Creating Opportunities for Domesticating and Commercializing Miombo Indigenous Fruit Trees in Southern Africa - Indigenous Fruit Trees in the Tropics: Domestication, Utilization and Commercialization - page 150

generation lines. The extent to which domesticated trees differ from their wild
progenitors depends on population size, the heritability of the desired trait under
selection, the mating system, the intensity of selection and the inherent
variability of the traits (Nyland, 1996; Simons, 1996; Cornelius et al ., 2006).
Domestication requires not just breeding, but also selection and management.
Simons (1996) has asserted that qualitative traits, such as fruit size, shape and
taste and tree form and precocity, are more strongly inherited from the selected
mother trees, such that 60% of the progeny might be similar to their parents.
Therefore cloning, after identifying and screening for a large number of superior
traits, is the best way to capture such genetic variation because it eliminates the
recombination or segregation of genes. This is the basis of participatory clonal
selection in the domestication programme of ICRAF.
Hamisy (2004) assessed the efficiency of tree selection using the technique of
random amplification of polymorphic DNA (RAPD). Two populations were used,
and data on 109 RAPD loci in 181 Uapaca kirkiana individuals showed a lack of
clear differentiation between populations and subpopulations, showing a high
level of genetic identity and thus indicating close relationships and the possibility
of gene flow between the populations. Farmer-mediated short-, medium- and
long-distance seed movements among populations may also explain the lack of
distinct genetic subdivisions. A similar, low amount of genetic variation was
observed in peach palm in Brazil (Cornelius et al ., 2006). However, Hamisy's
study confirmed that the elite trees selected from the wild using the participatory
approach showed more diversity than their wild counterpart sources and their
progenies. This was attributed to the extensive sampling used in the phenotypic
selection from natural forests, communal lands, homesteads and farms, which
enabled the capture of superior trees from wild populations (Akinnifesi et al .,
2006a) (Fig. 8.1). Large amounts of diversity between mother trees and their
progenies indicate the potential for using vegetative propagation techniques for
improvement to ensure genetic similarity as well as genetic diversity.
In practice, the use of seeds collected from isolated trees on the farm may
cause inbreeding (through parent-offspring or half-sib mating), hence the need
for proper selection guidelines for participatory domestication. A
comprehensive framework for breeding programmes has been described which
distinguishes four conceptually distinct breeding population components. First,
the 'base population' is chosen from which individuals are selected to be
carried forward to the next generation (Cornelius et al ., 2006). Secondly, the
'selected population' is the subset of the initial population that is to be carried
forward to the next generation. Thirdly, the 'breeding population' is composed
of trees that are used to produce the next generation, and may consist of all of
the selected population. Fourthly, the 'production population' is composed of
the trees used to produce propagules for commercial planting, or on-farm
planting in this case.
Cornelius et al . (2006) has articulated several ways in which these breeding
programmes could lead to genetic erosion and loss of genetic diversity: (i)
Because genetic variation depends on population size, intensive selection in the
breeding programme will lead to loss of diversity in the base population, and
possibly to cumulative genetic erosion if repeated in succeeding generations.