To evaluate the validity of serum pepsinogen levels as a screening tool for gastric cancer and adenoma, immunoradiometric assays of serum pepsinogen I level (PG I), II level (PG II) and esphagogastroduodenal endoscopies were done in 757 health examinees. Serum PG I level was higher in subjects with active duodenal ulcer (n=45, $75.2{pm}34.3{mu}g/l(mean{pm}standard;deviation)$, p<0.01) and gastroduodenal ulcers (n=8, $75.6{pm}19.8{mu}g/l$, p<0.05), and was lower in those with gastric adenoma(n=4, $37.7{pm}37.2{mu}g/l$, p<0.2) than those with normal, mild gastritis findings or ulcer scars (n=378, $56.6{pm}24.9{mu}g/l$. Serum PG II level was higher in subjects with active duodenal ulcer($17.2{pm}13.8{mu}g/l$, p<0.2), active gastro-duodenal ulcers ($18.3{pm}7.4{mu}g/l$, p<0.2) and gastric carcinoma (n=3, $23.8{pm}10.9{mu}g/l$, p<0.05) than those with normal, mild gastritis findings or ulcer scars $(14.5{pm}7.9{mu}g/l)$. Serum PG I/PG II ratio was higher in subjects with active duodenal ulcer($5.1{pm}1.6$, p<0.05) and was lower in those with chronic gastritis(n=107, $4.1{pm}1.7$, p<0.05), gastric polyp(n=19, $3.9{pm}1.4$, p<0.2), gastric adenoma(n=4, $2.1{pm}1.9$, p<0.01) and gastric carcinoma(n=3, $2.7{pm}1.2$, p<0.1) than those with normal, mild gastritis findings or ulcer scars ($4.5{pm}1.7$). Serum PG II level increased with age until 6th decade, whereas serum PG I/PG II ratio decreased with age in 378 subjects with normal, mild gastritis findings or ulcer scars. The screening criteria of serum PG I<$70{mu}g/l$ and PG I/PG II ratio<3.0 for detecting gastric cancer and adenoma gave a positive rate of 15.7%, sensitivity of 57.1% and specificity of 84.7%.