Two experiments were conducted to determine the effect of various forms of selenium (Se) on the activity of glutathione peroxidase (GSHPx) in liver, heart, kidney and eyes of the developing rat. In experiment 1, throughout mating, pregnancy and lactation, female rats consumed one of three diets: basal (< 0.05-mu-g Se/g); selenite (0.15-mu-g Se/g) and selenomethionine (0.15-mu-g Se/g). Some pups born to dams in the basal group were also given intraperitoneal doses of saline, selenite or selenomethionine. GSHPx activity was measured in tissues from fetuses, 7-d-old and 14-d-old nursing pups and the dams. In all tissues studied, GSHPx activity was highest in the 14-d-old pups whose mothers were in the selenomethionine group. Rat pups given intraperitoneal selenite (3-mu-g/kg body weight) had higher liver and kidney GSHPx activity than pups given the same amount of selenium as intraperitoneal selenomethionine. In experiment 2, all dams were fed the same basal diet, and pups were weaned to diets containing one of two levels of selenium (0.1 or 0.2-mu-g/g), one of three forms of selenium (selenite, selenomethionine or selenocystine) or no added selenium. After 14 d of repletion, the highest level of hepatic GSHPx activity occurred in the selenite group and the lowest in the basal diet group. After 21 d of repletion, renal GSHPx activity was lowest in the basal group followed by the selenocystine group. The highest tissue selenium concentration was found in kidney tissues of the selenocystine group. These data support the hypothesis that these dietary forms of selenium are differentially available for GSHPx activity.