Jejunal flora, bile acid deconjugation, and breath hydrogen (H2) and methane (CH4) excretion were studied in 22 Billroth II (BII)-operated patients with chronic postprandial symptoms, dumping (9), vomiting (7), pain (10), and diarrhoea (14). Sixteen were below 90% of desirable weight. Two control groups were included, one comprising 5 symptom-free, BII-operated volunteers and another comprising 12 healthy, unoperated volunteers. The numbers of bacteria recovered from jejunal secretions in the postgastrectomy patients did not differ significantly from those recovered in the symptom-free BII-operated controls but were significantly lower in the unoperated controls. Production of fermentation gas in anaerobic media supplemented with carbohydrates occurred in 17 of 22 postgastrectomy patients and in 4 of 5 BII-operated controls but in none of the unoperated controls. Bacterial bile acid deconjugating activity did not differ significantly between the postgastrectomy patients and the BII-operated controls but was significantly lower in the unoperated controls. Breath H2 excretion after glucose ingestion was significantly higher in the postgastrectomy patients than in both the BII-operated and the unoperated controls. The addition of pectin or guar gum to the glucose meal largely prevented postprandial symptoms and breath hydrogen excretion. Six out of 12 postgastrectomy patients treated with metronidazole recorded symptomatic effects, mainly on diarrhoea. Our findings indicate that jejunal bacterial overgrowth may be a major cause of the symptoms in some postgastrectomy patients. The tests available for demonstration of small-bowel bacterial overgrowth, perhaps with the exception of the glucose H2 breath test, did not differentiate satisfactorily between symptom-producing and non-symptom-producing abnormal jejunal flora. Thus these tests may seem to have a limited practical diagnostic value in such patients.