Investigations were made of the effects on nitrogen retention of continuous intravenous infusions of branched-chain amino acids (BCAA) with glucose. Eight wether lambs (35 kg) fed a full feed diet (2.68 Meal ME and 129 g CP kg^-1, DM basis) twice daily (50 g DM W^−0.75d⁻¹ ) at a level of 1.2 times their maintenance energy requirement were continuously infused intravenously (jugular) with L-leucine (Leu), L-isoleucine (Ile) and L-valine (Val). Infusion rates were 2.1, 1.5 and 1.7 g lamb⁻¹ d⁻¹, respectively, estimated to be about 50% of that absorbed from the gastrointestinal tract and/or required for muscle protein synthesis. Glycine was added to the infusion solutions to equalize the total N infused. Glucose (12 g W kg^−0.75d^-1) was also supplemented in the infusion solutions to maximise N-retention. Two 4 x 4 plus extra-period Latin squares were used. In square I, where the amino acids were infused individually, the treatments were: control, Leu, Ile and Val and in square II the treatments were: control, Leu, Leu + Ile and Leu + Ile + Val. Each period lasted 6 days. Digestibility coefficients of DM and OM were 75.8 and 77.1%, respectively, and were not affected by the treatment. Average N-digestibility was 67.1%; the Ile + Leu treatment increased N-digestibility (P < 0.03) compared to control. None of the individually infused amino acids significantly altered urinary-N (avg. 41% of total N-intake) or N-balance (avg. 22% of total intake). In both squares, Leu alone failed to affect urinary-N or N-retention. Nitrogen balance was statistically (P < 0.05) improved by 42%, when Leu was infused jointly with Ile. It is concluded that Leu infusion alone does not affect N-utilization in growing lambs, but when Leu is combined with Ile, N-utilisation is improved. The infusion of Val had little effect on N-utilization.