Aqueous ammonia in concentrations of 0.02 or 0.1% was continuously administered to rats to study its effect on the gastric mucosa histologically and cell kinetically. Furthermore, acetic acid ulcer, which is a model of chronic gastric ulcer, was experimentally induced in the stomachs of rats to assess the influence of 0.02% ammonia on the course of this ulcer. Male Donryu rats were divided into three groups given 0.02% ammonia, 0.1% ammonia or tap water. On several occasions (1, 3 and 5 days and 1, 4, 8, 12 and 24 weeks from the beginning of the experiment), the gastric mucosa in the fundic gland region and the antrum was examined histologically, and from the viewpoint of cell kinetics. The assessment in the 8th and 24th weeks employed the double labeling technique with bromodeoxyuridine and 3H-thymidine. The assessment on the other occasions used the flash labeling technique with bromodeoxyuridine. Both the 0.02% and 0.1% ammonia treatment groups showed a decrease in PAS-positive mucus and an enhanced cell cycling in the early stage of the experiment. After long periods of treatment, these groups showed a reduction in the gland height, a recovery in PAS-positive mucus and a suppression of cell cycle, suggesting direct toxicity of ammonia on the gastric mucosa. Although glandular atrophy was observed in these animals, infiltration of inflammatory cells was not observed. Thus, the relationship between ammonia and gastritis remained obscure. No ulcer developed in any group. Subsequently, we experimentally induced Ul-IV or Ul-V acetic acid ulcers in the stomachs of rats, according to the method of Okabe et al. (1971, 1972). These rats were divided into two groups given 0.02% ammonia or tap water. In the 4th and 8th weeks of the experiment, the stomachs of these rats were examined histologically and from the viewpoint of cell kinetics. The 0.02% ammonia treatment group showed a significant increase in the ulcer index (long diameter x short diameter; mm2) in the 4th and 8th weeks. This group also showed suppressed cell cycling of the regenerative epithelium and fibroblasts in the ulcer margin, suggesting direct toxicity of ammonia. Thus, healing of peptic ulcer was delayed by continuous administration of 0.02% ammonia.