The results of our experience with the technique of fine needle catheter jejunostomy and early postoperative feeding through the catheter over a 12-month period are presented. Of the 43 patiening with an elemental diet without complications, 3 (7 per cent) were not fed for different reasons and 13 (30 per cent) developed complications that led to temporary or permanent cessation of the feeding. One patient in this group died of inhalation pneumonia. The voluntary food intake of 12 of the patients who were fed by jejunostomy after major colorectal surgery was assessed daily for 2 weeks after operation and compared with that of 12 control patients who did not have nutritional support. No significant difference in voluntary food intake was found between the two groups. The changes in body composition and plasma proteins and the clinical outcome of 20 of the patients fed by jejunostomy after major colorectal surgery were also compared with those of 20 matched controls. Body weight and lean body mass (as assessed by total body potassium and arm muscle circumference) and plasma prealbumin fell significantly in the control patients but not in those fed by jejunostomy. However, plasma transferrin decreased in both groups and there was no significant difference in clinical outcome in terms of complication rate or duration of postoperative hospital stay. The study would suggest that this technique of jejunostomy feeding should be reserved for selected cases where it may prove to be of real value, rather than being used routinely.