ABSTRACT— The propensity of legumes to promote formation of intestinal gas was evaluated in healthy young men by measuring flatus passed from the rectum and the amount of bacterially formed gas in the exhaled breath. Previous experience has shown that flatulent test meals fed in the morning cause elevation of breath hydrogen concentration, flatus volume and flatus hydrogen and carbon dioxide content about 5–7 hr later. Test responses were measured against a baseline derived from feeding a bland, low-residue formula diet. In the present study, mature, dry lima beans were found to be as high in flatulence-inducing factor(s) (FF) as are California small white beans. Mung beans and soybeans caused the same excretion of hydrogen in the breath as did white beans, but only about 2/3 as much flatus. Peanuts were not gas-forming. When the dosage of soybeans was doubled, so that the carbohydrate contribution from the two varieties was more nearly equalized, soybeans caused as much passage of flatus as did white beans. Several products processed by methods that might be expected to alter FF concentration or activity were evaluated in the same way. Soybean and mung bean sprouts appeared to retain most of the FF present in the whole bean. Soybean curd (tofu) had little residual FF but another high-protein food, MPF, made from toasted soy grits, caused as much gas formation as an equal weight of soybean carbohydrate fed as whole beans. Enzymatic treatment of comminuted soybeans, designed to hydrolyze the constituent oligosaccharides, raffinose and stachyose, frequently cited as gas-formers, had only negligible effect on production of intestinal gas. Tempeh, made from soybean grits by mold fermentation, did not increase gas production over baseline values and caused a significant delay in the time of gas formation, suggestive of temporary suppression of intestinal bacteria. Ethanolic extraction of whole white beans reduced but did not eliminate their gas-forming quality.