ArticlePDF Available

Phylogenetic study of the tribe Potentilleae (Rosaceae), with further insight into the disintegration of Sibbaldia

Authors:

Abstract and Figures

Potentilleae, one of ten tribes of the Rosaceae, are mainly distributed in alpine regions of the northern hemisphere. The taxonomy of Potentilleae has been challenging due to extensive hybridization, polyploidization and/or apomixis characterizing several genera of Potentilleae, such as Alchemilla, Argentina and Potentilla. To help clarify relationships within Potentilleae, a phylogenetic analysis of the tribe with an emphasis on the polyphyletic genus Sibbaldia was conducted using nuclear ribosomal ITS and ETS and the plastid trnL-F and trnS-G spacer regions. In agreement with previous phylogenetic analyses, three major clades were identified in the present study–the subtribe Fragariinae, the genera Argentina and Potentilla. The 15 species of Sibbaldia were recovered in five distinct clades: three in subtribe Fragariinae, one in Argentina and the last in Potentilla. The recently established genus Chamaecallis, comprising a single species formerly treated in Sibbaldia that has intermediate floral character states with respect to Fragariinae and Potentilla, was recovered as sister to Drymocallis. Morphological character state reconstruction indicated that a reduction in the number of stamens (≤10) is a derived character state that has arisen multiple times in Potentilleae. Molecular dating analyses agreed with previously published estimates and suggested that crown group Potentilleae arose in the Middle to Late Eocene, with most generic-level divergences occurring in the Oligocene and Miocene.
Content may be subject to copyright.
Research Article
doi: 10.1111/jse.12243
Phylogenetic study of the tribe Potentilleae (Rosaceae),
with further insight into the disintegration of Sibbaldia
Tao Feng
1,2
, Michael J. Moore
3
, Min-Hui Yan
1,2
, Yan-Xia Sun
1
, Hua-Jie Zhang
1,2
, Ai-Ping Meng
1
, Xiao-Dong Li
1
,
Shu-Guang Jian
4
, Jian-Qiang Li
1
*, and Heng-Chang Wang
1
*
1
Key Laboratory of Plant Germplasm Enhancement and Specialty Agriculture, Wuhan Botanical Garden, Chinese Academy of Sciences,
Wuhan 430074, China
2
University of Chinese Academy of Sciences, Beijing 100049, China
3
Department of Biology, Oberlin College, Oberlin, Ohio 44074, USA
4
South China Botanical Garden, Chinese Academy of Sciences, Guangzhou 510650, China
*Authors for correspondence. Jian-Qiang Li, E-mail: lijq@wbgcas.cn. Tel. 86-27-87510330. Fax: 86-27-87510251. Heng-Chang Wang,
E-mail: hcwang@wbgcas.cn. Tel.:þ86-18971492385. Fax: 86-27-87510251.
Received 30 December 2015; Accepted 27 January 2017; Article first published online xx Month 2017
Abstract Potentilleae, one of 10 tribes of the Rosaceae, are mainly distributed in alpine regions of the Northern
Hemisphere. The taxonomy of Potentilleae has been challenging due to extensive hybridization, polyploidization,
and/or apomixis characterizing several genera of Potentilleae, such as Alchemilla,Argentina, and Potentilla. To help
clarify relationships within Potentilleae, a phylogenetic analysis of the tribe with an emphasis on the polyphyletic
genus Sibbaldia was carried out using nuclear ribosomal internal and external transcribed spacer regions and the
plastid trnL-F and trnS-G spacer regions. In agreement with previous phylogenetic analyses, three major clades were
identied in the present study: the subtribe Fragariinae, the genera Argentina, and Potentilla. The 15 species of
Sibbaldia were recovered in ve distinct clades: three in subtribe Fragariinae, one in Argentina, and the last in
Potentilla. The recently established genus Chamaecallis, comprising a single species formerly treated in Sibbaldia that
has intermediate oral character states with respect to Fragariinae and Potentilla, was recovered as sister to
Drymocallis. Morphological character state reconstruction indicated that a reduction in the number of stamens
(10) is a derived character state that has arisen multiple times in Potentilleae. Molecular dating analyses agreed
with previously published estimates and suggested that crown group Potentilleae arose in the Middle to Late
Eocene, with most generic-level divergences occurring in the Oligocene and Miocene.
Key words: Argentina,Chamaecallis, character evolution, molecular dating, polyphyly, Potentilla, Potentilleae, Sibbaldia.
1 Introduction
Molecular phylogenetic studies (Potter et al., 2002, 2007a;
Eriksson et al., 2003) have improved our understanding of the
backbone relationships of Rosaceae, which in many cases
differ distinctly from traditional taxonomic groupings based
on morphology (e.g., Hutchinson, 1964). Current phyloge-
netic-based classications recognize 10 tribes of Rosaceae
under three subfamilies (Potter et al., 2007a). Potentilleae
Sweet is one of the 10 tribes of Rosaceae and its taxonomy has
changed dramatically with respect to the intratribal classica-
tion over its history (Table 1; also see Eriksson et al., 1998),
with most changes concerning the taxonomic status of
several small genera, such as Argentina Hill, Comarum L.,
Dasiphora Raf. (Pentaphylloides Duhamel), Drymocallis Fourr.
ex Rydb., Duchesnea Smith, Horkelia Chamisso, Ivesia Torr.,
and Sibbaldiopsis Rydb.
The monophyly of Potentilleae in the modern sense was
rst established by Eriksson et al. (2003) using DNA
sequences from nuclear internal transcribed spacer (ITS)
and plastid trnL-F regions. The stem-based denition of
Potentilleae corresponded almost exactly to the tribe
Potentilleae sensu Hutchinson (1964; Table 1). More recent
systematic studies using molecular data (Potter et al., 2007a;
Dobe
s & Paule, 2010; T
opel et al., 2011) have supported the
monophyly of Potentilleae as dened by Eriksson et al.
(2003) and have recovered three main clades in Potentilleae,
corresponding to Fragariinae Torrey & Gray, Potentilla L., and
Argentina, although relationships among these three clades
have been conicting (T
opel et al., 2011; Eriksson et al., 2015;
Feng et al., 2015). The four subtribes proposed by Soj
ak
(1989, 2008)Potentillinae (Potentilla,Horkelia,Ivesia,Stel-
lariopsis Rydb., Tylosperma Botsch., Piletophyllum (Soj
ak)
Soj
ak), Fragariinae (Comarum,Farinopsis Chrtek & Soj
ak,
Dasiphora,Drymocallis,Sibbaldia L., Sibbaldiopsis,Fragaria L.,
Sibbaldianthe Juz., Potaninia Maxim., Schistophyllidium (Juz.
ex Fed.) Ikonn.), Chamaerhodotinae (Chamaerhodos Bunge.),
and Alchemillinae (Alchemilla L., Aphanes L., Lachemilla
J
SE Journal of Systematics
and Evolution
XXX 2017 | Volume 9999 | Issue 9999 | 115 © 2017 Institute of Botany, Chinese Academy of Sciences
Rydb.)are not all supported as monophyletic based on
these molecular data; for example, the recognition of
Chamaerhodotinae and Alchemillinae would make Fragar-
iinae paraphyletic (Dobe
s & Paule, 2010; Eriksson et al., 2015).
Fragariinae were rst described by Torrey & Gray (1840) and
generally treated as including all genera of Potentilleae except
for Potentilla (e.g., Bentham & Hooker, 1865; Focke, 1894).
Several taxa within Fragariinae that are now treated as
separate genera were sometimes included within Potentilla,
such as Comarum,Dasiphora, and Drymocallis (Table 1). In
addition, Alchemilla s.l. (including Aphanes and Lachemilla) and
Potaninia have in most cases been placed outside of
Potentilleae entirely (Table 1). Current molecular phylogenetic
studies (Eriksson et al., 2003, 2015; Potter et al., 2007a;
Lundberg et al., 2009) have given a clear circumscription to
Fragariinae, which contain 12 genera: Alchemilla s.l.
(ca. 1000 spp.), Chamaecallis Smedmark (1 sp.), Chamaerhodos
(58 spp.), Comarum (1 sp.), Dasiphora (ca. 4 spp.), Drymocallis
(ca. 30 spp.), Farinopsis (1 sp.), Fragaria (ca. 20 spp.), Potaninia
(1 sp.), Sibbaldia (26 spp.), Sibbaldianthe (3 spp.), and
Sibbaldiopsis (3 spp.).
Sibbaldia is a taxonomically complicated genus of
Potentilleae. It is relatively small with 1019 species
(Dixit & Panigrahi, 1981; Rajput et al., 1997; Dikshit & Panigrahi,
1998; Li et al., 2003), and occurs mostly in alpine regions of
East Asia, except for the circumboreally distributed
S. procumbens L. Sibbaldia is generally separated from
Potentilla rather articially by its reduced number of stamens
in traditional taxonomic treatments (Table 1; also see review
by Eriksson et al., 1997), with species of Sibbaldia having 5(10)
stamens and species of Potentilla possessing ca. 20 stamens.
Surprisingly, approximately 10 Malesian species and the North
American Potentilla pentandra Engelm. with ve stamens have
never been referred to Sibbaldia (Soj
ak, 2010). Moreover, the
species that have been historically referred to Sibbaldia are
morphologically heterogeneous, with notable variation in leaf
form, anther type, style insertion, presence or absence of a
oral disc, and ower color (Soj
ak, 2008). The only character
Table 1 Survey of the generic level taxonomic history of Potentilleae
Rydberg
(1898, 1908)
Wolf (1908) Schulze-Menz
(1964)
Hutchinson
(1964)
Takhtajan
(1997)
Kalkman (2004) Soj
ak (2008)
Potentilleae Potentilleae Potentilleae Potentilleae Potentilleae Potentilleae Potentilleae
Potentillinae Potentillinae
Potentilla Potentilla Potentilla Potentilla Potentilla Potentilla Potentilla
Argentina [Argentina Argentina ——[Argentina
Duchesnea Duchesnea] Duchesnea Duchesnea Duchesnea [Duchesnea Duchesnea]
Horkelia Horkelia Horkelia Horkelia Horkelia Horkelia Horkelia
——Horkeliella Horkeliella Horkeliella (Horkeliella)
(Ivesia) Ivesia Ivesia Ivesia Ivesia Ivesia
Comarella Comarella Comarella Comarella (Comarella
——Purpusia Purpusia Purpusia)
Stellariopsis Stellariopsis Stellariopsis Stellariopsis Stellariopsis
——Tylosperma
——Piletophyllum
Alchemillinae Fragariinae
Comarum [Comarum Comarum Comarum Comarum Comarum Comarum
——Farinopsis
Dasiphora Dasiphora Pentaphylloides ——Dasiphora
Drymocallis Drymocallis] Drymocallis Drymocallis Drymocallis
Sibbaldia Sibbaldia Sibbaldia Sibbaldia Sibbaldia Sibbaldia
Sibbaldiopsis [Sibbaldiopsis] Sibbaldiopsis Sibbaldiopsis] Sibbaldiopsis
Fragaria Fragaria Fragaria Fragaria Fragaria Fragaria Fragaria
——Sibbaldianthe
——PotaniniaPotaniniaPotaniniaPotaninia
——Schistophyllidium
Chamaerhodotinae
Chamaerhodos Chamaerhodos Chamaerhodos Chamaerhodos Chamaerhodos Chamaerhodos
Alchemilleae Alchemilla
Group
Alchemillinae
——Alchemilla AlchemillaAlchemilla Alchemilla Alchemilla
——Aphanes Aphanes (Aphanes Aphanes
——Lachemilla ——Lachemilla Lachemilla
——Zygalchemilla)
Parentheses around a group of generic names within a column indicate that, in that treatment, those genera were included in the
genus listed directly above them. Square brackets around the generic names within a column indicate that, in that treatment,
those genera were included in Potentilla. , Genus was not mentioned in the treatment. Genus was placed out of Potentilleae
in that treatment.
2 Feng et al.
J. Syst. Evol. 9999 (9999): 115, 2017 www.jse.ac.cn
shared in common by these taxa is the reduced number of
stamens. Recent morphological and molecular studies (Soj
ak,
2008; Dobe
s & Paule, 2010; Eriksson et al., 2015) have indicated
that Sibbaldia as traditionally circumscribed is highly polyphy-
letic. Dobe
s & Paule (2010) rst noted the polyphyly of
Sibbaldia with their nding that the cushion species Sibbaldia
tetrandra Bunge was nested in Potentilla, and therefore
proposed to use the name P. tetrandra (Bunge) J. D. Hooker.
Likewise, Lundberg et al. (2009) found that several Potentilla
species (Potentilla miyabei Makino, P. cuneifolia Bertol., and
P. tridentata Sol.) were closely related to Sibbaldia, and
subsequently Paule & Soj
ak (2009) transferred them to
Sibbaldia. Eriksson et al. (2015) undertook the rst compre-
hensive phylogenetic study of Sibbaldia and found that
Sibbaldia was resolved into ve separate clades of Potentil-
leae, namely Sibbaldia s.s., Sibbaldianthe, the Dasiphora clade,
Argentina, and the Himalayan clade of Potentilla. Despite
these advances, the phylogenetic positions of several East
Asian taxa of Sibbaldia have yet to be evaluated with
molecular data.
This study aims to provide further insights into the
phylogenetic relationships of Potentilleae using sequences
from two plastid (trnL-F and trnS-G) and two nuclear (ITS and
ETS) regions, with a focus on taxa traditionally included in
Sibbaldia. We include as many taxa of traditional Sibbaldia as
possible, including three East Asian species that were not
included in previous molecular phylogenetic studies. We also
use character state reconstructions to examine the evolution
and taxonomic value of key morphological characters used in
previous taxonomic treatments, and we use molecular dating
techniques to examine the diversication history of
Potentilleae.
2 Material and Methods
2.1 Taxon sampling
In total, 82 accessions representing 74 species and 15 genera
from Potentilleae were sampled. Taxon names, voucher
informationandGenBankaccessionnumbersaregivenin
Table 2. For some species with broad geographic distribu-
tions or signicant morphological variability, more than one
accession was sequenced. A total of 27 accessions
representing 15 species of Sibbaldia in the traditional sense
were sampled, including three East Asian species not
included in previous phylogenetic studies: Argentina
glabriuscula (Y
u&Li)Soj
ak, Potentilla tenuis (Hand.-Mazz.)
Soj
ak, and Sibbaldianthe sericea Grubov. All ve clades of
Potentilla recognized by T
opel et al. (2011) were represented
by at least two species. Likewise, our Potentilla sampling
also included species within all of the main clades of the
genus (clade C, clade D, clade E, clade F, clade G þI)
recovered by Dobe
s & Paule (2010). For the large genus
Alchemilla s.l., four species from early diverging lineages
were sampled. Exemplars of Duchesnea,Horkelia,andIvesia,
all of which have been merged into Potentilla based on
previous analyses (Dobe
s & Paule, 2010; T
opel et al., 2011),
were included, as was one exemplar of Schistophyllidium,
which was lumped into Sibbaldianthe by Eriksson et al.
(2015). Material was not available for Comarella,Purpusia
Brandegee, or Stellariopsis.Rosa majalis Herrm. and
Sanguisorba ofcinalis L. were selected as outgroups,
following Lundberg et al. (2009). In addition, 11 representa-
tives of the main clades of Rosaceae (one from Rosoideae
Juss. ex Arn., eight from Amygdaloideae Arn., and two from
Dryadoideae (Lam. & DC.) Sweet) were sampled for the
purposeofdivergencetimeestimation.
2.2 DNA isolation, amplification, and sequencing
Total genomic DNA was extracted from silica-dried leaf tissue
using the procedure of Doyle (1987) with the following
modications: 1015 mg dry leaf tissue was ground in 2.0-mL
tubes using a mini-beadbeater (Biospec, Beijing, China), and
the DNA pellet was washed twice with 75% ethanol.
The ETS region of 18S-26S nuclear ribosomal DNA was
amplied using the primers ETS1 and IGS8 (Oh & Potter, 2005).
The ITS1 (Urbatsch et al., 2000) and ITS4 (White et al., 1990)
primers were used to amplify the ITS, including the 5.8S rRNA
subunit. The plastid trnL-F spacer and trnL intron were
amplied using primers c and f of Taberlet et al. (1991). The
trnS-G spacer and trnG intron were amplied with the trnS
GCU
and 30trnG
UUC
primers of Shaw et al. (2005). Polymerase chain
reactions (PCRs) included 1.5 U Taq DNA polymerase
(Tiangen Biotech, Beijing, China), 1PCR reaction buffer,
1.5 mmol/L MgCl
2
, 0.2 mmol/L each dNTP, 0.2 mmol/L each of
forward and reverse primers, 1.0 mL total DNA, and ddH
2
O for
a total volume of 25.0 mL. The PCR products were checked for
length and concentrations on 2% agarose gels and sent to
Sangon (Shanghai, China) for commercial sequencing. All
regions were sequenced in both directions using both the
terminal PCR primers as well as internal primers (ITS2 (White
et al., 1990), trn-D and trn-E (Taberlet et al., 1991), and 5
́
trnG2G
and 5
́
trnG2S (Shaw et al., 2005)), excepting the ETS region,
where no internal primers were used. New sequences
generated for this study have been deposited in GenBank
under accession numbers KP875251 to KP875379 (Table 2).
2.3 Phylogenetic analyses
Sequences were assembled and manually edited using
BioEditor (Yang et al., 2003) and the SeqMan module
implemented in the dnastar package (DNAStar, Madison,
WI, USA). Sequences were initially aligned with muscle version
3.8.31 (Edgar, 2004) using the default settings followed by
manual adjustments in Se-Al version 2.0a11 (available at
http://tree.bio.ed.ac.uk/software/seal/). Ambiguously aligned
regions were excluded from ITS (positions 7482, 95126,
140146, 436449, 457466, 595625), ETS (8289, 159173,
211235, 301311, and 381387), trnL-F (371537, 545569,
890898, and 10601069), and trnS-G (360366, 489498,
816827, 882891, 9241350, 15071545, and 15681573).
Gapped sites with 50% missing data were excluded from
analyses using Phyutility version 2.6 (Smith & Dunn, 2008). In
total, 115, 85, 553, and 642 positions were excluded from the
ITS, ETS, trnL-F, and trnS-G datasets, respectively. The matrices
are available from TreeBASE (http://purl.org/phylo/treebase/
phylows/study/TB2:S17261).
Separate phylogenetic analyses were undertaken for each
region (ETS, ITS, trnL-F and trnS-G), as well as for combined
nuclear (ETS þITS) and plastid (trnL-F þtrnS-G) data using
both Bayesian inference and parsimony methods. Maximum
parsimony (MP) analyses were carried out with paupversion
4.0b10 (Swofford, 2003) using heuristic searches with 1000
Phylogeny of Potentilleae 3
www.jse.ac.cn J. Syst. Evol. 9999 (9999): 115, 2017
Table 2 List of taxa studied: Taxon, voucher specimen, and GenBank accession numbers for external transcribed spacer (ETS),
internal transcribed spacer (ITS), trnL-F, and trnS-G
Taxa Voucher GenBank accession no.
ETS ITS trnL/F trnS/G
Adenostoma fasciculatum Hook. &
Sarn.
Oh 970521 (DAV) DQ886358 AF348538
Alchemilla alpina L. T. Eriksson 805 (S) FJ422342/
FJ422343
AJ512217 FJ422305
Alchemilla alpina L. R. Eriksson s.n. (GH) U90816/
U90817
——
Alchemilla cryptantha Steud. ex A. Rich T. Eriksson 914 (S) FJ422344 FJ356153 FJ422283 FJ422306
Alchemilla mollis (Buser) Rothm. T. Eriksson s.n.(S) FJ422345 AJ511769 AJ512218 FJ422307
Alchemilla pentaphyllea L. B. Gehrke BG-E400 (ZH) FJ422346 FJ356154 FJ422284 FJ422308
Aremonia agrimonioides (L.) DC. Karlsson 94076 (LD) U90799 AJ512230/
AJ512231
FJ422310
Argentina anserina (L.) Rydb. Feng 116 (HIB) KP875251KF954772 KJ020644
Argentina anserina (L.) Rydb. Eriksson & Smedmark
44 (SBT)
FN421405 FN430824 FN561752 FN556670
Argentina glabriuscula (T.T. Y
u & C.L.
Li) Soj
ak 53
Feng 53 (HIB) KP875259KF954763 KJ020639 KP875356
Argentina glabriuscula (T.T. Y
u & C.L.
Li) Soj
ak 57
Feng 57 (HIB) KP875260KF954764 KJ020640 KP875358
Argentina leuconota (D. Don) Soj
ak Feng 108 (HIB) KP875257KF954771 KJ020641 KP875355
Argentina lignosa (Willd. in D.F.K.
Schltdl.) Soj
ak
M. T
opel MA132 (GB) FJ422369 FJ356171 FJ422299 FJ422332
Argentina lineate (Trevir.) Soj
ak Feng 1 (HIB) KP875254KP875291KP875354
Argentina micropetala (D. Don) Soj
ak Feng 8 (HIB) KP875252KF954771 KJ020641 KP875359
Argentina microphylla (D. Don) Soj
ak MA 144 (GB) FN421388 FN430809 FN556412 FN556679
Argentina peduncularis (D. Don) Soj
ak MA 173 (GB) FN421389 FN430820 FN561742 FN594721
Argentina phanerophlebia (Y
u & Li)
Feng & Wang
Feng 6 (HIB) KP875253KF954770 KJ020642 KP875360
Argentina songzhuensis T. Feng & H.
Wang
Feng 58 (HIB) KP875258KF954766 KJ020638 KP875357
Argentina stenophylla (Franch.) Soj
ak KGB 299 (GB) FN421381 FN555607 FN561738 FN556662
Argentina tapetodes (Soj
ak) Soj
ak Feng 93 (HIB) KP875255KF954769 KP875330KP875362
Argentina turfosa (Hand.-Mazz.) Soj
ak Feng 55 (HIB) KP875256KF954768 KP875331KP875361
Aronia arbutifolia (L.) Elliott Atha 1905-81 (NY) JQ392371 ——
Aronia arbutifolia (L.) Elliott Dickinson 2003-2 (TRT) ——JQ392186
Cercocarpus betuloides Nutt. Gao s.n. (DAV) DQ886355 AF348537
Chamaecallis perpusilloides (W.W. Sm.)
Smedmark
Feng 52 (HIB) KP875280KP875287KP875336KP875351
Chamaecallis perpusilloides (W.W. Sm.)
Smedmark
Feng 68 (HIB) KP875281KP875288KP875335KP875352
Chamaerhodos erecta (L.) Bunge Lackschewitz 11453 (GH) FJ422348 AJ512219 FJ422311
Chamaerhodos erecta (L.) Bunge Norlindh & Ahti
10161A (S)
U90794 ——
Chamaerhodos mongholica Bunge E. Rosenius 1028 (S) FJ422349 FJ356155 FJ422285 FJ422312
Chamaerhodos nuttallii Pickering ex
Rydb.
J.W. Moore 233133 (S) FJ422350 FJ356156 FJ422286
Chamaerhodos sabulosa Bunge Joel Eriksson 618 (S) FJ422351 FJ356157 ——
Comarum palustris L. M. Lundberg 17 (S) FJ356158 FJ422313
Comarum palustris L. T. Eriksson 659 (GH, S) ——AJ512237
Cotoneaster acutifolius Turczaninow SYS 5926 JQ405597 JQ405532
Dasiphora davurica (Nestl.) Kom. &
Aliss
M. Lundberg 24 (S) FJ42254 FJ356159 FJ422287 FJ422315
Dasiphora fruticosa (L.) Rydb. Feng 103 (HIB) KP875279KP875290KP875337KP875349
Dasiphora glabra (G. Lodd.) Soj
ak Feng 120 (HIB) KP875277KP875289KP875338KP875348
Continued
4 Feng et al.
J. Syst. Evol. 9999 (9999): 115, 2017 www.jse.ac.cn
Table 2 Continued
Taxa Voucher GenBank accession no.
ETS ITS trnL/F trnS/G
Dasiphora parvifolia
(Fisch. ex Lehm.) Juz.
Feng 119 (HIB) KP875278KF954762 KJ020646 KP875350
Dasiphora phyllocalyx Juz. T. Eriksson 757 (S) FJ356160 FJ422288 FJ422317
Dryas octopetala L. Aronsson s.n. (S) U90804 ——
Dryas octopetala L. Potter 011020-01 (DAV) ——DQ 851231
Drymocallis agrimonioides Rydb. M. Lundberg 15 (S) FJ42235 FJ422289 FJ422318
Drymocallis agrimonioides Rydb. Laferri
ere 2357 (A) U90787 ——
Drymocallis corsica (Soleirol ex Lehm.)
Kurtto
M. Lundberg 13 (S) FJ422357 FJ356161 FJ422290 FJ422319
Drymocallis glutinosa Rydb. M. Lundberg 5 (S) FJ422358 FJ356162 FJ42229 FJ422320
Drymocallis rupestris (L.) Soj
ak M. Lundberg 6 (S) FJ422359 FJ356163 FJ422292 FJ422321
Farinopsis salesoviana (Steph.) Chrtek
& Soj
ak
M. Lundberg 3 (S) FJ422353 ——FJ422314
Farinopsis salesoviana (Steph.) Chrtek
& Soj
ak
Eriksson & Vretblad
TE751 (S)
AJ511779 AJ512228
Filipendula vulgaris Moench Eriksson 821 (SBT) AJ416467 AJ416463
Fragaria chiloensis (L.) Mill. M. Lundberg 14 (S) FJ422360 FJ356164 FJ422293 FJ422322
Fragaria orientalis Losinsk. Feng 107 (HIB) KP875282KP875292KP875334KP875353
Fragaria viridis Weston M. Lundber 16 (S) FJ422364 FJ356166 FJ422295 FJ422326
Gillenia trifoliate (L.) Moench Boufford 21032 (YU) JQ392352 ——
Gillenia trifoliate (L.) Moench Packard 80-97 (YU) ——JQ392180
Horkelia bolanderi A. Gray Eriksson s.n. (SBT) FN421401 FN430789 FN556395 FN556664
Ivesia kingii S. Watson J. L. Reveal et al.
#4782 (GB)
FN421377 FN430787 FN561735 FN556666
Lyonothamnus floribundus Gray Oh 4988 (DAV) AY555315 AF348548
Potaninia mongolica Maxim Norlindh & Ahti
10348 (S)
FJ422366 AM286742 AM286743 FJ422328
Potentilla acaulis L. Feng 129 (HIB) ——KP875317KP875371
Potentilla alba L. MA 122 (GB) FN421355 FN430774 FN556379 FN556667
Potentilla alchemilloides Lapeyr. A. & A.-L. Anderberg
26 (S)
FJ422367 FJ356168 FJ422297 FJ422329
Potentilla argentea L. MA 143 (GB) FN421387 FN430808 FN561750 FN594719
Potentilla biflora Willd. ex Schltdl. 102 Feng 102 (HIB) KP875270KP875301KP875329KP875373
Potentilla biflora Willd. ex Schltdl. G46 Viereck 5042 (S) FN430826 FN561749 FN556673
Potentilla caulescens L. MA 133 (GB) FN421379 FN430819 FN556399 FN594714
Potentilla chinensis Ser. Feng 110 (HIB) KP875266KP875298KP875319KP875369
Potentilla clandestina Soj
ak Feng 25 (HIB) KP875274KP875308KP875327KP875378
Potentilla conferta Bunge Feng 127 (HIB) KP875264KP875296KP875320KP875368
Potentilla coriandrifolia D. Don Feng 133 (HIB) KP875269KP875302KP875363
Potentilla discolor Bunge Feng 118 (HIB) KP875262KP875299KP875321KP875370
Potentilla erecta (L.) Raeusch. MA 124 (GB) FN430780 FN556405 FN594699
Potentilla fragarioides L. Cult. in Hortus
Bergianus
FN555610 FN561747 FN557007
Potentilla griffithii Hook. f. Feng 44 (HIB) KP875261KP875293KP875316KP875372
Potentilla indica (Andrews) Wolf Feng 138 (HIB) KP875268KP875300KP875314KP875364
Potentilla kleiniana Wight & Arn. Feng 139 (HIB) KP875263KP875294KP875315KP875366
Potentilla multifida (Tausch) Wolf Feng 124 (HIB) KP875265KP875295KP875318KP875367
Potentilla purpurea (Royle) Hook. f. Feng 64 (HIB) KP875275KP875307KP875326KP875379
Potentilla reptans L. MA 131 (GB) FN421368 FN430815 FN561728 FN556657
Potentilla sischanensis Bunge ex Lehm. Feng 112 (HIB) KP875267KP875297KP875322KP875365
Potentilla stolonifera Lehm. ex Ledeb BE 1382: 1 (GB) FN421363 FN430814 FN556420 FN556654
Potentilla suavis Soj
ak Feng 37 (HIB) KP875276KP875305KP875323KP875374
Potentilla tenuis (Hand.-Mazz.) Soj
ak Feng 26 (HIB) KP875273KP875306KP875325KP875376
Potentilla tetrandra (Hook. f.) Bunge 89 Feng 89 (HIB) KP875271KP875303KP875328KP875375
Potentilla tetrandra (Hook. f.) Bunge 97 Feng 97 (HIB) KP875272KP875304KP875324KP875377
Continued
Phylogeny of Potentilleae 5
www.jse.ac.cn J. Syst. Evol. 9999 (9999): 115, 2017
replicates of random taxon addition, tree bisection
reconnection branch swapping, MulTrees on, saving a
maximum of 100 trees each replicate. All gaps were treated
as missing characters. Support was assessed using 1000
replicates of non-parametric bootstrap analysis (Felsenstein,
1985). Bayesian analysis was carried out using MrBayes
version 3.1.2 (Ronquist & Huelsenbeck, 2003). The Akaike
Information Criterion as implemented in jModelTest version
2.5.6 (Darriba et al., 2012) was used to select the suitable
model for each dataset (Table 3). Two independent analyses
were run for 2 000 000 generations with four chains (one cold
and three heated) each and the temp parameter set to 0.1.
Trees were sampled every 1000th generation. Convergence of
the chains was assessed using Tracer version 1.6. (available at
http://tree.bio.ed.ac.uk/software/tracer/). Based on this as-
sessment, the rst 25% of the sampled trees were discarded
and the remaining samples were summarized in a 50%
majority-rule consensus tree.
Table 2 Continued
Taxa Voucher GenBank accession no.
ETS ITS trnL/F trnS/G
Prunus laurocerasus L. EB 88 AF318724 AF318677
Rhodotypos scandens (Thunb.) Mak. UC Bot. Gard. 86.0616 AY177141 AF348566
Rosa majalis Herrm. T. Eriksson 641 (GH, S) FJ422371 U90801 AJ512229 FJ422333
Sanguisorba officinalis L. T. Eriksson 804 (S) FJ422372 AJ416465 FJ422334
Sanguisorba officinalis L. T. Eriksson s.n. (GH) U90797 ——
Sibbaldia parviflora Willd. M. Lundberg 4 (S) FJ422374 FJ356174 FJ422302 FJ422336
Sibbaldia procumbens L. Feng 131 (HIB) KP875284KP875310KP875339KP875345
Sibbaldia procumbens L. Feng S4 (HIB) KP875283KP875309KP875341KP875346
Sibbaldia semiglabra C. A. Mey. J. Klackenberg
82062-11 (S)
FJ422376 FJ356175 FJ422303 FJ422338
Sibbaldianthe adpressa (Bunge) Juz.
130
Feng 130 (HIB) KP875311KP875332KP875343
Sibbaldianthe adpressa (Bunge) Juz.
G11
V.A. Gusev 391 (S) FJ422377 FJ356176 FJ422304 FJ422339
Sibbaldianthe bifurca (L.) Kurtto & T.
Erikss.
Feng 113 (HIB) KF954761 KJ020645 KP875342
Sibbaldianthe sericea Grubov Feng 122 (HIB) KP875285KP875312KP875333KP875344
Sibbaldiopsis cuneifolia (Bertol.) Soj
ak
G5
M. Lundberg 39 (S) FJ422368 FJ356169 FJ422298 FJ422330
Sibbaldiopsis cuneifolia (Bertol.) Soj
ak
48
Feng 48 (HIB) KP875286KP875313KP875340KP875347
Sibbaldiopsis miyabei (Makino) Soj
ak Sten Bergman s.n. (S) FJ422370 FJ356172 FJ422300
Sibbaldiopsis tridentata (Sol.) Rydb. M. Lundberg 2 (S) FJ422379 ——FJ422341
Sibbaldiopsis tridentata (Sol.) Rydb. Eriksson & Smedmark
40 (S)
——AJ512236
Sibbaldiopsis tridentata (Sol.) Rydb. Hill 17146 (A) U90791 ——
Spiraea cantoniensis Nutt. Potter 970619-02 (DAV) DQ886362 AF348571
, Unavailable sequence. Newly generated sequences; otherwise, sequences were obtained from GenBank.
Table 3 Alignment and tree statistics for each of the phylogenetic analyses of the tribe Potentilleae (Rosaceae)
ITS ETS trnL-F trnS-G Combined
nuclear
Combined
plastid
No. of taxa 82 75 80 79 82 82
Aligned length 537 397 837 964 934 1801
No. variable
characters
205 (38%) 213 (54%) 276 (33%) 326 (34%) 418 (45%) 602 (33%)
No. parsimony
-informative characters
158 (29%) 174 (44%) 173 (21%) 175 (18%) 332 (36%) 348 (19%)
Tree length (steps) 636 564 444 493 1211 938
Consistency index 0.47 0.50 0.75 0.78 0.50 0.77
Retention index 0.83 0.85 0.93 0.93 0.84 0.93
Model GTR þIþG TVM þG TPM1uf þGþI GTR þG GTR þG TVM þG
ETS, external transcribed spacer; ITS, internal transcribed spacer.
6 Feng et al.
J. Syst. Evol. 9999 (9999): 115, 2017 www.jse.ac.cn
2.4 Character evolution
Three oral characters have commonly been used in
Potentilleae taxonomy: anther type, style insertion and
stamen number. The former two have been extensively
discussed in previous studies (Soj
ak, 2008; Dobe
s & Paule,
2010; T
opel et al., 2011; Eriksson et al., 2015). Here, we
examined the evolution of stamen numbers in the context of
the molecular phylogeny. The character states were scored
using fresh material and the literature (Li et al., 2003; Soj
ak,
2004, 2008, 2012): number of stamens ca. 20[0], 10[1].
Ancestral states were mapped onto the MP plastid tree
topology using the parsimony method as implemented in
Mesquite v.3.01 (Maddison & Maddison, 2015).
2.5 Divergence time estimation
Divergence times were estimated separately for the combined
nuclear data (ETS þITS) and the combined plastid data
(trnL-F þtrnS-G) using BEAST v.2.1.3 (Bouckaert et al., 2014).
For all analyses, the GTR þIþG(with four categories) model
of sequence evolution, a lognormal relaxed molecular clock
model, and the Yule prior were used to estimate divergence
times and the corresponding credibility intervals. Separate
age constraints were placed on Rosaceae (i.e., the root of
the tree) and the clade of Rosa þPotentilleae. Based on the
previous dating analyses of Rosales carried out by Chin et al.
(2014), the prior for the crown age of Rosaceae was
constrained using a normal distribution with a mean of
88 Ma and a standard deviation of 3, giving a 95% condence
interval of 8393 Ma. The oldest known fossil record of
the genus Rosa Rosa germerensis from the Eocene
(55.848.6 Ma) in North America (cf. Palaeobiology database,
http://paleodb.org) was used to calibrate the crown clade of
Rosa þPotentilleae. Leaf fragments of similar age were
reported by Hollick (1936) and have also been considered
to belong to Rosa. The age of the Rosa þPotentilleae crown
clade was constrained using a lognormal distribution with
offset values of 48 Ma, a mean of 1.0, and a standard deviation
of 1.0, giving a 95% condence interval of 48.662.2 Ma. The
Beast Markov chain Monte Carlo was run for 30 million
generations for both datasets, saving every 3000th tree.
Convergence of the chains was assessed using Tracer version
1.6. Samples from posterior distributions were summarized on
a maximum clade credibility tree with 50% burn-in using
TreeAnnotator version 2.1.2 (Bouckaert et al., 2014). Trees
were visualized using FigTree version 1.4.2 (available at http://
tree.bio.ed.ac.uk/software/gtree/).
3 Results
3.1 Data matrix and tree statistics
Characteristics of each data matrix and the corresponding
tree statistics are presented in Table 3.
3.2 Phylogenetic relationships within Potentilleae
Three major clades (Fragariinae, Argentina, and Potentilla)
were recovered within Potentilleae in both the nuclear and
plastid trees, although the positions of these clades varied
(Figs. 1, 2). All plastid trees resolved Argentina as sister to
Potentilla with very high support values (bootstrap
support ¼99%/posterior probability ¼1) in the combined
tree (Fig. 2), 98/1 in the trnL-F tree (not shown), and 97/1 in
the trnS-G tree (not shown). However, Argentina was sister to
Fragariinae in the ITS tree (82/1, not shown) and in the
combined nuclear data tree (77/0.89; Fig. 1). In the ETS tree
(not shown), Argentina, Fragariinae, and Potentilla formed a
polytomy.
Fragariinae sensu Eriksson et al. (2003) was resolved into
three main clades in both the nuclear and plastid trees: (i)
Fragaria; (ii) a clade (referred to as clade B) comprising
Dasiphora and related genera (the Dasiphora group recog-
nized in Eriksson et al., 2015); and (iii) a clade (referred to as
clade A) comprising the rest of Fragariinae: Alchemilla,
Comarum,Sibbaldia,Sibbaldianthe, and Sibbaldiopsis
(Figs. 1, 2). In the plastid phylogenetic tree (Fig. 2), clade A
was sister (100/1) to a clade of Fragaria þthe Dasiphora group
(0.91/1), whereas the relationships among these three clades
were not resolved in the nuclear phylogenetic tree (Fig. 1).
Relationships within the Dasiphora group were very similar
between the nuclear and plastid trees; all genera in the group
were monophyletic in both trees and only the position of
Chamaerhodos varied, but without strong support (Figs. 1, 2).
Within clade A, Sibbaldia and Sibbaldianthe were resolved as
sister groups in both the combined plastid and nuclear trees
but with relatively low support in the nuclear tree (/0.61;
Fig. 1) compared to that in the plastid tree (88/0.92; Fig. 2).
Farinopsis,Comarum, and Alchemilla formed a strongly
supported clade in the plastid tree (98/1, with no resolution
among these three genera; Fig. 2), but the relationships of
these genera collapsed into a polytomy with Sibbaldia,
Sibbaldianthe, and Sibbaldiopsis in the nuclear tree (Fig. 1).
Within Potentilla,allve major clades (Argentea, Ivesia,
Reptans, Alba, and Fragarioides) found by T
opel et al. (2011)
were also recovered in our trees and these roughly corre-
sponded to the clades C, D, E, F, and G þI recovered by Dobe
s&
Paule (2010). The relationships of these clades are almost
completely congruent between our study and previous studies
(Dobe
s & Paule, 2010; T
opel et al., 2011), except for the addition
of ve former Sibbaldia species, which formed a moderately to
strongly supported clade (the Himalayan clade) in both
analyses (/0.95 and 81/1.0 in nuclear and plastid trees,
respectively). The Himalayan clade was resolved as sister to all
the other Potentilla species with strong support (99/1.0) in the
plastid tree (Fig. 2), whereas it was nested in Alba in the nuclear
tree (Fig. 1). One representative of this clade, Sibbaldia
tetrandra, was also included in Dobe
s & Paule (2010), and
clustered with clade G þIþH(¼clade Alba) rather than being
sister to all other Potentilla species in the plastid phylogeny.
3.3 Polyphyly of Sibbaldia
The 15 taxa traditionally referred to Sibbaldia fell into ve
separate clades: (i) Sibbaldia s.s.; (ii) the Sibbaldianthe clade;
(iii) Chamaecallis (in the Dasiphora group); (iv) Argentina
(the Anserina cladeof Eriksson et al., 2015); and (v) the
Himalayan clade. The three species of Sibbaldia that were not
included in previous studies were recovered in three clades:
Sibbaldia sericea (Grub.) Soj
ak was recovered in Sibbaldianthe
close to Sibbaldianthe adpressa (Bunge) Juz.; Sibbaldia
glabriuscula was resolved in Argentina, and Sibbaldia tenuis
was recovered in the Himalayan clade.
Sibbaldia s.s., Sibbaldianthe, and Sibbaldiopsis displayed
complex relationships in the nuclear and plastid trees (clade C;
Phylogeny of Potentilleae 7
www.jse.ac.cn J. Syst. Evol. 9999 (9999): 115, 2017
Fig. 1. Bayesian 50% majority-rule consensus tree of Potentilleae based on combined analysis of nuclear ribosomal internal and
external transcribed spacer regions. Numbers above branches are maximum parsimony bootstrap values/Bayesian posterior
probabilities. The species formerly treated as Sibbaldia are shown in bold. Numbers following names refer to accession numbers.
Stamen number is reconstructed on the tree. A, Sibbaldiopsis cuneifolia (Bertol.) Soj
ak, anther with one theca (Th) that is not
interrupted at the apex of the anther, and lateral styles (St). B, Sibbaldianthe adpressa (Bunge) Juz., anther with one theca that is
not interrupted at the apex of the anther, and lateral styles. C, Argentina anserina (L.) Rydb., anther with two thecae that are
separated by connective at both apex and base of the anther, and lateral styles. D, Potentilla kleiniana Wight & Arn., anther with
two thecae that are separated by connective at both apex and base of the anther, and subterminal styles. E, Rosa chinensis Jacq.,
anther with two thecae and terminal styles. Left panels of C, D, and E, red arrows show the interruption between two thecae.
Right panels of A, B, C, D, and E, red arrows show the position of styles on achenes. Ov, ovary; Pe, petal.
8 Feng et al.
J. Syst. Evol. 9999 (9999): 115, 2017 www.jse.ac.cn
Fig. 2. Bayesian 50% majority-rule consensus tree of Potentilleae based on combined analysis of plastid trnL-F and trnS-G spacer
regions. Numbers above branches are maximum parsimony bootstrap values/Bayesian posterior probabilities. The species
formerly treated as Sibbaldia are shown in bold. Numbers following names refer to accession numbers.
Phylogeny of Potentilleae 9
www.jse.ac.cn J. Syst. Evol. 9999 (9999): 115, 2017
Fig. 3. Parsimony-based ancestral character state reconstructions for Potentilleae and outgroups plotted on the maximum
parsimony plastid tree topology for stamen number.
10 Feng et al.
J. Syst. Evol. 9999 (9999): 115, 2017 www.jse.ac.cn
Figs. 1, 2). Sibbaldia sensu Eriksson et al. (2015) was not
monophyletic in either tree; species of Sibbaldiopsis were
nested within it in both the nuclear and plastid trees. In the
nuclear tree (Fig. 1), two subclades of Sibbaldia (S. cuneata
Schouw ex Kunze þS. parviora Willd. and S. procumbens þS.
semiglabra C.A. Mey.) were recovered and formed a strongly
supported (99/1.0) trichotomous clade with Sibbaldiopsis
tridentata (Sol.) Rydb. Sibbaldiopsis tridentata also formed a
moderately supported (72/1.0) clade with Sibbaldia s.s. in the
plastid tree (Fig. 2). However, Sibbaldiopsis miyabei (Makino)
Soj
ak and S. cuneifolia (Bertol.) Soj
ak formed a larger, strongly
supported (99/1.0) clade with Sibbaldia (Fig. 2). In the nuclear
tree, the latter two species of Sibbaldiopsis formed a strongly
supported (98/1.0) clade with Sibbaldianthe (Fig. 1).
Sibbaldianthe was found monophyletic in the plastid tree
(Fig. 2). Finally, the two accessions of Sibbaldia procumbens
from Europe and East Asia, respectively, did not form a clade
in either tree (Figs. 1, 2).
3.4 Morphological character state reconstructions
Character state reconstructions revealed that the stamen
number has been reduced independently numerous times
from the ancestral state of numerous stamens (ca. 20) to 10
(Figs. 1, 3).
3.5 Divergence times
The topologies of Potentilleae in the maximum clade credibility
trees (Figs. S1, S2) derived from the beast analyses were largely
consistent with those from MrBayes (Figs. 1, 2). Lineage
divergence times (mean age and 95% highest posterior density)
estimated for the nodes of interest are summarized in Table 4.
The estimated ages of the main clades based on nuclear data
and plastid data weresimilar (Table 4; Figs. S1, S2). The earliest
divergence in crown group Potentilleae was estimated to have
occurred in the Middle Eocene, with the deepest divergences
within Potentilleae, including the crown groups of Fragariinae,
Argentina, and Potentilla, occurring from the latest Eocene to
Middle Miocene.
4 Discussion
4.1 Phylogeny and evolution of Potentilleae
Our results generally agree with those from previous
molecular analyses, most notably in nding three major
clades of Potentilleae: Fragariinae, Argentina, and Potentilla
(Dobe
s & Paule, 2010; T
opel et al., 2011; Eriksson et al., 2015).
Fragariinae (excepting Chamaecallis) is characterized by
anthers with one theca and basal to lateral styles, whereas
Argentina has anthers with two thecae and lateral styles and
Potentilla has anthers with two thecae and subterminal styles
(Soj
ak, 2004; 2008). The incongruencies between nuclear data
and plastid data regarding the phylogenetic position of
Argentina were already discussed by Eriksson et al. (2015) and
cannot be resolved based on our current data. Various ploidy
levels exist in Argentina (Ikeda & Ohba, 1999), including
diploids (e.g., A. glabriuscula and A. leuconota (D. Don) Soj
ak),
tetraploids (e.g., A. peduncularis (D. Don) Soj
ak) and
anisopolyploids (e.g., A. microphylla (D. Don) Soj
ak (2x, 4x),
A. anserina (L.) Rydb. (4x, 5x, 6x)). The morphological and
molecular conict inherent in Argentina may reect a complex
evolutionary history such as allopolyploidy in this genus (T
opel
et al., 2011). However, incomplete lineage sorting cannot be
ruled out based on the current data.
The divergence times of major lineages of Potentilleae
estimated from nuclear and plastid molecular dating analyses
are generally consistent with each other and with previous
analyses (Table 4; Figs. S1, S2). The Eocene origin of crown
group Potentilleae estimated in the beast analyses (Table 4;
Figs. S1, S2) agrees broadly with previous estimates of
diversication times in Potentilleae by Dobe
s & Paule (2010),
who included extensive sampling of Potentilla for three plastid
genes. The stem of the three major lineages was established in
the Eocene, however, their diversication happened much
later, mostly in the Miocene. For example, most genera in
Fragariinae originated in the Miocene (Figs. S1, S2). T
opel
(2010) indicated that the origin and diversication of ivesioid
Potentilleae (Ivesia clade in this study) coincided with the Late
Oligocene increased aridity in western North America. It is
possible that the development of cool and dry conditions in
northern temperate latitudes during the Miocene may have
played a main role in the diversication of Potentilleae. Dobe
s
& Paule (2010) suggested an Asian origin of Argentina in the
Oligocene, similar to our results. Argentina comprises 64
species disjunctively distributed in Asia (mainly in the
Sino-Himalaya region) and New Guinea. Only a few Asian
species and no New Guinean representatives have been
included in molecular phylogenies (Dobe
s & Paule, 2010; Feng
et al., 2015) to date, and hence it is not clear what the
relationship of the Asian taxa to the New Guinean taxa is. Our
Table 4 Lineage divergence times for clades of Potentilleae as estimated from beast based on combined nuclear DNA data and
combined plastid DNA data
Mean age (95% HPD) Ma
Phylogenetic split or age of clade base
Nuclear Plastid
Potentilleae 45.0 (37.451.0) 44.9 (36.152.4)
Fragariinae 34.9 (27.243.0) 32.9 (23.541.8)
Potentilla 25.7 (16.236.3) 22.6 (14.832.2)
Argentina 28.1 (18.639.4) 26.3 (17.635.7)
Sibbaldia þSibbaldianthe þSibbaldiopsis 19.4 (12.426.8) 23.0 (14.931.6)
Dasiphora group 23.4 (14.632.9) 17.8 (9.7026.7)
Fragaria þDasiphora group 29.7 (20.338.9) 26.0 (15.936.3)
HPD, highest posterior density.
Phylogeny of Potentilleae 11
www.jse.ac.cn J. Syst. Evol. 9999 (9999): 115, 2017
sampling of Argentina, except for the basal taxa A. lignosa
(Willd. in D. F. K. Schltdl.) Soj
ak and A. anserina, represents
taxa of limited geographic distribution in the high mountains
of the QinghaiTibet Plateau. Our analyses suggest that the
diversication of these Asian Argentina species, as well as
those in the Himalaya clade of Potentilla, occurred after the
Middle Miocene (Figs. S1, S2).
4.2 Taxonomy of Sibbaldia
Our results agree with the ndings of previous studies (Dobe
s
& Paule, 2010; Eriksson et al., 2015) that Sibbaldia in the
traditional sense is highly polyphyletic. Below we discuss the
taxonomic implications of our results.
Sibbaldia s.s. includes the type species S. procumbens.A
broadly dened S. procumbens, including S. cuneata,
S. parviora, and S. semiglabra, has been favored by recent
studies (Rajput et al., 1997; Soj
ak, 2008). Sibbaldia
procumbens s.l. occurs circumboreally, with discontinuous
populations in Europe, East Asia, and North America.
Correspondingly, the species consists of a series of relatively
pronounced forms, many of which have been described as
distinct taxa. Similar to Eriksson et al. (2015), we found some
phylogenetic structure among the accessions of Sibbaldia s.s.
For example, the two accessions of S. procumbens, one from
northern China and the other from Europe, were not sister in
either tree, which may indicate that S. procumbens should be
divided into multiple taxa. Our limited sampling precludes a
rm conclusion, however.
As currently circumscribed, Sibbaldianthe includes two
species formerly included in Sibbaldia:S. adpressa and
S. sericea. The latter was separated from the former by
M
es
ı
cek & Soj
ak (1969) on the basis of its pure white petals
that are longer than the sepals versus petals yellowish-green
and equal to or shorter than the sepals. However, the petal
characters used by M
es
ı
cek & Soj
ak (1969) are variable (also
see Rajput et al., 1997), and the main differences are in the
shape and indumentum of the leaets. Moreover,
Sibbaldianthe adpressa is a relatively widespread species
distributed from the Himalayas to Russia and is mainly found
on gravelly or rocky mountains, coastal slopes, and sandy
soils, whereas S. sericea is an arid-adapted endemic of the
Mongolian Plateau. In our results, the two accessions of
S. adpressa do not form a clade; instead, S. sericea is sister to
one of the accessions of S. adpressa in the plastid tree (Fig. 2)
and forms a polytomy with both accessions of S. adpressa in
the nuclear tree (Fig. 1). This may imply that S. sericea is a local
variety of S. adpressa, although the two accessions of
S. adpressa are relatively distant from one another phyloge-
netically in the plastid tree (Fig. 2). Alternatively, S. adpressa
may consist of multiple cryptic species and may be best
divided into multiple taxa. Again, further sampling will be
required to test these alternatives.
Three additional species formerly treated as Sibbaldia fall
within Argentina (A. phanerophlebia (Y
u & Li) Feng & Wang,
A. micropetala (D. Don) Soj
ak, and A. glabriuscula), in full
agreement with the morphological assessment of Soj
ak (2010)
based on the structure of the stipule auricles. Species of
Argentina possess ventral stipule auricles (Soj
ak, 2010), which
may be a good synapomorphy for the genus (Feng et al.,
2015). All three former Sibbaldia species share ventral stipule
auricles and are consistently resolved within Argentina with
strong support. The achenes of A. micropetala and relatives
A. phanerophlebia and A. emodi (H. Ikeda & H. Ohba) Soj
ak
(the last not sampled here) are characterized by a conspicu-
ous outgrowth on the ventral side, which is different from the
glabrous achenes of all the other species of Argentina. Soj
ak
(2008) initially described a new genus Piletophyllum for
A. micropetala based on this distinctive achene structure,
which he then abandoned in favor of placing this species in
Argentina based on the structure of the stipule auricles (Soj
ak,
2010). Our molecular analyses support Soj
ak (2010) and Feng
et al. (2015) in placing this and the other two species
(A. phanerophlebia and A. glabriuscula)inArgentina. The
former genus Tylosperma (Soj
ak, 2008) is supported as sister
to the other members of Argentina, which was also recovered
by the previous studies (T
opel et al., 2011; Eriksson et al., 2015).
Although the molecular data do not provide any arguments
about its taxonomic status, it is better treated within
Argentina because of the obvious morphological synapomor-
phies, such as ventral stipular auricles, lateral styles, anthers
with two thecae, and interruptedly pinnate leaves (Soj
ak,
2010).
Another clade of former Sibbaldia species comprises the
ve Himalayan taxa that fall within Potentilla (Figs. 1, 2). Of this
clade, Dobe
s & Paule (2010) had included Potentilla tetrandra
(formerly Sibbaldia tetrandra). The anther morphology of
these species supports their phylogenetic relationship (Soj
ak,
2004). Eriksson et al. (2015) resolved this clade as sister to the
Potentilla alba L. clade, albeit with very low support. Its
afliation with the Alba clade is moderately supported by our
nuclear phylogenetic tree (Fig. 1), but in the plastid tree (Fig. 2)
the Himalayan clade is sister to the remaining Potentilla
species, although support is generally low along the backbone
of Potentilla.Potentilla tenuis and P. suavis Soj
ak, former
Sibbaldia taxa that were not included by Eriksson et al. (2015),
are resolved within the Himalayan clade and hence our results
support their transfer to Potentilla. Other former Sibbaldia
species in the Himalayan clade include Potentilla sikkimensis
Prain (not sampled), P. purpurea (Royle) Hook. f.,
and P. clandestina Soj
ak. The last two species, together with
P. suavis, have sometimes been lumped into a single species,
P. purpurea (¼S. purpurea, Rajput et al., 1997), but this may
not seem warranted given the recovery of P. tenuis within a
clade containing these three taxa (Fig. 1). A sampling of a
wider range of populations will be necessary for delimiting
species in the P. purpurea complex.
Finally, Chamaecallis perpusilloides (W. W. Sm.) Smed-
mark is a morphologically isolated species that has been
formerly treated in Sibbaldia as well as in Potentilla because
of its anthers with two thecae (Soj
ak, 2008). Eriksson et al.
(2015) resolved it in Fragariinae near to Chamaerhodos and
Drymocallis andthusproposederectinganewgenus,
Chamaecallis. Our phylogenetic analyses support this
transfer and provide moderate to strong support for its
sister relationship to Drymocallis (Figs. 1, 2), unlike Eriksson
et al. (2015), who did not recover strong support for the
position of Chamaecallis within Fragariinae. Morphology
also supports the generic status of Chamaecallis:
C. perpusilloides is a cushion-like perennial herb with white
owers, trifoliolate leaves, and 10 or fewer stamens, which
is a combination of character states unique in Potentilleae.
Furthermore, the achenes of Chamaecallis have lateral
12 Feng et al.
J. Syst. Evol. 9999 (9999): 115, 2017 www.jse.ac.cn
styles like other Fragariinae, but the anthers have two
thecae that do not narrow at the top of the anther and
they sit close to each other, so that the split at the top
seems to be uninterrupted and only one theca seems to
exist (Soj
ak, 2008). Hence, Chamaecallis will be a key taxon
for understanding the evolution of oral development of
Potentilleae.
4.3 Taxonomic implications for other genera of Potentilleae
Sibbaldiopsis comprises three species, S. cuneifolia,S. miyabei,
and S. tridentata. These three species have long been placed
together (Wolf, 1908; Robertson, 1974; Soj
ak, 2004, 2008) due
to their homogeneous morphological characters, such as
ternate leaves, relatively long petals, numerous stamens
(ca. 20), and hairy achenes. Paule & Soj
ak (2009) transferred
all three species to Sibbaldia based on the plastid phylogeny
published by Lundberg et al. (2009) and Dobe
s & Paule (2010).
This transfer was also supported by the plastid tree presented
in this study (Fig. 2). However, the nuclear tree topology
disagreed with the plastid tree concerning the phylogenetic
position of Sibbaldiopsis, with S. cuneifolia and S. miyabei
clustered with Sibbaldianthe, and S. tridentata nested within
Sibbaldia (Fig. 1). Similar complex relationships among
Sibbaldia,Sibbaldiopsis, and Sibbaldianthe were also recov-
ered by previous studies (Lundberg et al., 2009; Eriksson et al.,
2015). The transfer of Sibbaldiopsis tridentata to Sibbaldia as
S. retusa (Paule & Soj
ak, 2009; Eriksson et al., 2015) is
supported by molecular phylogenies, but including the other
two species would make Sibbaldia polyphyletic in the context
of the nuclear phylogeny presented here (Fig. 1), and hence
may be premature. Instead, Eriksson et al. (2015) have
proposed to retain Sibbaldiopsis cuneifolia and S. miyabei
within Potentilla. These two species are morphologically
similar to each other, and intermediate between Sibbaldia and
Sibbaldianthe. Their ternate leaets resemble those
of Sibbaldia, and their large yellow owers with numerous
stamens are similar to those of Sibbaldianthe bifurca (L.)
Kurtto & T. Erikss. Our nuclear phylogeny groups both species
with Sibbaldia (Fig. 1), whereas the plastid phylogeny resolves
them as sister to Sibbaldianthe (Fig. 2). Moreover, both
species are polyploids: Sibbaldiopsis cuneifolia is tetraploid
(Ikeda, 1989) and S. miyabei is octoploid (Fedorov, 1974). Thus,
S. cuneifolia and S. miyabei may be allopolyploids originating
from hybridization between Sibbaldia and Sibbaldianthe.
Although S. cuneifolia and S. miyabei are similar to each
other, they do not form a clade in either tree and it is possible
that they were generated from the hybridization of different
parent taxa and/or independent episodes of allopolyploidiza-
tion. The complicated phylogenetic relationships, ploidy
levels, and morphological evolution of the clade containing
Sibbaldia s.s., Sibbaldianthe, and Sibbaldiopsis preclude a rm
taxonomic resolution of the generic status of these taxa given
our current data. Creating a single inclusive genus for clade C
(Figs. 1, 2) may prove to be the best taxonomic solution, as
also has been proposed by Eriksson et al. (2015), in which case
the name Sibbaldia has priority.
4.4 Morphological character evolution in Potentilleae
Currently, no single unique morphological feature that
denes the tribe Potentilleae is known. The tribe can be
distinguished from other tribes of Rosaceae by a set of
features, including herbaceous habit (versus shrubs in
Dasiphora,Potaninia,andsomeComarum), an enlarged
receptacle, pistils with lateral to basal styles (subterminal in
Potentilla), and the fruit being an achenetum or achene.
Potentilla is characterized by anthers with two thecae
(Fig. 1), that can be further divided into four types: large
narrow anthers, medium-sized anthers, large anthers, and
small anthers (Soj
ak, 2008). However, T
opel et al. (2011)
found no obvious correlation between the anther shape
categories and phylogenetic relationships in Potentilla.
However, at the genus level, anther shape has been
proposed to be taxonomically informative and has been
used to subdivide Potentilleae (Soj
ak, 2008). The subtribe
Fragariinae has a unique type of anther: anthers with one
theca, which is not found anywhere else in Rosaceae. A
morphological study (Soj
ak, 2008) implied that this type of
anther may have evolved from anthers with two thecae
through conuence of the two thecae at their apex (in some
cases also at the base). In the context of the molecular
phylogeny presented in this study and the previous studies
(Lundberg et al., 2009; Eriksson et al., 2015), the two thecae
state is the ancestral condition and one theca is derived, but
reversed to the ancestral state in Chamaecallis.
A reduced number of stamens (10) has long been used to
dene Sibbaldia (Dixit & Panigrahi, 1981; Rajput et al., 1997;
Dikshit & Panigrahi, 1998; Li et al., 2003).This was challenged by
a recent morphological study (Soj
ak, 2008) and a molecular
phylogenetic study (Eriksson et al., 2015). Our results support
these previous studies and indicate that the reduction in
stamen numbers is an apomorphy and has occurred indepen-
dently multiple times in Potentilleae (Figs. 1, 3). Potter et al.
(2007a) found a similar trend across all Rosaceae: in their
analyses, numerous stamens (>10) were recovered as the
ancestral state within Rosaceae, with a reduced number of
stamens (10) inferred to have appeared multiple times. In
Potentilleae, stamen number is stable in some genera
(Alchemilla 4(1), Chamaerhodos (5), Drymocallis (20), Dasi-
phora (20)), whereasit is variable in others (Argentina (20, 5,
and 510), Potentilla (20 and 520), and the newly circum-
scribed Sibbaldia (5 and 20)). Hence, stamen number alone
appears to be a character of limited use in delimiting genera in
Potentilleae, however, it may be valuable in combination with
other morphological characters for Potentilleae taxonomy.
Style insertion is another key character that traditionally has
been used in taxonomic treatments of Potentilleae (e.g.,
Rydberg, 1908; Hutchinson, 1964; Soj
ak, 2008). The styles of
Potentilleae range from subterminal to basal (Fig.1), with
Potentilla having subterminal styles, Argentina having lateral
ones, and Fragariinae having lateral to basal styles (Soj
ak,
2004; 2008). Within Potentilla, historically the subterminal
styles were further divided into four subcategories, namely
the long and threadlike Nematostylae, the cone-shaped
Conostylae, the club-shaped Gomphostylae and the short
threadlike Leptostylae (Wolf, 1908). Ancestral state recon-
struction by T
opel et al. (2011) found a correlation between
this subclassication of styles and phylogenetic relationships
within Potentilla. Furthermore, the style type is also a good
indicator of phylogenetic relationships in Potentilleae based
on the previous phylogenetic studies (Dobe
s & Paule, 2010;
Eriksson et al., 2015) and the phylogenetic trees presented in
this study (Figs. 1, 2). Due to the ambiguous phylogenetic
Phylogeny of Potentilleae 13
www.jse.ac.cn J. Syst. Evol. 9999 (9999): 115, 2017
position of Argentina, the ancestral state of style insertion in
Potentilleae is not clear (Eriksson et al., 2015). However, while
subterminal styles also frequently occur in Rosaceae outside
of Potentilleae, lateral to basal styles are rarely found outside
the tribe (Potter et al., 2007a). This fact, combined with
the overall position of Potentilleae within Rosaceae, indicates
that lateral styles can be safely assumed to be an apomorphy
for Potentilleae and may be a synapomorphy for Fragariinae
and Argentina. Thus, the evolution of style insertion within
Potentilleae can be inferred to have happened in the direction
of subterminal to lateral, then to basal.
Acknowledgements
This work was funded by NSFC (Grant Nos. 31070191 and
31370223) and NSTIPC (2013FY111200) and the SPRPCAS
(XDA13020500).
References
Bentham G, Hooker J. 1865. Genera plantarum. London: Reeve.
Bouckaert R, Heled J, K
uhnert D, Vaughan T, Wu C-H, Xie D, Suchard
MA, Rambaut A, Drummond AJ. 2014. BEAST 2: A software
platform for Bayesian evolutionary analysis. PLoS Computational
Biology 10: e1003537.
Chin S-W, Shaw J, Haberle R, Wen J, Potter D. 2014. Diversication of
almonds, peaches, plums and cherriesMolecular systematics
and biogeographic history of Prunus (Rosaceae). Molecular
Phylogenetics and Evolution 76: 3448.
Darriba D, Taboada GL, Doallo R, Posada D. 2012. jModelTest 2: More
models, new heuristics and parallel computing. Nature Methods 9:
772772.
Dikshit B, Panigrahi G. 1998. The family Rosaceae in India: Revisionary
studies on Potentilla L., Sibbaldia L.and Brachycaulos Dikshit et
Panigr. Dehra Dun: Bishen Singh Mahendra Pal Singh.
Dixit B, Panigrahi G. 1981. Revision of the genus Sibbaldia
L. (Rosaceae) in India. Proceedings of the Indian Academy of
Sciences 90: 253272.
Dobe
s C, Paule J. 2010. A comprehensive chloroplast DNA-based
phylogeny of the genus Potentilla (Rosaceae): Implications for its
geographic origin, phylogeography and generic circumscription.
Molecular Phylogenetics and Evolution 56: 156175.
Doyle JJ. 1987. A rapid DNA isolation procedure for small quantities of
fresh leaf tissue. Phytochemical Bulletin 19: 1115.
Edgar RC. 2004. MUSCLE: Multiple sequence alignment with high
accuracy and high throughput. Nucleic Acids Research 32: 17921797.
Eriksson T, Donoghue MJ, Hibbs MS. 1998. Phylogenetic analysis of
Potentilla using DNA sequences of nuclear ribosomal internal
transcribed spacers (ITS), and implications for the classication of
Rosoideae (Rosaceae). Plant Systematics and Evolution 211:
155179.
Eriksson T, Hibbs MS, Yoder AD, Delwiche CF, Donoghue MJ. 2003. The
phylogeny of Rosoideae (Rosaceae) based on sequences of the
internal transcribed spacers (ITS) of nuclear ribosomal DNA and
the trnL/F region of chloroplast DNA. International Journal of Plant
Sciences 164: 197211.
Eriksson T, Lundberg M, T
opel M,
Ostensson P, Smedmark JE. 2015.
Sibbaldia: A molecular phylogenetic study of a remarkably
polyphyletic genus in Rosaceae. Plant Systematics and Evolution
301: 171184.
Fedorov AA. 1969. Chromosome numbers of owering plants. Academy
of Sciences of the USSR. Leningrad: Komarov Botanical Institute
(Reprint 1974).
Felsenstein J. 1985. Condence limits on phylogenies: An approach
using the bootstrap. Evolution 39: 783791.
Feng T, Moore MJ, Sun Y, Meng A, Chu H, Li J, Wang H. 2015. A new
species of Argentina (Rosaceae, Potentilleae) from Southeast
Tibet, with reference to the taxonomic status of the genus. Plant
Systematics and Evolution 301: 911921.
Focke W. 1894. Rosaceae. In: Edgar A, Prantl K eds. Die nat
urlichen
Panzenfamilien. Leipzig: Engelmann. 161. (in German)
Hollick CA. 1936. The tertiary oras of Alaska. Washington DC: US
Goverment Printing Ofce.
Hutchinson J. 1964. The genera of owering plants. Oxford: Clarendon
Press.
Ikeda H. 1989. Chromosome numbers of the Himalayan Potentilla
(Rosaceae). Journal of Japanese Botany 64: 361367.
Ikeda H, Ohba H. 1999. A systematic revision of Potentilla L.
section Leptostylae (Rosaceae) in the Himalaya and adjacent
regions. BulletinUniversity Museum, University of Tokyo 39:
31117.
Kalkman C. 2004. Rosaceae. In: Kubitzki K ed. The families and genera
of vascular plants. Berlin: Springer. 343386.
Keck DD. 1938. Revision of Horkelia and Ivesia. Lloydia 1: 75142.
Kerr MS. 2004. A phylogenetic and biogeographic analysis of
Sanguisorbeae (Rosaceae), with emphasis on the Pleistocene
radiation of the high Andean genus Polylepis. Ph.D. Dissertation.
College Park: University of Maryland.
Li C, Ikeda H, Ohba H, Zhengyi W, Raven PH, Hong DY. 2003. Potentilla
L. In Wu ZY, Raven PH, Hong DY eds. Flora of China. Beijing:
Science Press; St. Louis: Missouri Botanical Garden Press. 9:
291328.
Linnaeus C. 1753. Species Plantarum. Stockholm: Salvius.
Linnaeus C. 1754. Genera plantarum. Stockholm: Salvius.
Love A. 1982. IOPB chromosome number reports LXXV. Taxon 31:
342368.
Lundberg M, T
opel M, Eriksen B, Nylander JA, Eriksson T. 2009.
Allopolyploidy in Fragariinae (Rosaceae): Comparing four DNA
sequence regions, with comments on classication. Molecular
Phylogenetics and Evolution 51: 269280.
Maddison WP, Maddison DR. 2015. Mesquite: a modular system for
evolutionary analysis. Version 3.04. Available from http://
mesquiteproject.org [accessed 20 Dec 2015].
M
es
ı
cek J, Soj
ak J. 1969. Chromosome counts of some Mongolian
plants. Folia Geobotanica et Phytotaxonomica 4: 5586.
Oh S-H., Potter D. 2005. Molecular phylogenetic systematics and
biogeography of tribe Neillieae (Rosaceae) using DNA sequences
of cpDNA, rDNA, and LEAFY. American Journal of Botany 92:
179192.
Paule J, Soj
ak J. 2009. Taxonomic comments on the genus
Sibbaldiopsis Rydb. (Rosaceae). Journal of the National Museum
(Prague), Natural History Series 178: 1516
Potter D, Eriksson T, Evans RC, Oh S, Smedmark J, Morgan DR, Kerr M,
Robertson KR, Arsenault M, Dickinson TA. 2007a. Phylogeny and
classication of Rosaceae. Plant Systematics and Evolution 266:
543.
Potter D, Gao F, Bortiri PE, Oh S-H, Baggett S. 2002. Phylogenetic
relationships in Rosaceae inferred from chloroplast matK and
trnL-trnF nucleotide sequence data. Plant Systematics and
Evolution 231: 7789.
14 Feng et al.
J. Syst. Evol. 9999 (9999): 115, 2017 www.jse.ac.cn
Potter D, Still SM, Grebenc T, Ballian D, Bo
zi
c G, FranjiæJ, Kraigher H.
2007b. Phylogenetic relationships in tribe Spiraeeae (Rosaceae)
inferred from nucleotide sequence data. Plant Systematics and
Evolution 266: 105118.
Rajput MT, Tahir SS, Hussain SZ, Spongberg S. 1997. The genus
Sibbaldia (Rosaceae). Pakistan Journal of Botany 29: 138.
Robertson KR. 1974. The genera of Rosaceae in the southeastern
United States. Journal of the Arnold Arboretum 55: 303662.
Ronquist F, Huelsenbeck JP. 2003. MrBayes 3: Bayesian phylogenetic
inference under mixed models. Bioinformatics 19: 15721574.
Rydberg PA. 1898. A monograph of the North American Potentilleae.
Memoirs from the Department of Botany of Columbia College 2:
1223.
Rydberg PA. 1908. Rosaceae. North American Flora 22: 293388.
Schulze-Menz G. 1964. Rosales. In: Melchior H ed. Englers Syllabus der
Panzenfamilien. Berlin: Borntraeger. 193242. (in German)
Shaw J, Lickey EB, Beck JT, Farmer SB, Liu W, Miller J, Siripun KC,
Winder CT, Schilling EE, Small RL. 2005. The tortoise and the hare
II: Relative utility of 21 noncoding chloroplast DNA sequences for
phylogenetic analysis. American Journal of Botany 92: 142166.
Smith SA, Dunn CW. 2008. Phyutility: A phyloinformatics tool for trees,
alignments and molecular data. Bioinformatics 24: 715716.
Soj
ak J. 1994. Notes on Potentilla (Rosaceae): XXII. X. The section
Dumosae. XI. The P. microphylla and P. stenophylla groups (sect.
Pentaphylloides). XII. Key to the taxa of P. sect. Pentaphylloides
(Anserina). Botanische Jahrb
ucher f
ur Systematik 116: 1181.
Soj
ak J. 2004. Potentilla L. (Rosaceae) and related genera in the
former USSR (identication key, checklist and gures) Notes on
Potentilla XVI. Botanische Jahrb
ucher f
ur Systematik 125: 253340.
Soj
ak J. 2008. Notes on Potentilla XXI. A new division of the tribe
Potentilleae (Rosaceae) and notes on generic delimitations.
Botanische Jahrb
ucher f
ur Systematik 127: 349358.
Soj
ak J. 2010. Argentina Hill, a genus distinct from Potentilla
(Rosaceae). ThaisziaJournal of Botany 20: 9197.
Soj
ak J. 2012. Potentilla L. (Rosaceae) and related genera in Asia
(excluding the former USSR), Africa and New GuineaNotes on
Potentilla XXVIII. Plant Diversity and Evolution 130: 7157.
Sweet R. 1825. British ower garden. London: James Ridgway.
Swofford DL. 2003. PAUP: Phylogenetic analysis using parsimony
(and other methods), version 4.0b10. Sunderland: Sinauer.
T
opel M. 2010. Phylogenetic and phyloclimatic inference of the
evolution of Potentilleae (Rosaceae). Ph.D Thesis. Gothenburg:
University of Gothenburg.
T
opel M, Lundberg M, Eriksson T, Eriksen B. 2011. Molecular data and
ploidal levels indicate several putative allopolyploidization events
in the genus Potentilla (Rosaceae). PLoS Currents 3: RRN1237.
Taberlet P, Gielly L, Pautou G, Bouvet J. 1991. Universal primers for
amplication of three non-coding regions of chloroplast DNA.
Plant Molecular Biology 17: 11051109.
Takhtajan AL. 1997. Diversity and classication of owering plants. New
York: Columbia University Press.
Torrey J, Gray A. 1840. Aora of North America: Containing abridged
descriptions of all the known indigenous and naturalized plants
growing north of Mexico, arranged according to the natural system.
New York: Wiley, Putnam.
Urbatsch LE, Baldwin BG, Donoghue MJ. 2000. Phylogeny of the
coneowers and relatives (Heliantheae: Asteraceae) based on
nuclear rDNA internal transcribed spacer (ITS) sequences and
chlorplast DNA restriction site data. Systematic Botany 25:
539565.
White TJ, Bruns T, Lee S, Taylor J. 1990. Amplication and direct
sequencing of fungal ribosomal RNA genes for phylogenetics. In:
Innis MA, Gelfand DH, Sninsky JJ, White TJ eds. PCR protocols:
A guide to methods and applications. London: Academic Press.
315322.
Wolf T. 1908. Monographie der Gattung Potentilla.Bibliotheca
Botanica 71: 1715. (in German)
Yang P, Craig PA, Goodsell D, Bourne PE. 2003. Bio Editor - simplifying
macromolecular structure annotation. Bioinformatics 19:
897898.
Supplementary Material
The following supplementary material is available online for
this article at http://onlinelibrary.wiley.com/doi/10.1111/
jse.12243/suppinfo
Fig. S1. Maximum clade credibility (MCC) trees derived from
BEAST, based on nuclear DNA sequences (external transcribed
spacer (ETS) þinternal transcribed spacer (ITS)). Branches
marked with an asterisk () indicate Bayesian posterior
probabilities 0.95. Node bars indicate 95% highest posterior
densities (HPD). Amygda., Amygdaloideae; C1, age constraint
of crown Rosaceae; C2, age constraint of the node joining
Rosa and its sister Potentilleae; Dry., Dryadoideae.
Fig. S2. Maximum clade credibility (MCC) trees derived from
BEAST, based on plastid DNA sequences (trnL-F þtrnS-G).
Branches marked with an asterisk () indicate Bayesian
posterior probabilities 0.95. Node bars indicate 95%
highest posterior densities (HPD). Amygda., Amygdaloideae;
C1, age constraint of crown Rosaceae; C2, age constraint of
the node joining Rosa and its sister Potentilleae; Dry.,
Dryadoideae.
Phylogeny of Potentilleae 15
www.jse.ac.cn J. Syst. Evol. 9999 (9999): 115, 2017
... Soják (1985Soják ( , 2008 divided the Potentilleae tribe into subtribes based on anther structure. Two of these, Potentillinae and Fragariinae, are resolved as well-supported sister clades in several phylogenetic analyses using DNA sequence data (Eriksson & al., , 2015Kurtto & Eriksson, 2003;Feng & al., 2015Feng & al., , 2017Xiang & al., 2017;. These studies also show that the two remaining subtribes of Soják (2008) are ingroups in Fragariinae (also see Lundberg & al., 2009). ...
... Töpel & al. (2011) found, and informally named, six clades with good support: The Anserina, Alba, Fragarioides, Reptans, Ivesioid, and Argentea clades. Apart from the Fragarioides clade, which was only resolved by their nuclear ribosomal data, these clades have also been recovered in other studies (Dobeš & Paule, 2010;Feng & al., 2017;Persson & al., 2020a), and not contradicted in some that lacked adequate sampling for properly resolving the major Potentillinae lineages (Eriksson & al., 2015;Xiang & al., 2017;Fig. 1. Phylogeny of the Potentilleae clade, assembled from previously published results (Eriksson & al., , 2015Lundberg & al., 2009;Dobeš & Paule, 2010;Töpel & al., 2011;Feng & al., 2017;Xiang & al., 2017;Persson & al., 2020a), with clades at the tips collapsed into grey triangles, and their approximate number of species. ...
... Apart from the Fragarioides clade, which was only resolved by their nuclear ribosomal data, these clades have also been recovered in other studies (Dobeš & Paule, 2010;Feng & al., 2017;Persson & al., 2020a), and not contradicted in some that lacked adequate sampling for properly resolving the major Potentillinae lineages (Eriksson & al., 2015;Xiang & al., 2017;Fig. 1. Phylogeny of the Potentilleae clade, assembled from previously published results (Eriksson & al., , 2015Lundberg & al., 2009;Dobeš & Paule, 2010;Töpel & al., 2011;Feng & al., 2017;Xiang & al., 2017;Persson & al., 2020a), with clades at the tips collapsed into grey triangles, and their approximate number of species. Genus names and clade names are indicated on branches or to the right. ...
... Eriksson et al. (1998) revealed early on that the traditional Potentilla was nonmonophyletic, it should be separated into several genera; while some previously recognized small groups such as Duchesnea Smith (1811: 372), Horkelia Chamisso (1827: 26) and Ivesia Torrey & Gary (1858: 72) should be included within Potentilla. With more phylogenetic studies conducted, many major well-supported clades such as Anserina, Alba, Fragarioides, Reptans, Ivesia, Argentea and Himalayan clades have been identified within the large genus (Töpel et al. 2011, Eriksson et al. 2015, Feng et al. 2017. Additionally, new genus such as Chamaecallis Smedmark (2014: 180) was proposed based on molecular evidence (Eriksson et al. 2015). ...
... Additionally, new genus such as Chamaecallis Smedmark (2014: 180) was proposed based on molecular evidence (Eriksson et al. 2015). All the evidence suggests a mixture of many genera in traditional Potentilla (Eriksson et al. 1998, Soják 2008, Eriksson et al. 2015, Feng et al. 2017. The diagnostic characters of Potentilla vary widely along with the phylogenetic and morphological studies, the number of stamens, the anther and stipular auricles shape has been proposed to be taxonomically informative (Soják 2008, Eriksson et al. 2015, Feng et al. 2017, which can be used to delimit species from close related and morphologically similar genera in Potentilleae, such as Potentilla, Argentina Hill (1756: 6) and Sibbaldia Linnaeus (1753: 284). ...
... All the evidence suggests a mixture of many genera in traditional Potentilla (Eriksson et al. 1998, Soják 2008, Eriksson et al. 2015, Feng et al. 2017. The diagnostic characters of Potentilla vary widely along with the phylogenetic and morphological studies, the number of stamens, the anther and stipular auricles shape has been proposed to be taxonomically informative (Soják 2008, Eriksson et al. 2015, Feng et al. 2017, which can be used to delimit species from close related and morphologically similar genera in Potentilleae, such as Potentilla, Argentina Hill (1756: 6) and Sibbaldia Linnaeus (1753: 284). ...
Article
Full-text available
Potentilla sunhangii, a new species of Potentilla, from the Jin-hou-ling Mountains, Shen-nong-jia Forest District of Hubei province, China, is described and illustrated. The wild population of the new species is at an elevation of approximately 2930 m and has leaves palmately 3 foliolate. It is morphologically much similar to P. saundersiana, whilst being easily distinguished by its dense array of glands around the plant and green on both surfaces of leaves. Our molecular phylogenetic analysis based on nuclear and chloroplast confirmed the species belongs to Potentilla, in Argentea clade.
... Potentilleae is one of the taxonomically ambiguous tribes in family Rosaceae. The taxonomy of Potentilleae has been consistently changing over the recent years in world-wide studies (Feng et al., 2017). In the study conducted by Eriksson et al. (1998), described Potentilla as a non-monophyletic genus, combining previously recognized genera such as Duchesnea, Horkelia and Ivesia under Potentilla. ...
... All the consensus sequences generated in the present study were submitted to the GenBank under the accession numbers MK605458-MK60573 and MK587724-MK587739. To examine the precise phylogenetic position of the study species, we adapted the phylogeny constructed in a study by Feng et al. (2017) (Table 01) using both nuclear and plastid genetic markers. We attempted to reconstruct the tribe: Potentilleae phylogeny using the sequences generated in the present study and sequences reported in previous studies (Feng et al., 2017). ...
... To examine the precise phylogenetic position of the study species, we adapted the phylogeny constructed in a study by Feng et al. (2017) (Table 01) using both nuclear and plastid genetic markers. We attempted to reconstruct the tribe: Potentilleae phylogeny using the sequences generated in the present study and sequences reported in previous studies (Feng et al., 2017). We constructed the multiple sequence alignments separately for ITS and trnL-trnF in MEGA v7. ...
Article
Full-text available
Purpose : The nomenclature of the wild strawberries inhabited in Sri Lanka is ambiguous. In Sri Lanka, this species is still named Duchesnea indica which needs a revision. Wild strawberries grow well in natural habitats of upcountry in Sri Lanka. Since the commercial strawberry cultivations gain a popularity in upcountry, the studies on wild strawberry is essential for crop improvement and management. Research Method : In the present study, we conducted extensive field sampling followed by a phylogenetic analysis with the DNA barcoding markers ITS and, trnL-F by using a representative sample of wild strawberry plants in Sri Lanka. The distribution of the species was identified using maximum entropy modeling approaches. Findings : Sri Lankan wild strawberry got placed at subtribe: Potentilla, and clade: Reptans and show a shallow divergence with the species Potentilla indica reported. Thus, we reposition the genus of wild strawberries in Sri Lanka from Duchesnea to Potentilla and hereafter name it as P. indica. The niche model analysis predicted a highly restricted distribution of Sri Lankan wild strawberry in Nuwara-Eliya district over an area of 166.36 km 2 in the altitude range of 1546-2524 m in a small climatic envelop highlighting the need for urgent conservation measures. Research Limitations : The pop-set for available in literature of P. indica is limited for comparison. Extensive studies based on DNA sequencing is needed for further validation. Originality / Value : Taxonomy, narrow distribution, need of conservation, and phylogenetic distance to Fragaria chiloensis, a progenitor species of cultivated strawberry, are defined for Sri Lankan wild strawberries.
... Phylogenetic Analyses. Potentilla-type ITS and partial ETS sequences retrieved from BAC clones and Illumina datasets of Erythronium were added into a recent backbone phylogeny of Potentilla (59). This phylogeny was supplemented with additional Potentilla taxa having publicly available ITS sequences that were potentially closely related to the alien ITS of Erythronium. ...
Preprint
Full-text available
The occurrence of horizontal gene transfer (HGT) in Eukarya is increasingly gaining recognition. Nuclear-to-nuclear jump of DNA between plant species at high phylogenetic distance and devoid of intimate association (e.g., parasitism) is still scarcely reported. Within eukaryotes, components of ribosomal DNA (rDNA) multigene family have been found to be horizontally transferred in protists, fungi and grasses. However, in neither case HGT occurred between phylogenetic families, nor the transferred rDNA remained tandemly arrayed and transcriptionally active in the recipient organism. This study aimed to characterize an alien eudicot-type of 45S nuclear rDNA, assumingly transferred horizontally to the genome of monocot European Erythronium (Liliaceae). Genome skimming coupled by PacBio HiFi sequencing of a BAC clone were applied to determine DNA sequence of the alien rDNA. A clear phylogenetic signal traced the origin of the alien rDNA of Erythronium back to the Argentea clade of Potentilla (Rosaceae) and deemed the transfer to have occurred in the common ancestor of E. dens-canis and E. caucasicum. Though being discontinuous, transferred rDNA preserved its general tandemly arrayed feature in the host organism. Southern blotting, molecular cytogenetics, and sequencing of a BAC clone derived from flow-sorted nuclei indicated integration of the alien rDNA into the recipient’s nuclear genome. Unprecedently, dicot-type alien rDNA was found to be transcribed in the monocot Erythronium albeit much less efficiently than the native counterpart. This study adds a new example to the growing list of naturally transgenic plants while holding the scientific community continually in suspense about the mode of DNA transfer. Significance Statement Ribosomal DNA is an essential component of all cellular genomes. In plants, accidental movement of rDNA via horizontal gene transfer has only been reported in sexually incompatible grasses (monocots) where it involved non-functional rDNA units. In this study, we propose that evolutionary trajectories of eudicots and monocots were bypassed by the jump of rDNA from a Potentilla species (Rosaceae) to a common ancestor of Erythronium dens-canis and E. caucasicum (Liliaceae). The alien eudicot-type rDNA appeared relatively well conserved in the examined host Erythronium genome, being able to be expressed while preserving its general tandemly repeated feature, evidences that have no match in earlier literature.
... The extract from Potentilla rugulosa leaves has been identified as a potential functional food supplement for preventing the development of obesity (Choi et al. 2020). Among the many studies that have been performed on Potentilla plants, very few studies have been performed on P. sischanensis, except for one phylogenetic study (Feng et al. 2017). Here, we report the complete chloroplast (cp) genome of P. sischanensis and analyze its phylogenetic relationship with other related species. ...
Article
Full-text available
Potentilla sischanensis Bunge ex Lehm. is a widespread perennial herb in north China. The plant has little yellow flowers, and the petioles are white-tomentose and sparsely villous. To determine the chloroplast genome, total genomic DNA was extracted from fresh leaves and sequenced. The complete chloroplast genome was assembled and annotated. The chloroplast genome of this plant is a circular form with a length of 156,240 bp, including a large single-copy region (LSC, 85,748 bp), a small single-copy region (SSC, 18,566 bp), and two inverted repeats (IRs, 25,963 bp). A total of 132 genes were predicted, comprising 87 encoded proteins, 8 rRNAs and 37 tRNAs. The evolutionary history indicates that P. sischanensis was grouped within Potentilla and formed a clade with Potentilla chinensis and Potentilla stolonifera with a 100% bootstrap support value. The complete cp genome can serve as a reference for future studies on molecular biology, evolution, population genetics, taxonomy and resource protection.
... based on maximum likelihood (ML) using RAxML 8.2.10 (Stamatakis, 2014), with 1,000 bootstrap replicates employed for estimating node support. Potentilla fruticosa (NC_036423) and Drymocallis saviczii (NC_050966) were downloaded from Genbank and used as outgroups (Eriksson et al., 1998(Eriksson et al., , 2003Potter et al., 2000;Feng et al., 2017;Dimeglio et al., 2014). In total, 36 cp genome sequences were aligned using a Geneious prime 2021.1.1 plugin MAFFT v.7.450 (Katoh and Standley, 2013), and the alignment was manually adjusted when necessary. ...
Article
Full-text available
Species identification is vital for protecting species diversity and selecting high-quality germplasm resources. Wild Fragaria spp. comprise rich and excellent germplasm resources; however, the variation and evolution of the whole chloroplast (cp) genomes in the genus Fragaria have been ignored. In the present study, 27 complete chloroplast genomes of 11 wild Fragaria species were sequenced using the Illumina platform. Then, the variation among complete cp genomes of Fragaria was analyzed, and phylogenetic relationships were reconstructed from those genome sequences. There was an overall high similarity of sequences, with some divergence. According to analysis with mVISTA, non-coding regions were more variable than coding regions. Inverted repeats (IRs) were observed to contract or expand to different degrees, which resulted in different sizes of cp genomes. Additionally, five variable loci, trnS-trnG, trnR-atpA, trnC-petN, rbcL-accD, and psbE-petL, were identified that could be used to develop DNA barcoding for identification of Fragaria species. Phylogenetic analyses based on the whole cp genomes supported clustering all species into two groups (A and B). Group A species were mainly distributed in western China, while group B contained several species from Europe and Americas. These results support allopolyploid origins of the octoploid species F. chiloensis and F. virginiana and the tetraploid species F. moupinensis and F. tibetica. The complete cp genomes of these Fragaria spp. provide valuable information for selecting high-quality Fragaria germplasm resources in the future.
... Hybridization, polyploidy, and apomixis are not rare among its species [4][5][6][7][8][9][10], which make the taxonomy of the group very complicated. The existing phylogenies of Potentilla and the Potentilleae tribe are based on relatively small subsets of taxa and are still far from comprehensive [2,[11][12][13][14][15]. Even in the cases where the taxa sets of Potentilleae were quite comprehensive [2,13], some groups of Potentilla s. str. ...
Article
Full-text available
The results of a molecular genetic study of Potentilla multifida agg. using two plastid markers (ndhC-trnV and psbA-trnH) and a nuclear ITS marker suggested that this group comprises a number of relatively young and incompletely differentiated species widely distributed in Northern Eurasia. The sequences were analyzed using tree-based (maximum likelihood) and network-based (statistical parsimony network) approaches. The plastid data suggested incomplete lineage sorting, characteristic of the group as a whole. The nuclear ITS results demonstrated quite a different pattern, with mostly conspecific accessions shaping monophyletic clades. The majority of the Potentilla sect. Multifidae species studied possess few, usually closely related plastid haplotypes, or are even monomorphic. In contrast, P. volgarica, a narrow endemic from the Volga River valley, presents plastid haplotypes belonging to two distantly related groups. Such a pattern of genetic diversity in P. volgarica may be explained by a long persistence of the species within an extremely small distribution range, on the right bank of the Volga River, most likely representing a contemporary refugium. The genealogy of plastid markers in P. volgarica suggests that this species is ancestral to P. eversmanniana, another narrow endemic from the S Urals.
Article
Chamaerhodos is a small genus with ca 7–8 species with a disjunct distribution in Asia and western North America. Due to limited sampling of species and genes in previous studies, little is known about the phylogenetic relationships among the species. Moreover, chloroplast genomic resources for Chamaerhodos have been limited. Herein, we conducted a comparative analysis of the complete chloroplast (cp) genomes of five Chamaerhodos species. The five cp genomes had a typical quadripartite structure with high conservation of gene content and gene order. These five cp genomes encoded an identical set of 129 genes, including 84 protein-coding genes, 37 tRNA genes and eight rRNA genes. Comparison of the boundaries between the IRs and single copy regions revealed only very slight boundary differences and the five cp genomes showed only little sequence divergence. Seven regions (trnR-UCU-atpA, trnS-GCU-trnG-UCC, TrnG-GCC-trnfM-CAU, trnL-UAA intron, trnH-GUG-psbA, trnF-GAA-ndhJ, matK-5′-trnK-UUU) were identified as excellent candidate markers. A total of 117, 108, 116, 117 and 118 perfect SSRs were detected in the cp genomes of Ch. altaica, Ch. canescens, Ch. erecta, Ch. sabulosa and Ch. trifida, respectively. A phylogenetic analysis support the monophyly of Chamaerhodos and place it as part of the Fragariinae clade sister to the clade of Dasiphora, Drymocallis and Potaninia. Within Chamaerhodos, Ch. canescens was inferred sister to a clade containing the other four species. The detailed characterization of the Chamaerhodos chloroplast genomes sets the foundation for further studies on the genus and its close relatives in the economically important Rosaceae.
Thesis
Full-text available
Premise and aims of the thesis: Eukaryotes that have more than the two standard sets of chromosomes are called polyploids. The process of going through a whole genome duplication is called polyploidization, and this is a common mechanism in plant speciation. There are two main types of polyploidy: autopolyploidy and allopolyploidy. Autopolyploidy is the result of a genome duplication within one species, and allopolyploidy is the result of a genome duplication following a hybridization between two different species. The genus Potentilla in the rose family (Rosaceae) is remarkable in that species range in ploidy level from diploid (2x) to hexadecaploid (16x). They are found all around the Northern Hemisphere, from lowland to mountain regions, and are generally characterized by yellow flowers and palmately compound leaves. However, taxonomists have long had problems with agreeing on which species should be included in the genus. Even today some authors exclude certain species from Potentilla, with the consequence of making it a non-monophyletic genus (i.e. not all of the descendants of their most recent common ancestor are included in the group). Even though the prevalence of polyploidy in plants is well-known, it has not been reflected in phylogenetic research. This is a problem, especially concerning allopolyploids, because the understanding we get of their evolutionary history is then much simplified. Many studies have used DNA sequences that often represent only one ancestral lineage (chloroplast, nuclear ribosomal), thus omitting parts of the species’ heritage. In previous phylogenetic analyses, a few major subclades were identified in Potentilla (informally named Alba, Anserina, Argentea, Fragarioides, Ivesioid and Reptans), but their relationships to one another differed depending on what type of DNA was studied. In addition, some species were found in different subclades in trees based on different DNA sequences. The fact that these sequences may be uniparentally inherited and that most of the species are polyploid led to an interpretation of an evolutionary history that involves hybridization and polyploidization in Potentilla. The type of DNA sequence best suited for investigating the evolutionary history of polyploids are low-copy nuclear DNA markers (LCN markers). They are present in each subgenome and inherited from both the maternal and the paternal parent. Thus, they have the potential to trace the relationships of each ancestral lineage of polyploids. LCN markers were in this thesis used for three different purposes in Potentilla: 1), to infer the relationships of the major subclades in the genus (Paper I); 2), to trace the putative hybrid origins of a number of North American polyploid species in the ‘Rivales group’ (Papers II and III); and 3), to assess the generic delimitation of Potentilla (Paper IV). Results and conclusions: A fully resolved and supported tree showing the major subclades in Potentilla was obtained after excluding the Fragarioides species from the dataset. Two of the clades, the Ivesioid and Reptans clades, showed signs of being of autopolyploid origin. In contrast, five of the six species in the Rivales group occurring in North America were inferred to be allopolyploids with ancestral lineages in the Argentea and Ivesioid clades. Thus, hybridization and polyploidization seem to have played a larger role later in the evolution of the genus, after the major clades diverged. Four lines of evidence – ploidy level, distribution of extant species, relationships seen in the gene trees, and a set of network analyses – indicated that precursors to three of the North American Rivales species have taken part in hybridizations that eventually formed a common ancestor for the high-ploidy Rivales species P. intermedia and P. norvegica. Parts of this population dispersed to Eurasia, while the rest remained in North America. Both lineages went through at least one more hybridization each and formed P. intermedia in Eurasia and P. norvegica in North America. Since many floras state that P. norvegica is of European origin, this will have implications for its assessment as native or introduced on both continents. The gene trees inferred in Papers I, II and III showed a network of gene flow between the Alba, Argentea, Fragarioides, Ivesioid and Reptans clades. Thus, the generic delimitation of Potentilla was set to include these clades, and excluding the Anserina clade. With this delimitation only six species, out of the ca 400 in the whole genus, had to be recombined to get new Potentilla names. Future perspectives: The LCN markers revealed relationships that could not have been found by the traditionally used chloroplast or nuclear ribosomal markers. This points to the importance of continuing using LCN markers when investigating the evolutionary history of polyploids. Additional markers are, however, needed to resolve some relationships, especially the putatively diploid Fragarioides species destabilizing the backbone phylogeny, and some species in the Rivales group of which we could not find all putative ancestral lineages. The High-Throughput Sequencing technique Target Capture could potentially generate enough data to solve these problems. Software programs that analyze reticulate evolution still struggle with species of high ploidy levels, and a good deal of manual preparation of analyses and interpretation of the results are still needed. In addition, a discussion is needed concerning criteria for species delimitation of allopolyploids. If the ancestral lineages are distantly related, this could have implications at even higher taxonomical levels, such as genera and families.
Article
Full-text available
Three species originally published as members of the genus Potentilla, later classified as Sibbaldiopsis, are here transferred in the genus Sibbaldia.
Article
Full-text available
The internal transcribed spacer (ITS) region of 18S-26S nuclear ribosomal DNA (rDNA) was sequenced in 65 taxa representing most coneflowers (i.e., species in Dracopis, Echinacea, Ratibida, and Rudbeckia) and other taxa representing 21 outgroup genera of tribe Heliantheae. Results of parsimony analysis of the rDNA dataset by itself and in combination with the cpDNA dataset uphold the hypothesis from an earlier cpDNA restriction site study that Echinacea is not closely related to the other three genera of coneflowers. The data support placement of Echinacea in subtribe Zinniinae. The remaining three coneflower genera represent a monophyletic lineage corresponding to subtribe Rudbeckiinae sensu H. Robinson. The rDNA data support two sublineages in Rudbeckia congruent with the two traditionally recognized subgenera, subg. Macrocline and subg. Rudbeckia. In subg. Macrocline, two geographic areas of diversification are indicated: southeastern and western United States. The widespread species R. laciniata is placed strongly with the western lineage of subg. Macrocline. The rDNA data support transfer of Dracopis to Rudbeckia subg. Macrocline, a relationship also supported by multiple morphological characters. The rDNA data do not confidently resolve the sister group of Rudbeckiinae from among the members of Heliantheae investigated. Multiple hypotheses are suggested for the outgroup taxa, such as expanding subtribe Zinniinae to include Echinacea and Trichocoryne, a genus previously regarded as belonging to subtribe Hymenopappinae (Heleneae or Heliantheae sensu lat.). Our findings further support expansion of subtribe Engelmanniinae to include Balsamorhiza, Borrichia, and Wyethia even though these taxa lack ray floret complexes and have fertile disc ovaries. We suggest that bioprospectors might usefully search among taxa of Zinniinae for bioactive substances similar to the immune stimulants of Echinacea.
Article
The Argentina group of the genus Potentilla is classified here as a distinct genus, supported by results of molecular studies, on the basis of an until recently unused morphological difference (ventral stipular auricles). Species of this group formerly included in Potentilla sect. Anserina are transferred to the separate genus Argentina.
Article
A monographic study of the genus Sibbaldia (Rosaceae) is carried out. A total of 10 species are recognized of which 9 are dominantly Asian species whereas 1 species (Sibbaldia procumfechs) is distributed in Asia, Europe and North America. Taxonomic criterion, key to the closely related genera and species along with distribution, ecological notes, speciemen citation, synonyms and nomenclatural notes are also given.
Article
This paper presents identification keys, citations of names accepted by the present author including synonyms and drawings of leaves of all extra-European species of Potentilla s.l. of the Old World excluding Asian species of the former USSR, which have been dealt with previously (Soják 2004).
Article
Potentilla lineata and allied species (Rosaceae) in the Himalaya and adjacent regions are revised. Potentilla festiva Soják, P. josephiana H. Ikeda & H. Ohba, P. lineata Trev., P. fallens Card. and P. polyphylla Wall. ex Lehm. are recognized. Potentilla josephiana is a new name for P. fulgens Wall. cx Hook. var. intermedia Hook. f. Four varieties are recognized in P. polyphylla: var. polyphylla; var. himalaica H. Ikeda & H. Ohba, var. nov.; var. interrupta (Yü & Li) H. Ikeda & H. Ohba, stat. & comb. nov.; and var. barbata Lehm. A polyploid series is found in this group. Four putative hybrids between the species are also recognized.