In vitro studies under short-circuited conditions suggest that amiloride, a diuretic agent which is thought to block apical membrane sodium entry, has significant effects on sodium absorption by the human colon. To evaluate this in vivo, we studied the effect of amiloride applied in concentrations of 10(-5) and 10(-4) M on sodium transport and potential difference (PD) in human colon during steady-state perfusion. Net sodium absorption was reduced 25% by amiloride and chloride absorption by 15%; potassium and bicarbonate secretion rates were enhanced. In other studies the colon was divided into a proximal and distal test segment by endoscopic introduction of a collection channel to the descending colon-sigmoid junction. Comparison of tritiated water absorption by the two segments indicated that the distal segment comprised approximately 20% of the total colon surface area. However, the distal test segment only accounted for 5-7% of total sodium, chloride, or water absorption; in contrast, 17-20% of total potassium or bicarbonate secretion occurred there. In the proximal test segment, amiloride reduced net sodium absorption by almost one third from 21.0 to 14.4 mmol/hr (P less than 0.02) but had no significant effect on PD. In the distal test segment, amiloride produced a 25% reduction in mean sodium absorption from 1.2 to 0.9 mmol/hr, but this reduction was not statistically significant; however, potential difference was significantly reduced by 33% (P less than 0.02). These results suggest that most sodium absorption in normal human colon in vivo is mediated by transport processes which are insensitive to these doses of amiloride.(ABSTRACT TRUNCATED AT 250 WORDS)