Reconstruction of pharyngoesophageal defects using free jejunal grafts has become an accepted technique of reconstruction. However, there are functional problems associated with the jejunal graft. We developed a canine model that allows us easy access to perform various studies on grafted jejunum, including videofluoroscopy and pressure manometry to determine baseline function. Using a microvascular technique, free jejunal grafts 10 to 30 cm in length were implanted in 11 mongrel dogs. The jejunal segments were implanted subcutaneously and exteriorized proximally and distally. The grafted dogs underwent videofluoroscopic studies. These studies revealed three different types of jejunal graft contractions of variable intensity: circumferential, longitudinal, and mixed. These contractions resulted in four patterns of barium movement: anterograde propulsion, retrograde propulsion, to-and-fro motion, and peristaltic propulsion. Videofluoroscopic studies were repeated on five dogs after an intravenous injection of metoclopramide (Reglan), which caused a significant short-term increase in the intensity of the basic jejunal contractions and barium propulsion. Pressure manometry studies using intraluminal pressure transducers were performed, revealing an inherent baseline contractility. Each dog has its own individual pattern of activity. The pressure generated by the contractions ranged from 5 to 350 mm Hg. Intravenous injection of Reglan produced a marked increase in pressure, but no change in the frequency of contractions. This study suggests that a free jejunal graft will maintain baseline motility. However, this graft may cause dysphagia by discoordination of contractions, retrograde propulsion of a bolus, or a sustained local contraction, demonstrating the clinical problems associated with free jejunal graft reconstruction of the cervical esophagus. Our results with Reglan suggest that it might be possible to improve the function of these grafts using pharmacologic agents.