To evaluate the effects of erythromycin on antroduodenal motility in children with chronic functional gastrointestinal symptoms, we studied 35 consecutive subjects referred for diagnostic motility studies. We recorded fasting motility for > 4 hr, then infused in random order either 1 or 3 mg/kg erythromycin intravenously over 1 hr and continued the study for another hour. Erythromycin induced phase III in 18 of 20 children who had phase III during fasting compared to only one of 15 who did not (P < 0.001). The antral motility index increased after erythromycin (1596 +/- 323 vs 436 +/- 242 mm Hg/30 min before erythromycin, P < 0.005) but the duodenal motility index did not change. The antral motility index was greater in children receiving 3 mg/kg than in those receiving 1 mg/kg (1968 +/- 391 vs 1226 +/- 285 mm Hg/30 min, P < 0.01), but duodenal motility indices did not differ. Only one child receiving the lower dose erythromycin complained of abdominal pain, nausea, or vomiting vs 9 of 19 the children receiving the higher dose (P < 0.02). In summary, in children with chronic functional gastrointestinal disorders, erythromycin rarely induced phase III in patients who did not have it during fasting. When different doses erythromycin are compared, 1 and 3 mg/kg are equally efficacious in inducing phase III episodes; the lower dose is associated with fewer side effects and the higher dose produces a higher antral motility index.