In the search for a superior alternative to conventional bowel preparation which often gives unsatisfactory results in children, we have introduced whole-gut irrigation for pediatric use. After a pilot study on 15 children during which adjustments on the technique were made, we settled on an intensive regimen with some notable modifications from conventional adult practice: the use of warm Hartmann's solution, a relatively large fluid load (mean volume 5.3 l/kg body weight, range 3.0 - 12.0 l/kg) and a rapid infusion rate (1.5 ml/kg/min). We then evaluated its safety, effectiveness, and acceptability prospectively on 45 patients undergoing colonoscopy or colorectal surgery, age ranging from 4 months to 11 years, with a mean of 3.9 years. Subjective complaints were mild and included nausea/vomiting, 12 cases (26.7 percent); abdominal colic, two (4.4 percent); and distending discomfort, three (6.7 percent). There was a mean weight gain of 4.0 percent but no gross electrolyte disturbances. Results of bowel preparation were satisfactory in 33 (73.3 percent), adequate in ten (22.2 percent) and poor in two (4.4 percent). Compared with our previous method, in which inadequate preparation occurred in 4/20 patients, (20 percent) by conventional measures, whole-gut irrigation represents a statistically significant improvement (P less than 0.05). In addition, whole-gut irrigation shortened hospital stay and obviated the traditional need of two to three days' dietary restrictions.