A comparison has been made of the kinetics of urea metabolism in sheep given conventional feeds or maintained wholly by intragastric infusions of volatile fatty acids, buffer, mineral and casein solutions. Daily nitrogen supply was 13.2 g/day in 'fed' sheep and 6.0 g/day in 'infusion' sheep. On each feeding system measurements were made at the basal (maintenance) level of intake and when the basal level was supplemented with infusions of urea into the abomasum (125 mmol/day) or into the rumen (300 mmol/day). Additional measurements were made when ammonium carbonate (100 mmol/day) was infused into the rumen, when cassava (150 g/day) was added to the diet of 'fed' sheep or when the input of casein to 'infusion' sheep was reduced by half. Urea kinetics were measured by means of a single intravenous injection of [14C]urea. Urea irreversible loss rate, urea pool size, urinary urea excretion, plasma urea concentrations and rumen ammonia concentrations were all significantly higher in sheep given conventional feeds than in those nourished by infusion but urea degradation in the gastrointestinal tract did not differ between the two methods of feeding. Regression analysis indicated only minor differences between the two feeding systems in the relationships between the various indices of urea metabolism. It is concluded that most of these findings can be attributed to differences in the quantity and nature of the nitrogen supplied in the basal diets and that the sheep nourished by infusion would be a suitable model for the study of factors involved in the control of urea recycling.