Phylogenetic placement of the enigmatic and critically endangered genus Saniculiphyllum (Saxifragaceae) inferred from combined analysis of plastid and nuclear DNA sequences

Abstract

Saniculiphyllum, a monotypic genus distributed in Southwest China, was thought to be extinct before our recent rediscovery. The taxonomic position of this genus has been enigmatic ever since its publication. It was originally treated as the only member of a distinct tribe Saniculiphylleae in the family Saxifragaceae. Some proposed a new family, Saniculophyllaceae, to accommodate this genus, although its affinities are clearly with members of Saxifragaceae. Here we analyzed six DNA regions, the nuclear ribosomal ITS and 26S rDNA and the plastid rbcL, matK, trnL-trnF, psbA-trnH genes, spacers, and intron to explore the phylogenetic position of Saniculiphyllum within Saxifragaceae. The combined nuclear and chloroplast dataset includes 63 ingroup species, representing all genera but Hieronymusia in the family. Results from likelihood, parsimony and Bayesian phylogenetic methods corroborate earlier results. Two clades of Saxifragaceae, the Heucheroid and Saxifragoid clades, were recovered. The topologies obtained from different analyses confirm the placement of Saniculiphyllum in Saxifragaceae, but our analyses reveal that Saniculiphyllum is embedded within the large Heucheroid clade. However, the closest relatives of Saniculiphyllum within the Heucheroid clade remain unclear. Combined with morphological data, our results suggest that Saniculiphyllum should best be regarded as a highly distinctive lineage within the Heucheroid clade of Saxifragaceae. Morphological novelties and conservation status of Saniculiphyllum are also presented.

Full text

This article appeared in a journal published by Elsevier. The attached
copy is furnished to the author for internal non-commercial research
and education use, including for instruction at the authors institution
and sharing with colleagues.
Other uses, including reproduction and distribution, or selling or
licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the
article (e.g. in Word or Tex form) to their personal website or
institutional repository. Authors requiring further information
regarding Elsevier’s archiving and manuscript policies are
encouraged to visit:
http://www.elsevier.com/copyright

Author's personal copy
Phylogenetic placement of the enigmatic and critically endangered genus
Saniculiphyllum (Saxifragaceae) inferred from combined analysis of plastid
and nuclear DNA sequences
Chun-Lei Xiang
a
, Matthew A. Gitzendanner
b
, Douglas E. Soltis
b
, Hua Peng
a
, Li-Gong Lei
a,
⇑
a
Key Laboratory of Biodiversity and Biogeography, Kunming Institute of Botany, Chinese Academy of Sciences, No. 132, Lanhei Road, Kunming, Yunnan 650201, PR China
b
Department of Biology, University of Florida, Gainesville, Florida 32611, USA
article info
Article history:
Received 12 July 2011
Revised 17 March 2012
Accepted 11 April 2012
Available online 24 April 2012
Keywords:
China
Endemic
Heucheroid clade
Saniculiphyllum guangxiense
Saxifragales
Phylogenetic analysis
abstract
Saniculiphyllum, a monotypic genus distributed in Southwest China, was thought to be extinct before our
recent rediscovery. The taxonomic position of this genus has been enigmatic ever since its publication. It
was originally treated as the only member of a distinct tribe Saniculiphylleae in the family Saxifragaceae.
Some proposed a new family, Saniculophyllaceae, to accommodate this genus, although its affinities are
clearly with members of Saxifragaceae. Here we analyzed six DNA regions, the nuclear ribosomal ITS and
26S rDNA and the plastid rbcL,matK,trnL-trnF,psbA-trnH genes, spacers, and intron to explore the phy-
logenetic position of Saniculiphyllum within Saxifragaceae. The combined nuclear and chloroplast dataset
includes 63 ingroup species, representing all genera but Hieronymusia in the family. Results from likeli-
hood, parsimony and Bayesian phylogenetic methods corroborate earlier results. Two clades of Saxifrag-
aceae, the Heucheroid and Saxifragoid clades, were recovered. The topologies obtained from different
analyses confirm the placement of Saniculiphyllum in Saxifragaceae, but our analyses reveal that Sanicu-
liphyllum is embedded within the large Heucheroid clade. However, the closest relatives of Saniculiphyl-
lum within the Heucheroid clade remain unclear. Combined with morphological data, our results suggest
that Saniculiphyllum should best be regarded as a highly distinctive lineage within the Heucheroid clade
of Saxifragaceae. Morphological novelties and conservation status of Saniculiphyllum are also presented.
Ó2012 Elsevier Inc. All rights reserved.
1. Introduction
Over the past two decades the traditionally recognized and
broadly defined Saxifragaceae s.l. (e.g., Cronquist, 1981) have been
shown to be polyphyletic (Chase et al., 1993; Morgan and Soltis,
1993; Soltis and Soltis, 1997; Soltis et al., 2000). As a result, the
family has experienced major revision (Johnson and Soltis, 1994,
1995; Soltis et al., 2001a; Soltis, 2007) and now it is generally trea-
ted as a number of smaller families. Recently, the APG classification
(APG III, 2009) clarified the circumscription of the family. Saxifrag-
aceae are now recognized as a modest sized family of 33 genera
and approximately 600 species, with a nearly worldwide distribu-
tion, but mainly found in temperate regions. Approximately one
half of the genera are monotypic (Soltis et al., 2001a; Soltis,
2007). Genera of Saxifragaceae from China include Astilbe Buch.-
Ham., Astilboides Engler, Bergenia Moench, Chrysosplenium L., Mitel-
la L., Mukdenia Koidzumi, Oresitrophe Bunge, Rodgersia A. Gray,
Saxifraga L., Tanakaea Franchet and Savat., Tiarella L., and the re-
cently described Saniculiphyllum C.Y. Wu and T.C. Ku (Wu and
Ku, 1992; Pan et al., 2001), which has been distinguished from
other genera of Saxifragaceae because of its unusual morphology,
including palmately lobed leaves, 3-5-carpellate and 3-5-loculed
ovary (Fig. 1).
The monotypic Saniculiphyllum, comprising the single species S.
guangxiense C.Y. Wu and T.C. Ku, is a clonal aquatic herb clinging to
the wet rocks or stone in brooks in southeast Yunnan Province and
Northwest of the Guangxi Zhuang Autonomous Region (Wu et al.,
2007). In the protologue to the description (Wu and Ku, 1992),
two specimens were cited. One was collected in 1989 (Guangxi
Zhuang Autonomous Region, Tianlin County), this specimen was
selected as HOLOTYPE (PE). Another specimen was collected in
1968 (Yunnan, Funing County, Lida), this specimen was a paratype
(KUN, PE). The species is considered to be endangered and was
actually thought to be extinct before the recent rediscovery of sev-
eral populations in Yunnan. The populations in Tianlin County of
Guangxi Zhuang Autonomous Region of southern China, from
where the type specimens of the species were collected in 1989,
had disappeared due to seasonal drying of the streams and plants
never been found in this area again.
1055-7903/$ - see front matter Ó2012 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.ympev.2012.04.010
⇑
Corresponding author. Fax: +86 871 5213916.
E-mail addresses: xiangchunlei@mail.kib.ac.cn (C.-L. Xiang), magitz@botany.u-
fl.edu (M.A. Gitzendanner), dsoltis@botany.ufl.edu (D.E. Soltis), hpeng@mail.ki-
b.ac.cn (H. Peng), ligonglei@mail.kib.ac.cn (L.-G. Lei).
Molecular Phylogenetics and Evolution 64 (2012) 357–367
Contents lists available at SciVerse ScienceDirect
Molecular Phylogenetics and Evolution
journal homepage: www.elsevier.com/locate/ympev

Author's personal copy
The systematic position of Saniculiphyllum has been enigmatic
ever since its publication. When establishing the genus, Wu and
Ku (1992) placed it within its own tribe, Saniculiphylleae within
Saxifragaceae, but the circumscription of the family Saxifragaceae
at that time was much broader compared to our current under-
standing. In the protologue (Wu and Ku, 1992), Saniculiphyllum
was considered to be related to Saxifraga and Mukdenia because
all have axile placentation. However, the number of petals, sepals,
stamens and habit of Saniculiphyllum is clearly different from the
latter two genera. Saniculiphyllum was also deemed to have a close
relationship to Chrysosplenium, but the latter genus has parietal
placentation. Compared with other genera such as Bolandra Gray,
Boykinia Nutt., Jepsonia Small, Peltoboykinia (Engler) H. Hara and
Sullivantia Torrey and Gray ex Gray, which all have axile placenta-
tion, Saniculiphyllum is easily distinguished by having a 10-lobed
floral disk, completely inferior ovary, short anther filament, and
creeping flat rhizomes (Fig. 1f, g, and i; Wu and Ku, 1992). Mabber-
ley (1997) elevated the genus to the rank of family, but as Sanicu-
lophyllum and Saniculophyllaceae. Wu et al. (2003) suggested that
the genus may be a highly specialized taxon within Saxifragaceae.
Based on its thick rhizomes and palmately lobed leaves, Soltis
(2007) thought that this genus might belong to the Darmera group
in Saxifragaceae. However, at the same time, he also stressed that it
was more appropriate at that point to consider the exact place-
ment of the genus within Saxifragaceae as unknown.
During the past two decades, tremendous progress has been
made in understanding the phylogeny of Saxifragaceae (Morgan
and Soltis, 1993; Johnson and Soltis, 1994; Soltis and Soltis,
1997; Soltis et al., 2001a,b). Saniculiphyllum, however, has never
been included in these comprehensive molecular phylogenetic
studies, because of its extremely rare occurrence and the unavail-
ability of material from which to obtain DNA. This Chinese ende-
mic remains one of the most taxonomically enigmatic members
of Saxifragaceae (Soltis, 2007). Furthermore, a better understand-
ing of the relationships of Saniculiphyllum may elucidate evolution-
ary processes across Saxifragaceae. The systematic position and
possible allies of Saniculiphyllum therefore require more rigorous
evaluation in a phylogenetic context.
Between 2008 and 2011, we found five new populations of San-
iculiphyllum guangxiense in Funing County of Yunnan in Southwest
Fig. 1. Photographs of Saniculiphyllum guangxiense in its natural habitat. (a and b) habitat of S.guangxiense in Funing County. (a) Group of individuals growing on cliff near
water; (b) the plants of S.guangxiense cling to stones in stream; (c) details of S.guangxiense showing the leaves palmately deeply lobed; (d and e) floral form of the species; (d)
cymes with 7–10 flowers; (e) flower, showing five red petals, five stamens, and three carpels; (f, g, and i) rhizomes and fibrous roots; (f) long and horizontal rhizomes cling to
stone; (g) densely fibrous roots; (i) details of fibrous roots under stereo microscope; (h) ovules; (j) details of seeds showing that these seeds generally hypogenetic. (For
interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
358 C.-L. Xiang et al. / Molecular Phylogenetics and Evolution 64 (2012) 357–367

Author's personal copy
China (Fig. 2). Unfortunately, we have not been able to find the spe-
cies again at the type collection site in Tianlin County of the Guan-
gxi Zhuang Autonomous Region in South China, despite our
repeated efforts. However, our discovery of new populations in
Yunnan Province enabled us to obtain DNA data to investigate
the systematic position of Saniculiphyllum.
In this study, we use nuclear [internal transcribed spacer (ITS),
and 26S] and chloroplast data [matK,rbcL,trnL-trnF (consisting of
the trnL intron and the trnL-trnF intergenic spacer), psbA-trnH]in
an attempt to elucidate the phylogenetic position of Saniculiphyl-
lum and provide new insights into the phylogeny of family Saxi-
fragaceae These sequences are widely used to assess the affinities
of taxonomically enigmatic angiosperm taxa (e.g., Herbert et al.,
2006; Chandler and Bayer, 2000; Bayer and Cross, 2002), as well
as phylogenetic relationships within Saxifragaceae (Hibsch-Jetter
et al., 1997; Soltis and Soltis, 1997; Morgan and Soltis, 1993; John-
son and Soltis, 1994, 1995; Soltis et al., 1996a,b, 2001a). Conse-
quently, a large data base of these sequences is now available
representing Saxifragaceae, permitting analyses across the phylo-
genetic diversity of the family.
Given the longstanding confusion regarding the relationship of
Saniculiphyllum to other genera within Saxifragaceae, the objec-
tives of this study were: (1) investigate the phylogenetic place-
ment of Saniculiphyllum in Saxifragaceae, (2) identify the major
lineage(s) that are related to Saniculiphyllum, (3) further contribute
to a comprehensive phylogenetic framework for the Saxifragaceae.
2. Materials and methods
2.1. Field work
Materials of Saniculiphyllum were collected from Ligong Village,
Lida Town, Funing County, Yunnan Province, Southwest China
(Fig. 2). Fresh leaves of two populations were collected for the
purpose of DNA extraction. Materials of the other ingroup and
outgroup taxa were either collected in the wild or obtained from
cultivated plants in Kunming Botanic Garden (Table 1). Voucher
specimens are deposited in the Herbarium of Kunming Institute
of Botany (KUN), Chinese Academy of Sciences.
2.2. Taxon sampling
Familial circumscriptions and nomenclature are based on the
treatment of the Saxifragaceae by Soltis et al. (2001a) and Soltis
(2007). Outgroup taxa were selected based on the well supported
placement of Iteaceae, Grossulariaceae, and Pterostemonaceae as
closest relatives to Saxifragaceae (Morgan and Soltis, 1993; Soltis
et al., 1993, 2001a; Johnson and Soltis, 1995). Therefore, one spe-
cies from Pterostemonaceae (Pterostemon rotundifolius Ramírez),
two species of Iteaceae (Itea virginica L., I.yunnanensis Franch.),
and five species of Grossulariaceae (Ribes glaciale Wall., R.maxi-
mowiczianum Kom., R.moupinense Franch. var. tripartitum (Batalin)
Jancz., R.soulieanum Jancz., R.tenue Jancz.) were selected as out-
group taxa.
The ingroup included 63 species, and included all genera of
Saxifragaceae except one, Hieronymusia Engler, a monotypic genus
restricted to remote areas of Argentina and Bolivia, for which we
have been unable to obtain suitable material. However, Hiero-
nymusia was actually included in Suksdorfia A. Gray by Gornall
and Bohm (1985) and is clearly closely related to that genus based
on morphology and chemistry. Therefore, following the Gornall
and Bohm (1985) treatment, all genera of Saxifragaceae were in-
cluded in our study.
In total, the study material consisted of 74 accessions represent-
ing 71 species, 25 of which were newly sequenced as part of this
study. Other sequences came from previous studies (Soltis et al.,
1991a, 1993, 1996a,b, 2001a; Soltis and Kuzoff, 1995; Johnson
and Soltis, 1995, 1998; Kuzoff et al., 1998, 1999; Conti et al.,
1999; Fishbein et al., 2001; Senters and Soltis, 2003; Okuyama
et al., 2005, 2008; Okuyama and Kato, 2009).
Fig. 2. Geographical distribution of Saniculiphyllum guangxiense in China. The solid circle is sample collecting site in Yunnan, and the filled box is the extinct distribution area
in Guangxi Zhuang Autonomous Region.
C.-L. Xiang et al. / Molecular Phylogenetics and Evolution 64 (2012) 357–367 359

Author's personal copy
Table 1
Voucher information and GenBank accession numbers for taxa used in this study. Vouch specimens are deposited in the following herbaria: K = Royalo Botanic Gardens, Kew, England; KUN = Kunming Institute of Botany, China;
OS = Ohio State University, United States; UC = University of California, United States; WA = University of Warsaw, Poland; WS = Washington State University, United States; UBC = University of British Columbia, Canada;
WTU = University of Washington, United States. If accession numbers of some sequences are not available from GenBank, then the source publications were listed. Regions not sampled are indicated by an en dash (–). A: Soltis et al.
(1993);B:Johnson and Soltis (1998);C:Johnson and Soltis (1995);D:Soltis and Kuzoff (1995);E:Soltis et al. (1996a);F:Soltis et al. (1996b). Asterisk: newly sequenced taxon in present study.
Taxon Voucher/herbarium Locality GenBank No./reference
matK rbcL psbA-trnH trnL-trnF ITS 26S
Ingroups
Bolandra oregana Grable 11668 (WS) Washington, United States L34117 U06209 AF374766 AF374809 U51255 AF374857
Boykinia rotundifolia Gornall 0101 (UBC) California, United States L34118 L11175 AF374767 AF374810 U51248 AF274638
Jepsonia parryi Rieseberg 1110 (WS) California, United States L34128 U06211 AF374765 AF374808 U51262 AF374856
Suksdorfia violacea Soltis & Soltis 2309 (WA) Washington, United States L34146 – – – U51257 –
Sullivantia oregana Gornall 11214 (WS) Oregon, United States L34113 U06219 AF374768 AF374811 U51258 AF274668
Telesonix hercheriformis Wolf 151 (WS) Wyoming, United States L34148 U06221 AF374764 AF374807 U51261 AF374855
Astilbe microphylla Botanical Gardens, Univ. Tokyo, Cult. (WS) Japan, Cult. CQ386964 – AF374771 AF374814 B AF374860
A. chinensisi⁄Yin et al. 1862 (KUN) Yunnan, China JN102180 JN102248 JN102204 JN102272 – JN102223
A. rivularis⁄Xiang 443 (KUN) Yunnan, China JN102181 JN102249 JN102205 JN102273 – JN102224
Saxifragopsis fragarioides UCBG 82-1325 (UC) California, United States AF37429 A AF374772 AF374815 B AF374861
Bensoniella oregona Lang, Soltis & Soltis, s.n. (WS) United States L34112 – – – AF158953 –
Conimitella williamsii Soltis & Soltis 1608 (WS) United States L34122 – – – AB292020 –
Elmera racemosa Soltis & Soltis 2179 (WS) Washington, United States L34124 U06210 AF374761 AF374804 D AF374853
Heuchera micrantha Soltis & Soltis 1949 (WS) United States C L01925 AF374763 AF374806 D AF374854
Lithophragma trifoliatum Kuzoff 95-03 (WS) California, United States – – – – AF158951 AF036501
Mitella nuda Johnson & Brunsfeld 1908 (WS) Precise locality unknown L34134 – AB492501 AB116715 AB163495 –
Tellima grandiflora Soltis & Soltis 2113 (WS) Alaska, United States L34149 U06222 AF374762 AF374805 D AF036500
Tiarella polyphylla Soltis 2555 (WS) Japan AB116692 – – AB116717 AF006834 –
Tolmiea menziesii Soltis & Soltis 1903 (WS) Oregon, United States L34152 U06223 AF374760 AF374803 D AF374852
Micranthes tolmiei WS 32167 (WS) Precise locality unknown AF115484 E AF374756 AF374799 B AF374849
M. integrifolia Soltis & Soltis 2253 (WS) Oregon, United States L20131 L01953 AF374758 AF374801 B AF274666
M. punctata Soltis & Soltis 2217 (WS) Washington, United States E A AF374757 AF374800 B AF374850
M. pallida⁄Ying et al. 1342 (KUN) Yunnan, China JN102182 JN102250 JN102206 JN102274 – JN102225
M. stellaris Horandl 2703 (WS) Precise locality unknown AF115493 AF374732 AF374759 AF374802 AF374827 AF374851
Chrysosplenium iowense Wendel s.n. (WS) Iowa, United States L34120 L19935 AF374747 AF374790 B AF274641
C. hydrocotylifolium⁄Lei 20090402 (KUN) Yunnan, China JN102183 JN102251 JN102207 JN102275 JN102226 –
C. lanuginosum var. gracile⁄Gan 1912 (KUN) Hubei, China JN102184 JN102252 JN102208 JN102276 JN102227 –
C. nepalense⁄Gan 1875 (KUN) Hubei, China JN102185 JN102253 JN102209 JN102277 JN102228 –
C. microspermum⁄Gan 2063 (KUN) Hubei, China – – JN102210 JN102278 JN102229 –
Peltoboykinia tellimoides Nikko Botanical Garden, cult. Japan L34138 U06213 AF374746 AF374789 AB248847 AF036499
Astilboides tabularis Palmengarten Botanical Gardens, Germany, Cult.
(WS)
Palmengarten Botanical Gardens, Germany, Cult. L34115 U06207 AF374750 AF374793 B AF374843
Bergenia purpurascens⁄Lei KBG01. (KUN) Yunnan, China JN102186 JN102254 JN102211 JN102279 JN102230 –
B. cordifolia Komarov Botanical Institute, Russia, Cult. (WS) Komarov Botanical Institute, Russia, Cult.
Darmera peltata University of California, Berkley (WS) University of California, Berkley, UC Cult. L34123 L11180 AF374752 AF374795 D AF374845
Mukdenia rosii University British Columbia, Botanical Gardens (WS) University British Columbia, Botanical Gardens,
Cult.
L34137 U06212 AF374751 AF374794 B AF374844
Oresitrophe rupifraga Beijing Botanical Gardens, China, Cult. (WS) Beijing Botancal Gardens, China, Cult. E E AF374749 AF374792 B AF374842
O. rupifraga⁄Lei KBG02 (KUN) Yunnan, China JN102187 JN102255 JN102212 JN102280 JN102231 –
Rodgersia aesculifolia⁄Yin et al. 1668 (KUN) Yunnan, China JN102188 JN102256 JN102213 JN102281 JN102232 –
Bergenia cordifolia Komarov Botanical Institute, Russia, Cult. (WS) L34116 U06208 AF374753 AF374796 D AF374846
R. pinnata Palmengarten Botanical Gardens, Germany, Cult.
(WS)
Palmengarten Botanical Gardens, Germany, Cult. L34139 U06214 AF374748 AF374791 U51264 AF374841
Leptarrhena pyrolifolia Soltis & Soltis 2237 (WS) Vancouver Island, Canada L34129 L11191 AF374769 AF374812 F AF374858
Tanakaea radicans Nikko Botanical Garden, Japan (WS) Japan L34147 U06220 AF374770 AF374813 U51263 AF374859
Saniculiphyllum guangxiense⁄Lei 20090403-1 (KUN) Yunnan, China JN102189 JN102257 JN102214 JN102282 JN102233 –
Saniculiphyllum guangxiense⁄Lei 20090403-5 (KUN) Yunnan, China JN102190 JN102258 JN102215 JN102283 JN102234 –
Cascadia nuttallii University of Washington 3446 (WTU) Oregon, United States AF115483 A AF374755 AF374798 B AF374848
Saxifragodes albowiana Arroyo et al. 941179 (WS) Chile AF374729 AF374731 AF374754 AF374797 AF374825 AF374847
Saxifraga aizoides Brochmann 92-78-1 (OS) Precise locality unknown E E AF374744 AF374787 AF087594 AF374839
S. balfourii⁄Yin et al. 2265 (KUN) Yunnan, China JN102191 JN102259 JN102216 JN102284 JN102235 –
360 C.-L. Xiang et al. / Molecular Phylogenetics and Evolution 64 (2012) 357–367

Author's personal copy
2.3. DNA extraction, PCR amplification and sequencing
Total DNA was obtained from freshly-collected and silica-gel-
dried leaf fragments. All accessions were identified using published
keys and compared to herbarium specimens. Total genomic DNA
was isolated using CTAB procedure of Doyle and Doyle (1987) as
modified by Soltis et al. (1991b) for fresh or frozen samples, or
the Bioteke’s plant mini Kit (Bioteke Corporation, Beijing, China)
following the manufacturer’s protocol. After extraction, the DNA
was resuspended in TE buffer and kept at 40 °C for further use.
PCR amplifications were performed using a Biometra T1 ther-
mocycler (Biometra, Göttingen, Germany). The 50
l
l volume poly-
merase chain reaction (PCR) contained 2
l
l DNA solution (adjusted
to approximately 20 ng), 5
l
l PCR reaction buffer, 5
l
l dNPT mix
(0.2 mM), 2
l
l of each primer, and 1.5 U Taq DNA polymerase
(Chenlü, Kunming, China).
Amplification and sequencing of the ITS region (ITS1, 5.8S rDNA,
ITS2) were performed with primers ITS4 and ITS5 (White et al.,
1990) or N-nc18S10 and C26A (Wen and Zimmer, 1996). PCR and
sequencing of the trnL intron and trnL-trnF intergenic spacer was
performed using the universal primers of Taberlet et al. (1991),
either as one fragment using primers ‘‘c’’ and ‘‘f’’ or as two separate
fragments using primers ‘‘c’’ and ‘‘d’’, and ‘‘e’’ and ‘‘f’’, respectively.
The primers used for amplifying and sequencing the psbA-trnH re-
gion were ‘‘psbA’’ and ‘‘trnH’’ as described in Hamilton (1999).
Amplifications were performed using a program consisting of
3 min at 94 °C followed by 35 cycles of 45 s denaturation (94 °C),
1 min annealing (53 °C) and 3 min extension (72 °C), ending with
a final 7 min extension at 72 °C. The PCR amplification protocols
were identical for all above three fragments.
The matK region was divided into three overlapping fragments
using the following PCR primer combinations for most taxa:
trnK-3914F and matK-1470R; trnK-710F and matK-2200R; matK-
1412F and trnK-2R, as sequencing primers. The base composition
of these primers is given in Johnson and Soltis (1994, 1995). For
some taxa, these primers failed to produce usable double-stranded
products, thus the primers matK-3268F, and matK-2200R of Okuy-
ama et al. (2005) were used. The amplification conditions were set
as follows: denaturation at 94 °C for 4 min, 30 cycles at 94 °C for
30 s, 55 °C for 30 s, 72 °C for 2 min, and a final extension of 7 min
at 72 °C.
Primers Z1 and Z-1351R were used for amplifying and sequenc-
ing the rbcL region (Chandler and Bayer, 2000). The PCR conditions
used were: 2 min at 94 °C, then 30 cycles with 45 s at 94 °C, 90 s at
45 °C and 90 s at 72 °C, and finally 5 min at 72 °C.
Amplified products were purified using a QIAquick PCR purifica-
tion Kit (BioTeke Corporation, Beijing, China) following the manu-
facturer’s instructions. Sequencing reactions were performed with
the dideoxy chain termination method running on an ABI PRISM
3730 automated sequencer. The same primers described above
for PCR were used for the sequencing reactions. All regions were
sequenced for both strands where there was an overlap of at least
70%. All sequences used in this study together with their GenBank
accession numbers or source publications are listed in Table 1.
2.4. Sequence analysis and phylogenetic reconstruction
Datasets for 26S rDNA and rbcL were easily aligned by eye. The
remaining datasets, which were much more difficult to align, were
aligned using an iterative strategy. The simultaneous alignment
and tree estimation program SATé (Liu et al., 2009; Yu and Holder,
2010) was used to obtain alignments and single gene phylogenetic
estimates for ITS, matK,psbA-trnH and trnL-trnF. SATé is not able to
perform alignments of multi-gene datasets, so each gene region
was analyzed individually. These alignments were then concate-
nated and analyzed with RAxML 7.2.7 (Stamatakis, 2006). As the
S. brunonis⁄Yin et al. 2169 (KUN) Yunnan, China JN102192 JN102260 – JN102285 JN102236 –
S. cernua Brochmann 92-31-22 (OS) Alaska, United States L34140 U06215 AF374736 AF374779 B AF374831
S. cymbalaria Ferguson 1994-04 (WS) Precise locality unknown E E AF374734 AF374777 AF087599 AF374830
S. fortunei Soltis 2519 (WS) Precise locality unknown E E AF374737 AF375780 AF374821 AF374832
S. hispidula⁄Yin et al. 1951 (KUN) Yunnan, China JN102193 JN102261 – JN102286 JN102237 –
S. mertensiana Saxifraga mertensiana Bong (WS) Oregon, United States L34142 U06216 AF374735 AF374778 AY231367 AF036498
S. oppositifolia Kew 1975-4135, Cult. (K) Royal Botanic Gardens, Kew E E AF374739 AF374782 B AF374834
S. osloensis Brochmann s.n. (WS) Precise locality unknown E E AF374745 AF374788 AF087608 AF374840
S. rotundifolia Gornall 0101 (UBC) Precise locality unknown L34118 L11175 AF374767 AF374810 U51248 AF374835
S. scardica Kew 361-84-03742, Cult. (K) Precise locality unknown E E AF374741 AF374784 B AF374836
S. spathularis BBG 001-91-75-10, Cult. (K) Precise locality unknown E E AF374742 AF374785 AF087596 AF374837
S. stolonifera 1⁄Lei s. n. (KUN) Kunming Botanic Garden, Cult. JN102194 JN102262 – JN102287 JN102238 –
S. stolonifera 2⁄Yin et al. 1276 (KUN) Yunnan, China JN102195 JN102263 JN102216 JN102288 JN102239 –
S. strigosa var. ramosa⁄Yin et al. 2109 (KUN) Yunnan, China JN102196 JN102264 – JN102289 JN102240 –
S. tricuspidata Parker s.n. (WS) Precise locality unknown E E AF374743 AF374786 AF087601 AF374838
S. diversifolia var.
haematophylla⁄
Yin et al. 1923 (KUN) Yunnan, China JN102197 JN102265 – JN102290 JN102241 –
S. hirculus Brochmann 93-219 (OS) Precise locality unknown E E AF374738 AF375781 AF374823 AF374833
Saxifragella bicuspidata Arroyo et al. 950914 (?) Chile AF374728 AF374730 AF374733 AF374776 AF374819 AF374829
Outgroups
Itea virginica D. M. E. Ware 94 (WILI) United States AF274618 L11188 AF374775 AF374818 AY231368 AF274650
I. yunnanensis⁄Xiang 444 (KUN) Yunnan, China JN102198 JN102266 JN102217 JN102291 JN102242 –
Pterostemon rotundifolius Jordan s.n. (WS) Mexico AF274630 L11203 AF374774 AF374817 AY231369 AF274663
Ribes aureum Soltis & Soltis 2220 (WS) Washington, United States L34153 L11204 AF374773 AF374816 AF426382 AF274665
R. glaciale⁄Yin et al. 1675 (KUN) Yunnan, China JN102199 JN102267 JN102218 JN102292 JN102243 –
R. maximowiczianum⁄Yin et al. 1601 (KUN) Yunnan, China JN102200 JN102268 JN102219 JN102293 JN102244 –
R. moupinense var. tripartitum⁄Yin & Dong 0438 (KUN) Yunnan, China JN102201 JN102269 JN102220 JN102294 JN102245 –
R. soulieanum⁄Yin & Dong 0434 (KUN) Yunnan, China JN102202 JN102270 JN10221 JN102295 JN102246 –
R. tenue⁄Yin et al. 1935 (KUN) Yunnan, China JN102203 JN102271 JN102222 JN102296 JN102247 –
C.-L. Xiang et al. / Molecular Phylogenetics and Evolution 64 (2012) 357–367 361

Author's personal copy
SATé alignments had a lot of gaps, we used Phyutility (Smith and
Dunn, 2008) to eliminate positions where more than 60% of the
taxa had gap characters for that position. This dataset was then
analyzed with RAxML with 1000 bootstrap replicates. SATé fre-
quently made little or no improvement over the initial MAFFT
alignment on single gene datasets even after 24 h of iteration over
alignment and tree building. Thus, we redid the alignments with
SATé using the topology of the tree obtained from the analysis of
the reduced dataset above as a starting topology in SATé. Effec-
tively this is similar to the iterative process SATé employs, but
making use of the information in the multi-gene phylogeny, rather
than being limited by each single gene topology. Alignments gen-
erated from the initial unguided alignment were different than
those obtained using the starting tree, though no formal statistics
were calculated. After this realignment in SATé, datasets were once
again concatenated and reanalyzed with RAxML, again with 1000
bootstrap replicates and six independent gene-based partitions
(estimating a single set of branch lengths). Data alignment and
trees were deposited in Treebase (Study number S12254).
The final SATé alignment was taken as the best available align-
ment of the genes and used for additional analyses of the dataset
with maximum parsimony (MP) (Swofford et al., 1996) and Bayes-
ian inference (BI) (Ronquist and Huelsenbeck, 2003). Maximum
parsimony analyses were performed with the heuristic search op-
tion of the computer package: Phylogenetic Analysis Using Parsi-
mony (PAUP

) version 4.0b10 (Swofford, 2003). Gaps were
treated as missing data. All unambiguous characters and charac-
ter-transformations were unordered and weighted equally. Sup-
port for clades was calculated via bootstrap analyses (Felsenstein,
1985) from 1000 replicates as described earlier (Li et al., 2009).
Bayesian inference (BI) based on Markov Chain Monte Carlo
methods (Yang and Rannala, 1997) was carried out with MrBayes
version 3.1.2 parallel version (Ronquist and Huelsenbeck, 2003).
Prior to analysis, the appropriate nucleotide sequence evolution
model for each marker was selected using the Akaike Information
Criterion (AIC) as implemented in jModeltest version 0.1.1 (Guidon
and Gascuel, 2003; Posada, 2008). The TVM + G model was shown
to best fit our data of matK,trnL-trnF,psbA-trnH, the GTR + I + G
model best fit rbcL, and GTR + G was best for ITS and the combined
datasets. Model parameters were estimated directly during the
runs. For each analysis, two simultaneous runs were performed,
starting from random trees for 2 10
6
generations, having three
heated and one cold chain. Markov chains were sampled every
100th generation.
At the end of the run we considered the sampling of the poster-
ior distribution to be adequate if the average standard deviation of
split frequencies was <0.01 (Ronquist et al., 2005). To assess
whether the MCMC chain reached stationarity we examined the
InL plots using Tracer V. 1.5.0 (Rambaut and Drummond, 2009).
Also, to visually check for convergence we used the program AWTY
online (Wilgenbusch et al., 2004). The states of the chain that were
sampled before stationarity (i.e., the ‘‘burn in’’ of the chain) were
discarded, and the posterior probability (PP) values were deter-
mined from the remaining trees.
The Shimodaira–Hasegawa (SH) test (Shimodaira and Hase-
gawa, 1999) was employed in the CONSEL package (Shimodaira
and Hasegawa, 2001) to compare the likelihood scores of trees de-
rived from two different data partitions (nuclear vs. plastid).
3. Results
3.1. Sequence and alignment
Sequence lengths were as follows in Saniculiphyllum guangx-
iense: 841 nucleotides (nt) for the trnL-trnF spacer, 274 nt for the
psbA-trnH spacer, 1402 nt for matK, 1296 nt for rbcL, and 666 nt
for the ITS1-5.8S-ITS2 region. The resulting combined and aligned
sequence matrix contained 7137 positions (including gaps) of
which 1350 positions belong to the trnL-trnF partition, 804 posi-
tions to the psbA-trnH partition, 1512 positions to the matK parti-
tion, 1461 positions to the rbcL partition, 819 positions to the ITS
partition, and 26S gene contributed 1191 bp. Of the 7137 nucleo-
tides 41% were variable (29% parsimony-informative) in the data-
set. Table 2 summarizes the properties contributed by each
aligned data partition.
3.2. Incongruence among data partitions
We divided the combined dataset into two process partitions:
chloroplast dataset (trnL-trnF,psbA-trnH,matK, and rbcL); and nu-
clear dataset (ITS and 26S). While the results of the SH test were
highly significant (P= 0.000), visual inspection indicates that there
are very few ‘‘hard’’ conflicts between the nuclear vs. plastid trees
(e.g., Bull et al., 1993; Mason-Gamer and Kellogg, 1996; Oh and
Potter, 2005; Quicke et al., 2007). In this study, two large clades,
Saxifragoid clade and Heucheroid clade, were well supported in
both plastid dataset (ML BS: 100, 75, Supplementary Fig. 1) and nu-
clear dataset (ML BS: 100, 100, Supplementary Fig. 2). Therefore,
we combined the datasets for simultaneous analyses.
3.3. Phylogenetic analyses
Phylogenetic relationships are very similar between the total
evidence trees obtained in the ML, MP and BI analyses (Supple-
mentary Figs. 3 and 4), although somewhat lower resolution was
obtained with MP and BI. The ML topology from the combined
dataset will therefore be the primary tree for discussion of phylo-
genetic relationships.
Analyses of the combined dataset yielded a single ML tree
(Fig. 3), and the topology obtained is basically congruent with a
previous study based on a smaller number of taxa (e.g. Soltis
et al., 2001a). The ingroup (Saxifragaceae) is well supported as
monophyletic (ML BS: 100, MP BS: 100, PP: 1.00; all values re-
ported in this order below). The ingroup consists of two large
Table 2
The statistics from analyses of the chloroplast and nuclear data sets for parsimony analysis.
Data matrix Aligned
positions
No. informative
sites
No. MPTs Tree Length CI (excluding
uninformative characters)
RI RC
matK 1512 494 36 1683 0.633 0.803 0.508
rbcL 1461 174 313 653 0.538 0.781 0.420
trnL-trnF 1350 363 110 1249 0.692 0.820 0.567
psbA-trnH 804 365 26 1645 0.535 0.687 0.367
Combined chloroplast data matrix 5127 1396 24 5582 0.566 0.734 0.416
ITS 819 481 24 3695 0.329 0.706 0.232
ITS + 26S 2010 657 45 4560 0.317 0.695 0.241
All combined data matrix 7137 2053 2 10,198 0.465 0.711 0.330
362 C.-L. Xiang et al. / Molecular Phylogenetics and Evolution 64 (2012) 357–367

Author's personal copy
clades in the combined molecular dataset. One of these is the Saxi-
fragoid clade (100, 100, 1.00) comprising the genera Saxifraga s.s.
and Saxifragella Engler. Many relationships within Saxifraga s.s.
are well supported. Saxifraga mertensiana,S.stolonifera, and S.for-
tunei form a well-supported clade (100, 100, 1.00) that is sister to
the remaining members of the clade (100, 100, 1.00). The next
branching member is Saxifragella, which is sister to the remainder
of the Saxifragoid clade, which is again well supported (100, 100,
1.00). Saxifraga tricuspidata then follows as sister to a well-sup-
ported (100, 100, 1.00) core group of taxa, as evidenced by Soltis
et al. (2001a).
The second major subclade of Saxifragaceae is the Heucheroid
clade which receives low support in the present study (73, 68,
1.00). Saniculiphyllum is embedded well within the Heucheroids.
Fig. 3. Maximum likelihood analysis based on a combined data set of trnL-trnF,psbA-trnH,matK,rbcL, ITS, and 26S sequences. All branches are drawn to scale. The clade
names for the family Saxifragaceae sensu stricto are those suggested by Soltis (2007); the geographical distribution of each taxon is shown on the tree. RAxML and MP BS
support values (>50%) are indicated above or below the branches as RAxML BS/MP BS. In Bayesian analysis, PP > 0.95 are indicated with thick branch. GenBank accession
numbers for the six genes of each species were presented in Table 1.
C.-L. Xiang et al. / Molecular Phylogenetics and Evolution 64 (2012) 357–367 363

Author's personal copy
In our study a well supported (100, 100, 1.00) Saxifragodes +Casca-
dia clade is sister to the remaining Heucheroids which form a
weakly supported clade (70, –, 0.96). Following Saxifragodes +Cas-
cadia a well supported (100, 100, 1.0) Boykinia group is then sister
to the remainder of the Heucheroids which also form a weakly sup-
ported clade (61, 91, 1.00). Following the Boykinia group, a well-
supported (100, 100, 1.00) Astilbe +Saxifragopsis clade is sister to
the remaining Heucheroids (59, –, –).
At this point in the Heucheroid topology, the two samples of
Saniculiphyllum formed a well-supported clade (100, 100, 1.00)
and are sister to the remaining Heucheroids, which form a clade
that is only weakly supported (64, –, 1.00). These remaining Heu-
cheroids comprise: Leptarrhena +Tanakaea (100, 100, 1.00), Dar-
mera group (99, 95, 1.00; Astilboides,Bergenia,Darmera Voss,
Mukdenia,Oresitrophe Bunge, and Rodgersia), Peltoboykinia group
(89, 98, 1.00; Chrysosplenium and Peltoboykinia), Micranthes group
[100, 100, 1.00; formerly Saxifraga section Micranthes (Haw.) D.
Don], and Heuchera group (100, 100, 1.00; Bensoniella C.V. Morton,
Conimitella Rydb., Elmera Rydb., Heuchera L., Lithophragma (Nutt.)
Torr. and A. Gray, Mitella,Tellima R. Br., Tiarella, as well as Tolmiea
Torr. and A. Gray) (Fig. 3).
4. Discussion
4.1. Overview of Saxifragaceae and placement of Saniculiphyllum
The present study is based on a combined analysis of four
molecular markers matK,psbA-trnH,rbcL,trnL-trnF from the chlo-
roplast genome, and two molecular markers ITS, 26S rDNA from
the nuclear genome; these have been used previously to infer rela-
tionships within family Saxifragaceae and among Saxifragales
(Fishbein et al., 2001; Soltis et al., 2001a). While our current study
represents increased taxonomic sampling (25 additional samples
were employed) from previous work (Soltis et al., 1991a,b, 1993,
2001a; Morgan and Soltis, 1993; Johnson and Soltis, 1995; Soltis
and Soltis, 1997), the phylogenetic tree obtained here is in close
agreement with previously published results based on molecular
data (Soltis et al., 2001a). Support values are also comparable, with
all groups that receive strong support in agreement with previous
studies relying on the same molecular markers (e.g., Soltis et al.,
2001a).
This major split determined here for Saxifragaceae based on
DNA sequence analyses (Saxifragoids vs. Heucheroids) agrees with
previous studies and is accompanied by some general morpholog-
ical differences. The Saxifragoid clade receives strong support (100,
100, 1.0), and Saxifraga is the core member of this clade (Soltis,
2007). As reviewed elsewhere (Webb and Gornall, 1989; Soltis,
2007), Saxifraga s.s. has a relatively uniform floral morphology.
The genus usually has actinomorphic flowers, 5 sepals, 5 petals,
10 stamens, and 2 carpels.
Saniculiphyllum is embedded within the second large subclade
of Saxifragaceae, the Heucheroid clade, which here receives low
support (73, 68, 1.00). The Heucheroid clade encompasses more
morphological variation than seen in the Saxifragoid clade (Soltis,
2007). For example, the Heucheroid clade includes actinomorphic
and zygomorphic forms, as well as variation in the number of se-
pals, petals, stamens and carpels.
4.2. Saniculiphyllum as a distinct lineage within Saxifragaceae
The distinctive morphology and phylogenetic placement of San-
iculiphyllum together suggest that the genus is not a member of any
of the well-supported subclades (groups) recognized previously
within Saxifragaceae (e.g., Heuchera,Boykinia,Darmera groups,
etc.; see Soltis, 2007). As noted, our analyses place Saniculiphyllum
as embedded well within the Heucheroids, a clade that receives
low support (73, 68, 1.00). Within the Heucheroid clade, Saxifrag-
odes +Cascadia, followed by the Boykinia group and then
Astilbe +Saxifragopsis, are subsequent sisters to the remainder of
the Heucheroids, which in turn form a weakly supported subclade
(59, –, –). Saniculiphyllum appears as sister to the remaining genera
of Heucheroids, a clade which also receives low support (64, –,
1.00). These remaining members of the Heucheroid clade comprise,
however, a number of well-supported subclades: Heuchera group
(100, 100, 1.00), Micranthes (100, 100, 1.00), Peltoboykinia/Chry-
sosplenium (100, 98, 1.00); Darmera group (100, 94, 1.00) (reviewed
in Soltis, 2007). Hence, relationships among the well-supported
subclades (groups) in the Heucheroid clade are poorly supported,
as was the case in Soltis et al. (2001a). This lack of resolution is
probably the result of a relatively rapid radiation (Soltis, 2007).
Thus, while we can conclude that Saniculiphyllum appears to be a
member of the Heucheroid clade, the relationship of the genus to
other Heucheroids remains uncertain.
The systematic position of Saniculiphyllum has been heavily de-
bated. When establishing the genus, Wu and Ku (1992) placed San-
iculiphyllum in subfamily Saxifragoideae (sensu Engler, 1930)asa
tribe (Saniculiphylleae). In the original description, they thought
the tribe Saniculiphylleae was closely related to tribe Saxifrageae
because both have axile placentation, but differ in the number of
petals, sepals, and stamens. Saniculiphyllum was also considered
to be related to Chrysosplenium in that some species of the latter
genus share some characters with Saniculiphyllum (e.g., short styles
exserted from a thick disk). But, Chrysosplenium is characterized by
parietal placentation which differs from Saniculiphyllum.
When all morphological data are considered, Saniculiphyllum is
a highly distinctive and unusual taxon, and it is quite different
from other members of Saxifragaceae in having a 10-lobed floral
disk, inferior ovary, short filament and creeping flat rhizomes, as
well as 3–5 carpels as reported in the present study. Because of
its distinctive characters, Mabberley (1997) elevated the genus to
the rank of a family as Saniculophyllaceae. Although the present
study clearly places the genus within Saxifragaceae (and within
the Heucheroids), the uncertain position of Saniculiphyllum within
the Heucheroids agrees with the highly distinctive morphology of
the genus. Results from our analyses suggest that Saniculiphyllum is
embedded within the Heucheroid clade and is sister to a clade
comprising five groups including Leptarrheana group, Darmera
group, Peltoboykinia group, Micranthes group, and Heuchera group,
but this position receives low support. Thus, Saniculiphyllum can be
regarded as a highly distinctive lineage within the Heucheroid
clade of Saxifragaceae based on a combination of evidence from
molecular phylogeny and morphological characters.
4.3. Comparison of topologies with earlier analyses
Despite strong overall similarities there are also some notewor-
thy differences from the trees we report here and the topology
recovered in Soltis et al. (2001a). There are two likely contributing
factors to differences between the topologies of Soltis et al. (2001a)
and those provided here. First, the present study has included 25
additional samples, including a distinctive genus, Saniculiphyllum.
The addition of Saniculiphyllum could impact relationships among
members of the Heucheroid clade, particularly given the poor
internal support in this part of the tree. Perhaps a more important
factor is sequence alignment, particularly alignment of the nuclear
ITS region, and the trnL-trnF, psbA-trnH plastid regions, as noted
below.
The total evidence topology of Soltis et al. (2001a) is highly sim-
ilar to that obtained here, but the relationships within the Heuche-
roid clade differ between the two studies. The Heucheroids form
two major subclades in Soltis et al. (2001a). The Boykinia group
364 C.-L. Xiang et al. / Molecular Phylogenetics and Evolution 64 (2012) 357–367

Author's personal copy
(BS = 100%) + Leptarrhena +Tanakaea group (BS = 100%) and
Astilbe +Saxifragopsis (BS = 100%) form one weakly supported ma-
jor subclade (BS = 58%). A clade of the Heuchera group
(BS = 100%), Darmera group (BS = 93%), Micranthes group
(BS = 100%), Cascadia +Saxifragodes (BS = 100%), and Peltoboyki-
nia +Chrysosplenium (BS = 100%) form a second weakly supported
major lineage within Saxifragaceae (BS = 75%). In contrast, in the
present study, the Heucheroids do not form two major subclades.
Differences between the total evidence topology presented here
and in Soltis et al. (2001a) are not well-supported. In both analyses
there are well-supported smaller clades (e.g., Heuchera,Darmera,
Micranthes,Peltoboykinia +Chrysosplenium groups, etc.), but rela-
tionships among these groups are generally poorly supported in
both studies.
The plastid tree of Soltis et al. (2001a) is very similar to the plas-
tid tree reported here. Heucheroid and Saxifragoid clades are
recovered in both analyses and both are well-supported. Relation-
ships within the Heucheroid clade differ in the placement of some
of the groups (e.g., Astilbe +Saxifragopsis), but few relationships
among these groups have support >50%. In the present study, San-
iculiphyllum is sister to Leptarrhena +Tanakaea, but with low sup-
port (BS = 68%); Saniculiphyllum was not included in Soltis et al.
(2001a). The matK and rbcL regions are straightforward in terms
of alignment and the major split of Heucheroids and Saxifragoids
was apparent even with matK alone and the two genes combined
revealed the two subclades with strong support (Soltis et al.,
1996b). However, the trnL-trnF, psbA-trnH plastid regions are very
difficult to align across the entire Saxifragaceae and whereas these
regions were aligned by eye in Soltis et al. (2001a), alignment in
the present study was done computationally. The close agreement
of the Soltis et al. (2001a) plastid tree and the plastid tree recov-
ered here is not surprising and likely reflects the strong matK +rbcL
framework.
The biggest difference between the present study and Soltis
et al. (2001a) is in the ITS + 26S tree. In the present study the Heu-
cheroids are not monophyletic. Instead, the Saxifragoid clade is sis-
ter to a subset of the Heucheroids: Darmera,Micranthes, and
Chrysosplenium +Peltoboykinia. This clade of Saxifragoids plus
some Heucheroids receives moderate support (BS = 84%). However,
ITS + 26S provided minimal resolution in Soltis (2007). Saxifragoids
were recovered with strong support (BS = 100); but Heucheroids
formed a clade but without BS support >50% and relationships
throughout the tree were poorly supported.
This difference in nuclear trees (ITS + 26S) is probably best
attributed to alignment differences with ITS. The 26S rDNA region
was straightforward to align, but is highly conserved and provides
little information at this level. The highly variable ITS region was
easy to align by eye within subgroups (e.g., within the Boykinia
group, or Darmera group, etc.), but very difficult to align among
these well-supported groups. In Soltis et al. (2001a) the alignment
was done by eye; in the present study, alignment was accom-
plished, as noted, with SATé. We may be pushing the limits of
the utility of the entire ITS region at this taxonomic depth in
Saxifragaceae.
4.4. The search for synapomorphies
Incongruence between morphology and molecular data is com-
mon in angiosperms, for example in Saxifraga (Vargas, 2000), as
well as in bryophytes, such as Leptodon D. Mohr (Sotiaux et al.,
2009) and Neckera Hedw. (Olsson et al., 2011). The Saxifragaceae
are a morphologically highly diverse group, variability of characters
such as carpels, sepals, petals, stamens, is generally considered to
be of taxonomic importance. Morphological characters supporting
the clades found in the molecular analyses are as yet unknown or
limited, although some morphological trends, which corroborate
the relationships based on DNA sequence date, are becoming appar-
ent (Nakazawa et al., 1997; Soltis et al., 1993, 2001a). While there
are some broad morphological trends within clades it remains dif-
ficult to identify morphological synapomorphies. The search for
morphological synapomorphies, especially in the Heucheroids
clade, remains a challenge given the enormous morphological var-
iation encompassed by members of this clade.
Earlier results on the morphological evolution in Saxifragaceae
(Soltis et al., 1991b, 2001a) showed that certain morphological
states have multiple origins. For example, at least three separate
instances of floral reduction (from 5 to fewer petals) occurred in
Saxifragaceae. In the present study our phylogenetic inferences im-
ply that several morphological character states, especially carpel
number that were held as unique and characteristic of Saniculiphyl-
lum, actually evolved independently. Most species of Saxifragaceae
standardly have 2 carpels, as do closely related genera (e.g., Itea,
Ribes). Five genera, Astilbe of Astilbe group, Conimitella,Micranthes,
Bergenia and Rodgersia of Heuchera group occasionally have 3 car-
pels. Only the genus Lithophragma of the Heuchera group, is charac-
terized by 3 carpels. However, based on our investigation of 491
flowers, Saniculiphyllum has three different carpel numbers. The
percentages of 3, 4 and 5 carpels are 28.72%, 45.62% and 25.66%
respectively, of which 4 and 5 carpels are unique to Saniculiphyllum
within Saxifragaceae (Fig. 4), while 2 carpels, which was reported
in the original description (Wu and Ku, 1992), could not be found
in our samples.
4.5. Distribution, ecology and conservation status of Saniculiphyllum
Saniculiphyllum grows only in wet habitats in Southwest China,
and water is a decisive environment factor. The water of the
streams have a pH of about 5.4 and an average temperature of
10–15 °C. Herbarium records indicate that the genus is confined
to Yunnan and Guangxi. The four known populations are confined
to an area about 10 square kilometers near a village in Funing
County of southwestern Yunnan of southwestern China. The type
and other early collections were from adjacent Guangxi Zhuang
Autonomous Region. But as noted plants had disappeared from
the type location due to continuing seasonal drought in southwest-
ern China in recent years. Plants of Saniculiphyllum generally cling
to stones in streams or wet, drippy rocks in water falls with dense
and fibrous adventitious roots. The vegetation type of the habitat of
S.guangxiense is broad-leaved evergreen forests dominated by
Cyclobalanopsis glaucoides Schottky (Fagaceae). Additional associ-
ated plants include Acorus calamus L. (Araceae), Elatostema sp.
(Urticaceae) and widespread cultivated species, Musa nana Lour.
(Musaceae).
Saniculiphyllum is highly threatened, and currently only known
from a single small area in Funing County from Yunnan Province
(Fig. 2). We found only five populations. Actually, five populations
were initially rediscovered, but one which was seen in 2009 at a site
nearest to the village disappeared in 2010 due to the polluted water
draining out from the village; now only four populations remained.
Hence, human activity has to be considered the biggest threat to the
survival of the species. The vegetation in this region is severely frag-
mented by roads and agricultural land. During our field surveys we
also visited another locality, Tianlin County from Guangxi Zhuang
Autonomous Region, where the type plants were collected. How-
ever, we were unable to find any plants at this site. It is hard to esti-
mate the exact number of plants that remain. Plants of the species
propagate through rhizomes and form clones on stones in the
streams or on cliffs near waterfalls. Two of the populations com-
prise perhaps 20–50 clones along streams. The two other popula-
tions occur near waterfalls and consist of a greater number of
clones. The recently published red-list of vascular plants from China
considers Saniculiphyllum to be in the Endangered IUCN category
C.-L. Xiang et al. / Molecular Phylogenetics and Evolution 64 (2012) 357–367 365

Author's personal copy
(EN; Wang and Xie, 2004). However, from our field experience we
believe that until additional extant populations are found, S.
guangxiense falls within the Critically Endangered category (CR,
based on criteria B2a) of the red list guidelines of the World Conser-
vation Union (IUCN, 2001).
Acknowledgments
The authors are grateful to Xi Lu, Zhi-Jian Yin, Hong-Jing Dong,
Qi-Liang Gan, and Yong-Zuo Shi for their assistance in sample collec-
tion. We also thank Dr. Amy Litt and two anonymous reviewers for
their constructive suggestions that greatly improved the paper. The
research was supported by the National Natural Science Foundation
of China (No. 30770152) granted to LGL and partially financed by the
Angiosperm Tree of Life Project (NSF EF-0431266) to DES.
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.ympev.2012.04.
010.
References
APG III, 2009. An update of the angiosperm phylogeny group classification for the
orders and families of flowering plants: APG III. Bot. J. Linn. Soc. 161, 105–121.
Bayer, R.J., Cross, E.W., 2002. A reassessment of tribal affinities of the enigmatic
genera Printzia and Isoetopsis (Asteraceae), based on three chloroplast
sequences. Aust. J. Bot. 50, 677–686.
Bull, J.J., Huelsenbeck, J.P., Cunningham, C.W., Swofford, D.L., Waddell, P.J., 1993.
Partitioning and combining data in phylogenetic analyses. Syst. Biol. 42, 384–
397.
Chandler, G.T., Bayer, R.J., 2000. Phylogenetic placement of the enigmatic western
Australian genus Emblingia based on rbcL sequences. Plant Spec. Biol. 15, 67–72.
Chase, M.W., Soltis, D.E., Olmstead, R.G., Morgan, D., Les, D.H., Mishler, B.D., Duvall,
M.R., Price, R.A., Hills, H.G., Qiu, Y.L., Kron, K.A., Rettig, J.H., Palmer, J.D., Manhart,
J.R., Sytsma, K.J., Michaels, H.J., Kress, W.J., Karol, K.G., Clark, W.D., Hedren, M.,
Gaut, B.S., Jansen, R.K., Kim, K.J., Wimpee, C.F., Smith, J.F., Furnier, G.R., Strauss,
S.H., Xiang, Q.Y., Plunkett, G.M., Soltis, P.S., Swensen, S., Williams, S.E., Gadek,
P.A., Quinn, C.J., Eguiarte, L.E., Golenberg, E., Learn Jr., 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. Ann. Mo. Bot.
Gard. 80, 528–580.
Conti, E., Soltis, D.E., Harding, T.M., Schneider, J., 1999. Phylogenetic relationships of
the Silver Saxifrages (Saxifraga, Sect. Ligulatae Haworth): implications for the
evolution of substrate specificity, life histories, and biogeography. Mol.
Phylogenet. Evol. 13, 536–555.
Cronquist, A., 1981. An Integrated System of Classification of Flowering Plants.
Columbia University Press, New York, USA.
Doyle, J.J., Doyle, J.L., 1987. A rapid DNA isolation procedure for small amounts of
fresh leaf tissue. Phytochem. Bull. 19, 11–15.
Engler, A., 1930. Saxifragaceae. In: Engler, A., Prantl, K. (Eds.), Die Natürlichen
Pflanzenfamilien, second ed., vol. 18a. Engelmann, Leipzig, Germany, pp. 74–226.
Felsenstein, J., 1985. Confidence limits on phylogenies: an approach using the
bootstrap. Evolution 39, 783–791.
Fishbein, M., Hibsch-Jetter, C., Soltis, D.E., Hufford, L., 2001. Phylogeny of
Saxifragales (Angiosperms: Eudicots): analysis of a rapid, ancient radiation.
Syst. Biol. 50, 817–847.
Gornall, R.J., Bohm, B.A., 1985. A monograph of Boykinia,Peltoboykinia,Bolandra, and
Suksdorfia (Saxifragaceae). Bot. J. Linn. Soc. 90, 1–71.
Guidon, S., Gascuel, O., 2003. A simple, fast, and accurate algorithm to estimate
large phylogenies by maximum likelihood. Syst. Biol. 52, 696–704.
Hamilton, M.B., 1999. Four primer pairs for the amplification of chloroplast
intergenic regions with intraspecific variation. Mol. Ecol. 8, 521–523.
Herbert, J., Chase, M.W., Moller, M., Abbott, R.J., 2006. Nuclear and plastid DNA
sequences confirm the placement of the enigmatic Canacomyrica monticola in
Myricaceae. Taxon 55, 349–357.
Hibsch-Jetter, C., Soltis, D.E., Macfarlane, T.D., 1997. Phylogenetic analysis of
Eremosyne pectinata (Saxifragaceae) based on rbcL sequence data. Plant Syst.
Evol. 204, 225–232.
IUCN, 2001. IUCN Red List Categories and Criteria. Version 3.1. IUCN, Gland. <http://
www.redlist.org>.
Johnson, L.A., Soltis, D.E., 1994. MatK DNA sequences and Phylogenetic
Reconstruction in Saxifragaceae s. str. Syst. Bot. 19, 143–156.
Johnson, L.A., Soltis, D.E., 1995. Phylogenetic inference in Saxifragaceae sensu stricto
and Gilia (Polemoniaceae) using matK sequences. Ann. Mo. Bot. Gard. 82, 149–
175.
Johnson, L.A., Soltis, D.E., 1998. Assessing congruence: empirical examples from
molecular data. In: Soltis, D.E., Soltis, P.S., Doyle, J.L. (Eds.), Molecular
Systematics of Plants II. Kluwer, Boston, USA, pp. 297–348.
Kuzoff, R.K., Sweere, J.A., Soltis, D.E., Soltis, P.S., Zimmer, E.A., 1998. The
phylogenetic potential of entire 26S rDNA sequences in plants. Mol. Biol. Evol.
15, 251–263.
Kuzoff, R.K., Soltis, D.E., Hufford, L., Soltis, P.S., 1999. Phylogenetic relationships
within Lithophragma (Saxifragaceae): hybridization, allopolyploidy, and ovary
diversification. Syst. Bot. 24, 598–615.
Li, Y.C., Yang, Z.L., Tolgor, B., 2009. Phylogenetic and biogeographic relationships of
Chroogomphus species as inferred from molecular and morphological data.
Fungal Divers. 38, 85–104.
Liu, K., Raghavan, S., Nelesen, S., Linder, C.R., Warnow, T., 2009. Rapid and accurate
large-scale coestimation of sequence alignments and phylogenetic trees.
Science 324, 1561–1564.
Mabberley, D.J., 1997. A Portable Dictionary of Vascular Plants. Cambridge
University Press, Cambridge.
Mason-Gamer, R.J., Kellogg, E.A., 1996. Testing for phylogenetic conflict among
molecular data sets in the tribe Triticeae (Gramineae). Syst. Biol. 45, 524–545.
Morgan, D.R., Soltis, D.E., 1993. Phylogenetic relationships among members of
Saxifragaceae sensu lato based on rbcL sequences data. Ann. Mo. Bot. Gard. 80,
631–660.
Nakazawa, M., Wakabayashi, M., Ono, M., Murata, J., 1997. Molecular
phylogenetic analysis of Chrysosplenium (Saxifragaceae) in Japan. J. Plant.
Res. 110, 265–274.
Oh, S.H., Potter, D., 2005. Molecular phylogenetic systematic and biogeography of
tribe Neillieae (Rosaceae) using DNA sequences of cpDNA, rRNA, and LEAFY.
Am. J. Bot. 92, 179–192.
Okuyama, Y., Kato, M., 2009. Unveiling cryptic species diversity of flowering plants:
successful biological species identification of Asian Mitella using nuclear
ribosomal DNA sequences. BMC Evol. Biol. 9, 105.
Okuyama, Y., Fujii, N., Wakabayashi, M., Kawakita, A., Ito, M., Watanabe, M.,
Murakami, N., Kato, M., 2005. Nonuniform concerted evolution and chloroplast
capture: heterogeneity of observed introgression patterns in three molecular
Fig. 4. Three types of carpels of Saniculiphyllum guangxiense. (a) Three carpels; (b) four carpels; (c) five carpels.
366 C.-L. Xiang et al. / Molecular Phylogenetics and Evolution 64 (2012) 357–367

Author's personal copy
data partition phylogenies of Asian Mitella (Saxifragaceae). Mol. Biol. Evol. 22,
285–296.
Okuyama, Y., Pellmyr, O., Kato, M., 2008. Parallel floral adaptations to pollination by
fungus gnats within the genus Mitella (Saxifragaceae). Mol. Phylogenet. Evol. 46,
560–575.
Olsson, S., Enroth, J., Buchbender, V., Hedenäs, L., Hutunen, S., Quandt, D., 2011.
Neckera and Thamnoryum (Neckeraceae, Bryopsida): paraphyletic assemblages.
Taxon 60, 36–50.
Pan, J.T., Gu, C.Z., Huang, S.M., Wei, Z.F., Jin, S.Y., Lu, L.D., Akiyama, S., Alexander, C.,
Bartholomew, B., Cullen, J., Gornall, R.J., Hultgård, U.M., Ohba, H., Soltis, D.E.,
2001. Saxifragaceae. In: Wu, C.Y., Rave, P.H. (Eds.), Flora of China, vol. 8. Science
Press/Missouri Botanical Garden Press, Beijing/St. Louis, USA, pp. 269–452.
Posada, D., 2008. JModelTest: phylogenetic model averaging. Mol. Biol. Evol. 25,
1253–1256.
Quicke, D.L., Jones, O.R., Epstein, D.R., 2007. Correcting the problem of false
incongruence due to noise imbalance in the incongruence length difference test.
Syst. Biol. 56, 496–503.
Rambaut, A., Drummond, A.J., 2009. Tracer Version 1.5.0. <http://
beast.bio.ed.ac.uk>.
Ronquist, F., Huelsenbeck, J.P., 2003. MrBayes 3: Bayesian phylogenetic inference
under mixed models. Bioinformatics 19, 1572–1574.
Ronquist, F., Huelsenbeck, J.P., van der Mark, P., 2005. MrBayes 3.1 Manual. <http://
mrbayes.csit.fsu.edu/manual.php>.
Senters, A.E., Soltis, D.E., 2003. Phylogenetic relationships in Ribes (Grossulariaceae)
inferred from ITS sequence data. Taxon 52, 51–66.
Shimodaira, H., Hasegawa, M., 1999. Multiple comparisons of log-likelihoods with
applications to phylogenetic inference. Mol. Biol. Evol. 16, 1114–1116.
Shimodaira, H., Hasegawa, M., 2001. CONSEL: for assessing the confidence of
phylogenetic tree selection. Bioinformatics 17, 1246–1247.
Smith, S.A., Dunn, C.W., 2008. Phyutility: a phyloinformatics tool for trees,
alignments and molecular data. Bioinformatics 24, 715–716.
Soltis, D.E., 2007. Saxifragaceae. In: Kubitzki, K. (Ed.), The Families and Genera of
Vascular Plants, vol. IX. Springer, Hamburg, Germany, pp. 418–435.
Soltis, D.E., Kuzoff, D.R., 1995. Discordance between nuclear and chloroplast
phylogenies in the Heuchera group (Saxifragaceae). Evolution 49, 727–742.
Soltis, D.E., Soltis, P.S., 1997. Phylogenetic relationships in Saxifragaceae sensu lato:
a comparison of topologies based on 18S rDNA and rbcL sequences. Am. J. Bot.
84, 504–522.
Soltis, D.E., Soltis, P.S., Collier, T.G., Edgerton, M.L., 1991a. Chloroplast-DNA variation
in Tellima grandiflora (Saxifragaceae). Am. J. Bot. 78, 1379–1390.
Soltis, D.E., Soltis, P.S., Collier, T.G., Edgerton, M.L., 1991b. Chloroplast DNA variation
within and among genera of the Heuchera group: evidence for chloroplast
capture and paraphyly. Am. J. Bot. 78, 1091–1112.
Soltis, D.E., Morgan, D.R., Grable, A., Soltis, P.S., Kuzoff, R., 1993. Molecular
systematics of Saxifragaceae sensu stricto. Am. J. Bot. 80, 1056–1081.
Soltis, D.E., Johnson, L.A., Looney, C., 1996a. Discordance between ITS and
chloroplast topologies in the Boykinia group (Saxifragaceae). Syst. Bot. 21,
169–185.
Soltis, D.E., Kuzoff, D.R., Gornall, R., Ferguson, K., 1996b. MatK and rbcL gene
sequence data indicate that Saxifraga (Saxifragaceae) is polyphyletic. Am. J. Bot.
83, 371–382.
Soltis, D.E., Soltis, P.S., Chase, M.W., Mort, M.E., Albach, D.C., Zanis, M., Savolainen,
V., Hann, W.H., Hoot, S.B., Fay, M.F., Axtell, M., Swensen, S.M., Nixon, K.C., Farris,
J.S., 2000. Angiosperm phylogeny inferred from a combined data set of 18S
rDNA, rbcL, and atpB sequences. Bot. J. Linn. Soc. 133, 381–461.
Soltis, D.E., Kuzoff, R.K., Mort, M.E., Zanis, M., Fishbein, M., Hufford, L., Koontz, J.,
Arroyo, M.K., 2001a. Elucidating deep-level phylogenetic relationships in
Saxifragaceae using sequences for six chloroplastic and nuclear DNA regions.
Ann. Mo. Bot. Gard. 88, 669–693.
Soltis, D.E., Tago-Nakazawa, M., Xiang, Q.Y., Kawano, S., Murata, J., Wakabayashi, M.,
2001b. Phylogenetic relationships and evolution in Chrysosplenium
(Saxifragaceae) based on matK sequence data. Am. J. Bot. 88, 883–893.
Sotiaux, A., Enroth, J., Olsson, S., Quandt, D., Vanderpoorten, A., 2009. When
morphology and molecules tell us different stories: a case-in-point with
Leptodon corsicus, a new and unique endemic moss species from Corsica. J.
Bryol. 31, 186–196.
Stamatakis, A., 2006. RAxML-VI-HPC: maximum likelihood-based phylogenetic
analyses with thousands of taxa and mixed models. Bioinformatics 22, 2688–
2690.
Swofford, D.L., 2003. PAUP

: Phylogenetic Analysis Using Parsimony (and other
methods), Version 4.0b10. Sinauer, Sunderland, Massachusetts, USA.
Swofford, D.L., Olsen, G.J., Waddell, P.J., Hillis, D.M., 1996. Phylogenetic inference.
In: Hillis, D.M., Moritz, C., Mable, B.K. (Eds.), Molecular Systematics, second ed.
Sinauer, Sunderland, Massachusetts, USA, pp. 407–514.
Taberlet, P., Gielly, L., Patou, G., Bouvet, J., 1991. Universal primers for amplification
of three non-coding regions of Chloroplast DNA. Plant Mol. Biol. 17, 1105–1109.
Vargas, P., 2000. A phylogenetic study of Saxifraga sect. Saxifraga (Saxifragaceae)
based on nr DNA ITS sequences. Plant Syst. Evol. 223, 59–70.
Wang, S., Xie, Y., 2004. China Species Red List, vol. l. Higher Education Press, Beijing,
China.
Webb, D.A., Gornall, R.J., 1989. A Manual of Saxifrages and Their Cultivation. Timber
Press, Portland.
Wen, J., Zimmer, E.A., 1996. Phylogeny and biogeography of Panax L. (the ginseng
genus, Araliaceae): inferences from ITS of nuclear ribosomal DNA. Mol.
Phylogenet. Evol. 6, 167–177.
White, T.J., Bruns, T., Lee, S., Taylor, J., 1990. Amplification and direct sequencing of
fungal ribosomal RNA genes for phylogenetic. In: Innis, M.A., Gelfand, D.H.,
Sninsky, J.J., White, T.J. (Eds.), PCR Protocols: A guide to Methods and
Applications. Academic Press, San Diego, CA, USA, pp. 315–322.
Wilgenbusch, J.C., Warren, D.L., Swofford, D.L., 2004. AWTY: A System for Graphical
Exploration of MCMC Convergence in Bayesian Phylogenetic Inference. <http://
ceb.csit.fsu.edu/awty>.
Wu, C.Y., Ku, T.C., 1992. A new tribe with a new monotypic genus of Saxifragaceae
(s.l.) from China. Acta Phytotax. Sin. 30, 193–196.
Wu, C.Y., Lu, A.M., Tang, Y.C., Chen, Z.D., Li, D.Z., 2003. The Families and Genera of
Angiosperms in China. Science press, Beijing.
Wu, C.Y., Sun, H., Zhou, Z.K., Peng, H., Li, D.Z., 2007. Origin and differentiation of
endemism in the flora of China. Frontiers Biol. China 2, 125–143.
Yang, Z., Rannala, B., 1997. Bayesian phylogenetic inference using DNA sequences: a
Markov chain Monte Carlo method. Mol. Biol. Evol. 14, 717–724.
Yu, J., Holder, M.T., 2010. SATé 1.0, University of Kansas, Lawrence, Kansan, USA.
C.-L. Xiang et al. / Molecular Phylogenetics and Evolution 64 (2012) 357–367 367