An experiment was conducted to investigate the metabolism of condensed tannin (CT) in sheep and goats offered a mixture of Digitaria decumbens (700 g/kg) and Desmodium intortum (300 g/kg) hay. Radioactive 14CO2 was used to label CT in young growing desmodium plants, [14C]CT was extracted, purified and infused intraruminally, and the metabolism of [14C]CT was followed in the rumen and lower digestive tract of both species. Digestion of DM, organic matter (OM), cell-wall constituents (CWC), N and the efficiency of rumen microbial synthesis were determined using a continuous intraruminal infusion of 51Cr EDTA, YbCl3 and Na235SO4. The measurements taken for sheep and goats respectively were: intake, 21 and 30 g/kg0.9 per d; digestibilities (g/g) of DM, 0.566 and 0.505; OM 0.578 and 0.508; neutral-detergent fibre, 0.584 and 0.532; and acid-detergent fibre, 0.535 and 0.435. None of these measurements was significantly different (P > 0.05) between animal species. There was an apparent net gain in Lignin across the rumen and whole intestinal tract for both animal species (19 and 29% for sheep and goats respectively). There were no significant differences between sheep and goats (P > 0.05) detected for any measurements of N excretion and utilization. The overall efficiency of N digestion and utilization was also similar between species. The routes of CT metabolism mere compared for both colorimetric estimates (butanol-HCl) of dietary CT (DCT) and the specific radioactivity of [14C]CT in digesta (abomasum) and excreta (urine and faeces) of both sheep and goats. [14C]CT showed total losses of 57 and 56% in sheep and goats respectively whilst losses of DCT of 71 and 70% were detected with butanol-HCl in sheep and goats respectively. The apparent losses of DCT across the rumen of sheep and goats were 12 and 9% whilst higher losses (49 and 42% for sheep and goats respectively) were observed for [14C]CT. Losses of DCT in the lower intestinal tract accounted for 69 and 71% of the total CT leaving the abomasum. By comparison, only 40 and 35% of CT was lost during intestinal passage in sheep and goats respectively. It was concluded that the infused free [14C]CT interacted with DCT and entered the protein and fibre-bound DCT pools. The loss of DCT during passage through the intestines was considered to be a consequence of either absorption of free CT or the degradation products of CT. It was assumed that free CT arose in the lower gastrointestinal tract from protein-CT and fibre-CT dissociation to be digested and/or absorbed. The higher recoveries of [14C]CT in faeces (32 and 35%) compared with DCT (27 and 26%) for sheep and goats respectively) were associated with the excretion of [14C] degradation products or conjugates which were not reactive to butanol-HCl. It was concluded that both methods (butanol-HCl and labelling CT with 14C) detected a substantial disappearance of CT (free, protein, and fibre-bound) during metabolism in the gastrointestinal tract in sheep and goats.