The suitability of diminishing return functions such as the monomolecular and rectangular hyperbola equations, specifically re-parameterized for analysing energy balance data, has been investigated in ruminants, pigs, broilers and turkeys. In the current study, these equations were applied to a total of six time course profiles with male and female broiler breeder pullets (Ross 308 between 1 and 21 weeks of age, n=63 for each sex) in order to provide estimates for energy and protein requirements for maintenance and growth. Non-linear regression was used to estimate parameters and combine them to determine other biological indicators. All statistical computations were implemented in SAS (SAS 2000) by means of the NLMIXED procedure. The best-performing model was identified based on model behaviour when fitting the data, biologically meaningful parameter estimates and statistical performance. In spite of similarity between the biological indicators calculated for each sex using parameter estimates of the two models, quantitative examination of their predictive ability by error measurement indices showed the monomolecular model was superior. The metabolizable energy (ME) and crude protein (CP) requirements for maintenance ranged from 365 to 432 kJ/kg live weight (LW)/d and from 4·3 to 4·9 g/kg LW/d, respectively. The average ME and CP utilization ranged from 11·30 to 12·10 kJ/g body weight (BW) gain and from 0·61 to 0·64 g/g CP intake, respectively. The results obtained using the models were within the range reported by different researchers. Utilization of ME and CP for gain was more efficient at low levels of energy and protein intake, and gradually decreased as the level of energy and protein intakes increased. In conclusion, the models described herein were considered advantageous because they could predict the magnitude and direction of the responses of growing broiler breeder pullets to dietary ME and CP intake, which can be used in choosing and developing special feeding programs to decrease production costs.
Email your librarian or administrator to recommend adding this journal to your organisation's collection.
* Views captured on Cambridge Core between September 2016 - 22nd June 2017. This data will be updated every 24 hours.