The selenium (Se) status of lambs grazing kikuyu (Pennisetum clandestinum) or ryegrass (Lolium multiflorum) in the KwaZulu-Natal Midlands was determined during four trials and their growth response to Se supplementation was measured. Two methods of Se supplementation were evaluated using a 2 x 2 factorial experimental design per season: a control, Se supplemented via fertilizing the pasture and/or injecting the lambs with a long-acting parenteral Se product. The Se concentrations in the whole blood of the unsupplemented groups at the end of the trial varied between 9 and 26 ng/ml, those in plasma between 5 and 15 ng/ml and in the livers between 96 and 163 ng/g DM. From these parameters it was concluded that the lambs were in state of Se deficiency, even though no symptoms of Se deficiency were observed. The lambs on the kikuyu did not grow well and even lost weight during the second summer. Despite the low weight gain during the first summer, a better (P < 0.05) growth response was obtained in the group receiving the parenteral Se supplementation above that of the unsupplemented group and the one grazing the Se fertilized kikuyu. The cumulative weight gains in both the Se fertilized groups during one ryegrass trial were higher (P < 0.05) than those in the control and parenteral supplemented groups. It was concluded that, although Se supplementation improved the growth of the lambs, the response varied between years. It was suggested that the poor growth of the lambs in some years was caused by factors other than a selenium deficiency and had to be overcome to obtain a positive response to Se supplementation.