The study was conducted to predict body weight from linear body measurements of indigenous non-descript chickens using principal component analysis (PCA). Linear body measurements for body weight (BW), body length (BoL), back length (BaL), neck length (NL), breast circumference (BrC), shank length (ShL), shank circumference (ShC), and toe length (TL) were taken from 360 indigenous non-descript hens. Phenotypic correlations among linear body measurements were all positive and significant (P <0.01). Body weight and BrC had the strongest correlation (P <0.01) coefficient of 0.71. Two principal components were extracted with a total variance of 63.9%. Principal component one (PC1) had the largest share of BrC (0.84), BW (0.83), and BoL (0.76) and ShC (0.62) in the order. Then principal component two (PC2) had high loadings in the order of TL (0.85), ShL (0.79), and BoL (0.62). The principal components based on a linear regression model could be a preferable tool to predict body weight of indigenous chickens raised under extensive production system. In conclusion, linear traits such as BrC, BoL and ShC could be useful tools to estimVilakazi BN_2020 Vol 13 (1)ate body weight of chickens.