The relationships between blood Se levels and glutathione peroxidase (GSH-Px) activities and the ability of cattle to produce antibodies in response to antigen challenges with sheep red blood cells (SRBC) and ovalbumin (OA) were examined in a two-phase experiment with yearling beef cattle. There were no treatment effects on weight gains, feed intake or efficiency of feed conversion in either phase. In phase 1, cattle fed the unsupplemented control diet had lower (P ≤ 0.05) blood Se levels and GSH-Px activities by week 8 than cattle fed Se-fertilized forage, but neither was different (P > 0.05) from those of cattle fed Se-enriched yeast or inorganic Se. Antibody titers resulting from injection of SRBC at week 4 of the experiment peaked about 3 wk after injection, then declined. Differences due to Se treatment were small. The same animals were used in phase 2, where half of the animals on each of the four treatments of phase 1 were fed a supplement with Se-enriched yeast and the other half were fed an unsupplemented control. Differences in blood Se levels and GSH-Px activities due to phase-1 treatments persisted throughout phase 2. Differences due to phase-2 treatments were apparent after 9 wk. The cattle were challenged with OA at week 12 and at week 17 with both OA and SRBC. Differences in antibody titer due to Se treatment were small but tended to be higher (P > 0.05) in the Se-supplemented cattle. These results support the suggestion that blood Se levels over 100 μg L−1 are needed to maintain optimum immunocompetence. Key words: Selenium, glutathione peroxidase, immunocompetence, cattle, antibody