Comparison of three methods of excreta collection used in estimation of energy and nitrogen excretion

Three methods of excreta collection: plastic trays (T), human colostomy bags affixed to the bird (B), and B containing 25 ml .5 N HCl (A) were compared in three experiments with adult, Single Comb White Leghorn cockerels. There were 128, 130, and 120 birds in Experiments 1, 2, and 3, respectively, from which 97, 84, and 105 excreta samples were obtained; most of the losses occurred among the B- and A-groups. Birds were fasted for 24 hr prior to the experiments and then precision-fed, by intubation, with 0, 5, 10, 15, 20, 25, 30, or 35 g of fish meal (Exp. 1) or soybean meal (Exp. 2). In the third experiment, the inputs were 30 g of corn, fish meal, soybean meal, and a 50:50 (wt/wt) blend of fish meal and soybean meal; control birds were fasted. Excreta were collected for 48 hr after feeding, and the outputs of excreta energy (FE + UE), excreta nitrogen (FN + UN), and (FE + UE) corrected to zero nitrogen balance (FEn + UEn) were measured. True metabolizable energy (TME) values and TME corrected to zero nitrogen balance (TMEn) were calculated. In Experiments 1 and 2, the regressions of (FE + UE), (FN + UN), and (FEn + UEn) on feed input had slopes which were independent of excreta collection methods (P greater than .05), indicating that estimates of TME and TMEn based upon the linear regression assay were not different (P greater than .05). In both experiments, the levels of excreta and energy output using T were greater (P less than .01) than when using B.