Prediction of body weights from linear body parameters of rabbit genotypes in the humid tropics

This study aimed to determine the association between body weight and linear body parameters in rabbits. A total of 93 kittens generated from crosses of New Zealand White (NZW) x NZW, Chinchilla (CHA) x CHA, NZW x CHA, and CHA x NZW genotypes of rabbit in a nested classification of a Completely Randomized Design were used for this study. Data on body circumference weight (BW) and linear body parameters namely, body length (BL), ear length (EL), heart girth (HG), head-to-shoulder (HS), length of hind limb (LHB), length of forelimb (LFB), tail length (TL) and thigh girth (TG) were taken bi-weekly from weeks 2 to 12. The regression procedure used was a stepwise multiple regression using the backward elimination method. The body weight was regressed on linear body parameters using the stepwise multiple regression procedure in SPSS software. The result showed that the regressions of body weight on linear body parameters are all positive and highly significant (p<0.01; 0.05) except for CHA x CHA in week 2. The values of the coefficient of determination (R2) ranged from 26 - 91% in NZW x NZW, 85 - 98% in NZW x CHA, 1 - 88% in CHA x CHA, and 92 - 99% in CHA x NZW. CHA x NZW had the highest R2 value of 94, 99, 94, 97, and 96% in weeks 2, 4, 6, 8 and 12, respectively. NZW x CHA also had a similar R2 value of 94, 96, and 98% in weeks 2, 8 and 10, respectively. For predicting body weight at weeks 2 and 4, LFB and TG were good predictors of body weight for NZW x NZW mating type while at weeks 6 and 10, LFB was a good predictor of body weight. Also, BL and HS proved to be the best predictors of body weight for NZW x CHA mating type. For CHA x CHA, HG and TG were good predictors of body weight while in CHA x NZW, BL was a good predictor of body weight at these periods except for weeks 10 and 12. This suggests that body weight increases with an increase in linear body measurements. Except CHA x CHA in weeks 2 and 12, all the multiple regression equations were significant; all the parameters had R2 values above 60% which shows that the linear model was a perfect fit. Hence, BL can be predicted with accuracy using CHA x NZW cross to facilitate the selection of rabbits for body weight-related purposes.