Fifty patients were admitted to a prospective controlled double blind trial for assessment of the effect of preoperative intramuscular cephazolin with oral metronidazole or placebo on the incidence of wound infection in patients requiring treatment by elective large intestinal resection. All patients were operated upon by one surgeon. After exclusion of two patients who died postoperatively 48 were analysed. No wound infection occurred in 22 patients receiving cephazolin with metronidazole. One (3.8%) infection developed in 26 patients receiving cephazolin with placebo. The difference did not attain statistical significance. Preoperative intramuscular cephazolin with or without metronidazole provides an effective and safe method of chemoprophylaxis for elective colorectal resection.