The experiment consisted of a 2 × 2 × 2 factorial design testing the two D+ and D− chicken lines selected for divergent digestion efficiency (fifth selection generation), xylanase (with or without) and ampicillin and collistin (with or without) supplementations. From 8 to 22 days, 144 chickens (18 birds per treatment) were fed a diet containing 55% wheat from a high-viscosity cultivar (Rialto). Effects of treatments were evaluated on individual growth performance (8 to 19 days), digestibilities of lipids and dry matter, dietary energy value (apparent metabolisable energy corrected to zero-nitrogen retention (AMEn)), digestive organ and breast sizes, and intestinal bile acids at 3 weeks of age. Individual variabilities were much lower in D+ than in D− birds for feed : gain ratios, digestibilities and AMEn values. In all cases, feed : gain ratios were lower in the D+ than in the D− line (P < 0.001), and D+ birds showed 22% to 86% higher values than in D− birds (P < 0.001) for digestibilities and AMEn. In D− birds, antibiotics but not xylanase supplementation had significant effects on lipid digestibility (P < 0.01) and AMEn (P < 0.05), whereas both supplements improved these parameters in D+ birds (P < 0.001 for both additives on lipids digestibility, P < 0.05 for xylanase and P < 0.01 for antibiotics on AMEn). Relative weights of gizzard and proventriculus, and gizzard : intestine weight ratio were higher in D+ than in D− birds, while relative weight of intestine was increased in D− birds compared with D+ birds. Antibiotics reduced intestine relative weight in D+ (P < 0.001) and D− (P < 0.01) lines. AMEn variations were efficiently predicted by the gizzard : intestine weight ratio. In conclusion, antibiotics were very efficient for improving growth performance, AMEn and digestibility values in both chicken lines. Xylanase was less efficient than antibiotics. Because of their low individual variabilities, D+ birds were much more efficient than D− ones for the detection of significant effects induced by xylanase supplementation. Differences between lines in feed : gain ratio, digestibilities and AMEn were reduced when xylanase and antibiotics were added together. Effects of xylanase supplementation and animal genetics on lipid digestibility could not be entirely explained in terms of intestinal bile acids. Other factors should be involved, especially for the lipid digestibility difference induced by animal genetics. The gizzard : intestine weight ratio was an efficient parameter for predicting AMEn variations due to animal genetics and additives.

Two different cannulation procedures (simple ileal ‘T’ cannula v. The post valve ‘T’ caecal cannula (PVTC)) and two indigestible markers (TiO2 v. Cr2O3) were studied with six male littermate pigs fitted with PVTC or simple ileal ‘T’ cannulae. Six diets were used, of which two were based on wheat and wheat bran and the other four were based on two barleys of different bushel weight without and with exogenous enzymes (ß-glucanase/xylanase). Proportional TiO2 and Cr2O3 recoveries in faeces were less than 1·00, the mean values for the six diets being 0·858 for TiO2 and 0·811 for Cr2O3. With both markers, recovery in faeces was lowest for the most digestible wheat-based diet (A). The ileal apparent digestibility (IAD) coefficients of dry matter (DM), crude protein (CP), energy and amino acids measured with Cr2O3 were significantly (P < 0·001) lower than those measured with TiO2. There was no difference in overall apparent digestibility of DM, CP and energy measured with simple ileal ‘T’ cannula and PVTC techniques. However, IAD of DM, energy and CP measured with the PVTC method were significantly higher than those measured with the simple ileal ‘T’ cannula method. The data also showed that the standard errors with the simple ileal ‘T’ cannula method were greater than when using the PVTC method. Diet significantly affected ileal and overall digestibility of nutrients (P < 0·001) with values being highest for diet A and least for the wheat bran-based diet (B). Overall digestibility (OD) of DM and energy were higher for the higher bushel weight barley-based diet (C) than for the normal bushel weight barley-based diet (E). Enzyme inclusion improved OD for both barley diets and ileal digestibility of energy (0·060) and CP (0·057) for the normal bushel weight barley.

The efficacy of an enzyme preparation added to barley and rye-based diets given to broiler chickens and containing both β-glucanase and arabinoxylanase activities, was investigated and compared with a well characterized commercial enzyme preparation containing predominantly P-glucanase activity.

Both enzyme preparations significantly improved body weight, cumulative food intake and food conversion efficiency for chickens given both barley- and rye-based diets. For the barley-fed chickens, on average a notable increase in body weight of 171 g at day 13 and 477 g at day 24 was noted, for both enzyme preparations. However, the preparation containing high amounts of both β-glucanase and arabinoxylanase activities was more effective in reducing the incidence of sticky droppings for rye-fed chickens, and, in comparison with previously published data, gave an optimal response at a lower supplementation rate. The results indicate that it is possible to reduce the protein content in enzyme-supplemented broiler chicken diets.