The oral ingestion of a meal or the delivery of nutrients directly to the stomach or duodenum stimulates water and ion absorption from the proximal jejunal lumen. To further investigate this phenomenon, this study tested two hypotheses: (1) direct jejunal nutrient delivery stimulates jejunal absorption, and (2) the signal for jejunal absorption requires intact enteric neurotransmission and will therefore be altered by mucosal neural blockade with the local anesthetic bupivacaine. Intestinal absorption studies (N = 52) were performed on eight dogs with 25-cm jejunal Thiry-Vella fistulas (TVF) and feeding jejunostomies. Luminal perfusion with [14C]PEG was used to calculate TVF absorption of H2O, Na+, and Cl-. Six groups were randomly studied over 4 hr. Each group incorporated a basal hour, a TVF or jejunostomy treatment hour, and an oral (groups 1 and 3) or a jejunal (groups 4 and 6) meal stimulus. The oral and jejunal meals were isocaloric and of identical composition. Groups 1-3 had saline (as a control) or 0.75% bupivacaine applied to the lumen of the TVF. Groups 5 and 6 had 0.75% bupivacaine application to the feeding jejunostomy. Both the oral and the jejunal meal stimuli resulted in a significant proabsorptive response in the TVF. TVF bupivacaine reduced basal absorption but did not diminish the meal-induced proabsorptive response. Treatment of the jejunostomy with bupivacaine caused no change in basal or postmeal absorption in the TVF.(ABSTRACT TRUNCATED AT 250 WORDS)