Effect of kernel density on the apparent and true metabolizable energy value of corn for chickens.

Five samples of corn selected to vary widely in kernel density (test weight per unit volume) were assayed for AMEn with male broiler chickens at 4 wk of age by regression analysis of a multilevel assay and for TMEn with adult White Leghorn roosters. The kernel densities (kilograms per hectoliter), AMEn, and TMEn (kilocalories per gram of DM) values of the corn were, respectively, 72, 3.68, and 3.96; 71, 3.72, and 3.95; 68, 3.66, and 3.90; 62, 3.64, and 3.88; 60, 3.54, and 3.68. Positive relationships were established between AMEn and TMEn (r = .974, P less than .01) or kernel density (r = .875, P less than .05). The assay for TMEn provided a simple, rapid and sensitive method for estimating the AMEn content of corn for chickens. The small (4%) variation in AMEn relative to a large (20%) variation in kernel density precludes the use of kernel density for estimating the AMEn content of corn.