|
|
Assessment of genetic diversity of economically and ecologically important tropical tree species of Central America and the Carribean: Implications for conservation, sustainable utilization and managementProject coordinator: Julia Wilson
Materials and MethodsNatural populations of trees were located in Central America, the Antilles and French Guyane. Collections of leaf, cambium, seed and scion material were made for each species as appropriate to the study and the tree locations were mapped. Genebanks and provenance trials were set up as appropriate. Herbarium samples were prepared and seed was stored. DNA was extracted from plant material for diversity studies and techniques developed and applied as appropriate. A wide range of different techniques was used according to the objectives of each study.
Results and ConclusionsA. Swietenia, Cedrela and Vochysia1. Reliable techniques for RAPD evaluation of all these species have been developed. 2. The results of the investigation into the levels and distribution of genetic diversity within populations of Cedrela odorata in Costa Rica revealed significant levels of genetic differentiation between populations sampled from the Atlantic/Southern Pacific and Northern Pacific regions of Costa Rica. The existence of such pronounced population differentiation has profound implications for the conservation and sustainable use of the species. Future conservation strategies should seek to conserve in situ populations from both regions of the country, or a substantial amount of the diversity within C. odorata may be lost. Further research is required to determine the significance of the genetic variation recorded here in the context of the entire species range, in order to assess the full distribution of the two groups that have been identified. It is also considered that moves should be made to rename these two groups as separate taxa; which will help to ensure that the diversity present within both of them is not lost. 3. Results from the study of populations of Swietenia macrophylla demonstrate that different populations of this species exhibit differing levels of diversity. Populations with a history of heavy logging show lower levels of diversity than undisturbed populations and diversity of populations varies according to their proximity to sites of putative Pleistocene Refugia. It is important that logging within the remaining populations of this species is strictly controlled, or valuable genetic resources will be lost. The diversity of trees examined has not clustered into groups according to their geographic location, but according to ecotypic differences, such as levels of precipitation and altitude, between the forests in which they have originated. The once nearly continuous rainforest in Central America has only been destroyed relatively recently. Therefore, it is not really so surprising that there is no real differentiation between recently fragmented populations according to the geographic distance between them. It would be of great interest to conduct a similar examination of in situ seedlings and saplings, rather than mature fruiting trees, to determine whether forest fragmentation has any effect on the diversity of the next generation of mahogany trees in the now highly fragmented Central American rainforest. Evidence provided by this project to government committees of the accelerating destruction of the natural forest, decreasing population size, and the increasing patchiness of distribution of S. macrophylla in Costa Rica has contributed directly to the conservation of this species: mahogany species are now protected in Costa Rica and further cutting is prohibited. In Nicaragua, evidence provided by this project has resulted in the Government placing a temporary order banning further mahogany extraction pending an inventory of the species. The project has also contributed to regional conservation by collection of seed and establishment of gene banks. Preliminary evidence from field trials indicates that there is substantial variation between progenies which requires further investigation for growth rate and disease resistance. This, coupled with evidence of the plasticity of S. macrophylla in the field and combined with the results of the molecular analysis highlight the substantial variation in the species, the value of conservation and provide indications as to the criteria which should be applied when developing conservation policies. This work is expected to continue to contribute to the development of conservation plans. 4. The study of Vochysia guatemalensis has revealed that it is possible to distinguish between individuals from different provenances using the frequencies of the RAPD bands and that the species is likely to be outcrossing, since there was more variation apparent within the provenances than between them. Furthermore, comparisons between groupings in the dendrogram derived from the RAPD data and various quantitative traits revealed a degree of correspondence between the two. The four main clusters in the dendrogram differed significantly for at least one quantitative trait.
B. Tabebuia heterophylla, Dicorynia guyanensis, Virola melinonii, Eperua grandiflora, Vouacapoua americana and Ocotea rubra1. Appropriate techniques for all species were developed. 2. Studies of Tabebuia heterophylla were conducted on 11 populations from 10 different islands in the lesser Antilles. Mating system analyses demonstrated that the species is predominantly outcrossing, but that there is significant variation in outcrossing rates between trees and even between different sides of the canopy in the same tree: some parts of the canopy can be completely selfed while other parts can be completely outcrossing. Controlled crosses demonstrated variation between trees in autoincompatibility. These results indicate the capacity of the species to reproduce in both high and low density situations. Studies of genetic diversity using RAPDs demonstrated significant differences in diversity between populations: populations from the northern islands exhibit higher levels of diversity than those of the southern islands; the northern islands showed less differentiation between them than the southern islands. The high levels of differentiation found suggest that geneflow between islands is extremely low and that populations on each island behave as isolated units. 3. Important interspecific differences in outcrossing rates were found between other species: Virola michelii, Chrysophyllum sanguinolentum and Eperua grandiflora appeared to be strictly outcrossing; Dicorynia guianensis and Vouacapoua americana demonstrated mixed mating systems and both species were tolerant of selfing. Temporal and spatial variations in outcrossing rates were investigated for Dicorynia guianensis: levels of differentiation between pollen clouds were stable between years for a given spatial subdivision but there was significant heterogeneity in between pollen clouds pollinating different trees. These differences decreased as the spatial scale increased from the single tree level to the compartment (25 ha) or stand (300 ha) level. Chloroplast DNA studies show clustering (<50 m) of haplotypes in the Paracou stand. These clusters are approximately the size of the patches of this species in the forest. No spatial differentiation was found for nuclear DNA polymorphism, suggesting that there is asymmetry of geneflow by seed and pollen. Pollen clouds of Virola michelii showed genetic differentiation at a scale between 2 and 3 ha, which is close to the neighbourhood size of male trees. For Vouacapoua americana differences in allelic frequencies of pollen clouds parallel those of adult trees in the same location. In the Paracou stand, differentiation of allele frequencies in adult trees occurs at a large spatial scale (up to 50 ha). 4. Virola michelii and Chrysophyllum sanguinolentum exhibited the highest level of phenotypic diversity (determined by RAPDs), followed by Eperua grandiflora and Dicorynia guianensis. Vouacapoua americana was substantially less polymorphic. When species were compared, outcrossing rate and phenotypic diversity were inversely correlated. Population size of a species was not correlated with diversity: Vouacapoua americana occured at the highest density (9.3 trees ha-1 >10 cm dbh) but had the lowest diversity. 5. The results obtained during this project have already inspired concrete actions to preserve genetic diversity for Tabebuia heterophylla. Because diversity is highly differentiated among the Lesser Antilles, additional collections have already been made in the northern islands (near Porto Rico) and will be undertaken in the southern part of the range. Furthermore the collections made for the diversity survey permitted the harvesting of scions from each island. More than 300 trees were grafted in a clonal bank near the research station at INRA Guadeloupe. Finally the fine study on sexual reproduction in Tabebuia led to the design of an indoor seed orchard which is planned to produce seed of better genetic quality than the actual natural stands. In the case of the five species of the Guianan rainforest, the combined analysis of the mating system and monitoring of genetic diversity identified outcrossing as a major cause of maintenance of genetic diversity. This observation needs to be confirmed on a larger set of species. If confirmed, it may lead to preventive actions to avoid severe modifications of outcrossing rates in silvicultural and logging operations. Because logging may change the spatial distribution of the trees, it may also affect the outcrossing rates and induce short term variation of diversity. FINAL SUMMARY REPORT Contract no. TS3*-CT94-03161 NOVEMBER 1994 - 31 OCTOBER 1997
|
|
Send mail to scav@ceh.ac.uk
with questions or comments about this web site.
|
