| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
What's this? |
Systematics |
2Laboratoire Dynamique de la Biodiversité, UMR 5172, 118 rte de Narbonne, 4R3, F-31062, Toulouse, France; 3Plant Biomechanics Group, Institute for Biology II, Botanical Garden of the Albert-Ludwigs-Universität, Schänzlestrasse 1, D-79104 Freiburg, Germany; 4Botanique et Bioinformatique de l'Architecture des Plantes, UMR 5120 CNRS, TA40/PS2, Boulevard de la Lironde, F-34398 Montpellier, France
Received for publication September 7, 2004. Accepted for publication May 4, 2005.
ABSTRACT
Lianas are common in the Apocynaceae s.l. and are predominant in the subfamily Secamonoideae. Shrub-like taxa are rare within this subfamily but occur in Malagasy genera such as Secamone, Secamonopsis, and Pervillaea. We explored the evolutionary appearance of shrub-like growth forms in Malagasy Secamonoideae through a molecular phylogeny using chloroplastic sequences. The phylogeny revealed several independent appearances of shrub-like growth forms within the Secamonoideae. Biomechanics and development of the shrub-like growth form were detailed in one species, Secamone sparsiflora, which has upright and self-supporting young stems that become procumbent in older stages of development. Biomechanical investigations revealed characteristics atypical of both lianas and self-supporting shrubs. Anatomical development in S. sparsiflora is initially similar to lianas in the same clade but shows potentially neotenic retention of juvenile wood development for most of the growth trajectory. The results suggest that evolution of lianescence can carry a high degree of specialization and developmental burden that might limit evolution back to self-supporting growth forms. Under certain geographic and ecological conditions, such as geographic isolation, xeric conditions and/or reduced biotic competition, escapes from lianescence to other growth forms can occur in some angiosperm groups via relatively simple heterochronic shifts of mechanically significant growth processes.
Key Words: Biomechanics • heterochrony • liana • molecular phylogeny • Secamone sparsiflora • Secamonoideae • shrub-like growth form
Plant growth forms and architecture can be described using qualitative features such as general morphology, ecology, and life history (e.g., Hallé et al., 1978
; Speck and Rowe, 1999
). In many cases, growth form states generally defined as trees, shrubs, or lianas may result from different developmental features and so might not be homologous. Studies using phylogenetic investigations to identify evolutionary processes that underlie changes in growth habit or overall architecture are relatively few (Bateman, 1994
, 1999
; Bateman and DiMichele, 1994
; Bateman et al., 1998
; Civeyrel and Rowe, 2001
; Speck et al., 2003
). Mechanical architectures of different plant growth forms can be characterized by trends in (1) flexural stiffness EI, a measure of the tangible resistance of a structure to bending forces, (2) axial second moment of area I, a measure of the shape and size of a stem section with reference to the central neutral axis in the plane of bending, and (3) structural Young's modulus Estr, equal to EI/I, a measure of the combined elastic stiffness of materials comprising a plant stem, independent of the size or geometric characteristics. Plants with self- or non-self-supporting growth forms can be characterized by variations in these properties of the stem during ontogeny (Speck, 1991
, 1994
; Rowe and Speck, 1996
, 2004
; Speck and Rowe, 1999
; Köhler et al., 2000
; Gallenmüller et al., 2001
; Isnard et al., 2003a
, b
; Speck et al., 2003
). Self-supporting plants such as trees and shrubs increase in Estr from young to older stems: both young individual plants and distal shoots of adults are relatively flexible, whereas older supporting trunks and branches are relatively stiff (Speck, 1991
, 1994
; Speck and Rowe 1999
). Lianas, however, show an opposite trend: "searcher shoots" are relatively stiff, whereas basal parts of the stem are more flexible (Speck, 1994
; Speck and Rowe, 1999
). Older stages of some lianas are well known for having highly compliant properties. These two growth forms represent extremes of a very complex range of mechanical architectures found in terrestrial plants. Another type of mechanical architecture includes what we refer to as "semi-self-supporting plants" in which the stem maintains unchanging and often relatively high values of structural Young's modulus in both young and old stages of growth. These plants might, to some extent, be considered as climbers but more often simply lean on the surrounding vegetation and are not usually firmly attached via twining or tendrils (Speck, 1994
; Rowe and Speck, 1998
; Speck and Rowe, 1999
; Gallenmüller et al., 2001
).
The evolution of lianoid growth forms has occurred many times in the course of plant evolution (Bhambie, 1972
; Putz, 1984
; Gentry, 1991
; Hegarty and Caballé, 1991
; Caballé, 1993
; Olson, 2003
) with perhaps half of all vascular plant families including climbers (Gentry, 1991
). Although scandent growth habits have evolved independently in many different plant groups, relatively few plant families such as the Convolvulaceae, Cucurbitaceae, Vitaceae and less inclusive taxa such as Mikania (Compositae) and Paullinia (Sapindaceae) have radiated almost exclusively as scandent habits (Gentry, 1991
). Many families of angiosperms dominated by trees and shrubs include lianescent members, but it is rare that basally lianescent or vinelike clades include large-bodied, woody self-supporting trees or shrubs (Speck et al., 2003
). Clades that have basally evolved a specialized lianoid habit may have developed constraints that limit evolution back to a fully self-supporting habit (Rowe and Speck, 2004
). Within the Apocynaceae s.l., there is a broad diversity of growth forms with trees and shrubs occurring predominantly in the basal subfamily Rauvolfioideae, and lianas mostly comprising the four remaining subfamilies Apocynoideae, Periplocoideae, Asclepiadoideae, and Secamonoideae.
The Secamonoideae (Endlicher, 1838
) is the smallest subfamily of the Apocynaceae sensu lato (Endress and Bruyns, 2000
); it contains seven genera (Secamone, Toxocarpus, Genianthus, Pervillaea, Secamonopsis, Calyptranthera, and Trichosandra) and has less than 200 species restricted to the Old World Tropics. Species of the Secamonoideae occur mainly in Madagascar (Klackenberg, 1992a
), where about half of the known species and genera are found, followed by southeast Asia and then Africa (Klackenberg, 1992b
, 2001
). Secamone is the largest genus with over 90 species occurring mainly in Madagascar (Klackenberg, 1992a
), Africa (Goyder, 1992
), Australia (Klackenberg, 1992b
), and Asia (Forster and Harold, 1989
; Klackenberg, 1992b
). The Secamonoideae are mostly lianas that climb by twining on trees and shrubs in semi-arid and tropical ecosystems. The group does not have specialized forms of trellis attachment other than twining leading stems (Baas et al., 1983
). Anomalous arrangements of secondary tissues described for many other tropical liana species (Fisher and Ewers, 1989
, 1992
; Caballé, 1993
, 1998
; Rowe and Speck, 1996
, 2004
; Speck et al., 1996
) occur in several climbing species of Apocynaceae s.l. (Solereder and Boodle, 1908
; Metcalf and Chalk, 1979
) but have not been observed in the subfamily Secamonoideae.
Recent studies on growth form diversity and systematics of the Secamonoideae (Civeyrel et al., 1998
; Civeyrel and Rowe, 2001
) indicate that rare nonclimbing and shrub-like forms are phylogenetically nested within this subfamily and especially within Malagasy genera such as Secamone, Pervillaea, and Secamonopsis. In Malagasy Secamonoideae, growth forms are predominantly lianas but some are shrub-like although not fully self-supporting such as Secamonopsis microphylla. This species is a nonclimbing member of the group with upright, nontwining aerial branches and procumbent or prostrate older stems (Civeyrel and Rowe, 2001
; Speck et al., 2003
). The bending mechanical properties of S. microphylla have a reduction in structural Young's modulus during development from young to old stems but not of the magnitude of true lianas, which can have decreases of up to 90% (Speck, 1994
; Rowe and Speck, 1996
; Speck and Rowe, 1999
; Civeyrel and Rowe, 2001
). The sister species of S. microphylla, Secamonopsis madagascariensis have a large and significant reduction in structural Young's modulus from young searcher stems to older woody twining axes of approximately 85% (Civeyrel and Rowe, 2001
). Furthermore, older stems of S. microphylla are narrower in diameter than those of S. madagascariensis. Secamone sparsiflora is another Malagasy shrub-like species and is present in arid and semi-arid zones. It does not generally twine but has upright young stems that become inclined and then procumbent after reaching a certain height in older stages of development.
In this study we investigate patterns of growth form evolution and underlying evolutionary processes within the Secamonoideae based on a cladistic analysis of nucleotide sequences from two regions of the chloroplast genome: (1) the maturase-encoding gene matK (Sugita et al., 1985
), which has been used previously to investigate relationships within the Apocynaceae (Endress et al., 1996
; Civeyrel et al., 1998
; Civeyrel and Rowe, 2001
), and (2) the trnT-trnL spacer (Taberlet et al., 1991
), which has been useful for delimiting genera within the Apocynaceae particularly within the Asclepiadoideae (Liede and Täuber, 2000
, 2002
; Liede, 2001
; Meve and Liede, 2001
, 2002
; Liede et al., 2002
). This study also focuses on one of the derived shrub-like species, Secamone sparsiflora Klack., in order to observe how the biomechanics of the stem and other developmental changes leading to a shrub-like growth form have occurred in this predominantly lianoid subfamily. Secamone sparsiflora Klack. is endemic and common in the open semi-arid vegetation at Isalo National Park in southwest, central Madagascar (22°37'S, 45°21'E). Geometric and mechanical properties of the main stem and branches are measured and correlated with anatomical development of the stem. We determine how the stem macroanatomy (bark, wood, and pith) and microanatomy (vessel, fiber wall, and wood ray densities) influence the mechanical variations observed. The mechanical architecture of S. sparsiflora is then discussed in an evolutionary context with reference to the sister group of lianas and other shrub-like forms in the Secamonoideae. We discuss the potential evolutionary constraints possibly limiting evolutionary transitions from mechanically specialized climbing lianas to self-supporting forms, and discuss possible developmental mechanisms by which such changes might be effected.
MATERIALS AND METHODS
Molecular phylogeny
DNA isolation, sequences analyzed, polymerase chain reaction (PCR), and sequencing
For the published sequences, total DNA was extracted as described in Civeyrel et al. (1998)
. Total DNA of the new sequences was isolated by grinding 0.2–0.4 g of herbarium material in liquid nitrogen and then treating samples with the Qiagen (Hilden, Germany) DNA easy plant kit. We examined nucleotide sequences from two regions of the chloroplast genome. The first region includes the maturase-encoding gene matK (1.5 kb) that lies within a 2.5 kb intron and interrupts two trnK exons (Sugita et al., 1985
). MatK has provided useful phylogenetic information for studies at the intrafamilial level in angiosperms (e.g., Xiang et al., 1998
; Hu et al., 2000
) and shows a suitable rate of mutation and resolution at the infrafamilial level (Chase et al., 2003
; Kimball et al., 2003
; Sauquet et al., 2003
; Treutlein et al., 2003
) that has been informative for relationships within the Apocynaceae (Endress et al., 1996
; Civeyrel et al., 1998
; Civeyrel and Rowe, 2001
). The second region of the chloroplast genome analyzed is a fragment of about 0.8 kb including the trnT-trnL spacer. This is a noncoding region of cpDNA for which universal primers exist (Taberlet et al., 1991
). For the gene matK, sequences of 18 additional species were added to the 26 species previously published (16 in Civeyrel et al., 1998
, and 10 in Civeyrel and Rowe, 2001
). For the spacer trnT-trnL, 42 new sequences were added to those published in Liede and Täuber (2002)
. The sources of plant material and GenBank accession numbers for matK and trnT-trnL from all taxa included in this paper are provided in the Appendix.
Double strands of matK were PCR-amplified using the primers matK 8F, matK 681F, matK 829R, matK 829F, and matK 1628R (Endress et al., 1996
; Civeyrel et al., 1998
). The trnT-trnL intergenic spacer of the chloroplast genome was PCR amplified using the primer pairs a and b designed by Taberlet et al. (1991)
. PCR amplifications were carried out using 1–8 µL total DNA, 5 µL 10x buffer, 3.5 mM MgCl2 (7 µL 25 mM stock), 0.2 mM for each dNTP (1 µL 10 mM stock), 0.2 µM for each primer (0.5 µL 50 µM stock), and 0.25 µL Taq DNA polymerase (5 units/µl). Sterile water was added to bring the final volume up to 50 µL. The addition of BSA (bovine serum albumin) was sometimes required in a few samples to suppress the effect of inhibitors. Thermal reactions were carried out as follows: (1) one initial cycle of 5 min at 95°C to ensure denaturation of double-stranded template DNA; (2) three cycle steps, which were repeated 40 times: denaturating step at 95°C for 1 min, annealing step of 50°C for 2 min, and extension step of 72°C for 3 min; and (3) a final extension step of 72°C for 5 min to complete unfinished DNA strands. PCR reactions were purified using the QIAquick spin columns following the manufacturer's protocols (Qiagen, Helden, Germany), and the purified products were eluted in 50 µL of sterile water for cycle sequencing.
Most taxa were sequenced at the Molecular Systematics Laboratory in Stockholm (Swedish Museum of Natural History) on an ABI prism 377 automated sequencer (Perkin Elmer, Applied Biosystems, Applera Europe corporation, Stockholm, Sweden) using standard ABI Prism Dye Terminator Cycle Sequencing Ready Reaction kit (Applied Biosystems, Foster City, California, USA). The remaining taxa were electrophoresed for sequencing in an ALFexpress DNA sequencer from Pharmacia (Amersham Pharmacia Biotech, Uppsala, Sweden), using the Amersham Thermo Sequenase Sequencing kit. Both strands of DNA, except for minor parts, were sequenced for most taxa using the same primers as those for amplification. Data files were assembled and edited using the Staden Package (Staden, 1996
). Ambiguous positions were coded using appropriate IUPAC (International Union of Pure and Applied Chemistry) ambiguity symbols to maximize retention of information.
Sequence alignment and phylogenetic analysis
Taxa from the subfamilies Secamonoideae, Asclepiadoideae, and Periplocoideae were chosen to constitute the ingroup, and taxa from the Apocynoideae and Rauvolfioideae were selected as the outgroup. The genera of Secamonoideae sampled in Madagascar include Pervillaea, Secamonopsis, and Secamone. Sequences were aligned using ClustalX version 1.8 (Thompson et al., 1997
) and then visually corrected. Indels were not coded in this study after alignment. The first and last 50 bases of each sequence were eliminated because sequences were not well determined near the primers.
Cladistic analyses were performed using the maximum-parsimony algorithm of the software package PAUP* version 4.0b10 (Swofford, 2002
) treating character states as unordered. For each data matrix (that of matK sequences and that of trnT-trnL sequences), 1000 replicates of random taxon addition were used to find islands of equally parsimonious trees through the tree-bisection-reconnection (TBR) swap algorithm. Only one tree per replicate was saved to minimize time spent swapping on islands of equally parsimonious trees. Complete swapping was performed on all 1000 trees saved in the 1000 replicates (this procedure should have found all possible trees at the length of the most parsimonious tree) with a tree limit of 5000 and MULTREES option on. Rounds of successive approximations weighting (hereafter SW; Farris, 1969
) were added, with characters reweighed according to their rescaled consistency index (CI) based on the best fit of characters on any of the trees. Reasons for using SW have been explained in a previous paper (Civeyrel et al., 1998
). Reweighting steps were repeated until the tree length remained unchanged over two successive iterations. Internal support was assessed using 1000 bootstrap replicates (Felsenstein, 1985
) retaining bootstrap percentages from 1–100%. Bootstrap percentages were compatible with the 50% majority rule consensus tree, with both (1) the simple taxon addition for heuristic search and sampling characters of equal probability but applying weight from SW, and with (2) nearest-neighbor interchange (NNI) swapping but holding 10 trees per step. All illustrated trees used ACCTRAN optimisation. The base weight of 1000 applied in SW was removed for tree presentation.
We employed the incongruence length test (ILD; Farris et al., 1995
) as implemented in PAUP* (partition homogeneity test) to test the null hypothesis that the phylogenetic information generated by the matK and trnT-trnL data sets are homogeneous. Invariant sites were removed for the test (Cunningham, 1997
), and 600 replications were performed with heuristic searches as described before. The strict consensus tree resulting from cladistic analysis of the combined data sets of matK and trnT-trnL is used to discuss relationships within the Secamonoideae. One of the most parsimonious trees, as similar as possible to the consensus tree, was selected to map growth form types and discuss their evolutionary patterns.
Sampling, biomechanics, and anatomy of Secamone sparsiflora
Individuals of Secamone sparsiflora were sampled near the Isalo National Park on privately owned land ("Berny Hotel," Ranohira) in January 2002. We follow the investigative protocols developed elsewhere (Rowe and Speck, 1996
; Isnard et al., 2003a
, b
); the mechanical architecture of a plant growth form is explored by (1) in situ field description of developmental stages, (2) overall geometric and mechanical characteristics (bending tests) along stems and branches of a given individual and from young to older stages of different individuals, (3) geometric and mechanical characteristics of selected individual stems and branches, taking into consideration the exact habit and orientation of axes characteristic of the growth form, and (4) the relative positions and proportions of different tissues from different stages of growth and their potential influence on mechanical properties.
Developmental stages of branching
We observed over 50 plants in natural positions of growth to characterize the habit and orientation of young, mature, and old axes. Three developmental stages were established in terms of position and orientation of the main stem: stage I, young apical axes, ultimate branches bearing leaves positioned either at the apex of the plant or developing from axillary buds; stage II, upright or unstable axes bearing apical stems of developmental stage I or other penultimate upright branches; stage III, old procumbent stems bearing axes of stage II and reiterative upright branches.
Bending tests
Three individual healthy plants were sampled to represent young self-supporting axes to older procumbent stems. Main leading stems and their branches were pruned from these plants and kept hydrated for a maximum of 2 days prior to bending tests. The distances from the base of the plant to each tested segment were recorded along the main axis and branches. Field protocols are described in detail elsewhere (Rowe and Speck, 1996
; Speck and Rowe, 1999
; Gallenmüller et al., 2001
; Isnard et al.,2003a
, b
). Three- and four-point bending tests were carried out on 77 stem segments using a portable bending apparatus. Tested axes were placed horizontally on the apparatus or, if slightly curved, were orientated with the convex surface facing downwards in three-point bending. A series of weights was applied manually to a pannier suspended from the center of the tested stem in three-point bending and to a two-armed pannier in four-point bending. Deflections were observed via a dissecting microscope mounted on the apparatus. Span distances of tested stems and weight increments were varied according to the size and resistance of the material tested and appropriate precautions were taken to avoid measuring shear in three-point bending (Vincent, 1990
; Niklas, 1992
; Rowe and Speck, 1996
). A span test was carried out to determine the minimum span-to-depth ratio required, below which the bending experiment would likely be significantly affected by shear. This indicated a minimum ratio of 20 for a woody axes. All segments tested with a ratio lower than 20 were eliminated from the analysis. Stems showing significant tapering above 12% from the mean diameter were also eliminated as were stems showing evidence of decay, damage or internal fracture. A total of 46 segments, of the 77 initially tested, were selected for data analysis. All stem segments tested were stored in alcohol during fieldwork and then transferred to FAA (1 part acetic acid 80%, 25 parts formalin 40%, 218 parts ethanol 50%).
Flexural stiffness, EI (N · mm2) was calculated from EI = l3/48b, where l (mm) is the length of the span and b is the slope derived from the force deflection curve where force, N, was plotted against maximum deflection (mm). Second moment of area, I (mm4), was established from the diameters of tested segments in three to five positions along the segment, and an elliptical model was used to determine the mean axial second moment of area I of each segment in relation to direction of the force applied (Vincent, 1990
; Niklas, 1992
). Structural Young's modulus, Estr (MN/m2), was calculated from the measured flexural stiffness EI and the mean axial second moment of area I by the formula Estr = EI/I.
Stems and branches were analyzed to determine general patterns of geometric and mechanical properties during ontogeny and along four representative axes differing in orientation. The latter included a young upright axis (4 contiguous segments), an inclined axis (3 contiguous segments), a procumbent axis (3 contiguous segments) and an older procumbent stem with reiterative upright axes (4 contiguous segments).
Anatomy
The macroanatomical organization was described in cross-sectional surfaces of 46 tested segments. These were first polished with fine-grade abrasive discs, soaked in 25% HCl for 30 s and irrigated in a solution of 3% phloroglucinol in 90% ethanol, staining lignified tissues red. Areas of bark, wood and pith were then delimited using a camera lucida, and their contributions to axial second moment of area were calculated from digitized outlines with an Optimas (Media Cybernetics, Silver Spring, Maryland, USA) macro kindly provided by Tancrède Almeras (UFR d'Arboriculture Fruitière, INRA, France).
The microanatomical study (wood structure) included 12 segments from four different plants to investigate trends in wood density in stages II and III. Specimens were chosen in which contributions of wood to second moment of area were greater than those of bark. Segments were softened in 40% hydrofluoric acid for 24 h then rinsed in running water for 24 h. Cross-sections 20–40 µm thick were cut on a wood microtome and stained with safranin and fast green. Cell lumens and wall thickness were measured on digital images taken from the center to the periphery of the wood cylinder using the software Optimas. Each sample area was standardized as a rectangle of 1000 x 800 µm and manually placed over well sectioned portions of the entire cross-section where cell walls and lumens were not oblique, torn or at varying levels in the plane of focus. Entire cross-sections were sampled via the rectangular areas at intervals of approximatly 1000 µm from the center to the periphery of the section. Measurements were carried out at varying magnifications necessary for resolving accurately the different sized surfaces: (1) vessel lumen density, i.e., total lumen area per entire rectangle (at 250x), (2) tracheid or fiber wall density, i.e., total wall area per unit of surface of 3–6 neighbouring fiber elements, based on means of five samples per rectangle (at 1200x), (3) wood ray density, i.e., total wood ray width per surface width, based on means of three samples per rectangle (750x). Wood density parameters were investigated in terms of (1) their variation across the stem cross-section and (2) as mean values of each parameter for each entire cross-section.
Means of geometrical and biomechanical parameters were statistically compared using a two-tailed t test to reveal significant variations from stage I to stage II for macroanatomical organization and then from stage II to stage III for both macroanatomical and microanatomical organization. Pearson correlation analyses were used to investigate changes in wood structure during development by comparing the means of each density parameter with the structural Young's modulus. Cross-sectional areas of tissues, cell lumens, and cell walls represent a proxy for physical density and provide developmental characteristics related to density that would be informative for developmental and systematic comparisons with other shrubs and lianas.
RESULTS
Molecular phylogeny and growth form evolution
matK analysis
Length of the sequences varied from 591 (for Ischnolepis graminifolia incomplete sequence) to 1481 bp (Fockea capensis). After alignment of the 44 sequences of matK, the matrix contained 1548 characters, including 15 insertions and deletions (indels), which occur as triplets (3–21 bp); none of these indels were coded in this analysis. The matrix included 146 informative characters (9.4%). Complete swapping from trees obtained after 1000 replicates gave 616 equally parsimonious trees 546 steps long, with CI = 0.8132 (0.6483 excluding uninformative characters) and RI = 0.8431. Two steps of SW analysis resulted in 168 trees with CI = 0.9494 (0.88515 excluding uninformative characters) and RI = 0.9567 (base weight of 1000 applied in SW). The strict consensus tree after complete swapping without weighting was identical to that of SW.
The monophyly of the three subfamilies represented in the ingroup is strongly supported with a bootstrap support (BS) of 100% for the Asclepiadoideae and Periplocoideae, and 88% for the Secamonoideae (Fig. 1). The Asclepiadoideae appears to be the sister group of the Secamonoideae (BS of 100%). Within the Secamonoideae, the monophyly of the two genera Pervillaea and Secamonopsis appears well supported with BS of 71% and 96%, respectively. The monophyly of the main genus Secamone in this analysis remains unclear due to the polytomy between the clade Secamonopsis, Secamone volubilis, S. oleaefolia, and a not fully resolved clade including 18 species of Secamone, which is well supported (BS = 92%). Within this latter clade, five monophyletic groups of taxa are well supported: (1) S. bosseri and S. cristata (BS = 99%), (2) S. buxifolia and S. ligustrifolia (BS = 100%), (3) S. glaberrima, S. unciata, S. sparsiflora, and S. tenuifolia (BS = 98%), in which (4) S. sparsiflora and S. tenuifolia are sister species (BS = 91%), and (5) a poorly resolved clade including S. cloiselii, S. ecoronata, S. minutifolia, and S. urceolata (BS = 85%) (Fig. 1).
|
The monophyly of the three subfamilies is well supported with a bootstrap support (BS) of 100% for the Periplocoideae, 82% for the Secamonoideae and 80% for the Asclepiadoideae with Asclepiadoideae as a sister group to the Secamonoideae (BS of 96%) (Fig. 1). Within the Secamonoideae, only the monophyly of Secamonopsis is strongly supported (BS of 95%) (Fig. 1). The position of Pervillaea tomentosa in a polytomy with the Secamone-Secamonopsis clade and the clade of the other Pervillaea species makes the validity of the genus Pervillaea questionable. The monophyly of the genus Secamone is also unclear in this analysis because of the significant polytomy between taxa of Secamone and the genus Secamonopsis. Within the genus Secamone, two clades appear well supported: (1) a clade including S. buxifolia, S. ligustrifolia, S. sparsiflora, S. tenuifolia, and S. unciata (BS = 86%), and (2) a clade including S. castanea, S. cloiselii, S. ecoronata, S. minutifolia, and S. urceolata (BS = 86%) (Fig. 1).
Data incongruence between data sets
Both data sets (matK and trnT-trnL) are homogeneous in terms of phylogenetic relationships (ILD test, P = 0.18; Farris et al., 1995
). The amount of incongruence between the data sets is therefore trivial, and an analysis of the combined data is appropriate. As a result, the discussion will be based only on the strict consensus tree from the SW analysis of the combined data sets (Fig. 2) to explore the relationships within the Secamonoideae. One of the most parsimonious trees from the SW analysis of the combined data sets (Fig. 3) will be used to discuss character transitions of growth forms.
|
|
The monophyly of the three subfamilies related to the ingroup is strongly supported with a BS = 100% for the Periplocoideae, 100% for the Asclepiadoideae and 93% for the Secamonoideae, which is the sister group of the Asclepiadoideae (BS = 100%) (Fig. 2). Within the Secamonoideae, a total of 11 groups can be distinguished with BS > 90%, including the following seven clades: (1) the group Pervillaea including P. decaryi, P. venenata, and P. phillipsonii is strongly supported with a BS = 99%, whereas the clade including P. tomentosa is less well supported (73%); (2) the Secamonopsis clade is strongly supported with BS = 100%, but its relationship with the other genera is not resolved; (3) a clade including Secamone cristata and S. bosseri (BS = 98%); (4) a clade including Secamone ligustrifolia and S. buxifolia (BS = 100%); (5) a clade including S. tenuifolia and S. sparsiflora (BS = 93%); (6) a clade including S. castanea, S. cloiselii, S. ecoronata, S. minutifolia, and S. urceolata (BS = 99%); and (7) a clade including S. oleaefolia and S. volubilis (BS = 94%) (Fig. 2).
Appearance of shrub-like growth forms in Secamonoideae
Optimizing growth forms onto one of the most parsimonious trees from the combined study (Fig. 3) indicates that twining lianas characterize most members of the Periplocoideae, Asclepiadoideae, and Secamonoideae. In the latter, the molecular phylogeny indicates five independent transitions from lianoid to "shrub-like" growth forms (Fig. 3). These transitions occur in different genera and clades of the Secamonoideae and appear to be confined to species in Madagascar. Secamone sparsiflora and its sister species S. tenuifolia show a very similar "shrub-like" habit to S. minutifolia, which is nested in another clade within the Secamonoideae among twining lianas. These three species co-occur in the same habitat on the Isalo plateau in southwest central Madagascar. The "shrub-like" clade including S. sparsiflora and S. tenuifolia is sister group to a lianescent clade including S. glaberrima and S. unciata, and this group of four species is sister to a further clade of two lianas S. ligustrifolia and S. buxifolia. Overall, similar "shrub-like" growth forms with upright young stages and older procumbent stems are found in more distant clades represented by Pervillaea venenata and Ischnolepis graminifolia (Periplocoideae). A different and more diminutive type of prostrate "shrub-like" growth form is seen in S. falcata, which is sister taxon to the clade of larger bodied, more upright "shrubs" of S. sparsiflora and S. tenuifolia. A very similar type of prostrate growth has been observed for Secamonopsis microphylla (Civeyrel and Rowe, 2001
; Speck et al., 2003
), which is sister species to Secamonopsis madagascariensis in the selected tree (Fig. 3). These two species form a clade, which is sister to Secamone and which, according to the results of the strict consensus tree (Fig. 2), has an unresolved position within the Secamonoideae. Interestingly, the more diminutive and prostrate "shrubs," S. falcata and S. microphylla, although phylogenetically separated, co-occur in an open chaparral-type vegetation at Tulear in coastal southwest Madagascar. These two species were not observed to occur with the larger more erect procumbent stems co-occurring on the Isalo plateau: S. minutifolia, P. venenata, S. sparsiflora, S. tenuifolia, and I. graminifolia.
Mechanical architecture of a "shrub-like" growth form: Secamone sparsiflora
Habit and orientation of main stem and branches
This species has a woody shrub-like growth form with leading axes reaching 1–3 m in height (Fig. 4A, B). The species produces two to three orders of branching (Fig. 4C, D). Ultimate branches bear linear to acicular slender leaves. Observations of over 50 individuals in the field indicate a complex development in terms of habit and orientation of young, mature and old axes. Young leading axes and branches are erect and self-supporting (Fig. 4A, B). Older stems assume a procumbent orientation on the ground and can further produce erect branches via reiterative leaders (Fig. 4A, C). A rare instance of twining behavior was recorded in the young branch of one individual out of over 50 plants observed (Fig. 4E).
|
|
|
; Speck and Rowe, 1999
). This graph visualizes relative changes in EI between stages of development relative to the first formed youngest stage. Values of EI are distributed just above or on the neutral line for upright or inclined segments (stage II) (Fig. 5A); older inclined or procumbent segments (oldest stage II and stage III) are distributed on the neutral line (Fig. 5A). This overall trend in EI is not typical of self-supporting plants in which older stages fall above the neutral line nor of many specialized lianoid climbers tested in which final values of old flexible stems fall below the neutral line (Speck, 1994
; Speck and Rowe, 1999
). Lowest values of Estr occur in young shoots (stage I) and old procumbent segments (stage III) with values of 2704 and 3906 MN/m2 respectively for the most flexible segments within these stages (Fig. 5B). Highest values of Estr are found in upright and inclined segments (stage II) with values of 7908 MN/m2 for the most rigid segment (Fig. 5B). Means of Estr show an initial increase from young shoots (stage I) to upright and inclined older stems (stage II) (t = 4.03, P < 0.01) (Table 1); there is a moderate decrease from the intermediate stage II to older procumbent axes (stage III) (t = 1.69, P > 0.05). Regarding overall stem development in terms of stem diameter rather than stem position, the group of 10 largest axes of stages II and III have a more marked decrease in Estr (circle in Fig. 5B). This suggests that late changes in Estr are related more to size and extent of secondary growth than to actual position and orientation during the transition from upright/inclined to procumbent stems.
Geometric and mechanical properties of different stem orientations
Geometric, mechanical, and developmental parameters are compared along four axes differing in orientation and architecture (Fig. 6): a young upright stem, an inclined stem, a procumbent stem, and an old procumbent stem with reiterative upright axes. All stems show marked increases towards the base in I and EI except for the entirely procumbent axis (Fig. 6A, B). This increase is most marked in the inclined and reiterative axes (Fig. 6B). Whatever the orientation of the stems, there is an initial increase of Estr below the apex and then a decrease at some point towards the base (Fig. 6C). These trends are similar to the data showing marginally higher values of Estr for the intermediate developmental stage II (Fig. 5B, Table 1).
|
|
|
|
|
Molecular phylogeny: strict consensus tree
The combined analysis of matK and trnT-trnL supports the monophyly of the three subfamilies: Secamonoideae, Asclepiadoideae, and Periplocoideae. The Secamonoideae is sister group of the Asclepiadoideae (Fig. 2), and the Asclepiadoideae and Periplocoideae are both strongly supported with bootstrap values of 100%. In this study, the Secamonoideae is strongly supported with a BS = 93%. In previous analyses, the subfamily was less well supported with bootstrap values of 71% (Civeyrel et al., 1998
) and 82% (Civeyrel and Rowe, 2001
), probably because of the larger number of outgroups and the inclusion of only one gene, matK.
Within the Secamonoideae, 11 clades are strongly supported with bootstrap values of over 90% (Fig. 2), including seven groups, which are discussed next.
Pervillaea
The group of three species of Pervillaea, P. decaryi, P. venenata, and P. phillipsonii is strongly supported with a bootstrap value of 99%. The inclusion of P. tomentosa, however, lowers the bootstrap value for the genus Pervillaea from 100% (Civeyrel and Rowe, 2001
) to 73%. This genus was reinstated as a separate genus by Klackenberg (1995)
who found that the type of P. tomentosa has characteristics that distinguish it from the genus Toxocarpus. Additional studies have reassigned taxa to Pervillaea from the genus Menabea (Klackenberg, 1996a
). Morphological studies have also shown close similarities between Pervillaea and Calyptranthera (Klackenberg, 1996b
, 1997
). Adding taxa from Toxocarpus and Calyptranthera to the phylogeny would test the validity of the genus Pervillaea, but the lack of fresh material from these taxa makes this difficult.
Secamonopsis
The species of Secamonopsis are strongly supported with a BS = 100%, but the relationships of this clade with other genera are not resolved. This result agrees with previous studies (Civeyrel and Rowe, 2001
), but the addition of the intergenic spacer has not strengthened the support for the group.
Secamone cristata group
The morphological group termed by Klackenberg S. cristata (1992a)
, including S. bosseri and various subspecies of S. cristata, characterized by long corolla tubes and dense inflorescences, is strongly supported with a BS = 98%.
Secamone ligustrifolia group
The morphological group termed S. ligustrifolia, including S. ligustrifolia and S. buxifolia (Klackenberg, 1992a
), characterized by large and fleshy coronal lobes with rounded back and a stigma head with large lower broader part, is strongly supported with a bootstrap value of 100%.
Secamone tenuifolia group
The S. tenuifolia group of Klackenberg (1992a)
, including S. tenuifolia and S. sparsiflora, is morphologically characterized by thick and finely puberulous corolla lobes and is also strongly supported with a bootstrap value of 93%. In this study, these two taxa are closely related to S. glaberrima, that is nested in this morphological group and S. unciata, which is different in habit and flower morphology from other taxa of the S. tenuifolia group. These four species show clear similarities in their pollinaria (Civeyrel, 1994
; Civeyrel and Rowe, 2001
).
Secamone castanea, S. cloiselii, S. ecoronata, S. minutifolia, and S. urceolata
The group including the five species, S. castanea, S. cloiselii, S. ecoronata, S. minutifolia, and S. urceolata, is strongly supported with a BS = 99%. Secamone castanea and S. urceolata were assigned to the S. humbertii group (Klackenberg, 1992a
), characterized by having pouchlike structures on the stamens. The other three species, S. cloiselii, S. ecoronata, and S. minutifolia, do not show clear similarities (Klackenberg, 1992a
), but the compact pouchlike structure of the stamens appears very similar to those of S. castanea and S. urceolata (Civeyrel, 1994
; Civeyrel and Rowe, 2001
).
Secamone oleaefolia and S. volubilis
Secamone oleaefolia and S. volubilis form a strongly supported clade with a BS = 94%. They do not share any clear morphological or biogeographical similarities. The first is distributed along the east coast and central plateau of Madagascar, and the second is endemic to the island of La Reunion. The group forms a polytomy with Secamonopsis and a clade including the other Malagasy species of Secamone. The validity of the main genus Secamone has been questioned in previous studies (Civeyrel and Rowe, 2001
) because of the position of S. volubilis as part of a polytomy with Secamonopsis and a clade including Malagasy Secamone. The addition of S. oleaefolia in this analysis creates a new clade with S. volubilis but forming a polytomy with Secamonopsis and the clade of Malagasy Secamone, making the validity of this latter genus unclear.
In conclusion, combining the matK and intergenic spacer trnT-trnL to investigate the relationships within the Secamonoideae provided a better resolution than previous studies (Civeyrel et al., 1998
; Civeyrel and Rowe, 2001
). Although the monophyly of Pervillaea and that of Secamonopsis are strongly supported, that of Secamone is still equivocal mainly because of the position of S. oleaefolia and S. volubilis. Future studies should include more informative genes and additional taxa with ambiguous systematic positions such as Calyptranthera and Toxocarpus.
Implications of the biomechanics of Secamone sparsiflora on growth form evolution
Secamone sparsiflora is a procumbent shrub-like form nested within a clade of lianas (Fig. 3). The biomechanical characteristics of Secamone sparsiflora are neither typical of self-supporting trees and shrubs nor of non-self-supporting lianas. Self-supporting young axes in Secamone sparsiflora are structurally comparable and possibly homologous to young searchers of lianas in the same clade such as those of S. ligustrifolia and S. buxifolia. These last two species also show relatively high values of Estr caused by the development of an initial dense, stiff wood (R. Lahaye et al., unpublished data). In older stages of development, S. sparsiflora shows only a slight decrease in Estr; on the other hand, the decrease in Estr is much more marked in lianas of the Secamonoideae (Speck et al., 2003
). Wood development in S. sparsiflora therefore appears to retain juvenile characteristics (neoteny) but also shows a significant decrease in tracheid or fiber wall area density during ontogeny that is correlated with a moderate decrease in Estr in stage III stems (the oldest) compared to stage II stems. In S. sparsiflora, Estr remains relatively high compared to many other tropical lianas (Speck, 1994
; Rowe and Speck, 1996
; Speck et al., 1996
; Köhler et al., 2000
), but this does not appear to sustain an upright habit for the entire growth trajectory. Self-supporting trees show a continuous increase of Estr during ontogeny, whereas self-supporting shrubs show an initial phase of increasing Estr followed by a plateau with a nearly constant Estr in old ontogenetic stages (Speck, 1991
, 1994
; Speck and Rowe, 1999
). As explained earlier, Estr of S. sparsiflora does not increase in older stages of development and only shows a slight decrease; it is therefore apparently not optimized for either self-supporting or lianoid growth.
The derived procumbent habit of S. sparsiflora cannot be interpreted as a complete "reversal" to a fully self-supporting habit but rather as a modification of an ancestral lianoid mechanical architecture that has produced a partially self-supporting growth form. Interestingly, evidence of relictual lianoid behavior in S. sparsiflora is suggested by one specimen from >50 individuals observed: one branch of a procumbent shrub was weakly twined around its main leading axis (Fig. 4E). This contrasts with many small-bodied lianas of the subfamily growing in open semi-arid chaparral habitats, which commonly twine and form trellises on themselves.
Twining lianas often have significant transitions in mechanical properties of the stem when the initial self-supporting phase of young individuals, or young searchers of mature plants have attached to a support. The transition is characterized by a marked decrease of Estr in older stems, often down to about 10% of the original value found in young stems (Speck, 1994
; Rowe and Speck, 1996
, 1998
; Speck and Rowe, 1999
). On the other hand, in S. sparsiflora, Estr of stage III decreases to only 85% of that from stage II. In many lianas, the decrease in Estr is often caused by profound anatomical changes in organization and mechanical properties between young and older stages, commonly known as anomalous wood development and described for many tropical liana species (Fisher and Ewers, 1989
, 1992
; Caballé, 1993
, 1998
; Rowe and Speck, 1996
, 2004
; Speck et al., 1996
). Lianas not having highly modified anomalous development can produce a more compliant wood in older stems with numerous vessels and less dense components of the wood including thinner tracheid or fiber cell walls (Hallé et al., 1978
; Caballé, 1986
, 1993
, 1998
; Putz and Holbrook, 1991
; Rowe and Speck, 1996
, 1998
; Speck and Rowe, 1999
). Initial observations indicate that lianoid species of Secamone show this latter type of non-anomalous wood development in older compliant stages, and a similar type is noted for the lianoid Secamonopsis madagascariensis (Civeyrel and Rowe, 2001
; Speck et al., 2003
). In the liana Secamone ligustrifolia, the stiffness in young stages is correlated with production of a dense wood (68% tracheid or fiber wall area density), which becomes less dense in older stages (58% tracheid or fiber wall area density) (R. Lahaye et al., unpublished data). The development of S. sparsiflora appears to show a modification of the type of lianoid developmental plan found in S. ligustrifolia. Production of dense wood facilitates an upright or ascending orientation during the initial phase of growth with heights up to 2–3 m. This phase is followed by a slight decrease in both wood density and structural Young's modulus occurring at the base of older ascending axes and procumbent stems that had exceeded their critical height.
Heterochrony, growth habit change, and stem biomechanics
Heterochronic processes have been previously considered as mediating evolutionary developmental changes of plant architectures and growth forms (e.g., Carlquist, 1962
; Bateman, 1994
; Bateman et al., 1998
; Civeyrel and Rowe, 2001
; Cronk, 2001
, 2002
; Speck et al., 2003
). In general developmental terms and in comparison with lianoid species such as S. ligustrifolia, S. sparsiflora could be interpreted as having a prolonged juvenile stage with dense wood production conferring a relatively long-lasting, self-supporting stage prior to instability and procumbence (Fig. 11). A putative ancestral form could be hypothesized as a liana similar to S. ligustrifolia or other lianoid members of Secamone. Despite an early self-supporting phase, the biomechanical data and wood density values, as well as the retention of a relatively narrow stem diameter, suggest that certain lianoid developmental patterns imposed mechanical constraints on growth habit evolution, particularly the transition to self-supporting architectures. In the lianoid growth trajectory, a relatively short, young, self-supporting growth phase is followed by a climbing phase and a production of compliant older stems (Fig. 11). This development also produces extremely long axes relative to stem diameter and for many lianoid groups, mature stems can be tens of meters long (e.g., S. ligustrifolia) or even hundreds of meters (Cremers, 1973
, 1974
; Caballé, 1986
, 1998
). Heterochronic modification of the developmental trajectory of a lianoid architecture could explain the marked change in overall size and length of the derived growth forms observed in procumbent shrubs such as S. sparsiflora. A neotenic developmental change in the production of dense juvenile stiff wood for a large part of the ontogenetic trajectory would have the effect of producing a relatively long self-supporting phase. In S. sparsiflora, our studies show that a less dense type of wood is produced toward the end of the development of upright stems and possibly engenders instability, observed in ascending, angled or leaning axes, and then procumbent older axes. Procumbent axes of S. sparsiflora do not exceed 3–4 m in length. This "shortening" results in procumbent shrubs rather than lianas with long stems and could be explained by a precocious offset of growth (progenesis) of apical meristems. Such processes are consistent with an interpretation of derived procumbent shrubs resulting from a combination of paedomorphic developmental changes resulting in a less complex and overall smaller plant body compared with a putative lianoid ancestor.
|
Island woodiness and island growth habit evolution
The presence of shrub-like growth forms within groups that are predominantly vines and lianas has been documented in a wide range of angiosperms such as Aristolochiaceae (Speck et al., 1997
, 2003
), Convolvulaceae (Carine et al., 2004
), Cucurbitaceae (Olson, 2003
) as well as in the Gnetales (Fisher and Ewers, 1995
). These exceptions have been recorded in continental areas and more rarely in oceanic islands. Radiations of floras in island situations are better known for the appearance of larger bodied and/or upright growth forms (Darwin, 1860
; Wallace, 1878
; Carlquist, 1965
, 1974
, 1980
; Knox et al., 1993
; Knox and Palmer, 1995
; Niklas, 1997
; Baldwin et al., 1998
; Panero et al., 1999
; Moore et al., 2002
; Mort et al., 2002
). Island-woodiness has been largely documented for herbaceous families that have evolved woody self-supporting growth forms from continental herbaceous ancestors (Campanulaceae: Knox et al., 1993
; Asteraceae: Knox and Palmer, 1995
; Kim et al., 1996
; Panero et al., 1999
; Carlquist, 2001
; Moore et al., 2002
; Boraginaceae: Böhle et al., 1996
; Crassulaceae: Mort et al., 2002
). The presence of shrub-like woody plants derived from woody climbing plants in Madagascar presents a somewhat different type of growth form change, possibly involving changes in organization related to a radiation in an isolated island situation.
The low variation of noncoding sequences among the Malagasy Secamonoideae (7.9% of the potentially parsimony-informative sites for the noncoding sequences trnT-trnL, and 9.7% for the coding gene matK) might also indicate a relatively recent radiation on the island and a rapid speciation. Other molecular studies of island plant groups (Knox and Palmer, 1995
; Böhle et al., 1996
; Kim et al., 1996
; Panero et al., 1999
; Moore et al., 2002
; Mort et al., 2002
) have documented similar low molecular variation among related island taxa, even in the face of strong morphological divergence (Baldwin et al., 1998
). The Secamonoideae have colonized most of the phytogeographical regions in Madagascar (Klackenberg, 1992a
). Shrub-like Secamonoideae occur independently in different clades and apparently only in the arid/semi-arid climate of southwest and southwest central Madagascar. Secamone sparsiflora is endemic to the Isalo plateau in southwest central Madagascar where two similar conspecific "shrub-like" species also occur: S. tenuifolia (sister species of S. sparsiflora) and S. minutifolia, which is nested in another lianescent clade. Transitions to procumbent shrubs in this area are also seen in other groups of the Secamonoideae with Pervillaea venenata, and in the Periplocoideae with Ischnolepis graminifolia. The type of growth form is similar in all five species with a young upright phase followed by a leaning and then procumbent orientation producing vertical reiterative shoots. The Isalo plateau is geographically isolated by approximately 100 km of semi-desert plains (Humbert, 1955
; White, 1983
). The area is characterized by rocky terrain and eroded sandy soils with vegetation restricted to localized sandy exposures. This open xeric environment could have favored evolution from woody vines to shrub-like growth forms possibly as a result of (1) reduced biotic competition in a more open environment, and (2) few and relatively small potential hosts for lianoid growth strategies.
In conclusion, the shrub-like S. sparsiflora could have evolved from a lianescent ancestor by relatively few but nevertheless profound morphological changes. If this evolutionary scenario is correct, relatively simple changes in the developmental plan of putative ancestral lianas in the Secamonoideae produced a range of types of procumbent shrubs largely differing in terms of size, growth form, and mechanical properties. This study complements recent research in which biomechanical and developmental characteristics have been explored to elucidate possible mechanisms and constraints of underlying patterns of growth form evolution in plants (Isnard et al., 2003b
; Speck et al., 2003
; Rowe and Speck 2004
).
FOOTNOTES
The authors thank M. Berny from Isalo, Madagascar, for permission to sample plants on his land. We thank Jens Klackenberg and Mari Källersjö, National Museum of Natural History, Stockholm, for collaboration in sampling herbarium specimens and molecular analyses. This research was supported by an ECLIPSE programme of the Center National de la Recherche Scientifique, France (2002); a HIGHLAT grant from the European Commission's programme for "Improving the Human Research Potential and the Socio-economic Knowledge Base" (IHP) for visits to the Museum of Natural History at Stockholm (2002–2003); an ATUPS grant from the University of Toulouse III (2003); an award to R.L. from the municipality of Gondrin, France, for fieldwork in Madagascar (2002); and a PROCOPE exchange grant awarded to N.R. and T.S (2000–2002). All of these financial aids and assistance are gratefully acknowledged. 
5 Author for correspondence (e-mail: lahaye{at}cict.fr
)
LITERATURE CITED
Baas P. E. Werker A. Fahn 1983 Some ecological trends in vessel characters. International Association of Wood Anatomists Bulletin, New Series 4: 141-159
Baldwin B. G. D. J. Crawford J. Francisco-Ortega S.-C. Kim T. Sang T. F. Stuessy 1998 Molecular phylogenetic insights on the origin and evolution of oceanic island plants. In D. E. Soltis, P. S. Soltis, and J. J. Doyle [eds.], Molecular systematics of plants II: DNA sequencing, 410–441. Kluwer, Norwell, Massachusetts, USA
Bateman R. M. 1994 Evolutionary-developmental change in the growth architecture of fossil rhizomorphic lycopsids: scenarios constructed on cladistic foundations. Biological Reviews 69: 527-597[CrossRef]
Bateman R. M. 1999 Integrating molecular and morphological evidence of evolutionary radiations. In P. M. Hollingsworth, R. M. Bateman, and R. J. Gornall [eds.], Molecular systematics and plant evolution, 432–471. Taylor and Francis, London, UK
Bateman R. M. W. A. DiMichele 1994 Saltational evolution of form in vascular plants: a neoGoldschmidtian synthesis. In D. S. Ingram and A. Hudson [eds.], Shape and form in plants and fungi, 61–100. Academic Press, London, UK
Bateman R. M. P. R. Crane W. A. DiMichele P. R. Kenrick N. P. Rowe T. Speck W. E. Stein 1998 Early evolution of land plants: phylogeny, physiology and ecology of the primary terrestrial radiation. Annual Review of Ecology and Systematics 29: 263-292[CrossRef][Web of Science]
Bhambie S. 1972 Correlation between form, structure and habit in some lianas. Proceedings of the Indian Academy of Sciences 75: 246-256
Böhle U.-R. H. H. Hilger W. F. Martin 1996 Island colonization and evolution of the insular woody habit in Echium L. (Boraginaceae). Proceedings of the National Academy of Sciences, USA 93: 11740-11745
Caballé G. 1986 Sur la biologie des lianes ligneuses en forêt gabonaise. Montpellier, France. Ph.D. dissertation, University Montpellier II Sciences et Technique du Languedoc, Montpellier, France
Caballé G. 1993 Liana structure, function and selection: a comparative study of xylem cylinders of tropical rainforest species in Africa and America. Botanical Journal of the Linnean Society 113: 41-60[CrossRef]
Caballé G. 1998 Le port autoportant des lianes tropicales: une synthèse des stratégies de croissance. Canadian Journal of Botany 76: 1703-1716[Web of Science]
Carine M. A. S. J. Russel A. Santos-Guerra J. Francisco-Ortega 2004 Relationships of the Macaronesian and Mediterranean floras: molecular evidence for multiple colonizations into Macaronesia and back-colonization of the continent in Convolvulus (Convolvulaceae). American Journal of Botany 91: 1070-1085
Carlquist S. 1962 Ontogeny and comparative wood anatomy of thorns of Hawaiian Lobeliaceae. American Journal of Botany 49: 413-419[CrossRef][Web of Science]
Carlquist S. 1965 Island life: a natural history of the islands of the world. Natural History Press, Garden City, New York, USA
Carlquist S. 1974 Island biology. Colombia University Press, New York, New York, USA
Carlquist S. 1980 Hawaii, a natural history. Geology, climate, native flora and fauna above the shoreline. Pacific Tropical Botanical Garden, Honolulu, Hawaii, USA
Carlquist S. 2001 Wood anatomy of the endemic woody Asteraceae of St. Helena. I. Phyletic and ecological aspects. Botanical Journal of the Linnean Society 137: 197-210[CrossRef]
Chase M. W. S. Knapp A. V. Cox J. J. Clarkson Y. Butsko J. Joseph V. Savolainen A. S. Parokonny 2003 Molecular systematics, GISH and the origin of hybrid taxa in Nicotiana (Solanaceae). Annals of Botany 92: 107-127
Civeyrel L. 1994 Variation et évolution des types polliniques du genre Secamone (Asclepiadaceae, Secamonoideae). Compte rendu de l'Académie des Sciences de Paris, Sciences de la Vie 317: 1159-1165
Civeyrel L. N. P. Rowe 2001 Phylogenetic relationships of Secamonoideae based on the plastid gene matK, morphology and biomechanics. Annals of the Missouri Botanical Garden 88: 583-602[CrossRef][Web of Science]
Civeyrel L. A. Le Thomas K. Ferguson M. W. Chase 1998 Critical reexamination of palynological characters used to delimit Asclepiadaceae in comparison to the molecular phylogeny obtained from plastid matK sequences. Molecular Phylogenetics and Evolution 9: 517-527[CrossRef][Web of Science][Medline]
Cremers G. 1973 Architecture de quelques lianes d'Afrique tropicale. Candollea 28: 249-280
Cremers G. 1974 Architecture de quelques lianes d'Afrique tropicale 2. Candollea 29: 57-110
Cronk Q. C. B. 2001 Plant evolution and development in a post-genomic context. Nature Reviews Genetics 2: 607-619[CrossRef][Web of Science][Medline]
Cronk Q. C. B. 2002 Perspectives and paradigms in plant evo-devo. In Q. C. B. Cronk, R. M. Bateman, and J. A. Hawkins [eds], Developmental genetics and plant evolution, 1–14. Taylor & Francis, London, UK
Cunningham C. W. 1997 Can three incongruence tests predict when data should be combined?. Molecular Biology and Evolution 14: 733-740[Abstract]
Darwin C. 1860 On the origin of species. London, UK
Endlicher S. 1838 Genera plantarum: Asclepiadeae. Vienna, Austria
Endress M. E. P. V. Bruyns 2000 A revisited classification of the Apocynaceae s.l. The Botanical Review 66: 1-56
Endress M. E. B. Sennblad S. Nilsson L. Civeyrel M. W. Chase S. Huysmans E. Grafström B. Bremer 1996 A phylogenetic analysis of Apocynaceae s.str. and some related taxa in Gentianales: a multidisciplinary approach. Opera Botanica Belgica 7: 59-102
Farris J. S. 1969 A successive approximations approach to character weighting. Systematic Zoology 18: 374-385[CrossRef]
Farris J. S. M. Källersjö A. G. Kluge C. Bult 1995 Testing significance of incongruence. Cladistics 10: 315-319[CrossRef][Web of Science]
Felsenstein J. 1985 Confidence limits on phylogenies: an approach using bootstrap. Evolution 39: 783-791[CrossRef][Web of Science]
Fisher J. B. F. W. Ewers 1989 Wound healing in stems of lianas after twisting and girdling injuries. Botanical Gazette 150: 251-265[CrossRef]
Fisher J. B. F. W. Ewers 1992 Xylem pathways in liana stems with variant secondary growth. Botanical Journal of the Linnean Society 108: 181-202
Fisher J. B. F. W. Ewers 1995 Vessel dimensions in liana and tree species of Gnetum (Gnetales). American Journal of Botany 82: 1350-1357[CrossRef][Web of Science]
Forster P. I. K. Harold 1989 Secamone R. Br. (Asclepiadaceae: Secamonoideae) in Australia. Austrobaileya 3: 69-78
Gallenmüller F. U. Müller N. P. Rowe T. Speck 2001 The growth form of Croton pullei (Euphorbiaceae)—functional morphology and biomechanics of a neotropical liana. Plant Biology 3: 50-61[CrossRef][Web of Science]
Gentry A. G. 1991 The distribution and evolution of climbing plants. In F. E. Putz and H. A. Mooney [eds.], The biology of vines, 3–49. Cambridge University Press, Cambridge, UK
Goyder D. J. 1992 Secamone (Asclepiadaceae subfam. Secamonoideae) in Africa. Kew Bulletin 47: 437-474[CrossRef]
Hallé F. R. A. A. Oldeman P. B. Tomlinson 1978 Tropical trees and forests: an architectural analysis. Springer-Verlag, Berlin, Germany
Hegarty E. E. G. Caballé 1991 Distribution and abundance of vines in forest communities. In F. E. Putz and H. A. Mooney [eds.], The biology of vines, 313–335. Cambridge University Press, Cambridge, UK
Hu J.-M. M. Lavin M. F. Wojciechowski M. J. Sanderson 2000 Phylogenetic systematics of the tribe Millettieae (Leguminosae) based on chloroplast trnK/matK sequences and its implications for evolutionary patterns in Papilionoideae. American Journal of Botany 87: 418-430
Humbert H. 1955 Les territoires phytogéographiques de Madagascar. L'Année Biologique 31: 439-448
Isnard S. T. Speck N. P. Rowe 2003a Growth habit and architecture of the sand dune-adapted climber Clematis flammula var. maritima L. Annals of Botany 91: 407-417
Isnard S. T. Speck N. P. Rowe 2003b Mechanical architecture and development in Clematis: implications for canalised evolution of growth forms. New Phytologist 158: 543-559[CrossRef][Web of Science]
Kim S.-C. D. J. Crawford J. Francisco-Ortega A. Santos-Guerra 1996 A common origin for woody Sonchus and five related genera in the Macaronesian islands: molecular evidence for extensive radiation. Proceedings of the National Academy of Sciences, USA 93: 7743-7748
Kimball R. T. D. J. Crawford D. H. Les E. Landolt 2003 Out of Africa: molecular phylogenetics and biogeography of Wolffiella (Lemnaceae). Biological Journal of the Linnean Society 79: 565-576[CrossRef]
Klackenberg J. 1992a Taxonomy of Secamone s.lat. (Asclepiadaceae) in the Madagascar region. Opera Botanica 112: 1-127
Klackenberg J. 1992b Taxonomy of Secamone (Asclepiadaceae) in Asia and Australia. Kew Bulletin 47: 595-612[CrossRef]
Klackenberg J. 1995 Malagasy Asclepiadaceae: reinstatement of the genus Pervillaea and two new combinations. Phytologia 78: 189-191
Klackenberg J. 1996a Revision of the Malagasy genus Pervillaea (Asclepiadaceae) and its phylogenetic relationship to Calyptranthera. Nordic Journal of Botany 16: 165-184
Klackenberg J. 1996b The new genus Calyptranthera (Asclepiadaceae) from Madagascar. Novon 6: 25-28
Klackenberg J. 1997 Revision of the Malagasy genus Calyptranthera (Asclepiadaceae). Adansonia 19: 21-37
Klackenberg J. 2001 Notes on Secamonoideae (Apocynaceae) in Africa. Adansonia 23: 317-335
Knox E. B. J. D. Palmer 1995 Chloroplast DNA variation and the recent radiation of the giant Senecios (Asteraceae) on the tall mountains of eastern Africa. Proceedings of the National Academy of Sciences, USA 92: 10349-10353
Knox E. B. S. R. Downie J. D. Palmer 1993 Chloroplast genome rearrangements and the evolution of giant lobelias from herbaceous ancestors. Molecular Biology and Evolution 10: 414-430[Web of Science]
Köhler L. T. Speck H.-C. Spatz 2000 Micromechanics and anatomical changes during early ontogeny of two lianescent Aristolochia species. Planta 210: 691-700[CrossRef][Web of Science][Medline]
Liede S. 2001 Subtribe Astephaninae (Apocynaceae-Asclepiadoideae) reconsidered: new evidence based on cpDNA spacers. Annals of the Missouri Botanical Garden 88: 657-668[CrossRef][Web of Science]
Liede S. A. Täuber 2000 Sarcostemma R. Br. (Apocynaceae-Asclepiadoideae)—a controversial generic circumscription reconsidered: evidence from trnL-F spacers. Plant Systematics and Evolution 225: 133-140[CrossRef][Web of Science]
Liede S. A. Täuber 2002 Circumscription of the genus Cynanchum (Apocynaceae-Asclepiadoideae). Systematic Botany 27: 789-800[Web of Science]
Liede S. U. Meve A. Täuber 2002 What is the subtribe Glossonematinae (Apocynaceae-Asclepiadoideae)? A phylogenetic study based on cpDNA spacer. Botanical Journal of the Linnean Society 139: 145-158[CrossRef]
Metcalf C. R. L. Chalk 1979 Anatomy of the dicotyledons, vol. 1. Oxford University Press, Oxford, UK
Meve U. S. Liede 2001 Reconsideration of the status of Lavrania, Larryleachia and Notechidnopsis (Asclepiadoideae-Ceropegieae). South African Journal of Botany 67: 161-168[Web of Science]
Meve U. S. Liede 2002 A molecular phylogeny and generic rearrangement of the Stapelioid Ceropegieae (Apocynaceae-Asclepiadoideae). Plant Systematics and Evolution 234: 171-209[CrossRef][Web of Science]
Moore M. J. J. Francisco-Ortega A. Santos-Guerra R. K. Jansen 2002 Choroplast DNA evidence for the roles of island colonization and extinction in Tolpis (Asteraceae: Lactuceae). American Journal of Botany 89: 518-526
Mort M. E. D. E. Soltis P. S. Soltis J. Francisco-Ortega A. Santos-Guerra 2002 Phylogenetics and evolution of the Macaronesian clade Crassulaceae inferred from nuclear and chloroplast sequence data. Systematic Botany 27: 271-288[Web of Science]
Niklas K. J. 1992 Plant biomechanics: an engineering approach to plant form and function. University of Chicago Press, Chicago, Illinois, USA
Niklas K. J. 1997 The evolutionary biology of plants. University of Chicago Press, Chicago, Illinois, USA
Olson M. E. 2003 Stem and leaf anatomy of the arborescent Cucurbitaceae Dendrosicyos socotrana with comments on the evolution of pachycauls from lianas. Plant Systematics and Evolution 239: 199-214[CrossRef][Web of Science]
Panero J. L. J. Francisco-Ortega R. K. Jansen A. Santos-Guerra 1999 Molecular evidence for multiple origins of woodiness and a new world biogeographic connection of the Macaronesian island endemic Pericallis (Asteraceae: Senecioneae). Proceedings of the National Academy of Sciences, USA 96: 13886-13891
Putz F. E. 1984 How trees avoid and shed lianas. Biotropica 16: 19-23[CrossRef][Web of Science]
Putz F. E. N. M. Holbrook 1991 Biomechanical studies of vines. In F. E. Putz and H. A. Mooney [eds.], The biology of vines, 73–97. Cambridge University Press, Cambridge, UK
Rowe N. P. T. Speck 1996 Biomechanical characteristics of the ontogeny and growth habit of the tropical liana Condylocarpon guianense (Apocynaceae). International Journal of Plant Sciences 157: 406-417[CrossRef]
Rowe N. P. T. Speck 1998 Biomechanics of plant growth forms: the trouble with fossil plants. Review of Paleobotany and Palynology 102: 43-62
Rowe N. P. T. Speck 2004 Hydraulics and mechanics of plants: novelty, innovation and evolution. In A. R. Hemsley and I. Poole [eds.], The evolution of plant physiology, 297–325. Academic Press, London, UK
Sauquet H. J. A. Doyle T. Scharaschkin T. Borsch K. W. Hilu L. W. Chatrou A. Le Thomas 2003 Phylogenetic analysis of Magnoliales and Myristicaceae based on multiple data sets: implications for character evolution. Botanical Journal of the Linnean Society 142: 125-186[CrossRef]
Solereder H. L. A. Boodle 1908 Systematic anatomy of the Dicotyledons, vol. I. Introduction, Polypetalae, Gamopetalae. Clarendron Press, Oxford, UK
Speck T. 1991 Changes of the bending mechanics of lianas and self-supporting taxa during ontogeny. In Natural structures. Principles, strategies, and models in architecture and nature, Proceedings of the II. International Symposium of the Sonderforschungsbereich 230, part I, 89–95. Stuttgart, Germany
Speck T. 1994 Bending stability of plant stems: ontogenetical, ecological and phylogenetical aspects. Biomimetics 2: 109-128
Speck T. N. P. Rowe 1999 A quantitative approach for analytically defining size, growth form and habit in living and fossil plants. In M. H. Kurmann and A. R. Hemsley [eds.], The evolution of plant architecture, 447–479. Royal Botanical Garden, Kew, UK
Speck T. N. P. Rowe F. Brüchert W. Haberer F. Gallenmüller H.-C. Spatz 1996 How plants adjust the "material properties" of their stems according to differing mechanical constraints during growth: an example of smart design in nature. In A. E. Engin [ed.], Bioengineering. Proceedings of the 1996 Engineering Systems Design and Analysis Conference, PD-vol. 77, 5: 233–241. American Society of Mechanical Engineers, New York, New York, USA
Speck T. C. Neinhuis F. Gallenmüller N. P. Rowe 1997 Trees and shrubs in the mainly lianescent genus Aristolochia s.l.: secondary evolution of the self-supporting growth habit?. In G. Jeronomidis and J. F. V. Vincent [eds.], Plant Biomechanics 1997: Conference proceedings I, 201–207. Centre for Biomimetics, The University of Reading, Reading, UK
Speck T. N. P. Rowe L. Civeyrel R. Claßen-Bockhoff C. Neinhuis H.-C. Spatz 2003 The potential of plant biomechanics in functional biology and systematics. In T. F. Stuessy, V. Mayer, and E. Hörandl [eds.], Deep morphology: toward a renaissance of morphology in plant systematics, 241–271. Koeltz Scientific Books, Königstein, Germany
Staden R. 1996 The Staden sequence analysis package. Molecular Biotechnology 5: 233-241[Web of Science][Medline]
Sugita M. K. Shinozaki M. Sugiura 1985 Tobacco chloroplast trnalys (uuu) gene contains a 2.5-kilobase-pair intron: an open reading frame and a conserved boundary sequence in the intron. Proceedings of the National Academy of Sciences, USA 82: 3557-3561
Swofford D. J. 2002 PAUP*: phylogenetic analysis using parsimony (* and other methods), version 4. Sinauer, 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][Web of Science][Medline]
Thompson J. D. T. J. Gibson F. Plewniak F. Jeanmougin D. G. Higgins 1997 The Clustal X windows interface: flexible strategies for sequence alignment aided by quality analysis tools. Nucleic Acids Research 25: 4876-4882
Treutlein J. F. S. Gideon B. E. Van Wyk M. Wink 2003 Phylogenetic relationships in Asphodelaceae (subfamily Alooideae) inferred from chloroplast DNA sequences (rbcL, matK) and from genomic fingerprinting (ISSR). Taxon 52: 193-207[CrossRef][Web of Science]
Vincent J. F. V. 1990 Structural biomaterials. Princeton University Press, Princeton, New Jersey, USA
Wallace A. R. 1878 Tropical nature and other essays. Macmillan, London, UK
White F. 1983 The vegetation of Africa. UNESCO, Paris, France
Xiang Q.-Y. D. E. Soltis P. S. Soltis 1998 Phylogenetic relationships of Cornaceae and close relatives inferred from matK and rbcL sequences. American Journal of Botany 85: 285-297[Abstract]
CiteULike Complore Connotea Del.icio.us Digg Facebook Reddit Technorati Twitter What's this?
This article has been cited by other articles:
|
|
 |
|
  L. Menard, D. McKey, and N. Rowe Developmental plasticity and biomechanics of treelets and lianas in Manihot aff. quinquepartita (Euphorbiaceae): a branch-angle climber of French Guiana Ann. Bot., June 1, 2009; 103(8): 1249 - 1259. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Isnard and N. P. Rowe The climbing habit in palms: Biomechanics of the cirrus and flagellum Am. J. Botany, December 1, 2008; 95(12): 1538 - 1547. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
0.05; **: P 
