1. An experiment was conducted with laying hens to determine to what extent sweet white lupins (Lupinus albus, cv Buttercup) could be used as a protein source in their diets. 2. Twenty five individually caged 20 week-old Hisex laying hens were used per treatment. The hens were housed in a convection house for the duration of the study which lasted 52 weeks. Two diets were formulated, one containing 300 g sweet lupins/kg and the other containing sunflower oil cake and fish meal as protein sources. These diets were blended to obtain seven diets with lupin inclusion rates of 0, 50, 100, 150, 200, 250 and 300 g/k. 3. Results indicated that up to 300 g lupins/kg diet can be fed to Hisex laying hens without a significant effect on egg production, egg mass, efficiency of food utilisation, egg shell thickness, Haugh units or yolk colour. Food intake was however, significantly (P<0.05) positively correlated with lupin inclusion rates.

Four rumen-fistulated Holstein cows were used to determine the ability of vermiculite to alter rumen fermentation, rumen fluid dilution rate, milk production, and nutrient utilization in a 4 x 4 Latin square experiment. Treatments consisted of 1) basal diet (70% concentrate:30% Eragrostis curvula hay), 2) basal + .6% NaHCO3 + 1.8% vermiculite, 3) basal + 1.2% NaHCO3, and 4) basal + 3.6% vermiculite. Feed intake and milk production were not affected, but the NaHCO3 treatments tended to increase milk fat production. The 1.2% NaHCO3 treatment increased rumen pH and fluid dilution rate, decreased molar percent propionate, and increased acetate:propionate ratio. Rumen NH3 N was not affected. Milk yield (kg/d) and milk fat (%) for the treatments were 1) 19.3, 3.33; 2) 19.4, 3.59; 3) 19.8, 3.62; and 4) 18.7, 3.32. Rumen fluid pH, dilution rate (%/h), and acetate:propionate ratio were 1) 5.75, 10.9, 2.08; 2) 5.89, 11.5, 2.16; 3) 5.95, 12.1, 2.38; and 4) 5.80, 11.6 and 1.82. Treatments had little effect on nutrient digestibilities and serum mineral concentrations. Vermiculite did not show any promise as a buffer but NaHCO3 proved to be an effective buffer in diets based on Eragrostis hay and 70% concentrate diets.

The true digestibility of protein from diets containing one of the protein-rich foodstuffs, sunflower oil cake, cottonseed oil cake, maize gluten meal and fish meal, was determined in chickens and in the small intestine of sheep. With the exception of maize gluten meal, a close correlation was found between digestibility in the two species (r = +0·98; P < 0·05) and in the sheep small intestine between the true digestibility of utilizable nitrogen and acid-detergent insoluble nitrogen content of the diet (r = -0·96; P < 0·05).