Skip to main content
NCBI home page
Annals of Botany logoLink to Annals of Botany
. 2005 Mar 24;95(7):1171–1177. doi: 10.1093/aob/mci128

Genetic Variation in Remnant Populations of Dalbergia nigra (Papilionoideae), an Endangered Tree from the Brazilian Atlantic Forest

RENATA ACÁCIO RIBEIRO 1, ANA CAROLINA SIMÕES RAMOS 1, JOSÉ PIRES DE LEMOS FILHO 2, MARIA BERNADETE LOVATO 1,*
PMCID: PMC4246900  PMID: 15790585

Abstract

• Background and Aims Dalbergia nigra, known as Brazilian rosewood, is an endangered tree species restricted to the Brazilian Atlantic Forest and has been intensively logged for five centuries due to its high-quality wood. The objective of the present study was to assess the genetic variation and structure in adults and saplings of the species from a large reserve of the Atlantic Forest, the Rio Doce State Park, and from two small surrounding fragments, one better preserved and another with a high degree of anthropogenic disturbance.

• Methods Analyses of genetic variation and structure were conducted by studying allozyme markers. Seven putative enzymatic loci were resolved, five of them being polymorphic.

• Key Results The mean numbers of alleles per locus (A) were 1·93 and 1·73, while the percentages of polymorphic loci (P) were 93 and 73 % for adults and saplings, respectively. Saplings from the fragment with high anthropogenic disturbance exhibited the lowest values of A and P. The fragment that constitutes a conservation area exhibited genetic variation similar to the population from the large reserve. The observed (Ho) and expected (He) heterozygosities were not significantly different among the three populations. Only sapling populations showed FST values (divergence among populations) significantly different from zero over all studied loci. The fragment with high anthropogenic disturbance exhibited considerable genetic divergence in relation to the above-cited populations.

• Conclusions The evaluated populations displayed mean levels of genetic variation intermediate to those expected for narrow and widespread species. The results suggest that fragments with similar area and geographical distance from a large protected reserve can exhibit different levels of genetic variation, depending on the degree of anthropogenic disturbance. The considerable genetic variation in the protected fragment points to the importance of adequate conservation of small fragments for the preservation of genetic variation in D. nigra.

Keywords: Allozymes, anthropogenic disturbance, Brazilian Atlantic Forest, Brazilian rosewood, conservation, Dalbergia nigra, endangered tree species, genetic variation, habitat fragmentation

INTRODUCTION

The genus Dalbergia (subfamily Papilionoideae) comprises more than 100 species of trees, shrubs and lianas distributed pantropically with centres of diversity in Amazonia and Indo-Asia (Polhill, 1981). In Brazil, 41 species of Dalbergia including two different varieties have been identified (de Carvalho, 1997). Many Dalbergia species are economically important due to their beautiful and valuable timber; however, over-exploitation and habitat fragmentation have caused these species to be in danger of extinction in the tropics (de Carvalho, 1997).

In the last decade, the development of strategies for conservation of endangered plant species has become a centre of interest for conservation biologists. At the forefront of conservation decisions on these species is the necessity to understand their genetic variation and structure of their populations, as well as the factors that influence the distribution of such variation (Milligan et al., 1994). Habitat fragmentation represents one of the major threats to biodiversity. In many parts of the world, fragmentation has transformed large regions of continuous forests into small and geographically isolated remnant fragments, often surrounded by human-modified areas (Turner, 1996; Young et al., 1996). Theoretically, population size reduction (bottleneck) after habitat fragmentation leads to an erosion of genetic variation and an increase of inter-population genetic differentiation through increased genetic drift, elevated inbreeding and reduced inter-population gene flow among plant populations (reviewed by Young et al., 1996).

One of the most endangered species of the genus Dalbergia is D. nigra (Vell.) Allemao ex Benth, known as Brazilian rosewood, a tropical tree restricted to the Brazilian Atlantic Forest, occurring from southern Bahia to northern São Paulo State (de Carvalho, 1997). The Atlantic Forest is one of the main tropical forests in the world and originally covered almost all the Brazilian coastland, in a total of 1·2 million km2 of native forest. After five centuries of human occupation and economic exploitation, it has been reduced to 7·5 % of the original area (Myers et al., 2000). The severe reduction and the large number of endemic species promoted the inclusion of this biome in the list of the 25 worldwide conservation hotspots by Conservation International (Myers et al., 2000). Ever since the occupation of its habitat D. nigra has been heavily logged as a source of wood for the manufacture of fine furniture and musical instruments (Lorenzi, 1992; Carvalho, 1994). Because of this over-exploitation and the deforestation of its natural habitat, D. nigra has become extremely rare in nature (de Carvalho, 1997), and extant stands are restricted to conservation areas (parks and reserves) and remote areas difficult to access (Costa et al., 1998). The species was included in the Official List of Threatened Brazilian Plants (IBAMA, 1992) and in the Appendix I of the Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES, 1992), causing the prohibition of its commercialization and international trade. Furthermore, D. nigra is listed as vulnerable according to the criteria of The World Conservation Union (IUCN) Red List of Threatened Species, which means that it is facing a high risk of extinction in the wild in the medium-term future (IUCN, 1994).

Adult trees of D. nigra can reach up to 35 m in height and 155 cm in diameter at breast height (dbh) (Leão and Vinha, 1975). The leaves are alternate, small and caduceus, and the racemes have hermaphrodite flowers with small and white-yellow petals (Carvalho, 1994). In south-east Brazil, it flowers in September–October and fruits containing two to four seeds ripen in November–December (Carvalho, 1994). Data on D. nigra's mating system are not available; however, studies of a congeneric species in Brazil (D. miscolobium) suggest outcrossing with an apparent self-incompatibility system (Gibbs and Sassaki, 1998). Pollen is likely to be dispersed by bees (personal observation) and seeds by wind (Carvalho, 1994), as was observed in a congeneric species in India (Mohana et al., 2001).

This study is part of the Brazilian Long Term Ecological Research Program which develops studies in the Atlantic Forest and Lacustrine Ecosystems of the Middle Rio Doce (19°28′ to 19°58′S and 41°50′ to 48°28′W) in Minas Gerais State, south-east Brazil. Some of the aims of this research are to investigate genetic variation in tree species, to determine the effects of anthropogenic actions on the biodiversity in the Middle Rio Doce and to propose conservation strategies. The region studied includes the biological reserve of Rio Doce State Park that constitutes a large remnant of the Brazilian Atlantic Forest (approx. 36 000 ha). The surrounding area of this reserve has been highly fragmented since the 1950s (A. Hirsch, pers. comm.), mainly due to Eucalyptus plantations and pasture, resulting in a large number of forest remnants. The reserve and surrounding remnants are considered priority areas for biodiversity conservation (Costa et al., 1998).

According to Turner and Corlett (1996), small fragments provide a safety net, protecting a significant number of species and assuring their genetic variation. Thus, the evaluation and protection of such fragments should be prioritized. The objective of the present study was to assess the genetic variation and structure of adult trees and saplings of D. nigra in populations from the Rio Doce State Park and from two small surrounding fragments, one of which is relatively well protected and the other exhibiting a high degree of anthropogenic disturbance. It is hypothesized that genetic variation will be lower in the populations from fragments than that from the reserve. Considering that D. nigra plants are long-lived and fragmentation in the region is recent, it is hypothesized that genetic variation will be lower in saplings than in adult trees from the fragmented areas.

MATERIALS AND METHODS

Populations and sampling

Three populations of D. nigra were sampled in the Rio Doce State Park biological reserve and its immediate surroundings (Fig. 1). The Campolina (CAM) population is located in a 130-ha area within the boundaries of the reserve in which there is no evidence of previous logging and which is considered a primary forest. The Santa Cruz (SC) and Areias (AR) populations are located in two distinct forest fragments near the reserve and are 51·5 and 40 ha, respectively, in area. The SC fragment is owned by a private company (Santa Barbara Ltda) and consists of vegetation similar to primary forest, surrounded by Eucalyptus plantations and pasture (Kageyama et al., 1999). The AR fragment is an area of high anthropogenic disturbance and is composed of species typical of secondary vegetation and surrounded by large areas of pasture (Kageyama et al., 1999). The geographical distances from CAM to SC and AR populations are approx. 27 and 34 km, respectively. The distance between SC and AR is about 8 km.

Fig. 1.

Fig. 1.

Open in a new tab

Populations of Dalbergia nigra sampled in Minas Gerais State, south-east Brazil. Thick lines represent the main rivers and grey areas represent forest (adaptation of map geoprocessed by A. Hirsch and E. C. Landau).

Leaves were collected from 20–30 individual adult trees and saplings in each population, totalling 136 individuals (63 adults and 73 saplings). Adult trees constitute the potentially reproductive class and measured from 25 to 53 cm dbh and 12 to 28 m high, whereas the saplings had not reached reproductive maturity and measured <9 cm dbh. To ascertain an adequate coverage of each population studied, random samplings were carried out across populations. Leaves were transported in liquid nitrogen and stored in the laboratory at −70 °C until enzyme extraction.

Electrophoresis

Leaf tissue was crushed in liquid nitrogen using extraction buffer (solution 1; Alfenas, 1998) with addition of PVP-40 (polyvinyl–pyrrolidone) and 0·2 % mercaptoethanol prior to the procedure. The enzyme extract was absorbed onto filter paper wicks and stored at −70°C until analysis. Horizontal electrophoresis was performed using 13 % starch gels on constant current of 40 mA at 5°C for 4–6 h. Twenty-one enzyme systems were tested using several combinations of gel and electrode buffers to find those that resulted in clear banding patterns and easy genetic interpretation. However, in the leaves of D. nigra the resolution of most of these enzyme systems was not adequate due to the high concentration of secondary compounds, which promote fast oxidation of the enzymatic extracts. The following five enzyme systems, involving seven putative genetic loci, produced interpretable results in D. nigra: diaphorase (EC 1.6.99, loci Dia-1 and Dia-2), isocitrate dehydrogenase (EC 1.1.1.41, locus Idh), malic enzyme (EC 1.1.1.40, locus Me), menadione reductase (EC 1.6.99.2, locus Mr) and phosphoglucoisomerase (EC 5.3.1.9, loci Pgi-1 and Pgi-2). These enzymes were assayed using a morpholine-citrate buffer (Clayton and Tretiak, 1972). Allozyme phenotypes were genetically interpreted according to conventional standard principles (Kephart, 1990; Alfenas, 1998).

Data analysis

Allelic frequencies and levels of genetic variation were calculated for adults and saplings of D. nigra for each population using GENEPOP (Raymond and Rousset, 1995) and BIOSYS (Swofford and Selander, 1981). The following statistics were computed: the mean number of alleles per locus (A), the percentage of polymorphic loci (P, no criterion), the observed heterozygosity (Ho) and the Hardy–Weinberg expected heterozygosity (He, Nei's unbiased estimate; Nei, 1978). Differences in allelic frequencies among populations were estimated by exact-tests for population differentiation (Haldane, 1954), and deviation in genotypic frequencies from the Hardy–Weinberg expectations by exact-test using the Markov Chain method (Guo and Thompson, 1992). Differences of Ho and He means for adults and saplings within and among populations were determined by t-test.

The population genetic structure was determined by Weir and Cockerham's methods (Weir and Cockerham, 1984) of calculating Wright's F-statistics (Wright, 1965) using TFPGA (Miller, 1997). To determine whether FIS and FIT values for each locus and over all loci were significantly different from zero, the chi-square statistics were used: χ2 = F(2N)(k − 1), with k(k − 1)/2 degrees of freedom, where N is the sample size and k the number of alleles (Weir, 1990). To determine the significance of the FST value per locus and over all loci, the chi-square statistics was also used: χ2 = (2N)FST(k − 1), with (k − 1)(n − 1) degrees of freedom, where n is the number of populations (Workman and Niswander, 1970). The 95 % confidence intervals to F-statistics were obtained by bootstrapping over loci and the variance estimates were found by jackknifing over loci (Weir and Cockerham, 1984). The population genetic structure and the genetic distance were determined in adults and saplings between pairs of D. nigra populations using FST analysis (Wright, 1965) and Nei's unbiased measures of genetic distance (Nei, 1978), respectively.

RESULTS

Across the 136 individuals scored, five polymorphic allozyme loci were identified (Dia-2, Idh, Me, Mr and Pgi-2) and a total of 10 alleles were recorded (Table 1). Dia-1 and Pgi-1 were monomorphic in the sampled populations and were not included in the analysis. The least frequent allele in Dia-2, Idh and Mr loci in the CAM and SC populations was even less frequent in the AR population, with adults and saplings exhibiting one and three monomorphic loci, respectively (Table 1). Using the exact tests for population differentiation, significant differences in allelic frequencies over all loci were detected, both in adults and saplings of D. nigra, between AR and CAM populations (P < 0·05; Table 2). Additionally, a difference was also detected between saplings of the AR and SC populations (P < 0·05). Significant differences were not detected between CAM and SC in adults and saplings, with the exception of the Me locus in saplings (Table 2).

Table 1.

Allelic frequencies for five polymorphic loci in adults and saplings of Dalbergia nigra populations

Populations
Locus
 Allele
 CAM
 SC
 AR
Dia-2
    Adults a 0·174 0·125 0·100
b 0·826 0·875 0·900
    Saplings a 0·304 0·233 0·026
b 0·696 0·767 0·974
Idh
    Adults a 0·152 0·125 0·000
b 0·848 0·875 1·000
    Saplings a 0·130 0·117 0·000
b 0·870 0·883 1·000
Me
    Adults a 0·370 0·325 0·575
b 0·630 0·675 0·425
    Saplings a 0·456 0·250 0·526
b 0·544 0·750 0·474
Mr
    Adults a 0·196 0·125 0·075
b 0·804 0·875 0·925
    Saplings a 0·326 0·283 0·000
b 0·674 0·717 1·000
Pgi-2
    Adults a 0·935 0·975 0·975
b 0·065 0·025 0·025
    Saplings a 0·956 1·000 1·000
b 0·044 0·000 0·000
Open in a new tab

The least frequent allele of each locus is underlined.

Table 2.

P-values of exact-test (Raymond and Rousset, 1995) for allelic frequency differences in adults and saplings between pairs of Dalbergia nigra populations

Locus
 CAM × SC
 CAM × AR
 SC × AR
Dia-2
    Adults 0·562 0·365 1·000
    Saplings 0·505 0·001* 0·006*
Idh
    Adults 0·766 0·015* 0·056
    Saplings 1·000 0·030* 0·040*
Me
    Adults 0·820 0·071 0·045*
    Saplings 0·037* 0·667 0·007*
Mr
    Adults 0·557 0·137 0·704
    Saplings 0·665 0·000* 0·000*
Pgi-2
    Adults 0·620 0·625 1·000
    Saplings 0·185 0·494 1·000
All loci
    Adults 0·937 0·024* 0·242
    Saplings 0·275 0·000* 0·000*
Open in a new tab
*

P < 0·05.

Genetic variation was estimated for adults and saplings in each population of D. nigra (Table 3). The mean number of alleles per locus (A) was 1·93 and 1·73 for adults and saplings, respectively. In the adults, the percentage of polymorphic loci (P) per population varied from 80 to 100 %, with a mean of 93 %. In the saplings, P varied from 40 to 100 %, with a mean of 73 %. Saplings from the AR fragment exhibited the lowest values of A and P. Observed mean heterozygosities (Ho) were 0·163 ± 0·02 (ranging from 0·130 to 0·209) and 0·172 ± 0·06 (ranging from 0·053 to 0·243) for adults and saplings in D. nigra populations, respectively. Expected mean heterozygosities (He) were 0·235 ± 0·04 (ranging from 0·176 to 0·296) and 0·243 ± 0·07 (ranging from 0·113 to 0·341) for adults and saplings, respectively. Similarly, the AR population showed the lowest genetic variation, being even lower in saplings than adults (Table 3). However, there was no significant difference by t-test in heterozygosities among populations (P < 0·05).

Table 3.

Genetic variation estimates for adults and saplings of Dalbergia nigra populations

Population
 N
 A
 P
 Ho (s.e.)
 He (s.e.)
Campolina
    Adults 23·0 2·00 100 0·209 (0·035) 0·296 (0·056)
    Saplings 23·0 2·00 100 0·243 (0·074) 0·341 (0·079)
Santa Cruz
    Adults 20·0 2·00 100 0·150 (0·032) 0·235 (0·064)
    Saplings 30·0 1·80 80 0·220 (0·085) 0·274 (0·077)
Areias
    Adults 20·0 1·80 80 0·130 (0·046) 0·176 (0·088)
    Saplings 20·0 1·40 40 0·053 (0·041) 0·113 (0·100)
Mean
    Adults 21·0 1·93 93 0·163 (0·024) 0·235 (0·035)
    Saplings 24·3 1·73 73 0·172 (0·060) 0·243 (0·068)
Open in a new tab

N, sample size; A, mean number of alleles per locus; P, percentage polymorphic loci; Ho, observed heterozygosity; He, expected heterozygosity (unbiased estimate, Nei, 1978); s.e., standard error.

Probabilities of the genotype frequency deviation from Hardy–Weinberg equilibrium were calculated for all polymorphic loci in adults and saplings of each population. Two loci (Idh and Me) were positive and differed significantly from zero (P < 0·05) over all populations in adults and saplings, which indicates a deficiency of heterozygotes. The other loci were consistent with Hardy–Weinberg equilibrium (P ≥ 0·05).

Wright's F-statistics (FIS, FST and FIT) for each polymorphic locus in adults and saplings of D. nigra are shown in Table 4. Mean inbreeding within populations (FIS) and total inbreeding (FIT) were positive and significantly different from zero over all loci (P < 0·05) both in adult and sapling populations (except the FIT value for locus Pgi-2 in saplings). However, only saplings showed values of FST significantly different from zero over all loci (P < 0·05). The mean FST values (level of divergence between populations) for adults and saplings were 0·017 and 0·080, respectively, and the values were significantly different as indicated by the 95 % confidence intervals (Table 4).

Table 4.

Estimates of Wright's F-statistics (Wright, 1965) for five polymorphic loci in adult and sapling populations of Dalbergia nigra

Locus
 FIS
 FST
 FIT
Dia-2
    Adults 0·133*** −0·015 0·120***
    Saplings −0·051** 0·091** 0·044*
Idh
    Adults 0·438*** 0·041 0·461***
    Saplings 0·576*** 0·021 0·585***
Me
    Adults 0·506*** 0·034 0·523***
    Saplings 0·579*** 0·059* 0·603***
Mr
    Adults 0·121*** 0·006 0·126***
    Saplings 0·034* 0·132*** 0·162***
Pgi-2
    Adults −0·027* −0·009 −0·036*
    Saplings −0·024* 0·025 0·001
Over all loci
    Adults 0·313*** 0·017 0·325***
    Saplings 0·280*** 0·080** 0·338***
Standard deviation
    Adults 0·132 0·013 0·140
    Saplings 0·196 0·021 0·168
95 % confidence intervals
    Adults 0·097–0·475 −0·007–0·034 0·093–0·492
    Saplings −0·016–0·565 0·046–0·114 0·094–0·586
Open in a new tab
*

P < 0·05;

**

P < 0·01;

***

P < 0·001.

The FST values for each pair of populations (Table 5) showed that adults and saplings in the SC population were genetically similar to their matches in the CAM population. However, the saplings in the AR population exhibited considerable genetic divergence in relation to the CAM and SC populations (FST = 0·123 and 0·147, respectively), whereas adults showed lower divergence in relation to these populations (FST = 0·040 and 0·044, respectively). Nei's coefficients (Nei, 1978) of genetic distances indicated similar results to those obtained with FST in relation to genetic divergence among populations (Table 5).

Table 5.

FST values (above diagonal) and genetic distances Nei's D (below diagonal) in adults and saplings between pairs of Dalbergia nigra populations

Populations
 CAM
 SC
 AR
Adults
    CAM − −0·023 0·040
    SC 0·000 − 0·044
    AR 0·013 0·013 −
Saplings
    CAM − 0·010 0·123
    SC 0·005 − 0·147
    AR 0·037 0·044 −
Open in a new tab

DISCUSSION

Species with restricted geographic distribution commonly present lower levels of genetic variation than widespread species (Hamrick and Godt, 1989). In spite of D. nigra having a narrow distribution and being considered in danger of extinction, the populations analysed showed levels of genetic variation (mean He = 0·235 and 0·243, for adults and saplings, respectively) intermediate to those expected for narrow species (HS = 0·215) and for widespread species (HS = 0·267) (Hamrick and Godt, 1989). Species with traits similar to D. nigra do not necessarily exhibit low genetic variation. High genetic variation has been reported for the endangered tree species Caesalpinia echinata (Cardoso et al., 1998) and the narrow endemic species Seseli farrenyi (López-Pujol et al., 2002), Cochlearia bavarica (Paschke et al., 2002) and Antirhea aromatica (González-Astorga and Castillo-Campos, 2004).

Habitat fragmentation may cause loss of population genetic variation through an accentuated reduction of population size (bottleneck), as an immediate effect of fragmentation, leaving only a small portion of the original gene pool after fragmentation (Frankel and Soulé, 1981; Barrett and Kohn, 1991; Young et al., 1996). Subsequently, populations which remain small and isolated for several generations loose alleles due to random genetic drift, reducing the levels of genetic variation within populations (Barrett and Kohn, 1991; Young et al., 1996). It is noteworthy that the least frequent alleles in the SC and CAM population are even less frequent in the saplings of the AR population. The AR population exhibited lower genetic variation compared with that observed in the SC population, probably due to a higher degree of anthropogenic disturbance, including timber exploitation which might result in reduction of population size. These findings suggest that not only the area of fragments but also the degree of their preservation is important to the maintenance of genetic variability.

The levels of heterozygosity (Ho and He) observed in the SC and AR populations were lower, but were not significantly different from the CAM population. Heterozygosity is slightly affected by low-frequency alleles and these are the most likely to be lost immediately after fragmentation (Taggart et al., 1990). In fact, relevant fragmentation in the region is recent and it was found that the least frequent allele in CAM was even less frequent in saplings from AR, in which fragment three loci were monomorphic. Studies on the levels of genetic variation in other tree species did not indicate any reduction in levels of heterozygosity in remnant populations either (Young et al., 1993; Coates and Hamley, 1999; White et al., 1999; Collevatti et al., 2001). In populations of D. nigra two allozyme loci (Idh and Me) were not in Hardy–Weinberg equilibrium. It seems that fragmentation has not affected the mating system in D. nigra, as all three populations showed departure from the Hardy–Weinberg equilibrium and their values of inbreeding were not significantly different in adults and saplings.

One of the theoretical effects expected from fragmentation is the increase of genetic divergence among populations due to genetic drift and reduction of gene flow (reviewed by Young et al., 1996). If this hypothesis is correct, considering the cumulative effect of genetic drift along generations (Hartl and Clark, 1997), higher genetic divergence among sapling populations that were established after fragmentation is expected than among adult probably dating from before fragmentation. In fact, the genetic divergence among populations evaluated by FST values was significant only in saplings, suggesting the development of a stronger genetic structure after fragmentation. The genetic divergence between pairs of D. nigra populations evaluated by FST and Nei's D values indicated that saplings from the fragment with a high degree of anthropogenic disturbance (AR population) formed the most divergent population. The development of divergence in the saplings of this population could appear surprising, given that the fragmentation occurred recently. However, the increase in FST values is small and can be due to more intense logging of this population as timber, significantly reducing the number of adult plants that contributed to the genetic pool of further generations. In fact, adult plants from AR also exhibited genetic divergence in relation to adults of the more preserved sites.

Effects of fragmentation on agents of pollination and seed dispersal can influence the gene flow in remnant populations (Young et al., 1996). As seeds are dispersed by wind, the intensity of which can increase after fragmentation (Saunders et al., 1991), a decrease of seed dispersal could not be expected in this case. In fact, an increase of gene flow was found after fragmentation in Acer saccharum, which is primarily a wind-pollinated and wind-dispersed species (Foré et al., 1992). Considering that D. nigra is probably bee pollinated (personal observation), the reduction of gene flow among their populations after fragmentation can be a consequence of modifications in the behaviour and availability of its pollination agent.

The fragmentation and deforestation of the Brazilian Atlantic Forest has occurred since colonial times in the 1500s. The forest fragmentation in the region of the present study is relatively recent (around 50 years) and this could explain the considerable genetic variation in the D. nigra populations found in the large remnant, Rio Doce State Park, and the SC fragment. However, the AR fragment that had a high degree of disturbance showed evidence of a reduction in genetic variation mainly in saplings and an increase in genetic divergence in relation to the other populations studied. Turner (1996) suggests that the greater pressure of human activity on wild life in forest remnants may contribute to the negative impact on tropical forest fragments. Considering that the AR and the SC fragments present a similar area and geographical distance to the large reserve, it is suggested that the differences in genetic variation and structure between them could be due to the difference in degree of anthropogenic disturbance. Contrary to the idea that only the preservation of large remnants can safeguard the biodiversity, recent research has shown that a significant number of forest species can persist for decades in small remnants (Turner and Corlett, 1996). The present results showing a relatively high genetic variation even in small fragments confirm the importance of their preservation, mainly for conservation of endangered species like D. nigra.

Acknowledgments

This study was supported by Brazil Long Term Ecological Research Program—Conselho Nacional de Desenvolvimento Tecnológico (PELD—CNPq/Brazil) and Fundação de Amparo à Pesquisa do Estado de Minas Gerais (FAPEMIG/Brazil). We thank Dr. Cesar Jacoby for revising the English version of this manuscript. We also thank Instituto Estadual de Florestas and Rio Doce State Park for providing facilities; Rosângela L. Brandão for all her help in the field; Edivani V. Franceschinelli and Marlene de Miranda for technical assistance in the laboratory; A. Hirsch and E. C. Landau for geoprocessing of the region map. R. A. Ribeiro has received a MSc fellowship from the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES/Brazil).

LITERATURE CITED

  1. Alfenas AC. 1998.Eletroforese de Isoenzimas e Proteínas Afins: Fundamentos e Aplicações em Plantas e Microrganismos. Viçosa, Brazil: Editora da Universidade Federal de Viçosa. [Google Scholar]
  2. Barrett SCH, Kohn JR. 1991. Genetic and evolutionary consequences of small population size in plants: implications for conservation. In: Falk DA, Holsinger KE, eds. Genetics and conservation of rare plants. New York: Oxford University Press, 3–30. [Google Scholar]
  3. Cardoso MA, Provan J, Powell W, Ferreira PCG, de Oliveira DE. 1998. High genetic differentiation among remnant populations of the endangered Caesalpinia echinata Lam. (Leguminoseae–Caesalpinioideae). Molecular Ecology 7: 601–608. [Google Scholar]
  4. Carvalho PER. 1994.Espécies Florestais Brasileiras: Recomendações Silviculturais, Potencialidades e Uso da Madeira. Brasília, Brazil: EMBRAPA-CNPF/SPI. [Google Scholar]
  5. CITES. 1992.CITES: Appendices I, II and III to the Convention on International Trade in Endangered Species of Wild Fauna and Flora. Washington: US Fish and Wildlife Service. [Google Scholar]
  6. Clayton JW, Tretiak DN. 1972. Amine-citrate buffers for pH control in starch gel electrophoresis. Journal of the Fisheries Research Board of Canada 19: 1169–1172. [Google Scholar]
  7. Coates DJ, Hamley VL. 1999. Genetic divergence and the mating system in the endangered and geographically restricted species, Lambertia orbifolia Gardner (Proteaceae). Heredity 83: 418–427. [DOI] [PubMed] [Google Scholar]
  8. Collevatti RG, Grattapaglia D, Hay JD. 2001. Population genetic structure of the endangered tropical tree species Caryocar brasiliense, based on variability at microsatellite loci. Molecular Ecology 10: 349–356. [DOI] [PubMed] [Google Scholar]
  9. Costa CMR, Herrmann G, Martins CS, Lins LV, Lamas IR. 1998.Biodiversidade em Minas Gerais: um atlas para sua conservação. Belo Horizonte, Brazil: Fundação Biodiversitas. [Google Scholar]
  10. de Carvalho AM. 1997. A synopsis of the genus Dalbergia (Fabaceae: Dalbergieae) in Brazil. Brittonia 49: 87–109. [Google Scholar]
  11. Foré SA, Hickey RJ, Vankat JL, Guttman SI, Shaefer RL. 1992. Genetic structure after forest fragmentation: a landscape ecology perspective on Acer saccharum Canadian Journal of Botany 70: 1659–1668. [Google Scholar]
  12. Frankel OH, Soulé ME. 1981.Conservation and evolution. Cambridge, UK: Cambridge University Press. [Google Scholar]
  13. Gibbs P, Sassaki R. 1998. Reproductive Biology of Dalbergia miscolobium Benth. (Leguminosae–Papilionoideae) in SE Brazil: the effects of pistillate sorting on fruit-set. Annals of Botany 81: 735–740. [Google Scholar]
  14. González-Astorga J, Castillo-Campos G. 2004. Genetic variability of the narrow endemic tree Antirhea aromatica Castillo-Campos & Lorence (Rubiaceae, Guettardeae) in a tropical forest of Mexico. Annals of Botany 93: 521–528. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Guo SW, Thompson EA. 1992. Performing the exact test of Hardy–Weinberg proportions for multiple alleles. Biometrics 48: 361–372. [PubMed] [Google Scholar]
  16. Haldane JBS. 1954. An exact test for randomness of mating. Journal of Genetics 52: 631–635. [Google Scholar]
  17. Hamrick JL, Godt MJW. 1989. Allozyme diversity in plant species. In: Brown AHD, Clegg MT, Kahler AL, Weir BS, eds. Plant population genetics, breeding and genetic resources. Sunderland, MA: Sinauer Associates, 43–63. [Google Scholar]
  18. Hartl DL, Clark AG. 1997.Principles of population genetics, 3rd edn. Sunderland, MA: Sinauer Associates. [Google Scholar]
  19. IBAMA. 1992.Lista Oficial de Espécies da Flora Brasileira Ameaçadas de Extinção. Instituto Brasileiro de Meio Ambiente e dos Recursos Naturais Renováveis, Secretaria de Meio Ambiente, Brazil. [Google Scholar]
  20. IUCN. 1994.Red List Categories, IUCN Species Survival Commission. Gland, Switzerland: IUCN. [Google Scholar]
  21. Kageyama PW, de Lacerda CMB, Gandara FB. 1999.3° Relatório para o IPGRI em Relação ao Projeto de Pesquisa. Estratégias e Parâmetros para Conservação Genética in situ de Florestas Tropicais no Brasil. São Paulo, Brazil: IPGRI, USP, CENARGEN. [Google Scholar]
  22. Kephart SR. 1990. Starch gel electrophoresis of plant isozymes: a comparative analysis of techniques. American Journal of Botany 77: 693–712. [Google Scholar]
  23. Leão AC, Vinha SG. 1975. Ocorrência do jacarandá no sul da Bahia. Cacau Atualidades, Ilhéus 12: 22–29. [Google Scholar]
  24. López-Pujol J, Bosch M, Simon J, Blanché C. 2002. Allozyme diversity in the tetraploid endemic Thymus loscosii (Lamiaceae). Annals of Botany 93: 323–332. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Lorenzi H. 1992.Árvores Brasileiras. Manual de Identificação e Cultivo de Plantas Arbóreas Nativas do Brasil. São Paulo, Brazil: Plantarum Ltda, Nova Udessa. [Google Scholar]
  26. Miller MP. 1997. Tools for populations genetic analyses (TFPGA) 1·3: a Windows program for the analysis of allozyme and molecular population genetic data. Computer software distributed by author. [Google Scholar]
  27. Milligan BG, Leebens-Mack J, Strand AE. 1994. Conservation genetics: beyond the maintenance of marker diversity. Molecular Ecology 3: 423–435. [Google Scholar]
  28. Mohana GS, Uma Shaanser R, Ganeshaiah KN, Dayanandan S. 2001. Genetic relatedness among developing seeds and intra fruit seed abortion in Dalbergia sissoo (Fabaceae). American Journal of Botany 88: 1181–1188. [PubMed] [Google Scholar]
  29. Myers N, Mittermeier RA, Mittermeier CG, da Fonseca GAB, Kent J. 2000. Biodiversity hotspots for conservation priorities. Nature 403: 853–858. [DOI] [PubMed] [Google Scholar]
  30. Nei M. 1978. Estimation of average heterozygosity and genetic distance from a small number of individuals. Genetics 89: 583–590. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Paschke M, Abs C, Schmid B. 2002. Relationship between population size, allozyme variation, and plant performance in the narrow endemic Cochlearia bavarica Conservation Genetics 3: 131–144. [Google Scholar]
  32. Polhill RM. 1981. Dalbergieae. In: Polhill RM, Raven PH, eds. Advances in legume systematics, Part 1. Kew: Royal Botanic Gardens, 233–242. [Google Scholar]
  33. Raymond M, Rousset F. 1995. GENEPOP: population genetics software for exact test and ecumenicism. Journal of Heredity 86: 248–249. [Google Scholar]
  34. Saunders DA, Hobbs RJ, Margules CR. 1991. Biological consequences of ecosystem fragmentation: a review. Conservation Biology 5: 18–32. [Google Scholar]
  35. Swofford DL, Selander RB. 1981. BIOSYS-1: a FORTRAN program for the comprehensive analysis of electrophoretic data in population genetics and systematics. Journal of Heredity 72: 281–283. [Google Scholar]
  36. Taggart JB, McNally SF, Sharp PM. 1990. Genetic variability and differentiation among founder populations of the pitcher plant (Sarracenia purpurea L.) in Ireland. Heredity 64: 177–183. [Google Scholar]
  37. Turner IM. 1996. Species loss in fragments of tropical rain forest: a review of the evidence (Review). Journal of Applied Ecology 33: 200–209. [Google Scholar]
  38. Turner IM, Corlett RT. 1996. The conservation value of small isolated fragments of lowland tropical rain forest. Trends in Ecology and Evolution 11: 330–333. [DOI] [PubMed] [Google Scholar]
  39. Weir BS. 1990.Genetic data analysis. Sunderland, MA: Sinauer Associates. [Google Scholar]
  40. Weir BS, Cockerham CC. 1984. Estimating F-statistics for the analysis of population structure. Evolution 38: 1358–1370. [DOI] [PubMed] [Google Scholar]
  41. White GM, Boshier DH, Powell W. 1999. Genetic variation within a fragmented population of Swietenia humilis Zucc. Molecular Ecology 8: 1899–1909. [DOI] [PubMed] [Google Scholar]
  42. Workman PL, Niswander JD. 1970. Population studies on southwestern Indian tribes. II. Local differentiation in the Papago. American Journal of Human Genetics 22: 24–29. [PMC free article] [PubMed] [Google Scholar]
  43. Wright S. 1965. The interpretation of population structure by F-statistics with special regard to systems of mating. Evolution 19: 395–420. [Google Scholar]
  44. Young A, Boyle T, Brown T. 1996. The population genetic consequences of habitat fragmentation for plants. Trends in Ecology and Evolution 11: 413–418. [DOI] [PubMed] [Google Scholar]
  45. Young AG, Merrian HG, Warwick SI. 1993. The effects of forest fragmentation on genetic variation in Acer saccharum Marsh. (sugar maple) populations. Heredity 71: 277–289. [Google Scholar]

Articles from Annals of Botany are provided here courtesy of Oxford University Press

RESOURCES