Among the patients with rectal cancer who entered Yonsei University Hospital for management from Jan. 1980 to Dec. 1990, we selected 23 subjects who were received surgical resection of tumor in rectum, and who proved to have liver metastasis during the diagnostic work-up, at the time of the operation, or within 3 months after starting definitive treatment. With those subjects, we investigated the role of radiation therapy by comparison of the treatment results of the patients without radiation therapy (S group) with those of the patients with radiation therapy to the primary site (S+R group). The local control rates of S group and S+R group were $64{\%}$ and $89{\%}$, and 2-year survival rates were $50{\%}$ and $78{\%}$, respectively. Although there was not statistically meaningful difference, local control rate and 2-year survival rate were higher in the group with radiation therapy to primary site than that without radiation therapy. The 2-year survival rates of the case with resection of the liver and the case without it were $63.6{\%}$ and $58.3{\%}$ respectively, which was not statistically significant. Also, the 2-year survival rate of the case with sustained local control was higher than that of the case with local failure, which was statistically significant ($76.5{\%}$ and $16.7{\%}$, p<0.005). From the above results, it is thought that radiation therapy to the primary site might improve the local control rate even in the patients with liver metastasis, which seems to be correlated to the higher survival rate.