The aim of this study was to determine what factors influence the direction of movement of canine jejunal chyme. In four dogs, pacing electrodes were implanted near each end of a 50-cm jejunal Vella loop, while recording electrodes and intraluminal pressure catheters were spaced along the loop. After recovery, the loop was perfused from either the proximal stoma (forward flow) or the distal stoma (reversed flow), and effluent was collected from the nonperfused stoma. The pacesetter potentials were paced electrically in a forward (aborad) or a reverse (orad) direction. During control conditions (forward flow-forward pacing), the mean transit time of liquids was 2.6 +/- 0.1 min (mean +/- SE) and the static volume of the loop was 8.8 +/- 0.3 ml. Reversing both direction of flow and direction of pacesetter potential propagation slowed transit (4.4 +/- 0.4 min; P less than 0.05) and increased loop volume (16.0 +/- 1.0 ml; P less than 0.01). Reversing flow with forward pacing resulted in even slower transit (7.5 +/- 1.2 min; P less than 0.05), maintained a large volume (33 +/- 10 ml), and increased basal pressure in the loop from -1.2 +/- 1.7 (control) to 2.3 +/- 1.6 mmHg (P less than 0.05). These observations suggest that the direction of propagation of jejunal pacesetter potentials influences the direction of movement of jejunal chyme but that other factors have a role as well.