This review summarizes what is known about the effect of diet on irritable bowel syndrome (IBS) symptoms emphasizing data from randomized, controlled clinical trials. Studies suggest that IBS symptoms in one quarter of patients may be caused or exacerbated by one or more dietary components. Recent studies indicate that a diet restricted in fermentable, poorly absorbed carbohydrates, including fructose, fructans (present in wheat and onions), sorbitol, and other sugar alcohols is beneficial, but confirmatory studies are needed. Despite a long history of enthusiastic use, fiber is marginally beneficial. Insoluble fiber may worsen symptoms. Some patients with IBS, especially those with constipation, will improve with increased intake of soluble fiber. Prebiotic fibers have not been adequately tested. Daily use of peppermint oil is effective in relieving IBS symptoms. The usefulness of probiotics in the form of foods such as live-culture yogurt and buttermilk for IBS symptoms is not established. In clinical practice, it is very difficult to establish that a patient's symptoms result from an adverse reaction to food. A double blind placebo-controlled food challenge is the most reliable method, but it is not suitable for routine clinical use. A modified exclusion diet and stepwise reintroduction of foods or trials of eliminating classes of food may be useful.