The bioavailability of Se from ground beef has been previously found in this laboratory to be greater than that of selenite or selenate when fed to female Fischer 344 rats. In the present study we examined the bioavailability of Se from various commercial portions of beef, the liver, striploin, round, shoulder and brisket. All beef was cooked, freeze-dried, finely powdered and mixed with the other dietary ingredients. The experimental diets were fed to the weanling Fischer 344 rats which had been subjected to dietary depletion of Se for 6 weeks. The bioavailability of Se from the beef diets was compared with that of Se as selenite or L-selenomethionine (SeMet) added to torula-yeast diets. Each experimental diet contained 0·10 mg Se/kg. After 8 weeks of dietary Se repletion, relative activity of liver glutathione peroxidase (EC 1·11·1·9; GSHPx) from the different dietary groups compared with that of control animals (100%) was (%): selenite 91, SeMet 122 (P < 0·05), liver 108, striploin 105, round 106, shoulder 106, brisket 103. Se recovery for liver GSHPx was generally highest from SeMet beef muscle = beef liver selenite. Muscle tissue deposition of Se was highest from SeMet beef muscle selenite = beef liver. In addition, the faecal excretion of Se was lowest from the SeMet dietary group and highest from the selenite dietary group. The experimental results suggest that all cuts of beef appear to be highly bioavailable sources of dietary Se when compared with selenite or L-SeMet. Selenjum bioavailability: Beef: Selenite: Selenomethionine