Variation in nutrient content of feedingstuffs rich in protein and reassessment of the chemical method for metabolizable energy estimation for poultry

Abstract

SummaryA series of five metabolism trials was made to determine apparent nitrogen digestibility and metabolizable energy (ME) contents of protein rich feedingstuffs. The mean nitrogen digestibilities of fish meal, groundnut, mustard, sesame and cottonseed cakes were 66, 69, 68, 57 and 40%, respectively. Corresponding values for metabolizable energy values were 1820, 2460, 2330, 1870 and 1530 kcal/kg, respectively. The metabolizable energy contents of coconut cake, niger cake and blood meal were 1190, 2360 and 2190 kcal/kg, respectively. The quantity of protein, its digestibility and crude fibre content in the cakes are the prime factors for this trend in MB. Simple and multiple regression equations were derived from biologically assayed metabolizable energy and chemically analysed energy-yielding nutrient contents of the feedingstuffs. The simple regression equation is:ME kcal/kg = 32·95 (% crude protein + % ether extract × 2·25+ % available carbohydrate)–29·20.The multiple regression equation is:ME kcal/kg = 370·29 + (24·47 × % crude protein)+ (65·77 × % ether extract)+ (44·07 × % available carbohydrate)- (8·15 × % crude fibre).The correlation coefficients of simple and multiple regression equations were 0·72 and 0·73, respectively, indicating that there is very little advantage for prediction in using the multiple regression equation. The usefulness of the equation for routine checking of poultry feeds for ME is apparent since the nutrients required to predict metabolizable energy can be analysed within a short period of time.