It has been suggested that protein synthesis in peripheral tissues: (1) responds in a curvilinear manner to increasing feed intake over a wide range of feeding levels; and (2) has a greater sensitivity to intake than protein breakdown. The aim of the present experiment was to test these hypotheses across the ovine hindlimb. Six growing sheep (6–8 months, 30–35 kg), with catheters in the aorta (two), posterior vena cava and jugular vein, received each of four intakes of dried grass pellets (0·5, 1·0, 1·5 and 2·5×maintenance energy; M) for a minimum of 7 d. A U-13C-labelled algal hydrolysate was infused intravenously for 10 h and from 3–9 h para-aminohippuric acid was infused to measure plasma flow. Arterial and venous plasma were obtained over the last 4 h and the concentrations and enrichments of thirteen 13C-labelled amino acids (AA) were determined by GC–MS. As intake increased, a positive linear response was found for plasma flow, arterial concentrations of the aromatic and branched-chain AA, total flow of all AA into the hindquarters and net mass balance across the hindquarters (except glycine and alanine). Based on two separate statistical analyses, the data for protein synthesis showed a significant linear effect with intake (except for phenylalanine, glycine and alanine). No significant curvilinear effect was found, which tends not to support hypothesis 1. Nonetheless, protein synthesis was not significantly different between 0·5, 1·0 and 1·5×M and thus the 2·5×M intake level was largely responsible for the linear relationship found. There was no significant response in protein breakdown to intake, which supports hypothesis 2.