The effect of acute (4.5 h) infusions into the mesenteric vein of an amino acid (AA) mixture, which simulated the composition of rumen microbial protein, on net transfers of NH3, urea and total AA across the portal-drained viscera (PDV) and liver in the ovine has been examined. Four wether sheep were surgically prepared with vascular catheters across the PDV and liver (Lobley et al. 1995) and were offered a basal diet of 1000 g grass pellets/d (approximately 1.4 × energy maintenance). Each animal was infused at weekly intervals with one of four dilutions of the AA mixture. These dilutions provided 0.44, 0.88, 1.32 and 1.84mmol AA-N/min infused, the lowest of which approximately doubled the net absorption of AA-N from the basal diet. Animals were treated with heparin to allow continuous collection of blood by peristaltic pump for 2 h preceding, and between 0.5–2.5 and 2.5–4.5 h after, the start of the AA infusions. Blood flow in the hepatic artery increased (100 v. 208 g/min; P = 0.002) in response to AA infusion, while hepatic portal venous flow decreased (2090 v. 1854 g/min; P = 0.006). The AA infusion also stimulated O2 uptake by the PDV (P < 0.001) and liver (P = 0.016). Absorption across the PDV and hepatic removal of NH3 were unchanged between basal and amino acid infusion conditions. Urea-N removal across the PDV was unaltered, but hepatic production increased (P < 0.001) with level of AA infusion. During infusions, net appearance of AA across the PDV was below the theoretical level. This may have been due to inhibition of AA uptake from the small intestine, and/or increased removal by the digestive tract of AA from the systemic circulation associated with greater arterial concentrations. Hepatic extraction of AA increased with level of infusion, both for total AA and those included in the infusate. Total hepatic urea-N production tended towards a maximum (estimated as 2 μmol N/g liver wet weight per min). The AA removed by the liver and not used for ureagenesis remained similar (170 μmol AA-N/min) between basal and AA infusions. This was presumed available for anabolic purposes (mainly synthesis of export proteins). The proportion of net AA-N appearance (absorption plus infused) across the PDV removed by the liver declined from 0.71 to 0.53 between basal and AA infusions. In contrast to findings from cattle (Wray-Cahen et al. 1997), increased AA infusion did not alter the net removal of glutamine across the liver. This may reflect differences between the studies in NH3: AA-N absorbed. Further differences between the cattle study and the current findings may relate to the different physiological state (pregnancy v. growth), which may alter the partition of AA between anabolic and catabolic fates.