by
Paulo Y. Kageyama and Inês de Souza Dias
ESALQ/USP Department of Silviculture
University of Sao Paulo, Caixa Postal 9
Piracicaba S.P. 13.400, Brazil
INTRODUCTION
From the far south to the north, Brazil's native forests are under intensive pressure for utilization, thus placing many existing species and populations in danger of extinction or genetic depletion.
Insufficient knowledge of the biology of the species, their potential for different end uses, and inter and intraspecific genetic variation, may lead to irreversible losses of a large number of species even before adequate studies on their variation and characteristics have been carried out.
Genetic concepts, which today are widely used in tree improvement programmes for a range of species, can successfully be applied in the management, improvement and conservation of Brazilian native species, provided due consideration is given to their biology and specific silvicultural requirements.
The present paper discusses research priorities in native Brazilian species, placing special emphasis on the following basic questions related to the genetic structure of their populations:
Reproductive biology;
Inter and intraspecific variation;
Conservation and genetic improvement.
REPRODUCTIVE BIOLOGY
Data on the reproductive biology, including information on breeding systems; pollination ecology; and characteristics of seed dispersal and the soil seed bank, are indispensable for gaining full understanding of a species. The population genetic structure of tree species is a function of these basic interrelated factors.
Breeding Systems
Investigations determining the breeding system of a species and gene flow within and between populations, should always be the first step in any genetic study.
The question whether a species is predominantly allogamous (outbreeding) or predominantly autogamous (inbreeding) will, at least partly, determine its population structure. However, the question of allogamy versus autogamy is not completely clear-cut, as individual as well as population to population differences in degrees of allogamy may exist even within the same species. Caution must therefore be exercised in drawing conclusions, and generalizations based on limited experiments must be avoided.
The amount and distribution of genetic variability, homozygosity, gene frequencies and their changes from one generation to another and genetic correlation between related individuals depend on the breeding system of the population under investigation (Frankel 1977).
The most widely used forest species have been found to favour out-breeding; Leucaena leucocephala is sometimes cited as one of the few autogamous forest species investigated. This is in contrast to earlier beliefs that most humid tropical forest species are self-compatible and self-pollinating (Corner 1954; Baker 1959; Federov 1966, cited by Bawa 1974). According to Bawa (1974) and Ashton (1976), the predominance of allogamy shows that the selection pressure in nature favours maximum genetic diversity in tropical tree species.
Bawa (1974) stated that the breeding system of plant species could be studied through observations on flowering patterns and the synchronization of flowering in dioecious species, and by comparing progenies from controlled crosses and controlled self-pollination, paying special attention to the degree of self-incompability in species and individuals.
In Brazil, Fonseca (1982) and Siqueira et al. (1982) have studied the breeding systems of Mimosa scabrella Benth, and Dipterix alata Vog, by looking at variance levels within and between progenies in progeny trials; the results strongly suggest that these two species are predominantly allogamous. The method of study is practical and fairly simple, and if proved reliable, can easily be adapted to studies of other species.
Studies on the breeding systems of dioecious plants can be carried out by examining flowering and anthesis and their interrelationships. Thus, if a species shows protandry or protogyny, autogamy is doubtful, although self-fertilization between flowers in different developmental stages in the same individual is still possible (Faegri and Pijl, 1976). A recent study by Crestana et al. (1982), which shows protandry in Esenbeckia leiocarpa Engl., is a good example of the use of this method in the determination of the breeding system of a native species. The preliminary results on Mimosa scabrella obtained by Catharino et al. (1982) showing probable protogyny, is another example of the application into practice of this method. Being a dioecious species, Araucaria angustifolia (Bert.) O. Ktze is, by nature, allogamous. However, this does not mean that there cannot be crossing between related individuals, thus resulting in population structures quite different from those produced by typical panmixis.
As can be deduced from the above, the population structure is directly related to the breeding system of a species. However, a straight-out prediction of the kind and degree of variation between and within populations is not possible based on this factor alone.
Pollination Ecology
It could seem that studies on the pollination ecology are highly specialized and of limited general interest. But once we understand the importance of interactions between plants and their pollinators in determining and maintaining the genetic structure of populations, it is clear that investigations in this field are indispensable.
The dispersal of pollen and seeds are fundamental in maintaining within-species diversity; according to Janzen (1980), pollination is the mechanism by which plants maximize gene flow between individuals. The same author lists some ecological characteristics of tropical species, such as the high percentage of species pollinated by animal vectors, large distances between individuals in outbreeding species, and a large number of complex interactions between plants and their pollinators.
According to Roche (1975), the inter-dependence between pollen vectors and tropical tree species is such that the disappearance of the vectors or a drastic reduction in their numbers could reduce the production of fertile seeds, leading to subsequent extinction of the species.
Slight changes in the pollination system can lead to isolation of sub-populations or keep potentially inter-breeding populations isolated from each other, even when they are sympatric (Faegri and Pijl 1976).
Studies of floral anthesis and foraging behaviour will enable us to draw conclusions on the type and efficiency of pollen vectors, time of receptivity of the stigma, distance of pollen flight or transfer, degree of inter-dependence between plant and pollen vectors (Crestana et al. 1982; Catharino et al. 1982).
In ex situ conservation stands, data on pollination ecology of the species are essential to ensure appropriate management aimed at maintaining integrity of the populations.
Knowledge of the pollination ecology will also be essential in more advanced tree breeding programmes, to improve both quantity and quality of seed production, through favouring conditions for efficient pollination.
Seed Dispersal
Seed dispersal is the first step determining the density and individual patterns of distribution of plant populations. Seed dispersal can also, in many cases, explain the transition from primary to secondary successional stages and the heterogeneous composition maintained in the tropical forest (Fleming and Heithaus 1981).
Together with pollen dispersal, seed dispersal determines the effective population size and influence genetic drift and related phenomena (Roche, 1975). The pollen and the seed produced by a population together represent the gene pool.
Over the thousands of years of their evolution, tropical forests have regenerated after natural disasters. A large gene pool was available to provide material for re-colonization of the areas, favouring rapidly colonizing species; these are today called pioneer or colonizing species (Gomez-Pompa et al. 1972).
Pioneer species are characterized by being light-demanding and fast-growing; they often have prolonged seed dormancy and efficient mechanisms of seed dispersal. Thus, according to Hall and Swaine (1980), they can remain dormant in the ground for a long time, until conditions for their establishment and growth become favourable.
In contrast, in climax species a natural “soil seed bank” is not common, and yearly seed production generally leads to the germination and immediate establishment of a new generation of seedlings. According to Gomez-Pompa et al. (1972), such seedlings can remain latent in the understory until some external factor changes the environmental conditions to favour their growth and development.
Only a few studies have been carried out in seed ecology of native Brazilian species, however, these have proved very valuable. Studies on the soil seed bank of bracatinga (Mimosa scabrella) (Carneiro et al. 1982) have shown that viable seeds can be stored up to a depth of 8 cm in the soil. Small changes in soil temperature cause the dormancy of the seeds to break and result in a rapid emergence of seedlings. This phenomenon explains the occurrence of almost pure and very dense natural bracatinga stands in southern Brazil.
Hall and Swaine (1980), have studied soil seed banks in Ghana, and were amazed at the large quantity of viable seeds especially of pioneer species found in the forest soil.
According to Fleming and Heithaus (1981), the relative distribution of plants in tropical forests depends to a large extent on the behaviour of the dispersing animals (mostly fruit-eating animals) and, thus, distribution can range from widely dispersed plants to thick clumps of plants growing close together in small areas.
These few examples show the considerable complexity in the regeneration of tropical forest species, and the importance of co-evolution between wildlife and plants; they also clearly demonstrate the need for more research in this field.
According to Hall and Swaine (1980), there are four main aspects of seed ecology in natural forests which are in urgent need of research: size and composition of soil seed banks found in different successional stages; patterns and frequency of seed production; length of seed viability; and the environmental conditions necessary for breaking of dormancy and seed germination in various species.
Research on these questions will provide data for appropriate, sustained management of the natural forests, with special emphasis on exploration and regeneration. It will also help us determine the genetic structure of the populations to be conserved and/or sampled for further studies.
INTER- AND INTRA-POPULATION VARIATION
Studies on phenotypic and genotypic variation in different characteristics between and within populations, will enable us to determine the genetic structure of a given species.
Studies of phenotypic variation of individuals and populations can be carried out in situ. Studies of genetic variation will, on the other hand, have to be carried out in identical test conditions for all individuals or populations under study, thus controlling the influence of the environment through appropriate experimental designs.
Phenotypic Variation
Phenotypic studies determine the variation which exists between populations and between individuals within populations; this variation is caused by a combination of genetic and environmental factors. The studies carried out by Fonseca (1982) on populations of Mimosa scabrella, investigating variation in seed characteristics, wood density and bark percentage, are a good example of such studies and their usefulness.
Fonseca, in his study, rightly argues that phenotypic studies should precede those on genetic variation, since they will give a preliminary indication of the genetic structure of a species and thereby assist the researcher in selecting the right sampling strategy for subsequent genetic studies.
If phenotypic studies indicate that no variation exists between and within populations for a given characteristic, it is probable that genetic variation will also be minimal. On the other hand, the presence of phenotypic variation does not, as such, provide evidence of the existence of corresponding, genetically controlled variation.
Phenotypic studies are reliable only when the age of the individuals under study can be determined and taken into consideration; this is generally difficult in natural populations. In the case of a pioneer species such as Mimosa scabrella where, by definition, regeneration generally is relatively even-aged, phenotypic studies in a natural population can, however, provide valuable results.
Genotypic Variation
The method most commonly used for genetic studies is that of provenance and progeny tests, in which seeds collected from representative individuals and/or populations are tested in laboratory, nursery or field conditions, using appropriate experimental designs.
Genetic testing through provenance and progeny tests can only be carried out when technical knowledge is available on the silviculture of a species (including seed handling, nursery techniques, establishment etc.), since such experiments involve the raising of seedlings and the establishment of plantation plots of progenies or provenances.
For the establishment of provenance trials, it is necessary to know the natural distribution of the species, so that adequate and correctly designed sampling can be carried out.
The number of populations included should be a function of the extent of the distribution range, and sampling should be carried out following important environmental gradients, for example changes in latitude, altitude, soils, exposure, etc. Depending on the rough patterns of variation which emerge through testing, sampling on a finer grid may be necessary at a later stage.
Regarding the number of individuals to be sampled within each population, this will depend on the degree of tree to tree variation. Generally, between 25 and 50 individuals should be sampled in allogamous tree populations. It is also recommended that individuals sampled are at a minimum distance of 100 m from each other, to minimize the possibilities of inclusion of close relatives in the sample (Shimizu, Kageyama and Higa 1982).
Of the native forest species in Brazil, few have been properly explored through systematic provenance studies. The species most throughly studied is the parana pine, Araucaria angustifolia. The results reported by Gurgel Filho (1980), Shimizu and Higa (1980), Kageyama and Jacob (1980), Monteiro and Romeiro (1980) and Fahler and Di Luccal (1980), have generally shown pronounced genetic differences between provenances of this species. More recent studies have been carried out on genetic differences between provenances of Dipterix alata (Siqueira et al. 1982) and Mimosa scabrella (Fonseca 1982), among others.
Only relatively recently have researchers in Brazil taken an active interest in progeny trials and in the testing of genetic variation within and between progenies. Results of such studies have been published by Kageyama and Jacob (1980) and Pires (1982) on Araucaria angustifolia; Siqueira et al. (1982) on Dipterix alata; Fonseca (1982) on Mimosa scabrella; and Kanashiro (1982) on Cordia goeldiana.
The value and reliability of progeny trials are dependent on a great number of factors related to sampling of the original populations; type of progenies used or created for trials; and statistical design and size of plots.
Increased attention is presently given in Brazil to provenance and progeny testing, and many native species have recently been included in trials of these types; our state of knowledge is thus steadily improving.
CONSERVATION AND GENETIC IMPROVEMENT
Conservation of Genetic Resources
The basis of conservation of genetic resources is the fact that maintenance of genetic variability between and within populations is essential for the survival and continued evolution of a species. The reduction of natural diversity will decrease the potential of a species to adjust to changes in the natural environment; it will also limit the possibilities of man to adapt it to meet changing human needs (Frankel 1977).
The strategy to be followed in the genetic conservation of a species is dependent on the genetic structure of its populations. As mentioned above, our understanding of the genetic structure will, in turn, depend on information available on the reproductive biology of the species concerned. A species or a population can be genetically conserved through conservation in situ or ex situ.
In situ conservation consists of maintaining samples of specific populations in their natural conditions. The in situ conservation areas of a given species can be either dispersed or continuous, large or small, depending on the patterns of variation found between and within the population to be conserved. In situ conservation is the ideal method of conservation for a range of species found in the mixed tropical forests, given the complexity and inter-dependence of the component species in such ecosystem.1
At the population level, the size of an in situ reserve will depend on the effective population size, and will be determined by the minimum number of interbreeding individuals needed to maintain and transmit to the next generation the full genetic variability of the original population.
Ex situ conservation is a strategy to be used where in situ conservation, for various reasons, is not practicable. This is the case e.g. when populations are under strong demographic or other pressure and their long-term conservation in situ, therefore, is unrealistic or impossible. Ex situ conservation can be done through the establishment of ex situ conservation stands, using reproductive material representative of the original population; or conserving available germplasm as seed, in long-term storage facilities.
For any conservation work, information from provenance and progeny tests is valuable in determining the genetic structure of the population to be sampled and conserved. However, sampling for conservation purposes and sampling for genetic tests on the development of a breeding population, should generally be done with different criteria, firstly because the few characteristics most often considered important in provenance and progeny tests established in connection with genetic improvement programmes (height, diameter, stem form, etc.), do not necessarily reflect patterns of variation in other characteristics in the natural populations; and secondly, because sampling for genetic improvement may be selective, whereas sampling for conservation should be random in order to capture maximum amounts of variation.
Germplasm banks can be useful tools of conservation in the short and medium terms for species having long-lived, “orthodox” seeds.
In Brazil, a great number of species in need of conservation have been identified. Many of them are well represented in in situ reserves, however, in many cases there is an urgent need to establish priorities in conservation and to draw up management plans for the reserves. The criteria for determining priorities are, as recommended by Guldager (1975), in order of importance: species or populations in danger of extinction; species of actual or potential socio-economic value; and species which are shy seeders or for which reproduction or regeneration is otherwise difficult.
In Brazil, highest priority should be given to the establishment of a series of in situ reserves of a given minimum size; these reserves should be established following significant environmental gradients. Subsequently, basic and applied research must be carried out in the reserves to ensure that goals established at the on-set are being achieved; otherwise, as Gomez-Pompa et al. (1972) assert, many species may disappear even before their biology has been clarified.
In some native species, such as Araucaria angustifolia (Timoni et al. 1980) and Dipterix alata (Siqueira et al. 1982), work has been carried out from the standpoint of ex situ conservation. Thus, provenance and progeny tests have been established to provide data on the genetic structure of their populations; these tests serve, to some extent, also as reserves of genetic material.
Genetic Improvement
Action towards genetic improvement of native forest species in Brazil can, in practice, be divided into two main categories: (i) selection of valuable populations; and (ii) individual selection within these populations. More sophisticated methods involving hybridization, induced mutations, etc. are not yet considered practicable.
Once again, it should be stressed that programmes for the production of genetically improved seed of a given species are only meaningful when the techniques for seed handling, nursery techniques, establishment and management of the species are known, and when its end uses and socio-economic value have been determined; basic aspects of reproductive biology must also be known prior to embarking on such programmes.
Selection of promising populations within a species, the first step in any improvement programme, will be based on provenance testing, as discussed earlier. The tests should be established on sites representative of potential plantation areas, and should be aimed at disclosing the genetic characteristics of the population and possible genotype-environment interactions. Native Brazilian forest species which are used in reforestation, such as Araucaria angustifolia, Mimosa scabrella and Cordia goeldiana, have recently been included in such trials.
Individual selection within populations, for the subsequent demarcation or establishment of seed stands and seed orchards, should generally be done only in plantation-grown material. Selection in natural stands is unreliable especially for growth characteristics (which have low heritability), as environmental influences related to spacing, age of trees, management or outside influences, generally vary in such stands.
In natural stands of pioneer species, which regenerate following disturbances such as fire and in which the trees subsequently are largely even-aged, within-population selection can, as an exception to the above principle, be effective especially for characteristics which are not greatly influenced by spacing (tree height, straightness, etc.). An example of such a species is Mimosa scabrella, of which extensive natural populations exist in the south of Brazil (Fonseca, 1982).
CONCLUSIONS
The application of genetic concepts to native species must be based on information on the genetic structure of existing populations. Such information can only be gained through investigations on the reproductive biology of the species and in-depth studies on inter-and intra-population variation patterns in them.
Basic biological and genetic data are indispensable for the management and utilization of natural stands and for their conservation and improvement.
BIBLIOGRAPHY
Ashton, P.S. 1976 An Approach to the Study of Breeding Systems, Population Structure and Taxonomy of Tropical Trees. In: Burley, J. and Styles, B.T. (Eds.). Tropical Trees: Variation, Breeding and Conservation. Academic Press, London. P. 35–42.
August, P.V. 1981 Fig Fruit Consumption and Seed Dispersal by Artibeus jamaicensis in the Llanos of Venezuela. Biotropica 13 (2): 70–6.
Bawa, K.S. 1974 Breeding Systems of Tree Species of a Lowland Tropical Community. Evolution 28: 85–92.
Carneiro, R.M. et al. 1982 Importância da dormência das sementes na regeneração da bracatinga. Circular técnica. IPEF, Piracicaba, (149), set.
Catharino, E.; Crestana, C.S.M. and Kageyama, P.Y. 1982 Biologia floral da bracatinga - Mimosa scabrella Benth. Silvicultura em São Paulo, São Paulo 16 (1): 525–531.
Crestana, C. de S.M.; Dias, I. de S. and Kageyama, P.Y. 1983 Biologia floral do guarantã (Esenbeckia leiocarpa Eng.). Silvicultura, São Paulo 8 (28): 35–8.
Faegri, K. and Van der Pijl, L. 1976 Principles of Pollination Ecology. Pergamon Press, Oxford. 291.
Fahler, J.C. and Di Luccal, C.M. 1979 Variación geografica de Araucaria angustifolia (Bert.) O. Ktze. Informe preliminar a los 5 años. IUFRO Meeting on Forestry Problems of the Genus Araucaria. Anais Curitiba, 21–28 outubro. pp.96–101.
FAO 1975 The Methodology of Conservation of Forest Genetic Resources. Report on a Pilot Study. FO:MISC/75/8. FAO, Rome.
FAO 1981 Report on the FAO/UNEP Export Consultation on In Situ Conservation of Forest Genetic Resources. FAO, Rome.
Fleming, T.H. and Heithaus, E.R. 1981 Frugivorous Bats, Seed Shadows and the Structure of Tropical Forests. Biotropica, 13: 45–53.
Fonseca, S.M. 1982 Variações fenotipicas e genéticas em bracatinga - Mimosa scabrella Benth. Dissertação de Mestrado, ESALQ/USP, Piracicaba, 86 pp.
Frankel, O.H. 1977 Philosophy and Strategy of Genetic Conservation in Plants. In: FAO/IUFRO Third World Consultation on Forest Tree Breeding. Canberra, Australia.
Gomez-Pompa, A.; Vasquez-Yanes and Guevara, S. 1972 The Tropical Rain Forest: A Nonrenewable Resource. Science 7: 762–5.
Guldager, P. 1975 Ex Situ Conservation Stands in the Tropics. In: The Methodology of Conservation of Forest Genetic Resources. FAO, Rome. pp. 85–103.
Gurgel Filho, O.A. 1980 Silvica da Araucaria angustifolia (Bert.) O. Ktze. IUFRO Meeting on Forestry Problems of the Genus Araucaria. Anais. Curitiba, 21–28 outubro. pp. 29–68.
Hall, J.B. and Swaine, M.D. 1980 Seed Stocks in Ghanaian Forest Soils. Biotropica 12 (4): 256–263.
Jackson, J.F. 1981 Seed Size as a Correlate of Temporal and Spatial Patterns of Seed Fall in a Neo-Tropical Forest. Biotropica 13 (2): 121–130.
Janzen, D.H. 1980 Ecologia vegetal nos trópicos. Série temas de biologia. EDUSP, São Paulo. pp. 79.
Kageyama, P.Y. and Jacob, W.S. 1980 Variação genética entre e dentro de populações de Araucaria angustifolia (Bert.) O. Ktze. IUFRO Meeting on Forestry Problems of the Genus Araucaria. Anais. Curitiba 21–28 outubro. pp. 83–86.
Kanashiro, M. 1982 Melhoramento genético de freijó (Cordia goeldiana Huber). Silvicultura, São Paulo 8 (28): 327–328.
Monteiro, R.F.R. and Speltz, R.M. 1980 Ensaio de 24 procedências de Araucaria angustifolia (Bert.) O. Ktze. IUFRO Meeting on Forestry Problems of the Genus Araucaria. Anais. Curitiba 21–28 outubro. pp. 181–200.
Pires, C.L.S. et al. 1983 Teste de progênie de Araucaria angustifolia (Bert) O. Ktze. Em Campos do Jordão. Silvicultura, São Paulo 8 (28): 440–445.
Roche, L.R. 1975 Biological Background. In: The Methodology of Conservation of Forest Genetic Resources. FAO, Rome. pp. 5–18.
Shimizu, J. and Higa, A.R. 1980 Variação genética entre procedências de Araucaria angustifolia (Bert.) O. Ktze, na região de Itapeva, estimada até o 6o ano de idade. IUFRO Meeting on Forestry Problems of the Genus Araucaria. Anais. Curitiba 21–28 outubro. pp. 78–82.
Shimizu, J.Y.; Higa, A.R. and Kageyama, P.Y. 1982 Procedimentos e recomendações para estudos com progênies de essências florestais. Documentos EMBRAPA/URPFCS, Curitiba (11): 1–34.
Siqueira, A.C.M.F. et al. 1983 Teste de progênie e procedência do cumbaru Dipterix alata Vog. Silvicultura, São Paulo 8 (28): 508–510.
Timoni, J. et al. Conservação genética da Araucaria angustifolia (Bert.) O. Ktze. IUFRO Meeting on Forestry Problems of the Genus Araucaria. Anais. Curitiba, 21–28 outubro. pp. 115–118.
1 Note by Editor: See e.g. FAO 1975, 1981
