Metabolisable energy content of diets and broiler performance as affected by wheat specific weight and enzyme supplementation.

1. Two experiments were performed with male, Ross broiler chicks to study the relationship between wheat specific weight and performance parameters and the effects of xylanase inclusion in diets of different wheat specific weight (SW). 2. In both studies wheat was incorporated at 667 g/kg and a protein concentrate based on soyameal 50, full fat soyameal, extrupro and herring meal and formulated to provide (g/kg) crude protein 370, crude oil 140, crude fibre 45, crude ash 110 was incorporated at 333 g/kg. 3. In Experiment 1, 3 wheats (69, 67, 57, SW respectively) without or with enzyme (Avizyme TX, Finnfeeds International) and in Experiment 2, 6 wheats (81, 77, 76, 70, 64, 61 SW respectively) were used and diets were fed ad libitum to individually caged birds. 4. In Experiment 1 the feeding period was 7 to 28 d and a total excreta collection was made from 14 to 20 d for determination of apparent metabolisable energy (AME) content. Experiment 2 started at 4 d with excreta collections from d 6 to 9, 10 to 12 and 13 to 19 inclusive for AME measurements. 5. In Experiment 1 there were no significant effects of wheat on dry matter (DM) intake but liveweight gain (LWG) and gain:food were significantly better for 67 SW than for 57 SW. Food enzyme inclusion improved LWG and gain:food but there was a significant wheat/enzyme interaction for gain:food with no improvement due to enzyme addition for 67 SW but a 5% improvement for 57 SW. There was no significant effect of wheat or enzyme addition on the measured AME contents of the diets. 6. In Experiment 2, although not statistically significant, there were large numerical differences in food intake between diets, with the lowest values occurring for 64 SW. For the complete period, wheat variety tended to affect LWG (P=0.057) with the best and worst results recorded for the 2 low SW wheats. Gain:food was significantly affected by wheat, the poorest values occurring with 64 SW and the best with 77 SW Diet AME content was also affected by wheat variety (P<0.01) but the low result obtained for 61 SW was not mirrored in the gain:food value. The lack of correlation between diet AME values and gain:food could not be attributed to differences in protein or amino acid content of the diets. 7. In Experiment 2 diet AME content increased from period 1.1 (6 to 9 d) to period 2 (13 to 20 d) by 7% but interpretation is hampered by the short adaptation period (2 d) prior to period 1.1. The AME values obtained in both studies during the 3rd week of age were towards the upper end of the range of values in the literature. 8. In Experiment 2 there was a good correlation (r2=0.95) between SW and wheat starch content and between SW and calculated wheat AME content (r2=0.40, P<0.001) but a poor correlation between SW and gain:food. There was also a high correlation between wheat AME content and starch content (r2=0.41, P<0.001). Taken together with the results of Experiment 1 it appeared that starch content is a better measure of wheat quality than specific weight.