Am. J. Bot. Cross-Journal Searching
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 Supplementary Data
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 (12)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Shepherd, K. A.
Right arrow Articles by Calladine, A.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Shepherd, K. A.
Right arrow Articles by Calladine, A.
Agricola
Right arrow Articles by Shepherd, K. A.
Right arrow Articles by Calladine, A.
(American Journal of Botany. 2004;91:1387-1397.)
© 2004 Botanical Society of America, Inc.

Systematics

Radiation of the Australian Salicornioideae (Chenopodiaceae)—based on evidence from nuclear and chloroplast DNA sequences1

K. A. Shepherd2,4, M. Waycott3 and A. Calladine3

2School of Plant Biology, Faculty of Natural and Agricultural Sciences, The University of Western Australia, 35 Stirling Highway, Crawley, Western Australia, 6009, Australia; 3School of Tropical Biology, James Cook University, Townsville, Queensland, 4811, Australia

Received for publication November 20, 2003. Accepted for publication May 13, 2004.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 LITERATURE CITED
 
In phylogenetic analyses of nuclear ITS and chloroplast trnL DNA sequences, the mostly endemic Australian genera; Halosarcia, Pachycornia, Sclerostegia, Tecticornia, and Tegicornia of the subfamily Salicornioideae (Chenopodiaceae) together form a monophyletic group, congruent with the hypothesis that they evolved from a common ancestor. However, limited genetic differentiation evident in both nrDNA and cpDNA sequences among these taxa suggests a possible rapid radiation. Based on fossil pollen records and climatic models of other authors, it is hypothesized that the expansion of the Australian endemic Salicornioideae most likely occurred during the Late Miocene to Pliocene, when increasing aridity caused the formation of extensive salt lakes along endorheic paleodrainage channels. Moreover, Australian Sarcocornia representatives were supported as monophyletic, nested within a paraphyletic Sarcocornia clade that also comprised European Salicornia in the ITS analysis. This suggests that Sarcocornia arrived in Australia subsequent to the ancestor of the Australian endemic genera most likely via long-distance dispersal.

Key Words: Chenopodiaceae • Halosarcia • ITS • phylogeny • rapid radiation • Salicornioideae • trnL


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 LITERATURE CITED
 
This is the first detailed molecular study that focuses on Australian members of the subfamily Salicornioideae Ulbr. (Chenopodiaceae Vent.), a taxonomically difficult group that has few diagnostic features at the generic level. The Salicornioideae are characterized by reduced leaves or fleshy node-like stems with a spiciform inflorescence of highly reduced flowers (Dalby, 1962 ; Scott, 1977 ; Kühn et al., 1993 ). Currently, 15 genera and approximately 80 species in two tribes, the Halopeplideae Ulbr. and the Salicornieae Dumort. (Kühn et al., 1993 ), are recognized (Table 1). They are globally distributed and commonly halophytic, naturally growing along coastlines and brackish waterways (Davy et al., 2001 ).


View this table:
[in this window]
[in a new window]
  		Table 1. The circumscription of the subfamily Salicornioideae Ulbr. sensu Kühn (1993) (except here Sarcocornia is recognized separately from Salicornia sensu Scott [1977] ), the number of species currently recognized within each genus, the number of taxa included in this study, and their global distribution

 
In Australia, there are 36 species and 13 subspecies in six genera; Halosarcia Paul G. Wilson, Sarcocornia Scott, Sclerostegia Paul G. Wilson, Pachycornia Hook.f., Tecticornia Hook.f., and Tegicornia Paul G. Wilson (Wilson, 1984 ). Of these, only four taxa extend beyond the continent. Halosarcia indica (Willd.) Paul G. Wilson subsp. indica found along the northern coastline of Australia is also present in southern Asia and Africa, while a second subspecies H. indica (Willd.) Paul G. Wilson subsp. leiostachya (Benth.) Paul G. Wilson is present in Malaysia and Indonesia (van Steenis, 1983 ). Another northern coastal species, Tecticornia australasica (Moq.) Paul G. Wilson, extends to southeast Asia, while Sarcocornia quinqueflora (Ung.-Sternb.) Scott subsp. quinqueflora is also present in New Zealand and Africa (Wilson, 1972 , 1980 ). In Australia, members of the subfamily are almost ubiquitous around the shorelines of inland salt lakes in the semi-arid and arid zones (Wilson, 1980 ). Accordingly, research interest in the group is increasing because of their potential use for the reclamation of salt-affected lands, an increasing problem in the semi-arid zone of Australia.

A close alliance between the Chenopodiaceae and Amaranthaceae has long been recognized because of a number of synapomorphic characters including similar floral characteristics, embryogeny, phytochemistry, sieve-element plastids, and pollen morphology (Kühn et al., 1993 ; Townsend, 1993 ; Judd and Ferguson, 1999 ). Molecular phylogenetic analyses confirm this close relationship (Chase et al., 1993 ; Downie et al., 1997 ; Meimberg et al., 2000 ; Savolainen et al., 2000 ; Soltis et al., 2000 ; Cuénoud et al., 2002 ; Kadereit et al., 2003 ). Currently, the Chenopodiaceae have been placed into synonymy under Amaranthaceae by some authors (Angiosperm Phylogeny Group, 1998 , 2003 ) although the exact phylogenetic relationships of this alliance still remain unclear. Accordingly, we are currently following the traditional taxonomic classification of these families.

Due to their characteristic morphological features, the Salicornioideae are easily recognized within Chenopodiaceae; however, the sister group relationship of the subfamily is unclear. A molecular study by Downie et al. (1997) , using cpDNA ORF2280 sequence data, placed Salicornia bigelovii Torr. sister to a clade comprising Suaeda Forsk. ex Scop., Kochia Roth and Camphorosma L. Additional cpDNA evidence based on the rbcL region by Kadereit et al. (2003) , which included a broader sampling of representatives of both Salicornioideae and Suaedoideae Ulbr., resolved that these subfamilies form a weekly supported clade with the monotypic genus Bienertia Bunge placed sister to the Salicornioideae. This is congruent with the findings of Schütze et al. (2003) based on internal transcribed spacer (ITS) sequence data; however, this study also showed that Bienertia placed sister to Suaeda in a phylogenetic analysis based on atpB-rbcL cpDNA data. It is clear that a wider sampling of species across this broad group and possible examination of more variable sequence regions may be required to truly clarify the sister group relationship of the Salicornioideae.

Infrageneric relationships and species delimitations are poorly understood within the Salicornioideae. Diagnostic characters are typically reduced and obscure, and plants have high levels of phenotypic plasticity under different spatial or seasonal conditions. This variability makes identifying taxa, even to the generic level, difficult. The varying treatments of the perennial Sarcocornia demonstrate some of the confusion evident within the Salicornioideae. In Europe, Sarcocornia is variously considered a separate genus sensu Scott (1977) (Stace, 1997 ; Davy et al., 2001 ) or synonymous with either Arthrocnemum Moq. sensu Moss (1954) (Aellen et al., 1967 ; Ball, 1993 ) or Salicornia L. (Freitag, 2000 ). In North America, Sarcocornia may be considered synonymous with Salicornia (Wiggins, 1980 ; Godfrey and Wooten, 1981 ; Wilken, 1993 ; Judd and Ferguson, 1999 ), as may Arthrocnemum subterminalis (Parish) Standl. (Wiggins, 1980 ; Wilken, 1993 ). However, in Australia, New Zealand, Africa, and Spain, Sarcocornia is still considered a distinct genus from both Salicornia and Arthrocnemum (Wilson, 1980 ; Connor, 1984 ; Castroviejo et al., 1990 ; O'Callaghan, 1992 ). Taxonomy of the Australian representatives of the Salicornioideae is complicated by the presence of large subspecies complexes, as well as putative interspecific and intergeneric hybrids (Wilson, 1980 ). Polyploidy is also evident at both the species and subspecies level (Shepherd and Yan, 2003 ).

Prior to Wilson's examination of the Australian Salicornioideae (Wilson, 1972 , 1980 , 1984 ), only four genera were recognized in the region; Sarcocornia, Arthrocnemum, Pachycornia, and Tecticornia. Wilson (1980) segregated most of the Australian Arthrocnemum representatives into the large genus Halosarcia, with five species placed in Sclerostegia and one in Tegicornia. Based on the presence of shared morphological features, Wilson (1980) suggested that the Australian, mostly endemic genera Halosarcia, Pachycornia, Sclerostegia, Tecticornia, and Tegicornia (hereafter referred to as endemic Australian genera) had a greater affinity to each other than to other Salicornioideae and accordingly, that they most likely evolved from a single common ancestor.

The primary aim of this study was to investigate the phylogenetic relationships of the endemic Australian genera Halosarcia, Pachycornia, Sclerostegia, Tecticornia, Tegicornia, and the Australian representatives of the global genus Sarcocornia using sequence data from both the nuclear ITS region and chloroplast trnL locus. Molecular phylogenetic techniques were used with the hope they could more readily elucidate the phylogenetic relationships within this group, which has such problematic morphological characters.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 LITERATURE CITED
 
Taxon sampling
Fifty-seven representatives were sampled from the six Salicornioideae genera currently recognized in Australia. We aimed to cover the geographic disjunctions among Australian species along with multiple representatives of taxa that had significant morphological plasticity (Appendix; see Supplementary Data accompanying the online version of this article). In addition, unusual undescribed taxa showing affinity to currently recognized species (affn.) and several putative hybrids (x) were included. Non-Australian representatives included Sarcocornia fruticosa (L.) Scott, Arthrocnemum macrostachyum (Moric.) Koch, and Salicornia ramosissima Woods from Spain; Salicornia procumbens Sm. from the Netherlands; and Sarcocornia pacifica (Standl.) Scott from Ensenada, Mexico.

This study was undertaken prior to the publication of the studies of Schütze et al. (2003) and Kadereit et al. (2003) ; therefore, the possible sister-group relationship of Bienertia cycloptera to the Salicornioideae was not known. Outgroup representatives were selected from other Chenopodiaceae subfamilies present in Western Australia and included Suaeda australis Moq. (subfamily Salsoloideae Ulbr., tribe Suaedeae Moq.), Atriplex codonocarpa Paul G. Wilson (subfamily Chenopodioideae Burnett, tribe Atripliceae Mey.), and Maireana triptera (Benth.) Paul G. Wilson (subfamily Chenopodioideae, tribe Sclerolaeneae Scott).

Laboratory procedures
Fresh shoot material was collected and either placed in silica gel or wrapped in foil and placed in dry ice and later freeze dried. A small number of samples were obtained from dried voucher specimens. DNA was isolated from approximately 0.02–0.1 g of dry tissue following a standard SDS (sodium dodecyl sulfate) extraction protocol (Waycott and Barnes, 2001 ) and using DNeasy Plant minikit, QIAGEN, Victoria, Australia, following the manufacturer's instructions, except for the final elution of 50 µL. DNA amplification of the nuclear ITS-1, 5.8S, ITS-2 rDNA region (hereafter, ITS) was conducted using the reverse primer (ITS-4) (White et al., 1990 ). However, a new primer, ITS-16, was designed upstream of White's ITS-1 (5'-CCGATTGAATGGTCCGGTGAAGTGCTCG-3') to enhance polymerase chain reaction (PCR) reproducibility. A number of cpDNA loci were tested for variability among Australian taxa but not analyzed because of the low phylogenetic signal. These included matK, ndhF, and the ATPB-rbcL intergenic spacer. The chloroplast region between the trnL-UAA 5' intron and the trnF exon including the trnL-trnF intergenic spacer (hereafter, trnL) using C and F primers (Taberlet et al., 1991 ) proved variable. However, in a number of Salicornioideae taxa, the presence of a long repeat of As at the 5' end resulted in problematic DNA sequencing. The subsequent design of a suitable primer KStrnL (5'-GAAGTTGTTCTAACAAATGGG-3') that annealed on the 3' side of the long repeat region improved results.

All PCR reactions were carried out in a total volume of 25 µL containing 13 µL H2O, 5 µL Q-Solution (supplied with Taq DNA polymerase, Qiagen [Victoria, Australia]), 2.5 µL PCR 10% buffer (containing 15 mmol/L MgCl2), 0.5 µL MgCl2 (25 mmol/L), 0.4 µL dNTP mix (containing 10 mmol/L of each dNTP), 1.2 µL of each primer (10 µmol/L), 0.2 µL Taq polymerase (5 units/µL), and 1 µL DNA template (approximately 20 ng total genomic DNA) in an MJ-Research PTC-200 thermocycler (Prolabo, France). For ITS, an initial denaturation for 5 min at 96°C was followed by 30 cycles of 30 s at 96°C, 15 s at 55°C, 45 s at 72°C, then a final step of 10 min at 72°C. For trnL, initial denaturation for 1.5 min at 94°C was followed by 35 cycles of 30 s at 94°C, 30 s at 55°C, 1 min at 72°C, then 5 min at 72°C. The PCR products were purified from a 1% TAE (Tris-acetic acid-EDTA) agarose gel using the QIAquick gel extraction kit (Qiagen). Cycle sequencing reactions were conducted in a total volume of 10 µL, with 4 µL of the BigDye v2 Terminator Chemistry (Applied Biosystems, Foster City, California, USA), approximately 80–100 ng of PCR product and 4 pmol primer, using the cycle-sequencing program of the manufacturer. Unincorporated nucleotides were removed using a Sephadex G-50 column, and the sample was then dried before being run on an ABI377 automated sequencer (Perkin-Elmer, Applied Biosystems, Foster City, California).

Chromatograms were edited in Sequencher 3.1 (GeneCodes, 1991–1998 ) and imported into Se-Al (Rambaut, 1995 ) for alignment. Sequence alignment was relatively unambiguous in trnL because of the low variation; therefore, sequences were aligned manually. Alignment was slightly more complicated in ITS as a result of the variation between the outgroup taxa and the ingroup, Salicornioideae. The ITS sequences were further aligned using Clustal X (Jeanmougin et al., 1998 ) with various gap-coding penalties to provide the most reliable alignment. The DNA sequences have been deposited with Genbank (http://www.ncbi.nlm.nih.gov/Genbank/GenbankSearch.html). Individual base positions were coded as unordered multistate characters. Gaps were treated as missing data. Informative indels were coded as present (1) or absent (0) according to Simmons and Ochoterena (2000) .

Phylogenetic analyses
Three separate data sets were analyzed, each of the two loci independently and then a combined analysis. Maximum parsimony (MP) and maximum likelihood (ML) analyses were conducted using PAUP* 4.0 (Swofford, 2002 ). To assess the degree of nonrandom structure attributed to phylogenetic signal, the g1 statistic indicating the skewness of tree length distributions of 1000 random trees was calculated (Hillis and Huelsenbeck, 1992 ). Equal weight parsimony analyses were conducted using heuristic search options with tree-bisection-reconnection (TBR) branch swapping, 1000 random addition sequence replications, gaps scored as missing, and including indel groups scored as present (1) and absent (0). Bootstrap support was calculated using 500 heuristic replications, with heuristic options set at simple addition sequence, TBR branch swapping.

Modeltest (Posada, 1998 ) was used for each of the data sets to determine the most appropriate evolutionary model for conducting maximum likelihood analysis. Because of the large number of equally parsimonious trees obtained for the trnL data, parsimony analyses were recalculated excluding coded gap regions and uninformative characters. The maximum likelihood analyses utilized the calculated Modeltest maximum likelihood settings with a heuristic search strategy comprising random addition sequence replications with TBR to completion. Bootstrap support for the maximum likelihood results was calculated as for maximum parsimony using the calculated Modeltest maximum likelihood settings. Further, an additional analysis using a partial trnL alignment (655 base pairs [bp], no gap coding) including Sarcocornia fruticosa (excluded from the full trnL analysis due to missing data), the Sarcocornia/ Salicornia taxa, Arthrocnemum macrostachyum, and Halosarcia indica (Willd.) Paul G. Wilson subsp. bidens (Nees) Paul G. Wilson was carried out as for other analyses and using Atriplex as the outgroup.

A partition homogeneity test to determine the congruency of the combined ITS and trnL data was performed with 100 replicates using a heuristic search option. Successive analyses were performed excluding taxa to determine the source of incongruence. The final combined ITS and trnL data set was analyzed with uninformative characters excluded using equally weighted parsimony analyses. Maximum likelihood and bootstrap analyses were performed using the same options as stated earlier.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 LITERATURE CITED
 
Maximum parsimony and maximum likelihood analyses
The aligned ITS data set included 59 taxa and 953 characters including coded gap regions. Of the total, 189 characters were variable but parsimony uninformative and 246 were informative. Phylogenetic analyses indicated nonrandom structure (g1 statistic = –0.902, significant at P = 0.01). The heuristic search yielded 98 most parsimonious trees of 1040 steps with a consistency index (CI) of 0.585, a retention index (RI) of 0.673, and a rescaled consistency index (RC) of 0.393 (Fig. 1). Bootstrap support for tree structure is indicated above supported nodes on a phylogram randomly chosen from the 98 most parsimonious trees (Fig. 1). ModelTest analysis identified the appropriate molecular evolution model as GTR + G + I, and the maximum likelihood analysis resulted in a tree of similar topology to the parsimony analysis. This tree had a –ln likelihood = 5487.55. Maximum likelihood bootstrap support values are indicated below the nodes (Fig. 1).



View larger version (44K):
[in this window]
[in a new window]
  		Fig. 1. Parsimony analysis of rDNA ITS data, a randomly chosen phylogram from 98 equally most parsimonious trees (1040 steps, 246 informative characters, consistency index [CI] = 0.585, retention index [RI] = 0.673, rescaled consistency index [RC] = 0.393). Bootstrap values >60% for maximum parsimony appear above the branch and for maximum likelihood, below. Supported clades are in boldface type

 
The trnL alignment for 44 taxa was 872 bp long. There were an additional 48 coded gap characters, so the final data set comprised 920 characters, of which only 79 were parsimony informative. The tree length skewness test confirmed that the results were due to nonrandom structure (g1 = –1.239, significant at P = 0.01). Maximum parsimony analysis including gap coding resulted in 142 700 equally parsimonious trees of 245 steps (CI = 0.882, RI = 0.868, and RC = 0.765). Maximum parsimony analysis with coded gap regions and uninformative characters excluded (leaving 51 parsimony-informative characters) resulted in 234 most parsimonious trees of 156 steps (CI = 0.891, RI = 0.899, and RC = 0.801) (Fig. 2). The maximum likelihood analysis using the GTR + I molecular evolution model (output from ModelTest) resulted in the tree length –ln likelihood of 476.52. Both analyses resulted in trees with a basal polytomy and only a few supported clades. Bootstrap support for both MP and ML analyses are indicated from a randomly chosen phylogram above and below the nodes respectively (Fig. 2).



View larger version (34K):
[in this window]
[in a new window]
  		Fig. 2. Parsimony analysis of cpDNA trnL data (uninformative characters and coded gap regions excluded), a randomly chosen phylogram from 234 equally most parsimonious trees (156 steps, 51 informative characters, CI = 0.891, RI = 0.899, RC = 0.801). Bootstrap values >60% for maximum parsimony appear above the branch and for maximum likelihood, below. Supported clades are in boldface type

 
The ITS and trnL data sets were both available for 43 taxa and the partition homogeneity test comparing these data sets indicated that they were significantly incongruent (P = 0.01). The successive removal of eight taxa, targeting putative hybrids and multiple representatives within species complexes, resulted in a nonsignificant P value of 0.182, indicating that the remaining data could be combined. The combined data included 35 taxa and comprised 1873 characters of which 232 were parsimony informative. Again the tree length skewness test confirmed nonrandom structure (g1 = –1.47, significant at P = 0.01). Maximum parsimony analysis excluding uninformative sites resulted in 90 equally most parsimonious trees (911 steps, CI = 0.708, RI = 0.678, RC = 0.480). Trees of equivalent topology were recovered from the ML analysis using the GTR + G + I evolutionary model (–ln likelihood = 6327.39) (Fig. 3). Bootstrap values are indicated on supported nodes for both parsimony (above) and maximum likelihood (below) on a phylogram randomly chosen from the 90 most parsimonious trees (Fig. 3). All supported nodes were present in all of the most parsimonious trees, as seen in the strict consensus tree of each analysis.



View larger version (27K):
[in this window]
[in a new window]
  		Fig. 3. Parsimony analysis of combined ITS and trnL data (uninformative characters excluded), a randomly chosen phylogram from 90 equally most parsimonious trees (911 steps, 232 informative characters, CI = 0.708, RI = 0.678, RC = 0.480). Bootstrap values >60% for maximum parsimony appear above the branch and for maximum likelihood, below. Supported clades are in boldface type

 
Phylogenetic reconstruction
The Australian endemic genera Halosarcia, Pachycornia, Sclerostegia, Tecticornia, and Tegicornia form a strongly supported clade (82% bootstrap support) in the ITS analysis (Fig. 1). In the ITS, trnL and combined analyses Sarcocornia and Salicornia form a very strong clade with 100, 95, 100% bootstrap support, respectively (Figs. 1–3). However, Sarcocornia may not be monophyletic because the Salicornia representatives are nested within Sarcocornia in both the ITS and the reduced analysis based on a subset of the trnL sequence data (Figs. 1, 4). The three Australian Sarcocornia species form a monophyletic group with strong bootstrap support in the ITS, trnL, and combined analyses (84, 92, and 96%, respectively) (Figs. 1–3). In the reduced trnL analysis, however, Sarcocornia quinqueflora subsp. quinqueflora is unresolved with respect to S. blackiana (Ulbr.) Scott and S. globosa Paul G. Wilson and the extra-Australian Sarcocornia and Salicornia species (Fig. 4). The position of Arthrocnemum macrostachyum is unresolved with respect to the Salicornia/Sarcocornia clade and the Australian endemic clade in the ITS analysis (Fig. 1). However, in the combined data set, it is resolved with moderate bootstrap support (66%) sister to the clade of the endemic genera (Fig. 3).



View larger version (18K):
[in this window]
[in a new window]
  		Fig. 4. Parsimony analysis of a subset of the cpDNA trnL data resolved a single most parsimonious tree (45 steps, 21 informative characters, tree scores equal 1). Maximum parsimony bootstrap values >60% appear above the branch

 
It is apparent from the ITS, trnL, and combined analyses that the individual Australian endemic genera are not supported as monophyletic (Figs. 1–3). The overall lack of structure in the endemic clade with short branch lengths results in a largely unresolved polytomy. In the ITS analysis, only a few terminal pairs of closely related species receive high bootstrap support such as Halosarcia leptoclada Paul G. Wilson subsp. inclusa Paul G. Wilson (KS888 and KS646) (100%); Sclerostegia disarticulata Paul G. Wilson (KS10) and S. disarticulata affn. (KS880) (90%); H. syncarpa Paul G. Wilson (KS721) and H. syncarpa affn. (KS879) (99%); and H. lepidosperma Paul G. Wilson (KS704 and KS726) (100%) (Fig. 1). The monotypic genus Tegicornia Paul G. Wilson is placed sister to H. lylei (Ewart et White) Paul G. Wilson (KS711) in the ITS and combined analysis with strong bootstrap support (79 and 94%, respectively). Tecticornia, represented by T. verrucosa Paul G. Wilson (KSQ6) in the ITS analysis, is positioned within the Australian endemic clade sister to H. peltata; however, this relationship is unsupported. A small clade comprising Sclerostegia tenuis (Benth.) Paul G. Wilson (KS482), S. moniliformis Paul G. Wilson (KS732), and S. sp. Hamersley Lake (K. Shepherd 887) (KS887) is well supported in the ITS analysis (Fig. 1). Two larger clades are also evident in the combined analysis. A Sclerostegia clade that includes Pachycornia triandra (F. Muell.) Black (KSQ2) and three Halosarcia species, H. entrichoma Paul G. Wilson (KS891), H. syncarpa affn. (KS879), and H. sp. Gunyidi (M. N. Lyons 2607) (KS897), received 75% bootstrap support. Also a Halosarcia clade comprising H. chartacea Paul G. Wilson, H. undulata Paul G. Wilson, H. auriculata Paul G. Wilson, H. calyptrata Paul G. Wilson, H. bulbosa Paul G. Wilson, H cymbiformis Paul G. Wilson MS, and H. sp. Angelfish Island affn. received 86% bootstrap support in the maximum likelihood analysis (Fig. 3).

Halosarcia indica is a widely distributed species that forms a well-supported clade, including two subspecies and H. sp. Yanneri Lake (S. van Leeuwen 3002), an undescribed species from the Little Sandy Desert in Western Australia, in both the ITS and trnL analyses (Figs. 1 and 2). Halosarcia sp. Yanneri Lake x (KS841) is a putative hybrid between H. indica and H. sp. Yanneri Lake (S. van Leeuwen 3002). In the ITS analysis, H. sp. Yanneri Lake x (KS841) was positioned between the two putative parent species, and in the trnL analysis, this taxon was placed sister to H. sp. Yanneri Lake within the H. indica clade. Another widely distributed species, H. halocnemoides, is not supported as monophyletic, because H. halocnemoides (Nees) Paul G. Wilson subsp. halocnemoides (KS740) was placed sister to H. fimbriata Paul G. Wilson (KS702) in the ITS analysis with 100% bootstrap support. Another putative hybrid H. halocnemoides x (KS763) collected near Kalgoorlie in Western Australia, which has a habit similar to H. halocnemoides and seed typical of H. pergranulata (Black) Paul G. Wilson, was placed within the H. halocnemoides clade with strong bootstrap support (100%). Moreover, Tegicornia uniflora x (GKA), a taxon that was thought by Wilson (1980) to be an intergeneric hybrid of an unknown Halosarcia species and Tegicornia uniflora Paul G. Wilson (of which the nearest population is located approximately 360 km away), was also placed sister to the H. halocnemoides clade with strong bootstrap support (95%). Incongruently, the trnL analysis places the putative T. uniflora x hybrid (GKA) sister to H. flabelliformis Paul G. Wilson (KS883) and Halosarcia halocnemoides (Nees) Paul G. Wilson subsp. longispicata Paul G. Wilson (IC3163) from the Northern Territory. In the combined ITS-trnL analysis with the putative hybrids and multiple representatives of a species removed, H. halocnemoides (Nees) Paul G. Wilson subsp. tenuis Paul G. Wilson (KS791) is placed sister to the remaining Australian species (Fig. 3).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
LITERATURE CITED
 
The phylogenetic hypotheses generated from both the nuclear ITS sequence data and the combined ITS and cpDNA trnL data strongly support the monophyly of the endemic Australian Salicornioideae genera; Halosarcia, Pachycornia, Sclerostegia, Tecticornia, and Tegicornia. This is congruent with Wilson's (1980) hypothesis that they evolved from a single common ancestor. The very low sequence variation among the endemic Australian genera has resulted in a largely unresolved polytomy with only terminal branches of closely related species receiving significant bootstrap support in the separate ITS and trnL analyses. It is evident that the generic boundaries as delimited by Wilson (1972 , 1980 , 1984 ) are not supported by this molecular study. The presence of two supported clades in the combined analysis, an expanded Sclerostegia clade comprising Pachycornia and some Halosarcia species and a smaller Halosarcia clade, suggest that previously unrecognized groups may be present. Further examination of more variable sequence regions in conjunction with a morphological phylogenetic analysis is required to clarify these relationships.

The three Sarcocornia species present in Australia are supported as a monophyletic group and are more closely related to the northern hemisphere Salicornia and Sarcocornia than to the Australian endemic genera. The longer branch that separates the well-supported Sarcocornia/Salicornia clade from the other Salicornioideae included in this study indicates an older divergence or more rapid evolution with subsequent recent speciation.

Wilson (1980) proposed that Arthrocnemum is a possible intermediate between Sarcocornia and the Australian endemic genera based on morphological characters shared with both groups. The monophyly of the Australian endemic genera supports their separation from Arthrocnemum (Wilson, 1980 , 1984 ); however, the phylogenetic position of Arthrocnemum macrostachyum is not fully resolved. In the combined analysis, A. macrostachyum is placed sister to the Salicornia/Sarcocornia and the endemic Australian clade, and accordingly, Wilson's (1980) hypothesis is not controverted. However, to resolve the relationships of Arthrocnemum, future molecular studies must include A. subterminalis from North America and representatives of the smaller, less well-known Salicornieae genera, such as Halostachys from Southwest and Central Asia and Heterostachys from South and Central America to confirm the monophyly of the Australian endemic genera clade and the possible sister relationship with Arthrocnemum. Moreover, representatives of the tribe Halopeplideae should also be included to fully understand the phylogenetic relationships of the Salicornioideae.

The lack of phylogenetic structure among the Australian species precludes us from inferring which lineage is most ancient. Three taxa from the endemic clade occur in the north coastal regions of Australia and extend beyond the continent to surrounding regions: Halosarcia indica subsp. indica, which colonizes tidal mudflats along tropical coasts bordering the Indian Ocean and the east coast of Africa; the closely related Halosarcia indica subsp. leiostachya present in Malaysia and Indonesia; and Tecticornia australasica (Moq.) Paul G. Wilson, which is also found in Papua New Guinea and Malaysia. Tecticornia australasica was not included in this study although nuclear DNA was obtained from T. verrucosa, an apparently closely related short-lived perennial from Western Australia. It is clear that H. indica and T. verrucosa remained unplaced with respect to the rest of the Australian group in the ITS analysis; therefore, neither are likely to be an early divergent within this monophyletic group. The long branch supporting the Salicornia/Sarcocornia clade suggests an early divergence, with the subsequent speciation of the Australian Sarcocornia lineage occurring more recently. Sarcocornia quinqueflora could be the founder for this group as it is the most widespread Sarcocornia species in Australia and is also found in New Zealand and southern Africa. Alternatively, S. quinqueflora may have an autochthonous origin derived from an ancestral species, which has subsequently dispersed.

The influence of including hybrids and polyploids in phylogenetic analyses is potentially problematic. Studies of known hybrids, particularly those formed between closely related taxa, suggest that they may align to one adult or in theory, be placed basally, having character traits of multiple origins (Funk, 1985 ; McDade, 1990 , 1992 ). According to McDade (1992) , major topological changes and low consistency indices may occur when hybrids are included in cladistic analyses. In addition, examining both cpDNA and nrDNA loci may provide evidence of unsuspected hybridization where conflicts between the data sets occur. In this study, topology overall remained consistent between all analyses, even when putative hybrid taxa were removed. Differences were observed in the terminal taxon pairing of two taxa, Halosarcia undulata that aligned with H. pruinosa in the ITS analysis and H. calyptrata in the trnL analysis and that of Tegicornia uniflora x (GKA) also aligned with different species. To resolve relationships among the remaining taxa, the incongruence between the two data sets was resolved with the removal of the putative hybrids and multiple representatives of terminal taxa. This suggests that hybridization may be occurring, and further study of these species is required to clarify their origin.

Polyploids and putative hybrids have been frequently reported in the Salicornioideae (Ball and Tutin, 1959 ; Dalby, 1962 ; Ball, 1967 ; Contandriopoulos, 1968 ; Hekmat-Shoar, 1978 ; Castroviejo and Coello, 1980 ; Wilson, 1980 ; Wolff and Jeffries, 1987 ; Castroviejo and Lago, 1992 ; O'Callaghan, 1992 ; Shepherd and Yan, 2003 ) although cytological studies have not confirmed if auto- or allopolyploids predominate. Autopolyploidy is certainly thought to occur; Halosarcia indica (Willd.) Paul G. Wilson subsp. bidens (Nees.) Paul G. Wilson triploids were suggested to be produced from tetraploids backcrossing to diploid progenitors because all three ploidy levels were detected within this subspecies (Shepherd and Yan, 2003 ). Clearly, further resolution of polyploidy and hybridization in this group would benefit our understanding of their origins and systematic relationships that require detailed studies of hybrid "taxa" and their progenitors.

The low level of sequence variation among the Australian endemic genera is unexpected considering the morphological diversity observed, particularly in Halosarcia, the largest genus of the Salicornioideae. This suggests either a slower than expected rate of evolution of the DNA sequences examined or, more likely, that the Australian lineage has radiated rapidly with insufficient time to accumulate divergence. Similar radiations have been observed in other families of this order. For example, molecular analyses of the Aizoaceae present in the Karoo region of southern Africa and the western American Portulacaceae (Klak et al., 2003 ) failed to resolve phylogenetic relationships at both the species and generic levels because of low sequence variation (Hershkovitz and Zimmer, 2000 ; Klak et al., 2003 ). Molecular analyses utilizing the ITS rDNA region of some of the larger Australian arid zone genera, Eucalyptus subgenus Symphyomyrtus (Steane et al., 1999 , 2002 ; Ladiges et al., 2003 ), Acacia subgenus Phyllodineae (Murphy et al., 2003 ), and Melaleuca (Brown et al., 2001 ), show low divergence despite significant morphological variation. Like the group studied here, these results suggest relatively recent and rapid radiations.

The origin of the order Caryophyllales is estimated on the basis of fossil evidence to be 83 million years before present (Ma) (Magallón et al., 1999 ) or 104–111 Ma using estimates of divergence in plastid and nuclear DNA sequence data (Wikström et al., 2001 ). The origin of the subfamily Salicornioideae is more recent because it is placed well within the Chenopodiaceae-Amaranthaceae, which are a derived lineage in the Caryophyllales (Downie et al., 1997 ; Angiosperm Phylogeny Group, 2003 ). Carolin (1982) argued that some elements of the arid flora may be ancient and postulated that Tertiary vicariance of the Australian Salicornioideae and other endemic Caryophyllales from related African and Eurasian elements accounted for the morphological diversity and unique nature of these endemic genera. The molecular phylogenetic evidence of this study suggests that a founder element of the Australian Salicornioideae arrived and proliferated relatively rapidly, and we propose that this occurred during the late Miocene to Pliocene. Palaeoclimatic studies based on geological evidence indicate that regions of Australia were possibly becoming more arid during the late Eocene (around 40–45 Ma) (Truswell and Harris, 1982 ) and thus, elements of arid flora could have established during this time. However, inland Australia only became extensively arid during the Pliocence/Pleistocene (less than 5 Ma), and this increasing aridity was followed by fluctuating wet and dry cycles throughout the Pleistocene (van de Graff et al., 1977 ; Frakes, 1999 ; Dodson and Ramrath, 2001 ). Examination of deep well cuttings in Central Australia indicated that Chenopodiaceous pollen does not appear in the fossil record until the late Miocene/Pliocene, approximately 6–9 Ma (Martin, 1998 ). It is likely that rapid changes in the climate and an increase in available saline habitats facilitated the recent radiation of the ancestral Salicornioideae element.

Floristic similarity at the generic and species level across disjunct arid regions of the world may be indicative of long-distance dispersal (Shmida, 1985 ). The successful dispersal of Salicornioideae species over long distances is supported by their establishment on every continent with the exclusion of Antarctica, and genera such as Sarcocornia and Salicornia are almost cosmopolitan. The Salicornioideae are likely to have arrived via long-distance dispersal in Australia in two separate events, with the ancestor of the endemic element establishing first, possibly around the late Miocene/early Pliocene and the Australian Sarcocornia lineage appearing at a later time. There is evidence of other Australian and New Zealand taxa that have arrived via long-distance dispersal and subsequently radiated during this period including Microseris (Asteraceae) (Vijverberg et al., 1999 ), Lepidium (Brassicaceae) (Mummenhoff et al., 2001 ), and Sophora sect. Edwardsia (Fabaceae) (Mitchell and Heenan, 2002 ).

Genera common to disjunct arid zones often include species that inhabit both coastal and desert areas (Shmida, 1985 ). Coastal halophytic species are preadapted to the harsh environments surrounding inland salt lakes as coastal areas comprise similar sandy, salty habitats. The founder element of Salicornioideae species were probably littoral, facilitating the dispersal of seeds or seedlings across great distances to Australia. Moreover, the corky nature of some Salicornioideae fruits allows them to float for up to 3 mo (Dalby, 1963 ), facilitating their dispersal over long distances. The arid zone within Australia extends to the coastline in the northwest of Western Australia and would provide ample opportunity for newly established, salt-adapted littoral species to subsequently disperse inland.

In Australia, minimal tectonic activity has resulted in a very subdued relief across the continent. Moreover, increased levels of precipitation during the Tertiary led to the formation of a vast interconnected endorheic palaeodrainage system that extended across the continent (Bettenay, 1962 ; Johnson, 1980 ). As the climate became more arid during the Miocene to Pleistocene, the ancient rivers began to contract and a vast network of modern salt lakes formed following this ancient system (van de Graff et al., 1977 ; De Deckker, 1983 ). These strings of salt lakes provide innumerable saline habitats available for colonization by salt-tolerant species. Moreover, these lakes can also intermittently reconnect along old flow lines after heavy rainfall periods (Bettenay, 1962 ), sometimes for hundreds of kilometers (van de Graff et al., 1977 ), facilitating the spread of seed. Further, strong winds can spread pollen and seed across large distances of the continent, increasing the chances of plants colonizing new habitats and subsequently undergoing further speciation.

Conclusions
The recent radiation of the Salicornioideae in Australia is most likely linked to the increased availability of new habitats as Australia became more arid. The increase in aridity and salinisation of the inland lake system probably began during late Miocene approximately 6–9 Ma, with widespread aridity present across the continent by 0.7 Ma during the Pleistocene. The inherent features of the plants promoted the rapid diversification across Australia: tolerance to extreme conditions, wind pollination, small, easily dispersed seeds, and labile genetic systems. The low sequence variation evident at both the ITS and trnL loci suggests that the endemic genera are very closely related. Further morphological and molecular studies are required to clarify current generic level relationships within the Australian representatives, to resolve the phylogenetic relationships within the subfamily, and to determine the origin of the Australian endemic lineage.


   FOOTNOTES
 
1 The authors thank Jeremy English for field assistance, Bernie Dudley for collecting plants from Spain, and Ian Clarke from the Melbourne Royal Botanic Gardens for collecting plants from the Northern Territory and South Australia. Thanks to Arthur Conacher and John Dodson from the School of Earth and Geographical Sciences for discussion about paleodrainage in Australia and to Grant Whiteman, Steve Hopper, and Kris Lemson for discussion and reviewing earlier drafts of the manuscript. K. S. would like to thank Tim Colmer and Terry Macfarlane for supervision, while undertaking a PhD project at The University of Western Australia, and Paul Wilson for enthusiastic discussion about the Australian Salicornioideae. This research was supported by an ARC Grant with linkage support from MERIWA, Normandy Mining Limited, Placer (Granny Smith), Acacia Resources, KCGM and the WA Herbarium. Back

4 kshepher{at}agric.uwa.edu.au Back


   LITERATURE CITED
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 LITERATURE CITED
 
Aellen P. J. Cullen M. J. E. Coode 1967 Arthrocnemum Moq. In P. H. Davis [ed.], Flora of Turkey and the East Aegean Islands, vol. 2, 320–321. Edinburgh University Press, Edinburgh, UK

Angiosperm Phylogeny Group. 1998 An ordinal classification for the families of flowering plants. Annals of the Missouri Botanical Garden 85: 531-553[CrossRef][ISI]

Angiosperm Phylogeny Group. 2003 An update of the Angiosperm Phylogeny Group classification for the orders and families of flowering plants: APG II. Botanical Journal of the Linnean Society 141: 399-436[CrossRef]

Ball P. W. 1967 Salicornia L. In P. H. Davis [ed.], Flora of Turkey and the East Aegean Islands, vol. 2, 321–323. Edinburgh University Press, Edinburgh, UK

Ball P. W. 1993 Arthrocnemum Moq. In D. A. Webb [ed.], Flora Europaea, vol. 1. Cambridge University Press, London, UK

Ball P. W. T. G. Tutin 1959 Notes on the annual species of Salicornia in Britain. Watsonia 4: 193-205

Bettenay E. 1962 The slat lake systems and their associated aeolian features in the semi-arid regions of Western Australia. Journal of Soil Science 13: 10-17[CrossRef][ISI]

Brown G. K. F. Udovicic P. Y. Ladiges 2001 Molecular phylogeny and biogeography of Melaleuca, Callistemon and related genera (Myrtaceae). Australian Systematic Botany 14: 565-585[CrossRef][ISI]

Carolin R. C. 1982 A review and critique of studies on the phytogeography of arid Australia. In P. J. M. Greenslade [ed.], Evolution of the flora and fauna of arid Australia, 119–123. Peacock Publications in association with the Australian Systematic Botany Society and ANZAAS (Australian and New Zealand Association for the Advancement of Science) South Australian Division, Frewville, South Australia, Australia

Castroviejo S. P. Coello 1980 Datos cariológicos y taxonómicos sobre las Salicorniinae A. J. Scott Ibéricas. Anales Jardín Botánico de Madrid 37: 41-73

Castroviejo S. E. Lago 1992 Datos acerca de la hibridación en el género Sarcocornia (Chenopodiaceae). Anales Jardín Botánico de Madrid 50: 163-170

Castroviejo S. M. Laníz G. L. González P. Montserrat F. M. Garmendia J. Paiva L. Villar 1990 Flora Iberica—Plantas vasculares de la Península Ibérica e Islas Baleares, vol. 2. Real Jardín Botánico, Madrid, Spain

Chase M. W. et al 1993 Phylogenetics of seed plants—an analysis of nucleotide sequences from the plastid gene rbcl. Annals of the Missouri Botanical Garden 80: 528-580[CrossRef][ISI]

Connor H. E. 1984 Gynodioecism in Sarcocornia quinqueflora (Salicornieae) in New Zealand. New Zealand Journal of Botany 22: 433-439[ISI]

Contandriopoulos J. 1968 A propos des nombres Chromosomiques des Salicornia de la région méditerranéenne. Bulletin du Museum D'Histoire Naturelle de Marseille 28: 45-52

Cuénoud P. V. Savolainen L. W. Chatrou M. Powell R. J. Grayer M. W. Chase 2002 Molecular phylogenetics of Caryophyllales based on nuclear 18S rDNA and plastid rbcL, atpB and MatK DNA sequences. American Journal of Botany 89: 132-144[Abstract/Free Full Text]

Dalby D. 1963 Seed dispersal in Salicornia pusilla. Nature 199: 197-198[CrossRef]

Dalby D. H. 1962 Chromosome number, morphology and breeding behaviour in the British Salicorniae. Watsonia 5: 150-162

Davy A. J. G. F. Bishop C. S. B. Costa 2001 Salicornia L. (Salicornia pusilla J. Woods, S. ramosissima J. Woods, S. europaea L., S. obscura P. W. Ball & Tutin, S. nitens P. W. Ball & Tutin, S. fragilis P. W. Ball & Tutin and S. dolichostachya Moss). Journal of Ecology 89: 681-707[CrossRef]

De Deckker P. 1983 Australian salt lakes: their history, chemistry and biota—a review. Hydrobiologia 105: 231-244[CrossRef][ISI]

Dodson J. R. A. Ramrath 2001 An Upper Pliocene lacustrine environmental record from south-western Australia—preliminary results. Palaeogeography, Palaeoclimatology, Palaeoecology 167: 309-320

Downie S. R. D. S. Katz-Downie K. Cho 1997 Relationships in the Caryophyllales as suggested by phylogenetic analyses of partial chloroplast DNA ORF2280 homolog sequences. American Journal of Botany 84: 253-273[Abstract]

Frakes L. A. 1999 Evolution of Australian environments. In H. S. Thompson [ed.], Flora of Australia, vol. 1, 163–203. ABRS/CSIRO (Australian Biological Resources Study/Commonwealth Scientific and Industrial Research Organisation), Melbourne, Australia

Freitag H. 2000 Salicornia L. In K. H. C. Baser [ed.], Flora of Turkey and the East Aegean Islands, supplement 2, 59–60. University Press, Edinburgh, UK

Funk V. A. 1985 Phylogenetic patterns and hybridization. Annals of the Missouri Botanical Garden 72: 681-715[CrossRef][ISI]

GeneCodes. 1991–1998 Sequencher. GeneCodes, Ann Arbor, Michigan, USA

Godfrey R. K. J. W. Wooten 1981 Chenopodiaceae (Goosefoot Family), aquatic and wetland plants of southeastern United States, 93– 95. University of Georgia Press, Athens, Georgia, USA

Hekmat-Shoar H. 1978 Contribution à l'étude cytotaxonomique de quelques halophytes d'Azerbaïdjan (Iran). Revue de Cytologie et de Biologie Vegetales—le Botaniste 1: 97-104

Hershkovitz M. A. E. A. Zimmer 2000 Ribosomal DNA evidence and disjunctions of Western American Portulacaceae. Molecular Phylogenetics and Evolution 15: 419-439[CrossRef][ISI][Medline]

Hillis D. M. J. P. Huelsenbeck 1992 Signal, noise and reliability in molecular phylogenetic analyses. Journal of Heredity 83: 189-195[Abstract/Free Full Text]

Jeanmougin F. J. D. Thompson M. Gouy D. G. Higgins T. J. Gibson 1998 Multiple sequence alignment with Clustal X. Trends in Biochemical Sciences 23: 403-405[CrossRef][ISI][Medline]

Johnson M. 1980 The origin of Australia's salt lakes. New South Wales Geological Survey—Records 19: 221-266

Judd W. S. I. K. Ferguson 1999 The genera of Chenopodiaceae in the southeastern United States. Harvard Papers in Botany 4: 365-416

Kadereit G. T. Borsch K. Weising H. Freitag 2003 Phylogeny of Amaranthaceae and Chenopodiaceae and the evolution of C4 photosynthesis. International Journal of Plant Sciences 164: 959-986[CrossRef]

Klak C. T. A. Hedderson H. P. Linder 2003 A molecular systematic study of the Lampranthus group (Aizoaceae) based on the chloroplast trnL-trnF and nuclear ITS and 5S NTS sequence data. Systematic Botany 28: 70-85[ISI]

Kühn U. V. Bittrich R. Carolin H. Reitag I. C. Hedge P. Uotila P. Wilson 1993 Chenopodiaceae. In V. Bittrich [ed.], The families and genera of vascular plants. Flowering plants—dicotyledons, vol. 2, 253–280. Springer-Verlag, Berlin, Germany

Ladiges P. Y. F. Udovicic G. Nelson 2003 Australian biogeographical connections and the phylogeny of large genera in the plant family Myrtaceae. Journal of Biogeography 30: 989-998[ISI]

Magallón S. P. R. Crane P. S. Herendeen 1999 Phylogenetic pattern, diversity, and diversification of eudicots. Annals of the Missouri Botanical Garden 86: 297-372[CrossRef][ISI]

Martin H. A. 1998 Late Cretaceous–Cainozoic palynology of the Poonarunna No. 1 well, Central Australia. Transactions of the Royal Society of South Australia 122: 89-138[ISI]

McDade L. 1990 Hybrids and phylogenetic systematics I. Patterns of character expression in hybrids and their implications for cladistic analysis. Evolution 44: 1685-1700[CrossRef][ISI]

McDade L. A. 1992 Hybrids and phylogenetic systematics II. The impact of hybrids on cladistic analysis. Evolution 46: 1329-1346[CrossRef][ISI]

Meimberg H. P. Dittrich G. Bringmann J. Schlauer G. Heubl 2000 Molecular phylogeny of Caryophyllidae s.l. based on matK sequences with special emphasis on carnivorous taxa. Plant Biology 2: 218-228[CrossRef][ISI]

Mitchell A. D. P. B. Heenan 2002 Sophora sect. Edwardsia (Fabaceae): further evidence from nrDNA sequence data of a recent and rapid radiation around the southern oceans. Botanical Journal of the Linnean Society 140: 435-441[CrossRef]

Moss C. E. 1954 The species of Arthrocnemum and Salicornia in Southern Africa. Journal of South African Botany 20: 1-22

Mummenhoff K. H. Brüggeman J. L. Bowman 2001 Chloroplast DNA phylogeny and biogeography of Lepidium (Brassicaceae). American Journal of Botany 88: 2051-2063[Abstract/Free Full Text]

Murphy D. J. J. T. Miller R. J. Bayer P. Y. Ladiges 2003 Molecular phylogeny of Acacia subgenus Phyllodineae (Mimosoideae: Legumonosae) based on DNA sequences of the internal transcribed spacer region. Australian Systematic Botany 16: 19-26[CrossRef][ISI]

O'Callaghan M. 1992 The ecology and identification of the southern African Salicornieae (Chenopodiaceae). South African Journal of Botany 58: 430-439[ISI]

Posada D. K. A. Crandall 1998 Modeltest: testing the model of DNA substitution. Bioinformatics 14: 817-818[Abstract/Free Full Text]

Rambaut A. 1995 SeAl. Sequence alignment program. University of Oxford, Oxford, UK

Savolainen V. et al 2000 Phylogeny of the eudicots: a nearly complete familial analysis based on rbcL gene sequences. Kew Bulletin 55: 257-309[CrossRef]

Schütze P. H. Freitag K. Weising 2003 An integrated molecular and morphological study of the subfamily Suaedioideae Ulbr. (Chenopodiaceae). Plant Systematics and Evolution 239: 257-286[CrossRef][ISI]

Scott A. J. 1977 Reinstatement and revision of Salicorniaceae J. Agardh (Caryophyllales). Botanical Journal of the Linnean Society 75: 357-374

Shepherd K. A. G. Yan 2003 Chromosome number and size variations in the Australian Salicornioideae (Chenopodiaceae)—evidence of polyploidisation. Australian Journal of Botany 51: 1-12[CrossRef][ISI]

Shmida A. 1985 Biogeography of the desert flora. In I. Noy-Meir [ed.], Hot deserts and arid shrublands. Elsevier, New York, New York, USA

Simmons M. P. H. Ochoterena 2000 Gaps as characters in sequence-based phylogenetic analysis. Systematic Biology 48: 369-381

Soltis D. E. et al 2000 Angiosperm phylogeny inferred from 18S rDNA, rbcL and atpB sequences. Botanical Journal of the Linnean Society 133: 381-461[CrossRef]

Stace C. A. 1997 New flora of the British Isles. Cambridge University Press, London, UK

Steane D. A. G. E. McKinnon R. E. Vaillancourt B. M. Potts 1999 ITS sequence data resolve higher level relationships among the eucalypts. Molecular Phylogenetics and Evolution 12: 215-223[CrossRef][ISI][Medline]

Steane D. A. D. Nicolle G. E. McKinnon R. E. Vaillancourt B. M. Potts 2002 Higher-level relationships among the eucalypts are resolved by ITS-sequence data. Australian Systematic Botany 15: 49-62

Swofford D. L. 2002 PAUP. Phylogenetic analysis using parsimony. Sinauer, Sunderland, Massachusetts, USA

Taberlet P. L. Geilly G. Pautou J. Bouvet 1991 Universal primers for amplification of three non-coding regions of chloroplast DNA. Plant Molecular Biology 17: 1105-1109[CrossRef][ISI][Medline]

Townsend C. C. 1993 Amaranthaceae. In V. Bittrich [ed.], The families and genera of vascular plants, vol. 2, 70–91. Springer-Verlag, Berlin, Germany

Truswell E. M. W. K. Harris 1982 The Cainozoic palaeobotanical record in arid Australia: fossil evidence for the origins of an arid-adapted flora. In P. J. M. Greenslade [ed.], Evolution of the flora and fauna of arid Australia. Peacock Publications in association with Australian Systematic Botany Society and ANZAAS South Australian Division, Frewville, South Australia, Australia

van de Graff W. J. E. R. W. A. Crowe J. A. Bunting M. J. Jackson 1977 Relict early Cainozoic drainages in arid Western Australia. Zeitschrift für Geomorphologie N.F 21: 379-400

van Steenis C. G. G. J. 1983 Chenopodiaceae. In C. G. G. J. van Steenis [ed.], Flora Malesiana, 557. Martinus Nijhoff, The Hague, The Netherlands

Vijverberg K. T. H. M. Mes K. Bachman 1999 Chloroplast DNA evidence for the evolution of Microseris (Asteraceae) in Australia and New Zealand after long-distance dispersal from western North America. American Journal of Botany 86: 1448-1463[Abstract/Free Full Text]

Waycott M. P. A. G. Barnes 2001 AFLP diversity within and between populations of the Caribbean seagrass Thalassia testudinum (Hydrocharitaceae). Marine Biology (Berlin) 139: 1021-1028[CrossRef]

White T. J. T. Burns S. Lee J. Taylor 1990 Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. In T. J. White [ed.], PCR protocols: a guide to methods and applications, 315–322. Academic Press, San Diego, California, USA

Wiggins I. L. 1980 Chenopodiaceae, flora of Baja California, 100–110. Stanford University Press, Stanford, California, USA

Wikström N. V. Savolainen M. W. Chase 2001 Evolution of the angiosperms: calibrating the family tree. Proceedings of the Royal Society of London, series B 268: 2211-2220[Medline]

Wilken D. H. 1993 Chenopodiaceae. In J. C. Hickman [ed.], The Jepson manual: higher plants of California. University of California Press, Berkeley, USA

Wilson P. G. 1972 A taxonomic revision of the genus Tecticornia. Nuytsia 1: 277-288

Wilson P. G. 1980 A revision of the Australian species of Salicornieae (Chenopodiaceae). Nuytsia 3: 1-154

Wilson P. G. 1984 Chenopodiaceae. In A. George [ed.], Flora of Australia, vol. 4, 277–309. Australian Government Publishing Service, Canberra, Australian Capital Territory, Australia

Wolff S. L. R. L. Jeffries 1987 Morphological and isozyme variation in Salicornia europaea (s.l.) (Chenopodiaceae) in northeastern North America. Canadian Journal of Botany 65: 1410-1419




This article has been cited by other articles:


Home page
 J Exp BotHome page
E. V. Voznesenskaya, H. Akhani, N. K. Koteyeva, S. D. X. Chuong, E. H. Roalson, O. Kiirats, V. R. Franceschi, and G. E. Edwards
Structural, biochemical, and physiological characterization of photosynthesis in two C4 subspecies of Tecticornia indica and the C3 species Tecticornia pergranulata (Chenopodiaceae)
J. Exp. Bot., April 4, 2008; (2008) ern028v1.
[Abstract] [Full Text] [PDF]

Home page
 ANN BOT (LOND)Home page
K. A. SHEPHERD, T. D. MACFARLANE, and T. D. COLMER
Morphology, Anatomy and Histochemistry of Salicornioideae (Chenopodiaceae) Fruits and Seeds
Ann. Bot., May 1, 2005; 95(6): 917 - 933.
[Abstract] [Full Text] [PDF]