelium - the continuous mucus layer in the colon keeps the surface epithelium out of contact with bacteria and the ileo-caecal valve ensures that the distal small intestine is relatively microbe free. Most interaction takes place at the Peyer's patches in the distal ileum and their smaller colonic equivalents, the lymphoid follicles. Peyer's patches are overlain by a ‘dome' epithelium, 5% of whose cells are specialised M (microfold) epithelial cells, which act as the major portal of entry for bacteria. There are no goblet cells in the dome epithelium and M cells have a very sparse glycocalyx allowing easy microbial interaction. It is intriguing that the typical age range for the onset of Crohn's disease (CD) is similar to the age at which the number of Peyer's patches is greatest. Peyer's patches are commonly the sites of the initial lesions in CD and the ‘anti-pancreatic' antibody associated with CD has been shown to have as its epitope the glycoprotein 2 that is the receptor for type-1 bacterial fimbrial protein (fimH) on M cells. There are many reasons to believe that the mucosal barrier is critically important in the pathogenesis of inflammatory bowel disease (IBD). These include (i) associations between both CD and ulcerative colitis (UC) with genes that are relevant to the mucosal barrier; (ii) increased intestinal permeability in unaffected relatives of CD patients; (iii) increased immune reactivity against bacterial antigens, and (iv) animal models in which altered mucosal barrier, e.g. denudation of the mucus layer associated with oral dextran sulphate in rodents, induces colitis. Whilst some IBD patients may have genetic factors leading to weakening of the mucosal barrier, it is likely that environmental factors may be even more important. Some may be subtle and indirect, e.g. the effects of stress on the mucosa barrier, whilst others may be more obvious, e.g. the effect of pathogen-related gastroenteritis, known often to act as trigger for IBD relapse. We have also been very interested in the potentially harmful effects of ingested detergents - either by contamination of cutlery by inadequate rinsing or via ingestion of processed foods containing permitted emulsifiers. In vitro and ex vivo studies show that even very small trace amounts of these surfactants can greatly increase bacterial translocation. Implications for therapy are not yet so obvious. We advise our IBD patients to avoid processed foods containing emulsifiers and to rinse their dishes well - whilst accepting that there is no direct evidence yet to support this. Therapies that aim to enhance the mucosal barrier have yet to come to market, but trials of enteric-delivered phosphatidylcholine in UC are promising. The faecal concentration of mucus-degrading bacterial enzymes (particularly proteases, sulphatases and sialidases) correlates with disease activity in UC, and these represent good targets for therapy.