A single pass perfusion system was used in anaesthetised, restrained rats to examine the effect of changing the composition of the perfusion fluid on the damage caused to the colonic epithelium by deoxycholic acid. Damage to the colonic surface was monitored with light microscopy, transmission and scanning electron microscopy and with measurements of deoxyribonucleic acid and carbohydrate in the perfusate. New scoring techniques for monitoring alterations in surface epithelium of light microscopy sections were used. The damaging effect of 5 mM deoxycholic acid to the colonic epithelium is inhibited by lowering the pH of the perfusion fluid from 7.9 to 5.5, or by increasing the calcium concentration from 0 to 4 mM. This inhibition is shown to be because of a decreased amount of bile acid in solution. Thus it is not the total concentration of deoxycholic acid in the colon that is responsible for the colonic damage, but the concentration in solution. Although extrapolation to the human situation must be made with caution, the concentration of bile acid in solution in the faecal water may be more relevant to colonic mucosal damage than total bile acid concentration.