Germplasm Supply, Propagation and Nursery Management of Miombo Fruit Trees - Indigenous Fruit Trees in the Tropics: Domestication, Utilization and Commercialization - page 343

(asexual) propagation has shown to be most relevant to the objective of
germplasm improvement and domestication of miombo indigenous fruit trees for
various reasons. The first reason is the need to achieve desirable fruit and tree
traits in a reasonable time, such as superior fruit size, improved taste and increased
pulp content (Akinnifesi et al ., Chapters 8 and 21, Tchoundjeu et al ., Chapter 9,
this volume). This involves the development of cultivars with high productivity,
high harvest index, superior quality and improved food value and acceptability
(Leakey et al ., 2005; Akinnifesi et al ., 2006). The second reason is the need to
obtain fruits and products through improved precocity, i.e. a shorter period before
first fruiting. In a priority-setting exercise in the region, farmers indicated the desire
to cultivate indigenous fruit trees with early-bearing traits and superior fruit traits
(Maghembe et al. , 1998). The third reason is the need to overcome the natural
variability that could affect fruit quality and acceptability to consumers and
markets. It is also necessary to devise innovative ways of meeting the increasing
demand for quality planting material by farmers and the development community.
Tree germplasm is unequivocally the primordial input to agroforestry
(Simons, 1996). Early efforts to domesticate trees have followed the example of
the forestry approach based on seeds, in which the emphasis is on timber. This
problem is particularly apparent with miombo fruit trees, which are little known
in the commercial sector. Therefore, adequate seed is simply not available. Even
where collections exist of some species there is general under-representation of
the genetic diversity.
19.2.1 Germplasm collection
Germplasm collection is a key step in any process to domesticate trees.
Germplasm collection aims at capturing the natural genetic variation in a priority
species to enable selection and improvement in their ultimate domestication
(Dawson and Were, 1997). In any germplasm collection strategy it is important
to standardize collection practices, to sample as much variation as possible and
to prevent introducing biases in the evaluation stage. Most agroforestry trees are
grown from seed, while fruit trees are often propagated vegetatively. In collecting
germplasm of miombo fruit tree species, different strategies have been employed
or recommended (Ndugu et al. , 1995; Dawson and Were, 1997; Mulawarman
et al. , 2003). It is recommended that germplasm collection should pay attention
to the following: (i) trees should have abundant fruits (at least 500) to provide
sufficient seed for multilocation trials; (ii) farmers should determine which trees
are superior; (iii) at least 25 trees should be sampled per provenance; (iv) trees
should be separated by at least 50 m to avoid coancestry; and (v) seed should
be kept separate by using the open-pollinated family system (Kwesiga et al. ,
2000). In 1995, ICRAF trained a total of 30 germplasm collectors from
Botswana, Malawi, Mozambique, Namibia, Swaziland, Tanzania, Zambia and
Zimbabwe using funds received through the Southern Africa Development
Community (SADC), Southern Tree Seed Centre Network (STSCN), and
facilitated the collection and exchange of over 10 t of germplasm of Uapaca
kirkiana and Sclerocarya birrea provenances (Kwesiga et al. , 2000).