Evaluation of growth performance of the Kenya dual Purpose goat composites from birth to yearling

Abstract

Growth data from 6,800 progeny of straight breds, intermediate crosses between Toggenburg (T), Anglo-Nubian (N), Small East African (E) and Galla (G) breeds in equal proportions and the resulting composites from 1982 to 2004 were used to evaluate growth performance from birth to yearling. The aim of the study was to assess the effects of genetic and non genetic factors on early growth performance as well as estimate the additive and non additive genetic parameters. A fixed effects least squares model (Harvey, 1990) was used to estimate the effects of non genetic factors. Pedigree Viewer programme of the Derivative-Free-Restricted Maximum Likelihood, DFREML, (Meyer, 1998) was used for pedigree analyses and calculation of inbreeding coefficient for individual animals. Additive and non additive genetic parameters were simultaneously estimated by fitting an animal model using DFREML algorithm (Meyer, 1998), in which linear coefficients for non additive effects were substituted for genetic group and fitted as regression covariates in the model. Two-breed crosses performed best in all traits while the straight breds had the least performance. Mean birth weight (BWT) for the two composites was the same (2.69 ± 0.49 and 2.69 ± 0.46 for composite 1 and 2 respectively). Mean weaning weight (WWT) for the composites was 9.89 ± 2.38 kg and 9.29 ± 2.17 kg with mean pre-weaning average daily gains (ADG1) of 55.51 ± 9.39 and 52.10 ± 9.13 g/day in composite 1 and 2 respectively. Mean yearling weights (YWT) were 17.32 kg, 18.67 kg and 17.85 kg with mean post-weaning average daily gains (ADG2) of 36.72, 40.42, and 35.88 g/day for the straight breds, two- and four-breed crosses respectively. Sex, type of birth, period and season of kidding, age of the dam and period and season of weaning significantly Xl affected early growth traits in varying degrees (P < 0.05 - P < 0.001) in the different genotypes. The mean proportion of inbred animals over the study period was 7.04 %, with an average inbreeding coefficient of 3.6 % and insignificant (P>0.05) annual change of - 0.001 ± 0.0001. The effect of inbreeding on the growth traits was not significantly different from zero (P>0.05). Heritability estimates ranged from 0.10± 0.02 to 0.24 ± 0.01 for post-weaning and pre-weaning average daily gains respectively. Phenotypic correlations between BWT and WWT and between BWT and ADG1 were equal and negative (-0.02 ± 0.001) while the respective genetic correlations were positive but very low (0.15 ± 0.04 for BWT and WWT and 0.001 ± 0.02 for BWT and ADG1, respecti vel y). Anglo-Nubian had the greatest effect on WWT (+1.69 kg), YWT (+1.74 kg) and ADG1 (+ 13.28 g/day). Between the indigenous breeds, Galla breed performed better in all traits except YWT. Effect of individual heterosis was greater in BWT (+0.05 kg) and ADG2 (+0.36 and +3.04 g/day for YWT and ADG2 respectively). Maternal heterotic effects were positive in pre-weaning traits (+0.04 kg and +0.39 gI day for WWT and ADG 1 respectively. The study establishes that early growth performance of KDPG composites is affected by environmental influences such as nutrition which should be appropriately adjusted for optimal performance. The effect of retained recombination loss has been reducing the performance of the composites hence the programme did not optimize on the potential benefits of crossbreeding.