In a survey of patients treated with coloanal anastomosis for rectal cancer, 81 of 90 eligible patients responded to a questionnaire evaluating current anorectal function. Time from operation to assessment ranged from 1.3 to 12.3 years (median: 4.3 years). The median stool frequency was two per day; 22% of patients reported four or more stools per day. In the patients surveyed, fecal continence was complete in 51%, incontinence to gas only in 21%, minor leak in 23%, and significant leak in 5%. Complete evacuation of the neorectum was problematic in 32%. Overall function was excellent in 28%, good in 28%, fair in 32%, and poor in 12%. The impact of treatment variables on functional outcome was assessed by univariate and multivariate analyses. No surgical technique correlated with improved or impaired outcome. Time since surgery (reduced stool frequency) and use of postoperative adjuvant radiotherapy (increased stool frequency, increased difficulty with evacuation) did appear to influence functional outcome. We conclude that the functional results of coloanal anastomosis are good but not optimal. Continued investigation of the effects of surgical technique and adjuvant therapy is warranted.