Controversy continues regarding the initial management of civilian colon injuries. The main issues are the safety of colostomy versus the desirability of primary repair and the role of exteriorized repair. From 1979 through 1984, 727 patients with injuries to the colon were treated at a large urban trauma center. Ninety-seven per cent of injuries were caused by penetrating wounds. Ten patients died in the operating room prior to repair of the colon wound. For patients who survived long enough to have their injury treated, 52.4% were treated by primary repair, 32.9% were treated with colostomies, and 14.6% were treated with exteriorized repair. Of the factors that have been stated to influence decision making, the extent of the colon injury was the most important. Location, number, and type of associated injuries, fecal contamination, and shock were less important. However, none of these latter factors mandated performance of a colostomy. The overall mortality rate for the series was 9.9%. Forty-one out of 70 deaths occurred within the first 48 hours and were due to shock and hemorrhage. The mortality rate for primary repair was significantly lower than that for colostomies (p less than 0.01). The presence of shock and age greater than 40 were significant factors influencing mortality (p less than 0.01). Mortality also was directly related to the number and type of associated abdominal injuries. Abdominal abscess also occurred significantly less often in patients treated with primary repair than in those with colostomies (p less than 0.01). The use of exteriorized repair was successful in avoiding colostomy in 59% of patients. Primary repair can be performed with minimal morbidity and mortality and should be the mainstay of treatment for civilian colon injuries.