This study was carried out with 240 indigenous day-old chickens comprising of mixed sex frizzle feathered, naked neck, normal feathered and Noiler raised to 16 weeks of age to predict the body weight of 4 stains of Nigerian local and improved chickens using multiple linear regression and allometry function/model fitted for body weight and linear body measurements of the 4 stains of chicken. Data generated was subjected to multiple linear regression, allometry analysis and prediction using SAS 9.2 (Version 2008). Result of Prediction using multiple linear regression and allometry functions fitted into body weight and linear body measurements of the four breeds of local chickens in this study from 4 to 16 weeks revealed that wing length, shank length, shank circumference, drumstick length, drumstick circumference, nose to shoulder length, shoulder to tail length, body length, height at withers and breast girth could be the best variables used to predict body weight of frizzle feathered, naked neck, normal feathered and noiler chickens.

Keywords : Prediction, Allometry, Fitted, Function, Multiple, Regression, Strain