The effects of chronic progesterone treatment on gastric tumorigenesis were examined in 6-week-old male SD rats. The rats were castrated, progesterone or testosterone pellets were implanted, and, starting one week after the operation, 100 mg/liter of N-methyl-N'-nitro- N-nitrosoguanidine (MNNG) was administered in the drinking water for 16 weeks. Every 2 months the pellets were changed. Group 1 animals received castration plus MNNG while Groups 2 and 3 also received progesterone and testosterone, respectively. In the Group 4 case, progesterone and testosterone were administered alternately for 2-month periods and in Group 5 MNNG was given to intact animals. All survivors were killed one year after the start of MNNG treatment. In Group 1 the incidence of gastric tumors was significantly decreased as compared with the Group 5 value. The Group 2 incidence, in contrast, was similar to that in Group 5, and the size of the observed gastric tumors was massively increased. The area of the pyloric gland mucosa was also greater than in other groups. Testosterone treatment was associated with a less pronounced increase in tumor size and a recovery in incidence. The results indicate that progesterone may exert a promoting influence on gastric tumor development.