Statistical Modeling of Body Weight and Body Dimensions in Four Strains of Meat-type Chickens.

Abstract

This study was conducted to determine relationship between carcass traits and linear body parameters and establish equations for predicting carcass traits using linear body parameters in four strains of meat-type chickens. A total of four hundred (400) day old chicks of one hundred (100) each of Cobb 500, Marshall, Arbor acre and Noiler strains were managed intensively on deep liter system at the Poultry Unit of the Livestock Teaching and Research Farm, Department of Animal Science, University of Maiduguri, Borno State. Data on shank length (cm) shank circumference (cm), keel length (cm) , thigh length (cm), back length (cm), chest circumference (cm) and drumstick length (cm) were taken on weekly basis up to nine (9) weeks of age using tailor's measuring tape. Carcass, thigh, drumstick, wing and breast weights were the carcass traits examined using weighing scale in (g) Data were analyzed using SAS(2000). The correlation among carcass traits and linear body parameters were determined using same software. Data were also subjected to simple and multiple linear regression analyses. The better relationship was determined between thigh weight and shank length in Cobb 500 (R2 =0.87), carcass weight and back length (R2 =0.86) in Marshall, breast weight and thigh length (R2 =0.72) in Arbor acre and carcass weight and shank length (R2 =0.70)in Noiler. The predictive equations showed that there were significant (P<0.01) relationships between body weight and linear body parameters. High R2values recorded in this study show that the predictive equations could be used to predict carcass traits with high degree of accuracy, with the multiple regression models being more efficient than the simple regression model.