A comparative case-referent study was conducted using data from the Hospital-Based Epidemiologic Research Program at Aichi Cancer Center (HERPACC) (Nagoya, Japan), with the aim of clarifying the effect of family history on gastric cancer by subsite. Our study comprised 995 histologically confirmed gastric cancer cases (180 cardia, 430 middle, 365 antrum and 20 unclassified) and a total of 43,846 non-cancer outpatients at Aichi Cancer Center Hospital between 1988 and 1995. Logistic regression was used to calculate odds ratios (ORs) for family history of gastric cancer and other cancers, adjusted for age, year and season at first hospital visit, habitual smoking, habitual alcohol drinking, regular physical exercise, preference for salty food and raw vegetable intake. In both genders, a positive family history of gastric cancer was associated with a moderate, but statistically significant increase in risk of gastric cancer [OR = 1.51, 95% confidence interval (95% CI) = 1.29–1.76], while no association was observed between the risk of gastric cancer and a family history of other cancers [OR = 0.97, 95% CI = 0.84–1.13]. OR increased for the middle and antrum parts of gastric cancer, but an increment for the cardiac part was observed only in those with a maternal history of gastric cancer. Our results suggest that the risk of gastric cancer in relation to family history varies by subsite and, furthermore, that the subsite-specific risk of gastric cancer is linked to a maternal history of gastric cancer. Int. J. Cancer 76:801–805, 1998.© 1998 Wiley-Liss, Inc.