Peach palm - World Agroforestry Centre

3 0 Peach palm. Bactris gasipaes Kunth
frequency), 0.141 ± 0.035 observed mean heterozygosity, 0.191 + 0.047 expected
heterozygosity under Hardy-Weinberg equilibrium. The observed heterozygosity
is lower than values reported for other palms, and may be partially explained by
the history of the germplasm that was analyzed in Hawaii (limited genetic base).
Although sample size was adequate for genetic analysis, larger samples and more
systematic analyses are necessary to elucidate the relationships among landraces.
In addition, the inheritance of isozyme phenotypes should be determined.
Morphological, chemical, isozyme and DNA differences may be useful for
identifying plants that, when crossed, would produce more variable progenies,
possibly resulting in heterotic effects on growth and yield. However, the lack of
significant correlation between isozyme heterozygosity and other traits, if
generally true, suggests that expectations may be modest (Clement 1995a).
6.2 Existing germplasm banks
Several ex situ field germplasm banks were established in the early 1950s, but not
all of them were maintained. Collecting and genebank establishment started at
Centro Agronómico Tropical de Investigación y Enseñanza (CATIE), Turrialba,
Costa Rica in the 1960s. In the late 1970s, considerable efforts were made to
enlarge the collection. Since the 1970s, interest in peach palm has increased, bringing
financing for international germplasm collections, and stimulating a more sustained
interest in peach palm among participating countries. Germplasm has been collected in
Bolivia, Brazil, Colombia, Costa Rica, Ecuador, Nicaragua, Panama, Peru and Venezuela.
Some collections followed a targeted sampling strategy, attempting to select
phenotypically superior individuals in the field, while most collections included both
targeted phenotypes and randomly selected individuals. Although significant, these
collections represent only a small area in each country where cultivated peach palm is
thought to occur, with extensive areas yet to be explored and collected.
There are now germplasm banks at experimental stations in Brazil, Colombia, Costa
Rica, Ecuador, Nicaragua, Panama, Peru and Venezuela (Table 6). Most of these are located
within the natural range of cultivated peach palm (except those in São Paulo and Bahia,
Brazil and Tovar, Venezuela). They are not replicated experimentally, and accessions
are generally represented by only nine plants (3 to >30 in some cases), with
interplant spacing of 5 m. Most of them have not been characterized and evaluated
and, although most of the germplasm banks are being maintained, many accessions
are being lost owing to decreasing plant vigour and offshoot production.
Nearly all of these germplasm banks have both local and international
accessions, and some accessions are present in several banks. In 1983-84, the US
Agency for International Development funded extensive germplasm collecting in
Brazil, Colombia, Ecuador and Peru (Clement and Coradin 1988), organized by two
Brazilian research institutes (Instituto Nacional de Pesquisas a Amazônia, Centro
Nacional de Recursos Genéticos e Biotecnologia). The collections included both
targeted phenotypes and randomly selected individuals in numerous populations