There is considerable evidence that the level of deoxycholic acid in the bile influences biliary cholesterol saturation. Deoxycholic acid is formed in the colon and absorbed slowly. Hence changes in colonic transit rate might influence biliary deoxycholic acid and the cholesterol saturation of bile. When 14 constipated subjects took standardised senna tablets for six weeks in a dose sufficient to lower mean whole gut transit time from 134 to 54 hours, deoxycholic acid as a proportion of biliary bile acids fell from 25.9 +/- 8.6 to 17.2 +/- 8.3% (p less than 0.0001) and deoxycholic acid pool measured by isotope dilution fell from 0.64 +/- 0.34 to 0.45 +/- 0.29 g (p less than 0.0001). In those subjects (n = 8) whose bile was initially supersaturated with cholesterol, the saturation index fell from 1.40 +/- 0.22 to 1.20 +/- 0.19 (p = 0.02). Conversely, when 12 normal volunteers took loperamide capsules sufficient to cause symptomatic constipation and to prolong mean transit-time from 48 to 103 hours, the deoxycholic acid pool increased from 0.40 +/- 0.24 to 0.57 +/- 0.17 g (p = 0.008). The percentage deoxycholic acid did not alter significantly, because the estimated total bile acid pool expanded (from 1.98 +/- 0.61 to 2.81 +/- 0.48 g; p less than 0.001), presumably because of loperamide slowing down small bowel transit. Despite this expansion of the bile acid pool, loperamide increased the cholesterol saturation index from 1.10 +/- 0.31 to 1.20 +/- 0.32 (p = 0.01). Changes in colonic transit rate alter the size of the deoxycholic acid pool and bile cholesterol saturation. These findings suggest that constipation or slow colonic transit might increase the chance of supersaturated bile and hence of gall stones.