We used Japanese monkeys (Macaca fuscata) to establish an experimental model in order to clarify the pathogenicity of Helicobacter pylori in gastric and duodenal disorders. A suspension (5 ml; 10(9) CFU/ml) of H. pylori cells isolated from humans was sprayed around the antrum of the stomach of each of 12 of 17 animals with an endoscope. The remaining five animals were not inoculated; they served as a control group. On days 7, 14, and 28 after inoculation, the gastric mucosa samples were examined grossly and were biopsied for microscopic examination with an endoscope. H. pylori was recovered from 7 of the 12 inoculated animals (58%), and infiltration by neutrophils and monocytes was observed histologically. Macroscopic gastritis with erythema and erosions were noted for five of these animals. On day 28 after inoculation, five animals in the infected group were treated with ampicillin. In two infected but untreated animals, the bacteria persisted for more than 6 months. The result of the gastritis scoring of the antral mucosa and the ammonia concentration in the gastric secretion were significantly higher (P < 0.01 to 0.001) for the infected group than for the control group; however, these values decreased to levels comparable to those for the control group after treatment with ampicillin. Urease activity was positive in gastric biopsy specimens from five of the seven animals in the infected group after 7 days and from four of these animals after 14 days but was negative in all specimens from animals in the control group. The level of antibody (immunoglobulin G) in serum for the infected group was elevated but changed very little for the control group. These results suggest that this M. fuscata model can be used to study H. pylori infection and that H. pylori can induce gastritis.