1. Four steers with simple rumen and abomasal cannulas were given diets consisting of ground and pelleted alkali-treated straw, rolled barley and tapioca supplemented with urea (diet U) or containing single-cell protein (diet SCP), maize-gluten meal (diet MGM) or rapeseed meal (diet RSM) in place of some of the tapioca. The isoenergetic diets were given in a 4 x 4 Latin square design in eight feeds/d at 3-h intervals and provided sufficient metabolizable energy to support a growth rate of approximately 0.5 kg/d. Chromic oxide and polyethylene glycol were given as markers and appropriate samples taken from the rumen and abomasum. Flows (g/d) at the abomasum of organic matter and nitrogenous and carbohydrate constituents were calculated. 2. Rumen ammonia levels were similar with all three protein supplements at about 9 mmol/l, which was significantly lower (P less than 0.05) than that in animals on diet U (16 mmol/l). Rumen liquid outflow rates (/h) were 0.099, 0.139, 0.125 and 0.160 for diets U, SCP, MGM and RSM respectively; the difference between diet U and diet RSM was significant (P less than 0.05). Corresponding values for Cr2O3 outflow rates were 0.027, 0.032, 0.027 and 0.030/h respectively, which did not differ significantly from each other. 3. RNA, 35S and diaminopimelic acid (DAP) were used as microbial markers. Efficiencies of microbial-N (MN) synthesis, expressed as g MN/kg apparently digestible organic matter, truly digestible organic matter or carbohydrate fermented, were generally not significantly affected by the diet and averaged 29, 22 and 29 respectively based on mean RNA and 35S markers. Corresponding values derived from DAP of 22, 16 and 21 g MN/kg respectively were all significantly (P less than 0.001) lower. Using 35S as microbial marker, MN flows at the abomasum as a proportion of non-ammonia-nitrogen flow were 0.78, 0.64, 0.51 and 0.78 for diets U, SCP, MGM and RSM respectively. Derived true rumen degradability values (g/g intake) of the total dietary N were 0.91, 0.79, 0.69 and 0.90 for diets U, SCP, MGM and RSM respectively. Protein supplement degradabilities for single-cell protein, maize-gluten meal and rapeseed meal were 0.73, 0.51 and 0.98 respectively. 4. Mouth-to-abomasum digestibility coefficients of the main neutral-sugar components of dietary polysaccharides were 0.68, 0.63 and 0.61 for arabinose, xylose and cellulose-glucose on diet U.(ABSTRACT TRUNCATED AT 400 WORDS)