A unique opportunity to evaluate the method of chemical lymph node clearance for colorectal cancer exists at Ferguson Hospital. Lymph node clearance has been used at the institution since 1977, and this retrospective analysis was undertaken to ascertain its validity there. Furthermore, the node positive group was evaluated to ascertain if the current staging system (Turnbull, 1967) is prognostically accurate for the Dukes' C group. Specifically evaluated for possible prognostic variance was the survival of those patients whose tumors demonstrated partial bowel wall penetration and only one to four positive nodes, a "C1 subset," previously reported to have favorable prognosis. Eight hundred sixty-four cases of colon and rectal cancer treated surgically from 1977 to 1982 were analyzed. There was a mean of 27 lymph nodes retrieved per specimen and a mean of 4.5 positive nodes per Dukes' C specimen. There were 43 C1 and 201 C2 cases with five-year survival rates of 73 and 38 percent, respectively. The results of chemical clearance at Ferguson Hospital were found to be comparable with that of other centers using chemical clearance and superior to hand dissection. The C1 subset clearly is noted to have prognostic advantage and should occupy a separate designation in any staging system.