Improvement strategies applicable to non-industrial trees with a clear focus on: end-use target group of beneficiaries, amount of planting material required, the biology of the species and the amount and structure of variation are discussed. Whichever strategy is used it is paramount that material is properly evaluated in trials and that diverse and productive germplasm is released. Cognizance of farmer's needs and criteria should be taken given that they are generally risk averse and try to stabilise production rather than solely maximise production. The consequences of release of planting stock need to be considered in relation to whether improvement will be recurrent or a single event. The establishment of breeding seedling orchards that perform the combined functions of resource population, breeding population, progeny test and seed production area would appear to offer distinct advantages. Decision strategies for improvement of non-industrial trees are discussed in relation to the features outlined.
Simons, A.J. Genetic improvement of non-industrial trees. Agroforestry Systems (1992) 18 (3) 197-212. [DOI: 10.1007/BF00123317]