To investigate the selenium status during long-term dietary supply of selenium yeast, 30-day-old male rats were fed for 379 days a methionine-adequate low-selenium diet supplemented with 0.2mgSe/kg (selenium-adequate diet) or 1.5mgSe/kg (high-selenium diet) in the form of selenium yeast that contained 60% of the element as l-selenomethionine. Their selenium load was determined at several intervals by neutron activation analysis of the selenium concentrations in the main selenium body pools, skeletal muscle and liver. After 64 days the tissue selenium concentrations plateaued in both groups and then stayed at that level. Compared with the selenium-adequate group, elevated tissue selenium concentrations were found in the high-selenium group, but the increase by a factor of 3.5 in the muscle and by a factor of 2.3 in the liver was smaller than the 7.5-fold increase in the selenium intake. In the selenium-adequate group about 50% of the muscle selenium and 30% of the liver selenium and in the high-selenium group about 85% of the muscle selenium and 70% of the liver selenium were estimated to be present in non-selenoprotein forms. During selenium depletion the liver glutathione peroxidase activity in the high-selenium group remained unaffected for 4 weeks and then decreased more slowly than that in the selenium-adequate group. From these results it can be concluded that selenium incorporated from the selenium yeast diet into non-selenoprotein forms can serve as an endogenous selenium source to maintain selenoprotein levels in periods of insufficient selenium supply.