Purpose: The prognosis for gastric cancer with peritoneal seeding is very poor, and the role of surgical intervention is limited. We evaluated the effect of radical removal of primary and metastatic lesions on survival in gastric cancer with peritoneal seeding. Materials and Methods: From May 1989 to March 1999 at Kosin University Gospel Hospital, 115 patients revealed gastric cancer with peritoneal seeding but without liver or lung metastasis and without follow-up loss. The study group included 86 patients who underwent surgery for radical removal of primary gastric and metastatic peritoneal lesions. The control group included 29 patients who experienced incomplete removal of primary or metastatic lesions. Both groups received intraoperative intraperitoneal chemotherapy using mytomycin or cisplatin, and 25 patients underwent postoperative intravenous chemotherapy. Results: The median survival times in the study and the control groups were 13 months and 4 months, respectively (p<0.0001). The 1-year, 2-year, and 5-year survival rates were, respectively, $50.6\%,;18.1\%$, and $11.3\%$ in the study group and $14.8\%,;3.7\%$ and $0\%$ in the control group (p<0.0001). In the study group, neither postoperative intravenous chemotherapy nor microscopic invasion of the resection margin had any effect on survival, but intraoperative intraperitoneal chemotherapy and degree of peri-toneal seeding, especially the amount of peritoneal seeding, had an effect on survival. In the control group, neither intraperitoneal nor intravenous chemotherapy had any effect on survival, but resection of the primary gastric lesion improved survival. Conclusion: Radical removal of primary gastric and metastatic peritoneal lesions improved the survival rate for gastric cancer with peritoneal seeding. However, a randomized prospective study is needed to correctly evaluate the effect of intraperitoneal or intravenous chemotherapy.