ª 2004 The American Genetic Association
Journal of Heredity 2004:95(3):234–243
DOI: 10.1093/jhered/esh043
High Genetic Diversity in Sarracenia
leucophylla (Sarraceniaceae),
a Carnivorous Wetland Herb
Z-F. WANG, J. L. HAMRICK,
AND
M. J. W. GODT
From the South China Institute of Botany, Chinese Academy of Sciences, Guangzhou 510650, P. R. China (Wang);
Department of Plant Biology, University of Georgia, Athens, GA 30602 (Hamrick and Godt); and
Department of Genetics, University of Georgia, Athens, GA 30602 (Hamrick).
Address correspondence to Mary Jo W. Godt at the address above, or e-mail: [email protected].
Abstract
Eighteen allozyme loci were used to examine genetic diversity in 10 natural populations of Sarracenia leucophylla Raf., a pitcher
plant restricted to the southeastern United States. One ex situ population propagated for restoration in Georgia was also
analyzed. S. leucophylla is an insect-pollinated, outcrossing perennial wetland herb that is threatened over much of its
geographic range. Fifteen loci (83.3%) were polymorphic, with a mean number of alleles of 3.33. Compared to species
having similar life-history traits and to previously analyzed Sarracenia species, S. leucophylla displayed unexpectedly high genetic
diversity. For example, genetic diversity within the species (Hes) was 0.224 and mean population genetic diversity (Hep) was
0.183. Although small S. leucophylla populations maintained less genetic diversity than larger ones, these differences were not
statistically significant. Nonetheless, this suggests that small populations may have lost rare alleles. Statistically significant
genetic differentiation among populations was found (h ¼ 0.192, P , .01), although it was not atypical considering the
species’ life-history characteristics. A significant correlation (P , .01) between genetic and geographic distance was found,
indicating an isolation-by-distance effect. However, the correlation coefficient for this relationship was low (r ¼ 0.46),
suggesting that factors other than gene flow play a prominent role in the geographic distribution of genetic diversity within
the species. The ex situ population captured most of the allozyme variation found in its source population.
It is widely accepted that genetic variation is of fundamental
importance for species’ conservation (Barrett and Kohn
1991; Ellstrand and Elam 1993; Gilpin and Soulé 1986;
Hamrick and Godt 1996a; Karron 1997; Lande 1999).
Because many small populations generally have low genetic
diversity, their capacity to adapt to environmental change
may be diminished and their ability to survive over the long
term may be compromised (Lande 1999). Small populations
are also prone to inbreeding and genetic drift (Ellstrand and
Elam 1993; Karron 1997; Lande 1999). Inbreeding typically
reduces population fitness through the increased expression
of recessive deleterious alleles as homozygosity increases
(Lande 1999). Genetic drift is expected to randomly reduce
genetic variation within small populations. In particular, lowfrequency alleles, which can be associated with population
fitness, are likely to be lost (Barrett and Kohn 1991). In small
populations, the random loss of self-incompatibility alleles
may also directly reduce the reproductive capacity of
individuals (DeMauro 1993; Karron 1997; Young et al.
1999) and may lead to population extinction (Barrett and
234
Kohn 1991; Holsinger et al. 1999). These genetic factors,
particularly when combined with demographic stochasticity,
may cause small populations to fall into ‘‘extinction
vortices,’’ feedback loops that result in the number of
individuals becoming smaller and smaller until populations
become extinct (Gilpin and Soulé 1986). Although populations are expected to lose genetic diversity at a rate dependent on their effective population size, not all small
populations are genetically depauperate (Ellstrand and Elam
1993; Frankham 1997; Gitzendanner and Soltis 2000; Godt
and Hamrick 1998). Factors such as species’ life-history
traits, biogeography, and gene flow also play critical roles in
determining the current genetic composition of populations
(Hamrick and Godt 1996a,b; Holsinger et al. 1999). Therefore, understanding genetic factors that contribute to extinction risks for particular species is critically important for
their conservation (Godt and Hamrick 1998; Hamrick and
Godt 1996a).
Sarracenia leucophylla Raf., the white-topped pitcher plant,
is restricted to the southeastern United States (Juniper et al.
Wang et al. Genetic Diversity in Sarracenia leucophylla
range. Despite its threatened status, S. leucophylla remains
popular in the cut flower and horticultural trades, and large
numbers of pitchers are illegally collected (Robbins 1998).
Successful in situ and ex situ conservation programs can
only be achieved if genetic diversity within and among natural
populations is understood. Knowledge of genetic diversity
and structure can prioritize natural populations for in situ
conservation and permit the choice of genetically robust
populations for sources of propagules for ex situ conservation (Barrett and Kohn 1991; Hamrick and Godt 1996a). The
main objectives of our study were to assess allozyme variation within and among natural populations of S. leucophylla,
examine relationships between genetic variation and population size, provide indirect estimates of historical levels of
gene flow among populations, and examine associations
between genetic and geographic distance. We also compared
genetic diversity in an ex situ population propagated for
restoration with its source population to determine whether
the ex situ population captured a representative portion of
the genetic diversity of the wild population.
Materials and Methods
Species Biology
Figure 1. Approximate historical distribution of
S. leucophylla (shaded) (Schnell 1976, Juniper et al. 1989) and
collection locations (dots). UPGMA dendrogram of the 11
S. leucophylla populations based on Nei’s (1978) unbiased genetic
distance.
1989; Schnell 1976) (Figure 1). It is found in moist habitats
such as bogs, swamps, and sandy savannahs that are usually
acidic and low in nutrients (Robbins 1998). Due to habitat
destruction and wetland alteration, S. leucophylla is considered
at risk by the International Union for Conservation of
Nature and Natural Resources (IUCN), with the following
designations for the states in which it is found: ‘‘extinct/
endangered’’ in Georgia, ‘‘rare’’ in Alabama, ‘‘vulnerable’’ in
Mississippi (Walter and Gillett 1998), and ‘‘endangered’’ in
Florida (Chafin L, Florida Natural Areas Inventory, personal
communication). Currently there is only one extant
population (numbering less than 100 plants) in Georgia
(Patrick T, Georgia Natural Heritage Program, personal
communication). According to Robbins (1998), S. leucophylla
is imperiled in Mississippi (with 6–20 occurrences, or fewer
than 3000 individuals) and vulnerable in Alabama and
Florida (with 21–100 occurrences, or fewer than 10,000
individuals). S. leucophylla is listed in Appendix II of the
Convention on International Trade in Endangered Species
of Wild Fauna and Flora (CITES). The global heritage rank
status (December 1997) for the species is G3, which indicates
that the species is vulnerable to extinction over its entire
Sarracenia leucophylla is a carnivorous perennial herb. Its
hollow, trumpet-shaped leaves (pitchers) arise from a rhizome, grow to 1 m in height (Patrick et al. 1995), and form
a rosette. Pitcher plants spread clonally via short rhizomes,
giving older, larger plants a ‘‘clump-like’’ appearance. Pitchers
are green at the base and become white toward the top
(hood). The hood is generally white with red and/or green
veins. Pitchers are pitfall traps containing enzymes that digest
insects and small animals. These prey provide nutrients to the
plants, which grow in nitrogen-poor soils (Juniper et al. 1989;
Schnell 1976; Slack 1979; Swenson 1977).
Sarracenia leucophylla flowers are maroon colored, umbrella
shaped, and 6–7 cm in diameter (Patrick et al. 1995). The
species flowers from March to April (Patrick et al. 1995). It is
insect pollinated, primarily by bumblebees, and has a floral
structure that encourages outcrossing (Schnell 1976; Slack
1979; Walcott 1935). Seeds lack specialized external
morphological features and are dispersed primarily by gravity
and water (Schnell 1976).
The regeneration and persistence of S. leucophylla is linked
to fire, which reduces woody plant competition and helps
maintain open, sunlit habitats (Juniper et al. 1989; Schnell
1976). Fire suppression is a major threat to the survival of
many Sarracenia species. Currently many southeastern Sarracenia populations are restricted to roadsides and power line
rights-of-way where woody vegetation is controlled and open
habitat is maintained (Godt MJW, personal observation).
Collection and Genetic Analysis
Leaf samples (pitchers) were collected from 10 natural
populations of S. leucophylla (Figure 1) and from an ex situ
population located at the Atlanta Botanical Garden (Atlanta,
235
Journal of Heredity 2004:95(3)
GA) (Table 1). Precise population locations are not given
because of the species’ threatened status. Forty-eight plants
were sampled from each population by removing a pitcher,
or a portion of one. In several populations, the pitchers of
some plants had senesced. In this case, phyllodia (basal
leaves) were sampled in lieu of pitchers. Leaves were placed
in plastic bags and stored on ice during transport to the
laboratory. Population sizes were roughly estimated in the
field by inspection.
In the laboratory, the leaves were crushed under liquid
nitrogen using mortars and pestles. Clean ocean sand was
added to the mortars to facilitate crushing. An extraction
buffer (Mitton et al. 1979) was added to solubilize and
stabilize the enzymes. Enzyme extracts were absorbed onto
chromatography paper wicks that were placed in 96-well
microtest plates and stored in an ultracold freezer (708C)
until used in electrophoretic analyses.
Four electrode buffer systems and 11 enzyme stains
resolved 18 loci on 12% starch gels. The 11 enzymes stained
were aldolase (ALD), aspartate aminotransferase (AAT),
diaphorase (DIA), fluorescent esterase (FE), isocitrate
dehydrogenase (IDH), malate dehydrogenase (MDH), 6phosphogluconate dehydrogenase (6-PGDH), phosphoglucoisomerase (PGI), shikimate dehydrogenase (SKDH), triose
phosphate isomerase (TPI), and uridine diphosphoglucose
pyrophosphorylase (UGPP). Buffer systems (following) were
numbered as in Soltis et al. (1983). Ald, Idh, and Skdh were
resolved with buffer system 4; Dia, Pgi-1, Pgi-2, Tpi-1, Tpi-2,
Ugpp-1, and Ugpp-2 were resolved with buffer system 6. A
modified buffer system 8 (recipe available from the authors
upon request) was used to resolve Aat-1, Aat-2, and Fe. Mdh1, Mdh-2, Mdh-3, 6Pgdh-1, and 6Pgdh-2 were resolved with
buffer system 11. Enzyme stains followed Soltis et al. (1983),
except for AAT and DIA, which followed Cheliak and Pitel
(1984), and UGPP, which is described in Manchenko (1994).
Loci and alleles were numbered consecutively from the most
anodal form. Two individuals from the Atlanta Botanical
Garden ex situ collection had anomalous banding patterns.
These plants were suspected to be hybrids and were excluded
from the analyses.
Data Analysis
Genetic diversity was described by allele frequencies, the
percentage of polymorphic loci (P ), the mean number of
alleles per locus ( A ), and per polymorphic locus ( A P ), the
mean effective number of alleles per locus ( Ae; Kimura and
Crow 1964), observed heterozygosity (Ho), and unbiased
expected heterozygosity (He; Nei 1978). All calculations were
made using POPGENE 1.31 (Yeh et al. 1999), except AP,
which was calculated using GDA 1.1 (Lewis and Zaykin
2001). Within the text, genetic diversity parameters subscripted with an ‘‘s’’ indicate species values, while those
subscripted with a ‘‘p’’ indicate population means. A
multilocus fixation index (FIS) was calculated for each
population (Weir 1996) using GDA 1.1.
Deviations from Hardy-Weinberg equilibrium for each
locus and population were tested by the Markov chain
236
Table 1. Sarracenia leucophylla collection sites and population
size estimates
a
Population
Location
Estimated
population sizea
EX
GA
FL1
FL2
FL3
FL4
FL5
AL1
AL2
AL3
MS
Atlanta Botanical Garden, GA
Americus, GA
Tate’s Hell 1, FL
Tate’s Hell 2, FL
Pine Log State Forest, FL
Eglin Air Force Base, FL
Blackwater State Forest, FL
AL 61, AL
Roscoe Rd, AL
Deer Park, AL
Kertz State Forest, MS
Ex situ population
Small
Medium
Large
Small
Medium
Medium
Large
Large
Large
Medium
Small, ,100 individuals; medium, 100–1000 individuals; large, .1000
individuals.
method of exact probability (Guo and Thompson 1992)
using GENEPOP 3.3 (Raymond and Rousset 1995). Markov
chain parameters for statistical tests were 1000 batches with
10,000 iterations per batch.
The degree of genetic structuring for each locus was
investigated by Wright’s F-statistics (Wright 1978) as
calculated with Weir and Cockerham’s (1984) estimators
using FSTAT 2.9.3. Wright’s F-statistics (1978) partition
total heterozygosity deficiencies (FIT) into components due
to deficiencies within populations (FIS) and subdivision
among populations (FST). FIT, FIS, and FST correspond to F,
f, and h, respectively, in Weir and Cockerham’s (1984)
notation. Confidence intervals (95% and 99%) were obtained
by bootstrap resampling the data set 15,000 times. Nei’s GST
value (1973), a measure of genetic differentiation among
populations, was also calculated using FSTAT 2.9.3.
Nei’s unbiased genetic distances and identities (1978) were
calculated for all population pairs using POPGENE 1.31
(Yeh et al. 1999). A dendrogram based on the unweighted
pair group method with arithmetic mean (UPGMA) was
generated using POPGENE 1.31.
An indirect estimate of historical levels of gene flow was
computed from h according to the equation Nm ¼ 0.25(1 h)/h (Wright 1931). Relationships between pairwise population estimates of gene flow (Nm) and geographic distances
were examined to test for isolation by distance by Mantel’s
tests (Mantel 1967) as described by Bohonak (2002). A
significant negative correlation between gene flow and
geographic distance implies that distance is a predictor of
population differentiation (Slatkin 1993).
Results
The 18 loci had a total of 53 alleles in the 11 populations
studied. Three of the loci (Aat-2, Ald, and Skdh) were
monomorphic in all populations. Polymorphic loci had two
to four alleles, except for Pgi-2 and Tpi-1, which had seven
alleles each. For most polymorphic loci (e.g., 6Pgdh-1, 6Pgdh-
Wang et al. Genetic Diversity in Sarracenia leucophylla
Table 2.
Sarracenia leucophylla allele frequencies for 15 polymorphic loci
Allele
6Pgdh-1
6Pgdh-2
Aat-1
Dia
Fe
Idh
Mdh-1
Mdh-2
Mdh-3
Pgi-1
Pgi-2
Tpi-1
Tpi-2
Ugpp-1
Ugpp-2
3
4
5
3
4
5
3
4
5
2
3
4
5
2
3
4
5
4
5
3
4
4
5
3
4
3
4
1
2
3
4
5
6
7
1
2
3
4
5
6
7
3
4
2
3
4
5
3
4
5
EX
GA
FL1
FL2
FL3
FL4
1.000
1.000
1.000
1.000
1.000
1.000
0.957
0.043
0.098
0.902
0.844
0.156
0.115
0.885
0.990
0.010
0.031
0.969
1.000
0.990
0.010
0.021
0.708
0.271
0.594
0.406
0.229
0.458
0.313
1.000
1.000
1.000
1.000
1.000
0.958
0.031
0.837
0.043
0.120
1.000
0.833
0.063
0.104
1.000
0.083
0.365
0.552
1.000
0.344
0.044
0.611
1.000
0.063
0.125
0.813
1.000
0.076
0.924
1.000
0.156
0.844
1.000
0.542
0.458
1.000
0.406
0.594
1.000
1.000
0.022
0.978
0.033
1.000
1.000
0.010
0.990
1.000
1.000
0.011
1.000
FL5
AL1
AL2
AL3
MS
0.146
0.844
0.010
0.010
0.823
0.167
0.031
0.948
0.021
0.359
0.630
0.011
0.146
0.854
0.271
0.698
0.031
0.073
0.906
0.021
0.979
0.021
0.865
0.135
0.948
0.052
1.000
0.927
0.073
0.823
0.177
0.854
0.146
0.115
0.813
0.073
1.000
1.000
0.021
0.979
0.010
0.990
0.042
0.958
0.094
0.208
0.698
1.000
0.063
0.479
0.458
1.000
0.333
0.271
0.396
1.000
0.021
0.240
0.208
0.531
1.000
0.094
0.906
1.000
0.115
0.885
1.000
0.292
0.708
1.000
0.792
0.208
1.000
0.396
0.604
1.000
1.000
0.043
0.957
1.000
1.000
0.125
0.875
0.021
0.979
0.271
0.729
0.104
0.896
0.073
0.927
0.281
0.271
0.448
0.781
0.219
0.542
0.458
0.927
0.073
0.115
0.885
0.198
0.802
0.104
0.292
0.604
0.917
0.083
0.646
0.354
0.781
0.219
0.010
0.990
0.021
0.979
0.969
0.031
0.010
1.000
0.021
0.042
0.967
0.989
1.000
0.104
0.896
0.938
0.073
0.802
0.125
1.000
1.000
0.938
0.389
0.033
0.256
0.044
0.233
0.022
0.022
0.740
0.260
0.052
0.542
0.385
0.021
0.194
0.097
0.292
0.056
0.278
0.083
0.010
0.115
0.875
0.031
0.969
0.533
0.056
0.133
0.067
0.156
0.663
0.033
0.163
0.011
0.087
0.056
0.479
0.521
0.052
0.073
0.865
0.010
0.021
0.979
0.043
0.531
0.469
0.146
0.313
0.542
0.063
0.772
0.677
0.228
0.323
0.565
0.435
0.478
0.448
0.552
0.323
0.522
0.677
0.848
0.152
0.896
0.104
0.112
0.725
0.138
0.025
0.896
0.104
0.104
0.896
1.000
NA
NA
NA
NA
NA
NA
NA
0.073
0.927
0.021
0.021
0.948
0.010
0.531
0.469
0.240
0.094
0.656
0.010
1.000
1.000
0.098
0.500
0.691
0.402
0.255
0.053
0.542
0.458
0.323
0.375
0.302
0.979
0.021
0.067
0.022
0.844
0.044
0.022
0.490
0.510
0.043
0.277
0.681
1.000
0.990
0.010
0.542
0.458
0.152
0.446
0.402
0.990
0.010
0.188
0.813
See Figure 1 for population locations.
NA indicates populations that had low enzyme activity for a particular locus.
Bold numbers indicate unique alleles (i.e., alleles restricted to one population).
2, Aat-1, Dia, Idh, Mdh-2, Mdh-3, Pgi-1, and Ugpp-2) the
frequency of the common allele was greater than 0.8. Idh and
Mdh-2 were polymorphic in only two populations. Unique
alleles (i.e., alleles restricted to a single population) were
detected in the GA, FL3, FL4, FL5, and AL2 populations
(Table 2). In addition, the AL3 and MS populations shared
alleles at three loci (Idh, Mdh-2, and Ugpp-2) not found in
other populations (Table 2). For Dia, these two populations
237
Journal of Heredity 2004:95(3)
Table 3.
Genetic diversity statistics for 11 S. leucophylla populations and the species
Population
P (%)
A
AP
Ae
Ho
He
FIS
EX
GA
FL1
FL2
FL3
FL4
FL5
AL1
AL2
AL3
MS
Meana
Speciesa
55.6
50.0
44.4
33.3
50.0
55.6
50.0
61.1
72.2
83.3
77.8
58.0
83.3
1.61
1.56
1.61
1.59
1.78
1.83
1.94
2.11
2.17
2.33
2.22
1.91
2.94
2.10
2.11
2.38
2.67
2.56
2.50
2.89
2.82
2.62
2.60
2.57
2.57
3.33
1.21
1.23
1.20
1.19
1.39
1.31
1.30
1.49
1.57
1.50
1.42
1.36
1.51
0.121
0.139
0.089
0.077
0.192
0.143
0.162
0.214
0.227
0.240
0.215
0.170 (0.057)
—
0.134
0.146
0.111
0.109
0.190
0.158
0.173
0.211
0.240
0.259
0.230
0.183 (0.053)
0.224
—
0.048
0.194
0.299
0.010
0.098
0.065
0.012
0.054
0.074
0.067
0.072
—
The genetic parameters are the percentage of polymorphic loci (P ), the mean number of alleles per locus (A) and per polymorphic locus (AP ), the mean
effective number of alleles per locus (Ae), observed heterozygosity (Ho), unbiased expected heterozygosity (He), and the fixation index (FIS).
Standard deviations are in parentheses.
a
Ex situ population was not included.
similarly shared an allele not found in any other population
except AL2.
Although 15 of the 18 loci (83.3%) were polymorphic
within the species, only 58.0% were polymorphic within
populations (Table 3). The mean number of alleles per locus
within populations (A) ranged from 1.56 (GA) to 2.33 (AL3),
with an overall population mean of 1.91. The mean number
of alleles per polymorphic locus ( AP ) was 2.57 across
populations, and ranged from 2.10 (EX) to 2.89 (FL5). The
mean effective number of alleles per locus (Ae) ranged from
1.19 (FL2) to 1.57 (AL2). The effective number of alleles per
locus was 1.51 for the species and averaged 1.36 across
populations. The highest expected heterozygosity (He) was
found for AL3 (0.259) and lowest for FL2 (0.109), with
a population mean of 0.183. For the species, expected
heterozygosity was 0.224. Compared to natural populations, the ex situ population maintained intermediate levels
of genetic diversity (Table 3). Population fixation indexes
were positive with the exceptions of populations FL3
and AL1.
No statistically significant differences in genetic diversity
were detected between population size classes, although
smaller populations tended to maintain somewhat less
genetic diversity (Figure 2). This was evident for the
percentage of polymorphic loci and the number of alleles
per polymorphic locus, reflecting somewhat diminished
allelic richness in smaller populations.
Among the 10 natural populations and 18 loci, significant
deviations (P , .05) from Hardy-Weinberg expectations
were detected for 19 of 96 tests. All but two (Aat-1 in FL3
and Pgi-1 in AL1) of these deviations were positive, indicating heterozygote deficiencies. Most of the 19 significant
deviations were associated with Tpi-1 and Ugpp-1. Seven
populations and six populations had significant positive
deviations from Hardy-Weinberg equilibrium at Tpi-1 and
Ugpp-1, respectively. Pooled populations (or loci) displayed
238
significant deviations at four loci (Fe, Pgi-2, Tpi-1, and Ugpp-1)
and six populations (FL1, FL2, FL4, AL1, AL3, and MS).
F-statistic estimates (Weir and Cockerham 1984) and
associated confidence intervals over all loci are summarized
in Table 4. The overall f (¼ Wright’s FIS) was 0.073, and it
was not significantly different from zero. The per-locus h
values ranged from 0.016 (Dia) to 0.304 (Tpi-1). The overall h
was 0.192, with the 99% confidence interval from bootstrapping over loci ranging from 0.139 to 0.248, which
indicated highly significant genetic differentiation. The
indirect estimate of gene flow, Nm (the number of migrants
per generation), was 1.05.
Nei’s (1978) unbiased genetic identities ranged from
0.896 (between the EX and AL1 populations) to 0.996
(between the EX and GA populations) (Table 5). Considering only the natural populations, the highest genetic
identity was found between FL4 and AL2 (0.979), and the
lowest between FL3 and AL1 (0.901). Mean genetic identity
across the 11 populations was 0.942 [standard deviation
(SD) ¼ 0.023].
The UPGMA dendrogram revealed some clustering of
geographically adjacent populations, but some anomalous
clustering as well (Figure 1). Not surprisingly, the ex situ
population (EX) clustered most closely with its source
population (GA) at the lowest genetic distance. The sole
extant GA population (and its derivative ex situ population),
which is the most geographically isolated population, was the
most genetically distinct. The three most westerly populations (MS, AL3, and AL1) clustered together, with the closest
association among the three found between the two most
geographically adjacent populations, MS and AL3. Anomalously the AL2 and FL4 populations clustered closely
together, and the FL2 population (12 km from FL1) was
the most distinct from the other populations, except those
from Georgia.
Wang et al. Genetic Diversity in Sarracenia leucophylla
Table 4. F-statistics (Weir and Cockerham 1984) at 15
polymorphic loci in S. leucophyllaa
Figure 2. Comparison of genetic diversity for three S.
leucophylla population size classes. Genetic diversity parameters
were the proportion of polymorphic loci (P ), mean number of
alleles per locus (A ), and polymorphic locus (AP ), mean
effective number of alleles per locus (Ae), observed
heterozygosity (Ho), and expected heterozygosity (He).
Significant negative correlations between gene flow and
geographic distance were found both in observed and logtransformed data (Figure 3). The largest correlation was r ¼
0.46, and the smallest was r ¼ 0.36. Significant correlations
suggest an isolation-by-distance effect, whereby populations
that are geographically closer exchange more genes than
more distant populations.
Discussion
Genetic diversity
Generally plant species with restricted geographic ranges
maintain less genetic diversity than more widespread species
(Hamrick and Godt 1989). Considering its geographic
distribution, S. leucophylla may be considered a narrowly
distributed species for comparative purposes (Hamrick and
Godt 1989), although the species is far less common than
distribution maps suggest. Nonetheless, compared to mean
genetic diversity found in narrowly distributed species, S.
leucophylla maintains high levels of genetic diversity (Table 6).
For example, the percentage of polymorphic loci (Ps) in S.
leucophylla was 83.3%, whereas mean Ps ¼ 45.1% for 101
narrowly distributed species (Hamrick and Godt 1989). Mean
genetic diversity (Hes) was 0.137 for these 101 narrowly
distributed plants (Hamrick and Godt 1989) and 0.224 for S.
leucophylla. Indeed, genetic diversity for S. leucophylla was higher
than means found for species having regional or widespread
distributions (Hamrick and Godt 1996b).
Rare species generally maintain less variation than more
widespread species within the same genus (Gitzendanner and
Soltis 2000; Karron 1997), although exceptions occur (Godt
and Hamrick 1998). The genus Sarracennia is restricted to
North America, with most of the taxa found in the
southeastern United States. Only Sarracennia purpurea, with
a range extending from Newfoundland to Florida and as far
a
Locus
F
h
f
6Pgdh-1
6Pgdh-2
Aat-1
Dia
Fe
Idh
Mdh-1
Mdh-2
Mdh-3
Pgi-1
Pgi-2
Tpi-1
Tpi-2
Ugpp-1
Ugpp-2
Over loci
Bootstrapping
over loci
(95% CI)
Bootstrapping
over loci
(99% CI)
0.083
0.241
0.077
0.006
0.246
0.126
0.193
0.012
0.068
0.006
0.329
0.496
0.097
0.403
0.044
0.251
Lower: 0.127,
Upper: 0.354
0.173
0.175
0.138
0.016
0.167
0.162
0.219
0.165
0.073
0.122
0.064
0.304
0.166
0.180
0.111
0.192
Lower: 0.149,
Upper: 0.235
0.109
0.080
0.071
0.023
0.095
0.043
0.034
0.213
0.006
0.146
0.283
0.276
0.084
0.272
0.075
0.073
Lower: 0.040,
Upper: 0.169
Lower: 0.099,
Upper: 0.383
Lower: 0.139,
Upper: 0.248
Lower: 0.068,
Upper: 0.195
Ex situ population was not included.
west as northeastern British Columbia, is found outside of
the southeastern United States. S. leucophylla is found only in
four southeastern states (Georgia, Mississippi, Florida, and
Alabama). Despite this difference in their distributions, S.
leucophylla has higher genetic diversity than S. purpurea, as
estimated for the species, and as population means (Table 6).
Genetic diversity is also higher for S. leucophylla than for other
previously analyzed Sarracenia species (Table 6), with the
exception of the Sarracennia rubra complex. Unfortunately the
phylogenetic relationship of S. leucophylla within Sarracenia is
not well understood (Bayer et al. 1996). A robust phylogeny
might provide insight into differences in genetic diversity
within the group. For example, if S. leucophylla was an
ancestral species, higher genetic diversity might be expected
within this species compared to others.
Sarracenia species are known to successfully hybridize in
nature, and several congeners co-occurred with S. leucophylla
at some of our collection sites. Hybridization and introgression can lead to increased genetic diversity within
species. However, Sarracenia hybrids (at least early generation
hybrids) are easily recognized by the intermediate sizes,
shapes, and colors of their pitchers. Although we observed
Sarracenia hybrids at several sites, they were not common, and
we avoided collecting pitchers with aberrant morphologies.
Thus, although past hybridization and introgression may
have contributed to the high genetic diversity within S.
leucophylla, our study provides no pertinent evidence.
Population Size
In contrast to several previous studies (Cruzan 2001; Godt et
al. 1996; Van Treuren et al. 1991; Young et al. 1999), we did
239
Journal of Heredity 2004:95(3)
Table 5.
Matrix of Nei’s (1978) unbiased genetic identities between S. leucophylla populations
Population
EX
GA
FL1
FL2
FL3
FL4
FL5
AL1
AL2
AL3
MS
EX
GA
FL1
FL2
FL3
FL4
FL5
AL1
AL2
AL3
MS
****
0.996
0.909
0.922
0.918
0.926
0.921
0.896
0.940
0.918
0.903
****
0.916
0.941
0.929
0.926
0.937
0.902
0.947
0.932
0.911
****
0.928
0.945
0.960
0.948
0.952
0.966
0.954
0.954
****
0.942
0.933
0.941
0.910
0.956
0.961
0.945
****
0.966
0.956
0.901
0.966
0.930
0.927
****
0.956
0.931
0.979
0.933
0.944
****
0.945
0.975
0.956
0.944
****
0.975
0.968
0.970
****
0.972
0.974
****
0.979
****
not find statistically significant differences in genetic diversity
among populations of different sizes. Nonetheless, there was
a tendency for smaller populations to be less genetically
diverse (Figure 2), suggesting that they may have lost alleles.
The lack of statistical significance with population size and
genetic diversity may be due to the relatively small number
of populations within each size class, imprecise estimates
of population size, and the evolutionary history of the
populations themselves.
The low genetic diversity found in population FL2 and its
distinctness from a nearby population (FL1) are notable and
was unexpected, since FL2 was a relatively large population
located only 12 km from FL1. A possible explanation for the
low genetic diversity estimates for FL2 is that the highly
variable Tpi-1 locus was excluded from the analysis for this
population due to poor enzyme activity and resolution.
However, when Tpi-1 is excluded from the overall analysis,
FL2 still had comparatively low genetic variation. The
establishment of FL2 from a small number of founders, or
transient population bottlenecks, may account for the
population’s low genetic diversity.
Heterozygote deficiencies were found for several S.
leucophylla populations (Tables 3 and 4). These may have been
due to selfing (although Sarracenia species are considered to
be highly outcrossed) (Schnell 1976; Slack 1979), biparental
inbreeding (mating between relatives) or population substructure (i.e., a Wahlund effect) (Hedrick 1985). Increased
inbreeding is expected in smaller populations (Ellstrand and
Elam 1993), but for S. leucophylla, heterozygote deficiencies
were not related to population size. For example, significant
deviations from Hardy-Weinberg expectations were not
found for the small GA population; the other small
population (FL3) had a small, but nonsignificant excess of
heterozygotes. In contrast, significant heterozygote deficiencies were found for four of the larger populations (FL2, AL1,
AL2, and AL3). We suspect that these deficiencies were
largely related to biparental inbreeding or population
substructure. The largest deviation was found for FL2,
which was also one of the most genetically distinct
populations; it may have experienced a bottleneck in the
past. If so, inbreeding may have played a larger role in the
deviations found in this population. Over all populations, FIS
240
was small (0.073) and nonsignificant, indicating that on the
whole the species is randomly mating.
Isolation by Distance and Genetic Differentiation
According to generalizations proposed by Slatkin (1993), the
observed pattern of genetic structure in S. leucophylla suggests
that populations are in equilibrium between gene flow and
genetic drift. However, the number of sampled populations
is insufficient to draw definitive conclusions [see Packer and
Owen (2001)]. In addition, the low correlation coefficient
(r ¼ 0.46; R2 ¼ 0.21) indicates that about 79% of the
variation in gene flow was related to factors other than
geographic distance. This was also reflected in some
anomalous clustering of populations in the UPGMA
phenogram. For example, although population FL2 was
geographically close to FL1, it did not cluster with it, but was
nearly as genetically distinct as the geographically disjunct
Figure 3. The relationship between pairwise (between
populations) estimates of gene flow (Nm) and geographic
distance (in kilometers) between S. leucophylla populations
(the linear regression was highly significant: Nm ¼ 7.97 3
103 km þ 2.99, R2 ¼ 0.214, P , .002).
Wang et al. Genetic Diversity in Sarracenia leucophylla
Table 6.
traits
Comparisons of genetic diversity and genetic differentiation among Sarracenia species and species having similar life-history
Species
a
Sarracenia jonesii
Sarracenia oreophilaa
Sarracenia rubrab
S. rubra ssp. alabamensisb
S. rubra ssp. rubrab
Sarracenia purpureac
Sarracenia leucophylla
Narrowly distributed plants (N ¼ 101)b
Short-lived herbs (N ¼ 236)b
S. jonesiia
S. oreophilaa
S. rubrab
S. rubra ssp. alabamensisb
S. rubra ssp. rubrab
S. purpureac
S. leucophylla
Narrowly distributed plants (N ¼ 115)b
Short-lived herbs (N ¼ 236)b
Ps (%)
As
APs
Aes
Hes
GST
60.9
59.1
93.3
80.0
73.3
60.9
83.3
45.1
45.0
2.04
1.91
2.87
2.27
2.40
2.13
2.94
1.83
1.78
2.71
2.54
3.00
2.58
2.91
2.86
3.33
2.84
2.73
1.14
1.15
1.45
1.44
1.35
1.38
1.51
1.17
1.18
0.086
0.082
0.225
0.209
0.177
0.189
0.224
0.137
0.136
0.190
0.133
0.091
0.086
0.141
0.549
0.174
Pp (%)
Ap
APp
Aep
Ho
Hep
28.3
27.8
76.7
53.3
52.1
23.4
58.0
30.6
29.0
1.33
1.34
2.33
1.73
1.66
1.28
1.91
1.45
1.41
2.15
2.22
2.75
2.36
2.27
2.14
2.57
2.47
2.39
1.10
1.11
1.39
1.37
1.27
1.09
1.36
1.13
1.13
0.062
0.063
0.161
0.183
0.139
0.051
0.170
0.061
0.060
0.193
0.187
0.142
0.055
0.183
0.105
0.103
Genetic parameters are the percentage of polymorphic loci (P ), the mean number of alleles per locus (A) and per polymorphic locus (AP ), the mean
effective number of alleles per locus (Ae), observed heterozygosity (Ho), expected heterozygosity (He), and the proportion of total genetic diversity found
among populations (GST). The subscript ‘‘s’’ indicates species’ values, while the subscript ‘‘p’’ indicates population means.
a
From Godt and Hamrick (1996).
b
From Godt and Hamrick (1998) and unpublished data.
c
From Godt and Hamrick (1999).
GA population. Moreover, geographically distant populations (i.e., FL4 and AL2) clustered together. However, the
significant correlation between geographic distance and gene
flow indicates that isolation by distance has played a role in
the genetic structure of this species. Other factors (e.g,
selection, founder effects, loss of intervening populations,
and possibly human-mediated gene flow) have probably also
influenced S. leucophylla’s overall genetic structure.
Given the indirect estimate of gene flow (Nm ¼ 1.05), it is
unlikely that genetic drift has played a major historical role in
determining the genetic structure of this species. A migration
rate of Nm ¼ 1.0 is theoretically necessary to counter
population divergence due to genetic drift (Wright 1931).
Despite this gene flow estimate, we found significant
genetic divergence (h ¼ 0.192, P , .01) among S. leucophylla
populations. The GST value (0.17) indicated that about 83%
of the total genetic variation was found within S. leucophylla
populations. Such population divergence is typical for
narrowly distributed outcrossing plants (mean GST ¼
0.169) (Hamrick and Godt 1996a). Population divergence
among S. leucophylla populations fell within the range found
for other Sarracenia species (Table 6), with S. purpurea, the
most widespread species, exhibiting the most intraspecific
divergence. Nei’s genetic identity measures, which takes into
consideration shared monomorphic and polymorphic loci,
were generally high and indicated little population divergence
(Table 5).
Conservation Implications
Fortunately genetic diversity within S. leucophylla does not
appear to be seriously compromised in comparison to other
Sarracenia taxa. Nonetheless, diminishing population numbers and decreasing population sizes will have a detrimental
impact on genetic diversity within the species. Although not
statistically significant, a trend for diminished allelic richness
with decreasing population size is already evident.
Immediate threats to the persistence of S. leucophylla are
loss and alteration of its habitats, competition from other
species, and collection for commercial or personal purposes.
Several sampled populations were clearly in decline, and are
perhaps ecologically irretrievable due to the buildup of
woody vegetation. For example, the AL2 population was
barely persisting (as adult plants) in a sea of grass and
shrubbery. The FL1 population was also in decline because
of inadequate frequency of burning. As woody vegetation
encroaches, pitcher plants cease flowering and recruitment
is halted. Although adult pitcher plants may persist for
a number of years under such conditions, these populations
are almost inevitably doomed without human intervention.
Three populations (AL1, AL2, and AL3) were in vulnerable,
unprotected, roadside sites. Active ecological management
(including prescribed burns or other means of controlling
woody vegetation) and increased legal protection will undoubtedly be required if this species is to survive in the wild.
241
Journal of Heredity 2004:95(3)
From a genetic standpoint, the loss of populations may
lead to loss of alleles from the species. Furthermore, gene
flow will become increasingly limited as populations become
extinct, leading to further genetic losses due to drift.
Reestablishment of populations from seed via natural
dispersal appears unlikely over reasonable time frames due
to the distances between populations and the decreasing
availability of appropriate pitcher plant habitat. All these
factors contribute to species’ endangerment.
The magnitude of the loss of genetic variation will
depend on which populations are extirpated. Our data
reinforce the point made by Schoen and Brown (1991) that
species with higher levels of genetic structure have higher
variance in genetic diversity within populations (i.e., He). S.
leucophylla has only a moderate level of genetic structure (h ¼
0.192), but He ranged from 0.109 to 0.259 (CV ¼ 0.292).
Since levels of He are not strongly associated with population
size and/or any other geographic factor, establishment of
effective conservation strategies is dependent on empirical
studies such as this.
The ex situ population maintained at the Atlanta
Botanical Garden captured most of the genetic variation
found in its source population. It is a good safeguard against
extinction of the sole remaining Georgia population, and
would serve as a good source of propagules should this
population require restoration. Ex situ protection of other
populations, especially those with unique alleles (e.g., FL3,
FL4, FL5, and AL2) should be undertaken, as well as in situ
measures to ensure their viability in the wild.
Acknowledgments
We thank Robert Godt for assistance in collecting population samples and
recording field data. Linda Chafin (Florida Natural Areas Inventory), Carol
Denhof (Atlanta Botanical Gardens, GA), Asa Haddock (Tate’s Hell State
Forest, FL), Bruce Haggedorn and Dennis Teague (Eglin Air Force Base,
FL), Tom Arrington (Blackwater River State Forest, FL), Tom Beitzel and
Daniel Young (Pine Log and Point Washington State Forests, FL), Ron
Weizel (Mississippi State Heritage Program), Al Schotz (Alabama State
Heritage Program), Fred Nations (Alabama botanist), and Scott Phipps
(Weeks Bay National Research Reserve, AL) provided location information;
many of these individuals also helped locate populations in the field. For
laboratory assistance during ‘‘trial runs’’ and during enzyme extractions we
thank the following students: Cecile Deen, Emily Gilbert, Tiffany Walker,
Grant Farmer, Angel Smith, Tereza Bercikova, and Katerina Bercikova. A
special thanks to John Spencer (the Shade Tree Mechanic) who expeditiously repaired our collection van in Eight Mile, Alabama. This study was
funded in part by a grant from the Turner Foundation to the Georgia
Plant Conservation Alliance. Z.-F. Wang was supported by the Chinese
Midwest Environmental Project of the Chinese Academy of Sciences from
the World Bank.
References
Barrett SCH and Kohn JR, 1991. Genetic and evolutionary consequences of
small population size in plants: implications for conservation. In: Genetics
and conservation of rare plants (Falk DA and Holsinger KE, eds). New
York: Oxford University Press; 3–30.
Bayer RJ, Hufford L, and Soltis DE, 1996. Phylogenetic relationships in
Sarraceniaceae based on rbcL and ITS sequences. Syst Bot 21:121–134.
242
Bohonak AJ, 2002. IBD (isolation by distance): a program for analyses of
isolation by distance. J Hered 93:153–154.
Cheliak WM and Pitel JA, 1984. Techniques for starch gel electrophoresis
of enzymes from forest tree species. Petawawa National Forestry Institute,
Information Report P1-X-42. Chalk River, Ontario, Canada: Canadian
Forest Service, Agriculture, Canada.
Cruzan MB, 2001. Population size and fragmentation thresholds for the
maintenance of genetic diversity in the herbaceous endemic Scutellaria
montana (Lamiaceae). Evolution 55:1569–1580.
DeMauro MM, 1993. Relationship of breeding system to rarity in the
lakeside daisy (Hymenoxys acaulis var. glabra). Conserv Biol 7:542–550.
Ellstrand NC and Elam DR, 1993. Population genetic consequences of
small population size: implications for plant conservation. Annu Rev Ecol
Syst 24:217–242.
Frankham R, 1997. Do island populations have less genetic variation than
mainland populations? Heredity 78:311–327.
Gilpin ME and Soulé ME, 1986. Minimum viable populations: processes of
species extinction. In: Conservation biology: the science of scarcity and
diversity (Soulé ME, ed.). Sunderland, MA: Sinauer Associates; 19–34.
Gitzendanner MA and Soltis PS, 2000. Patterns of genetic variation in rare
and widespread plant congeners. Am J Bot 87:783–792.
Godt MJW and Hamrick JL, 1996. Genetic structure of two endangered
pitcher plants, Sarracenia jonesii and Sarracenia oreophila (Sarraceniaceae). Am J
Bot 83:1016–1023.
Godt MJW and Hamrick JL, 1998. Allozyme diversity in the endangered
pitcher plant Sarracenia rubra ssp. alabamensis (Sarraceniaceae) and its close
relative S. rubra ssp. rubra. Am J Bot 85:802–810.
Godt MJW and Hamrick JL, 1999. Genetic divergence among infraspecific
taxa of Sarracenia purpurea. Syst Bot 23:427–438.
Godt MJW, Johnson BR, and Hamrick JL, 1996. Genetic diversity and
population size in four rare southern Appalachian plant species. Conserv
Biol 10:796–805.
Guo SW and Thompson EA, 1992. Performing the exact test of HardyWeinberg proportions for multiple alleles. Biometrics 48:361–372.
Hamrick JL and Godt MJW, 1989. Allozyme diversity in plant species. In:
Plant population genetics, breeding and genetic resources (Brown AHD,
Clegg MT, and Weir BS, eds). Sunderland, MA: Sinauer Associates; 43–63.
Hamrick JL and Godt MJW, 1996a. Conservation genetics of endemic plant
species. In: Conservation genetics: case histories from nature (Avise JL and
Hamrick JL, eds). London: Chapman & Hall; 281–304.
Hamrick JL and Godt MJW, 1996b. Effects of life history traits on genetic
diversity in plant species. Philos Trans R Soc Lond B 351:1291–1298.
Hedrick PW, 1985. Genetics of populations. Boston: Jones and Bartlett.
Holsinger KE, Mason-Gamer RJ, and Whitton J, 1999. Genes, demes, and
plant conservation. In: Genetics and the extinction of species: DNA and
the conservation of biodiversity (Landweber LF and Dobson AP, eds).
Princeton, NJ: Princeton University Press; 23–46.
Juniper BE, Robins RJ, and Joel DM, 1989. The carnivorous plants.
London: Academic Press.
Karron JD, 1997. Genetic consequences of different patterns of
distribution and abundance. In: The biology of rarity: causes and consequences of rare-common differences (Kunin WE and Gaston K J, eds).
London: Chapman & Hall; 174–189.
Kimura M and Crow JF, 1964. The number of alleles that can be
maintained in a finite population. Genetics 49:725–738.
Lande R, 1999. Extinction risks from anthropogenic, ecological, and genetic
factors. In: Genetics and the extinction of species: DNA and the
conservation of biodiversity (Landweber LF and Dobson AP, eds). Princeton, NJ: Princeton University Press; 1–22.
Wang et al. Genetic Diversity in Sarracenia leucophylla
Lewis PO and Zaykin D, 2001. Genetic data analysis: computer program
for the analysis of allelic data, version 1.0 (d16c). Available at http://
lewis.eeb.uconn.edu/lewishome/software.html.
Soltis DE, Haufler CH, Darrow DC, and Gastony GJ, 1983. Starch gel
electrophoresis of ferns: a compilation of grinding buffers, gel and electrode
buffers, and staining schedules. Am Fern J 73:9–27.
Manchenko GP, 1994. Handbook of detection of enzymes on electrophoretic gels. Ann Arbor, MI: CRC Press.
Swenson AA, 1977. Cultivating carnivorous plants. Garden City, NY:
Doubleday.
Mantel N, 1967. The detection of disease clustering and a generalised
regression approach. Cancer Res 27:209–220.
Van Treuren R, Bijlsma R, Van Delden W, and Ouborg NJ, 1991. The
significance of genetic erosion in the process of extinction. I. Genetic
differentiation in Salvia pratensis and Scabiosa columbaria in relation to
population size. Heredity 66:181–189.
Mitton JB, Linhart YB, Sturgeon KB, and Hamrick JL, 1979. Allozyme
polymorphisms detected in mature needle tissue of ponderosa pine. J Hered
70:86–89.
Nei M, 1973. Analysis of gene diversity in subdivided populations. Proc
Natl Acad Sci USA 70:3321–3323.
Nei M, 1978. Estimation of average heterozygosity and genetic distance
from a small number of individuals. Genetics 89:583–590.
Packer L and Owen R, 2001. Population genetic aspects of pollinator decline.
Conserv Ecol 5:4[online]; http://www.consecol.org/vol5/iss1/art4.
Patrick TS, Allison JR, and Krakow GA, 1995. Protected plants of Georgia.
Social Circle, GA: Georgia Department of Natural Resources; 155–156.
Walcott MV, 1935. Illustrations of North American pitcher plants.
Washington, DC: Smithsonian Institution Press.
Walter KS and Gillett HJ, 1998. 1997 IUCN red list of threatened plants.
Compiled by the World Conservation Monitoring Centre. Gland, Switzerland: IUCN–The World Conservation Union.
Weir BS and Cockerham CC, 1984. Estimating F-statistics for the analysis
of population structure. Evolution 38:1358–1370.
Weir BS, 1996. Genetic data analysis II. Sunderland, MA: Sinauer
Associates.
Wright S, 1931. Evolution in Mendelian populations. Genetics 16:97–159.
Raymond M and Rousset F, 1995. GENEPOP (version 1.2) population
genetics software for exact tests and ecumenicism. J Hered 86:248–249.
Wright S, 1978. Variability within and among natural populations, vol. 4.
Chicago: University Of Chicago Press.
Robbins CS, 1998. Examination of the U.S. pitcher-plant trade with a focus
on the white-topped pitcher-plant. Traffic Bulletin Excerpts, vol. 17, no. 2.
Yeh FC, Yang R-C, and Boyle T, 1999. POPGENE version 1.31—Microsoft Windows-based freeware for population genetic analysis. Quick users’
guide. Edmonton: University of Alberta.
Schnell DE, 1976. Carnivorous plants of the United States and Canada.
Annville, PA: Lebanon Valley Offset Company.
Schoen DL and Brown AHD, 1991. Intraspecific variation in population
gene diversity and effective population size correlates with the mating
system in plants. Proc Natl Acad Sci USA 88:4494–4497.
Slack A, 1979. Carnivorous plants. Cambridge, MA: MIT Press.
Slatkin M, 1993. Isolation by distance in equilibrium and nonequilibrium
populations. Evolution 47:264–279.
Young AG, Brown AHD, and Zich FA, 1999. Genetic structure of
fragmented populations of the endangered daisy Rutidosis leptorrhynchoides.
Conserv Biol 13:256–265.
Received August 23, 2003
Accepted March 11, 2004
Corresponding Editor: Halina Knap
243