Molecular Ecology (2003) 12, 3201– 3212
doi: 10.1046/j.1365-294X.2003.01986.x
Restoration of genetic diversity in the dry forest tree
Swietenia macrophylla (Meliaceae) after pasture
abandonment in Costa Rica
Blackwell Publishing Ltd.
M . C É S P E D E S ,* M . V . G U T I E R R E Z ,† N . M . H O L B R O O K ‡ and O . J . R O C H A *§
*Programa Regional de Maestría en Biología, Universidad de Costa Rica, San José, Costa Rica; †Estación Experimental Agrícola ‘Fabio
Baudrit’, Universidad de Costa Rica, Alajuela, Costa Rica; ‡Department of Organismic and Evolutionary Biology, Harvard University,
Cambridge, USA; §Escuela de Biología, Universidad de Costa Rica, Ciudad Universitaria ‘Rodrigo Facio’, San Pedro de Montes de Oca,
San José, Costa Rica
Abstract
We studied the levels of genetic diversity of Swietenia macrophylla (big leaf mahogany) in
five successional plots in the Santa Rosa National Park, Guanacaste, Costa Rica. We
selected sites with different lengths of time since the last major disturbance (typically fire):
6, 9, 15 and 20 years. In addition, we also included a patch of mature forest that had experienced selective logging and other human activity in the past 100 years. Genetic diversity
was assessed using five polymorphic DNA microsatellite loci. We found a total of 21 alleles
in the five loci examined, in which the number of alleles present varied among the five sites
studied. Allelic diversity varied between sites ranging from 20 to 14 alleles, and our data
revealed that earlier successional sites have more alleles than older sites. There was significant heterogeneity in allele frequencies between sites; however, genetic differentiation
between populations was low (FST = 0.063) indicating that most of the variation was found
within sites and extensive gene flow between sites. In addition, our analysis also showed
that genetic diversity of adult trees does not solely determine the diversity of seedlings and
saplings found around them, also supporting the existence of extensive gene flow. The
impact of these findings for the design of conservation strategies for tropical dry forests
trees is discussed.
Keywords: change in land use, dispersal, ecological succession, gene flow, mahogany, metapopulation, microsatellite markers, over-exploitation, pollinators
Received 16 May 2002; revision received 3 December 2002; accepted 22 August 2003
Introduction
Tropical dry forests are considered the most threatened of
the major tropical forest types ( Janzen 1988). In Central
America, only ≈ 1% of the original area covered with dry
forest remains as such ( Janzen 1988; Sader & Joyce 1988;
Maas 1995; Sanchez-Azofeifa 1996; Kramer 1997). In Costa
Rica, most tropical dry forests were replaced with
grasslands for cattle raising between 1940 and 1975 (León
et al. 1982). Typically, pastures were established using
Hyparremia rufa ( Jaragua grass), a grass introduced from
Africa. This fire-tolerant grass allowed cattle ranchers to
maintain their pastures by burning the fields during the
Correspondence: O. J. Rocha. E-mail: [email protected]
© 2003 Blackwell Publishing Ltd
long dry season. However, more recently, extensive areas
of pasture have been abandoned because of the low price
of beef on the international market. These pastures are
likely to experience accelerated successional change following
abandonment, but accidental and provoked burning of
grassland during the dry season is a major threat to the
restoration of tropical dry forests ( Janzen 1988).
Fire is frequently cited as the most important factor controlling the dynamics of succession in tropical dry forests
( Janzen 1988). Typically, fires start by the combustion of
highly flammable fire-tolerant Jaragua grass and extend
into forested areas ( Janzen 1986). The main consequences
of these seasonal fires are: (i) the destruction of existing forest patches, (ii) modification of the edaphic and aerial
microclimate necessary for site recovery, and (iii) suppression of regeneration due to fire-induced mortality of
3202 M . C É S P E D E S E T A L .
seedlings and saplings ( Janzen 1986). Therefore, in order for
plant succession to occur, it is important to prevent fires.
Few studies have examined early plant succession in
abandoned pasture fields in tropical regions (Uhl et al.
1981; Uhl & Jordan 1984; Uhl 1988; Nepstad et al. 1990;
Kapelle et al. 1995; Zahawi & Augspurger 1999). Zahawi
& Augspurger (1999) studied early succession in four
abandoned unseeded fields of different age in Ecuador. They
found that, in general, species richness increased over
time; secondary forest with 25 – 40 years since abandonment had the highest diversity, whereas an open site with
5 – 8 years since abandonment had the lowest diversity. Uhl
& Jordan (1984) studied the dynamics of early succession
in Amazonia after forest cutting and burning, and found 56
tree species taller than 2 m after 5 years. More than half
of these were primary forest species. However, Kapelle
et al. (1995) reported that, in a successional gradient in a
montane forest in Costa Rica, early and late successional
forests are more diverse than primary forest. Moreover, they
concluded that secondary forest in the upper montane may
be as diverse as tropical lowland forest. However, the level
of genetic variation found in abandoned pastures as succession progresses has not been studied.
Metapopulation studies provide a new approach to
study the ecology and the genetics of populations at the
landscape level, which may be of special interest for conservation biology (McCullough 1996; Hanski 1999). This is
particularly true when populations function as a product
of local dynamics and the regional processes of migration, extinction and colonization (Hanski & Gilpin 1991;
Gillman 1997). In most cases, plant metapopulations occur
in habitats that represent temporal niches along successional
series (Poschlod 1996; Gillman 1997). They may also occur
in spatial niches of the natural landscape with small and
isolated areas, such as in tree fall or wind fall areas in
forests (Gillman 1997; Valverde & Silverton 1997) and
areas that experience fire occasionally (Menges et al. 1998).
Human activity also creates metapopulation scenarios, e.g.
by intensive land use, changes in agricultural practices,
and urban development that may result in fragmentation
of the landscape and temporal degradation of suitable
habitats (Rocha et al. 1997, 2002). Dispersal processes,
which play a key role in the survival of metapopulations,
have changed markedly since humans cultivated the landscape, and also during the development of the man-made
landscape (Poschlod 1996).
Metapopulation dynamics also affects the genetic structure and the level of population diversity and differentiation (Giles & Goudet 1997a). Gene flow is an important
factor in the genetic structure at the landscape level. Metapopulation structure may break down under high disturbance rates (Husband & Barret 1996; Valverde & Silvertown
1997; Ouborg et al. 1999). Because plants are relatively
immobile, present strong spatial structure and restricted
dispersal, habitat fragmentation and patch isolation may
lessen the probabilities of recolonization events (Giles &
Goudet 1997a; Gillman 1997; Ouborg et al. 1999). Thus, isolation by distance may result in a disruption of the structure of the metapopulation events (Giles & Goudet 1997a;
Gillman 1997). However, some authors have shown that
gene flow appears to be high for tropical plants even in
severely disturbed landscapes (Hamrick et al. 1991; Rocha
& Aguilar 2001a; White et al. 2002). But in most cases, ascertaining the implications on species or nature conservation
is only partially possible because of the lack in knowledge
about interactions between populations. Therefore, it is
currently more useful to study some important aspects of
the metapopulation concept, such as dispersibility and dispersal processes, to give this rather theoretical concept a
more practical basis in order to use it for nature conservation purposes.
Swietenia macrophylla (big leaf mahogany) is one of the
timber species considered to be threatened by the international timber trade. In general, mahoganies are among the
most commercially important tropical timber tree species —
dominating international trade in the areas where they are
native (Lamb 1966; Chalmers et al. 1994). Big leaf mahogany has been logged from Mexico and Central America
since the 17th century (Snook 1996). In Costa Rica, it is considered a threatened species by most foresters, as it is currently found in only a few locations ( Jiménez 1996). This is
particularly true in the dry areas of western Costa Rica
where mahogany has been harvested intensively.
In this study, we describe the levels of genetic variation
and differentiation in four populations of S. macrophylla
in abandoned pastures with different times since the
occurrence of the last disturbance. In addition, we compare
levels of variation found in these four sites with that found
in a selectively logged mature forest. We discuss the role
of long-distance dispersal of pollen and seeds for the conservation of genetic diversity.
Materials and methods
Study species
Swietenia macrophylla is a commercially important tree
naturally distributed from southern Mexico southward to
Colombia, Venezuela, and parts of the upper Amazon and
its tributaries in Peru, Bolivia and Brazil. The usual habitat
of this species are the dry, moist and wet lowland tropical
and subtropical forests up to 1400 m in altitude (Whitmore
1983). In Costa Rica it is found in the northwestern
province of Guanacaste and Los Chiles in the northern part
of the Province of Alajuela (Holdridge et al. 1997).
Adult trees can reach up to 45 m in height and 2 m
in diameter above the heavy buttresses (Chudnoff 1984;
Holdridge et al. 1997; Pennington & Sarukhan 1998).
© 2003 Blackwell Publishing Ltd, Molecular Ecology, 12, 3201– 3212
R E S T O R A T I O N O F G E N E T I C D I V E R S I T Y O F S W I E T E N I A 3203
Leaves are glabrous, paripinnate and with six to eight leaflets per leaf. In dry areas, trees of S. macrophylla loose their
leaves during the dry season for longer than trees in wet
areas. Flowers are minute, appear to be perfect, but in fact
only develop anthers or stigma, and typically appear with
new leaves formation at the end of the dry season. Fruits
are erect strongly wooded capsules up to 20 cm long, and
take a year to mature. Each fruit contains about 40 seeds.
Upon maturation, some of the valves of the capsule are
dropped and the winged seeds are exposed for wind dispersal (Whitmore 1983; Chudnoff 1984; Holdridge et al.
1997; Pennington & Sarukhan 1998).
Regeneration of mahogany is favoured by local disturbance. It has been reported that Mayan slash-and-burn
agriculture produced conditions favourable for its regeneration, as it improved seed germination and seedling
survival (Whitmore 1983). Big leaf mahogany has become
the wood of choice to replace Honduran mahogany S.
humilis and Caribbean mahogany S. mahogani, both now considered commercially extinct throughout much of their
ranges. Its populations have also been reduced throughout
its natural distribution, particularly in Central America
(Whitmore 1983).
Study site
This study was conducted at Santa Rosa National Park,
located in northwestern Costa Rica (10°45′ to 11°00′ N and
85°30′ to 85°45′ W) (Fig. 1). This park is part of the Guanacaste Conservation Area (ACG). According to the ecological
map of Costa Rica ( Tosi 1969), two life zones are present in
the area; however, this study was conducted in an area
classified as moist transition dry forest (Holdridge 1967;
Hartshorn 1983). The park includes a mosaic of forests of
different ages and abandoned pastures (Hartshorn 1983;
Janzen 1986; Gerhardt 1993).
The climate is highly seasonal, with a well-defined dry
season from late November to mid May. Annual rainfall is
between 800 and 2600 mm, with a mean of 1423.4 mm.
Mean annual temperature is 25.7 °C, and mean annual relative humidity is 81% (Rojas Jiménez 2001).
The vegetation in this region is a complex series of
habitats and vegetation types. It is further complicated by a
long history of man-induced disturbance, including forest
clearing for the establishment of pastures, selective logging, fragmentation of the landscape, cattle grazing and
annual burning of pastures to prevent woody encroachment and colonization (Hartshorn 1983). Since the establishment of Santa Rosa National Park, an intensive fire
control programme has been in place, and this has resulted
in a mosaic of different successional stages differing in the
time since the occurrence of the last fire episode.
Sample collection
We collected leaves from 20 individual trees and saplings
from 5 successional sites in the Santa Rosa National Park,
Guanacaste, Costa Rica. Leaf material was collected from
each individual and placed in a resealable zip-lock plastic
bag with at least 20 g of self-indicating silica gel to dehydrate the tissue while preserving the integrity of the DNA.
On return to the laboratory, the dry leaf material was
stored in paper bags, and kept at room temperature in the
dark until DNA extraction. The selection of the sites was
based on the occurrence of fire during the last 20 years in
the park, and on information provided by park personnel.
We selected sites with the following times since last disturbance: 6, 9, 15 and 20 years. These four sites had experienced
multiple fires over the last 50 years; therefore, the time
since the last disturbance refers specifically to fire episodes.
The sites have been named Pitilla I, Pitilla II, Las Mesas and
Principe, respectively. In addition, we also selected a mature
forest. This mature forest has experienced relatively low
levels of human disturbance ( Janzen 1983; Burnham 1997).
All individuals collected were marked and mapped. In
addition, we collected leaf samples of all reproductive
adults found within 1 km of the successional sites.
DNA extraction
Fig. 1 Location of study sites.
© 2003 Blackwell Publishing Ltd, Molecular Ecology, 12, 3201–3212
Total genomic DNA was extracted from leaves of S. macrophylla using a modification of the protocol described by
Karp et al. (1997) and Lodhi et al. (1994). Approximately
250 mg of leaf tissue was ground in liquid nitrogen with a
pestle in a 1.5-mL Eppendorf tube. The ground tissue was
mixed with 600 µL of 0.1 m Tris–HCl pH 8.0, 1.0 m NaCl,
2% CTAB, 2% PVP and 1% β-mercatoethanol extraction
3204 M . C É S P E D E S E T A L .
Locus
Repeated sequence
Primer (5′−3′)
Fragment size (bp)
mac44
(AC)13
184–196
mac49
(GT)27
mac52
(A)10(AC)11
mac59
(AAT)5(CAT)3
mac83
(CT)10(AC)14
TCCCTTCCAGCTTFTTCCTG
TGGCCCTTTCTCGATGTC
AGGGTCTTTGTTGATGAATACC
ATCATCGCCGTTGATACATT
GCGATGAATGTTGTACACCG
AGCAGCCCGGTACTAGCATGT
TGGAGTAAAGTCGAGGGCTG
GGCTGGATATGGCACTTGTT
CCTGAAAGCTTAATGGCCAC
TCAGCCTTTGAGGGATGAAC
Table 1 Oligonucleotide primer sequences,
repeat motifs, number of alleles, and product
sizes of the five 5 microsatellite loci used in
this study*
105–151
234–252
197–212
165–175
*Taken from White & Powell (1997).
buffer. The mix was incubated at 60 °C for 40 min with
agitation every 10 min. Samples were centrifuged for 10 min
at 12 000 g at 10 °C. The supernatant solution was placed
in another tube, and 1 vol. of chloroform–octhanol (24:1)
was added, mixed gently and centrifuged for 10 min. The
supernatant was transferred to another tube where 1/10 vol.
of 3 m NaAC pH 5.2 and 0.6 vol. of ice-cold isopropanol
were added. The mix was kept at 5 °C for 10–20 h. DNA
was precipitated with 70% ethanol, air dried and later
resuspended in 100 µL of 1.0 m TE buffer.
Genetic analysis
We studied the levels of genetic variation using five microsatellite marker loci developed by White & Powell (1997).
The forward and reverse primers for each microsatellite locus
are presented in Table 1. The polymerase chain reactions
(PCRs) were performed using a Rapidcycler (Idaho Technology) in a total volume of 20 µL containing 3.3 µL of 10×
PCR buffer (50 µm Tris–HCl pH 8.3, 2,5 mg/mL of BSA
and 10 µm MgCl2), 2 µL (0.3 units) Taq DNA polymerase
(Promega), 2.6 µL 0.32 mm dNTP, 20 ng of template DNA,
and 9.7 µL DNA free water. We also added 1.6 µL 0.4 µm
solution of forward and reverse primers. PCR conditions
were: 94 °C for 1 min, followed by 30 cycles of 5 s at 94 °C,
10 s at 55 °C and 35 s at 72 °C. A final extension step of
4 min at 72 °C was added after the last cycle. PCR products
were visualized on 40 cm polyacrilamide gels using silver
staining (Promega). The different alleles for each locus
were identified according to their molecular mass, where
the most anodal allele was arbitrarily identified as ‘1’, with
the remaining alleles identified sequentially (Fig. 2).
Data analysis
Genetic diversity at each site was quantified by means of
the number of alleles per locus (A), the effective number of
alleles per locus (AE), observed heterozygosity (HO) and
Nei’s expected heterozygosity (HE) (Nei 1973), for each
Fig. 2 Silver-stained polyacrilamide gel showing the six alleles
found in locus mac49.
locus and averaged over all loci. The significance of allele
frequency differences between locations was assessed
using the exact test (Weir 1996). The effective number of
alleles was estimated as the reciprocal of the homozygosity
(Hartl & Clark 1989). In addition, genetic differentiation
was determined using the infinite allele model FST (Weir &
Cockerham 1984). The degree of relatedness between locations, based on Nei’s genetic distances, was represented
in a tree using upgma. Bootstrap sampling of loci tested
the robustness of clusters of the tree, using 103 simulations.
These analyses were conducted using the program popgene
1.31 (Yeh et al. 1999).
Results
Allele richness
We found relatively high levels of genetic diversity among
the sampled individuals considering the degree of isolation
of the reproductive trees, and the level of fragmentation of
the forest. For the 5 microsatellite loci surveyed, there were
21 alleles identified in the 89–96 individual sampled
(Table 2). The level of variation was also high at each
successional site, as there were between 14 and 20 alleles in
each location (Table 2). We also observed more than one
allele in each locus examined, and most alleles were
present in more than one site (Table 2). Furthermore, all
alleles were present at frequencies higher than 1%.
© 2003 Blackwell Publishing Ltd, Molecular Ecology, 12, 3201– 3212
R E S T O R A T I O N O F G E N E T I C D I V E R S I T Y O F S W I E T E N I A 3205
Table 2 Allele frequency of five microsatellite loci of Swietenia macrophylla, for each of the five successional sites sampled at Santa Rosa
National Park, Guanacaste, Costa Rica
Population
Locus
Pitilla I
(6 years)
Pitilla II
(9 years)
Las Mesas
(15 years)
Principe
(20 years)
Mature
forest (100 years)
mac49
n (95)
1
2
3
4
5
6
18
0.4722
0.0833
0.0278
0.0278
0.3056
0.0833
19
0.4474
0.1842
0.0526
0.0263
0.2895
0.0000
20
0.3750
0.1750
0.1250
0.0250
0.0000
0.0000
19
0.1842
0.3421
0.3158
0.0000
0.0000
0.0000
19
0.5000
0.0000
0.0000
0.0000
0.0000
0.0000
mac59
n (89)
1
2
3
20
0.7000
0.1250
0.1750
20
0.7250
0.0750
0.2000
26
0.8125
0.0625
0.1250
17
0.6471
0.1176
0.2353
16
0.7188
0.1250
0.1563
mac83
n (91)
1
2
3
20
0.7000
0.0750
0.2250
19
0.8684
0.0000
0.1316
18
0.6944
0.0000
0.3056
17
0.6471
0.0000
0.3529
17
0.7059
0.0000
0.2941
mac44
n (90)
1
2
3
20
0.7250
0.0500
0.2250
17
0.8529
0.0000
0.1471
18
0.6944
0.0000
0.3056
17
0.6471
0.0000
0.3529
18
0.6389
0.0000
0.3611
mac52
n (96)
1
2
3
4
5
6
20
0.5750
0.1250
0.0750
0.0750
0.0000
0.1500
20
0.0750
0.4500
0.2500
0.1750
0.0250
0.0250
20
0.2750
0.2750
0.2250
0.1000
0.0000
0.1250
17
0.2750
0.5000
0.2000
0.0000
0.0250
0.0000
19
0.2368
0.3421
0.2632
0.0526
0.0000
0.1053
Table 3 Mean number of alleles (Na), mean effective number of alleles per locus (Ne), mean observed heterozygosity (HO), and Nei’s mean
expected heterozygosity (HE) for the five loci in the five populations of Swietenia macrophylla from Santa Rosa National Park, Guanacaste,
Costa Rica. Standard deviations are shown in parentheses
Site
Na
Ne
HO
HE
Pitilla I (6 years)
Pitilla II (9 years)
Las Mesas (15 years)
Principe (20 years)
Mature forest (> 100 years)
4.00 (1.41)
3.60 (1.82)
3.40 (1.52)
3.00 (1.00)
2.80 (1.30)
2.22 (0.59)
2.27 (1.11)
2.70 (1.44)
2.22 (1.45)
2.24 (0.93)
0.50 (0.33)
0.45 (0.34)
0.52 (0.34)
0.58 (0.25)
0.49 (0.34)
0.53 (0.11)
0.47 (0.23)
0.54 (0.21)
0.54 (0.09)
0.51 (0.14)
We found that the mean number of alleles per locus
varied between 4.0 and 2.8 (Table 3). However, the variation
in the mean effective number of alleles is smaller, ranging
between 2.22 and 2.70. Overall, there is a decrease in the
number of alleles found at each site as time since the last
© 2003 Blackwell Publishing Ltd, Molecular Ecology, 12, 3201–3212
disturbance progresses. In the early successional stages,
Pitilla I (6 years) and Pitilla II (9 years) we found more alleles
(20 and 18 alleles, respectively) than in the mature forest
(14 alleles). We found fewer alleles in the older of the two
intermediate successional sites; namely, Principe (20 years,
3206 M . C É S P E D E S E T A L .
with 15 alleles) and Las Mesas (15 years, with 17 alleles)
(Table 2).
Allele frequency
There is significant heterogeneity in allele frequencies
between sites for at least two loci. For loci mac49 and mac52
the homogeneity test of the allele frequencies across sites
revealed significant variation (χ2 = 66.84, P < 0.0001 and
χ2 = 52.56, P < 0.0001, respectively). For example, some
alleles appeared in only one or two sites, but were found in
relatively high frequencies (see allele 6 at locus mac49 or
allele 2 at locus mac83 in Table 2). In addition, for two loci,
mac49 and mac52, the most frequent allele also varies from
one location to another. But for the other three loci, the
most frequent allele is the same in all locations.
Mean HO ranged from 0.45 to 0.58, and HE ranged from
0.47 to 0.54. One locus, mac49, failed to conform with
Hardy–Weinberg equilibrium at three locations. These
deviations might result from the relative small sample size
in relation to the number of potential genotypes.
Genetic diversity
Mean FST for all loci was only 0.0631, indicating a low level
of genetic differentiation and high levels of gene flow
between sites (Table 4). The range of variation of the values
of FST for each locus was also low, with significant levels of
genetic exchange between sites. We also found that the
mean FIS was very low, indicating that within-site structure
(inbreeding or Wahlund effect) was not significant (Table 4).
However, some loci had high values of FIS, indicating
deficiency or excess of heterozygotes. Once again, these
values might result from the relatively small sample size in
relation to the number of potential genotypes.
Genetic differentiation among the five sites was calculated using Nei’s unbiased estimator of genetic distances
(Nei 1978). The genetic distance between sites varied
between 0.0272 and 0.1397 (Pitilla II–Las Mesas and Pitilla
I–Principe, respectively, Table 5). The population from
Principe was the most genetically distant from all other
populations in the study and resolved apart from the main
cluster in the dendrogram. Geographical distance is not
correlated with genetic distance. For example, the two
early successional sites, Pitilla I and Pitilla II, are adjacent,
but had a genetic distance of 0.0889. Similarly, El Principe
and mature forest, also adjacent to each other, presented a
genetic distance of 0.1656. In contrast, two other adjacent
sites, Las Mesas and the mature forest presented a low value
of genetic distance (0.0280). Moreover, the lowest genetic
distance was found between Pitilla I and Las Mesas (0.0272).
Genetic distances between the five sites in the Santa Rosa
National Park are represented in the tree shown in Fig. 3.
Relationship between adult trees and saplings
Our analysis also revealed that the reproductive trees
found in the vicinity of the sampling sites do not determine
the allelic diversity found in each site (Table 6a). Our data
show that often there are fewer alleles in the reproductive
trees than those found in the sampling plots. For example,
for locus mac49 we found six alleles at Pitilla I; however,
the adult trees in the vicinity bear only three alleles. In
addition, we also found that some alleles present in the
adult tree population were not found in our sample of
saplings and juveniles. This is the case for loci mac59 and
mac44. There is also significant heterogeneity in the allele
frequencies of seedling and saplings growing in each site
compared with that of the reproductive adult tree in the
vicinity (Table 6b).
The distribution of the sampled individuals may play an
important role in explaining these findings. Figure 4 shows
the distribution of all marked individuals in each of the five
Table 5 Nei’s unbiased estimates of genetic distance among the
populations of Swietenia macrophylla in five sites at the Santa Rosa
National Park, Guanacaste, Costa Rica
Location
Pitilla I
Pitilla II
Las Mesas
Principe
Pitilla I
Pitilla II
Las Mesas
Principe
Mature Forest
****
0.0889
0.0426
0.1805
0.0476
****
0.0272
0.1077
0.0394
****
0.0507
0.0309
****
0.1656
Table 4 Summary of F-statistics and gene flow for all loci
Locus
FIS
FIT
FST
Nm*
mac49
mac59
mac83
mac44
mac52
Mean
−0.5221
0.3046
0.2148
0.2735
0.0817
0.0155
−0.3385
0.3123
0.2356
0.2962
0.1494
0.0776
0.1206
0.0110
0.0265
0.0313
0.0737
0.0631
1.8221
22.3998
9.1817
7.7320
3.1424
3.7105
*Nm = 0.25(1 – FST)/FST.
Fig. 3 Dendrogram of the genetic relationship between sites.
© 2003 Blackwell Publishing Ltd, Molecular Ecology, 12, 3201– 3212
R E S T O R A T I O N O F G E N E T I C D I V E R S I T Y O F S W I E T E N I A 3207
(a) No. of alleles
Area I
Site
Juveniles
Pitilla I
Pitilla II
Adult trees
Area II
Site
Juveniles
Las Mesas
Principe
Mature forest
Adult trees
mac49
mac59
mac83
mac44
mac52
6
5
3
3
3
2
3
2
4
3
2
3
5
4
2
3
3
3
2
2
2
2
2
2
4
4
2
3
3
0.3000
0.2500
0.2500
0.2000
0.8500
0.1000
0.0500
0.6500
0.0000
0.3500
0.6000
0.2000
0.2000
0.1111
0.1111
0.4444
0.2778
0.0556
0.5000
0.2500
0.1667
0.0833
0.5000
0.0833
0.2500
0.1667
0.6667
0.0000
0.3333
0.5833
0.2500
0.1667
0.6000
0.3000
0.0000
0.0000
0.0000
0.1000
5
5
5
4
5
Table 6 (a) Comparison of the number of
alleles in the juveniles (seedlings and
saplings) of each site and the reproductive
adult trees in the vicinity. (b) Allele
frequencies for all loci examined among the
reproductive adult trees found in the
vicinity of the successional sites. Adult
trees are divided in two groups, those close
to sites Pitilla I and Pitilla II (Area I), and
those close to Las Mesas, Principe and
Mature forest (Area II). Asterisks indicate
heterogeneity in the allele frequency of
adult trees and juveniles at each area for
each locus (χ2 test, P < 0.01)
(b) Allele frequencies
Area I
Area II
Allele
1
2
3
4
5
Allele
1
2
3
4
5
6
locations surveyed according to size. It is clear that in the
early stages of succession most trees are found in the smaller
size categories, while in the later stages of succession more tall
trees are found. In addition, most of the sampled individuals
found in the mature forest died during the study period.
Discussion
Allele richness
Our results indicate that there are high levels of allelic
diversity in the populations of Swietenia macrophylla at all
sites considered in this study. Multiple reasons can be
invoked to explain such levels of genetic diversity reported.
First, the levels of allelic diversity, along with the high
levels of heterozygosity and the low values of FIS found in
this study, strongly suggest that this is a predominantly
outcrossing species. Other studies have also reached similar
conclusions, for example, Gillies et al. (1999) studying the
genetic diversity in Mesoamerican populations of S. macrophylla using random amplified polymorphic DNA (RAPD)
markers found that 80% of the variation was maintained
within populations. Based on their findings, they proposed
that S. macrophylla is a predominantly outcrossing species.
However, there is no direct estimation of the actual rate of
outcrossing in this species using progeny analyses.
© 2003 Blackwell Publishing Ltd, Molecular Ecology, 12, 3201–3212
Considering the distance between trees of S. macrophylla
isolated in pastures, for this species to be a predominant
outcrosser would require extensive movement of pollen.
White & Boshier (2000) demonstrated that extensive pollen
movement is likely in a highly fragmented landscape.
They showed that at extreme isolation, S. humilis trees
in pastures could receive pollen from trees located up to
4.5 km away. They also concluded that fragmentation does
not result in reproductive isolation, but rather in increased
levels of long-distance pollen flow. Moreover, White et al.
(2000) proposed that the increase in pollen flow found in
S. humilis counteracts the negative effects proposed for
habitat destruction and fragmentation. Similar findings
were reported by Rocha & Aguilar (2001b) for the dry forest tree Enterolobium cyclocarpum, where trees in pastures
showed outcrossing rates similar to that of trees in continuous forests. In addition, seeds produced on trees in
pastures are sired by more pollen donors than those from
trees in continuous forest (also see Apsit et al. 2001). However, the changes in pollinator behaviour required to explain
the increase in pollen flow are not known (but see Dicks 2001).
The majority of tropical rainforest species investigated
to date appeared to be outcrossers with extensive gene
flow (Ashton 1969; Bawa et al. 1985; Hamrick & Murawski
1991; Hamrick et al. 1991; Hall et al. 1994a,b, 1996; Doligez
& Joly 1997a). Isozyme studies conducted to determine the
3208 M . C É S P E D E S E T A L .
Fig. 4 Height distribution of the samplings
and trees found in each of the five locations.
mating system of these species further support the predominance of outcrossing among tropical rain forest trees
(O’Malley & Bawa 1987; Doligez & Joly 1997a; Nason &
Hamrick 1997). Doligez & Joly (1997a) recently reviewed
the outcrossing rates reported for 28 species of tropical forest trees in natural populations. However, very little is
known about how mating systems are modified in highly
fragmented or degraded areas.
The decrease in genetic diversity as the age of the stand
increases needs to be studied in more detail (Table 2). Even
though this finding might be confounded with a site effect,
it may also indicate the recruitment of new reproductive
individuals as time progresses, and/or the effects of nat-
ural selection causing differential mortality among saplings
as the structure of the community changes. In general,
there are significant differences in species composition,
and the vertical structure of the plant communities sampled in this study (Pacheco 1998). These changes in the
structure of the plant community could result in selective
pressures against certain genotypes, reducing genetic
diversity in the older sites where the canopy is more dense.
Once again, these results indicate the role that life history
traits, such as the duration of the juvenile phase, play in
determining the levels of population variation and differentiation (Whitlock & McCauley 1990; Mariette et al. 1997;
Auterlitz et al. 2000).
© 2003 Blackwell Publishing Ltd, Molecular Ecology, 12, 3201– 3212
R E S T O R A T I O N O F G E N E T I C D I V E R S I T Y O F S W I E T E N I A 3209
Genetic diversity
Despite the significant levels of heterogeneity in allelic
frequencies, the levels of genetic differentiation between populations was low (FST = 0.0631). It has been argued
that fragmentation and severe bottlenecks do not greatly
affect diversity, as both typically result in the loss of rare
alleles (Giles & Goudet 1997a,b; Lawrence & Marshall
1997). Because genetic diversity is measured using allele
frequencies, common alleles have a higher contribution to
diversity than uncommon and rare alleles. Therefore, the
loss of uncommon and rare alleles hardly affects diversity
(Giles & Goudet 1997a; Lawrence & Marshall 1997). We
found that alleles found in lower frequencies (< 0.01) were
less likely to be found in all sites. Moreover, the similarity
in the effective allele number, estimated as the reciprocal of
homozygosity, was similar at all sites indicating that, on
average, two alleles were over-represented at each site.
The similarity in genetic diversity between the populations
sampled indicates that there is no clear relationship genetic
between distance and geographical distance (Fig. 3).
Genetic similarity results from two main factors: (i) variation in the presence or absence of alleles between sites, and
(ii) variation in allele frequencies between sites (Table 2).
Two adjacent populations; namely, Pitilla I (6 years) and
Pitilla II (9 years) are not the most similar populations. On
the contrary, Pitilla II is more similar to Las Mesas (15 years),
located almost 4 km away. However, these two sites are
more similar to each other with respect to species composition, vertical structure and light environment than they
are to Pitilla I (Pacheco 1998). These findings suggest that
changes in environmental factors might play a key role in
the likelihood of germination, establishment and survival
of seedlings of S. macrophylla in each site. Whitmore (1983)
reported that open areas provide conditions more favourable for the growth and survival of seedlings and saplings
of this species. This suggests that each site is capturing a
different sample of the genetic diversity, as the number of
reproductive trees varies over time, and optimal conditions for the establishment of new individuals also change
as succession progresses. For this reason, it should be
expected that sites with similar community structure
should also be more similar in their genetic composition.
Relationship between adult trees and saplings
Our results clearly indicate that movement of pollen is not
limited to adjacent trees, as alleles absent among mature
trees in the vicinity of each site are found in seedlings and
saplings. Moreover, we also found significant heterogeneity
in allele frequencies between adult trees and juveniles, indicating that even the most abundant alleles show significant
differences in their frequencies (Table 6). These results
support the notion of extensive gene flow in fragmented
© 2003 Blackwell Publishing Ltd, Molecular Ecology, 12, 3201–3212
habitats proposed by other authors (White et al. 1999; White
& Boshier 2000). These findings strongly suggest that
metapopulation models could best be used to describe the
genetic structure of plants in highly fragmented landscapes.
It could be argued that in fragmented landscapes, the trees
in fragments and the remnant trees in pastures are not
isolated from each other, on the contrary, there is extensive
gene flow between all these subdivisions. In addition,
these data also illustrate how the few scattered trees left in
the pastures and the forest remnant serve as traps for
genetic variation, as pollen from distant trees sire their
seeds and contribute to enrich the local diversity of alleles.
Moreover, Rocha & Aguilar (2001b) also showed that
Enterolobium cyclocarpum trees left in pastures, and isolated
by at least 500 m from other conspecific trees, have a high
outcrossing rate, and mate with more trees than those in
the continuous forest. They also claim that trees in pastures
are important for promoting gene flow between forest
fragments, as they facilitate movement of pollinators
between fragments (also see Dicks 2001).
The occurrence of local extinction and recolonization
further supports the existence of a metapopulation structure among these populations (Sork et al. 1999). Accelerated recolonization of newly abandoned pastures by
S. macrophylla can only result from long-distance dispersal
of seeds. Moreover, high levels of genetic diversity, as
reported here, also require extensive movement of pollen
between trees. Under these circumstances it is reasonable
to conclude that human-induced changes at the landscape
level have not changed dispersal processes in a negative
way, thus suggesting a strong resilience in the mechanisms
of pollen and seed movement during the development of
the man-made landscape. Wade & McCauley (1988) proposed that the genetic effects of extinction and recolonization depend on the number of individuals that recolonize
the site and twice the number of migrants, and demonstrated that genetic differentiation might increase or
decrease depending on these values. Furthermore, in this
case, the level of genetic differentiation is typical of that of
other tropical trees in natural forests, indicating that the
contraction and expansion of this population did not
diminish gene flow. In general, these findings point out
that genetic diversity, in a nonequilibrium metapopulation
such as that studied here, is not only the result of gene flow,
but also depends on the patterns of colonization and the
life cycle of the species (Whitlock & McCauley 1990; Mariette
et al. 1997; Austerlitz et al. 2000).
It is important to note that, in addition to the negative impact of habitat fragmentation and degradation,
S. macrophyla has been logged in Central America since the
17th century (Snook 1996). The area that is now part of the
Santa Rosa National Park was logged long before the park
was established (Hartshorn 1983). The impact of current
logging practices on the genetic diversity of tropical trees
3210 M . C É S P E D E S E T A L .
has been examined by other authors (Hall et al. 1994b;
Doligez & Joly 1997b). Hall et al. (1994b) argued that selective logging of Carapa guianensis in Costa Rica did not
reduce genetic variation. They proposed that the high
population density and synchronous flowering promote
outcrossing and maintain high levels of genetic variation
within populations. In contrast, Doligez & Joly (1997b)
reported that the outcrossing estimates for Carapa procera in
logged plots in French Guiana are lower than in undisturbed plots, and argued that the decrease in density is
probably not the only cause of the decrease in outcrossing
rates on logged plots. Similar reasoning was proposed by
Gillies et al. (1999), who showed that increasing logging
significantly decreased genetic diversity in populations
of S. macrophyla. They also argued that the low levels of
population differentiation found in S. macrophylla have probably resulted from extensive gene flow. They also pointed
out that greater population differentiation should occur in
highly disturbed habitats, where changes in the insect
pollinator’s behaviour might leads to limited gene flow, or
where seed dispersal distances may be reduced. However,
the results reported here fail to support that notion, suggesting that the changes in pollinator’s behaviour favoured
long-distance gene flow as proposed by White et al. (2002).
We found that despite the intensity of logging experienced
in the past, genetic diversity was not reduced, as indicated
by allele richness and observed heterogeneity.
The results presented here illustrate a great potential for
the regeneration of important timber species after severe
episodes of over-exploitation, and habitat fragmentation
and degradation due to grazing and fire. It has been proposed that the reduction of continuous habitat into several
smaller spatially isolated patches represents a significant
threat to the long-term survival of many plant species
(Saunders et al. 1991; Nason et al. 1997). It has been argued
that reproductive isolation is one of the important consequences that fragmentation of the landscape may have on
the biota that remain in the smaller patches (Saunders et al.
1990). Moreover, reproductive isolation is often associated
with reduction in the size of plant populations which, in
turn, may result in drastic loss of genetic variability due to
drift, reduced gene flow, elevated inbreeding and inbreeding depression (Templeton et al. 1990; Young & Merriam
1994). However, here we argue that there are high levels of
gene flow between the population of S. macrophylla considered in this study. Moreover, our findings agreed
with those of Gillies et al. (1999), White & Boshier (2000),
and Rocha & Aguilar (2001a,b) demonstrating high levels
of gene flow, and changes in pollinator behaviour that
allow the long-distance movement of pollen. However,
Rocha & Aguilar (2001a,b) also demonstrated that despite
the similarity in outcrossing rates between trees in pastures and trees in continuous forest, the progeny of trees
from the latter are more vigorous than that of the former.
Acknowledgements
The authors thank M.E. Zaldivar, A.G. Stephenson, L. Castro,
E. Castro, G. Aguilar, M. Artavia and Ana Araya for advice,
comments, and/or criticism on a previous version of this manuscript. We also thank D. Charlesworth, A.C.M. Gillies, and two
anonymous referees for their comments and suggestions. The
Guanacaste Conservation Area (ACG) and the staff of Parque
Nacional Santa Rosa for their collaboration. This work was supported by the International Plant Genetic Resources Institute and
the Center for International Forestry Research (grants 96/073, 97/
052, 98/049, and 00/066), a University of Costa Rica grant (VI-11191-223) to OJR, and a Andrew W. Mellon Foundation grant to MG,
NMH, and OJR.
References
Apsit V, Hamrick JL, Nason JD (2001) Breeding neighborhood size
of a Costa Rican dry forest species. Heredity, 92, 415–420.
Ashton PS (1969) Speciation among tropical forest trees: some
deductions in the light of recent evidence. Biological Journal of the
Linnean Society, 1, 155–196.
Austerlitz F, Mariette S, Machón N, Gouyon PH, Godelle B (2000)
Effects of colonization processes on genetic diversity: differences between annual plants and tree species. Genetics, 154,
1309–1321.
Bawa KS, Perry DR, Beach JH (1985) Reproductive biology of tropical
lowland rain forest trees. I. Sexual systems and incompatibility
mechanisms. American Journal of Botany, 72, 331–345.
Burnham RJ (1997) Stand characteristics and leaf litter composition of a hectare in Santa Rosa National Park, Costa Rica.
Biotropica, 29, 384–395.
Chalmers KJ, Newton AC, Waugh R, Wilson J, Powell W (1994)
Evaluation of the extent of genetic variation in mahoganies
(Meliaceae) using RADP markers. Theoretical and Applied Genetics,
89, 504–508.
Chudnoff M (1984) Tropical Timbers of the World. USDA Forest Service, Ag. Handbook no. 607. USDA, Washington, DC, USA.
Dicks C (2001) Genetic rescue of remnant tropical trees by alien
pollinators. Proceedings of the Royal Society of London, 268, 2391–
2396.
Doligez A, Joly HI (1997a) Mating system of Carapa procera
(Meliaceae) in French Guiana tropical forest. American Journal of
Botany, 84, 461–470.
Doligez A, Joly HI (1997b) Genetic diversity and spatial structure
within a natural stand of a tropical forest tree species, Carapa
procera (Meliaceae), in French Guiana. Heredity, 79, 72–82.
Gerhardt K (1993) Tree seedling development in tropical dry
abandoned pastures and secondary forest in Costa Rica. Journal
of Vegetation Sciences, 4, 95–102.
Giles BE, Goudet J (1997a) Genetic differentiation in Silene dioica
metapopulations: estimation of spatial–temporal effects in a
successional plant species. American Naturalist, 149, 507–526.
Giles BE, Goudet J (1997b) A case study of genetic structure in
a plant metapopulation. In: Metapopulation Biology; Ecology,
Genetics and Evolution (eds Hanski IA, Gilpin ME), pp. 429– 454.
Academic Press, San Diego.
Gillies ACM, Navarro C, Lowell AJ, Hernández M, Wilson J,
Cornelius JP (1999) Genetic diversity in Mesoamerican populations
of mahogany (Swietenia macrophylla), assessed using RAPDs.
Heredity, 83, 722–732.
© 2003 Blackwell Publishing Ltd, Molecular Ecology, 12, 3201– 3212
R E S T O R A T I O N O F G E N E T I C D I V E R S I T Y O F S W I E T E N I A 3211
Gillman M (1997) Plant population ecology. In: Plant Genetic
Conservation: the in situ Approach (eds Maxted N, Ford-Lloyd BV
Hawkes JG), pp. 114 –131. Chapman & Hall, London.
Hall P, Chase MR, Bawa KS (1994a) Low genetic variation but high
population differentiation in a common tropical forest tree
species. Conservation Biology, 8, 471– 482.
Hall P, Orrell LC, Bawa KS (1994b) Genetic diversity and mating
system in a tropical tree, Carapa guianensis (Meliaceae). American
Journal of Botany, 81, 1104 –1111.
Hall P, Walker S, Bawa KS (1996) Effects of forest fragmentation on
diversity and mating systems in a tropical tree Pithecellobium elegans. Conservation Biology, 10, 757–768.
Hamrick JL, Godt MJW, Murawski DA, Loveless MD (1991) Correlation between species traits and allozyme diversity: implications for conservation biology. In: Genetics and Conservation of
Rare Plants (eds Falk DA, Holsinger KE), pp. 75 – 86. Oxford University Press, New York.
Hamrick JL, Murawski DA (1991) Levels of allozyme diversity in
populations of uncommon neotropical tree species. Journal of
Tropical Ecology, 7, 395 –399.
Hanski I (1999) Metapopulations. Oxford University Press, Oxford.
Hanski I, Gilpin M (1991) Metapopulation dynamics: brief history
and conceptual domain. Biological Journal of the Linnean Society,
42, 3–16.
Hartl DL, Clark AG (1989) Principles of Population Genetics, 2nd
edn. Sinauer Associates, Sunderland, MA.
Hartshorn GS (1983) Plants. In: Natural History of Costa Rica (ed.
Janzen DH), pp. 118–183. Chicago University Press, Chicago.
Holdridge LR (1967) Life Zone Ecology. Tropical Science Center,
San José, Costa Rica.
Holdridge LR, Poveda LJ, Jiménez Q (1997) Arboles de Costa Rica,
Vol. I, 2nd edn. Tropical Science Center, San José, Costa Rica.
Husband BC, Barrett SCH (1996) A metapopulation perspective in
plant population biology. Journal of Ecology, 84, 461– 469.
Janzen DH (1983) Natural History of Costa Rica. Chicago University
Press, Chicago.
Janzen DH (1986) Seed dispersal by fruits-eating birds and mammals. In: Seed Dispersal (ed. Murray DR), pp. 123 –189. Academic
Press, San Diego.
Janzen DH (1988) Tropical dry forest: the most endangered major
tropical ecosystem. In: Biodiversity (ed. Wilson EO), pp. 130–137.
National Academy Press, Washington, DC.
Jiménez Q (1996) Árboles maderables en peligro de extinción en Costa
Rica. Editorial INCAFO, San José, Costa Rica.
Kapelle M, Kennis PAF, de Vries RAJ (1995) Changes in diversity
along successional gradients in a Costa Rican upper montane
Quercus forest. Biodiversity and Conservation, 4, 10 –34.
Karp A, Ingram DS, Isaac PG (1997) Molecular Tools for Screening
Biodiversity: Plants and Animals. Kluwer Academic, Dordrecht,
The Netherlands.
Kramer EA (1997) Measuring landscape changes in remnant tropical
dry forest. In: Tropical Forests Remnants: Ecology, Management and
Conservation of Fragmented Communities (eds Laurance WF,
Bierregaard RO Jr), pp. 386 – 399. Chicago University Press, Chicago.
Lamb FB (1966) Mahogany of Tropical America: Its Ecology and Management. University of Michigan Press, Ann Arbor.
Lawrence MJ, Marshall DF (1997) Plant population genetics. In:
Plant Genetic Conservation: the in Situ Approach (eds Maxted N,
Ford-Lloyd BV, Hawkes JG), pp. 114 –131. Chapman & Hall,
London.
León JC, Barboza C, Aguilar J (1982) Desarrollo tecnológico de la
ganadería de carne. CONICIT, San José, Costa Rica.
© 2003 Blackwell Publishing Ltd, Molecular Ecology, 12, 3201–3212
Lodhi MA, Ye GN, Weeden NF, Reisch BI (1994) A simple and efficient method for DNA extraction from grapevine cultivars and
vitis species. Plant Molecular Biology Reporter, 12 (1), 6–13.
Maas JM (1995) Conversion of tropical dry forest to pasture and
agriculture. In: Seasonally Dry Tropical Forest (eds Bullock SH,
Mooney HA, Medina E), pp. 399–422. Cambridge University
Press, Cambridge.
Mariette S, Lefranc M, Legrand P, Taneyhill D, Frascaria-Lacoste
N, Machon N (1997) Genetic variability in wild cherry populations in France: effects of colonization processes. Theoretical and
Applied Genetics, 94, 904–908.
McCullogh DR (1996) Metapopulations and Wildlife Conservation.
Island Press, Washington, DC.
Menges ES, Hawkes CV, Platt WJ, Peet RK (1998) Interactive
effects of fire and microhabitat on plants of Florida scrub. Ecological Applications, 8, 935–946.
Nason JD, Aldrich PR, Hamrick JL (1997) Dispersal and the
dynamics of genetic structure in fragmented tropical tree
populations. In: Tropical Forest Remnants: Ecology Management
and Conservation of Fragmented Communities (eds Laurance WF,
Bierrgaad RO Jr), pp. 304 –320. University of Chicago Press,
Chicago.
Nason JD, Hamrick JL (1997) Reproductive and genetic consequences of forest fragmentation: two case studies of neotropical
trees. Journal of Heredity, 88, 264–276.
Nei M (1973) Analysis of genetic diversity in subdivided populations. Proceedings of the National Academy of Sciences of the USA,
70, 3321–3323.
Nei M (1978) Estimation of average heterozygosity and genetic distance from a small number of individuals. Genetics, 89, 583 –590.
Nepstad DC, Uhl C, Serrao EAS (1990) Surmounting barriers to
forest regeneration in abandoned, highly degraded pastures: a
case study from Paragominas, Pará, Brazil. In: Alternatives to
Deforestation: Steps Toward Sustainable Use of the Amazon Rainforest (ed. Anderson AB), pp. 215–229. Columbia University Press,
New York.
O’Malley DM, Bawa KS (1987) Mating systems of a tropical rain
forest tree species. American Journal of Botany, 82, 501–506.
Ouborg NJ, Piquot J, van Groenendael JM, van Groenendael JM
(1999) Population genetics, molecular markers and the study of
dispersal in plants. Journal of Ecology, 87, 551–568.
Pacheco A (1998) Inventario de la composición florística de un bosque seco
ubicado en el Parque Nacional Santa Rosa. Informe de practica de especialidad. Instituto Tecnológico de Costa Rica, Cartago, Costa Rica.
Pennington TD, Sarukhan J (1998) Arboles tropicales de México. Universidad Nacional Autónoma de México Fondo de Cultura
Económica, México D.F. México.
Poschlod P (1996) The metapopulation concept — a consideration
from the plant ecological viewpoint. Zietschrift für Okologie und
Naturschultz, 5, 161–185.
Rocha OJ, Aguilar G (2001a) Variation in the breeding behavior of
the dry forest tree Enterolobium cyclocarpum (guanacaste) in
Costa Rica. American Journal of Botany, 88, 1600–1606.
Rocha OJ, Aguilar G (2001b) Reproductive biology of the dry
forest tree Enterolobium cyclocarpum (guanacaste) in Costa Rica:
a comparison between trees left in pastures and trees in continuous forest. American Journal of Botany, 88, 1607–1614.
Rocha OJ, Degreff J, Barrantes D, Castro E, Macaya G (2002) Metapopulation dynamics of lima beans (Phaseolus lunatus) in the
Central Valley of Costa Rica. In: Managing Plant Genetic
Resources (eds Engels JMM, Brown AHD, Jackson MT), pp. 205 –
215. CAB International, Wallingford.
3212 M . C É S P E D E S E T A L .
Rocha OJ, Macaya G, Baudoin JP (1997) Causes of local extinction
and recolonization, determined by 3 years of monitoring wild
populations of Phaseolus lunatus L. in the Central Valley of Costa
Rica. Plant Genetic Resources Newsletter, 112, 44 – 48.
Rojas Jiménez K (2001) Fenología de la copa y del sistema de raices
finas de Enterolobium cyclocarpum (guanacaste) un árboil de
brotación temprana en el bosque tropical seco. MSc Thesis,
Sistema de Estudio de Posgrado, Universidad de Costa Rica.
Sader SA, Joyce AT (1988) Deforestation rates and trends in Costa
Rica. Biotropica, 20, 11–19.
Sanchez-Azofeifa GA (1996) Assessing land use/cover changes in
Costa Rica. PhD Dissertation, University of New Hampshire,
New Hampshire.
Saunders DA, Hobbs RJ, Margules CR (1990) Biological consequences of ecosystem fragmentation: a review. Conservation
Biology, 5, 18–32.
Snook LK (1996) Catastrophic disturbances, logging and the ecology of mahogany (Swietenia macrophylla King): grounds for
listing a major tropical timber species in CITES. Botanical Journal
of the Linnean Society, 122, 35 – 46.
Sork VL, Nason J, Campbell DR, Fernández JF (1999) Landscape
approaches to historical and contemporary gene flow in plants.
Trends in Ecology and Evolution, 14, 219 –224.
Templeton AR, Shaw K, Routman E, Davis S (1990) The genetic
consequences of habitat fragmentation. Annals of the Missouri
Botanical Garden, 77, 13 –27.
Tosi JA (1969) Mapa ecológico de Costa Rica. Tropical Science Center,
San José, Costa Rica.
Uhl C (1988) Restoration of degraded land in the Amazon basin.
In: Biodiversity (ed. Wilson EO), pp. 326–332. National Academy
Press, Washington, DC.
Uhl C, Clark K, Clark H, Murphy P (1981) Early plant succession
after cutting and burning in the upper Rio Negro region of the
Amazon Basin. Journal of Ecology, 69, 631– 649.
Uhl C, Jordan CF (1984) Succession and nutrient dynamics following
forest cutting and burning in Amazonia. Ecology, 65, 1476–1490.
Valverde T, Silvertown J (1997) A metapopulation model for
Primula vulgare, a temperate forest understorey herb. Journal of
Ecology, 85, 193–210.
Wade MJ, McCauley DE (1988) Extinction and recolonization:
their effects on the genetic differentiation of local populations.
Evolution, 42, 995–1005.
Weir BS (1996) Genetic Data Analysis II. Sinauer Associates, Sunderland, MA.
Weir BS, Cockerham CC (1984) Estimating F-statistics for the analysis of population structure. Evolution, 38, 1358–1370.
White G, Boshier H (2000) Fragmentation in Central American dry
forests: genetic impacts on Swietenia humilis (Meliaceae). In:
Genetics, Demography and Viability of Fragmented Populations (eds
Young AG, Clarke GM), pp. 293–311. Cambridge University
Press, Cambridge.
White G, Boshier H, Powell W (1999) Genetic variation within a
fragmented population of Swietenia humilis Zucc. Molecular Ecology, 8, 1899–1909.
White G, Boshier H, Powell W (2002) Increased pollen flow
counteracts fragmentation in a tropical dry forest: example from
Swietenia humilis Zuccarini. Proceedings of the National Academy
of Science, 99, 2038–2042.
White G, Powell W (1997) Isolation and characterization of microsatellite loci in Swietenia humilis (Meliaceae): an endangered
tropical hardwood species. Molecular Ecology, 6, 851–860.
Whitlock MC, McCauley DE (1990) Some population genetic
consequences of colony formation and extinction: genetic
correlation within founding groups. Evolution, 44, 1717–1724.
Whitmore JL (1983) Swietenia macrophylla and S. humilis (Caoba,
Mahogany). In: Natural History of Costa Rica (ed. Janzen DH),
pp. 331–333. Chicago University Press, Chicago.
Yeh FC, Yang R, Boyle T (1999) POPGENE, Version 1.31: Microsoft
Window-Based Freeware for Population Genetic Analysis. University of Alberta, Edmonton, Canada.
Young A, Merriam HG (1994) Effects of forest fragmentation on
the spatial genetic structure of Acer saccharum Marsh (sugar
maple) populations. Heredity, 72, 201–208.
Zahawi RA, Augspurger CK (1999) Early plant succession in abandoned pastures in Ecuador. Biotropica, 31, 540–552.
The authors are population biologists interested in a diverse range
of topics related to conservation of biological diversity in the
Neotropics. In particular, we are interested in the impact of
man-made disturbances, such as habitat fragmentation and
degradation, and over-exploitation, on genetic diversity of tropical
plants. This study is part of a project that examines the changes in
species composition, community structure, and physiological
traits of plants during the process of succession that follows
pasture abandonment. In this context, restoration of genetic
diversity is an important component.
© 2003 Blackwell Publishing Ltd, Molecular Ecology, 12, 3201– 3212