| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Systematics |
2Department of Biology, University of Missouri-St. Louis, St. Louis, Missouri 63121 USA; The Missouri Botanical Garden,St. Louis, Missouri 63166 USA; 3Botanical Gardens, The University of Tokyo, Tokyo 112-0001 Japan
Received for publication September 11, 2003. Accepted for publication January 22, 2004.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Key Words: Areae • Arisaema • Beringia • biogeography • molecular clock • Pinellia • sex change • Typhonium
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
; Gusman and Gusman, 2002
; Li et al., in press
). About 140 occur in the Himalayas, southern India, Sri Lanka, China, Korea, Japan, and Southeast Asia; one species ranges from Nepal, across Saudi Arabia and Oman to East Africa; five or six species are endemic in East Africa; and three are endemic to North America (Jack-in-the-pulpit, green dragon, and a close relative; Fig. 1). Arisaema typically occurs in cool temperate environments, including montane grasslands, with a few species thriving near timberline at 4200–4500 m in the Himalayas and at 3200 m in East Africa (on Mt. Ruwenzori). As expected from the geographic range of the genus, most species are deciduous and overwinter via underground tubers or rhizomes. Only about 38 species are evergreen. The sexually produced diaspores are berries that are spread by birds, and asexual propagation occurs via tuber offsets, stolons, or rhizomes. Pollination usually is by fungus gnats (Mycetophilidae) and sciarid gnats (Sciaridae), but pollination by nectar-seeking small bees may also play a role (Murata et al., 1993
; Vogel and Martens, 2000
).
|
; Freeman et al., 1980
). It evolved an unknown number of times in animals and plants, with Arisaema one of the best-documented cases in plants (Schlessman, 1988
). Sex expression in Arisaema is dependent on nutrient status, with several reversals possible during a plant's life. Large plants produce staminate and pistillate flowers or only pistillate flowers and function as hermaphrodites or females. Small plants produce only staminate flowers. As demonstrated experimentally, and well-known to Arisaema growers, size (thus sex) is influenced by resources accumulated in the root storage organs, with effects carrying over for several years and species apparently very long-lived (20 years or more; Bierzychudek, 1984
).
All Arisaema are sex changers, although literature before 1990 sometimes cited A. flavum as consistently producing male and female flowers simultaneously (bibliography available at http://www.umsl.edu/
biosrenn/). However, Murata (1990c)
discovered sex-changing populations of A. flavum, described as subspecies tibeticum, leaving the remaining two subspecies of A. flavum as the only known non-sex-changing entities in Arisaema. Remarkably, subspecies tibeticum is diploid, while the two non-sex-changing subspecies are tetraploid and set viable seeds after selfing (automatic selfing is prevented in sex-changing species of Arisaema). This suggests a return from environmental sex determination to simultaneous bisexuality concomitant with polyploidy and regular selfing. Because of its unusual male flowers (with a single stamen vs. the usual 3–5) and the absence of sex change in two of its three subspecies, A. flavum has been seen as "phenetically most primitive" (Grayum, 1990
), while others have seen it as derived (Li, 1981
; Vogel and Martens, 2000
). The monophyly and phylogenetic position of this species and the possibility of a secondary loss of environmental sex determination are among the questions investigated here.
Except for its unusual sexual strategy, Arisaema shares its reproductive and vegetative characters with Pinellia, a genus of six species in temperate East Asia (G. Zhu and H. Li, unpublished manuscript), and Typhonium, a heterogeneous assemblage of about 50 species from tropical Asia, southeast Australia, and tropical Africa (T. venosum; Sriboonma et al., 1994
; Hetterscheid and Boyce, 2000
; Hetterscheid et al., 2001
; species numbers are moot since Typhonium is paraphyletic; Renner and Zhang, in press
). At least 10 species have been transferred between these genera, for example, A. hirsutum S. Y. Hu [Typhonium hirsutum (S. Y. Hu) J. Murata et Mayo], A. submonoicum Gagnep. [T. horsfieldii (Miq.) Steenis], and A. tripartitum Engl. [Pinellia tripartita (Blume) Schott], illustrating the genera's indistinct boundaries. Testing the monophyly of Arisaema was therefore another important goal of our study.
Morphological and biogeographic work on Arisaema has suggested that it reached East Africa from Asia (Li, 1981
; Grayum, 1990
) and that it entered North America twice. The latter is implied by the traditional placement of the American species in different sections. (The status of the Mexican species, A. macrospathum Benth., as either a distinct species or as a subspecies of A. dracontium has long been unclear [but see Gusman, 2000
].) In spite of their placement in different sections, the North American species are thought to hybridize (Sanders and Burk, 1992
; K. Clay, Indiana University, personal communication), perhaps arguing for relatively recent divergence. The African and Arabian species of Arisaema have been revised (Mayo and Gilbert, 1986
), but remain poorly collected, and their sectional assignment has been difficult (Murata, 1984
, 1990b
). Mayo and Gilbert (1986
; also Mayo, 1993
) have stressed the morphological similarities that exist between African/Arabian and Indian/Nepalese species pairs; they specifically mention the species pair A. consanguineum (Himalaya, northern Thailand) and A. mildbraedii (Africa). Although our sampling of African species is limited, we were able to obtain material of this particular pair and thus could test Mayo and Gilbert's hypothesis.
To better understand the relationships of Asian Arisaema to species occurring in Africa and North America and to gain insight into the evolutionary plasticity of the sex change strategy and the leaf and inflorescence characters that form the basis for current classifications (Hara, 1971
; Murata, 1984
, 1990b
; Gusman and Gusman, 2002
; Li et al., in press
), we here analyze chloroplast sequences obtained for 81 accessions representing 77 of the approximately 150 recognized species of Arisaema.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
). We therefore rooted our trees with a representative sample of species of Pinellia and Typhonium. We specifically included species that have been moved between Arisaema, Pinellia, and Typhonium, namely A. hirsutum S. Y. Hu [Typhonium hirsutum (S. Y. Hu) J. Murata et Mayo], A. submonoicum Gagnep. [T. horsfieldii (Miq.) Steenis], and A. tripartitum Engl. [Pinellia tripartita (Blume) Schott].
Total genomic DNA was isolated from silica-dried leaves using DNeasy kits (QIAGEN, Valencia, California, USA) or the method of Doyle and Doyle (1987)
. DNA amplification by the polymerase chain reaction (PCR) was performed according to the protocol described in Zhang and Renner (2003)
. To amplify the chloroplast trnL intron and adjacent spacer before the trnF gene, we used the universal primers c, d, e, and f of Taberlet et al. (1991)
. The chloroplast rpl20-rps12 intergenic spacer between the ribosomal protein genes S12 and L20 was sequenced using primers rpl20 and rps12 of Hamilton (1999)
.
Amplified fragments were purified by running the entire product on a low melting-point agarose gel and then recovering the DNA with QIAquick Gel Extraction Kits (QIAGEN). Cycle sequencing of the purified PCR products used the BigDye Terminator Cycle Sequencing kit (Applied Biosystems [ABI], Norwalk, Connecticut, USA) according to the manufacturer's suggested protocol. The dye was removed by 2 µL of 3 mol/L NaOAc (pH 4.6) and 50 µL ethanol precipitation. Samples were then run on the ABI 377 automated sequencer of the Department of Biology at the University of Missouri-St. Louis. Both strands were sequenced and used to generate consensus sequences in Sequencher (version 4.1.2, GeneCodes, Ann Arbor, Michigan, USA), which was also used for the manual sequence alignment of all sequences. Alignment was unproblematic except for a long stretch of up to 218 base pairs (bp) in the trnL intron. The aligned sequences were copied into NEXUS files for phylogenetic analysis.
Phylogenetic analyses and molecular clock dating
Tree searches were conducted with version 4.0b.10 of PAUP* (Swofford, 2002
) or MrBayes version 3.0B4 (Huelsenbeck and Ronquist, 2001
; Ronquist and Huelsenbeck, 2003
). DNA insertions or deletions were treated as missing data and excluded from most parsimony, maximum likelihood (ML), and Bayesian analyses. Parsimony analyses used heuristic searching, 10 random sequence taxon addition replicates, with 100 trees in memory, and tree bisection-reconnection (TBR) swapping. Clade support was assessed via nonparametric bootstrapping, using PAUP. Bootstrap analyses under parsimony used 200 replicates, with closest taxon addition and one tree held in memory. The "collapse," but not the "steepest descent," option of PAUP were in effect during all searches, and character changes were interpreted under ACCTRAN optimization.
Maximum likelihood (ML) analyses were performed using the general time reversible (GTR) + Gamma (G) + Pinv model, which uses a discrete gamma model (in our case with four rate categories) to approximate the continuous gamma distribution (Yang, 1994
; implemented in PAUP) and also assumes a proportion of invariable sites. Base frequencies were the empirically observed ones. The shape parameter of the gamma distribution can be estimated efficiently from any of numerous equally parsimonious trees because its dependence on tree topology is minor (Yang and Kumar, 1996
; Sullivan et al., 1999
), which greatly reduces the computational demands of maximum likelihood searches.
Bayesian analyses used the GTR + G model. One cold and three incrementally heated Markov chain Monte Carlo (mcmc) chains were run for 100 000 or 1 million cycles, with trees sampled every 10th or 100th generation, using random trees as starting points and a temperature parameter value of 0.2 (the default in MrBayes). For each data set, mcmc runs were repeated at least twice as a safeguard against spurious results. The first 5000 or 7000 trees were discarded as burn-in, and the remaining tree used to construct Bayesian consensus trees. Examination of the log-likelihoods and the observed consistency between runs suggested that these burn-in periods were sufficiently long for chains to have become stationary.
For the molecular clock dating, we first performed a likelihood ratio test (LRT; Felsenstein, 1981
) that compared the likelihood scores with and without the clock assumption on one of the 26 equally parsimonious trees obtained from the combined data (excluding all gaps and rooted with just Pinellia tripartita). When the data did not reject the clock (Results), we used an ultrametric (clock-enforced) tree to obtain age estimates for the biogeographically interesting disjunctions. In addition, we used cross-validated penalized likelihood (Sanderson, 2002
; implemented in the Unix shareware program r8s, http://ginger.ucdavis.edu/r8s/), because this software enables one to incorporate multiple calibration points and minimal and/or maximal ages, rather than a single, fixed calibration point, with the program then calculating the most likely ages of nodes given the remaining constraints and substitutions in the data set. Penalized likelihood is a semiparametric approach that allows different substitution rates between ancestral and descendent branches, but then reduces the resulting enormous number of more or less arbitrary alternatives by assigning a penalty that increases with the abruptness of rate change between adjacent branches. The penalty (or "smoothing" parameter) is calculated by sequentially removing part of the data (one branch at a time), re-estimating the remaining model parameters, and using the fitted model parameters to predict the data that were removed (i.e., the expected number of substitutions on the pruned branch). In the case of clocklike or almost clocklike data, the smoothing value is high, resulting in invariable rates between ancestors and descendants (Sanderson, 2002
). Trees were studied in TreeEdit (http://evolve.zoo.ox.ac.uk/software/TreeEdit/; Rambaut and Charleston, 2000
).
A calibration for genetic distances in Arisaema comes from infructescences from the mid-Miocene Latah Formation near Spokane (18–16 [million years] ago; Knowlton, 1926
) that closely match infructescences of living North American A. triphyllum.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Phylogenetic analyses
The trnL-F spacer sequences contain too little signal to yield much resolution for the 81 ingroup accessions (Fig. 2). It is apparent, however, that many species group according to section, albeit usually without statistical support. (The 11 sections recognized by Murata [1984
, 1990b]
are Arisaema, Clavata, Decipientia, Dochafa, Fimbriata, Franchetiana, Nepenthoidea, Pedatisecta, Sinarisaema, Tenuipistillata, and Tortuosa. All are here represented by their type species, except section Decipientia of which we could not obtain type species material and instead included material of A. rhizomatum, the only other species placed in this section.) When we sequenced more than one accession per species (A. ciliatum, A. flavum, and A. speciosum), conspecific accessions differed by one or two substitutions in the trnL intron, the trnL-F spacer, or the rpl20-rps 12 spacer, with the three regions not differing consistently in information content at this level.
|
|
|
2 = 2(3437.4240–3421.1720) = 35.50, P < 0.1; 26 df). The data could thus be modeled as clocklike. Setting the split between A. triphyllum and its Asian sister species A. amurense to 18 my old yields a substitution rate of 0.000081 substitutions per site per my (0.00146 divided by 18). This gives estimates of 31 ± 10 my for the divergence between Nepalese-Tibetan A. costatum and East African A. schimperianum and of 49 ± 12 my for the split between Asian A. heterophyllum and the North American A. dracontium/A. macrophyllum clade. Cross-validated penalized likelihood, for which we constrained the split between A. triphyllum and A. amurense to minimally 18 or 20 my old (in different runs) and the root of the genus to maximally 60 my old, yielded younger ages, namely 19 my for the split between A. costatum and A. schimperianum and 32 my for the split between A. heterophyllum and A. dracontium/A. macrophyllum. The divergence between the diploid A. flavum subsp. flavum and its tetraploid relatives ssp. abbreviatum and tibeticum was dated to 13 my by the ultrametric clock and to 8 my by penalized likelihood. |
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
, b
), evolved once in the common ancestor of extant Arisaema. The only taxa that lack the sex change strategy are subspecies abbreviatum and flavum of A. flavum and both are tetraploid (Murata, 1990c
). The third subspecies of A. flavum, subsp. tibeticum, is diploid and changes sex from year to year just like the remaining species in the genus. Subspecies tibeticum occurs from Bhutan to Sichuan, ssp. abbreviatum occurs from Nepal to Oman, and the typical form of ssp. flavum is thought to range from Oman, Yemen, and Saudi Arabia to Ethiopia (our material, verified by P. Boyce, however is from Pakistan). It is plausible that the two tetraploid subspecies, concomitant with their spread into drier habitats, reverted from sex change and the correlated dependence on cross-pollinating insects to simultaneously bisexual spathes, which provide independence from pollinating insects because they permit automatic selfing (with self-pollen falling on receptive stigmas; Vogel and Martens, 2000
). Most species of Arisaema are pollinated by humidity-loving fungus gnats and sciarid gnats (Vogel and Martens, 2000
), but the subspecies of A. flavum are unusual in offering a sugary secretion in its inflorescences (Murata et al., 1993
; J. Murata, personal observations). The exact roles of insect pollination vs. automatic selfing in natural populations are unknown.
While the monophyly of Arisaema appears clear, the group's relationships to its closest relatives, Pinellia and Typhonium, remain unresolved. A study based on mitochondrial and chloroplast loci for the clade of 16 genera to which Arisaema belongs also was unable to resolve the trichotomy of Arisaema, Pinellia, Typhonium (Renner and Zhang, in press
), but revealed that all genera of Areae (Arum, Biarum, Dracunculus, Eminium, Helicodiceros, and Theriophonum) are embedded in Typhonium. Based on overall phenetic resemblance, Grayum (1990
, p. 682) suggested that Arisaema was sister to Areae. By contrast, Murata (1990a)
saw Arisaema as sister to Pinellia, based on a uniquely shared leaf-folding pattern in which all of the lateral leaflets are folded downwards in bud. However, the downward folding is lost in three of the 11 sections of Arisaema and is found in at least one species of Typhonium transferred there from Arisaema by Murata. Additional sequencing is needed to resolve the Arisaema-Pinellia-Areae (Typhonium) trichotomy.
Major groups within Arisaema
Current understanding of morphological evolution in Arisaema, and most of the sectional classification, go back to Engler (e.g., 1879; our Fig. 2), who gave much weight to leaf shape (diagrams of most species' leaf shape: Gusman and Gusman, 2002
). Based on new anatomical and morphological data from living material, Hara (1971)
, Murata (1984
, 1990b
), and Gusman and Gusman (2002)
made major modifications to Engler's system, and the molecular data to some extent support their newly circumscribed sections, albeit without statistical support and generally with one or two species falling outside the species cluster around the type species. This is the case for sections Arisaema, Clavata, Dochafa, Fimbriata (which includes parts of Tortuosa), Franchetiana, Pedatisecta, and Sinarisaema. Section Nepenthoidea is supported in the combined spacer data (Fig. 3), but is not represented in the spacer + intron data (Fig. 4). Sections Decipientia and Tenuipistillata are represented by only one species (Fig. 3).
The only section that is clearly polyphyletic is Tortuosa. Tortuosa was based on pedate or rarely simple, trifoliolate, or radially 5–7-foliolate leaves and a sessile spadix appendage that is long exerted from the spathe (see keys in Murata, 1984
; Gusman and Gusman, 2002
). The section is especially variable in chromosome number (Watanabe et al., 1998
), and all its characters also appear elsewhere in the genus. Based on our results, its traits will need to be reanalyzed or weighted differently to define a monophyletic subgroup around the type species of sect. Tortuosa.
A surprising finding is the placement of A. jinshajiangense in a clade with members of sect. Sinarisaema (Fig. 2). Arisaema jinshajiangense has leaves that originate in a quincuncial spiral and are trifoliolate (have just three leaflets), while most other species in Sinarisaema have leaves arranged in a spirodistichous spiral and with numerous radiate leaflets (all leaflets emerge from the same central point; see drawings in Gusman and Gusman, 2002
, p. 298–299). Gusman and Gusman (2002
, p. 144) include A. jinshajiangense in sect. Clavata but its trifoliolate leaves are also exceptional in that section, which usually has pedate leaves (p. 136–137). The unexpected placement of A. jinshajiangense with species that have quite different leaves may point to a greater plasticity in phyllotaxis than heretofore assumed for the genus.
Morphologically distinct sections that are supported by our data, such as section Arisaema, with strongly fused anther cells, a long spadix appendage that is exserted from the spathe, and erect fruiting peduncles, are usually also geographically coherent (sect. Arisaema includes 17 species that are restricted to the Sino-Himalayan region). The finding that the African species A. mildbraedii and A. schimperianum apparently belong in sect. Arisaema, where they are related to Nepalese/ Tibetian A. costatum, is all the more surprising. The relevant sequences all have unique motifs that differ from the remaining 74 species sequenced, which excludes contamination. The material of A. mildbraedii was obtained from Kenya, while the material of A. schimperianum (originally from Ethiopia) was contributed by J. Murata. The grouping of these two species with A. costatum is mainly due to a uniquely shared mutation from A to C at trnL-F nucleotide position 109. As pointed out by Mayo and Gilbert (1986)
, morphological similarities exist between African/Arabian species of Arisaema and certain Nepalese species, and these authors specifically mentioned the species pair A. mildbraedii and A. consanguineum (Himalaya, northern Thailand). Our data (Fig. 2) instead show A. mildbraedii grouping with A. costatum and A. schimperianum, while A. consanguineum groups with members of Sinarisaema in agreement with its morphology (Murata, 1984
). More generally, however, the data support Mayo and Gilbert's assessment of close ties between some African and Asian species (next section), and unpublished observations by Murata on cultivated A. schimperianum have revealed that its leaf arrangement is quincuncial, which would fit with the quincuncial leaf arrangement typical of section Arisaema and would provide some morphological support for the placement of this East African species in section Arisaema. Additional sampling of Indian and African species is needed for a fuller exploration of the connections between East Africa, India, and Asia.
Temporal and spatial range expansion in Arisaema
From the molecular clock estimates, the divergence between African A. schimperianum and Nepalese/Tibetan A. costatum occurred between 31 to 19 my ago, with the semiparametric approach (penalized likelihood) favoring the younger date and the ultrametric approach the older date. That penalized likelihood yields younger ages is mainly due to the constraint of the root node to maximally 60 my old. The justification for this constraint lies in the highly derived position of Arisaema, Pinellia, and their relatives, the Areae, in the family tree of Araceae (Cabrera et al., in press
). Alternatively, one can rely on the estimates from the ultrametric approach. Either estimate fits with the considerably moister settings that existed in the southeastern parts of the Arabia peninsula during the Oligocene and Miocene (Whybrow and Hill, 1999
, and references therein). Mammalian faunas of Afro-Arabia underwent a marked transition near the Oligocene/Miocene boundary at approximately 24 my, when many endemic taxa were replaced by migrants from the Northern Hemisphere (Kappelman et al., 2003
). Clearly, these age estimates should be taken with a grain of salt because they are influenced by our taxon sampling density, which as discussed is lowest for Indian and African species.
The problem of taxon sampling is less acute in our placement of North American Jack-in-the-pulpit, A. triphyllum, and green dragon, A. dracontium and its Mexican relative, A. macrospathum. Arisaema triphyllum belongs to section Pedatisecta, with some 32 species endemic in China, Korea, and Japan, and is most similar to the Asian A. amurense from which it can be distinguished only by details of the inflorescence (Murata, 1984
). Arisaema dracontium and A. macrospathum belong in section Tortuosa, which in addition has one species in India and 11 in China, Korea, and Japan (three extending into Taiwan, Thailand, and Burma). American A. dracontium and Mexican A. macrospathum are especially similar to Asian A. heterophyllum (Engler, 1879
, p. 546–548 and more recent authors). The placement of the three American species in different Asian clades by the molecular data agrees with their traditional morphological placement in different sections and is unlikely to change with additional sampling. The 18-my-old fossil Arisaema infructescence discovered near Spokane (Knowlton, 1926
) further supports that Arisaema ranged across Beringia, possibly into the Early Miocene, although the fossil of course does not indicate when connections between Asian and American populations were severed. A new insight from the molecular clock estimates is that the origin of today's A. dracontium/A. macrophyllum clade may date back to the Oligocene (whether estimated under semiparametric or ultrametric clocks) and may be quite a bit older than the A. triphyllum line. Arisaema is not the only genus inferred to have crossed Beringia repeatedly; additional Asian-centered genera with multiple links to North America include Magnolia (Qiu et al., 1995
), Nyssa (Wen and Stuessy, 1993
), Panax (Wen, 2001
), Trillium (Kato et al., 1995
), and Aralia (Wen, 1999
, for a review).
Future work in the genus will depend on the addition of a suitably fast-evolving, perhaps nuclear, gene region. In our experience and that of others, working with the internal transcribed spacer region (ITS) in Arisaema is problematic (Buzgo et al., 2002
), with multiple paralogous copies commonly found as a result of slow or ineffective concerted evolution.
|
FOOTNOTES |
|---|
The authors thank T. Barkman, J. Bogner, P. Bruggeman, A. M. Chambers, Binghui Chen, G. Gusman, W. Hetterscheid, B. W. Magrys, J. McClements, H. He, C. Obunyali, P. Schmidt, J. W. Waddick, J. C. Wang, and the botanical gardens of Mainz and Munich for leaf material; B. Genton and A. Weerasooriya for help with sequencing; and G. Gusman for support throughout this study—without his material, this work would not have been possible. Funding for this research came from the University of Missouri system and the University of Missouri-St. Louis. 
4 Present address: Systematische Botanik, Ludwig-Maximilians-Universitaet, Menzinger-Str. 67, D-80638 Muenchen, Germany (renner{at}umsl.edu
)
5 Present address: Department of Biology, Colorado State University, Fort Collins, CO 80523
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Buzgo M. Y.-L. Qiu M. W. Chase 2002 ITS in Arisaema (Araceae). AIBS abstract, published at http://www.botany2002.org/viewer.shtml
Cabrera L. I. G. A. Salazar M. W. Chase S. J. Mayo In press Phylogenetics of Araceae and Lemnaceae: evidence from multiple plastid DNA data sets. Proceedings of the Monocots III meeting, Rancho Santa Ana Botanical Garden, Claremont, California, USA
Charnov E. L. J. Bull 1977 When is sex environmentally determined?. Nature 266: 828-830[CrossRef][Medline]
Doyle J. J. Doyle 1987 A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochemical Bulletin 19: 11-15
Engler A. 1879 Araceae. In A. and C. de Candolle. [eds.], Monographieae Phanerogamarum, vol. 2, 1–681. Wolf and Sons, Munich, Germany
Felsenstein J. 1981 Evolutionary trees from DNA sequences: a maximum likelihood approach. Journal of Molecular Evolution 17: 368-376[CrossRef][ISI][Medline]
Freeman D. C. K. T. Harper E. L. Charnov 1980 Sex change in plants: old and new observations and new hypotheses. Oecologia 47: 222-232[CrossRef][ISI]
Grayum M. H. 1990 Evolution and phylogeny of the Araceae. Annals of the Missouri Botanical Garden 77: 628-697[CrossRef][ISI]
Gusman G. 2000 Observations on Arisaema macrospathum Benth. (Araceae). Aroideana 23: 41-47
Gusman G. L. Gusman 2002 The genus Arisaema. A monograph for botanists and nature lovers. Ganter Verlag, Ruggell, Liechtenstein
Hamilton M. B. 1999 Four primer pairs for the amplification of chloroplast intergenic regions with intraspecific variation. Molecular Ecology 8: 521-523[Medline]
Hara H. 1971 A revision of the eastern Himalayan species of the genus Arisaema (Araceae). In H. Hara [ed.], Flora of eastern Himalaya, 2nd Report, 321–354. University of Tokyo Press, Tokyo, Japan
Hetterscheid W. L. A. P. C. Boyce 2000 A reclassification of Sauromatum Schott and new species of Typhonium Schott (Araceae). Aroideana 23: 48-55
Hetterscheid W. L. A. D. Sookchaloem J. Murata 2001 Typhonium (Araceae) of Thailand: new species and a revised key. Aroideana 24: 30-55
Huelsenbeck J. P. F. R. Ronquist 2001 MrBayes: Bayesian inference of phylogeny. Bioinformatics 17: 754-755
Kappelman J. et al. 2003 Oligocene mammals from Ethiopia and faunal exchange between Afro-Arabia and Eurasia. Nature 426: 549-552[CrossRef][Medline]
Kato H. S. Kawano R. Terauchi M. Ohara F. H. Utech 1995 Evolutionary biology of Trillium and related genera (Trilliaceae). I. Restriction site mapping and variation of chloroplast DNA and its systematic implications. Plant Species Biology 10: 17-29
Knowlton F. H. 1926 Flora of the Latah formation of Spokane, Washington, and Coeur d'Alene, Idaho. U.S. Geological Survey, Reston, Virginia. Professional Paper 140A: 1-81
Li H. 1981 Himalayas-Hengduan Mountains–the center of distribution and differentiation of the genus Arisaema. In Li Dongsheng [ed.], Geological and ecological studies of Qinghai-Xizang Plateau, vol. 2, 1321–1327. Science Press, Beijing, China
Li H. G. Zhu J. Murata In press Arisaema. In C. Y. Wu and P. Raven [eds.], Flora of China, vol. 00, 00–00. Science Press, Beijing, China and Missouri Botanical Garden, St. Louis, Missouri, USA
Mayo S. J. 1993 Aspects of aroid geography. In W. George and R. Lavocat [eds.], The Africa–South America connection, 44–58. Clarendon Press, Oxford, UK
Mayo S. J. N. G. Gilbert 1986 A preliminary revision of Arisaema (Araceae) in tropical Africa and Arabia. Kew Bulletin 41: 261-278[CrossRef]
Murata J. 1984 An attempt at an infrageneric classification of the genus Arisaema (Araceae). Journal of the Faculty of Science, University of Tokyo, section 3, Botany 13: 431-482
Murata J. 1990a Developmental patterns of pedate leaves in tribe Areae (Araceae-Aroideae) and their systematic implication. Botanical Magazine (Tokyo) 103: 339-343[CrossRef][ISI]
Murata J. 1990b Present status of Arisaema systematics. Botanical Magazine (Tokyo) 103: 371-382[CrossRef][ISI]
Murata J. 1990c Three subspecies of Arisaema flavum (Forssk.) Schott (Araceae). Japanese Journal of Botany 65: 65-73
Murata J. C. Ren H. Murata S. Ogawa 1993 Arisaema flavum, a honey-producing Araceae. Abstract 1203, 15th International Botanical Congress, Yokohama, Japan
Qiu Y.-L. M. W. Chase C. R. Parks 1995 A chloroplast DNA phylogenetic study of the eastern Asia–eastern North America disjunct section Rytidospermum of Magnolia (Magnoliaceae). American Journal of Botany 82: 1582-1588[CrossRef][ISI]
Rambaut A. M. M. Charleston 2000 TreeEdit version 1.0 alpha 4–61. Computer program available at http://evolve.zoo.ox.ac.uk/software.html?id=TreeEdit University of Oxford, Oxford, UK
Renner S. S. L.-B. Zhang In press Biogeography of the Pistia clade (Araceae) based on cp and mt DNA sequences and Bayesian divergence time inference. Systematic Biology
Ronquist F. J. P. Huelsenbeck 2003 MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics 19: 1572-1574
Sanders L. L. C. J. Burk 1992 A naturally-occurring population of putative Arisaema triphyllum subsp. stewardsonii x A. dracontium hybrids in Massachusetts. Rhodora 94: 340-347[ISI]
Sanderson M. J. 2002 Estimating absolute rates of molecular evolution and divergence times: a penalized likelihood approach. Molecular Biology and Evolution 19: 101-109
Schlessman M. A. 1988 Gender diphasy ("sex choice"). In J. Lovett Doust and L. Lovett Doust [eds.], Plant reproductive ecology: patterns and strategies, 139–153. Oxford University Press, New York, New York, USA
Sriboonma D. J. Murata K. Iwatsuki 1994 A revision of Typhonium (Araceae). Journal of the Faculty of Science, University of Tokyo, section 3, Botany 15: 255-313
Sullivan J. D. L. Swofford G. J. P. Naylor 1999 The effect of taxon sampling on estimating rate heterogeneity parameters of maximum-likelihood models. Molecular Biology and Evolution 16: 1347-1356[ISI]
Swofford D. L. 2002 PAUP*. Phylogenetic analysis using parsimony (*and other methods), version 4. Sinauer Associates, Sunderland, Massachusetts, USA
Taberlet P. L. Gielly 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]
Vogel S. J. Martens 2000 A survey of the function of the lethal kettle traps of Arisaema (Araceae), with records of pollinating fungus gnats from Nepal. Botanical Journal of the Linnean Society 133: 61-100[CrossRef]
Watanabe K. T. Kobayashi J. Murata 1998 Cytology and systematics in Japanese Arisaema (Araceae). Journal of Plant Research 111: 509-521[CrossRef][ISI]
Wen J. 1999 Evolution of eastern Asian and eastern North American disjunct distributions in flowering plants. Annual Review of Ecology and Systematics 30: 421-455[CrossRef][ISI]
Wen J. 2001 Evolution of the Aralia-Panax complex (Araliaceae) as inferred from nuclear ribosomal ITS sequences. Edinburgh Journal of Botany 58: 183-200[CrossRef]
Wen J. T. F. Stuessy 1993 The phylogeny and biogeography of Nyssa (Cornaceae). Systematic Botany 18: 68-79[CrossRef][ISI]
Whybrow P. J. A. Hill [Eds.] 1999 Fossil vertebrates of Arabia. Yale University Press, New Haven, Connecticut, USA
Yang Z. 1994 Maximum likelihood phylogenetic estimation from DNA sequences with variable rates over sites: approximate methods. Journal of Molecular Evolution 39: 306-314[CrossRef][ISI][Medline]
Yang Z. S. Kumar 1996 Approximate methods for estimating the pattern of nucleotide substitution and the variation of substitution rates among sites. Molecular Biology and Evolution 13: 650-659[Abstract]
Zhang L.-B. S. S. Renner 2003 The deepest splits in Chloranthaceae as resolved by chloroplast sequences. International Journal of Plant Sciences 164: (5 Suppl.) S383-S392[CrossRef]
Read all eLetters
eLetters:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
