H. S. Bayley

1. Six 45 kg pigs with re-entrant ileal cannulas were used in two 3 × 3 Latin-square design experiments to study the site of absorption of protein and amino acids. Semi-purified diets containing soya-bean meal (SBM), rapeseed meal (RSM) or no protein source (protein-free) were offered at the rate of 1 kg dry matter/d. 2. Flow-rates of ileal contents for 24 h collection periods, corrected for recovery of marker, were 3135, 3127 and 1243 ml (SE 390) for SBM, RSM and protein-free diets respectively. 3. Amounts of dry matter digested in the small intestine were 730, 669 and 809 g/d for SBM, RSM and protein-free diets respectively, all values being significantly different ( P < 0·001). 4. Nitrogen intakes were 32·6, 29·9 and 5·9 g/d, and amounts digested in the small intestine were 25·7, 20·2 and 1·6 g/d for SBM, RSM and protein-free diets respectively, all values being significantly different ( P < 0·001). Amounts digested in the large intestine were 2·6, 3·7 and 0·7 g/d. 5. Total amino acid intakes and amounts collected at the ileum and in the faeces were (g/d): SBM, 177, 24 and 18; RSM, 149, 28 and 22; protein-free 3, 9 and 12. Digestibility in the small intestine was higher for SBM than RSM for seventeen of the eighteen amino acids estimated. Greater quantities of arginine, methionine, cystine and tyrosine were voided in the faeces than passed through the ileal cannulas for pigs receiving the SBM and RSM diets. For those receiving the protein-free diet this was true for each amino acid except proline. 6. Significant differences were found between all diets in the concentration of some amino acids in ileal and faecal amino-N, and endogenous protein secretions did not mask the differences between diets. 7. Differences in digestibility between SBM and RSM were greater at the ileum than in the faeces. Amino acid fermentation in the large intestine obscured or reduced differences between SBM and RSM.

Growth rate, feed efficiency, and mortality were not adversely affected by reducing the level of herring meal in an open formula dry-pellet diet fed to rainbow trout (Salmo gairdneri) in aquaria through which 80–90% of the effluent water was recirculated. The herring meal was reduced from 35 to 18%, and the level of soybean meal increased from 10 to 39%. The fish gained 5.5 and 5.4 kg/100 fish from 8 to 40 wk of age on the high and low herring meal diets, respectively. Deletion of brewers' yeast, corn fermentation extractives, and whey powder from the diet with the higher level of herring meal increased weight gain to 7.0 kg/100 fish. When the fermentation by-products were removed from the diet with the lower level of fish meal the gain was reduced to 4.8 kg/100 fish. All the diets contained 4% of soybean or rapeseed oil or a marine oil; the type of oil did not influence the gains, but the conversion of feed to gain was most efficient for the diet containing the marine oil. The apparent digestibilities of the proximate components of three of the diets suggested that the fermentation by-products were not as digestible to the fish as the other components of the diet.