Search

Agroforestry ecosystems may be an important resource for conservation and sustainable use of tropical trees, but little is known of the genetic diversity they contain. Inga edulis, a widespread indigenous fruit tree in South America, is used as a model to assess the maintenance of genetic diversity in five planted vs. five natural stands in the Peruvian Amazon. Analysis of five SSR (simple sequence repeat) loci indicated lower allelic variation in planted stands [mean corrected allelic richness 31.3 (planted) and 39.3 (natural), P = 0.009]. Concerns regarding genetic erosion in planted Amazonian tree stands appear valid, although allelic variation on-farm is still relatively high

Nine microsatellites were used to screen 131 samples of Barren Brome (Anisantha sterilis: synonym Bromus sterilis) collected from within the fields of three English farms [from Oxfordshire (Oxon), Leicestershire (Leics) and Wiltshire (Wilts)] and eight seeds taken from samples of each of 10 farms across England, UK. Most individuals (approximately 97%) were homozygous. Polymorphism occurred at all nine loci in all three farms sampled at the field scale, and at most loci for nine of the other 10 farm samples. Between three and 11 alleles were found per locus. Gene diversity (D = 1 - summation operator p(i)2) ranged from 0.088 to 0.760. Polymorphism occurred among individuals within and among fields, and farms. Some alleles were found in only one farm. On the basis of the alleles at all nine loci in the 211 sampled plants, a total of 92 (44%) different genotypes was identified. Clustering analysis using the unweighted pair group method with arithmetic averages (UPGMA) for the combined Oxon, Wilts and Leics samples did not cluster them into their respective farms. Similarly, a phenogram of samples from all 10 farms showed considerable mixing of individuals with respect to farm origins. Identification of genotypes on field plans showed evidence of both spatial localization and mixing. Previous reports have suggested that A. sterilis is strictly inbreeding with little intrapopulation variation at the genetic level. Our data reveal that A. sterilis exists as numerous separate and genetically different lines, which are maintained by inbreeding but which very occasionally outcross. Possible explanations for this pattern of high genetic diversity are discussed.

Genetic variation was assessed in the two bush mango species, Irvingia gabonensis and I. wombolu, valuable multipurpose fruit trees from central and west Africa that are currently undergoing domestication. A total of 130 individuals sampled from Cameroon, Nigeria and Gabon were analysed using 74 random amplified polymorphic DNAs (RAPDs). Significant genetic integrity was found in the two morphologically similar species (among-species analysis of molecular variance [AMOVA] variance component 25.8%, P < 0.001), with no evidence of hybridization, even between individuals from areas of sympatry where hybridization was considered probable. Results suggest that large-scale transplantation of either species into new habitats will probably not lead to genetic introgression from or into the other species. Therefore, subsequent cultivation of the two species should not be hindered by this consideration, although further studies on the potential for hybridization/introgression between these species would be prudent. Significant genetic differentiation of both species (among-countries within species, nested AMOVA variance component 9.8%, P < 0.001) was observed over the sampled regions, and genetic similarity of samples decreased significantly with increasing geographical distance, according to number of alleles in common (NAC) analysis. 'Hot spots' of genetic diversity were found clustered in southern Nigeria and southern Cameroon for I. wombolu, and in southern Nigeria, southern Cameroon and central Gabon for I. gabonensis. The possible reasons for this distribution of genetic variation are discussed, but it may reflect evolutionary history, as these populations occur in areas of postulated Pleistocene refugia. The application of these results to domestication programmes and, in the light of extensive deforestation in the region, conservation approaches, is discussed.

An understanding of the processes that determine the observed patterns of genetic variation in natural plant populations is an important factor in the management of biodiversity. Pollen-mediated gene dispersal is recognized as a major determinant of population genetic structure. Here, the utility of simple sequence repeat (SSR) analysis was investigated for the measurement of pollen-mediated gene transfer by paternity exclusion in a restricted, fragmented and endangered population of the insect-pollinated tropical leguminous tree Gliricidia sepium located in Guatemala. Data at a single SSR locus, which revealed six allelic variants, were employed to generate minimum distance curves of pollen dispersal. Combined data from all six alleles indicated that a minimum of 1.8% of transfer events occurred over a distance of greater than 75 m. However, this value represents an underestimate because of the exclusion approach employed for analysis. Considering the four rarest alleles in the population only (combined frequency = 0.196), which provides a less biased indicator of gene transfer, a minimum of 6.1% of pollen movements could be attributed to greater than 75 m. One extreme example of gene transfer of over 275 m was recorded. Estimates of pollen transfer suggest a homogenizing effect on genetic structure over the spatial scale of the study population and provide an important indicator for the genetic management of natural and exotic stands of G. sepium. This study provides the first example of SSR analysis being employed to estimate directly pollen movement in a natural stand of any tree species.

Agroforestry ecosystems may be an important resource for conservation and sustainable use of tropical trees, but little is known of the genetic diversity they contain. Inga edulis, a widespread indigenous fruit tree in South America, is used as a model to assess the maintenance of genetic diversity in five planted vs. five natural stands in the Peruvian Amazon. Analysis of five SSR (simple sequence repeat) loci indicated lower allelic variation in planted stands [mean corrected allelic richness 31.3 (planted) and 39.3 (natural), P = 0.009]. Concerns regarding genetic erosion in planted Amazonian tree stands appear valid, although allelic variation on-farm is still relatively high.

Nine microsatellites were used to screen 131 samples of Barren Brome (Anisantha sterilis: synonym Bromus sterilis) collected from within the fields of three English farms [from Oxfordshire (Oxon), Leicestershire (Leics) and Wiltshire (Wilts)] and eight seeds taken from samples of each of 10 farms across England, UK. Most individuals (approximately 97%) were homozygous. Polymorphism occurred at all nine loci in all three farms sampled at the field scale, and at most loci for nine of the other 10 farm samples. Between three and 11 alleles were found per locus. Gene diversity (D = 1 - summation operator p(i)2) ranged from 0.088 to 0.760. Polymorphism occurred among individuals within and among fields, and farms. Some alleles were found in only one farm. On the basis of the alleles at all nine loci in the 211 sampled plants, a total of 92 (44%) different genotypes was identified. Clustering analysis using the unweighted pair group method with arithmetic averages (UPGMA) for the combined Oxon, Wilts and Leics samples did not cluster them into their respective farms. Similarly, a phenogram of samples from all 10 farms showed considerable mixing of individuals with respect to farm origins. Identification of genotypes on field plans showed evidence of both spatial localization and mixing. Previous reports have suggested that A. sterilis is strictly inbreeding with little intrapopulation variation at the genetic level. Our data reveal that A. sterilis exists as numerous separate and genetically different lines, which are maintained by inbreeding but which very occasionally outcross. Possible explanations for this pattern of high genetic diversity are discussed.

Gliricidia sepium provides a variety of products important for rural communities in tropical countries. Native populations in Meso-America currently form an important source of seed for distribution to farmers, but concerns centre on mechanisms which may lead to their genetic erosion, including anthropogenic dispersal and subsequent introgression from the related species, G. maculata. Populations of Gliricidia were examined genetically using approaches based on the polymerase chain reaction to test for interspecific hybridization and introgression between G. sepium and G. maculata. Analysis involved 13 RAPD and two RFLP-PCR markers which were identified to have species-diagnostic distributions. Data from both approaches corresponded and indicated three locations where multilocus genotypes were consistent with an hybrid origin. Data at one of these sites was consistent with introgression following hybridization. The hybrid origin of populations was supported by the intermediate geographical location of these sites to ‘pure’ populations of each species. Analysis of maternally inherited organellar DNA, which involved the detection of SSCPs in mitochondrial DNA amplification products, allowed further delineation of genetic structure among Gliricidia populations. Mitochondrial data indicated a high degree of organelle differentiation between sampled locations and identified G. sepium- and G. maculata-diagnostic haplotypes. This data supported the interpretation of genetic structure based on RAPDs and RFLP-PCR. In addition, cytonuclear analysis allowed the directionality of gene transfer during the formation of hybrid populations to be described. Despite evidence for the occurrence of interspecific hybridization and introgression in Gliricidia, important resource populations of G. sepium on the Pacific coast appear to have retained their genetic integrity. Implications in terms of the conservation and utilization of genetic resources within the genus are discussed.

Randomly amplified polymorphic DNA (RAPD) markers were used to analyse genetic diversity within and between Hordeum spontaneum populations sampled from Israel. Nei's index of genetic differentiation was used to partition diversity into within and between population components. Fifty-seven per cent of the variation detected was partitioned within 10 H. spontaneum populations. Using principal component and multiple regression analysis, part of the variation detected between populations was seen to be associated with certain ecogeographical factors. Fifty-eight per cent of the distribution of the phenotypic frequencies of three RAPD phenotypes detected using a single primer in 20 H. spontaneum populations could be accounted for by four ecogeographical variables, suggesting adaptive variation at certain RAPD loci.