Two experiments were carried out using three wethers each. Sheep were fitted with rumen and re-entrant duodenal cannulas. In Experiment 1, diets included 10, 30 and 50% ground straw and barley plus sunflower oil meal in the ratio 9:1. Straw was used as a diluting substance in order to obtain diets with similar net energy contents but different dry matter levels and thus different ruminal dilution rates. In Experiment 2, diets contained ground barley and dehydrated alfalfa in the ratio 2:1 plus 0, 24 and 32% polyvinylchloride (PVC). PVC was used as ballast to increase dry matter intake (DMI) and ruminal dilution rates without changing the amount of nutrients in the diets. All diets were pelleted. Intake of net energy and nitrogen was quite similar for the three diets within each experiment. Dry matter intake, however, was different. Protozoa disappeared almost completely when sheep received the diet with 10% straw and all diets in Experiment 2 because of low pH values in the rumen. There were no significant differences in volatile fatty acids (VFA) and ammonia concentration in the rumen fluid. Rumen volume and liquid and solid dilution rate increased with increasing DMI. Organic matter digestibility in the forestomachs and in the whole digestive tract decreased linearly with increasing DMI and shifted from the reticulo-rumen to the intestine. It appears that the depression of digestibility was highly related to DMI despite the fact that the energy level did not change and this is likely to be a result of increased ruminal dilution rates. Microbial protein passing to the duodenum increased proportionally to DMI. The results of Experiment 2 showed clearly that the efficiency of microbial protein synthesis ( E mps) in the rumen increased with increases in the level of feeding. These results could not be affected by other factors such as the level of energy intake, concentrate: roughage ratio, energy and N sources because the intake of feed ingredients and nutrients for all three rations was quite similar. Comparison of the three microbial markers (diaminopimelic acid (DAPA), purine nitrogen, urine purine derivative excretion) showed that microbial protein entering the duodenum was underestimated by DAPA only when the number of protozoa in rumen fluid was high. The estimation of microbial protein reaching the duodenum on the basis of urinary purine derivatives excretion did not differ significantly from the estimation by purine N in duodenal digesta. Therefore, it is possible to avoid invasive cannulation technique without reducing the accuracy of estimation of microbial protein entering the small intestine.