全 文 ：植 物 分 类 学 报 44 (2): 113–125(2006) doi:10.1360/aps050065
Acta Phytotaxonomica Sinica http://www.plantsystematics.com
———————————
Received: 13 April 2005 Accepted: 5 September 2005
Supported by the National Natural Science Foundation of China, Grant Nos. 39870079, 30170062, and the Ph.D. Programs
Foundation of Ministry of Education of China, Grant No. 20020335104.
* Present address: College of Life Science and Technology, Tongji University, Shanghai 200092.
** Author for correspondence. E-mail:
A phylogenetic analysis of the Smilacaceae based on
morphological data
1CHEN Shi-Chao* 1QIU Ying-Xiong 1WANG Ai-Li 2Kenneth M. CAMERON 1FU Cheng-Xin**
1(Laboratory of Systematic and Evolutionary Botany and Biodiversity, College of Life Sciences, Zhejiang University,
Hangzhou 310058, China)
2(The Lewis B. and Dorothy Cullman Program for Molecular Systematics Studies, The New York Botanical Garden,
Bronx, New York 10458, USA)
Abstract In order to construct phylogenetic relationships of the family Smilacaceae, 79
species representing all genera and sections of the family from all over the world were used for
cladistic analysis, with Philesia and Lapageria of Philasiaceae as outgroups. A matrix of 36
binary and 11 multistate morphological characters was constructed. The first parsimonious
analysis yielded numerous most parsimonious trees of 275 steps, and then 82 most
parsimonious trees with 40.729 steps length were produced by reweighting characters using
maximum values of rescaled consistency (RC), which remained stable after the second
reweighting. There is a moderate resolution but generally poor bootstrap support for the
phylogenetic relationships in the family. From the strict MP tree and the NJ tree, Ripogonum is
a sister to Smilax, but may be better treated as a separate family. Heterosmilax is considered to
be derived from ancestral Smilax and occurs as the most derived clade of the Smilacaceae, and
thus it is reasonable to retain it as a section in Smilax s.l. Within Smilax most currently
recognized sections are polyphyletic, but some major lineages are distinct. The herbaceous
Smilax species form a monophyletic clade, and it is also the case for the South American
species. Their positions in the cladogram are intriguing and may help to shed light on
evolutionary and biogeographic history of the family.
Key words Smilacaceae, Smilax, Heterosmilax, Ripogonum, morphology, cladistics,
phylogeny.
The family Smilacaceae was separated by Hutchinson (1973) from the polyphyletic
family Liliaceae s.l., and this treatment was followed by a number of botanists (Dahlgren,
1975; Dahlgren et al., 1985; Cronquist, 1981, 1988; Thorne, 1983; Takhtajan, 1987; Conran,
1998; Judd, 1998; Judd et al., 1999). It comprises ca. 375 species (Heywood et al., 1993) with
a wide distribution in tropics and subtropics. The family is characterized by reticulate leaf
venation, paired petiolar tendrils, dioecious flowers, umbellate inflorescences and mostly
woody, climbing habit. Although there is a general agreement on the familial status of the
Smilacaceae, the controversy on its circumscription and infrafamilial phylogenetic
relationships still exists among botanists. Dahlgren (1975) included four genera in the family,
viz., Smilax L., Heterosmilax Kunth., Pseudosmilax Hayata and Ripogonum J. R. Forst. & G.
Forst., whereas Takhtajan (1997) recognized only two genera, Smilax (including
Pseudosmilax) and Heterosmilax. Conran & Clifford (1985) suggested for the first time that
Ripogonum be placed in a family of its own based on the phenetic studies, but this suggestion
has not been widely accepted. Recent molecular data (Chase et al., 2000; Cameron & Fu, in
press) indicated that Ripogonum was a sister to Smilax. However, Rudall and her coworkers
Acta Phytotaxonomica Sinica Vol. 44 114
(Rudall et al., 2000) constructed a combined tree from molecular and morphological data,
showing that Smilax is a sister group to the Philesia-Ripogonum complex. Therefore, the
circumscription and generic relationships of the family remain to be clarified. The
Smilacaceae is here circumscribed as containing three genera, Smilax, Heterosmilax, and
Ripogonum following Koyama (1960, 1984), Hutchinson (1973), Heywood et al. (1993),
Mabberley (1997), Takhtajan (1997), and Conran (1998).
Smilax is a core genus of the family with ca. 350 species (Takhtajan, 1997) or ca. 200
species (Cameron & Fu, in press) and worldwide distribution, but most diverse in Asia and
America. Heterosmilax, with ca. 12 species, is a small southeastern Asian genus closely allied
to Smilax, while Ripogonum with six species occurs only in eastern Australia, New Guinea,
and New Zealand. In his attempt at monographic study of the family, Koyama (1960)
proposed a classification of six sections in Smilax, and later he (Koyama, 1984) divided
Heterosmilax into two sections. Holmes (2002) suggested without explanation that
Heterosmilax be reduced to a section within Smilax. Conran and Clifford (1985) considered
Ripogonum distantly related to the rest of the Smilacaceae based on phenetic analyses. The
taxonomic treatments of the family and genera in the past several decades (Wang et al., 1978;
Andreata, 1997; Judd, 1998; Chen et al., 2000) involved mainly local areas, and thus the
family has never been monographed in its entirety and is in need of monographic and
phylogenetic study.
As a part of phylogenetic research of the Smilacaceae, this paper attempts to construct a
phylogeny of the family through cladistic analysis based on morphological data. The paper
describes variability of morphological characters in 79 species with two genera (Philesia
Comm. ex Juss. and Lapageria Ruiz & Pav.) of the family Philesiaceae as outgroups.
1 Material and methods
1.1 In-group and outgroups
Seventy-nine species of the three genera were selected as operational taxonomic units
(OTUs) or terminal taxa. The present analysis was based on the study of herbarium specimens
from the Herbaria of Institute of Botany, the Chinese Academy of Sciences (PE); Kunming
Institute of Botany, the Chinese Academy of Sciences (KUN); Zhejiang University, China
(HZU); New York Botanical Garden, the United States (NY); Muséum National dHistoire
Naturelle, Paris, France (PCU); and University of Tokyo, Japan (TI). Living material in the
field in China and the United States and transplanted materials growing in the Botanical
Garden of Zhejiang University were also used. The information of partial characters was
taken from the literature. The voucher specimens are listed in Table 1.
Two outgroups were chosen based on the system of APGII (2003). In addition,
Philesiaceae is positioned as sister to Smilacaceae following the cladistic analyses on
monocots using morphological and rbcL sequence data performed by Chase et al. (1995).
Thus, two species respectively from the genera Philesia and Lapageria of Philesiaceae were
chosen as the outgroups. All the trees in this paper were rooted with this family.
1.2 Morphological characters
In order to minimize the noise due to developmental plasticity, every effort was made by
using mature organs for each character. Forty-seven vegetative and floral characters,
including some palynological and cytological characters, were selected in this analysis, of
which 36 were scored as binary and 11 as multi-state (Table 2). In a few cases, some
characters were scored as missing when unavailable. Winclada program (Nixon, 1999) was
employed to record raw data and create data matrix for tree-building calculation. The
complete data matrix is shown in Table 3.
No. 2 CHEN S-C et al.: A phylogenetic analysis of the Smilacaceae based on morphological data 115
Table 1 Source of materials for observation
Taxon Locality Voucher
Heterosmilax chinensis Wang Without precise locality, Sichuan (四川), China W. C. Cheng (郑万钧) 11002 (PE)
H. gaudichaudiana (Kunth)
Maxim.
Without precise locality, Guangdong (广东),
China
S. Y. Hu (胡秀英) 12870 (PE)
H. japonica Kunth Taishun (泰顺), Zhejiang (浙江), China C. X. Fu (傅承新) 96118 (HZU)
H. seisuiensis (Hayata) Wang &
Tang
Without precise locality, Taiwan (台湾), China Anonymous s.n. 82439 (PE)
H. yunnanensis Gagnep. Xichou (西畴), Yunnan (云南), China S. Z. Wang (王守正) 0192936 (KUN)
Ripogonum album R. Br. Without precise locality, New Guinea Morobe & Distrikt 488 (NY)
R. discolor F. Muell. Sydney, Australia C. X. Fu (傅承新) 98020 (HZU)
Smilax aberrans Gagnep. Without precise locality, Guizhou (贵州), China C. X. Fu (傅承新) 20037 (HZU)
S. arisanensis Hayata Huaping (花坪), Guangxi (广西), China C. X. Fu (傅承新) 91052 (HZU)
S. aspera L. Without precise locality, Italy P. Delprete 7164 (NY)
S. aspericaulis Wall. Gongshan (贡山), Yunnan (云南), China Dulongjiang Exped. (独龙江队) 4707
(KUN)
S. astrosperma Wang & Tang Mt. Wuzhishan (五指山), Hainan (海南), China X. R. Liang (梁向日) 64347 (PE)
S. auriculata Walt Without precise locality, North Carolina, USA B. Wrunctan 1197 (NY)
S. australis R. Br. Without precise locality, Australia L. Pv. 1196 (NY)
S. basilata Wang & Tang Wenshan (文山), Yunnan (云南), China S. C. Chen (陈士超) 0302029 (ZHU)
S. biumbellata Koyama Without precise locality, Yunnan (云南), China B. Y. Qiu (邱炳云) 52520 (PE)
S. bona-nox L. Without precise locality, North Carolina, USA K. M. Cameron 1053 (NY)
S. bracteata Presl Funing (富宁), Yunnan (云南), China H. T. Tsai (蔡希陶) 0193111 (KUN)
S. brasiliensis Spreng. Without precise locality, Brazil Anonymous s.n. 24826 (PCU)
S. californica A. Gray Without precise locality, Virginia, USA FW93 CF state 1913 (NY)
S. campestris Griseb. Without precise locality, Brazil From literature (Andreata, 1997)
S. chapaensis Gagnep. Mt. Qingchengshan (青城山), Sichuan (四川),
China
C. X. Fu (傅承新) 88302 (HZU)
S. china L. Pingbian (屏边), Yunnan (云南), China C. W. Wang (王启无) 88478 (PE)
S. chingii Wang & Tang Mt. Jinfoshan (金佛山), Nanchuan (南川),
Chongqing (重庆), China
C. X. Fu (傅承新) 92001 (HZU)
S. cissoides Mart. ex Griseb. Without precise locality, Brazil From literature (Andreata, 1997)
S. cocculoides Warb. ex Diels Mt. Fanjingshan (梵净山), Guizhou (贵州),
China
C. X. Fu (傅承新) 91307 (HZU)
S. corbularia Kunth Malipo (麻栗坡), Yunnan (云南), China K. M. Feng (冯国楣) 4871 (PE)
S. darrisii Lévl. Nanchuan (南川), Chongqing (重庆), China K. Y. Lang (郎楷永) 870 (PE)
S. davidiana A. DC. Mt. Tianmushan (天目山), Zhejiang (浙江),
China
S. C. Chen et al. (陈士超等) 0306132
(HZU)
S. densibarbata Wang & Tang Pingbian (屏边), Yunnan (云南), China C. W. Wang (王启无) 85452 (PE)
S. discotis Warb. Mt. Fengyangshan (凤阳山), Zhejiang (浙江),
China
C. X. Fu (傅承新) 89144 (HZU)
S. ecirrhata Wats. Without precise locality, USA J. A. Steyermark 40460 (NY)
S. ferox Wall. ex Kunth Without precise locality, Yunnan (云南), China B. Y. Qiu & S. T. Li (邱炳云, 李生唐)
0193635 (KUN)
S. glabra Roxb. Simao (思茅), Yunnan (云南), China Anonymous s.n. 6312, 0194040 (KUN)
S. glauca Walter Without precise locality, New York, USA C. X. Fu 9987 (HZU)
S. glauco-china Warb. ex Diels Guilin (桂林), Guangxi (广西), China S. C. Chen (陈士超) 0304023 (HZU)
S. havanensis Jacq. Without precise locality, Cuba MMB770 (NY)
S. hayatae Koyama Without precise locality, Taiwan (台湾), China Taipei Empire University (台北帝国大
学) 19308 (PE)
S. hemsleyana Craib Wenshan (文山), Yunnan (云南), China C. X. Fu (傅承新) 93043 (HZU)
S. herbacea L. Without precise locality, North Carolina, USA K. M. Cameron 11 (NY)
S. hispida Raf. Gillies Lake, USA P. V. Krotkov 7268 (NY)
S. jamesii G. A. Wallace Without precise locality, California, USA ARK 3725 (NY)
S. kraussiana Meisn. Without precise locality, Gabon JMR 3151 (NY)
S. lanceifolia Roxb. Pingbian (屏边), Yunnan (云南), China C. W. Wang (王启无) 71996 (PE)
S. lanceolata L. Without precise locality, Mexico E. Matuda 3988 (NY)
S. laurifolia L. Without precise locality, Carolina, USA P. Krotkov 5683 (NY)
S. lebrunii Lévl. Xichou (西畴), Yunnan (云南), China S. C. Chen (陈士超) 03101417 (HZU)
Acta Phytotaxonomica Sinica Vol. 44 116
Table 1 (continued)
Taxon Locality Voucher
S. lunglingensis Wang & Tang Pingbian (屏边), Yunnan (云南), China B. Y. Qiu (邱炳云) 52403 (PE)
S. lutescens Vell. Without precise locality, Brazil From literature (Andreata, 1997)
S. megacarpa A. DC. Without precise locality, Cambodge, Thailand V. N. Laos (PCU)
S. menispermoidea A. DC. Pingbian (屏边), Yunnan (云南), China B. Y. Qiu (邱炳云) 53633 (PE)
S. microphylla C. H. Wright Mt. Jinfoshan (金佛山), Chongqing (重庆),
China
C. X. Fu (傅承新) 92027 (HZU)
S. minarum A. DC. Without precise locality, Brazil From literature (Andreata, 1997)
S. muscosa Toledo Without precise locality, Brazil From literature (Andreata, 1997)
S. myrtillus A. DC. Gongshan (贡山), Yunnan (云南), China Dulongjiang Exped. (独龙江队) 6680
(KUN)
S. nervo-marginata Hayata Baishanzu (百山祖), Zhejiang (浙江), China S. C. Chen et al. (陈士超等) 0308004
(HZU)
S. nigrescens Wang & Tang Without precise locality, Sichuan (四川), China S. C. Chen (陈心启) 02023 (PE)
S. nipponica Miq. Kaili (凯里), Guizhou (贵州), China Qiannan Exped. (黔南队) 1183 (KUN)
S. oblongifolia Pohl ex Griseb. Without precise locality, Brazil From literature (Andreata, 1997)
S. ocreata A. DC. Pingbian (屏边), Yunnan (云南), China C. W. Wang (王启无) 73366 (KUN)
S. outanscianensis Pamp. Without precise locality, Hubei (湖北), China Y. Liu (刘瑛) 530 (PE)
S. perfoliata Lour. Yangshan (阳山), Guangdong (广东), China L. Deng (邓良) 1543 (HZU)
S. planipes Wang & Tang Pingbian (屏边), Yunnan (云南), China C. X. Fu (傅承新) 93015 (HZU)
S. polycephala Wang & Tang Without precise locality, Yunnan (云南), China Sino-USSR Exped. (中苏考察团) 2504
(PE)
S. polycolea Warb. ex Diels Nanning (南宁), Guangxi (广西), China S. H. Chun (陈少卿) 14559 (PE)
S. pottingeri Prain Malipo (麻栗坡), Yunnan (云南), China K. M. Feng (冯国楣) 12162 (PE)
S. pumila Walter Without precise locality, Florida, USA K. M. Cameron 1065 (NY)
S. riparia A. DC. Fengcheng (凤城), Liaoning (辽宁), China C. X. FU (傅承新) CF9902 (HZU)
S. rotundifolia Bello Without precise locality, New York, USA W. Stevens 23665 (NY)
S. scobinicaulis C. H. Wright Mt. Emeishan (峨眉山), Sichuan (四川), China C. X. Fu (傅承新) 92051 (HZU)
S. sieboldii Miq. Without precise locality, Zhejiang (浙江), China M. B. Deng (邓懋斌) 3914 (PE)
S. spicata Vell. Without precise locality, Brazil From literature (Andreata, 1997)
S. stans Maxim. Zhenkang (镇康), Yunnan (云南), China T. T. Yu (俞德浚) 11317 (PE)
S. syphilitica Willd Without precise locality, Brazil From literature (Andreata, 1997)
S. trinervula Miq. Without precise locality, Jiangxi (江西), China K. C. Kuan (关克俭) 77058 (PE)
S. tsinchengshanensis Wang Nanchuan (南川), Chongqing (重庆), China K. H. Shing (邢公侠) 937 (PE)
S. vanchingshanensis (Wang &
Tang) Wang & Tang
Nanchuan (南川), Chongqing (重庆), China F. T. Wang (汪发缵) 10914 (PE)
S. vitiensis A. DC. Without precise locality, Fiji T. G. Yuncker 15512 (NY)
S. walteri Pursh Without precise locality, North Carolina, USA KMC1058 (NY)
Lapageria rosea Ruiz & Pav. Without precise locality, Chile MFD3686-6 (NY)
Philesia buxifolia Lam. ex Poir. Without precise locality, Chile KMC01 (NY)
1.3 Phylogenetic analysis
Maximum parsimony (MP) and neighbor-joining (NJ) trees were constructed using
PAUP* 4.0b10 program (Swofford, 2002). Cladistic analysis of phylogenetic relationships
was conducted by using Wagner parsimony and applying heuristic search with tree bisection-
reconnection (TBS) branch-swapping, multiple parsimony (MUL-PARS), and simple
stepwise taxon application of 1000 replications. During the first tree-construction process, all
characters were unordered and equally weighted with multi-states interpreted as uncertainty,
and “?” treated as missing information. Subsequently, reweighting of the characters by
maximum values of rescaled consistency indexes (RC) was performed in order to reduce the
misleading effect of homoplasious characters. Phenetic analysis was also performed with the
PAUP* program by using the neighbor-joining clustering method, with a mean character
difference as a distance measure. Supporting values were calculated by performing bootstrap
(bts) analysis of 1000 heuristic search replicates with the TBR branching swapping algorithm
(Felsenstein, 1985).
No. 2 CHEN S-C et al.: A phylogenetic analysis of the Smilacaceae based on morphological data 117
Table 2 List of characters and character states used in data matrix
1. Habit : 0—woody, 1—herbaceous
2. Stems aboveground: 0—perennial, 1—annual
3. Stem: 0— vines, 1—erect
4. Cross section of stem：0—round, 1—square
5. Wing on stem: 0—absent, 1—present
6. Prickles on stem: 0—absent, 1—present
7. Shape of prickles on stem: 0—blackish needle-like,
1—triangular or obtuse or absent
8. Verrucae on stem: 0—absent, 1—present
9. Tendril: 0—absent, 1—present
10. Leaf margin: 0—regular, 1—irregular (minutely
roughened under a lens)
11. Leaf color when dry: 0—yellowish or greenish, 1—
blackish
12. Phyllotaxis: 0—alternate, 1—opposite or whorled
13. Abscission zone of leaf on petiole: 0—at middle,
1—at top, 2—at bottom
14. Leaf blade white hairy abaxially: 0—absent,
1—present
15. Number of main veins: 0—seven, 1—five, 2—three
16. Wings on petiole: 0—absent, 1—narrow, 2—
semicircular, 3—broadly clasping node
17. Wing length/petiole length: 0—≤1/2, 1—>1/2
18. Flowers: 0—bisexual, 1—monosexual
19. Flowers borne on stem: 0—solitary, 1—as an
inflorescence
20. Inflorescence: 0—single umbels, 1—panicle, 2—
spike
21. Rachis articulation: 0—absent, 1—present
22. Rachis arising: 0—at young branch, 1—in axil of
leaf
23. Scale-like prophyll at base of peduncle: 0—absent,
1— present
24. Receptacle: 0—obviously bulgy, 1—not bulgy
25. Number of flowers per umbel: 0—≤10, 1—>10
26. Filament length: 0—≥2.5 mm, 1—<2.5 mm
27. Cross section of peduncle: 0—round, 1—compressed
28. Rachis length: 0—obviously shorter than petiole,
1—subequal to petiole, 2—longer (obviously longer than
petiole)
29. Bract on peduncle: 0—absent, 1—present
30. Stamen length/perianth length: 0—subequal, 1—1/2
31. Number of stamens in male flowers: 0—three, 1—six,
2—more than six
32. Number of staminodes in female flower: 0—six, 1—
three, 2—absent
33. Mature stamens: 0—spreading into a plane, 1—otherwise
34. Filament status: 0—free, 1—partly connate, 2—connate
35. Tepals: 0—free, 1—connate
36. Tepals when mature in male flower: 0—recurved, 1—
erect
37. Flowering time: 0—March–July, 1—August–December
38. Shape of berries: 0—globose, 1—prolate
39. Carpopodium: 0—straight, 1—recurved
40. Berry diameter: 0—small (≤1 cm), 1—big (>1 cm)
41. Berry color: 0—black, 1—red
42. Tepal color: 0—greenish-yellow, 1—red-brown
43. Pollen aperture: 0—monosulcate, 1—inaperturate, 2—
pseudo-aperturate
44. Pollen ornamentation: 0—reticulate, 1—spinulate, 2—
rugulate
45. Density of spinules on pollen: 0—low (<90 per 100 µm2),
1—high (>90 per 100 µm2)
46. Length of spinules on pollen: 0—long (>3 µm), 1—short
(<3 µm)
47. Basic numbers of chromosomes (n=): 0—16, 1—13, 15
2 Results
Cladistic analysis of the matrix with 47 morphological characters gave more than 154000
equally most parsimonious trees of 275 steps length with RC=0.141 (consistency index
CI=0.218, retention index RI=0.647). Only six clades were found in the strict consensus tree
(not shown). The low RC value indicates a high level of homoplasy in the variation pattern of
morphological characters. Farris (1969, 1989) suggested that homoplasious characters be
downweighted to obtain reliable result and proposed successive reweighting of characters by
RC to reduce the ambiguity caused by homoplasious characters. Twice reweighting characters
by maximum values of RC yielded 82 most parsimonious trees with 40.729 steps length
(CI=0.513, RI=0.840, RC=0.431) which remained stable after the second reweighting. A strict
consensus of 82 reweighted equally parsimonious trees is shown in Fig. 1. The unweighted
and reweighted trees revealed essentially the same clades of species clusters except position
of a few species.
Acta Phytotaxonomica Sinica Vol. 44 118
Table 3 Data matrix employed in the cladistic and phenetic analysis of Smilacaceae
5 10 15 20 25 30 35 40 45
. . . . . . . . .
Heterosmilax chinensis 00000 00110 01011 10110 00001 11100 01111 00010 21121 10
H. gaudichaudiana 00000 00010 01011 10110 00001 01200 0–121 0?010 00121 10
H. japonica 00000 00010 01110 10110 00001 01110 01111 01010 00121 10
H. seisuiensis 00000 00010 01111 10110 00001 11000 2–101 00010 00121 10
H. yunnanensis 00000 00010 01100 10110 00001 01100 20111 00010 00121 10
Ripogonum album 00000 00000 00011 00012 00011 00200 1–100 10011 10000 1?
R. discolor 00000 10000 00011 00012 00011 00200 1–100 10011 10000 1?
Smilax aberrans 00100 00000 01112 11110 01011 10201 10100 10001 00110 00
S. arisanensis 00000 10010 01111 11110 00111 10200 11100 00010 00110 00
S. aspera 00000 10010 01100 10112 00101 10201 10100 01010 10110 00
S. aspericaulis 00000 10110 01011 10111 00101 00110 11100 11110 01110 00
S. astrosperma 00000 10010 01012 10110 00101 00011 11100 1?010 00110 0?
S. auriculata 00000 10010 01011 10110 00001 01000 10100 00010 00110 0?
S. australis 00000 10010 01012 10110 10101 11100 1?100 0?010 00110 0?
S. basilata 00000 10010 01112 11110 00100 10000 1?120 00011 00110 0?
S. biumbellata 00000 00010 01110 10110 00000 10200 11100 00010 01110 00
S. bona-nox 00000 11110 01000 10110 00001 01200 1?100 10010 10110 0?
S. bracteata 00000 10010 01011 10111 00101 00110 11100 11010 ?1110 00
S. brasiliensis 00000 10010 01001 10110 01001 10001 10000 2?010 ?0210 0?
S. californica 00000 11011 01010 10110 00001 01200 10100 1?010 00110 0?
S. campestris 00000 00010 01212 10110 00001 10201 12000 2?010 ?0210 0?
S. chapaensis 00000 10110 01011 10110 10101 00010 10100 11010 00110 00
S. china 00000 10010 01011 11110 01001 00200 10100 10010 10110 00
S. chingii 00000 10010 01011 11110 01010 00200 10100 10010 10110 00
S. cissoides 00000 10010 01111 10110 00001 10101 10000 2?010 ?0210 0?
S. cocculoides 00100 00000 01012 10110 10100 10110 1–120 00010 00110 00
S. corbularia 00000 00010 01111 10110 00001 11101 11121 01010 10121 00
S. darrisii 00100 00000 11112 10110 00010 10201 1?100 00010 00110 0?
S. davidiana 00000 10010 01111 21110 01001 00200 11100 10010 10110 00
S. densibarbata 00000 10010 01011 10110 10101 00111 10100 00010 00110 0?
S. discotis 00000 10010 01111 21110 01000 00200 11100 00010 00110 00
S. ecirrhata 11100 00010 01110 10110 00001 01200 10100 10010 00110 01
S. ferox 00000 10000 01112 21110 01011 00100 10100 10010 10110 00
S. glabra 00000 00010 01112 10110 00001 10001 11110 01010 00121 00
S. glauca 00000 11010 01111 11110 00010 11200 10100 00010 00110 0?
S. glauco-china 00000 10010 01111 11110 01001 00200 11100 00010 00110 00
S. havanensis 00000 10010 01101 10110 01001 11001 10100 1?010 00110 0?
S. hayatae 00000 00000 01112 10110 00010 10201 1?100 00010 10110 0?
S. hemsleyana 00000 00010 01010 10110 10101 00110 11110 10010 ?0110 0?
S. herbacea 11000 00010 01110 10110 00001 01200 10100 10010 00110 01
S. hispida 00000 11011 01110 10110 00001 00200 1?100 10010 00110 00
S. jamesii 11100 00010 01211 10110 00001 01200 10100 10010 00110 01
S. kraussiana 00000 10010 01111 10110 10101 01110 1?100 0?010 ?0110 0?
S. lanceifolia 00000 10010 01011 10110 10101 00010 10100 11010 00110 00
S. lanceolata 00000 10010 01111 10110 00001 01100 11100 11010 00110 0?
S. laurifolia 00000 10011 01012 10110 00001 00100 10100 01010 00110 0?
S. lebrunii 00000 10110 01111 11110 01000 00100 10100 10010 10110 00
S. lunglingensis 00000 10010 01011 11111 00101 00110 11100 10010 00110 00
S. lutescens 00000 00010 01012 10110 00001 10100 1?000 2?010 ??210 0?
S. megacarpa 00000 00010 11111 10110 10100 00110 11100 11011 10110 0?
S. menispermoidea 00000 00010 01111 11110 00000 10201 10120 00010 01110 00
S. microphylla 00000 10010 01111 11110 00001 10001 11100 01010 00121 00
S. minarum 00000 00110 01011 10110 00011 10201 12000 2?010 ??210 0?
S. muscosa 00000 10111 01111 10110 01001 10000 1?000 2?010 ??210 0?
S. myrtillus 00101 10000 01012 11110 00010 10201 1?100 01010 00110 00
S. nervo-marginata 00000 00110 01110 10110 00000 11201 1?100 00010 ?0121 00
S. nigrescens 00000 00010 11111 11110 00001 10201 10100 00010 00110 0?
S. nipponica 11000 00010 01110 10110 00001 01200 10100 10010 00110 01
S. oblongifolia 00100 00000 01111 10110 00001 10100 11000 2?010 ?0210 0?
S. ocreata 00000 10010 01011 31111 00101 00111 12110 10010 10110 00
S. outanscianensis 00000 10010 01011 10110 01010 00100 10100 10010 00110 00
S. perfoliata 00000 10010 01011 31111 00101 00111 12100 10010 10110 0?
S. planipes 00000 00010 01111 11110 10110 00110 11100 11011 ?0110 0?
S. polycephala 00011 00010 01111 10112 00101 10211 1–100 01011 00110 00
S. polycolea 00000 10010 01011 10110 01000 00200 10100 00010 10110 00
No. 2 CHEN S-C et al.: A phylogenetic analysis of the Smilacaceae based on morphological data 119
Table 3 (continued)
5 10 15 20 25 30 35 40 45
. . . . . . . . .
S. pottingeri 00000 00010 11011 10110 00001 11200 1–100 00010 00110 0?
S. pumila 00000 00110 01211 10110 00001 11000 10100 11110 10110 0?
S. riparia 11000 00010 01110 10110 01001 01200 10100 10010 00110 01
S. rotundifolia 00000 10010 01010 10110 00000 00100 11100 10011 00110 0?
S. scobinicaulis 00000 11010 11110 11110 00010 00000 11100 10010 00110 00
S. sieboldii 00000 11011 01110 10110 00010 00200 11100 10010 00110 00
S. spicata 00011 00010 01111 10110 01001 10100 10100 2?010 ?0210 0?
S. stans 00100 00000 01111 11110 00010 10201 10100 00010 00110 00
S. syphilitica 00000 10110 01012 30110 10101 10210 1?000 2?010 ?0210 0?
S. trinervula 00000 10010 01012 10110 01010 00100 10100 10010 10110 00
S. tsinchengshanensis 00100 00000 01112 11110 01110 10201 11100 01010 01110 00
S. vanchingshanensis 00000 00010 01011 10110 10111 00110 10110 11010 ?0110 00
S. vitiensis 00000 00010 01211 10110 00001 00100 2?100 1?01? ?0110 0?
S. walteri 00000 10010 01011 10110 00010 00100 10100 10110 10110 0?
Lapageria rosea 00000 00000 01111 0000– –00–– 00100 1–100 0?011 ?11?? ?1
Philesia buxifolia 00000 00000 01112 0000– –00–– 00100 1–100 0?011 ?11?? ?1
Phenetic neighbour-joining analysis on the same matrix yielded a tree of 257 steps length
and with CI=0.213 and RC=0.676 (Fig. 2). The neighbour-joining and parsimonious trees
revealed similar clades of species, which differed in the position of some Smilax species.
In the strict consensus tree (Fig. 1) and the NJ tree (Fig. 2), the Ripogonum was basal to
the whole cladogram, in which R. album and R. discolor formed a monophyletic clade with
100% and 99% bootstrap values respectively. The two species are characterized by opposite
or whorled leaves (character 12), hermaphrodite flowers (character 18), and monosulcate and
reticulate pollen (character 43, 44).
All species of Smilax and Heterosmilax appeared in a clade on both the strict consensus
tree (Fig. 1) and the NJ tree (Fig. 2), with a high bootstrap value (>80%). This clade was well
circumscribed and characterized by a number of characters, especially two synapomorphies,
i.e., a pair of tendrils (character 9) and dioecism (character 18). On the strict tree, the clade of
five Heterosmilax species was supported by a high bootstrap value (84%), and merged in a
bigger clade including S. corbularia, S. nervo-marginata, S. glabra and S. microphylla with
89% bootstrap value, but the value of the bigger clade reduced to 37% in the NJ tree.
The herbaceous species, disjunct in East Asia and North America, always formed a
strongly supported clade on both the strict tree and the NJ tree (bootstrap values 93% and
86% respectively), with two synapomorphies, herbaceous habit (character 1) and annual stem
(character 2).
In both the strict tree and the NJ tree, the woody species from South America joined in a
clade. They share characters of petioles with narrow wings (character 16) and tepals of mature
male flowers spreading into a plane (character 33). This clade was supported by 69%
bootstrap value in the strict tree, but 23% in the NJ tree.
In the present analysis, many characters had a high level of homoplasy with low
consistency indices. Some portions of the trees were unresolved with low bootstrap values,
and much resolution was lost within the strict tree.
3 Discussion
In recent years, most botanists adopted a narrow concept of the Smilacaceae with only
three genera, Smilax, Heterosmilax, and Ripogonum (Thorne, 1992; Heywood et al., 1993;
Takhtajan, 1997), but Cronquist (1981, 1988) accepted a broader definition, including Smilax,
Ripogonum and Philesiaceae, Luzuriagaceae, and Petermanniaceae. However, as indicated in
the recent phylogenetic analysis of molecular data by Chase and his coworkers (Chase et al.,
1993, 1995, 2000), the family as circumscribed by Cronquist (1988) is very probably
Acta Phytotaxonomica Sinica Vol. 44 120
polyphyletic. In this study we follow the narrow concept of the family.
Fig. 1. The strict consensus tree of 82 equally most-parsimonious cladograms obtained from phylogenetic analysis of
morphological data for Smilacaceae. The tree is oriented with Lapageria rosea and Philesia buxifolia (Philesiaceae) as the
designated outgroups. Numbers on branches are bootstrap values (Bootstrap values below 50% are not shown) and
geographic distribution and the previous section for each species are indicated.
No. 2 CHEN S-C et al.: A phylogenetic analysis of the Smilacaceae based on morphological data 121
Fig. 2. The neighbor-joining tree obtained from phylogenetic analysis of morphological data for Smilacaceae. Numbers
above branches are bootstrap values (Bootstrap values below 50% are not shown).
Acta Phytotaxonomica Sinica Vol. 44 122
3.1 Phylogenetic position Ripogonum
The placement of Ripogonum in the Smilacaceae has been questioned by some botanists
(Conran & Clifford, 1985; Conran, 1987, 1989; Takhtajan, 1997). Based on 56 morphological
characters, Conran & Clifford (1985) found that Ripogonum is in phenetics closer to the
Dioscoreaceae than to Smilax and Heterosmilax. The recent ITS analysis of Smilax (extending
to the whole family) by Cameron & Fu (in press) shows that Ripogonum is a sister to the
remaining two genera Smilax and Heterosmilax. The present result shows that Ripogonum
species form a clade with a high bootstrap value (99%) in the strict tree (Fig. 1), and is also
sister to Heterosmilax and Smilax. Furthermore, in the NJ tree (Fig. 2), the branch separating
Ripogonum from Smilax is more than five times longer than the branch between S. vitiensis
and the remaining species of Smilax, which implies a somewhat isolated position of
Ripogonum within the family. The genus with six species is found only in Eastern Australia,
New Guinea and New Zealand and is characterized by monosulcate and reticulate pollen,
bisexual flowers, opposite/whorled leaves, and absence of paired tendrils. The separation of
Ripogonum as an independent family by Conran & Clifford (1985), followed by Takhtajan
(1987), is also supported by our study.
The APGII authors (APGII, 2003) mentioned that the Ripogonaceae and Philesiaceae
may be included in the Smilacaceae on account of their spinose pollen. Actually, pollen of
Ripogonum is monosulcate and reticulate rather than inaperturate and spinose according to
our palynological investigation of the Smilacaceae (Chen et al., unpublished). Fay (in press)
proposed that the three families would be best treated separately since they do not form a
monophyletic group within the combined molecular analyses.
3.2 Phylogenetic position Heterosmilax
Heterosmilax was established by Kunth (1850) and accepted in a monographic treatment
of de Candolle (1878), based on possession of a fused perianth. Its taxonomy was
subsequently revised by Koyama (1984). However, the generic status of Heterosmilax has
recently been questioned by some workers (Judd, 1998; Holmes, 2002). In the strict tree (Fig.
1), although all the species of Heterosmilax form a clade, they are nested in Smilax and form a
larger clade with four species of Smilax, which indicates that they are not in a position to
deserve a separate genus. A few species of Smilax (e.g. S. corbularia) have a partially united
perianth (Chen et al., 2000) and gradation in pollen morphology exists between Heterosmilax
and Smilax (Chen et al., unpublished). These two facts make the establishment of
Heterosmilax as a genus unjustifiable. Judd (1998), Judd et al. (1999), Takhtajan (1997) and
Holmes (2002) maintained the group in Smilax, which was confirmed by ITS analysis
(Cameron & Fu, in press) and this study.
In the strict tree, all the species of Heterosmilax form a most derived clade, which
implicates that evolutionary direction is from Smilax to Heterosmilax.
3.3 Phylogenetic relationships within Smilax
Smilax is a core genus in the family with ca. 200 species, including both woody and
herbaceous species. Koyama (1960) recognized six sections in his taxonomic treatment of
eastern Asian Smilax: sect. Pleiosmilax, sect. Smilax, sect. Macranthae, sect. Coilanthus, sect.
China, and sect. Coprosmanthus (Nemexia). However, a recent study on the genus using ITS
sequences (Cameron & Fu, in press) shows that Koyama’s system is possibly artificial. The
strict tree presented here also shows that Koyama’s sections, except for sect. Coprosmanthus
(Nemexia), are not supported as monophyletic groups (Fig. 1). The section Coprosmanthus
(Nemexia) is characterized by herbaceous habit and annual stem, and is a sister to all woody
Smilax (except S. vitiensis). This result is consistent with the results of ITS and rpL16
analyses (Fu et al., 2005).
The two species in sect. Smilax, S. aspera and S. polycephala, also form a small clade.
No. 2 CHEN S-C et al.: A phylogenetic analysis of the Smilacaceae based on morphological data 123
However, this result conflicts with the ITS analysis, which indicates that S. aspera, the most
widely distributed species in the genus, is a sister to the entire family, and with the most
primitive position (Cameron & Fu, in press). The exact position of S. aspera in the family
needs further investigation. In sect. Pleiosmilax, only one species from Fiji, S. vitiensis, was
analyzed in the present study, and thus it cannot be judged as monophyletic or not. In the
strict tree, this Pacific species possessing numerous stamens is at the basal position of Smilax,
which is unexpected. At any rate, we still presume that the feature of numerous stamens has
evolved at least twice in the Smilacaceae, since three species of Heterosmilax also have
numerous stamens. Sect. Coilanthus is closely related to Heterosmilax according to the
previous studies mentioned by Koyama (1984), which is supported in this work, although the
bootstrap value is not exceptionally high. In sect. Macranthae, most species form a clade,
which implies somewhat close relationships between species although bootstrap values do not
support this.
It is immature at the present time to subdivide naturally Smilax into sections based on
morphology, though there are a few clades of species that can be defined by habit or
vegetative features.
3.4 Geographic implications
Cameron & Fu (in press) have indicated clear differentiation between Old World and
New World species in the Smilax on their ITS tree. However, our morphological tree shows a
more complicated distribution pattern. It presents several mixed clades which nest species
from different continents, but only two of them are supported by bootstrap values (as
indicated by arrows in Fig. 1). One is the herbaceous S. riparia-S. herbacea clade with 93%
bootstrap value, which is the evidence of migration from the Old World to the New World. Fu
et al. (2005) have discussed the historical biogeography of this clade in detail. They attribute
the disjunct distribution to Pleistocene vicariance events separating contiguous
transcontinental populations following migration from East Asia to North America, as
documented for a number of flowering plant groups (Boufford & Spongberg, 1983).
The other mixed clade with 62% bootstrap value contains S. sieboldii, S. scobinicaulis, S.
hispida, S. californica and S. bona-nox. All these species (except S. bona-nox) together with a
fossil taxon S. lamarensis Knowlton from Wyoming were presumed to have experienced
migration from Asia to North America by Norton (1916). On the contrary, Cameron & Fu (in
press) conclude according to their ITS analysis that the presence of S. sieboldii and S.
scobinicaulis in Asia was a result of migration from North America. Our study also shows
that the two Chinese species, S. sieboldii and S. scobinicaulis, were derived from North
American species S. hispida, S. californica and S. bona-nox (see Fig. 1). Furthermore, S.
californica, the most northwestern American species, has a basal position in this mixed clade,
which favors Cameron and Fu’s hypothesis based on ITS analysis (Cameron & Fu, in press).
This study gives an outline of phylogenetic relationships within the Smilacaceae, and
thus further works on the Smilacaceae are badly needed. We are now generating more
molecular data adding to the phylogenetic analysis.
Acknowledgements We are grateful to the curators of PE, KUN, HZU, NY, PCU and TI
for access to specimens, and to Prof. HONG De-Yuan (Institute of Botany, the Chinese
Academy of Sciences) for his critical reading and improvement of the manuscript, and to
Prof. ZHANG Xiao-Ping (Anhui Normal University) for valuable comments on an earlier
draft of the manuscript. We would like to thank Dr. DENG Ming who kindly assisted in many
ways when the first author twice visited the Kunming Institute of Botany, the Chinese
Academy of Sciences.
Acta Phytotaxonomica Sinica Vol. 44 124
References
Andreata R H P. 1997. Revisão das espécies brasileiras do gênero Smilax L. (Smilacaceae). Pesquisas
Botânica 47: 7–244.
APGII. 2003. An update of the Angiosperm Phylogeny Group classification for the orders and families of
flowering plants: APGII. Botanical Journal of the Linnean Society 141: 399–436.
Boufford D E, Spongberg S A. 1983. Eastern Asian-Eastern North American phytogeographic relationship—a
history from the time of Linnaeus to the twentieth century. Annals of the Missouri Botanical Garden 70:
423–439.
Cameron K M, Fu C-X. A nuclear rDNA phylogeny of Smilax (Smilacaceae). In: Columbus J T, Friar E A,
Porter J M, Prince L M, Simpson M G eds. Monocots: Comparative Biology and Evolution. Vol. 2.
Rancho Santa Ana Botanic Garden. Claremont: California Press. (in press)
Chase M W, Duvall M R, Hills H G, Conran J G, Cox A V, Eguiarte L E, Hartwell J, Fay M F, Caddick L R,
Cameron K M, Hoot S. 1995. Molecular phylogenetics of Lilianae. In: Rudall P J, Cribb P J, Cutler D F,
Humphries C J eds. Monocotyledons: Systematics and Evolution. London: Royal Botanic Gardens, Kew.
1: 109–137.
Chase M W, Soltis D, Olmstead R G, Morgan D, Les D H, Mishler B, Duvall M R, Price R A, Hills H G, Qiu
Y-L, Kron K A, Rettig J H, Conti E, Palmer J D, Manhart J R, Sytsma K J, Michaels H J, Kress W J,
Karol K G, Clark W D, Hedrén M, Gaut B S, Jansen R K, Kim K-J, Wimpee C F, Smith J F, Furnier G R,
Straus S H, Xiang Q-Y, Plunkett G M, Soltis P S, Swensen S M, Williams S E, Gadek P A, Quinn C J,
Eguiarte L, Golenberg E, Learn G H, Graham S W, Barrett S C H, Dayanandan S, Albert V A. 1993.
Phylogenetics of seed plants: an analysis of nucleotide sequences from the plastid gene rbcL. Annals of
the Missouri Botanical Garden 80: 528–580.
Chase M W, Soltis D E, Soltis P S, Rudall P J, Fay M F, Hahn W H, Sullivan S, Joseph J, Givinish T J,
Systma K J, Pires J C. 2000. Higher-level systematics of the monocotyledons: An assessment of current
knowledge and a new classification. In: Wilson K L, Morrison D A eds. Monocots: Systematics and
Evolution. Collingwood, Victoria: CSIRO Publishing. 3–16.
Chen S-C (陈心启), Koyama T, Liang S-Y (梁松筠). 2000. Smilax L. and Heterosmilax Kunth. In: Wu Z-Y
(吴征镒), Raven P H eds. Flora of China. Beijing: Science Press; St. Louis: Missouri Botanical Garden
Press. 24: 96–117.
Conran J G. 1987. A phenetic study of the relationships of Drymophila R. Br. within the reticulate-veined
Liliiflorae. Australian Journal of Botany 35: 283–300.
Conran J G. 1989. Cladistic analyses of some net-veined Liliiflorae. Plant Systematics and Evolution 168:
123–141.
Conran J G. 1998. Smilacaceae. In: Kubitzki K ed. The Families and Genera of Vascular Plants. Vol. 3.
Flowering Plants. Monocotyledons: Lilianae (except Orchidaceae). Berlin: Springer-Verlag. 417–422.
Conran J G, Clifford H T. 1985. The taxonomic affinities of the genus Ripogonum. Nordic Journal of Botany 5:
215–219.
Cronquist A. 1981. An Intergrated System of Classification of Flowering Plants. New York: Columbia
University Press. 1262–1263.
Cronquist A. 1988. The Evolution and Classification of Flowering Plants. 2nd ed. New York: New York
Botanical Garden Press. 555–556.
Dahlgren R M T. 1975. A system of classification of the angiosperms to be used to demonstrate the
distribution of characters. Botaniska Notiser 128: 119–147.
Dahlgren R M T, Clifford H T, Yeo P F. 1985. The Families of the Monocotyledons. Structure, Evolution and
Taxonomy. Berlin: Springer-Verlag.
de Candolle A. 1878. Smilacacées. Monographia Phanerogamarum 1: 1–217.
Farris J S. 1969. A successive approximations approach to character weighting. Systematic Zoology 18:
374–385.
Farris J S. 1989. The retention index and the rescaled consistency index. Cladistics 5: 417–419.
Fay M F. Phylogenetics of Liliales and Liliaceae based on combined molecular analyses (abstract). In:
Columbus J T, Friar E A, Porter J M, Prince L M, Simpson M G eds. Monocots: Comparative Biology
and Evolution. Vol. 2. Rancho Santa Ana Botanic Garden. Claremont: California Press. (in press)
Felsenstein J. 1985. Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39:
783–791.
Fu C-X, Kong H-H, Qiu Y-Y, Cameron K M. 2005. A molecular phylogenetic study of the herbaceous, East
Asian-North American disjunct species of Smilax sect. nemexia (Smilacaceae). International Journal of
No. 2 CHEN S-C et al.: A phylogenetic analysis of the Smilacaceae based on morphological data 125
Plant Science 166: 301–309.
Heywood V H, Moore D M, Stearn W T. 1993. Flowering Plants of the World. Updated ed. New York: Oxford
University Press. 320.
Holmes W C. 2002. Smilacaceae. In: Flora of North America Editorial Committee ed. Flora of North America.
Vol. 26. New York: Oxford University Press.
Hutchinson J. 1973. The Families of Flowering Plants Arranged According to a New System Based on Their
Probable Phylogeny. 3rd ed. Oxford: Clarendon Press.
Judd W S. 1998. The Smilacaceae in the southeastern United States. Harvard Papers in Botany 3: 147–169.
Judd W S, Campbell C S, Kellog E A, Sterens S P F. 1999. Plant Systematics, a Phylogenetic Approach.
Massachusetts: Sinauer Associates Inc. Publishers. 183–184.
Koyama T. 1960. Materials toward a monograph of the genus Smilax. Quarterly Journal of Taiwan Museum
13: 1–61.
Koyama T. 1984. A taxonomic revision of the genus Heterosmilax (Smilacaceae). Brittonia 36: 184–205.
Kunth. 1850. Enum. Plant 5: 270.
Mabberley D J. 1997. The Plant-Book. 2nd ed. Cambridge: Cambridge University Press. 858–859.
Nixon K C. 1999. Winclada (beta) ver. 0.9.9. New York, Published by the Author.
Norton J F. 1916. The eastern and western migration into North America of Smilax. Journal of the Washington
Academy of Science 6: 281–283.
Rudall P J, Stobart K L, Hong W P, Conran J G, Furness C A, Kite G C, Chase M W. 2000. Consider the
lilies: systematics of Liliales. In: Wilson K L, Morrison D A eds. Monocots: Systematics and Evolution.
Victoria: CSIRO Publishing, Collingwood. 347–359.
Swofford D L. 2002. PAUP*, phylogenetic analysis using parsimony (* and other methods). Version 4.0b10.
Massachusetts: Sinauer Associates.
Takhtajan A L. 1987. Systema Magnoliophytorum. Leningrad: Nauka. 438.
Takhtajan A L. 1997. Diversity and Classification of Flowering Plants. New York: Columbia University Press.
Thorne R F. 1983. Proposed new realignments in the angiosperms. Nordic Journal of Botany 3: 85–117.
Thorne R F. 1992. An updated phylogenetic classification of the flowering plants. Aliso 13: 365–389.
Wang F-T (汪发缵), Tang T (唐进), Chen S-C (陈心启), Liang S-Y (梁松筠). 1978. Smilax L. and
Heterosmilax Kunth. In: Flora Reipublicae Popularis Sinicae (中国植物志). Beijing: Science Press. 15:
189–245.
菝葜科基于形态学证据的系统发育分析
1陈士超* 1邱英雄 1王艾丽 2Kenneth M. CAMERON 1傅承新**
1(浙江大学生命科学学院, 植物系统进化与生物多样性实验室 杭州 310058)
2(The Lewis B. and Dorothy Cullman Program for Molecular Systematics Studies, The New York Botanical Garden, Bronx, New
York 10458, USA)
摘要 对全世界范围分布的菝葜科Smilacaceae的79个代表种(包括了全部的属和组), 以分布于南美洲
的Philesia Comm. ex Juss.和Lapageria Ruiz & Pav.属为外类群, 选取包括花粉和染色体性状在内的47个
广义的形态学性状进行了分支分类系统发育分析, 同时以表征分类的方法构建了距离树(NJ)辅助分析,
首次对世界分布的菝葜科各属间及属内的系统发育关系作了探讨。(1)Ripogonum与菝葜属Smilax +肖菝
葜属Heterosmilax互为姐妹群, 但是距离较远, 支持将类菝葜属(新拟中文名)Ripogonum独立为科的观
点; (2)肖菝葜属在菝葜科内处于较为进化的分支上, 并与菝葜属土茯苓组sect. Coilanthus的部分种组成
一个具较高支持率(88%)的单系分支, 分析表明肖菝葜属并非是一个好属, 应归入菝葜属; (3)菝葜属6
个组的划分大都没有得到支持, 只有东亚北美间断分布的草本菝葜组sect. Nemexia的单系得到很好的
支持(93%); (4)分布于南美洲巴西的种类聚为一个单系类群, 表明它们可能有共同的起源, 但由于取样
局限, 南美洲种类的系统地位有待进一步研究。
关键词 菝葜科; 菝葜属; 肖菝葜属; 类菝葜属; 形态学; 分支分析; 系统发育