Am. J. Bot. Join BSA Today!
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (6)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Coates, D. J.
Right arrow Articles by Hamley, V. L.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Coates, D. J.
Right arrow Articles by Hamley, V. L.
Agricola
Right arrow Articles by Coates, D. J.
Right arrow Articles by Hamley, V. L.
(American Journal of Botany. 2003;90:997-1008.)
© 2003 Botanical Society of America, Inc.

Population Biology

Evolutionary patterns and genetic structure in localized and widespread species in the Stylidium caricifolium complex (Stylidiaceae)1

David J. Coates2, Stephen Carstairs and Vicki L. Hamley

Science Division, Department of Conservation and Land Management, Locked Bag 104, Bentley Delivery Centre, Western Australia, 6983, Australia

Received for publication November 15, 2002. Accepted for publication February 18, 2003.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 LITERATURE CITED
 
The Stylidium caricifolium (Stylidiaceae) complex consists of seven currently recognized species and a taxon of putative hybrid origin endemic to southwest Western Australia. These taxa vary in geographical distribution from widespread, extending over a range of 500 km, to extremely localized, covering a range of only 0.5 km. Patterns of allozyme variation were investigated in 61 populations covering all taxa and two closely related species. Measures of genetic diversity were consistently lower and in some cases significantly lower in four rare and geographically restricted taxa compared with their widespread relatives. In contrast, genetic diversity in two other localized taxa was comparable or higher than in the widespread taxa. The level of divergence among populations was moderate to high, with a significant trend of higher FST values for the widespread species to lower values for the geographically restricted and rare taxa. Phylogenetic relationships and levels of divergence indicate that most taxa are probably relictual rather than recently evolved. Geographical localization and rarity in this complex can be attributed to a range of factors associated with habitat specificity, historical and ecological processes that characterize the southwest region, and mode of origin.

Key Words: allozymes • chromosome variation • genetic diversity • population structure • rarity • Stylidiaceae • Stylidium caricifolium • southwest Australia • triggerplant


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 LITERATURE CITED
 
Narrow geographic range is one of a number of key factors that characterize rare species (Rabinowitz, 1981 ; Fiedler and Ahouse, 1992 ; Fiedler, 1995 ; Gaston, 1997 ) and often indicates increased vulnerability to extinction. Predictions regarding the genetic consequences of a restricted geographic range generally follow genetic theory for small populations occupying a narrow range of environments. Genetic processes associated with restricted geographic range and rarity in plants have received significant attention over the last decade (see Karron, 1987 ; Hamrick and Godt, 1989 ; Barrett and Kohn, 1991 ; Ellstrand and Elam, 1993 ; Gitzendanner and Soltis, 2000 ). Theoretically, geographically restricted and rare plants might be expected to show low levels of genetic variation both within species and within populations due to selection under a narrow range of environmental conditions and genetic drift and inbreeding in small isolated populations (Barrett and Kohn, 1991 ; Ellstrand and Elam, 1993 ). Low genetic variation within geographically restricted species may also be due to founder events associated with recent speciation (Gottleib, 1981 ; Loveless and Hamrick, 1988 ). Assuming an isolation by distance model, geographically restricted species might also show less genetic structure across their range than widespread species (Hamrick and Godt, 1989 ; Godt and Hamrick, 1999 ). Broad comparisons between geographically restricted and widespread species have indicated that some trends are evident, with geographically restricted species tending to have less genetic variability than widespread species (Hamrick and Godt, 1989 ). More specific studies comparing geographically restricted and rare species with their widespread congeners show that the former often have significantly lower levels of genetic variation (Karron, 1987 ; Gitzendanner and Soltis, 2000 ). However, it is also evident that there are notable inconsistencies. A number of more recent studies provide evidence of conflicts with these trends, with some geographically restricted and rare species having levels of genetic diversity similar to their widespread congeners (Edwards and Wyatt, 1994 ; Young and Brown, 1996 ; Ge et al., 1999 ).

The flora of southwest Western Australia provides a unique opportunity to investigate evolutionary and genetic processes associated with restricted geographical range given that a significant proportion of the high floristic diversity in this region consists of rare and geographically restricted species. Fewer species are widespread with a high proportion of species having localized distributions and often occurring in allopatric or parapatric replacement series with close relatives (Hopper, 1992 ; Cowling et al., 1994 ; Gibson et al., 2000) . Although some of these are no doubt geographically restricted and rare due to recent widespread land clearing and habitat destruction, many appear to be rare due to evolutionary and ecological processes that characterize this region.

The flora of southwest Western Australia is recognized for its remarkable diversity and endemism and is one of the world's vascular flora biodiversity hotspots (Mittermeier et al., 1998 ). A range of historical processes are thought to have contributed to both rarity and local endemism. The region is characterized by a diverse array of evolutionary patterns, combining refugial species in higher rainfall areas with fragmented relictual species and suites of newly derived taxa in the more arid areas (Hopper et al., 1996 ). Compared with many other floras, particularly in North America and Europe, this flora has persisted for an extremely long period, probably well into the Cretaceous, without any large-scale extinction episodes associated with glaciation (Hopper et al., 1996 ). In addition, widespread climatic and habitat instability have been experienced in this region since the late Tertiary leading to cyclic expansion and contraction of the mesic and arid zones (Hopper, 1979 ; Hopper et al., 1996 ). Significant recent speciation is postulated to have occurred during this period particularly through the semiarid transitional rainfall zone of the South-West Botanical Province (Hopper, 1979 ). However, there is also increasing palaeobotanical evidence that a relatively high proportion of the flora is more ancient (McLoughlin and Hill, 1996 ; McLoughlin and McNamara, 2001 ). Thus it seems likely that a second consequence of the climatic instability since the late Tertiary has been the extinction of many relictual species throughout much of their range. These species now persist in geographically restricted and fragmented or disjunct population remnants.

Apart from palaeobotanical support for the antiquity of sections of the flora, population genetic studies also have provided evidence for relatively ancient fragmented population systems within some species and species complexes. Recent gene flow between populations either by long-distance seed dispersal or pollen movement probably has been limited or absent for a long period. A likely consequence is significant genetic differentiation between populations, with any phylogenetic similarity being due to common ancestry rather than any ongoing process of genetic exchange. Evidence for this comes primarily from isozyme studies (Broadhurst et al., 1999 ; Coates, 2000 ) and cpDNA studies (Byrne et al., 1999 , 2001 , 2002 ) in fragmented geographically restricted and rare species. These studies show significant levels of genetic divergence between populations and indicate that genetic processes associated with historical eco-geographic barriers to gene flow have led to the formation of distinct evolutionary lineages within many of these species.

Currently 221 taxa of Stylidium are recognized from Australia and southeast Asia (Wagstaff and Wege, 2002 ). Of these, some 70% occur only in the South-West Botanical Province of Western Australia, highlighting this area as the major center of triggerplant evolution (James, 1979 ). Atkins (2001) lists 55 taxa as rare and or geographically restricted. The Stylidium caricifolium complex consists of seven closely related species and a taxon of putative hybrid origin endemic to the South-West Botanical Province. These species vary in geographical distribution from widespread to extremely localized and rare, in some cases consisting of one or a few populations covering a few kilometers. This study was initiated to assess whether genetic variation and genetic structure are reduced in rare and geographically restricted species compared with closely related widespread species and to investigate evolutionary patterns within and between chromosomally different taxa in this complex.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 LITERATURE CITED
 
Species, distributions, and habitat
Seven species are currently recognized within the Stylidium caricifolium complex—S. affine, S. caricifolium, S. chiddarcoopingense, S. maritimum, S. nungarinense, S. sejunctum, and S. wilroyense (see Lowrie et al., 1998 , 1999 )—which covers a significant proportion of the South-West Botanical Province in Western Australia (Fig. 1). They are all relatively short-lived, insect-pollinated herbaceous perennials. Stylidium flowers are protandrous and are characterized by a touch-sensitive column that "triggers" when contacted by the insect, effecting pollination before resetting. Pollinator behavior can lead to high levels of geitonogamous self-pollination (Banyard and James, 1979 ; Burbidge and James, 1991 ). However, effective outcrossing is maintained in most perennial triggerplants because of a significant reduction in seed set following self-pollination due to post-zygotic seed abortion (see Burbidge and James, 1991 ).



View larger version (48K):
[in this window]
[in a new window]
  		Fig. 1. Distributions, karyotypes, and sampled populations for taxa in the Stylidium caricifolium complex and two closely related taxa, S. pilosum and S. albomontis. Because Stylidium sp. is chromosomally polymorphic, a representative karyotype is not given for it (Clackline) (see Coates and James, 1996 )

 
Species in the complex are found in a diversity of habitats, from coastal dune systems (S. maritimum), to high rainfall eucalypt forest (S. affine), through to semi-arid eucalypt woodland and shrublands (S. sejunctum). All species have allopatric or parapatric distributions, with S. caricifolium and S. affine the only exceptions that have relatively large zones of parapatry where hybridization has also been documented (Coates and James, 1996 ). A number of taxa are extremely closely related. Carlquist (1969) suggested that S. affine, S. caricifolium, and S. nungarinense differ only at the subspecific level and that S. maritimum is conspecific with S. affine. Another taxon, Stylidium sp. (Clackline), thought to be a stabilized hybrid derivative of S. affine and S. caricifolium, also is known from a geographically restricted area between the two putative parents (Coates and James, 1996 ). With the exception of S. nungarinense and S. chiddarcoopingense, species within the complex have very different chromosome complements (Fig. 1) including, in a number of cases, different chromosome numbers (Coates, 1982 , 1995 ; Coates and James, 1996 ). Stylidium nungarinense and S. chiddarcoopingense show only minor karyotypic differences.

Three species have relatively widespread distributions (S. affine, S. caricifolium, and S. nungarinense [Fig. 1]) and are generally abundant where they occur. There is a clear disjunction in the distribution of S. nungarinense, with the Wongan Hills populations isolated from the main distribution to the southeast (Fig. 1). The other species in the complex show geographically restricted distributions. Of these, S. maritimum has the largest geographic range but occurs in a series of fragmented populations restricted to a narrow strip of coastal dunes. Stylidium sejunctum is restricted to a series of banded ironstone hills, while the most restricted taxa, S. chiddarcoopingense and S. wilroyense, have geographic ranges of only a few kilometers. All the geographically restricted taxa could be regarded as rare based on their limited geographic range, few populations, and restricted habitat (see Fiedler and Ahouse, 1992 ; Gaston, 1997 ). Stylidium albomontis and S. pilosum, two other species related to the S. caricifolium complex, were included in this study primarily to provide outgroup comparisons for the phylogenetic studies.

Isozyme electrophoresis
Immediately prior to flower opening and anthesis, buds were collected from 10 to 50 plants from 57 populations covering all taxa within the complex. Bud material also was collected from an additional four populations, two each from two other species (S. albomontis and S. pilosum), considered to be closely related to the S. caricifolium complex. These species were included to provide outgroup comparisons for the phylogenetic studies.

The triggerplant column consisting of fused anthers and stigma was dissected from a single bud from each plant and prepared for isozyme analysis. A modification of the grinding buffer of Systma and Schaal (1985) gave the best results (50 mg/mL PVP, 0.8 mmol/L NAD, 0.4 mmol/L NADP, 1.0 mmol/L EDTA, 1.0 mmol/L ascorbic acid, 0.01 gm bovine serum albumin, 1 gm sucrose in 10 mL Tris-HCl PH 7.5, and add 20 µL ß-mercaptoethanol). Our isozyme methods using the Helena Laboratory cellulose acetate plate electrophoresis system were described previously by Coates (1988) .

Ten enzyme systems were assayed: aspartate aminotransferase (AAT, Enzyme Commission [E.C.] number 2.6.1.1), alcohol dehydrogenase (ADH, E.C. 1.1.1.1), glucose-6-phosphate dehydrogenase (G6PDH, E.C. 1.1.1.49), glucose phosphate isomerase (GPI, E.C. E.C. 5.3.1.9), glutamate dehydrogenase (GDH, E.C. 1.4.1.2), isocitrate dehydrogenase (IDH, E.C. 1.1.1.42), leucine aminopeptidase (LAP, 3.4.17.1), malate dehydrogenase (MDH, 1.1.1.37), phosphoglucomutase (PGM, E.C. 2.7.5.1), and phosphogluconate dehydrogenase (PGD, E.C. 1.1.1.44). Sixteen zones of activity were scored, and each zone was assumed to represent a single locus. Sixteen polymorphic loci were detected (Aat-1, Aat-2, Adh-1, Adh-2, G6pdh-1, Gpi-1, Gpi-2, Gdh-1, Idh-1, Lap-1, Mdh-1, Mdh-2, Pgm-1, Pgm-2, Pgd-1, and Pgd-2).

Data analysis
Data analyses were performed using the computer program POPGENE (Yeh and Boyle, 1997 ). The mean number of alleles per locus (A), percentage polymorphic loci (Pp), observed heterozygosity (Ho), and gene diversity (expected panmictic heterozygosity) (He) were estimated at the population level and species level. Fixation indices (FIS) (Wright, 1978 ) were estimated to examine population deviation from random mating. Differences in genetic variation between taxa at the population levels were determined using t test for Pp and nested analysis of variance (ANOVA) for A, He, and Ho with loci nested within populations. Bonferroni simultaneous tests were used for pairwise comparisons between taxa for A, He, and Ho.

Interpopulation divergence was estimated using Nei's (1978) unbiased genetic distance (D) and FST using the estimator {theta} (Weir and Cockerham, 1984 ). In the case of S. nungarinense, D was calculated separately for the two disjunct population groups. Separate estimates of FST were made for each taxon, including the two S. nungarinense population groups, as described by Weir and Cockerham (1984) . The relationship between the geographic range and interpopulation divergence in a taxon, based on FST and D, was investigated by linear regression analysis.

Patterns of isolation by distance were investigated in the two most widely distributed taxa S. caricifolium and S. affine. The patterns observed should reflect the relative influences of gene flow and drift as populations become more geographically separated. The approach proposed by Hutchinson and Templeton (1999) was used to investigate the level of association between genetic and geographic distance and whether the populations within the species have achieved equilibrium. Measures of FST and straight-line geographic distances were determined for all pairwise comparisons between the 18 S. caricifolium populations and 14 S. affine populations. Because the pairwise components of the distance matrices are not independent, the statistical significance of the association between geographic distance and FST was tested by a Mantel randomization test based on 5000 iterations (Mantel, 1967 ). To determine whether the degree of scatter shown in scatterplots increases with geographic distance and that the populations are in drift gene flow equilibrium (see Hutchinson and Templeton, 1999 ), the residuals from the linear regression of FST on geographic distance were correlated with geographic distance. Again the significance of the association was determined by a Mantel test based on 5000 iterations.

Both UPGMA and maximum likelihood methods were used to determine genetic relationships between all populations and taxa. The UPGMA analysis was based on Nei's unbiased genetic distance. Using the program GENDIST in the PHYLIP computer package (PHYLIP 3.4c; Felsenstein, 1995 ) a pairwise distance matrix was generated from gene frequency data and a UPGMA phenogram constructed using NEIGHBOR. A maximum likelihood tree, using gene frequency data, was constructed using the program CONTML in PHYLIP. Robustness of the UPGMA and maximum likelihood trees was determined by constructing 100 bootstrap replicates of the distance matrix for the UPGMA and the gene frequency data for the maximum likelihood tree with SEQBOOT in PHYLIP. These bootstrapped data sets were then run through NEIGHBOR and CONTML, respectively, and then a majority-rule consensus tree was generated in the PHYLIP program CONSENSE for each method.

To further investigate genetic relationships between populations within selected taxa, and any association with geographical patterns, UPGMA and continuous character maximum likelihood analyses were carried out on populations of the three most widely distributed species: S. affine, S. caricifolium, and S. nungarinense. The best tree topology (shortest tree length) produced by the previous two methods was then evaluated using a modified Wagner method (FREQPAR) (Swofford and Berlocher, 1987 ) that applies the principal of parsimony to gene frequency data. Because the FREQPARS program has a limited ability to search for the tree with the shortest length, Swofford and Berlocher (1987) indicate that the program is more useful in its current form as a means of evaluating tree topologies produced by other methods. Although other trees were generated and tested in FREQPARS, the UPGMA or maximum likelihood trees always gave the shortest tree length for population relationships within the three species.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 LITERATURE CITED
 
Genetic variation within populations and species
Most allozyme loci were polymorphic within taxa and in many cases polymorphic within populations. The exceptions were Gdh-1 and Idh-1, which showed fixed differences between some taxa and the occasional polymorphism within a few populations of the widely distributed species S. affine, S. caricifolium, and S. nungarinense.

There were no significant differences between two most widespread taxa, S. caricifolium and S. affine, in the mean population estimates for A, Ho, and He (Table 1), although Pp was significantly higher (P < 0.05) in S. caricifolium (Table 2). The only geographical trend observed was moderately lower levels of genetic diversity in the more peripheral southeastern populations (35, 36, 37; Fig. 1) of S. caricifolium and the geographically isolated population (16; Fig. 1) of S. affine.


View this table:
[in this window]
[in a new window]
  		Table 1. Estimates of genetic diversity and the fixation index (FIS) for 61 populations covering all taxa within the Stylidium caricifolium complex and two related taxa. Both population mean and species/taxon estimates are given. Popn, population number given in Fig. 1; N, mean sample size per locus; A, mean number of alleles per locus; Pp, percentage of polymorphic loci; He, gene diversity (expected panmictic heterozygosity); Ho, observed heterozygosity. Standard errors are given in parentheses

 

View this table:
[in this window]
[in a new window]
  		Table 2. Comparisons of genetic variation between taxa at the population levels based on t test for Pp, and nested analysis of variance (ANOVA) for A, He and Ho. Bonferroni simultaneous tests were used for pairwise comparisons between taxa for A, He and Ho. Levels of significance are indicated by NS, (P > 0.05); *, (P < 0.05); **, (P < 0.01); ***, (P < 0.001)

 
Based on the phylogenetic studies S. nungarinense appears to consist of two separate taxa, the widespread S. nungarinense (southern) and the rare and geographically restricted S. nungarinense (Wongan). No significant differences were observed in any single locus diversity measure between S. nungarinense (southern) compared with the other widespread taxa, S. affine and S. caricifolium.

In the rare and geographically restricted taxa, genetic diversity measures revealed two distinct patterns (Tables 1 and 2). Population levels of genetic diversity were comparable in S. chiddarcoopingense, S. maritimum, S. sejunctum, and S. nungarinense (Wongan) but consistently lower than in the widespread S. affine, S. caricifolium, and S. nungarinense (southern). The one exception was a significantly higher polymorphism in S. chiddarcoopingense compared with S. sejunctum. In S. chiddarcoopingense Ho was significantly lower than in S. nungarinense (southern). In S. maritimum, He and Ho were significantly lower than in S. affine, and A, He, and Ho significantly lower than in both S. caricifolium and S. nungarinense (southern). In S. sejunctum Ho was also significantly lower than in S. nungarinense (southern).

In contrast, genetic diversity levels in populations of the two remaining geographically restricted taxa S. sp (Clackline) and S. wilroyense were similar and consistently higher than in other taxa. With the exception of S. nungarinense (southern), some or all genetic diversity measures were significantly higher in S. sp. (Clackline) than in all other taxa (Table 2). In S. wilroyense, A was significantly higher than in S. affine, S. nungarinense (Wongan), S. maritimum, and S. sejunctum, while He and Ho were significantly higher than in S. maritimum.

Population level estimates of genetic variability in the two outgroup species generally were lower than taxa in the S. caricifolium complex. However, sample sizes for each of the two populations studied in these species were small and would need to be increased before robust comparisons could be made with other taxa.

The species/taxon level estimates A, Pp, and He (Table 1) were consistently higher for the widespread taxa, S. affine, S. caricifolium, and S. nungarinense (southern), compared with the rare and geographically restricted taxa, with the exception of S. sp (Clackline). Again taxon level estimates for S. albomontis and S. pilosum were lower than taxa in the S. caricifolium complex. Although this may reflect a real trend in these species, population sample sizes would need to be improved.

With the exception of population 26 in S. caricifolium and population 61 of S. pilosum, all population fixation indices (FIS) were equal to or greater than zero. There was generally a broad range in FIS values, particularly in the widespread S. caricifolium (–0.05 to 0.38) and S. affine (0.08 to 0.39). The values ranged across species from a mean of 0.31 in S. chiddarcoopingense to 0.14 in S. maritimum and to 0.11 in the more distantly related S. pilosum. The highest fixation indices were found in populations of the geographically restricted species S. chiddarcoopingense and S. sejunctum, with values consistently higher across all populations of these two taxa. In contrast, lower FIS values were found in populations of all other rare taxa, particularly in S. maritimum and S. wilroyense.

Population genetic structure
Differentiation between populations of S. nungarinense was extremely high (Table 3). It is largely attributed to major allele frequency differences at two loci (Lap-1 and Pgd-2) and fixed allele differences at three loci (Adh-1, Adh-2, and Aat-1) between the two disjunct population groups S. nungarinense (Wongan) and S. nungarinense (southern). In comparison, the average levels of differentiation between populations within the other species were moderate. There was a significant correlation between the level of divergence among populations within taxa and geographic range for both {theta} (R2 = 0.636, P = 0.018) and D (R2 = 0.567, P = 0.031) with higher values for the more widespread taxa S. caricifolium, S. affine, and S. nungarinense (southern) to lower values for the more geographically restricted taxa S. chiddarcoopingense, S. maritimum, S. nungarinense (Wongan), S. sejunctum, and S. sp (Clackline) (Table 3).


View this table:
[in this window]
[in a new window]
  		Table 3. Population divergence based on {theta} (FST) and D (Nei's [1978] unbiased genetic distance) for eight taxa in the Stylidium caricifolium complex

 
The scatter diagram for populations covering the range of S. caricifolium indicated a highly significant positive relationship between FST and geographical distance (Fig. 2; R2 = 0.5497, P < 0.001). A significant positive correlation also exists between the residuals, obtained from a linear regression of FST against geographic distance and geographic distance (R2 = 0.1894, P = 0.027) indicating a greater scatter of pairwise FST–geographical distance points as geographical distance increases between populations.



View larger version (14K):
[in this window]
[in a new window]
  		Fig. 2. Scatterplots of FST estimates (Weir and Cockerham, 1984 ) against geographic distance (kilometers) for Stylidium caricifolium and S. affine

 
In S. affine, the scatter diagram and regression analysis indicated a significantly positive association between FST and geographic distance (Fig. 2; R2 = 0.2244, P = 0.019). However, the scatter diagram also showed two clusters of points, one at pairwise geographic distances less than 100 km and the other at pairwise geographic distances more than 200 km. These distance groups appear to correspond to pairwise comparisons between populations within vs. pairwise comparisons between populations among the two geographically and phylogenetically distinct population groups within S. affine (Figs. 3–5). There was no correlation between the residuals, obtained from a linear regression of FST against geographic distance and geographic distance (R2 = 0.0643, P = 0.253). This indicates that there is no relationship between the scatter of pairwise FST–geographical distance points as geographical distance increases between populations.



View larger version (20K):
[in this window]
[in a new window]
  		Fig. 3. Relationships between populations within Stylidium caricifolium, S. affine, S. nungarinense, and S. chiddarcoopingense based on the best tree topology (shortest length tree) evaluated using a modified Wagner method, which applies the principle of parsimony to gene frequency data (Swofford and Berlocher, 1987 )

 


View larger version (22K):
[in this window]
[in a new window]
  		Fig. 5. Continuous character maximum likelihood tree for populations and taxa in the Stylidium caricifolium complex based on gene frequency data. Bootstrap values >50% are shown on branches

 
Phylogenetic geographical distribution associations between populations within the three widespread species S. affine, S. caricifolium, and S. nungarinense (Fig. 3) provided some indication of a north–south pattern of divergence among populations, although the degree of separation differed markedly between the three species. In S. caricifolium there was no clear phylogenetic pattern associated with geographic distribution except for the separation of populations northwest and southwest of the range. In S. affine, with the exception of population 17, there was separate clustering of the northern and southern populations, with the isolated population 16 showing intermediate affinities.

Two distinct population clusters were evident in S. nungarinense corresponding to the two disjunct population groups, S. nungarinense (southern) and S. nungarinense (Wongan). The level of divergence between these population groups (D = 0.296), allopatric distribution and habitat differences suggest that they should probably be considered separate taxa. The closely related S. chiddarcoopingense populations also were included in this phylogeny and indicated that both S. nungarinense taxa are related less to each other than they are to S. chiddarcoopingense, while the level of divergence (D = 0.108) between S. chiddarcoopingense and the S. nungarinense (southern) indicated that these two taxa are more closely related.

Phylogenetic patterns between all populations and species
With the exception of S. nungarinenese, the UPGMA (Fig. 4) and maximum likelihood tree (Fig. 5) showed that populations of each species cluster together. As expected, populations of S. albomontis and S. pilosum formed a separate group from the S. caricifolium complex taxa with 100% and 99% bootstrap support in the UPGMA and maximum likelihood trees, respectively. Divergence between S. albomontis and S. pilosum and all taxa within the complex generally was very high (Table 4). Stylidium wilroyense was the most phylogenetically distinct taxon within the complex with levels of divergence between it and other taxa (Table 4) ranging from D = 0.597 (S. affine) to D = 0.877 (S. maritimum). Populations of S. maritimum also clustered separately from all other taxa in the UPGMA but were placed in a group with S. nungarinense and S. chiddarcoopingense in the maximum likelihood tree. Again the level of divergence between S. maritimum and all other taxa was relatively high with the lowest level (D = 0.376) with S. nungarinense (Wongan). Stylidium nungarinense and S. chiddarcoopingense form a discrete cluster from the other taxa with the separate clustering and relative positions of the S. nungarinense (southern), S. nungarinense (Wongan), and S. chiddarcoopingense populations the same in both the UPGMA and maximum likelihood trees and as described previously for the population phylogenetic studies based on the FREQPAR analysis (Fig. 3). The S. sejunctum populations are all placed as a separate group between the S. nungarinense taxa and S. affine and S. caricifolium.



View larger version (25K):
[in this window]
[in a new window]
  		Fig. 4. UPGMA dendrogram of populations and taxa in the Stylidium caricifolium complex based on Nei's (1978) unbiased genetic distance. Bootstrap values >50% are shown on branches

 

View this table:
[in this window]
[in a new window]
  		Table 4. Genetic divergence between taxa in the Stylidium caricifolium complex based on Nei's (1978) unbiased genetic distance (D)

 
Populations of the putative hybrid taxon, S. sp. (Clackline), formed a separate cluster intermediate in position, in both the UPGMA and maximum likelihood trees, between the putative parents S. affine and S. caricifolium. Divergence between S. sp. (Clackline) and the putative parents S. affine (D = 0.175) and S. caricifolium (D = 0.178) was comparable to the level of divergence between S. affine and S. caricifolium (D = 0.186).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
LITERATURE CITED
 
Geographic distribution and patterns of genetic variation
Taxa within the S. caricifolium complex show a range of geographic distributions but are phylogenetically closely related and are characterized by the same pollination system, similar seed dispersal mechanisms, self-compatiblility, and frequent geitogamous self-pollination (Coates, 1981 ; Burbidge and James, 1991 ). These characteristics make it possible to readily investigate theoretical predictions of lower genetic diversity and reduced genetic structure for rare and geographically restricted taxa (Hamrick and Godt, 1989 ; Barrett and Kohn, 1991 ; Ellstrand and Elam, 1993 ) in this complex, while at the same time minimizing any confounding effects that may be associated with phylogenetic differences and differences in life-history attributes and breeding system (see Karron, 1987 ; Karron et al., 1988 ; Gitzendanner and Soltis, 2000 ).

Population-based estimates of genetic diversity are relatively high for all taxa (Table 1) when compared to other outcrossing animal-pollinated and short-lived herbaceous perennials (Hamrick and Godt, 1989 ) but indicate significant differences in genetic variability between geographically restricted taxa and more widespread taxa. These findings are consistent with those of Hamrick and Godt (1989) , who found that geographic range accounted for a major proportion of variation in both species and populations. Thus the two most widespread and common species, S. caricifolium and S. affine, showed consistently, and in a number of cases significantly, higher levels of genetic diversity than the geographically restricted species S. chiddarcoopingense, S. maritimum, and S. sejunctum. In comparisons with the other relatively widespread species, S. nungarinense also showed a similar pattern, although they were complicated by the phylogenetic evidence suggesting that this species consists of two different taxa: the more widespread S. nungarinense (southern) and the rare and geographically localized S. nungarinense (Wongan). Genetic diversity levels were significantly lower in S. nungarinense (Wongan) than in S. nungarinense (southern) and the other widespread species, but comparable to the geographically restricted species S. chiddarcoopingense, S. maritimum, and S. sejunctum. Lower levels of genetic diversity in S. chiddarcoopingense, S. maritimum, S. sejunctum, and S. nungarinense (Wongan) are likely due to a number of factors. Fluctuations in population size and repeated bottlenecks associated with ongoing Pleistocene climatic instability may be primary determinants. Habitat specificity also is a likely factor, with three taxa confined to breakaways (Jutson, 1934 ) and rocky slopes associated with granite outcrops (S. chiddarcoopingense), banded ironstone (S. sejunctum), and laterites (S. nungarinense [Wongan]), while S. maritimum is restricted to coastal dune systems. In addition, the close phylogenetic relationship between S. chiddarcoopingense, S. nungarinense (Wongan), and S. nungarinense (southern) suggests that lower levels of genetic diversity in the two rare and geographically restricted taxa may be due to founder events associated with relatively recent divergence and isolation.

While the relationship between lower levels of genetic diversity and restricted geographic distribution are likely to be a consequence of habitat specificity and stochastic processes associated with population size fluctuations, inbreeding also may be important (Barrett and Kohn, 1991 ; Ellstrand and Elam, 1993 ). The only species in the complex to show no reduction in seed set due to post-zygotic seed abortion following self-pollination (Coates, 1981 ), S. maritimum, has the lowest levels of genetic diversity of any of the taxa. This suggests that inbreeding associated with selfing may be a significant determinant of low levels of genetic diversity in this species. Interestingly, despite having an overall deficit of heterozygotes, the magnitude of the deficit was low compared with other taxa. Consistently higher fixation indices were observed in populations of other rare and geographically restricted species, S. chiddarcoopingense and S. sejunctum. Inbreeding is also likely to be contributing to the reduced levels of genetic diversity observed in the small isolated populations of both of these species.

Despite reduced levels of genetic diversity in geographically restricted taxa this trend is not consistent across all such taxa in the complex. Two other rare and geographically restricted taxa, S. wilroyense and S. sp. (Clackline), show comparable or in a number cases significantly higher levels of genetic diversity than the two most widespread species S. affine and S. caricifolium. A number of explanations have been given for unexpectedly high levels of genetic variation in rare and geographically restricted species. These include recent origin and retention of high levels of variation from a widespread progenitor (Gottlieb et al., 1985 ), hybridization (Smith and Pham, 1996 ), maintenance of relatively large populations (Young and Brown, 1996 ) and long-term stability of populations occupying refugia (Lewis and Crawford, 1995 ), and relatively recent fragmentation of previously widespread species (Karron, 1991 ).

Morphological, cytogenetic (Coates, 1981 ; Coates and James, 1996 ), and phylogenetic analyses based on allozyme data strongly support the notion that S. sp. (Clackline) is of hybrid origin involving past hybridization between S. affine and S. caricifolium, with subsequent limited range expansion. Increased genetic variability in this taxon appears to be due to combinations of alleles found in both putative parental taxa and the presence of novel alleles that have probably arisen post-hybridization or as part of the hybridization process (Coates and James, 1996 ).

In contrast, there is no clear explanation for the high levels of genetic variation in S. wilroyenese. Phylogenetically this species is the most distinctive in the complex, although there is no indication that it differs in terms of life-history characteristics or breeding system. It is known only from one population of some 200 plants in a heavily cleared landscape. Perhaps S. wilroyense had a wider distribution, with all other populations eliminated by land clearing over the last 50 yr. While recent loss of habitat probably has been a major factor causing rarity in this species, it is also possible that it always had a restricted distribution but persisted in relatively large populations. Any recent reduction in population size following land clearing may not have been great enough to cause a detectable loss in genetic variation. It is also possible that insufficient time has elapsed since land clearing for drift and inbreeding effects associated with reduced population size to have any significant effect. A similar explanation has been given for the lack of any detectable reduction in genetic variation in small fragmented populations of the rare Lambertia orbifolia (Coates and Hamley, 1999 ).

While a reduction in genetic diversity is often evident in rare and geographically restricted species compared with widespread species, a trend to less genetic structure has not been observed generally in comparisons between rare and widespread congeners (Gitzendanner and Soltis, 2000 ) or in broader comparisons between species with contrasting geographic ranges (Hamrick and Godt, 1989 ). Explanations for this include confounding effects due to differences in life history, taxonomic biases, different evolutionary histories, and sampling strategies (see Hamrick and Godt, 1989 ; Godt and Hamrick, 1999 ). In this study the level of divergence within taxa was moderate, with a significant trend from higher FST values for the more widespread taxa S. caricifolium, S. affine, and S. nungarinense (southern), to lower values for the more geographically restricted taxa S. maritimum and S. sejunctum, and lowest values for the most geographically restricted taxa S. chiddarcoopingense, S. sp. (Clackline), and S. nungarinense (Wongan). These findings provide support for theoretical predictions, based on an isolation by distance model, that genetic structure should decrease with a decrease in geographic range (Godt and Hamrick, 1999 ).

An important consideration in assessing population genetic structure within a species is the evaluation of isolation by distance and the individual effects of gene flow and drift. Even within a species there may be regional differences in the presence and patterns of isolation by distance (Hutchison and Templeton, 1999 ). It seems likely that comparisons in genetic structure between closely related taxa, whether they be geographically restricted or widespread, will also be confounded by isolation by distance patterns. Although we have not investigated isolation by distance in all taxa in this complex, largely because of sampling difficulties in the rare and geographically restricted taxa, we did find evidence for isolation by distance in the two most widespread species, S. caricifolium and S. affine. However, the relative influences of gene flow and drift appear to differ between populations of these two species. In S. caricifolium there was a highly significant positive relationship between FST and geographical distance and a significantly greater scatter of pairwise FST–geographical distance points as geographical distance increased between populations. This indicated that the S. caricifolium populations are in gene flow drift equilibrium (Hutchison and Templeton, 1999 ). In contrast, in S. affine, despite the significant positive relationship between FST and geographical distance, there was no evidence for gene flow drift equilibrium, although this condition could be confounded by the presence of two geographically and phylogenetically distinct population groups (Fig. 3) within this species.

The findings from this study highlight a number of general issues raised by various authors in relation to rarity and its causes (Fiedler and Ahouse, 1992 ; Kunin and Gaston, 1993 ; Fiedler, 1995 ). Rarity results from a wide array of causes, and patterns of rarity cannot be generalized in any simplistic fashion. The phylogenetic and biogeographical data indicate a number of different causes of rarity in the S. caricifolium complex and thus different expected outcomes in relation to the maintenance of levels and patterns of genetic diversity. Low levels of genetic diversity in four of the six rare and geographically restricted taxa can be attributed to one or a combination of population size effects, habitat specificity, recent speciation, and breeding system. Higher levels of genetic diversity in the other two rare taxa can be explained by hybridization and possibly large population sizes prior to recent land clearing and habitat destruction. In contrast the pattern of genetic divergence among populations within taxa follows a consistent trend based on geographical distribution and appears not to be influenced by other factors that may be contributing to rarity in any one taxon.

Evolutionary patterns, chromosome change, and speciation
Chromosome differences and generally high levels of allozyme divergence between taxa support the recognition of the seven currently recognized species in the S. caricifolium complex. These studies also support earlier findings (Coates and James, 1996 ) that there are two additional taxa. One is thought to be a stabilized hybrid derivative (S. sp. [Clackline]) and the other a biogeographically distinct form of S. nungarinense in the Wongan Hills area. The level of divergence between taxa combined with phylogenetic, cytogenetic, and biogeographic evidence raises a number of issues in relation to the evolution of this complex and Stylidium in general.

The extensive speciation in Stylidium in the South-West Botanical Province has been attributed to factors such as adaptive diversification associated with complex soil mosaics (Carlquist, 1969 ) and versatility in the pollination mechanisms (Carlquist, 1969 ; Armbruster et al., 1994 ). At a broader level, significant climatic oscillations throughout the Pleistocene also have been strongly implicated in Stylidium evolution (Coates and James, 1996 ). Based largely on earlier suggestions concerning the relatively recent origin of a significant component of the high vascular plant diversity found in the southwest region (Hopper, 1979 ), speciation and diversification in the genus has generally been considered to be relatively recent (Coates and James, 1996 ).

Triggerplant evolution is also characterized by widespread chromosome repatterning with haploid chromosome numbers ranging from n = 5 to n = 16 and species with the same haploid number usually having distinctive karyotypes (James, 1979 ; Coates, 1981 , 1982 ). This indicates that even closely related species are separated by multiple chromosome rearrangements. Given the level of chromosome change within the genus, James (1979) and Coates and James (1996) postulated that chromosome change has played a significant role in Stylidium evolution and speciation. This led to the suggestion that recent speciation postulated in the genus may have been rapid and associated with chromosomal repatterning (Coates and James, 1996 ), similar to the saltational speciation model proposed by Lewis (1966) . Recent speciation associated with chromosome change might be expected to result in low levels of genetic divergence between closely related species (Gottleib, 1984 ; Sites and Moritz, 1987 ; Crawford, 1989 ). However, this was not evident in this study (Table 4), with most species pairs in the complex showing relatively high levels of divergence.

An alternate evolutionary pattern more compatible with recent molecular and palaeobotanical observations on other plant groups in southwest Australia is that speciation and intraspecific differentiation in this complex has occurred over a much greater period of time. CpDNA studies on closely related species pairs in the genus Acacia indicate levels of divergence into the Pliocene (Byrne et al., 2001 , 2002 ), while at the intraspecific level high allozyme divergence has been reported for disjunct populations of Lambertia orbifolia (Coates and Hamley, 1999 ) with cpDNA nucleotide divergence (0.223%), again suggesting Pliocene separation between the two population groups (Byrne et al., 1999 ). Similarly, population genetic studies on 34 species in southwest Australia reviewed in Coates (2000) show that 22 geographically localized and rare species have historically fragmented or disjunct populations with a relatively high proportion of genetic diversity apportioned between populations, compared with the average estimate for flowering plants (Hamrick and Godt, 1989 ). In relation to this complex, it is interesting to note that the coastal dune endemic S. maritimum probably evolved in conjunction with the development of coastal dune systems that appeared in the late Pliocene, while the phylogenetic evidence suggests that S. wilroyense may be older.

In conclusion, patterns of genetic variation in the S. caricifolium complex can be attributed to a range of factors associated with the mode of origin of the taxon, geographical distribution, habitat specificity, and historical and ecological processes that characterize the southwest region. Most taxa in the complex appear not to be evolved recently with the probable exceptions of S. chiddarcoopingense and S. nungarinense (Wongan). Reduced levels of genetic variation and less genetic structure are a feature of a number of rare and geographically restricted taxa. Yet there are notable exceptions: the most phylogenetically distinct species in the complex, S. wilroyense, has some of the highest levels of genetic variation in its single population. Although the findings from this study generally support predictions that rare and geographically restricted species will have reduced genetic variation and less genetic structure, they also indicate that causes and patterns of rarity and geographical localization are often complex. These findings highlight the potential for further studies in this ancient flora that may provide useful comparisons to more recently evolved floras in relation to the causes of rarity and their genetic consequences. Such studies may not only improve our understanding of evolutionary processes in highly endemic floras, but also better inform the management and conservation of genetic resources of many rare and often threatened plant species.


View this table:
[in this window]
[in a new window]
  		Table 1. Continued

 

   FOOTNOTES
 
1 The authors thank Linda Broadhurst, Margaret Byrne, Peggy Fiedler, and Colin Yates for comments and discussion. Back

2 Author for reprint requests (davidc{at}calm.wa.gov.au ) Back


   LITERATURE CITED
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 LITERATURE CITED
 
Armbruster W. S. M. E. Edwards E. M. Debevec 1994 Floral character displacement generates assemblage structure of Western Australian triggerplants (Stylidium). Ecology 75: 315-329[CrossRef][ISI]

Atkins K. J. 2001 Declared rare and priority flora list for Western Australia. Department of Conservation and Land Management, Perth, Western Australia, Australia

Barrett S. C. H. J. R. Kohn 1991 Genetics and evolutionary consequences of small population size in plants: implications for conservation. In D. A. Falk and K. E. Holsinger [eds.], Genetics and conservation of rare plants, 3–30. Oxford University Press, New York, New York, USA

Banyard B. J. S. H. James 1979 Biosystematic studies in the Stylidium crassifolium species complex (Stylidiaceae). Australian Journal of Botany 27: 27-37[CrossRef]

Broadhurst L. H. D. J. Coates B. H. Tan 1999 Genetic diversity in the monospecific Western Australian endemic, Geleznowia verrucosa Turz (Rutaceae). Heredity 82: 292-299

Burbidge A. H. S. H. James 1991 Postzygotic seed abortion in the genetic system of Stylidium (Angiospermae: Stylidiaceae). Journal of Heredity 82: 219-328

Byrne M. B. MacDonald D. J. Coates 1999 Divergence in the chloroplast genome and nuclear rDNA of the rare Western Australian plant, Lambertia orbifolia. (Proteaceae). Molecular Ecology 8: 1789-1796[CrossRef][Medline]

Byrne M. B. MacDonald D. J. Coates 2002 Phylogeographical patterns in chloroplast DNA variation within Acacia acuminata (Leguminosae: Mimosoidae) complex in Western Australia. Journal of Evolutionary Biology 15: 576-587[CrossRef][ISI]

Byrne M. G. Tischler B. MacDonald D. J. Coates J. McComb 2001 Phylogenetic relationships between two rare acacias and their common, widespread relatives in south-western Australia. Conservation Genetics 2: 157-166[CrossRef]

Carlquist S. 1969 Studies in Stylidiaceae: new taxa, field observations, evolutionary tendencies. Aliso 7: 13-64

Coates D. J. 1981 Chromosome, morphometric and breeding system studies in the Stylidium caricifolium species complex (Stylidiaceae). Australian Journal of Botany 29: 397-417[CrossRef]

Coates D. J. 1982 Chromosome variation and species relationship in the scale-leaved triggerplants (Stylidium section Squamosae). Australian Journal of Botany 30: 121-130[CrossRef]

Coates D. J. 1988 Genetic diversity and population structure in the rare Chittering grass wattle, Acacia anomala. Australian Journal of Botany 36: 273-286[CrossRef]

Coates D. J. 1995 Chromosome re-patterning, genetic diversity and defining conservation units in Western Australian triggerplants (Stylidium). In P. E. Brandham and M. D. Bennett [eds.], Kew Chromosome Conference IV. Royal Botanic Gardens Kew, Richmond, Surrey, UK

Coates D. J. 2000 Defining genetic units for conservation in a rich and fragmented flora: implications for management of threatened species in south-west Australia. Australian Journal of Botany 48: 329-339[CrossRef]

Coates D. J. V. L. Hamley 1999 Genetic divergence and mating systems in the endangered and geographically restricted species Lambertia orbifolia Gardner (Proteaceae). Heredity 83: 418-427[CrossRef][ISI][Medline]

Coates D. J. S. H. James 1996 Chromosome repatterning, population genetic structure and local speciation in south west Australian triggerplants (Stylidium). In S. Hopper, J. Chappill, M. Harvey, and A. George [eds.], Gondwanan heritage: past, present and future of the Western Australian biota, 276–286. Surrey Beatty, Sydney, Australia

Cowling R. M. E. T. F. Witkowski A. V. Milewski K. R. Newbey 1994 Taxonomic, edaphic and biological aspects on narrow plant endemism on matched sites in mediterranean South Africa and Australia. Journal of Biogeography 21: 651-664[CrossRef][ISI]

Crawford D. J. 1989 Enzyme electrophoresis and plant systematics. In D. E. Soltis and P. S. Soltis [eds.], Isozymes in plant biology, 146–164. Dioscorides Press, Portland, Oregon, USA

Edwards A. L. R. Wyatt 1994 Population genetics of the rare Asclepias texana and its widespread sister species, A. perennis. Systematic Botany 19: 291-307[CrossRef][ISI]

Ellstrand N. C. D. R. Elam 1993 Population genetic consequences of small population size: implications for plant conservation. Annual Review of Ecology and Systematics 24: 217-242[CrossRef][ISI]

Felsenstein J. 1995 PHYLIP (phylogeny inference package), version 3.5c. http://evolution.genetics.washington.edu/phylip.html

Fiedler P. L. 1995 Rarity in the Californian flora: new thoughts on old ideas. Madroño 42: 127-141

Fiedler P. L. J. J. Ahouse 1992 Hierarchies of cause: toward an understanding of rarity in vascular plant species. In P. L. Fiedler and S. K. Jain [eds.], Conservation biology, 23–48. Chapman and Hall, New York, New York, USA

Gaston K. J. 1997 What is rarity?. In W. E. Kunin and K. J. Gaston [eds.], The biology of rarity, 30–41. Chapman and Hall, London, UK

Ge S. K.-Q. Wang D.-Y. Hong W.-H. Zhang Y.-G. Zu 1999 Comparisons of genetic diversity in the endangered Adenophora lobophylla and its widespread congener, A. potaninii.. Conservation Biology 13: 509-513[CrossRef][ISI]

Gibson N. G. Keighery B. Keighery 2000 Threatened plant communities of Western Australia. 1. The ironstone communities of the Swan and Scott Coastal Plains. Journal of the Royal Society of Western Australia 83: 1-11

Gitzendanner M. A. P. S. Soltis 2000 Patterns of genetic variation in rare and widespread congeners. American Journal of Botany 87: 783-792[Abstract/Free Full Text]

Godt M. J. W. J. L. Hamrick 1999 Genetic divergence among infraspecific taxa of Sarracenia purpurea. Systematic Botany 23: 427-438[CrossRef][ISI]

Gottleib L. D. 1981 Electrophoretic evidence and plant populations. Progress in Phytochemistry 7: 1-45

Gottleib L. D. 1984 Isozyme evidence and problem solving on plant systematics. In W. F. Grant [ed.], Plant biosystematics, 343–357. Academic Press, Orlando, Florida, USA

Gottleib L. D. S. I. Warwick V. S. Ford 1985 Morphological and electrophoretic divergence between Layia discoidea and L. glandulosa. Systematic Botany 10: 133-148

Hamrick J. L. M. J. W. Godt 1989 Allozyme diversity in plant species. In A. H. D. Brown, M. T. Clegg, A. L. Kahler, and B. S. Weir [eds.], Plant population genetics, breeding and genetic resources, 43–63. Sinauer Associates, Sunderland, Massachusetts, USA

Hopper S. D. 1979 Biogeographical aspects of speciation in the southwest Australian flora. Annual Review of Systematics and Ecology 10: 399-422[CrossRef]

Hopper S. D. 1992 Patterns of plant diversity at the population and species levels in south-west Australian meditteranean ecosystems. In R. J. Hobbs [ed.], Biodiversity of Mediterranean ecosystems in Australia, 27–46. Surrey Beatty and Sons, Chipping Norton, New South Wales, Australia

Hopper S. D. M. S. Harvey J. A. Chappill A. R. Main B. Y. Main 1996 The Western Australian biota as Gwondwanan heritage: a review. In S. D. Hopper, J. A. Chappill, M. S. Harvey, and A. S. George [eds.], Gondwanan heritage: past, present and future of the Western Australian biota, 1–46. Surrey Beatty and Sons, Chipping Norton, New South Wales, Australia

Hutchison D. W. A. R. Templeton 1999 Correlation of pairwise genetic and geographic distance measures: inferring relative influences of gene flow and drift on the distribution of genetic variability. Evolution 53: 1898-1914[CrossRef][ISI]

James S. H. 1979 Chromosome numbers and genetic systems in the triggerplants of western Australia (Stylidium; Stylidiaceae). Ausralian Journal of Botany 27: 17-25[CrossRef]

Jutson J. T. 1934 The Physiography (geomorphology) of Western Australia. Geological Survey, Bulletin No. 95. Perth, Western Australia, Australia

Karron J. D. 1987 A comparison of levels of genetic polymorphism and self-compatibility in geographically restricted and widespread plant congeners. Evolutionary Ecology 1: 47-58

Karron J. D. 1991 Patterns of genetic variation and breeding systems in rare plant species. In D. A. Falk and K. E. Holsinger [eds.], Genetics and conservation of rare plants, 87–98. Oxford University Press, New York, New York, USA

Karron J. D. Y. B. Linhart C. A. Chaulk C. A. Robertson 1988 Genetic structure of populations of geographically restricted and widespread species of Astragalus (Fabaceae). American Journal of Botany 75: 1114-1119[CrossRef][ISI]

Kunin W. E. K. J. Gaston 1993 The biology of rarity: patterns, causes and consequences. Trends in Ecology and Evolution 8: 298-301[CrossRef]

Lewis H. 1966 Speciation in flowering plants. Science 152: 167-172[Free Full Text]

Lewis P. O. D. J. Crawford 1995 Pleistocene refugium endemics exhibit greater allozyme diversity than widespread congeners in the genus Polygonella (Polygonaceae). American Journal of Botany 82: 141-149

Loveless M. D. J. L. Hamrick 1988 Genetic organization and evolutionary history in two American species of Cirsium. Evolution 42: 254-265[CrossRef][ISI]

Lowrie A. D. J. Coates K. F. Keneally 1998 A taxonomic review of the Stylidium caricifolium complex (Stylidiaceae), from south-west Western Australia. Nuytsia 12: 43-58

Lowrie A. D. J. Coates K. F. Keneally 1999 Stylidium chiddarcoopingense (Stylidi