PHENOTYPIC CHARACTERIZATION AND IDENTIFICATION OF FARMERS BREEDING OBJECTIVE TRAITS OF INDIGENOUS CHICKEN ECOTYPES AROUND DIRE DAWA, ETHIOPIA

Abstract

This study was conducted to phenotypically characterize and identify farmers’ breeding objective traits of indigenous chicken ecotypes around Dire Dawa. In total, 120 households participated in the interviews, which were conducted using a structured questionnaire and for phenotypic characterization, four hundred eighty individual chickens (163 female and 77 male from midland, 87 male and 153 female from lowland) were characterized under field conditions to describe qualitative and quantitative traits following standard chicken descriptors. Descriptive statistics (frequency procedure), Generalized Linear Model (GLM) procedures, and multivariate statistics were used to analyze the data. The result revealed that the mean flock size of chickens per household was 13.76 and 10.3 (ranging from 2-25) in lowland and midland agroecology, respectively. Farmers' primary reason for culling was poor productivity chickens followed by old age. Furthermore, plumage colour, sicken, and bad temperament of hens and cocks. Sale for income (1st), egg production (2nd), and home consumption (3rd) were the purposes for which farmers rear chickens. Source of income ranked (1st), home consumption (2nd), and hatching chickens (3rd) were purposes of egg production. The effective population size (Ne) estimated in lowland, and midland altitudes were 4.66, and 3.88, respectively whereas the rate of inbreeding per generation (∆F) was 0.10, and 0.12, respectively. Hens are predominantly selected on the bases of egg production, followed by mothering ability, hatchability, large body size, colour, and brooding ability while cocks are predominantly selected based on body size, plumage colour, comb type, disease tolerance, growth, and fertility. Most of the local chickens had single (77.29 %,) comb type followed by rose (9.79 %,) double, buttery 6.88%, and 6.4% overall proportion agroecology. The average body weight of local hens in lowland and midland ecotypes were1.29±0.02 and 1.26±0.02kg respectively, while the respective values for mature males were 1.55±0.02 and 1.62±0.03 kg. Morph metric measurements indicated that significant differences (P<0.05) were observed between agroecology concerning shank length, comb length, wattle length, neck length, keel length, and comb height. In all parameters, male shows a higher significance (P<0.05) value than female xv chicken due to the hormonal effect. The degree of linear association among the variables measured by the Pearson correlation coefficient (r) and their statistical significance of r ranged from 0.13 to 0.80 for females and from 0.15 to 0.85 for males. The multivariate analysis result showed that two PC were extracted which accounted for 92 % of the total variation. All squared Mahalanobis’ distances obtained between ecotypes for the female and male populations were significant (p<0.05), In the male, the shortest distance (0.46) was observed between the lowland and midland district while in the female the longest distance (4.03) was observed between the lowland and midland agroecology.This shows that female ecotypes have distinct and measurable group differences across the district. Quantitative variables varied between sex groups and correct classification percentages were calculated separately for female and male sample populations. In the case of females, the overall average error count estimate was 14.56 % for all observations from all sites, which means that 85.44 % of the samples were correctly classified. The error count estimate for male populations was lower than for female populations (17.07%) having an overall average hit ratio of 82.13%.The correct classification ranged from 83.03 to 87.73% in the case of the female population and 79.22 to 86.21 % in the case of the male population. In contrast from midland, the highest correct classification percentages were calculated for lowland in both sexes, indicating that the sample populations from the lowland were more homogeneous on the quantitative variables as can be observed from their respective high hit ratios, while birds from midland were more heterogeneous. In conclusion, there is the diversity of indigenous chicken population and farmers’ preference for specific traits that may invite to design community-based genetic improvement.