Analysis of Alternative Pure-breeding Structures for Sheep in Smallholder and Pastoral Production Circumstances in the Tropics

The key issue in this study was to technically compare, through stochastic simulation, different breeding programmes that vary in the level of interaction between breeders and producers. The breeding structures considered were: (i) a single closed nucleus providing seed-stock to village flocks, (ii) a group of commercial flocks running a co-operative (‘ram circle’) breeding programme with no nucleus, (iii) an interactive two-tier open nucleus breeding scheme, comprising a nucleus and a commercial tier - the best males are used within the nucleus while the remainder migrate to the commercial flocks, with no female migration, and (iv) as scheme iii but with female migration between tiers. For the latter two schemes, 100% of the nucleus animals are distributed over village flocks every 3 years. The nucleus is then replaced by a new batch of selected males and females from the village flocks obtained through ‘interactive cycling screening’, based on ‘picking the best phenotype’ in the commercial flocks. Single trait selection was considered, and based on estimated breeding value, using either best linear unbiased prediction (BLUP) or the individual’s phenotype as a deviation from contemporaries in the same flock, year and season. The results showed that genetic merit increased slightly and inbreeding decreased significantly with increase in nucleus size. For instance, with BLUP selection and trait measurement on both sexes, and first record established at year 2, a nucleus size of 100 dams with 50 dams mated to each sire resulted in genetic merit of 0.118 units and an average inbreeding coefficient of 0.119 while that with 500 dams gave a response of 0.134 with an average inbreeding coefficient of 0.037. Running one closed nucleus had a 6-24% advantage over a ‘ram circle’ in terms of genetic gain. Decreasing the dam to sire ratio was a simple way to avoid inbreeding in breeding schemes of small size, with very little compromise towards genetic gain or even an increase in the longer term. Relative to a two-tier nucleus (scheme i), cyclic screening of commercial animals for use in the nucleus gave an almost optimum genetic response, while the villagers acquired superior breeding stock in return as an incentive to participate in genetic improvement. Participation of farmers offers them a sense of ownership of the breeding programme, and is likely to make it more sustainable in the long-term. This study provides insight into the advantages and disadvantages of designed breeding structures, especially the ‘interactive cyclic’ breeding schemes, which should be useful in deciding breeding programmes to adopt for sheep in developing countries in the tropics. Keywords: Sheep, Breeding Structures, Selection, Tropic