Australian Journal of Botany

Transcription

1 CSIRO PUBLISHING Australian Journal of Botany Volume 47, 1999 CSIRO Australia 1999 An international journal for the publication of original research in plant science All enquiries and manuscripts should be directed to Australian Journal of Botany CSIRO PUBLISHING PO Box 1139 (15 Oxford St) Collingwood Telephone: Vic. 366 Facsimile: Australia simone.farrer@publish.csiro.au Published by CSIRO PUBLISHING for CSIRO Australia and the Australian Academy of Science

2 Aust. J. Bot., 1999, 47, Monophyly of Genera and Species of Characeae based on rbcl Sequences, with Special Reference to Australian and European Lychnothamnus barbatus (Characeae: Charophyceae) Richard M. McCourt AE, Kenneth G. Karol B, Michelle T. Casanova C and Monique Feist D ADepartment of Botany, Academy of Natural Sciences, Philadelphia, PA 1913, USA. BLaboratory of Molecular Systematics, National Museum of Natural History, MSC, MRC-534, Smithsonian Institution, Washington, DC 256, USA. CBotany Department, University of New England, Armidale, NSW 2351, Australia. DLaboratoire de Paléobotanique, Institut des Sciences de l Evolution, Université des Sciences, Place Bataillon, 3495 Montpellier, France. E Corresponding author. Abstract Sequences for the chloroplast-encoded large subunit of the Rubisco gene (rbcl) were used to test the monophyly of multiple isolates within species, and multiple species within genera, of green algae in the Characeae (Class Charophyceae). Parsimony and maximum likelihood analyses supported the monophyly of genera and most species, with the exception of a paraphyletic assemblage comprising isolates of two species, dioecious Chara connivens Salzm. ex A.Br. and monoecious C. globularis Thuill., which together constitute a monophyletic group. The rbcl data support the independent evolution of either monoecious or dioecious sexual systems in the two connivens-globularis, clades. Comparisons of disjunct isolates of the monotypic Lychnothamnus barbatus (Meyen) Leohn. revealed nearly identical rbcl sequences in isolates from Croatia, Germany and Australia, although all three sequences were unique. The variation exhibited by these isolates was similar to variation between isolates within species of Chara and Lamprothamnium from different continents. The limited variation may be due to dispersal of thalli or oospores between continents; however, the rarity of known intercontinental transfers of Characeae in the last two centuries suggests that the Australian population is probably not an exotic from Europe. Lychnothamnus barbatus populations in Australia and elsewhere thus merit continued protected status. Introduction The monospecific charalean genus Lychnothamnus occurs in widely separated freshwater habitats on three continents (Wood and Imahori 1965). Although this alga sometimes forms dense stands in shallow water, L. barbatus (Meyen) Leonh. is not abundant or common throughout its range. Indeed, collecting this species is exceedingly difficult and some molecular studies have had to employ herbarium specimens as sources of DNA (McCourt et al. 1996). Sequences for rbcl (the plastid-encoded large subunit of the enzyme ribulose-1,5- bisphosphate carboxylase oxygenase [Rubisco]) (McCourt et al. 1996) and the small subunit of ribosomal DNA (18S rdna) (Meiers et al. 1999) have been published for European specimens of L. barbatus. This report presents rbcl sequences for L. barbatus from a restricted and possibly endangered population in Australia and for an additional European L. barbatus, and discusses the implications of these molecular data for phylogenetic relationships of the three populations. By analysing the sequences along with new rbcl sequences from other genera and conspecific isolates we also test the monophyly of genera and selected species in the Characeae. Lychnothamnus barbatus has been collected from sites in Europe, India, China and Australia (Jao 1947; Pal et al. 1962; Wood and Imahori 1965). The species has been recorded from two sites in south-east Queensland, Australia. Specimens were collected more than 36 years ago CSIRO /99/3361

3 362 R. M. McCourt et al. from a population in Warrill Creek by R. D. Wood in 196 (Wood 1972), and no further collections were made until specimens were collected from Warrill Creek in October 1996, and from one other site in December 1996 in the same region (Casanova 1997). The herbarium records and the specimens collected by Wood indicate that the species formed robust, perennial stands in deep pools in at least two locations in Warrill Creek in 196. The Warrill Creek population in October 1996 consisted of a few weak specimens in a single location, and was decimated 2 months later (December 1996) when the site was subjected to high, turbid flows. Lychnothamnus barbatus is rare in Australia. Under the criteria of risk of extinction (Chalson and Keith 1995) the species would be classified as critical (there are fewer than 25 mature individuals, the geographic range less than 4 km, and there is a pattern of decline). The locality and catchments of habitats of L. barbatus in Australia are heavily utilised for cattle grazing and irrigated horticulture (Thomas 1973). These processes represent threats to the alga s survival in Australia. The species has been listed under the Australian Endangered Species Protection Act (1992). The act provides protection for native Australian species, and determination of whether L. barbatus is native to Australia, or is an exotic introduced from Europe, is important in deciding whether the species should remain protected under that act. Analysis of the genetic similarity of European and Australian L. barbatus may help determine whether the specimens collected in Australia are actually derived from European L. barbatus. Given the rarity of specimens, extensive population genetic analysis was not feasible for this study. Instead we chose to obtain sequence data from the rbcl gene and compare specimens from three populations: one in Croatia, one in Germany and one in Australia. Differences between these specimens from three populations of L. barbatus were compared to differences among multiple isolates of species in other characean genera (Chara, Nitella, Tolypella and Lamprothamnium). This taxon sample expands on the data set of McCourt et al. (1996), who used a representative species of each genus and two of Tolypella. The sequence data thus provide a test of the monophyly of some species and genera, as well as a test of the original hypothesis of relationships proposed for genera and tribes (Wood and Imahori 1965). The rbcl gene was chosen because of the availability of sequence data from other extant Characeae, in particular for multiple representatives of several widespread species of Chara, and the gene s rate of change, which suggested that intraspecific differences might be present but limited in number. Materials and Methods Species sampled are included in Table 1. Nomenclature follows the Microspecies Index in Wood and Imahori (1965, pp ). Species identifications were based on morphological observations, not on tests of potential interbreeding; the latter sometimes falsifies morphospecies identifications (Proctor 197), and crossing experiments were beyond the scope of this study. Species chosen were widely distributed so that comparisons of sequence differences between disparate populations could be compared. New rbcl sequences for 18 species in five genera of Characeae are provided. Outgroup taxa were chosen from other charophycean green algae and land plants, which are well supported as members of the same clade containing the Characeae (Graham et al. 1991; Mishler et al. 1994; McCourt 1995). Thalli were cleaned of possibly contaminating epiphytes under a dissection microscope prior to DNA extraction. Methods for DNA extraction and purification, rbcl amplification, and direct sequencing of the double-stranded amplification product followed the techniques described in McCourt et al. (1996). Sequences were determined visually from autoradiographs and entered directly into a Nexus file (see description of Paup* below). Sequences were proofread against the original autoradiographs. Primer sequences for amplification and sequencing were those used in McCourt et al. (1996) and are available upon request. None of the sequences contained indels or introns, consistent with prior studies of rbcl in the Characeae (Manhart 1994; McCourt et al. 1996), and sequences were easily aligned. Sequences were usually 1354 nt in length, although a 134-nt fragment of the sequence for Nitella praelonga P/CR7 was not determined. Phylogenetic analysis was performed by using the test version of Paup* (4.d64PPC version) with permission of the author D. Swofford (Smithsonian Institution, Washington, DC). Phylogenetic relationships

4 rbcl Phylogeny of Genera and Species in Characeae 363 Table 1. Species used in the analysis of rbcl sequences Nomenclature follows the Microspecies Index of Wood and Imahori (1965) Species Isolate Collection Genbank source accession Chara connivens JRM AB Manhart (1994) X-774 B Carmona, Spain AF9716 F14 C NE Spain AF97161 X-214 B Israel AF97162 C. globularis F124C C Eastern Germany AF97163 X-692 B Tas., Australia AF97164 X-751 B Alberta, Canada AF97165 C. vulgaris X-152 B Denmark AF97166 F11 C Southern France AF97167 C. rusbyana X-66 B Argentina AF97168 LG D Unknown AF97169 Lamprothamnium papulosum F137 C Southern France AF9717 X-695 B Tas., Australia U27534 Nitellopsis obtusa F131B C Western Germany U2753 Lychnothamnus barbatus Croa E Croatia U27533 Aus F Australia AF97171 Ger C Berlin, Germany AF97172 Nitella praelonga P/CR7 B Costa Rica AF97173 N. translucens F18 C Western France AF97745 Nitella sp. JRM AB Manhart (1994) L13482 N. opaca F146 C Western Poland AF97174 Tolypella prolifera F142 C Netherlands U27532 F15 C Maine et Loire, France AF97175 T. nidifica F138 C Southern France U27531 T. glomerata F131A C Western Germany AF97176 Coleochaete orbicularis JRM A Manhart (1994) Marchantia polymorpha JRM A Manhart (1994) Oryza sativa JRM A Manhart (1994) Pseudotsuga menziesii JRM A Manhart (1994) Ophioglossum engelmanii JRM A Manhart (1994) Psilotum nudum JRM A Manhart (1994) Spriogyra maxima JRM A Manhart (1994) A Sequence from Manhart (1994). B Thalli provided by Dr V. Proctor, Texas Tech University, Lubbock, TX, USA. C Thalli provided by M. Feist, Université de Montpellier, Montpellier, France. D Thalli provided by L. Graham, University of Wisconsin, Madison, WI, USA. This dioecious isolate has also been referred to as C. zeylanica (L. Graham, pers. comm.), but V. Proctor has identified it as C. rusbyana. E From herbarium specimen of W. Krause (see McCourt et al. 1996). F Thalli collected by M. Casanova. were inferred by maximum parsimony (MP) and maximum likelihood (ML) (Swofford et al. 1996). For parsimony analyses, heuristic searches were performed by the MULPARS, TBR and steepest descent options. All sites were weighted equally and character states were unordered. The input order of taxa was shuffled randomly in replications. Bootstrap re-sampling methods (Felsenstein 1985) were used ( replications) in parsimony analyses to assess robustness of inferred clades. To choose the appropriate ML model, likelihood scores under the various models in Paup* were determined for the

5 364 R. M. McCourt et al. trees resulting from the MP search. As described in Swofford et al. (1996), likelihood ratios of tree scores under the various models were compared by a likelihood ratio test. This procedure yielded the choice of the general time-reversible model incorporating site-specific rate heterogeneity (GTRss). Pairwise distance comparisons were calculated by the mean character distance formula of Paup*, in which the absolute distance is divided by the total weight of the characters for which data are not missing. Results A single most parsimonious tree was found in the MP search (Fig. 1). The three C. connivens isolates at the top of Fig. 1. possessed identical sequences and their branches are collapsed in the tree shown. The ML analysis found three equally optimal trees, and the only differences were in the ordering of the three C. connivens isolates with identical sequences (JRM, X774, and F14) at the top of the tree in Fig. 1. The topology of the tree in Fig. 1 in terms of relationships of genera is identical to that for the equal-weights parsimony analysis of McCourt et al. (1996). Monophyly of the family Characeae was strongly supported, as was the topology of internal branches, with two exceptions. Bootstrap support for the sister relationship of Nitella and genera of the tribe Chareae was weak (6%). Therefore, basal relationships within the family were less clearly resolved. Strong bootstrap support (> 9%) was found for the monophyly of all the minor genera, including L. barbatus; monophyly of the more species-rich genera Nitella and Chara was less strongly supported (57 and 65%, respectively). Monophyly of isolates within species was strongly supported, with the exception of a paraphyletic group containing two well-supported clades, each with isolates of C. connivens and C. globularis. The three isolates of Lychnothamnus barbatus form a monophyletic group with % bootstrap support. Each species isolate has a unique sequence, with few interspecific differences. The Croatian Lychnothamnus exhibits one autapomorphy, while the Australian and German isolates share one synapomorphy. Sequences of the latter pair of species differ by one base change, an autapomorphy for the Australian Lychnothamnus; the German isolate exhibits no autapomorphies. Sequence variation among isolates of Lychnothamnus was comparable to that for conspecific isolates of other genera, for example multiple isolates of Chara and Lamprothamnium show two to nine base pair differences for isolates from different continents (Table 2). Pairwise distance measurements for the taxa (Table 2) show that differences among isolates within species of Chara, Lamprothamnium and Lychnothamnus (tribe Chareae) were never more than 1%, and differences between species were only slightly higher. Differences between species within Nitella and Tolypella (tribe Nitelleae) ranged from 4 to 8.6%. Discussion The phylogeny inferred from this enlarged sample of taxa is congruent with that in McCourt et al. (1996) based on rbcl sequences from a representative species from each genus and two of Tolypella. The tribe Chareae is monophyletic, while the relationships of the three basal clades (Tolypella, Nitella and tribe Chareae) are not clearly resolved. Monophyly of the genera is generally well supported, although Chara and Nitella, the two most speciesrich genera, were less strongly supported. Lamprothamnium isolates are monophyletic and constitute the sister taxon to a monophyletic Chara. The latter finding conflicts with the phylogenetic reconstruction based on sequence data from the small subunit of rdna, a nuclear-encoded gene (Meiers et al. 1997, 1999). The support for the internal branch leading to Chara and Lamprothamnium is not strong when sequences from additional species of Chara are included in the analysis (McCourt and Karol, unpubl. data), but the monophyly of both Chara and Lamprothamnium are still supported. The conflict between 18S rdna and rbcl sequence analyses regarding the status of these two genera remains a mystery. Monophyly of the species isolates was supported, except for C. connivens and C. globularis which formed a paraphyletic cluster. For these two species the sequences differences among

6 rbcl Phylogeny of Genera and Species in Characeae 365 isolates were comparable to the small differences (< 1%) found for conspecifics of other species in the Chareae. The paraphyly of several isolates of the dioecious C. connivens and the generally monoecious C. globularis is intriguing for several reasons. Wood and Imahori (1965) combined this monoecious and dioecious species pair (plus other microspecies) into one species, C. globularis Thuill., em. R.D.W. These authors classified these two microspecies as Chara. connivens JRM 3 C. connivens X C. connivens F C. globularis F124C C. globularis X C. connivens X C. globularis X692 7 C. vulgaris X152 C. vulgaris F11 12 C. rusbyana X66 C. rusbyana LG 1 12 Lamprothamnium F137 1 Lamprothamnium X Nitellopsis obtusa F131B 1 12 Lychnothamnus Croa 1 1 Lychnothamnus Aust 74 Lychnothamnus Ger Nitella praelonga Nitella sp. JRM N. translucens F18 46 N. opaca F Tolypella prolifera F142 3 T. prolifera F15 57 T. nidifica F138 T. glomerata F131A Coleochaete 66 Marchantia Oryza Pseudotsuga Ophioglossum Osmunda 36 Plenasim Psilotum Spirogyra Tribe Chareae Tribe Nitelleae Fig. 1. Single most parsimonious tree found in the MP search. Branch lengths shown above and bootstrap support ( replications) below each branch. Tribe designations Chareae and Nitelleae are based on Wood and Imahori (1965). Tree length = 1626 steps; CI =.5185 (.4815 excluding uninformative characters). The tree was obtained by using the heuristic search option of Paup* (4.d64 PPC version), TBR, MULPARS, collapse zero length branches, and steepest descent options in effect. A total of random taxon additions were conducted. The tree is based on 473 parsimony informative characters. Maximum likelihood analysis using GTRss model in Paup* found this tree and two other equally optimal topologies that differed only in the placement of the three C. connivens species sharing an identical sequence (JRM, X774 and F14).

7 366 R. M. McCourt et al. Table 2. Pairwise distances between isolates and species Nitellopsis was omitted because only one population clone was sampled. Below diagonal: total character differences. Above diagonal: sequence dissimilarity expressed as mean character differences (adjusted for missing data) Chara 1 C. connivens JRM C. connivens X C. connivens F C. connivens X C. globularis F124C C. globularis X C. globularis X C. vulgaris X C. vulgaris F C. rusbyana X C. rusbyana LG Lamprothamnium L. papulosum F L. papulosum X Lychnothamnus L. barbatus Croa L. barbatus Aus L. barbatus Ger 2 1 Nitella N. praelonga Nitella sp. JRM N. translucens F N. opaca F Tolypella T. prolifera F T. prolifera F T. nidifica F T. glomerata F131A 87 89

8 rbcl Phylogeny of Genera and Species in Characeae 367 closely related forms within the variety globularis of a much more broadly defined species of C. globularis. These authors noted that the two forms are very similar except for a minor difference in branchlet morphology. Proctor (198) disagreed strongly with this combining practice, which Wood and Imahori (1965) applied to many other monoecious dioecious species pairs, although in the case of globularis and connivens, Proctor (1975) agreed there was some slight support for combining the two on the basis of limited success in interbreeding the two species under laboratory conditions. But whereas Wood and Imahori (1965) combined these microspecies (and numerous others) as part of a broad, morphologically defined species, Proctor s (1975) breeding experiments suggested the two may constitute a single biological species separate from other taxa. Proctor (1975) postulated that C. connivens or closely related dioecious forms may have given rise independently to several monoecious lines of C. globularis via relatively minor changes in the life cycle, although he stated that there is no experimental evidence for this (Proctor 198, p. 231). The rbcl data provide just such evidence and strongly support Proctor s hypothesis. Successful interbreeding between isolates of these two species would corroborate the species boundary implied by the rbcl data. Sampling of more variable genes, such as the internally transcribed (ITS-1 and ITS-2) regions of the nuclear ribosomal RNA genes, might provide further evidence regarding the relationships of isolates within species (Meiers 1997) and the ability of separate populations to interbreed (Coleman and Mai 1997). These further studies may warrant combining of the two described species into one, or a redefinition of each species based on molecular and breeding data. It is noteworthy that sequence differences between species of tribe Nitelleae sensu Wood & Imahori (Nitella and Tolypella) were much greater than sequence differences between species in the Chareae (i.e. the latter show much shorter branch lengths). One might attribute this difference to the smaller number of Nitella and Tolypella species sampled, i.e. the branches involved are missing many taxa that would shorten the internodes. However, although additional unsampled species of Nitella and Tolypella might break up their long branches, the total number changes in Nitella and Tolypella exceeds by far the number within the Chareae as a whole. Unless there are some highly divergent and unsampled species of Chara, we hypothesise that the divergence of Chara species may represent a more recent radiation (descent from a more recent common ancestor); alternatively, Chara may have experienced a different rate of sequence evolution in this clade. Comparison of Australian and European Isolates of Lychnothamnus There are several possible explanations for the disjunct distributions of European and Australian specimens of L. barbatus, which are mirrored by that of other monoecious species (e.g. C. braunii, Lamprothamnium papulosum and Nitellopsis obtusa (Proctor 197; Grant and Proctor 1972)). One explanation for the disjunction is that L. barbatus was recently introduced by human transport into Australia from Europe. Another one is that the widely separate populations of L. barbatus are remnants of a formerly widespread species distributed throughout several continents, including Eurasia, South-east Asia and Africa (Soulié-Märsche 1981). There is no published record of fossil Lychnothamnus (nor of any fossil charophyte) in Australia, but research underway at the University of New South Wales may provide more data on the charophyte fossil record on this continent (A. Garcia, pers. comm.). This broad distribution of L. barbatus may be the result of dispersal of a monoecious species via nonhuman means (i.e. water birds (Proctor 1962; Proctor et al. 1967)). Alternatively, the disjunct distributions may represent residual populations following break-up of Gondwana some 15 million years ago near the end of the Jurassic Period (Wilford and Brown 1994). The nearly identical rbcl sequences in the three L. barbatus isolates suggest a close relationship between the populations. Evidence suggesting that L. barbatus is an exotic European species in Australia includes the morphological similarity among the Australian specimens and those illustrated from other continents (Pal et al. 1962; Wood 1972), along with

9 368 R. M. McCourt et al. the colonisation of the Fassifern region by German immigrants in the 186s (McHugh 1978). However, several lines of evidence argue against the hypothesis that the Australian population is a product of recent European invasion. First, the specimens from the different continents are not morphologically identical (they differ in the length of stipulodes and bract cells, the number of branchlets in a whorl and the degree of cortication (Garcia and Casanova, unpubl. data)). Second, the environments in which the populations occur on both continents are different (Europe has cold, deep, clear lakes; south-east Queensland has subtropical, warm, ephemeral streams). Third, disjunct distributions such as that of L. barbatus are well documented among aquatic plants in general (Sculthorpe 1967) and especially among charophytes (Proctor 197). Finally, circumstantial evidence suggests that the likelihood of transporting viable charophyte thalli or oospores between continents is low. Vernon Proctor of Texas Tech University (pers. comm.) has noted that there is evidence of but one probable introduction of a European species of Chara into North America. Furthermore, no European Lychnothamnus has been known to establish a viable population in North America. Further sampling of L. barbatus and other Characeae from Asia and South-east Asia would test hypotheses of dispersal between areas where this species has historically occurred. If not an exotic, what is the explanation for the current distribution of L. barbatus? The possibility exists that L. barbatus is capable of intercontinental dispersal. This species and Lamprothamnium papulosum and Chara muelleri exhibit similar broad distributions (Wood and Imahori 1965; V. Proctor, pers. comm.). All three species are monoecious and thus more likely to be capable of successful dispersal to new areas (Proctor 198). Gondwanan L. barbatus may have dispersed to Europe via Africa after the continental break-up (Wilford and Brown 1994). Australian L. barbatus would then be a descendant of a broadly distributed species in Gondwana, where lacustrine and fluvial habitats were common in the region that would become Queensland. In any case, the populations of L. barbatus in Australia and elsewhere represent natural populations worthy of conservation efforts. In conclusion, rbcl sequence data suggest that the three L. barbatus isolates are genetically distinct, and the sequence divergence levels are comparable to those for isolates of species of Chara and other genera sampled from different continents. Further studies with more variable genes (e.g. ITS-1 and ITS-2) are planned. The data are consistent with what one would expect from a native Australian population of a cosmopolitan species of the Characeae. The responsible action, given the declining status of L. barbatus worldwide, is to assume that the species is native to Australia and to continue to list it under the Endangered Species legislation. Acknowledgments The authors thank V. Proctor (Texas Tech University, Lubbock, Texas) for providing samples of Characeae from his culture collection, for many stimulating discussions on the evolution of these algae, and for a critical review of this manuscript. The authors thank Linda Graham (University of Wisconsin, Madison) for the sample of C. rusbyana. Dave Swofford of the Laboratory of Molecular Systematics, Smithsonian Institution, generously allowed us permission to use Paup* for our analyses. The research was supported by NSF grant DEB to RMM, the Sloan Foundation, and the University Research Council of DePaul University, Chicago. References Australian Endangered Species Protection Act (1992). Commonwealth of Australia (Australian Government Publishing Service: Canberra.) Casanova, M. T. (1997). Report on instream flora survey for Department of Natural Resources, Queensland. University of New England, Armidale. Chalson, J. M., and Keith, D. A. (1995). A risk assessment scheme for vascular plants: pilot application to the flora of New South Wales, Project No. 45, Commonwealth Endangered Species Program. Australian Nature Conservation Agency, Canberra.

10 rbcl Phylogeny of Genera and Species in Characeae 369 Coleman, A. W., and Mai, J. C. (1997). Ribosomal DNA ITS-1 and ITS-2 sequence comparisons as a tool for predicting genetic relatedness. Journal of Molecular Evolution 45, Felsenstein, J. (1985). Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39, Graham, L. E., Delwiche, C. F., and Mishler, B. D. (1991). Phylogenetic connections between the green algae and the bryophytes. Advances in Bryology 4, Grant, M. C., and Proctor, V. W. (1972). Chara vulgaris and C. contraria: patterns of reproductive isolation for two cosmopolitan species complexes. Evolution 26, Jao, C. C. (1947). Some Characeae from Kunming, Yunnan. Botanical Bulletin Academica Sinica 1, 44. Manhart, J. (1994). Phylogeny of green plants based on rbcl sequences. Molecular Phylogenetics and Evolution 3, McCourt, R. M. (1995). Green algal phylogeny. Trends in Ecology and Evolution 1, McCourt R. M., Karol K. G., Guerlisquine M., and Feist, M. (1996). Phylogeny of extant genera in the Family Characeae (Division Charophyta) based on rbcl sequences and morphology. American Journal of Botany 83, McHugh, R. (1978). Boonah State School Boonah State School Parents and Citizens Association, Boonah, Queensland. Meiers, S. T. (1997). Phylogeny of the Characeae (Charophyta) and biogeography of Chara (Characeae) inferred from ribosomal genes. Ph.D. Dissertation. Louisiana State University, Baton Rouge, LA, USA. Meiers, S. T., Rootes, W. L., Proctor, V. W., and Chapman, R. L. (1997). Phylogeny of the Characeae (Charophyta) inferred from organismal and molecular characters. Archiv für Protistenkunde 148, Meiers, S. T., Proctor, V. W., and Chapman, R. L. (1999). Phylogeny and biogeography of Chara (Charophyta) inferred from 18S rdna sequences. Australian Journal of Botany 47, Mishler, B. D., Lewis, L. A., Buchheim, M. A., Renzaglia, K. S., Garbary, D. J., Delwiche, C. F., Zechman, F. W., Kantz, T. S., and Chapman, R. L. (1994). Phylogenetic relationships of the green algae and bryophytes. Annals of the Missouri Botanical Garden 81, Pal, B. P., Kundu, B. C., Sundaralingham, V. S., and Venkataraman, G. S. (1962). Charophyta. (Indian Council of Agricultural Research: New Delhi.) Proctor, V. W. (1962). Viability of Chara oospores taken from migrating water birds. Ecology 43, Proctor, V. W. (197). Taxonomy of Chara braunii: an experimental approach. Journal of Phycology 4, Proctor, V. W. (1975). The nature of charophyte species. Phycologia 14, Proctor, V. W. (198). Historical biogeography of Chara (Charophyta): an appraisal of the Braun Wood classification plus a falsifiable alternative for future consideration. Journal of Phycology 16, Proctor, V. W., Malone, C. R., and de Vlaming, V. L. (1967). Dispersal of aquatic organisms: viability of disseminules recovered from the intestinal tract of captive killdeer. Ecology 48, Sculthorpe, C. D. (1967). The Biology of Aquatic Vascular Plants. (Edward Arnold: London.) Soulié-Märsche, I. (1981). Charophytes. In Holocene Lake Deposits and Paleoenvironments in Central Sahara, Northeastern Mali. (Eds N. Petit-Maire and J. Riser.) Palaeogeography, Palaeoclimatology, Palaeoecology 35, Swofford, D. L., Olsen, G. J., Waddell, P. J., and Hillis, D. M. (1996). Phylogenetic inference. In Molecular Systematics. 2nd Edn. (Eds D. M. Hillis, C. Moritz and B. K. Mable.) pp (Sinauer: Sunderland, Massachusetts.) Thomas, T. C. (1973). Boonah Shire Handbook. (Queensland Department of Primary Industries: Brisbane.) Wilford, G. E., and Brown, P. J. (1994). Maps of late Mesozoic Cenozoic Gondwana break-up: some palaeogeographical implications. In History of the Australian Vegetation: Cretaceous to Recent. (Ed. R. S. Hill.) pp (University Press: Cambridge.) Wood, R. D. (1972). Characeae of Australia. (J. Cramer: Lehre.) Wood, R. D., and Imahori, K. (1965). A revision of the Characeae. First Part: Monograph of the Characeae. (Verlag von J. Cramer: Weinheim.) Manuscript received 23 September 1997, accepted 26 November