The objective of this study was to develop an experimental model to measure both fractional and absolute rates of protein synthesis in the small intestine of lambs. Six male castrate lambs (∼6 months old, mean liveweight 26 kg) were offered, via continuous feeders, 900 g/day lucerne chaff. They were prepared with catheters in the cranial mesenteric vein (CMV), femoral artery (FA), jugular vein and abomasum, and a blood flow probe around the CMV. Cr-EDTA (0·139 mg Cr/ml, ∼0·2 ml/min) was infused abomasally for 24 h and L-[2,6-3H]phenylalanine (Phe) (441±33·8 μCi into the abomasum) and L-[U-14C]phenylalanine (43·9±4·08 μCi into the jugular) were also infused during the last 8 h. Blood from the CMV and FA was sampled during isotope infusions. At the end of infusions, lambs were killed and tissue and digesta samples removed from four sites along the small intestine (SI). Transfers of labelled and unlabelled Phe were measured between SI tissue, its lumen and blood, enabling both fractional and absolute rates of protein synthesis and gain to be estimated. The total SI protein pool was 84 (±1·7) g and fractional gain rate was 7·5 (±5·5)% per day. Mean protein synthesis and fractional synthesis rates (FSR) were calculated from the mean retention of 14C and 3H in SI tissue. FSR tended to increase caudally along the SI (although P > 0·05). The choice of free Phe pool for estimating precursor specific radioactivity (SRA) for protein synthesis had a significant effect on FSR. Assuming that tissue free Phe SRA represented precursor SRA gave a mean FSR of 129 (±24)% per day. Corresponding estimates for free Phe SRA in the FA and CMV were 20 (±1·8) and 30 (±3·1)% per day respectively. The correct value for protein synthesis was therefore in doubt, although indirect evidence suggested that blood SRA (either FA or CMV) may be closest to true precursor SRA. This evidence included (i) comparison with flooding dose estimates of FSR, (ii) comparison of 3H[ratio ]14C Phe SRA in free Phe pools with this ratio in SI protein, and (iii) the proportion of SI energy use associated with protein synthesis. Advantages of the present experimental model compared to other methods included (i) measurements of both protein synthesis and gain, and hence, all components of turnover, (ii) measurement of absolute as well as fractional rates of synthesis and gain, (iii) inclusion of proteins which are synthesised and exported, and (iv) concurrent measurement of protein synthesis and energy utilization by the small intestine.