Abstract
A 21-day experiment was conducted to study the effect of reference diet type and assay method on apparent metabolizable energy (AME) and nitrogen-corrected (AMEn) of soybean meal (SBM) and canola meal (CM). Broilers (n = 240) were allocated to 10 treatments with eight replicates/treatment and three birds/replicate. Treatments included corn-SBM or corn-CM reference diets (RD). To each RD, 300 or 450 g/kg of SBM or CM were added to make a total of eight test diets. For the difference method, AME of SBM and CM substituted at 300 g/kg in corn-CM RD gave greater AME values compared to inclusion in the corn-SBM RD. The AMEn of SBM increased with increase in inclusion level in the corn-CM RD but AMEn of CM decreased with increased inclusion level of CM in the corn-SBM RD. For the regression method, AME and AMEn of the test feedstuffs were greater with corn-CM RD compared with corn-SBM RD. The AME of SBM was not affected by assay method, whereas AME of CM was lower when determined using the regression method. In conclusion, both the reference diet type and assay method influenced assayed AME and AMEn value of test protein feedstuffs and should be considered in cross-studies comparisons.