| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
2Landcare Research, P. O. Box 69, Lincoln 8152, New Zealand; and 3Section of Evolution and Ecology, University of California, Davis, California 95616
Received for publication July 21, 1998. Accepted for publication January 21, 1999.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Key Words: Australia • Carmichaelia • Clianthus • Fabaceae • Montigena • New Zealand • phylogeny • Swainsona.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
). Carmichaelieae comprise five genera and 
30 species of morphologically distinct trees, shrubs, and lianas that are largely endemic to New Zealand, whereas Galegeae comprise some 20 genera and over 3000 species (Hutchinson, 1964;
Polhill, 1981
). One genus, Astragalus, contains more than half of the species diversity in this temperate radiation (Sanderson and Wojciechowski, 1996
).
Carmichaelieae are related to tribe Galegeae through the monotypic New Zealand genus Clianthus and the large Australian genus Swainsona (Polhill, 1981
). Clianthus formerly included two species, but Thompson (1990)
removed the Australian Clianthus formosa transferring it to Swainsona. Swainsona is a large genus of 84 species in Australia that also included a single New Zealand species, Swainsona novae-zelandiae (Allan, 1961
). Swainsona novae-zelandiae was recently placed in the monotypic genus Montigena, based on substantial differences in growth habit and anatomy between it and other species of Swainsona (Heenan, 1998b
). In a recent phylogenetic analysis of morphological and anatomical characters, Montigena forms a monophyletic group along with Clianthus and Carmichaelia (Heenan, 1998a
).
Traditionally, tribe Carmichaelieae included Carmichaelia, Chordospartium, Corallospartium, Notospartium, and Streblorrhiza (Hutchinson, 1964;
Polhill, 1981
). However, the phylogenetic study of Heenan (1998a)
supports an expanded circumscription of the genus Carmichaelia that includes the segregate genera Chordospartium, Corallospartium, and Notospartium (Heenan, 1998c
). There are presently 23 endemic species of Carmichaelia in New Zealand and a single species on Lord Howe Island (Heenan, 1998c
; Green, 1994
). The eastern portion of the South Island of New Zealand is a center of diversity for the group with 15 species endemic to this region (Heenan, 1995
, 1996). Most species occur in drought- or frost-prone areas, predominantly in alluvial habitats. Few species of Carmichaelia actually occur on the mountains.
The disparity in species diversity among the genera of Carmichaelieae and Galegeae may partly reflect their taxonomic history. Early treatments were plagued by inadequate collections and limited understanding of intraspecific variation. One result has been nomenclatural ambiguity and instability. In an effort to develop a classification scheme that reflects the phylogeny of the group, the present study builds upon the previous studies of Lavin, Doyle, and Palmer (1990)
, Liston and Wheeler (1994)
, Liston (1995)
, Sanderson and Liston (1995)
, Sanderson and Wojciechowski (1996)
, and Heenan (1998a)
. We identify well-supported clades and provide these with taxon names, and these are defined by reference to a common ancestor and all of its descendants (DeQueiroz and Gauthier, 1992, 1994; Cantino, Olmstead, and Wagstaff, 1997
). We use the inferred phylogeny to explore the origins and patterns of diversification in the New Zealand endemics, and the following questions are posed: (1) Are the phylogenetic relationships presented here congruent with those inferred by Heenan (1998a)
? Are the inferred relationships consistent with the traditional tribal classification of Polhill (1981)
? (2) What are the origins of the New Zealand endemics? What are their closest relatives? Are the modes of speciation and patterns of diversification in New Zealand taxa similar to those in Australian Swainsona? Does the level of sequence divergence differ between Australian and New Zealand taxa? (3) What factors account for the disparity in species richness between Carmichaelia and Montigena novae-zelandiae? Do morphological novelties influence speciation or extinction rates? Are there differences in the time of establishment in New Zealand?
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
, Astragalus alpinus, A. vogelii, A. sinicus, A. complanatus, Colutea istria, Eremosparton flaccidum, Lessertia brachystachya, Sphaerophysa salsula, and Sutherlandia frutescens were selected as outgroups, and the resulting phylogenetic trees were rooted along the long branch leading to A. alpinus. The complete data set is available upon request from the corresponding author.
|
. The two internal transcribed spacers (ITS-1, ITS-2) and the 5.8S rDNA gene were amplified by the polymerase chain reaction (PCR) using the following temperature profile (97°C for 1 min, 1 cycle; linked to 94°C for 1 min; 48°C for 45 sec; 72°C 1 min, repeat 30 cycles; linked to 72°C for 5 min, 1 cycle; hold at 5°C). Primer sequences and our amplification and sequencing strategy follow Wagstaff and Garnock-Jones (1998)
. Excess primers and unincorporated nucleotides were removed from the PCR products by spin column centrifugation (QIAquick PCR purification kit, QIAGEN Inc., 28159 Avenue Stanford, Valencia, California 91355). The purified DNA templates were labelled with dRhodamine dye terminator by cycle sequencing with the following temperature profile (96°C for 30 sec; 50°C for 15 sec; 60°C for 4 min, repeat 25 cycles; hold at 5°C). Excess dye terminators were removed by alcohol precipitation in the presence of 3.0 M sodium acetate pH 5.2. The fluorescent-labelled DNA samples then were sent to the Waikato DNA sequencing facility. Both the forward and reverse DNA strands were sequenced to insure the reproducibility of our results.
Data analysis
The DNA sequences were manually aligned and then analyzed using PAUP* version 4.0 (Swofford, 1995
) and MacClade version 3.06 (Maddison and Maddison, 1996
). Parsimony and maximum likelihood were used as optimality criteria to evaluate trees and to assess the robustness of our phylogenetic hypothesis to differences in the assumptions of these two approaches. Parsimony analysis selects the hypothesis of relationships that minimize the number of evolutionary steps (tree length) required to explain the data, whereas maximum likelihood selects the hypothesis that maximizes the probability (-log likelihood) of observing the data given a model of evolution (Swofford et al., 1996
).
The large number of taxa included in our survey made a heuristic search strategy necessary for the parsimony and maximum likelihood analyses. The FILTER TAXA option in MacClade was used to identify redundant taxa. Taxa were considered redundant even if their states were not identical, as long as a resolution of missing or ambiguous data could make them identical. The parsimony analysis was conducted using RANDOM addition sequence with 100 replicates, TBR branch swapping, MULPARS in effect, STEEPEST DESCENT, with all character changes weighted equally. Gaps were treated as missing data. The most parsimonious trees were evaluated under the Hasegawa-Kishino-Yano (HKY-85) model (Hasegawa, Kishino, and Yano, 1985
) to estimate parameters for maximum likelihood analysis (Swofford et al., 1996
). Relative rate (Wu and Li, 1985
; Sanderson, 1997
) and global likelihood ratio tests (Felsenstein, 1988
) were conducted to assess rate constancy.
Support for the inferred clades is given by the number of synapomorphies as indicated by branch length and by bootstrap values (Felsenstein, 1985
). We conducted 100 bootstrap replications excluding uninformative sites with one RANDOM ADDITION sequence for each bootstrap replication, TBR branch swapping, and MULPARS in effect. The criteria suggested by Cantino, Olmstead, and Wagstaff (1997)
were used to define well-supported clades. These are clades supported by 50% bootstrap values that are also supported by nonmolecular evidence, clades supported by 70% bootstrap values that are neither supported nor in confict with nonmolecular evidence, or clades that are supported by >90% bootstrap. The TOPOLOGICAL CONSTRAINT option in PAUP* (Swofford, 1995
) was used to assess the degree of conflict with previous phylogenetic studies.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
The ITS sequences are characterized by unequal nucleotide frequencies, a transition/transversion bias, and rate heterogeneity across sites. The distribution of variable sites in the ITS region is shown in Fig. 1. The proportion of nucleotide bases exhibit a G/C bias; the empirical frequencies are A = 0.21126, C = 0.26048, G = 0.28528, and T = 0.24298. A transition/transversion ratio of 1.320148 was estimated via maximum likelihood using the HKY-85 model (Hasegawa, Kishino, and Yano, 1985
) with the proportion of sites assumed to be invariable = 0.155353, and the rates for variable sites assumed to follow a gamma distribution with shape parameter = 0.813092.
|
|
|
|
|
The global likelihood ratio test does not support rate constancy (Fig. 4), and many of the relative rate lineage comparisons also reject constancy (Table 4), sometimes in strong terms. This is not surprising considering variability in branch lengths (Figs. 3, 4). The relative rate tests appear to be useful to identify particular lineages that are exceptionally distinct in their rates of molecular evolution. Thus, within much of the New Zealand clade there is rate constancy, but this perhaps has something to do with the low variability of Carmichaelia and, hence, the very low power of the test because of the few character differences between taxa.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
). We feel that information content, clarity, and stability are three key features of a useful classification, and these are best accomplished within a phylogenetic context. The inferred phylogeny presented here has implications for classification and enhancing our understanding of the origins and patterns of diversification of New Zealand legumes.
Classification
The results presented here along with the earlier results of Sanderson and Wojciechowski (1996)
indicate that the traditional tribal classification of Galegeae and Carmichaelieae proposed by Polhill (1981)
does not accurately convey evolutionary relationships or levels of species diversity. Tribe Galegeae is clearly polyphyletic. Glycyrrhiza (subtribe Glycyrrhizinae) and Galega (subtribe Galeginae), the type for the tribe, are distinct from other members of tribe Galegae. Glycyrrhiza forms a distinct lineage at the base of a large temperate clade of legumes. Galega is nested in a clade that includes members of tribes Hedysareae and Vicieae (Sanderson and Wojciechowski, 1996
). Continued recognition of Galega officinalis as the type would require the recognition of a more inclusive Galegeae. Subtribes Astragalinae and Coluteinae are at best paraphyletic, and tribe Carmichaelieae is nested within subtribe Coluteinae (Sanderson and Wojciechowski, 1996
).
Even within the well-supported New Zealand and Australian clade there are problems of poor resolution and conflict with previous studies. The conflict centers chiefly on the paraphyly or monophyly of Swainsona. Our results suggest that Swainsona is paraphyletic by exclusion of the New Zealand taxa Montigena, Clianthus, and Carmichaelia. The New Zealand taxa appear in two distinct lineages (Fig. 2), whereas they form a well-supported clade in Heenan's (1998a)
study. When a topological constraint was enforced that placed Montigena in a clade with Carmichaelia and Clianthus, we recovered trees of 357 steps, five steps longer than the maximum parsimony trees of 352 steps that place Montigena among Australian Swainsona. When monophyly of Swainsona including Montigena was enforced, we recover trees of 353 steps, one step longer than the maximum parsimony trees. These results are in agreement with Thompson (1993)
who included Montigena novae-zelandiae in group 9 along with S. lessertiifolia.
Little doubt remains that the New Zealand and Australian taxa together form a clade (Figs. 2–4), and it is one of sufficient distinctiveness and intrinsic interest to warrant a name being attached. However, if we are to remain wedded to the Linnaean system, we are stuck with choosing the rank of that taxon. If this clade remains Carmichaelieae, then what of the remaining dozen or so genera in the Astragalean clade, which is equally well supported and has Carmichaelieae nested within it? We could not recognize it as a tribe—it is now paraphyletic. Nor is it reasonable to recognize several subclades within it as tribes—they are not well resolved or especially distinct from each other. This is also a problem even if we keep Carmichaelieae as a tribe with more or less its current circumscription. This is a classic case in which the utility of the Linnaean hierarchy is called into question.
Perhaps an alternative to a Linnaean classification is a rank-free phylogenetic classification. Such a classification offers nomenclatural clarity in the sense that taxon names are explicitly associated with clades by definition (Cantino, Olmstead, and Wagstaff, 1997
). Under a phylogenetic system a single name is applied to a clade; thus there is no redundancy. The recognition of monotypic taxa in a traditional classification identifies taxa that are identical in context to a single taxon at the next lower rank (e.g., subtribe Galeginae and the monotypic genus Galega refer to the same clade. As our knowledge of relationships improves, there is a corresponding need to recognize additional ranks to reflect phylogeny (Cantino, Olmstead, and Wagstaff, 1997
).
Until a consensus is reached regarding the utility of phylogenetic classification, we recommend that the circumscription of Pollhill's (1981) tribe Carmichaelieae should be expanded to include Clianthus, Montigena, and Swainsona (formerly tribe Galegeae). However, it should be recognized at a lower rank. We suggest the name Carmichaelinae, which denotes a subtribe in a Linnaean classification. This name also corresponds to the least inclusive clade that contains Carmichaelia, Clianthus, Montigena, and Swainsona.
Synapomorphies of this clade are style brush (Polhill, 1981
), polyploid chromosome number 2n = 32 (Goldblatt, 1981
), and molecular characters (Liston and Wheeler, 1994
; Sanderson and Liston, 1995
; Sanderson and Wojciechowski, 1996
). We would conserve Carmichaelia as the name of the type, a name that is widely recognized. Unfortunately this proposed change only resolves some of the potential problems of nomenclature within the Astragalean clade. We acknowledge that many taxa cannot be unambiguously placed within a tribal and subtribal classification scheme because of the lack of resolution so far obtained. In addition, large genera such as Astragalus and Oxytropis are placed in monotypic subtribes, leaving a problem with redundant names.
Origins and patterns of diversification in New Zealand
Because of its unique combination of geographic and temporal isolation, level of habitat diversity, climate, and geologic history, New Zealand hosts a diverse and largely endemic flora (Wardle, 1991
). It includes a disproportionately high number of large diverse genera such as Carmichaelia, as well as morphologically isolated genera that include only a single species like Clianthus and Montigena (Fenner and Lee, 1997
). Although high habitat diversity provides an appropriate theater for speciation to occur, some genera may possess or acquire traits that confer an ability to undergo rapid evolutionary change. Some radiations may have resulted from the appearance of new ecological opportunities as well as the acquisition of traits that produce a propensity to speciate (Fenner and Lee, 1997
). The disparity in species richness between Carmichaelia and Montigena could result from differences in the time of establishment, differential extinction, or the evolution of some novel adaptation that has facilitated speciation.
Analysis of ITS sequences support a northern hemisphere origin for Carmichaelinae independent from the South African endemics Sutherlandia and Lessertia (Figs. 2–4). The basal lineage of Carmichaelinae forms a star phylogeny. Four lineages are well supported by the sequence data, but relationships among them are unclear. We postulate an initial radiation that accompanied increased aridity in Australia and New Zealand during the Tertiary. Though fossils are not abundant, they first appear during the mid-Tertiary. Representatives of the Astragalaean clade were widely distributed in temperate regions by the Oligocene (Axelrod, 1992
), and Carmichaelia first appears in New Zealand in late Pliocene Waipaoa Series (Oliver, 1928
). However, the lack of rate constancy among the ITS sequences makes precise inferences about divergence times in Carmichaelinae problematic (Table 4; Fig. 4).
The results support two distinct New Zealand lineages that have independent origins among Australian Swainsona (Fig. 3). The first lineage includes the monotypic genus Clianthus, 23 species of Carmichaelia in New Zealand, and C. exsul, which must have recently dispersed from New Zealand to Lord Howe Island. Lord Howe is a volcanic island 6 million years old (McDougall, Embleton, and Stone, 1981
). The second lineage, Montigena novae-zelandiae, is currently recognized as a monotypic genus (Heenan, 1998b
). All members of Carmichaelinae are colonizers that invade disturbed habitats on shallow, poorly developed soils. They form symbiotic nitrogen-fixing relationships with the root-nodulating Rhizobium. The New Zealand strains of Rhizobium are acid-producing in contrast to the many Australian woody legumes that grow in acid heathland and form a symbiosis with alkali-producing strains (Greenwood, 1978
). This characteristic would most certainly enhance the establishment of founder populations in New Zealand.
Clianthus puniceus emerges in an unresolved polytomy at the base of Carmichaelia (Figs. 2–4). Along with Carmichaelia stevensonii and C. muritai, it exhibits morphological characteristics that are ancestral within Carmichaelinae. Clianthus differs from Carmichaelia in being a leafy evergreen shrub. Its large red flowers are specialized for bird pollination and have a large standard and keel petals, shorter wing petals, and drooping inflorescences. The fruit is a follicle (Godley, 1979
). Some of the anatomical features of Clianthus are intermediate between Swainsona and Carmichaelia [e.g., vessels with and without helical thickenings (Heenan, 1997b
)]. We suggest that Clianthus evolved in the temperate and equable climatic conditions present during the Miocene; it lacks the specialized xeromorphic features of Carmichaelia. Currently it is found in nonforested rocky habitats in the North Island that are exposed to direct sunlight (Shaw and Burns, 1997
). Historical records indicate that Clianthus was never a common plant, and the range of the species has contracted substantially from that recorded during the 19th century. Humans have played a significant role in its recent decline through seed and plant collection, habitat destruction, and the introduction of herbivores or adventives. It is now exceedingly rare and vulnerable in the wild (Shaw and Burns, 1997
).
The Marlborough endemics Carmichaelia stevensonii and C. muritai share a basal position along with Clianthus (Figs. 2–4). They exhibit several characters associated with drought or cold tolerance. The seedling and juvenile forms are usually leafy, but they are predominantly leafless at maturity, the stems are photosynthetic, and their wood exhibits well-developed xeromorphic features (Heenan, 1997a, b
). Their evolution may correspond with the onset of the Kaikoura Orogeny and the formation of the Marlborough Fault System during the early Miocene through to the late Miocene (Suggate, Stevens, and Te Punga, 1978
; Browne, 1992
). The climate of eastern Marlborough progressively became more dry at this time and other temperate xerophytes arose (Mildenhall and Pocknall, 1984
).
The rapid adaptive radiation and high level of endemism in the eastern South Island occurred relatively recently in the evolutionary history of Carmichaelia (Figs. 2–4). We suggest that it accompanied a later phase of the Kaikoura Orogeny and episodes of glaciation. By the late Pliocene, the Kaikoura Orogeny had substantially uplifted the Southern Alps (Suggate, Stevens, and Te Punga, 1978
; Adams, 1979
), cooler climatic conditions were present (Kennett, Watkins, and Vella, 1971
; (Kemp, 1978
), and subarid areas expanded in the interior of the South Island (Raven, 1973
). This would have been a time of severe disturbance; glaciation and erosion created a variety of new habitats such as extensive scree slopes, outwash plains, terrace complexes, alluvial plains, and glacial moraine.
The leafless shrubby habit with photosynthetic shoots along with specialized xeromorphic anatomical features that are characteristic of Carmichaelia stevensonii and C. muritai were preadaptive and may have triggered the burst of evolution observed in the bulk of Carmichaelia. Carmichaelia are mostly long-lived woody trees, shrubs, or lianas that are specialized to tolerate drought or cold. They are generally leafless at maturity and often have flattened stems with terminal innovation shoots and persistent pith. Photosynthesis occurs mainly in the stems, which range from slightly flattened to strap-like and >1 cm wide in C. williamsii. Their vessels are arranged in diagonal aggregations and have helical thickenings (Heenan, 1997b
, 1998a). In contrast, the Australian species of Swainsona are mostly annuals or short-lived perennials that avoid seasonally dry climates by dying back to ground. They are leafy, have basal innovation shoots, and lack persistent pith. The narrow ray parenchyma and the vessel elements are usually solitary or in small groups and lack helical thickenings.
The level of genetic divergence among the New Zealand taxa does not appear to be correlated with the level of phenotypic diversity. A similar pattern is observed in other island floras. While Clianthus and Montigena are genetically distinct entities, there is very little sequence divergence in Carmichaelia (Table 3) despite the morphological diversity among these species. It appears that reproductive isolation in Carmichaelia is primarily the result of geographical isolation and ecological specialization. For example, C. williamsii occurs in open Metrosideros excelsa forests on the northern offshore islands and has large yellow flowers adapted to bird pollination; the dwarf C. nana occurs on recent alluvial deposits, whereas other dwarf species such as C. monroi occur in grassland or on rock outcrops; C. kirkii is a sprawling liana known from scattered populations throughout the eastern and central areas of the South Island; and C. arborea is a large shrub that is most common in high rainfall areas along and west of the main divide in New Zealand (Heenan, 1995, 1996
).
Clianthus puniceus and Montigena novae-zelandiae apparently have strong internal barriers that prevent hybridization, whereas hybrids in several combinations among species of Carmichaelia are known from the wild and have been created in a test garden, including taxa formerly placed in separate genera (Heenan, 1998c
). The pollen stainability and seed set of the F1 garden hybrids are high and F2 progeny have been produced, which indicate high levels of compatibility and fecundity among the various crosses. Hybridization may have led to the formation of several new species (Heenan, 1998b
). The new gene combinations that result yield a myriad of diverse, perplexing phenotypes. In contrast, species of Astragalus are characterized by allopatry and local endemism with little hybridization (Sanderson, 1991
).
Montigena novae-zelandiae is the only New Zealand legume restricted to scree (Heenan, 1998b
). Its succulent fleshy summer-green leaves that are arranged in loose rosettes or cushions, superficial cells with reddish pigments, blue-grey waxy bloom, and shallow well-developed underground rhizome with strong root development are characteristic of plants that inhabit screes. Unlike Clianthus, Montigena may be a recent arrival. The extensive scree slopes that it inhabits eroded after the uplift of the Southern Alps. Its restricted distribution and lack of radiation on to similar habitats, such as glacial moraine and river gravel alluvium, also suggest a recent radiation following dispersal from Australia. Its sister species, Swainsona galegifolia, is widely distributed in coastal areas of eastern Australia and on the dividing range from tropical Queensland to northern Victoria (Thompson, 1993
). We suggest that Montigena novae-zelandiae has been evolutionarily constrained by the inhospitable environment in which it is found; it is a specialist, and there has been selection against the evolution of new species.
While analysis of ITS sequences provides strong support for monophyly of Carmichaelinae, more work is required to assess its relationships to other members of the temperate herbaceous clade. Our molecular studies have identified several well-supported clades that are difficult to characterize with morphological characters. We feel further molecular studies of Carmichaelinae would benefit from an increased sample of Swainsona, perhaps employing a more rapidly evolving DNA sequence such as the ETS (external transcribed spacer) region of nuclear DNA. This could lead to a more highly resolved phylogeny to assess patterns of diversification in New Zealand and Australia.
|
FOOTNOTES |
|---|
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Allan, H. H. 1961 Flora of New Zealand, vol. I, Government Printer, Wellington, New Zealand.
Axelrod, D. I. 1992 In P. S. Herendeen and D. L. Dilcher [eds.], Advances in legume systematics, part 4, 259–279. Royal Botanical Garden, Kew.
Browne, G. H. 1992 The northeastern portion of the Clarence Fault: tectonic implications for the late Neogene evolution of Marlborough, New Zealand. New Zealand Journal of Geology and Geophysics 35: 437–445.
Cantino, P. D., R. G. Olmstead, and S. J. Wagstaff. 1997 A comparison of phylogenetic nomenclature with the current system: a botanical case study. Systematic Biology 46: 313–331.[CrossRef][ISI]
De Queiroz, K., and J. Gauthier. 1992 Phylogenetic taxonomy. Annual Review of Ecology and Systematics 23: 449–480.[ISI]
———, and ———. 1994 Toward a phylogenetic system of biological nomenclature. Trends in Ecology and Evolution 9: 27–31.
Doyle, J. J., and J. L. Doyle. 1987 A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochemistry Bulletin 19: 11–15.
Felsenstein, J. 1985 Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39: 783–791.[CrossRef][ISI]
———. 1988 Phylogenies from molecular sequences: inferences and reliability. Annual Review of Genetics 22: 521–565.[CrossRef][ISI][Medline]
Fenner, M., and W. Lee. 1997 A comparative study of the distribution of genus size in twenty angiosperm floras. Biological Journal of the Linnaean Society 62: 225–237.
Godley, E. J. 1979 Legume and follicle. New Zealand Gardener, Feb.: 11, 22.
Goldblatt, P. 1981 Cytology and the phylogeny of the Leguminosae. In R. M. Polhill and P. H. Raven [eds.], Advances in legume systematics, part 2, 427–463. Royal Botanic Gardens, Kew.
Green, P. S. 1994 Flora of Australia, vol. 49. Australian Government Publishing Service, Canberra.
Greenwood, R. M. 1978 Rhizobia associated with indigenous legumes of New Zealand and Lord Howe Island. In M. W. Loutit and J. A. R. Miles [eds.], Microbial ecology, 402–403. Springer-Verlag, New York, NY.
Hasegawa, M., H. Kishino, and T. Yano. 1985 Dating of the human-ape splitting by a molecular clock of mitochondrial DNA. Journal of Molecular Evolution 9: 366–369.
Heenan, P. B. 1995 A taxonomic revision of Carmichaelia (Fabaceae–Galegeae) in New Zealand (part I). New Zealand Journal of Botany 33: 455–475.[ISI]
———. 1996 A taxonomic revision of Carmichaelia (Fabaceae–Galegeae) in New Zealand (part II). New Zealand Journal of Botany 34: 157–177.
———. 1997a Heteroblasty in Carmichaelia, Chordospartium, Corallospartium, and Notospartium (Fabaceae–Galegeae) from New Zealand. New Zealand Journal of Botany 35: 243–249.[ISI]
———. 1997b Wood anatomy of the Carmichaelia (Fabaceae–Galegeae) complex in New Zealand. New Zealand Journal of Botany 35: 395–415.[ISI]
———. 1998a Phylogenetic analysis of the Carmichaelia complex, Clianthus, and Swainsona (Fabaceae), from Australia and New Zealand. New Zealand Journal of Botany 36: 21–40.
———. 1998b Montigena (Fabaceae–Galegeae), a new genus endemic to New Zealand. New Zealand Journal of Botany 36: 41–51.[ISI]
———. 1998c An emended circumscription of Carmichaelia (Fabaceae–Galegeae), with new combinations, a key, and notes on hybrids. New Zealand Journal of Botany 36: 53–63.[ISI]
Hutchinson, J. 1964 Genera of flowering plants. Oxford University Press, Oxford.
Kemp, E. M. 1978 Tertiary climatic evolution and vegetation history in the southeast Indian Ocean region. Palaeogeography, Palaeoclimatology, and Palaeoecology 24: 169–204.
Kennett, J. P., N. D. Watkins, and P. Vella. 1971 Palaeomagnetic chronology of Pliocene-Early Pleistocene climates and the Plio-Pleistocene boundary in New Zealand. Science 171: 276–279.
Lavin, M., J. J. Doyle, and J. D. Palmer. 1990 Evolutionary significance of the loss of the chloroplast DNA inverted repeat in the Leguminosae subfamily Papilionoideae. Evolution 44: 390–402.[CrossRef][ISI]
Lee, A. T. 1948 The genus Swainsona. Contributions from the New South Wales National Herbarium 1: 131–271.
Liston, A. 1995 Use of polymerase chain reaction to survey for the loss of the inverted repeat in the legume chloroplast genome. In M. D. Crisp and J. J. Doyle [eds.], Advances in legume systematics, part 7, 31–40. Royal Botanic Gardens, Kew.
———, and J. A. Wheeler. 1994 The phylogenetic position of the genus Astragalus (Fabaceae): Evidence from the chloroplast genes rpoC1 and rpoC2. Biochemical Systematics and Ecology 22: 377–388.[CrossRef]
Maddison, W. P., and D. R. Maddison. 1996 MacClade version 3.06: analysis of phylogeny and character evolution. Sinauer, Sunderland, MA.
McDougall, I., J. J. Embleton, and D. B. Stone. 1981 Origin and evolution of Lord Howe Island, south west Pacific Ocean. Journal of the Geological Society of Australia 28: 155–176.[ISI]
Mildenhall, D. C., and D. T. Pocknall. 1984 Palaeobotanical evidence for changes in Miocene and Pliocene climates in New Zealand. In J. C. Vogel [ed.], Late Cainozoic palaeoclimates of the southern hemisphere. A. A. Balkema, Rotterdam.
Oliver, W. R. B. 1928 The flora of the Waipaoa Series (Late Pliocene) of New Zealand. Transactions and Proceedings of the New Zealand Institute 59: 287–303.
Raven, P. H. 1973 Evolution of subalpine and alpine plant groups in New Zealand. New Zealand Journal of Botany 11: 177–200.
Polhill, R. M. 1981 Tribe 17. Carmichaeliae Hutch. (1964). In R. M. Polhill and P. H. Raven [eds.], Advances in legume systematics, part I, 364–366. Royal Botanic Gardens, Kew.
Sanderson, M. J. 1991 Phylogenetic relationships within North American Astragalus L. (Fabaceae). Systematic Botany 16: 414–430.[CrossRef][ISI]
———. 1997 r8s, version. 0.10. A computer program available by anonymous FTP at loco.ucdavis.edu.
———, and A. Liston. 1995 Molecular phylogenetic systematics of Galegeae, with special reference to Astragalus. In M. D. Crisp and J. J. Doyle [eds.], Advances in legume systematics, part 7, 331–350. Royal Botanic Gardens, Kew.
———, and M. F. Wojciechowski. 1996 Diversification rates in a temperate legume clade: are there "so many species" of Astragalus (Fabaceae). American Journal of Botany 83: 1488–1502.[CrossRef][ISI]
Shaw, W. B., and B. R. Burns. 1997 The ecology and conservation of the endangered endemic shrub, kowhai ngutukaka Clianthus puniceus in New Zealand. Biological Conservation 81: 233–245.[CrossRef][ISI]
Suggate, R. P., G. R. Stevens, and M. T. Te Punga [eds.]. 1978 The geology of New Zealand, vol. II. Government Printer, Wellington, New Zealand.
Swofford, D. L. 1995 Phylogenetic analysis using parsimony (PAUP* version 4.0). Sinauer, Sunderland, MA.
———, G. J. Olsen, P. J. Waddell, and D. M. Hillis. 1996 Phylogenetic Inference In D. M. Hillis, C. Moritz, and B. K. Mable [eds.], Molecular systematics, 2nd ed. Sinauer, Sunderland, MA.
Thompson, J. 1990 New species and new combinations in the genus Swainsona (Fabaceae) in New South Wales. Telopea 4: 1–5.
———. 1993 A revision of the genus Swainsona (Fabaceae). Telopea 5: 427–581.
Wagstaff, S. J., and P. J. Garnock-Jones. 1998 Evolution and biogeography of the Hebe complex (Scrophulariaceae) inferred from ITS sequences. New Zealand Journal of Botany 36: 425–437.[ISI]
Wardle, P. 1991 Vegetation of New Zealand. Cambridge University Press, Cambridge.
Wu, C-I., and W-H. Li. 1985 Evidence for higher rates of nucleotide substitution in rodents than man. Proceedings of the National Academy of Sciences, USA. 82: 1741–1745.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
