ArticlePDF Available

Towards a phylogenetically informed taxonomy of Cyrtandra (Gesneriaceae) in the Solomon Islands

Article

Towards a phylogenetically informed taxonomy of Cyrtandra (Gesneriaceae) in the Solomon Islands

Abstract and Figures

The objective of this study was to build on a previously published molecular phylogenetic hypothesis for Cyrtandra to identify major clades of the genus in the Solomon Islands. Species of Cyrtandra in the Pacific are represented by locally uncommon, poorly known and rarely collected taxa, a large number of which are at particular risk of extinction due to high rates of deforestation. We propose that research in Cyrtandra has progressed to the point that regional taxonomic revisions can be conducted effectively following procedures developed for Cyrtandra in Malesia and Hawaii. We identify four distinct clades in the Solomon Islands, each characterized by suites of morphological characters that can be used to circumscribe taxa for revision. Both genetically sampled taxa for the phylogenetic analysis and unsampled taxa are evaluated based on morphological traits that differentiate these four clades. A hypothesis of relationship is proposed as a model for future taxonomic revision of the genus in this region that includes 24 species and two additional taxa possibly new to science. A diagnostic key to the clades currently recognized in the Solomons and an annotated checklist of recognized species are also included as a foundation for future work in the archipelago.
Content may be subject to copyright.
TOWARDS A PHYLOGENETICALLY INFORMED TAXONOMY OF
CYRTANDRA (GESNERIACEAE)IN THE SOLOMON ISLANDS
JOHN R. CLARK*
National Tropical Botanical Garden (PTBG), 3530 Papalina Road, Kalaheo, HI 97641.
Rancho Santa Ana Botanic Garden (RSA), 1500 North College Avenue, Claremont, CA 91711.
Marie Selby Botanical Gardens (SEL), 811 South Palm Avenue., Sarasota, FL 34236.
Email: jclark@ntbg.org
HANNAH J. ATKINS
Royal Botanic Garden Edinburgh, 20a Inverleith Row, Edinburgh, EH3 5LR, United Kingdom.
GEMMA L.C. BRAMLEY
Royal Botanic Gardens, Kew, Richmond, Surrey, TW9 3AB, United Kingdom.
DIANA D. JOLLES
Rancho Santa Ana Botanic Garden, 1500 North College Avenue, Claremont, CA 91711.
ERIC H. ROALSON
School of Biological Sciences, Washington State University, 339 Abelson Hall, Pullman, WA 99164.
WARREN L. WAGNER
National Museum of Natural History, Smithsonian Institution, PO Box 37012, Washington, DC 20013.
ABSTRACT. The objective of this study was to build on a previously published molecular phylogenetic
hypothesis for Cyrtandra to identify major clades of the genus in the Solomon Islands. Species of Cyrtandra in
the Pacific are represented by locally uncommon, poorly known and rarely collected taxa, a large number of
which are at particular risk of extinction due to high rates of deforestation. We propose that research in
Cyrtandra has progressed to the point that regional taxonomic revisions can be conducted effectively following
procedures developed for Cyrtandra in Malesia and Hawaii. We identify four distinct clades in the Solomon
Islands, each characterized by suites of morphological characters that can be used to circumscribe taxa for
revision. Both genetically sampled taxa for the phylogenetic analysis and unsampled taxa are evaluated based on
morphological traits that differentiate these four clades. A hypothesis of relationship is proposed as a model for
future taxonomic revision of the genus in this region that includes 24 species and two additional taxa possibly
new to science. A diagnostic key to the clades currently recognized in the Solomons and an annotated checklist
of recognized species are also included as a foundation for future work in the archipelago.
Key words: island biogeography, systematics, dispersal-mediated allopatry, floristics, southeast Asia
INTRODUCTION
The flowering plant family Gesneriaceae in-
cludes over 3000 species found throughout much
of the world’s tropics with centers of diversity in
the tropical Americas, Africa, southeast Asia and
the Pacific (Wiehler 1983, Burtt & Wiehler 1995,
Weber 2004, Weber et al. 2013). Gesneriads have
received great scientific attention in the last two
decades resulting in substantial advances in our
knowledge of this taxonomically challenging
family (see Mo¨ller & Clark 2013; Weber et al.
2013). Of genera in the family, Cyrtandra
J.R.Forst. & G.Forst. is perhaps the most chal-
lenging taxonomically, owing to a complex
morphological diversification as well as the great
geographic range and sheer number of reported
species (Atkins et al. 2013).
Cyrtandra is the largest gesneriad genus (700 or
more species; see Atkins et al. 2013) and has the
greatest range for any genus in the family,
extending from the Nicobar Islands in the Indian
Ocean to the distant Hawaiian and Marquesan
islands in the Pacific (Burtt 2001, Kiehn 2001,
Atkins et al. 2013). The broad distribution and
large number of species in Cyrtandra is in marked
contrast to other gesneriads, including the genera
Aeschynanthus Jack (,185 spp.), Boea Comm. ex.
Lam. (,14 spp.), Coronanthera Vieill. ex
C.B.Clarke (,11 spp.) and Epithema Blume
Selbyana selb-31-02-07.3d 3/12/13 12:49:18 166
Selbyana 31(2): 166–183. 2013.
166
(,20 spp.), which reach their easternmost extent
in the Solomon Islands (Skog & Boggan 2007).
Species of Cyrtandra are commonly found in
wet upland tropical forests (Backer & Van den
Brink 1965; Gillett 1967, 1973; Fosburg & Sachet
1981; Burtt 1990, 2001; Smith 1991; Cronk et al.
2005) but in some instances are also found
growing at sea level (e.g., C. samoensis on the
island of Niue; Gillett 1973). Morphologically, the
genus is quite diverse, exhibiting a wide range of
flower, fruit and habit types (Clarke 1883,
Hillebrand 1888, Schlag-Edler & Kiehn 2001).
Cyrtandra species are commonly understory
shrubs or trees but can also be small terrestrial
herbs, woody vines and sometimes epiphytic
subshrubs (Burtt 2001). Flowers are commonly
white but can be yellow, pink or red and are born
in the axils of terminal stem leaves. Occasionally
inflorescences are borne along the woody trunks or
even on exposed roots (e.g., C. kaulantha
H.St.John & Storrey from Oahu; Wagner et al.
1990, 1999). Fruits are hard, normally indehiscent
capsules or, as is often the case in the Pacific and
in a few species in Malesia, fleshy berries (Hilliard
et al. 2003).
Tropical plant families like the Gesneriaceae are
under threat of extinction from deforestation
(Krupnick & Kress 2005), and warrant intensified
in situ and ex situ study and conservation efforts
(Pence 2002). This is particularly true for the
genus Cyrtandra; despite the exceptional geo-
graphic range for the genus, frequently species are
regionally restricted endemics rarely extending
beyond archipelagos (Samuel et al. 1997). More
often, species of Cyrtandra are restricted to single
islands or even confined areas within islands
(Atkins et al. 2001; Bramley et al. 2003, 2004a,
2004b; Bramley 2005; Cronk et al. 2005). As such,
many species of Cyrtandra are threatened and can
be summarily eliminated through destructive
practices such as logging. Species of Cyrtandra
are also quite susceptible to disturbance, and even
areas where only selective logging has taken place,
or livestock grazing has occurred, densities of the
genus are greatly reduced or completely eliminat-
ed (authors this paper; pers. obs.).
Given the severity of loss in the genus, urgency
exists to systematically sample, interpret relation-
ships and apply accepted taxonomic concepts to
both genetically sampled and unsampled taxa (see
Emerson 2002, Hopper 2000). This methodology
has been applied, in part, in Malesia to revise
section Dissimiles C.B.Clarke (Bramley 2005) and
as a basis for taxonomic revision in Hawaii (see
Clark et al. 2009). Sixty morphologically unique
species of Cyrtandra are now recognized in
Hawaii based on a phylogenetically informed
understanding of relationships in that archipelago
(Wagner et al. in prep.). Applying these strategies
in other areas where Cyrtandra is found will aid in
both defining and revising taxonomy in this poorly
circumscribed genus, and also in identifying
monophyletic clades for future systematics work,
field colleting prioritization and conservation
assessment (Atkins et al. 2013).
Among areas warranting intensified study for
Cyrtandra, the interface between continental
Southeast Asia and the remote Pacific, an area
ranging from the Solomon Islands to the west, and
east into the Pacific as far as Samoa, has been
implicated as a center of origin for the Pacific
clade of Cyrtandra (Clark 2008; Clark et al. 2008,
2009). A general west to east trend is common
among Pacific islands floras and areas like Fiji
share as much as 90%of their plant genera with
New Guinea and neighboring islands (Mueller-
Dombois & Fosberg 1998). It is possible that other
species have similar evolutionary histories tied to
this region (Takhtajan 1969, 1986; Evenhuis &
Bickel 2005; Balke et al. 2009; see also Cowie &
Holland 2006 on dispersal). Results from biogeo-
graphic research (Clark et al. 2009) support the
hypothesis that once the barrier to Pacific dispersal
had been overcome (presumably a result of fleshy
fruit dispersed by shore birds), Cyrtandra was well
suited for island hopping (see Cronk et al. 2005),
and continued to do so across the Pacific. As such,
interpreting affinities for Pacific Cyrtandra all
require a greater understanding of phylogenetic,
taxonomic and biogeographic relationships in the
genus in this region (Clark 2008).
The purpose of the current study is to present a
new phylogeny for Cyrtandra inclusive of spec-
imens collected on a 2009 expedition to the
Solomon Islands. The first and only revision of
the genus in this archipelago was conducted by
Gillett (1975). Recent fieldwork in the Solomons
by one of the authors (JRC) suggests that
additional undescribed species of Cyrtandra exist
there and current taxon delimitations are in need of
amendment.
The archipelago represents a manageable sys-
tem to both apply and refine techniques for
addressing this genus taxonomically and can be
used in the future to address similar issues in
geographically more complex and less taxonom-
ically treated regions including Borneo and New
Guinea (see Atkins et al. 2013). The phylogeny
presented in this study is meant to demonstrate the
minimum number of estimated clades of Cyrtan-
dra in the Solomon Islands, and to define these
monophyletic clades using morphology as a basis
for future taxonomic revision. A diagnostic key to
these four clades is provided and an annotated
checklist of recognized species is included with
details on distribution and clade affinities, where
appropriate.
Selbyana selb-31-02-07.3d 3/12/13 12:49:19 167
CLARK ET AL.: CYRTANDRA OF THE SOLOMON ISLANDS 167
Selbyana selb-31-02-07.3d 3/12/13 12:49:19 168
TABLE 1. Taxon sampling list for 54 individuals sampled in the current study including four outgroup species (Aeschynanthus spp.). Specimens are organized alphabetically
by species. ID numbers are J. R. Clark’s DNA extraction numbers and are here used for reference. GenBank accession numbers are included for all taxa for each of the
three genic regions analyzed in the current study. C. 5Cyrtandra;A. 5Aeschynanthus.ITS5internal transcribed spacer regions 1 and 2, including the 5.8S subunit;
ETS 5external transcribed spacer region; psbA-trnH 5chloroplast sequence data.
Species ID No. Collector and no.; voucher Origin ITS ETS psbA-trnH
A. hartleyi P.Woods C0217 Clark 826; PTBG New Guinea KF148661 KF148649 KF148673
A. longicaulis Wall. ex R.Brown C0056 MSBG 1974-2207-W Indonesia EU919959 EU919898 EU920018
A. solomonensis P.Woods C0203 Clark 767; PTBG Solomon Islands, Isabel KF148657 KF148645 KF148669
A. tricolor Hook. C0055 MSBG 1974-1760-W Indonesia EU919958 EU919897 EU920017
C. anthropophagorum Seem. ex A.Gray C0114 Clark 688; PTBG Fiji, Viti Levu EU919987 EU919926 EU920042
C. aurantiicarpa G.W.Gillett C0076 Clark 655; PTBG Samoa, Savai’i EU919971 EU919910 EU920030
C. baileyi F.Muell. in Bailey, F.M. DD6 Costion 1687; CNS Australia, Queensland KF148662 KF148650 KF148674
C. aff. bidwillii C.B.Clarke C0130 Wood 11072; PTBG Society Islands, Hua Hine GQ475176 GQ475089 GQ475139
C. biserrata H.St.John C0153 Wood 11386; PTBG Hawai’i, Moloka’i GQ475194 GQ475107 GQ475157
C. calpidicarpa (Rock) H.St.John & Storey C0053 Clark 584; PTBG Hawai’i, O’ahu GQ475168 GQ475080 GQ475123
C. compressa C.B.Clarke C0074 Clark 652; PTBG Samoa, Savai’i EU919970 EU919909 EU920029
C. confertiflora (Wawra) C.B.Clarke C0159 Roalson 1584-01; WS Hawai’i, Kaua’i GQ475200 GQ475113 GQ475163
C. cordifolia Gaudich. C0048 Clark 579; PTBG Hawai’i, O’ahu EU919955 EU919894 EU920014
C. erectiloba G.W.Gillett C0179 Clark 745; E, K, PTBG, U.S. Solomon Islands, Isabel KF148652 KF148640 KF148664
C. falcifolia C.B.Clarke C0141 Kiehn 940823-4/3; WU Samoa, U’polu GQ475184 GQ475097 GQ475147
C. feaniana F.Br. C0059 Price 200; PTBG Marquesas, Hiva Oa EU919960 EU919899 EU920019
C. ferruginea Merr. C0137 Cubey and Scott 226; E Philippines, Luzon GQ475181 GQ475094 GQ475144
C. filibracteata B.L.Burtt C0205 Clark 770; E, K, PTBG, U.S. Solomon Islands, Kolombangara KF148658 KF148646 KF148670
C. fulvovillosa Rech. C0183 Clark 786; E, K, PTBG, U.S. Solomon Islands, Kolombangara KF148654 KF148642 KF148666
C. grayana Hillebr. C0103 Clark 666; PTBG Hawai’i, Maui EU919982 EU919921 EU920039
C. hawaiensis C.B.Clarke C0155 Roalson 1569-04; WS Hawai’i, O’ahu GQ475196 GQ475109 GQ475159
C. kauaiensis Wawra C0026 Clark 556A; PTBG Hawai’i, Kaua’i EU919940 EU919879 EU919999
C. kealiae Wawra ssp. urceolata
W.L.Wagner & Lorence
C0054 Perlman 18805; PTBG Hawai’i, Kaua’i EU919957 EU919896 EU920016
C. kusaimontana Hosok. C0033 Flynn 5995; PTBG Micronesia, Kosrae EU919945 EU919884 EU920004
C. laxiflora H.Mann C0157 Roalson 1574-01; WS Hawai’i, O’ahu GQ475198 GQ475111 GQ475161
C. leucantha A.C.Sm. C0116 Clark 693; PTBG Fiji, Viti Levu EU919988 EU919927 GQ475136
C. longifolia (Wawra) Hillebr. ex C.B.Clarke C0023 Clark 551; PTBG Hawai’i, Kaua’i EU919939 EU919878 EU919998
C. macrotricha G.W.Gillett C0180 Clark 747; PTBG Solomon Islands, Isabel KF148653 KF148641 KF148665
C. cf. mesilauensis B.L.Burtt C0140 Cubey and Scott 229; E Indonesia, Borneo GQ475183 GQ475096 GQ475146
C. munroi C.N.Forbes C0151 Oppenheimer H120638; BISH Hawai’i, Lana’i GQ475192 GQ475105 GQ475155
C. occulta A.C.Smith C0117 Clark 694; PTBG Fiji, Viti Levu EU919989 EU919928 GQ475137
C. cf. occulta A.C.Smith C0119 Clark 702; PTBG Fiji, Viti Levu EU919990 EU919929 GQ475138
C. pendula Blume C0098 Wiriadinata, H. 12716; U.S. Indonesia EU919979 EU919918 EU920037
C. picta Blume C0097 Wiriadinata, H. 12715; U.S. Indonesia EU919978 EU919917 EU920036
C. pogonantha A.Gray C0081 Clark 660; PTBG Samoa, U’polu GQ475173 GQ475086 GQ475129
C. pulchella Rich ex A.Gray C0029 Lorence 8525; PTBG Samoa, Tau EU919941 EU919880 EU920000
C. richii A.Gray C0073 Clark 651; PTBG Samoa, Savai’i GQ475171 GQ475084 GQ475127
C. samoensis A.Gray C0030 Lorence 8633; PTBG Samoa, Ofu EU919942 EU919881 EU920001
168 SELBYANA Volume 31(2) 2013
MATERIALS AND METHODS
For the phylogenetic analysis, 50 ingroup taxa
(Cyrtandra) and four outgroup taxa (Aeschy-
nanthus) were sampled (Table 1). This study
builds on two previously published studies, Clark
et al. (2008, 2009), and incorporates 12 new
samples including eight from the Solomon Islands
representing at least five species, two of which are
possibly new to science. Pacific taxa in the current
study include representatives from all major
supported clades reported by Clark et al. (2009).
For a distribution map of the genus Cyrtandra see
Figure 1 in Clark et al. (2008) and for current
estimated numbers of species by region, see
Atkins et al. (2013).
DNA isolation and purification from silica-dried
leaf material, genic region amplification and
sequencing were performed using protocols de-
scribed by Clark et al. (2008). Cycle sequence
products were purified using standard gel column
purification methods and then analyzed on an
automated DNA Sequencer at facilities at the
University of Florida’s Interdisciplinary Center
for Biotech Research. Forward and reverse se-
quencing reactions were conducted for each taxon
sampled. Sequence chromatograms were proofed,
edited and contigs were assembled using Geneious
6.0.5 (Biomatters, Ltd., New Zealand). Edited
contigs were then aligned using MUSCLE (v.
3.8.31; Edgar 2004) as implemented in SeaView
(Gouy et al. 2010) with further editing by eye to
correct minor alignment issues. The nuclear
ribosomal internal transcribed spacer region, in-
cluding ITS1, ITS2 and the 5.8S subunit, the 59end
external transcribed spacer region (ETS) and the
chloroplast psbA-trnH region, were amplified using
protocols described in Clark et al. (2008).
Nucleotide substitution model parameters were
estimated using DT-ModSel (Minin et al. 2003).
Aligned sequences were analyzed using maximum
likelihood (ML) and Bayesian inference (BI)
methods. Maximum likelihood analyses were
performed using PAUP* 4.0b10 (Swofford 2002)
implementing heuristic searches with TBR branch
swapping and initial starting trees generated using
neighbor-joining reconstruction. Bootstrap support
indices were estimated for each node using 100
heuristic bootstrap replicates executed over 100
random addition cycles with 10 trees saved per
cycle (Hillis & Bull 1993). Bayesian analyses
were performed using MrBayes v. 3.1 (Huelsen-
beck & Ronquist 2001). Four chains were run for
30,000,000 generations each, sampled every
10,000 generations. Multiple independent BI
analyses were run. The first 20%of trees were
excluded from summary as burn-in. Posterior
probabilities were compared between two identi-
cal runs to ensure convergence had been reached.
Selbyana selb-31-02-07.3d 3/12/13 12:49:19 169
Species ID No. Collector and no.; voucher Origin ITS ETS psbA-trnH
C. samoensis A.Gray C0031 RP 71221; PTBG Tonga EU919943 EU919882 GQ475122
C. samoensis A.Gray C0142 Kiehn 940819-1/1; WU Samoa, U’polu GQ475185 GQ475098 GQ475148
C. samoensis A.Gray C0144 Kiehn 940819-2/1; WU Samoa, U’polu GQ475187 GQ475100 GQ475150
C. serratifolia H.J.Atkins & Cronk C0136 Cubey and Scott 225; E Indonesia, Sulawesi GQ475180 GQ475093 GQ475143
C. sp. C0095 Wiriadinata, H. 12709; U.S. Indonesia, Java EU919976 EU919915 EU920034
C. sp. ‘‘A’’ C0173 Clark 724; PTBG Solomon Islands, Isabel KF148651 KF148639 KF148663
C. sp. ‘‘B’’ C0209 Clark 787; PTBG Solomon Islands, Kolombangara KF148659 KF148647 KF148671
C. sp. ‘‘Java’’ C0211 Clark 856 (5MSBG
2009-0654A); PTBG
Java KF148660 KF148648 KF148672
C. subulibractea G.W.Gillett C0184 Clark 788; E, K, PTBG, U.S. Solomon Islands, Kolombangara KF148655 KF148643 KF148667
C. subulibractea G.W.Gillett C0191 Clark 805; K, PTBG Solomon Islands, Vangunu KF148656 KF148644 KF148668
C. tahuatensis Fosberg & Sachet C0061 Perlman 18399; PTBG Marquesas, Fatu Hiva EU919962 EU919901 EU920021
C. thibaultii Fosberg & Sachet C0062 Meyer 2541; PTBG Marquesas, Ua Pou EU919963 EU919902 EU920022
C. thibaultii Fosberg & Sachet C0065 Wood 10428; PTBG Marquesas, Ua Pou EU919966 EU919905 EU920025
C. tintinnabula Rock C0012 Perlman 17676; PTBG Hawai’i, Hawai’i EU919930 EU919869 GQ475114
C. urvillei C.B.Clarke C0034 Lorence 7838; PTBG Micronesia, Kosrae EU919946 EU919885 EU920005
C. wainihaensis H.Le´ v. C0021 Clark 549; PTBG Hawai’i, Kaua’i EU919937 EU919876 EU919996
TABLE 1. Continued.
CLARK ET AL.: CYRTANDRA OF THE SOLOMON ISLANDS 169
In the previous studies (Clark et al. 2008, 2009),
ML analyses were conducted on individual gene
trees and then compared with one another to
assess compatibility of genic regions for combined
analysis. No well-supported branches ($70%
bootstrap support) among the various topologies
were in conflict and thus the three genic regions
were combined and analyzed. This precedent was
followed for the current study.
Clade assignments were based on phylogenetic
results from the above described analysis. Mor-
phological assessments were made for Solomon
Islands species in supported clades (with $70%
ML bootstrap and/or $90%Bayesian posterior
probability) by qualitatively comparing characters
of species within and among monophyletic clades
and then using these to define characters repre-
sentative of each. Characters used include gross
vegetative morphology, inflorescence location and
type (e.g., cauliflorous vs. axillary, bracteate vs.
nonbracteate). A diagnostic key was then con-
structed to morphologically define identified
clades and also to serve as a tool to hypothesize
relationships of taxa not genetically sampled in
this study. In other words, the key provided is used
to assign Solomon Islands species not sampled
genetically to form a hypothesis of relationship for
further study. Genetically unsampled species were
assessed for inclusion in one of the defined clades
using their descriptions in the literature (Gillett
1973, 1974, 1975) along with personal observa-
tions of type and other representative material
including live specimens and photographs.
An annotated checklist of currently recognized
species from the Solomon Islands was compiled
from the literature (principally Gillett 1975, and
literature cited therein). The Solomons are inter-
preted as a biogeographic continuum beginning in
the eastern islands of the Bismarck Archipelago
and extending into the southeastern islands of the
Solomons including those geologically aligned
with Vanuatu. This area corresponds with Muel-
ler-Dombois and Fosberg’s Western Melanesia
Region (1998).
The Bismarck Archipelago is defined to include
islands east of but not including New Guinea and
ending north of Bougainville Island. The Solomon
Islands include Bougainville (a UN-designated
provisional independent state) and all those islands
politically affiliated with the Solomons including
the Santa Cruz group. While Vanuatu is not treated
in this study, at least one species overlaps with the
Solomons and is included for reference as the
Banks Islands and southward.
Types and vouchers were examined, when
possible, either physically or virtually via web-based
image libraries maintained by relevant herbaria. For
species observed in the wild by one of the authors
(JRC) as well as additional herbarium specimens in
various herbaria, comments on distribution and
affinities are included where appropriate.
RESULTS
Phylogenetics
Aligned sequences were 788 bp for ITS,
including the 5.8S subunit, 494 bp for ETS and
466 bp for psbA-trnH (excluding the 59end
inversion, as described in Clark et al. 2009), for
a total of 1748 bp of aligned sequence data.
Nucleotide substitution parameters were estimated
for the combined genic region matrix as the K80
plus gamma substitution model with a transition to
transversion ratio of 1.429 and a gamma shape of
0.541. Combined ML analysis of ITS, ETS and
psbA-trnH genic regions using this model resulted
in one most likely tree (FIGURE 1).
Support values across the topology, as indicat-
ed with ML bootstrap values (BS) and Bayesian
posterior probabilities (BI), illustrate a well-
supported phylogenetic hypothesis. These results
parallel relationships inferred in the previous
studies (Clark et al. 2008, 2009). Southeast Asian
taxa are situated in a grade of taxa leading to
the Pacific clade (FIGURE 1). Four supported
clades are recovered for sampled Solomon Islands
specimens. These are distributed along the
Southeast Asian grade (clade 1,100%BS,
.99%PP; and clade 2,100%BS, .99%PP), sister
to the monophyletic Pacific clade (clade 3,100%
BS, .99%PP) and within the Pacific clade sister
to the South Pacific clade (clade 4, 90%BS,
.99%PP). While clades 1 and 2 could represent
either one or two distinct clades for discussion
based on their sister relationship, sufficient
morphological differences exist between the two
that we here recognize them as separate. See
below for further rationale.
A long branch separating the monophyletic
Pacific clade from the Southeast Asian grade was
recovered in previous studies. In the current study,
clade 3 containing two unidentified/undescribed
species, are supported as sister to the Pacific clade
(100%BS; .99%PP), effectively dividing this
former long branch in half. Although this relation-
ship is supported in both bootstrap and Bayesian
support indices, the Pacific clade is only supported
in the Bayesian analysis (FIGURE 1). The southeast
Asian grade includes clades of Javan/Bornean,
Javan, Australia/Solomon Islands, Solomon Is-
lands, Philippines, and Javan/Sulawesian.
Fijian and Samoan geographic areas are poly-
phyletic within the Pacific clade (FIGURE 1). The
Marquesas species are divided along two major
supported clades but their relationship to one
another remains unresolved leaving the possibility
that Marquesan species of Cyrtandra might repre-
Selbyana selb-31-02-07.3d 3/12/13 12:49:19 170
170 SELBYANA Volume 31(2) 2013
Selbyana selb-31-02-07.3d 3/12/13 12:49:19 171
FIGURE 1. Maximum Likelihood phylogeny for the combined analysis of ITS, ETS and psbA-trnH genic regions.
Support indices are shown on supported nodes in the format bootstrap/posterior probability. Line weights correspond
with the branch length scale in the upper left; outgroup scaled for emphasis. Numbers 1–4 reference the clades
discussed in the text. Additional clades and grades referenced in the text are indicated by name. Numbers following
taxon names are those referenced in the text and in Table 1. A. 5Aeschynanthus;C. 5Cyrtandra.
CLARK ET AL.: CYRTANDRA OF THE SOLOMON ISLANDS 171
sent more than one introduction to that archipelago.
These Marquesan clades are grouped in a Bayesian-
supported clade with a supported clade of Samoan
species of Cyrtandra. These latter species represent
a unique clade in the Pacific characterized by
distinctive orange fruit, a trait which is not
represented in any species outside this clade in
the entire genus (Clark in prep).
Diagnostic Key to Phylogenetic Clades
The four monophyletic Solomon Islands clades
can be distinguished from one another based on a
suite of vegetative and reproductive characters
(FIGURE 2). It is important to note that this key is
meant to be diagnostic for the purposes of on-
going research on Cyrtandra in the Solomon
Islands and is not comprehensive nor is it
designed for species-level identification. A few
described specimens do not readily key out well
to any of the following four groups; details on
these are discussed in the annotated checklist that
follows. It is possible that these represent one or
more additional evolutionary lines of Cyrtandra
that occur in this biogeographically complex
region.
1. Leaves anisophyllous or pseudo-alternate,
having one leaf reduced to a bract-like
scale, sometimes alternately anisophyl-
lous/opposite (on the same plant), occa-
sionally strictly opposite.
2. Plants understory shrubs, weakly
branched, generally less than 3 meters
tall; inflorescences of one to few
flowers, often axillary, corollas white
to cream colored, never red. … Clade 1.
Examples in the phylogeny:
C. baileyi,C. erectiloba
2. Plants large, woody shrubs or trees,
often greater than 3 meters when
mature; inflorescences of numerous
flowers, variously cauliflorous or ax-
illary on secondary leafy branches,
corollas red, pink, occasionally strip-
ed or white (rarely greenish to yel-
Selbyana selb-31-02-07.3d 3/12/13 12:49:21 172
12
34
FIGURE 2. Example species photographed in the wild representing the four clades discussed in the text. Insets are
flower details for each species. Numbers correspond with the clades referenced in the text. 1. Cyrtandra erectiloba,
Isabel Island; 2. C. filibracteata, Kolombangara Island; 3. C. sp. ‘‘A,’’ Isabel Island; 4. C. fulvovillosa, Kolombangara
Island. All photographs by J.R. Clark.
172 SELBYANA Volume 31(2) 2013
lowish). ……… Clade 2. Examples in the
phylogeny: C. filibracteata, C. subulibractea
1. Leaves not anisophyllous or rarely so
being only weakly and never pseudo-
alternate, normally opposite or whorled.
3. Inflorescences compact and congested,
flowers one to many in dense, bracteate
cymes in the axils of leaves; leaves
opposite or often appearing whorled at
or near the stem apex, at times restricted
to the upper portion of fleshy (to woody)
stems. …… Clade 3. Examples in the
phylogeny: C. sp. ‘‘A,’’ C. sp. ‘‘B’’
3. Inflorescences spreading and open,
rarely if ever congested, flowers few
to many, in spreading, open cymes,
bracts if present, small and deciduous,
born in leaf axils along younger por-
tions of stem; leaves present along the
entirety of stems, excluding older
woody portions, often with consistently
asymmetrical or unequal bases. ……
Clade 4. Examples in the phylogeny:
C. fulvovillosa,C. macrotricha
All species treated in this study are summarized
in Table 2.
Annotated Checklist of Cyrtandra of the Solomon
Islands and Surrounding Regions
What follows is an account of known taxonomy,
disposition of type material, hypothesized rela-
tionships based on genetic and morphological data
(‘Solomon Islands Clade’’), listing of additional
specimens observed by one or more of the authors,
and anecdotal accounts on the distribution and
affinities of the 26 species in and/or near the
Solomon Islands.
Cyrtandra J.R.Forst. & G.Forst., Char. Gen. Pl. 3,
pl. 3. 1775. TYPE:Cyrtandra biflora J.R.Forst.
& G.Forst.
1. Cyrtandra atherocalyx G.W.Gillett, Kew Bull.
30: 393. 1975. TYPE: Bismark Archipelago––
New Ireland: Danfu River, 8 km upstream from
Danfu Bridge, near Manga, 220 m, 2 February
1970, Sands 771 (Holotype: K; Isotypes: E!, K,
LAE, UC).
Soft-stemmed shrub to 3 m, sap oxidizing to
orange (Clark 729), leaves opposite, often con-
gested; inflorescences bracteate, axillary, congest-
ed, calyces greenish, pubescent, corollas green-
ish-cream to yellow, actinomorphic, fruits white,
Selbyana selb-31-02-07.3d 3/12/13 12:49:39 173
TABLE 2. Species list for 26 taxa treated in the annotated checklist. Species are organized here according to
phylogenetic lineages as described in the text. Order in the checklist is alphabetical; number of each species as
treated in the text is listed here for convenience.
Species Solomon Islands clade Order in Checklist
Cyrtandra erectiloba G.W.Gillett 1 6
Cyrtandra wariana Schltr. 1 24
Cyrtandra filibracteata B.L.Burtt 2 7
Cyrtandra fuscovellea K.Schum. 2 10
Cyrtandra hentzelmaniana G.W.Gillett 2 11
Cyrtandra heteronema G.W.Gillett 2 13
Cyrtandra magentiflora G.W.Gillett 2 17
Cyrtandra purpurifolia G.W.Gillett 2 18
Cyrtandra subulibractea G.W.Gillett 2 20
Cyrtandra terraguilelmii K.Schum. 2 21
Cyrtandra cylindrocalyx G.W.Gillett 3 4
Cyrtandra laciniata G.W.Gillett 3 14
Cyrtandra sericifolia G.W.Gillett 3 19
Cyrtandra trichocalyx G.W.Gillett 3 22
Cyrtandra sp. ‘‘A’’ 3 25
Cyrtandra sp. ‘‘B’’ 3 26
Cyrtandra cominsii Hemsl., 4 2
Cyrtandra floribunda K.Schum. 4 8
Cyrtandra fulvovillosa Rech. 4 9
Cyrtandra herbacea G.W.Gillett 4 12
Cyrtandra macrotricha G.W.Gillett 4 16
Cyrtandra atherocalyx G.W.Gillett UNCLASSIFIED 1
Cyrtandra comocarpa G.W.Gillett UNCLASSIFIED 3
Cyrtandra dinocalyx G.W.Gillett UNCLASSIFIED 5
Cyrtandra lutescens G.W.Gillett UNCLASSIFIED 15
Cyrtandra valviloba G.W.Gillett UNCLASSIFIED 23
CLARK ET AL.: CYRTANDRA OF THE SOLOMON ISLANDS 173
subglobose. Bismarck Archipelago (Malakata, New
Ireland, New Britain), Solomon Islands (Choiseul,
Wagina, Kolombangara, New Georgia, Isabel,
Rendova, Guadalcanal, San Cristobal, Rennel).
Solomon Islands Clade UNCLASSIFIED.
Cyrtandra atherocalyx is characterized by
linear, acuminate inflorescence bracts (Gillett
1975) and is also unique in having nearly
cylindrical corollas that tend to be yellowish. In
preliminary phylogenetic analyses, a specimen
tentatively identified to this species (Clark 729)
was included (data not shown). This specimen was
resolved sister to Cyrtandra ferruginea Merr. from
the Philippines. If supported, this species repre-
sents a fifth clade in the archipelago. Due to an as
yet unidentified error resulting in an inordinately
long branch in the phylogeny, it was excluded
from final analysis for this paper. Additional
sequencing will be required to better ascertain
affinities between this species and others in the
region.
A peculiar feature of Cyrtandra atherocalyx is
the sap oxidizes to an orange color upon exposure
to air (J.R. Clark pers. obs.). This feature is
apparently shared by C. dinocalyx G.W.Gillett
from New Ireland (see that species, this paper) and
suggests a possible relationship between the two
species.
Other specimens examined: Solomon Islands
San Cristobal: Waiamura, 100 m, 11 August 1932,
Brass 2636 (BISH; note: Gillett (1975) has the
specimen listed as ‘‘8 November 1932’’ but as this
would post date the following specimen of a
higher collection number, we speculate this is in
error); Waiamura, 50 m, September 1932, Brass
3147(BISH). Isabel: Mt. Kubonitu, 11 May 2009,
Clark 729 (PTBG)
2. Cyrtandra cominsii Hemsl., in Kew Bull.
Misc. Inf. 1895:137. 1895. TYPE: Vanuatu
Santa Maria Island (Banks Islands), Comins 288
(Holotype: K!).
Shrub to 4 m, leaves opposite, bases unequal to
acute; inflorescences axillary, bracteate open
cymes, calyces greenish white with lobes divided
nearly to base, corollas white to cream to pale
yellow. Solomon Islands (Malaita, Rennell,
Ulawa, San Cristobal), Vanuatu (Banks Islands).
Solomon Islands Clade 4
Cyrtandra cominsii is widely dispersed and
variable across its range. This species and other
taxa in clade 4 are at least superficial similar to the
Cyrtandra samoensis complex of species (see
Clark et al. 2009) first hypothesized by Gillett
(1974, 1975) to be closely related species. This
complex includes a widespread assemblage of
morphologically similar but likely distantly related
species. Similarities in this group are here hypoth-
esized to be a result of conserved plesiomorphic
characters and not a close phylogenetic relationship.
As such, assignment of genetically unsampled taxa
to this clade is questionable at best until phyloge-
netics-based information becomes available.
In the original description, Hemsley (1895)
indicates that plants can have pseudoalternate
leaves. This has not been verified in any
specimens we have observed and was not
mentioned by Gillett (1973). This detail may have
been a result of erroneous observation made by the
Rev. R.B. Comins whom the species is named and
from whom herbarium material and associated
notes were received by Hemsley.
Gillett commented on the similarities and diffi-
culties in distinguishing between C. cominsii and C.
fulvovillosa Rech. (see the latter species, this paper),
particularly in the southeastern Solomon Islands.
Indeed, some specimens assigned by Gillett to this
species (e.g., Brass 2693) might be more appropri-
ately called C. fulvovillosa based on calyx shape and
division of calyx lobes (2/3 the length rather than
entirely divided as is typical for C. cominsii).
Other specimens examined: Bismarck Archipel-
agoNew Britain: Gasmata, 6 miles east of
Fullerborn Harbour, 300 m, 8 May 1973, Wo-
mersley 55342 (US; K). Solomon IslandsSan
Cristobal: Anganawai, 11 October 1968, Runikera
10881 (K); Balego, Nagonago, 100 m, 19 August
1932, Brass 2693 (A, BISH!). Malaita: near the
Mannu Anchorage, edge of beach rain-forest, in
coral sand, 25 August 1945, White 98 (K).
Rennell: West Lavangu Village, 21 May 1969,
Gafui 14973 (K). Ulawa: lowland rainforest, 4
October 1932, Brass 2944 (A, BISH, BM).
Vanikoro (Santa Cruz Group): upper Lawrence
River Valley, 28 April 1955, Hadley 75 (BISH).
3. Cyrtandra comocarpa G.W.Gillett, Kew Bull.
30: 387. 1975. TYPE: Bismarck Archipelago
New Britain: Bialla, 4–8 km inland, on vertical
face of gorge running through upland rain
forest, 120 m, 3 July 1973, Gillett 2555
(Holotype: K; Isotypes: A, E!, LAE, UC).
Shrub to 2 m, leaves opposite to anisophyllous;
cauliflorous, bracteate inflorescences, calyces
white, corollas green, fruits white. Known only
from the type locality. Solomon Islands Clade
UNCLASSIFIED
Material was not available at the time of this study.
4. Cyrtandra cylindrocalyx G.W.Gillett, Kew
Bull. 30: 394. 1975. TYPE: Solomon Islands
Bougainville Island: along trail between Arawa
and Panguna, near the summit ridge of the
Crown Prince Range, 200–300 m, 20 July
Selbyana selb-31-02-07.3d 3/12/13 12:49:39 174
174 SELBYANA Volume 31(2) 2013
1973, Gillett 2571 (Holotype: K; Isotypes: E!,
LAE, UC).
Virgate, pubescent shrub 1–2 m high; inflores-
cences axillary, conspicuous bracts, calyces light
green, corollas greenish white, fruit white, fleshy.
Known only from the type locality. Solomon
Islands Clade 3
This species demonstrates morphological simi-
larities with Cyrtandra hawaiensis C.B.Clarke
and other superficially similar species in having
bracteate inflorescences and leaf arrangement and
shape. However, a close relationship between
these species is not supported (FIGURE 1).
5. Cyrtandra dinocalyx G.W.Gillett, Kew Bull.
30: 375. 1975. TYPE: Bismarck Archipelago
New Ireland: 8 km west and upstream of Danfu
River Bridge, near Manga, 900 m, 14 February
1970, Sands 863 (Holotype: K; Isotypes: A,
CANB, E!, L, LAE, UC).
Small tree to 4 m high, sap oxidizing to orange,
leaves opposite; inflorescences axillary, bracts
present, often bilobed, calyces leathery, greenish-
cream to flushed pink, corollas funnelform, pale
lime-green, fruit ovoid, color not specified.
Restricted to New Ireland. Solomon Islands
Clade UNCLASSIFIED
Placement of this species is currently not
possible; Cyrtandra dinocalyx might represent an
additional clade, possibly inclusive of C. ather-
ocalyx. Gillett (1975) commented on similarities
between C. dinocalyx and several others west of
the Solomons in New Guinea. The relatively large,
coriaceous calyces are also superficially similar to
cyrtandras known from the Marquesas (Wagner et
al. submitted). Other features such as general habit,
leaf shape, and indumentum are suggestive of
characteristics representative of clade 3, this paper.
6. Cyrtandra erectiloba G.W.Gillett, Kew Bull.
30: 407. 1975. TYPE: Solomon Islands
Bougainville Island: near summit of Arawa-
Panguna highway, 975 m, 19 July 1973, Gillett
2569 (Holotype: K; Isotypes: A, BRI, E!, L,
LAE, UC). FIGURE 2.1.
Erect, branching shrub to 3 m, strongly
anisophyllous/pseudo-alternate leaves; calyces
corolloid, white, corolla white, stigmas bilobed.
Bismarck Archipelago (New Ireland), Solomon
Islands (Bougainville, Kolambangara, Isabel, Van-
gunu, San Cristobal). Solomon Islands Clade 1
This species is representative of a widely
dispersed group of species common throughout
New Guinea, Bismarck Archipelago and Solomon
Islands first recognized by Gillett (1975). Repre-
sentatives are also present in north Queensland,
Australia (C. baileyi; DD6 sampled in the
phylogeny) and Vanuatu (C. efaitensis Guill.;
sensu Gillett 1975). Cyrtandra wariana Schltr. is
scarcely distinguishable from C. erectiloba and is
principally differentiated in having a discoid rather
than bilobed stigma. Ranges for the two species
overlap in the Bismarck Archipelago.
Other specimens examined: Bismarck Archipel-
agoNew Ireland: Namatanai Sub-district, Hans
Meyer Range, Danfu river Valley, 2 February
1970, Sands 782 (BISH). Solomon Islands
Isabel: along trial to island summit, 1200 m, 13
May 2009, Clark 745 (E, K, PTBG, U.S.; 5
C0179 in phylogeny). Kolombangara: Forest
Management Area L5, interior to KFPL Field
Station along stream camp, 1000 m, 2 June 2009,
Clark 789 (E, K, PTBG, U.S.). Vangunu: inland
from Western coast logging camp, 600 m, 10 June
2009, Clark 801 (E, K, PTBG, U.S.).
Two additional specimens examined from the
southeastern Solomons are tentatively assigned to
this species: Solomon IslandsSan Cristobal:
S of Rarasi Village, 365–457 m, 10 March 1966,
Kondo Loc. 50 s.n. (BISH); Ngonangonamela,
above Archane, 1 May 1972, Powell BSIP 19447
(CSIRO, BISH).
Both of these specimens from San Cristobal are
distinctive in having opposite leaves and not
pseudoalternate as is the case in C. erectiloba.
The Kondo Loc. 50 s.n. specimen looks overall
similar to C. erectiloba, although the flowers are
rather different; neither the calyx nor the corolla
are urceolate and appear more elongate-bilabiate;
they are also larger (,2 cm) than is typical for the
species (,2 cm).
The latter specimen, Powell BSIP 19447, is even
more distinctive. The leaves are anisophyllous but
the smaller leaf of each pair is still substantial insize
and not at all trending towards ‘‘bract-like.’’ The
flowers are also larger than typical for C. erectiloba,
and are bilabiate, similar in many respects to
Cyrtandra efatensis From Vanuatu. Gillett (1974,
1975) hypothesized C. erectiloba, C. efatensis, and
C. wariana and other similar species represent a
complex throughout the region. Of these three
species, specimens observed to date are strictly
pseudoalternate. The San Cristobal specimens
might thus represent yet another unique member
of this complex worthy of species-level rank.
7. Cyrtandra filibracteata B.L.Burtt, Kew Bull.
Misc. Inf. 1936: 463. 1936. TYPE: Solomon
IslandsBougainville Island: Siwai, July 1930,
Waterhouse 132 (Holotype: K; Isotypes: A,
BISH!). FIGURE 2.2.
Tree or shrub to 6 m, opposite to anisophyllous
leaves; cauliflorous, bracts villiform, calyces
corolloid, green, white or pink, corollas pink to
Selbyana selb-31-02-07.3d 3/12/13 12:49:39 175
CLARK ET AL.: CYRTANDRA OF THE SOLOMON ISLANDS 175
red or striped red and white/pinkish. Solomon
Islands (Bougainville, Shortland Island, Mono
Island, Treasury Island, Choiseul, Baga Island,
Kolombangara, New Georgia, Isabel, Guadalca-
nal). Solomon Islands Clade 2
Widespread and conspicuous throughout its
range, this species is closely allied with Cyrtandra
subulibractea and other red-flowered species
found distributed broadly in the Bismarck Archi-
pelago and the Solomons. This species in partic-
ular shows characters suggestive of bird pollina-
tion including showy, red, tubular corollas, and
cauliflorous inflorescences. Honeyeaters in the
genus Myzomela (Meliphagidae) are suspected
pollinators (C. Filardi pers. comm.).
Other specimens examined: Solomon Islands
Bougainville: Korowai Hills, west of Toiumonapu
Plantation, 60 m, Van Royen N.G.F. 16375
(BISH). Kolombangara: NE Coast, Shoulder Hill
Area, 250–400 m, Regalado & Sirikolo 777
(BISH); KFPL Forest Management Area R2,
260 m, 26 May 2009, Clark 770 (E, K, PTBG,
U.S.; 5C0205 in phylogeny); KFPL Forest
Management Area L5 along trail to crater,
1000 m, 31 May 2009, Clark 782 (E, K, PTBG,
U.S.). Vangunu: upland from logging camp, 900 m,
13 June 2009, Clark 806 (E, K, PTBG, U.S.).
8. Cyrtandra floribunda K.Schum., Fl. Schutz-
geb. Su¨dsee: 377. 1905. TYPE: Papua New
GuineaNew Guinea: Bismarck Mtns.,
1200 m, January 1902, Schlechter 14070
[Holotype destroyed (?); Isotype WRSL!].
Cyrtandra rhynchotechoides Hatusima, Bot. Mag.
Tokyo 57: 121 (1943).
Shrub to 1 m, leaves opposite, bases strongly
unequal, inflorescences axillary, branched cyme of
many flowers (as many as 50), calyces corolloid,
funnelform, corollas white to greenish white,
funnelform. Bismarck Archipelago (New Guinea,
New Britain). Solomon Islands Clade 4
This species is conspicuous in its copious
number of flowers per inflorescence. Cyrtandra
floribunda overlaps in range with C. fulvovillosa
Rech. in the Bismarck Archipelago. If phyloge-
netic analysis supports inclusion in Clade 4 as here
hypothesized (which is nested within the Pacific
clade), C. floribunda would represent a verified
‘‘back’’ dispersal to New Guinea from the Pacific
islands.
Other specimens examined: Papua New Guin-
eaNew Guinea: MI. road to Mt. Suckling, 488 m,
Katik LAE 56326 (BISH). New Britain: East New
Britain, Gasmata Sub-district, 6 miles east of
Fullerborn Harbour, 300 m, Womersley et al. LAE
55342 (BISH).
9. Cyrtandra fulvovillosa Rech., Repert. Spec.
Nov. Regni Veg. 11: 185. 1912. TYPE: Papua
New GuineaNew Britain: in rain forest near
Toma, October 1905, Rechinger 4993 [Holo-
type W] FIGURE 2.4.
Cyrtandra malaccocaulos Schlechter, Bot. Jahrb.
Syst. 58: 313 (1923).
Herb to shrub, up to 3 m, leaves opposite, bases
unequal; inflorescences axillary, simple to several-
branching cymes, flowers 2–15 or more, calyces
deciduous, greenish-white, corollas white, funnel-
form, fruit white. Papua New Guinea (New
Guinea), Bismarck Archipelago (New Britain)
Solomon Islands (Treasury Island, Vella Lavella,
Ganongga, Gizo, Kolombangara, Vangunu, Tete-
pare, Guadalcanal). Solomon Islands Clade 4
A relatively common but variable species in the
Solomon Islands. Cyrtandra fulvovillosa overlaps
extensively with C. cominsii in habit and morphology
as well as range in the southeastern Solomon Islands.
Other specimens examined: Papua New Guin-
eaNew Guinea: Kani Mts., 700 m, 23 June
1908, Schlechter 17849 (WRSL, isotype of C.
malaccocaulos). Solomon IslandsKolomban-
gara: Forestry Division L5, interior of crater north
of field station, 600 m, 31 May 2009, Clark 786 (E,
K, PTBG, SEL, U.S.; 5C0183 in phylogeny). Gizo:
1 mile north of Gizo Town, Kondo 507(BISH).
Vangunu: inland from western coast logging camp,
1000 m, 13 June 2009, Clark 808 (PTBG).
10. Cyrtandra fuscovellea K. Schum., Fl.
Schutzgeb. Su¨dsee: 379. 1905. TYPE: Papua
New Guinea––New Guinea: Torricelli Mtns.,
600 m, April 1902, Schlechter 14552 [Holo-
type destroyed (?), Isotype: WRSL!].
Shrub to 1 m, leaves anisophyllous or pseudo-
alternate, inflorescences axillary, calyces green,
corollas red; fruit white. Papua New Guinea (New
Guinea), Bismarck Archipelago (New Britain).
Solomon Islands Clade 2
From observations of the type, this species
shares many traits with others in our clade 2
including the red corollas. The low-growing habit
is similar to species in clade 1 but might represent a
convergence in form rather than a shared, derived
trait. Its affinities and distribution in New Guinea
and the Bismarck Archipelago suggest more than a
superficial relationship among the red-flowered
cyrtandras.
11. Cyrtandra hentzelmaniana G.W.Gillett, Kew
Bull. 30: 403. 1975. TYPE: Solomon Islands
Bougainville: Panguna - Arawa Trail, 1 km
above Panguna, Crown Prince Range, 920 m,
20 July 1973, Gillett 2570 (K holotype, A,
BRI, E!, LAE, UC isotypes).
Selbyana selb-31-02-07.3d 3/12/13 12:49:39 176
176 SELBYANA Volume 31(2) 2013
Shrub or tree to 7 m, leaves pseudoalternate,
inflorescences axillary on woody branches, caly-
ces corolloid, red, pink or white; corollas red, pink
or white, fleshy; fruit white. Solomon Islands
(Bougainville, Baga Island, Kolombangara, Gua-
dalcanal). Solomon Islands Clade 2
As described by Gillett (1975), this broadly
circumscribed species appears to share many
characters with an assortment of species throughout
the Solomons, most similar to Cyrtandra hetero-
nema G.W.Gillett, described below, C. filibracteata
and C. subulibractea G.W.Gillett. However, the
isotype (E) observed differs dramatically in having
markedly hirsute corollas, similar to many species
of Cyrtandra found throughout Malesia. Many
species complexes are known to exist in the genus
and rather unsatisfactory taxonomies have been
proposed to account for these relationships (Burtt
2001). Depending on geographic structure in C.
heintzelmaniana, recognition of regional subspe-
cies or varieties might be warranted although such
structure might not be expected.
Other specimens examined: Solomon Islands
Bougainville: Koniguru, Buin, 900 m, 17 August
1930, Kajewski 2103 (BISH).
12. Cyrtandra herbacea G.W.Gillett, Kew Bull.
30: 382. 1975. TYPE: Solomon Islands
Bougainville: 6 mi (10 km) north of Buin,
145 m, 4 September 1964, Craven & Schodde
402 (Holotype: K; Isotype: LAE!).
Weak-stemmed herb to 0.5 m, leaves charac-
teristically pubescent; inflorescences pubescent,
6–40 flowers; calyces green; corollas white,
funnelform; fruit pale green. Solomon Islands
(Bougainville). Solomon Islands Clade 4
This species is perhaps the most vegetatively
specialized of the Solomon cyrtandras, exhibiting
a consistently herbaceous habit. Gillett based his
recognition of this species on this unique habit and
also on the smallish flowers that differentiate it,
although tenuously, from Cyrtandra floribunda.
13. Cyrtandra heteronema G.W.Gillett, Kew
Bull. 30: 402. 1975. TYPE: Solomon Is-
landsBougainville: Panguna-Arawa trail,
north of summit, ridge, Crown Prince Range,
near Kupei, 1000 m, 20 July 1973, Gillett 2578
(Holotype: K; Isotypes: BRI, E!, LAE, UC).
Shrub to 2 m, leaves opposite or pseudoalter-
nate, inflorescences axillary, calyces and corollas
both red; anthers conspicuously excerted; fruit
white. Solomon Islands (Bougainville). Solomon
Islands Clade 2
Gillett (1975) chose to recognize this species as
separate from Cyrtandra heintzelmaniana based
principally on stamen filament length and pres-
ence or absence of an indumentum on the inner
corolla tube (present on C. heteronema). However,
as mentioned under C. heintzelmaniana, a wide
range of variability exists. For example, Kajewski
1733 is most similar to C. heteronema, but lacks
significant pubescence on the inner calyx lobes.
Either inclusion of this species in the former is
warranted or, as suggested, further subdivision of
C. heintzelmaniana may be warranted to standard-
ize species concepts here.
Other specimens examined: Solomon Islands
Bougainville: Kupei Gold Field, 1000 m, 17 April
1930, Kajewski 1733 (BISH).
14. Cyrtandra laciniata G.W.Gillett, Kew Bull.
30: 388. 1975. TYPE: Bismarck Archipelago
New Britain: gorge below Yau Yau Village,
Nakanai Range, 750 m, 5 July 1973, Gillett
2559 (Holotype: K; Isotypes A, BRI, E!, LAE,
UC).
Shrub to 3.5 m, leaves opposite, fleshy,
inflorescences congested, bracteate, flowers often
concealed in decaying plant matter; plants highly
variable as circumscribed. Bismarck Archipelago
(New Ireland, New Britain), Solomon Islands
(Kolombangara, Isabel, Guadalcanal). Solomon
Islands Clade 3
Two species sampled in this study, labeled
Cyrtandra sp. ‘‘A’’ and C. sp. ‘‘B,’’ share many
characteristics with the description for C. laciniata
including common ranges. However, these differ
in significant ways as detailed under their respective
descriptions. It is possible that Gillett has defined a
continuum of species ranging from New Guinea to
the Southern Solomon Islands. Additional taxonomic
work will be required to resolve these relationships.
15. Cyrtandra lutescens G.W.Gillett, Kew Bull.
30: 386. 1975. TYPE: Bismarck Archipelago
New Britain: Mt. Bango, 275 m, 19 May 1968,
Henty & Lelean 29451 (Holotype: E!; Isotype:
LAE).
Shrub to 3 m, leaves opposite, bases unequal;
inflorescences axillary on older woody stems, on
stout peduncles, calyces deciduous, corollas white
to green, bilabiate. Bismarck Archipelago (New
Britain). Solomon Islands Clade UNCLASSI-
FIED
While many characters including vegetative traits
such as leaf shape and overall habit align this
species with Clade 4, its exclusive range in New
Britain, coupled with the stoutly peduncled, bracte-
ate inflorescences, suggest affinities to Clade 3.
16. Cyrtandra macrotricha G.W.Gillett, Kew
Bull. 30: 383. 1975. TYPE: Solomon Is-
landsGuadalcanal: Vulolo, Tutuve Mt.,
Selbyana selb-31-02-07.3d 3/12/13 12:49:39 177
CLARK ET AL.: CYRTANDRA OF THE SOLOMON ISLANDS 177
1000 m, 17 April 1931, Kajewski 2531
(Holotype: A; Isotypes: BM, G, P).
Herb to tree, smallish in stature, although up to
6 m according to Gillett (1975, and specimens
referenced therein), leaves opposite, bases unequal
to cordate; inflorescences axillary, branching
cymes of 25–30 flowers, calyces deciduous,
corollas funnelform, white to greenish white; fruit
a white berry. Solomon Islands (Guadalcanal,
Isabel, San Cristobal). Solomon Islands Clade 4
As one of the sampled specimens in our
phylogenetic analysis, this species is verified as
part of Clade 4. As such, a clade of no fewer than
two distinct Solomon Islands species is supported
as nested within the monophyletic Pacific clade
(BS 90%;PP.99%).
Other specimens examined: Solomon Islands
Isabel: inland from coast on trail to Mt. Kubonito,
near permanent camp near stream, 650 m, 14 May
2009, Clark 747 (PTBG). San Cristobal: Hinua-
haoro, 900 m, Brass 3033 (BISH).
17. Cyrtandra magentiflora G.W.Gillett, Kew
Bull., 30: 384. 1975. TYPE: Solomon Is-
landsBougainville: near Koniguru No. 1,
ca. 11 miles (18 km) north of Buin, 270 m, 20
July 1964, Craven & Schodde 50 (Holotype:
CANB; Isotypes: E!, LAE).
Tree to 9 m, leaves anisophyllous; inflorescenc-
es cauliflorous, bracts linear, calyces concealed,
corollas crimson-purple, bilabiate, fruits not
known. Solomon Islands (Bougainville). Solomon
Islands Clade 2
This species is most likely allied with Cyrtan-
dra filibracteata. Greater scrutiny of the type is
required to ascertain what characters can be used
to warrant continued recognition of this species.
Known only from the type specimens.
18. Cyrtandra purpurifolia G.W.Gillett, Kew
Bull. 30: 400. 1975. TYPE: Solomon Islands
Bougainville: Arawa trail, below Kupei, north
side of Crown Prince Range, 20 July 1973,
Gillett 2579 (Holotype: K; Isotypes: A, BRI,
E!, L, LAE, UC).
Shrub or branching tree to 2 m (6 m), leaves
pseudoalternate, distinctively purple underneath;
inflorescences axillary, bracteate with linear
bracts, flowers few, calyces small and deciduous,
corollas pink, fruit white. Solomon Islands (Bou-
gainville, Guadalcanal). Solomon Islands Clade 2
In many respects, this species is allied with both
clades 1 and 2. However, the colored corollas are
suggestive of the latter. The species is not always
distinguishable from Cyrtandra subulibractea
specimens aside from the distinctive purple
undersides of the leaves.
Other specimens examined: Solomon Islands.
Bougainville: Kupei goldmine, 900 m, 5 April
1930, Kajewski 1621 (BISH). Guadalcanal: Tutive
Mt., 1200 m, 13 May 1931, Kajewski 2659 (BISH).
19. Cyrtandra sericifolia G.W.Gillett, Kew Bull.
30: 392. 1975. TYPE: Solomon Islands
Bougainville: Summit ridge of Crown Prince
Range, between Arawa and Panguna above
Kupei, 1270 m, 20 July 1973, Gillett 2575
(Holotype K; Isotypes: E!, LAE, UC).
Shrub to 3 m, leaves opposite, glabrous at
maturity; inflorescences axillary, congested, up to
20 flowers, calyces and corollas tubular, the latter
cream white, fruit white. Solomon Islands (Bou-
gainville). Solomon Islands Clade 3
Based on specimens observed and Gillett’s
description, this species is aligned most likely
with clade 3. Apparently a widespread group in the
Solomons, species in this clade are superficially
similar to Cyrtandra hawaiensis C.B.Clarke from
Hawaii (Wagner et al. 1990) and several species
from Fiji (Smith, 1991, Clark et al., 2009). All
possess common features such as large leaf blades
often restricted to the stem apex and conspicuous-
ly bracteate inflorescences. Phylogenetic data,
however, suggest these species are not closely
related (FIGURE 1).
Other specimens examined: Solomon Islands
Bougainville: Koniguru, Buin, 950 m, 12 August
1930, Kajewski 2079 (BISH).
20. Cyrtandra subulibractea G.W.Gillett, Kew
Bull. 30: 409. 1975. TYPE: Solomon Islands
Bougainville: Crown Prince Range, Panguna,
Arawa trail, 1040 m, 20 July 1973, Gillett
2572 (Holotype: K; Isotypes: A, BRI, E!,
LAE, UC).
Shrub or tree to 5 m, gynodioecious, anisophyl-
lous, inflorescences axillary, 1–few flowers each,
calyces petaloid, green to red, corollas pale
yellow, white to pink, fruit white. Solomon Islands
(Bougainville, Kolombangara, Vangunu, Isabel,
Guadalcanal, Malaita, Rennell). Solomon Islands
Clade 2
Plants identifiable to this species were often
seen growing sympatrically with Cyrtandra fili-
bracteata on Kolombangara and Vangunu islands
(J.R. Clark pers. obs.) and, as noted earlier by
Gillett (1975), discerning between the two is
problematic. Never-the-less, the distinctiveness of
the two, when taken in context of ‘‘typical’’ for
each species, combined with more rigorous multi-
island sampling and phylogenetic analysis (J.R.
Clark current research; data not shown), support
the recognition of the two species as distinct. The
occurrence of gynodioecy, while uncommon in
Selbyana selb-31-02-07.3d 3/12/13 12:49:40 178
178 SELBYANA Volume 31(2) 2013
Cyrtandra, is not unheard of; Cyrtandra longifolia
(Wawra) Hillebrand ex C.B.Clarke has been
documented as such on Kauai, Hawaii (Wagner
et al. in prep.).
Several specimens observed including Kajewski
1742 and Kajewski 2711 were identified as having
white corollas. The variability in corolla color is
noteworthy in this species although geographical
structuring in this variation is not currently known.
This, coupled with variability in vegetative
characters warrants further scrutiny, particularly
in how they overlap with taxa considered part of
clade 1 which are exclusively white-flowered.
Other specimens examined: Solomon Islands
Bougainville: Kupei Gold Field, 1000 m, 18 April
1930, Kajewski 1742 (BISH); Kugumaru, Buin,
150 m, 7 June 1930, Kajewski 1854 (BISH).
Kolombangara: KFPL Forestry Area L5, interior
of crater from Field Station, 1000 m, 31 May
2009, Clark 788 (E, K, PTBG, U.S.; 5C0184 in
phylogeny). Vangunu: near Mbope, 500 m, 12
June 2009, Clark 805 (K, PTBG; 5C0191 in
phylogeny). Isabel: Tiratona (sic.), 170 m, 29
December 1932, Brass 3398 (BISH). Guadalcanal:
Sorvorhio Basin, 185 m, 23 January 1932,
Kajewski 2711 (BISH).
21. Cyrtandra terraeguilelmi K.Schum., Bot.
Jahrb. Syst. 9: 217. 1888. TYPE: Bismarck
ArchipelagoNew Guinea: Sattleberg, near
Finschhafen, July 1886, Hollrung 203 [Holo-
type destroyed (?); Isotype: WRSL!].
Shrub to 2.5 m, pseudoalternate leaves; inflo-
rescences axillary, bracteate, calyces fleshy, cam-
panulate, pink or red, corolla red, fleshy, densely
pubescent, fruit white. Papua New Guinea (New
Guinea), Bismarck Archipelago (New Britain).
Solomon Islands Clade 2
This species, restricted to New Guinea and New
Britain, shares many features in common with the
Solomon Islands species Cyrtandra filibracteata
and C. subulibractea, both of which were sampled
in this molecular analysis. Taxonomic revision in
this clade will require a thorough sampling of New
Guinea species; Bismarck Archipelago and Solo-
mon Islands members of this group are possibly
not restricted to these archipelagos and conceiv-
ably extend into New Guinea. Note: this species
was listed as ‘‘Cyrtandra terra-guilelmii’’ in
Gillett’s 1974 treatment.
Other specimens examined: Bismarck Archipel-
agoNew Britain: West Nakanai, Lairu Cr., near
Gule, Cape Hoskins, 23 August 1954, Floyd 6596
(BISH).
22. Cyrtandra trichocalyx G.W.Gillett, Kew
Bull., 30: 396. 1975. TYPE: Bismarck Archi-
pelagoBougainville: Isina Village, 425 m,
1 February 1967, Lavarack & Ridsdale
31280-A (Holotype: E!; Isotype: LAE).
Herbaceous shrub to 3 m, leaves opposite, bases
rounded to acute, inflorescences axillary, congest-
ed, calyces greenish, corollas white, fruit greenish
white to yellowish. Bismarck Archipelago (New
Britain) and Solomon Islands (Bougainville).
Solomon Islands Clade 3
The distinctive congested, bracteate inflores-
cences align this species with clade 3.
23. Cyrtandra valviloba G.W.Gillett, Kew Bull.
30: 384. 1975. TYPE: Bismarck Archipelago
New Britain: NNE. slope Mt. Ulawon, 400 m,
20 February 1971, Lelean & Stevens 51236
(Holotype: K; Isotype: LAE).
Branching shrub to 3 m, woody, leaves opposite,
cauliflorous, bracteate, bracts large, nearly as wide
as long, calyces cylindrical, corollas pale green,
nearly actinomorphic, fruit white, ovoid. Bismarck
Archipelago (New Britain, New Ireland). Solomon
Islands UNCLASSIFIED
Material was not available for study of this
species.
24. Cyrtandra wariana Schltr., Bot. Jahrb. Syst.
58: 360. 1923. TYPE: Papua New Guinea
New Guinea: near Udu, on the Waria River
(near Morobe), 300 m, 12 March 1908,
Schlechter 17448 [Holotype destroyed (?);
Isotypes: P, WRSL!].
Branching shrub to 3 m, woody, leaves
pseudoalternate, rarely anisophyllous; inflores-
cences axillary, bracteate, bracts small and
inconspicuous, calyces urceolate and inflated,
corolloid, corollas white to pale green, stigma
discoid, fruit white. Papua New Guinea (New
Guinea), Bismarck Archipelago (New Britain,
New Ireland). Solomon Islands Clade 1
Cyrtandra wariana shares many features in
common with C. erectiloba, as mentioned previ-
ously. Their ranges overlap in the Bismark
Archipelago and intact flowers are required to
discern between the two. The feature of discoid
stigma in this species vs. bilobed in C. erectiloba
warrants additional study.
Additional taxa sampled and identified here, but
not currently described:
25. Cyrtandra sp. ‘‘A’’ FIGURE 2.3.
Sparsely branching shrub or small tree to 3 m,
often monopodial with leaves restricted to the
upper half or near the apex, leaves opposite, often
congested near the apex and appearing whorled,
petioles up to 70 cm, distinctively white to
Selbyana selb-31-02-07.3d 3/12/13 12:49:40 179
CLARK ET AL.: CYRTANDRA OF THE SOLOMON ISLANDS 179
whitish-green, blades elliptic to acuminate, to
50 cm long, inflorescences axillary in bracteate,
congested cymes with stout pedicels, corollas five-
lobed, campanulate, fruit a white ellipsoid berry.
Solomon Islands Clade 3
This undescribed species appears to be related to
Cyrtandra laciniata but differs in its distinctive
leaves and petioles and characteristic monopodial
habit. It is relatively common throughout its range
and is therefore somewhat surprising that no
historical collections examined to date are of this
species. More field and herbarium studies are
warranted prior to formal description of this species.
Specimens examined: Solomon IslandsKo-
lombangara: KFPL Management Area R2, 260 m,
26 May 2009, Clark 772 (K, PTBG); KFPL Forest
Management Area L5, near camp along stream, 2
June 2009, Clark 790 (PTBG). Isabel: on trail to
Mt. Kubonitu, 800 m, Clark 722 (PTBG); Mt.
Kubonitu, 800 m, 14 May 2009, Clark 724 (PTBG
5C0173 in phylogeny).
26. Cyrtandra sp. ‘‘B’’ (not shown)
Sparsely branching shrub, leaves opposite,
hirsute, inflorescences in congested bracteate
clusters in the leaf axils, bracts distinctively linear,
long acuminate, flowers white, campanulate, fruit
a white berry. Solomon Islands Clade 3
Cyrtandra sp. ‘‘B’’ appears to be of an
undescribed taxon similar to but differing from
C. laciniata and C. sp. ‘‘A’’ in its uniformly
linear, acuminate bracts. While no material
examined to date appears to be conspecific with
this species, a more exhaustive examination of
herbarium material is warranted. The specimen is
here referenced for purposes of documenting the
voucher for this study and as a foundation for
future work.
Specimens examined: Solomon IslandsKo-
lombangara: interior of crater north of Forest
Management Area L5, 31 May 2009, Clark 787
(PTBG 5C0209 in phylogeny).
DISCUSSION
With the addition of new samples from the
Solomon Islands, a greater resolution and knowl-
edge of distribution patterns has been gained. It
was first hypothesized by Clark et al. (2009) that
the Solomon Islands might harbor cyrtandras that
are intermediate between Malesian and Pacific
clades. Our results support this hypothesis with
clade 3 being resolved conclusively as sister to the
Pacific clade and thus separating the former long
branch between these as was resolved in all
previous studies (Cronk et al. 2005; Clark et al.
2008, 2009). The discovery of clade 4 being firmly
nested within the Pacific clade suggests a potential
‘‘back’’ dispersal ability in some groups where
clades from decidedly Pacific cyrtandras can and
do colonize and become established in older, more
established ecosystems. It is not conclusively
known if these clades will also be found in more
central continental origin areas like Borneo and
New Guinea, although species hypothesized to be
conspecific are known from there (Gillett 1975).
Further sampling and study in these areas is
greatly needed to better understand the biogeo-
graphic history of this genus.
The phylogenetic placement of four distinct
clades of Cyrtandra in the Solomon Islands is not
all that unexpected in the archipelago. Gillett
(1975) suggested as many as eight independent
introductions of Cyrtandra are found there. The
islands’ proximity to more mainland centers of
diversity for the genus, in particular New Guinea,
intuitively increases the probability of multiple
dispersal events (Carlquist 1981). As such, distinct
evolutionary lines are expected to occur here,
similar to other areas in the western range for the
genus (Atkins et al. 2001) and in contrast with
remote island areas such as Hawaii and possibly
the Marquesas (Wagner et al. 1990; Wagner et al.
submitted). Of the 26 taxa considered in this study,
two were assigned to clade 1, eight to clade 2, six
to clade 3, five to clade 4, and five remained
unclassified (Table 2).
Clade 1 and clade 2 in this study, while forming
a single well supported clade, are here treated as
two recognizable groups. Although species in both
clades are similar in many ways including a trend
towards anisophylly, separation along flower color
and hypothesized pollination syndrome is war-
ranted. Clade 1 appears to be a small sampling of a
much wider dispersed group of cyrtandras found
throughout the western range for the genus
including Borneo, New Guinea, the Philippines,
and the Solomon Islands (Gillett 1975, Burtt 2001,
J.R. Clark pers. obs.). Representatives of this clade
also appear to have potential to disperse to other
more remote regions including C. baileyi (sampled
in this study) from North Queensland, Australia
and C. efatensis from Vanuatu. The unique
morphology of specimens from San Cristobal
Island also suggests a propensity for regional
variation in this complex despite its overall
extensive range.
The red-flowered clade 2 is separated from
clade 1 and all other clades known to occur in the
Solomons and the Pacific by the red flowers and
apparent bird pollination (Cronk et al. 2005). This
evolutionary transition to ornithophily is not
common in Cyrtandra to begin with; while red-
pigmented flowers do occur in some species, the
vast majority are either white, cream or pale
Selbyana selb-31-02-07.3d 3/12/13 12:49:40 180
180 SELBYANA Volume 31(2) 2013
yellow (Burtt 2001). A parsimonious hypothesis
assumes a limited shift to this syndrome over time
(Barret 1998, Grant 1998) and further supports
the monophyly of this clade. Clade 1 is either the
sister clade of the ornithophilous clade 2, or,
perhaps more likely given the current limited
sampling, either could be a nested subclade within
a broadly dispersed larger clade. Regardless,
additional sampling and character state reconstruc-
tion analyses are warranted to better understand the
evolution of these two related groups of Cyrtandra
and this apparent shift to bird pollination.
Clade 3, here represented by two as yet
unidentified/undescribed species, is phylogeneti-
cally situated as sister to the entire Pacific clade of
Cyrtandra (FIGURE 1). In previous studies, a long
branch connected the southeast Asian grade of
species with the monophyletic Pacific clade (Clark
et al. 2008, 2009). Discovery of a clade that
effectively splits this branch in half further re-
fines our biogeographic knowledge of the east-
ward expansion of Cyrtandra into the Pacific.
Cyrtandra laciniata and other species from the
Bismarck and Solomon Islands archipelagos are here
hypothesized to belong in clade 3; all of these share
among other characters, congested, bracteate axillary
inflorescences. While these species are morphologi-
cally similar to other Pacific cyrtandras including C.
occulta from Fiji and C. hawaiensis from Hawaii,
phylogenetic data suggest no close relationship.
However, in the latter two mentioned species, Wagner
et al. (1990, 1999) hypothesized a relationship
between these based on said morphological charac-
teristics. This hypothesis is supported in our study. In
the analysis, C. occulta and a related Fijian species
form a clade that is resolved sister to the Hawaiian
clade (FIGURE 1). It remains unclear as to what
significance these shared, apparently pleisiomorphic,
traits, outside of the fleshy berries common to most all
Pacific cyrtandras, have, if anything, in regard to the
expansion of the genus throughout the Pacific.
Much like clade 3, clade 4 includes species that
share superficial similarities with taxa that are not
all that closely related to one another. Cyrtandra
fulvovillosa and C. macrotricha (clade 4, sampled
here) have been hypothesized to be part of the
Cyrtandra samoensis complex of species (sensu
Clark et al. 2009; see Gillett 1973, 1974).
However, clade 4 is actually resolved as sister to
the South Pacific clade, in which C. samoensis
A.Gray is firmly nested among other morpholog-
ically dissimilar species of Cyrtandra.
CONCLUSIONS AND FUTURE DIRECTIONS
Understanding the evolutionary and taxonomic
relationships in Cyrtandra are complicated by a
broad geographic range, large number of species,
and a recent radiation of these species in the
Pacific and possibly elsewhere, potentially con-
founded by hybridization (Atkins et al. 2001,
Cronk et al. 2005, Clark et al. 2009). In addition,
fresh or silica gel-dried plant material is usually
required for successful DNA extraction and
analysis thus requiring extensive fieldwork to
complete molecular studies (Clark et al. 2008).
Despite a concerted recent effort to sample
Cyrtandra across its range (principally by authors
in this paper), sampling is far from comprehensive
and many clades are only partially or not at all
sampled. Even when sampling is near complete,
species-level relationships cannot be resolved
using commonly used markers (Atkins et al.
2001; Bramley et al. 2004a, 2004b; Clark et al.
2008, 2009). A recent study suggests using next-
generation sequencing and a large number of genic
regions is also unsuccessful in finding molecular
solutions to resolving these relationships (Pillon
et al. 2013). To advance our understanding in
Cyrtandra it becomes currently necessary to use a
combination of molecular and morphological
approaches as applied in this study.
In the Solomons, we have chosen to use this
phylogenetically-informed assessment for defining
clades and subsequent morphological categoriza-
tion of genetically sampled and unsampled
species. This approach is warranted in that it has
been successfully applied in well-sampled areas
such as Hawaii to arrive at a more satisfactory
taxonomy than using morphology or genetics
alone (Wagner et al. in prep). The challenge
remains to accurately interpret synapomorphic
characters verses pleisiomorphic ones. It has been
shown that some clades, including the Cyrtandra
samoensis complex (sensu Clark et al. 2009), include
a disparate group of species that are only superfi-
cially similar. Clades 3 and 4 in the current study
both exemplify this phenomenon. Geographical
proximity of similar species must also be consid-
ered; based on our current understanding of species
in the Pacific, those taxa that share many
morphological similarities and occur within com-
mon or neighboring areas are more consistently
resolved as closely related. These caveats have been
used to interpret relationships in the Solomons
based on our current knowledge. As such, these
hypotheses form the basis for ongoing research in
the Pacific and serve as a proof of concept for
methodology described here. Future advances in
knowledge will require ongoing field sampling and
study combined with molecular and morphological
analysis to both establish and refine taxonomy (see
Atkins et al. 2013).
ACKNOWLEDGEMENTS
The kind peoples of the Solomon Islands and all
of Melanesia and Polynesia are graciously ac-
Selbyana selb-31-02-07.3d 3/12/13 12:49:40 181
CLARK ET AL.: CYRTANDRA OF THE SOLOMON ISLANDS 181
knowledged for their assistance, expertise and
hospitality. John J. Schenk and one anonymous
reviewer are acknowledged for expert reviews
which greatly improved this manuscript. Bruce K.
Holst diligently edited and formatted the taxo-
nomic treatment in this paper.
Funding: Institute for Museum and Library
Services Grant No. MA-06-09-0198-09 to JRC;
National Museum of Natural History, the Smith-
sonian; The National Tropical Botanical Garden
McBryde Research Endowment for Hawaiian and
Pacific Botany; a generous private donation from
Ann Vidaver to fund field work in the Solomon
Islands (The Ann Vidaver Expedition to the
Solomon Islands, 2009); additional private dona-
tions from Marjorie Schmiel and Jeanne Katzen-
stein. Assistance: The Ministry of Forestry and
Environment, Solomon Islands (Gordon Konair-
amo, Fred Pitisopa, Clement Tavie, Gideon Solo,
Edwin Koveke, Patteson Tofu, Peter Kosui,
Robert Olisae); Kolombangara Forest Products
Limited (KFPL; Simon LaGassicke, Vaeno Vig-
ulu, Derol Sikua, Tim Bula); The Ministry of
Natural Resources and Environment, The Govern-
ment of Samoa (Pati Liu, Toni Tipama’a); The
University of the South Pacific, Fiji (SPRH;
Marika Tuiwawa, Alivereti Naikatini, Senilolia
‘Fiona’ Tuiwawa, Isaac Rounds); National Trop-
ical Botanical Garden (PTBG; Tim Flynn, David
Lorence, Steve Perlman, Natalia Tangalin, Chip-
per Wichman, Ken Wood), Bernice P. Bishop
Museum (BISH; Napua Harbottle, Clyde Imada);
The National Museum of Natural History (US;
Rusty Russell, Larry Skog, Jun Wen); Virginia
Commonwealth University (Greg Plunkett [now at
NY]); U.S. National Parks Service, Haleakala NP,
Maui (Bill Haus, Patti Weigen), The Marie Selby
Botanical Gardens (Bruce K. Holst, Harry E.
Luther), Maui Land and Pineapple Co. (Randy
Bartlett, Hank Oppenheimer, [now Maui Nui
Coordinator, Hawaii Plant Extinction Prevention
Program]); The New England Tropical Conserva-
tory (Scott Hoover), Rancho Santa Ana Botanic
Garden (RSA; Lucinda McDade), the Royal
Botanic Gardens Edinburgh (E; David Middleton,
Toby Pennington); University of Vienna (WU;
Michael Kiehn); Australian Tropical Herbarium
(CNS; Mark Harrington). Additional acknowl-
edgements: Anisapi, Bruce Baldwin, Chris and
Wendy Booth, Chris Filardi, George Hadley,
Umamoa La’auoleola, Joel Lau, Keith and Lanu
Martin, Maria and Ron Rova, Art Whistler. We
would also like to thank herbaria world-wide for
the concerted effort to digitize herbarium collec-
tions for virtual use by the international botanical
community. This effort has provided unparalleled
access to the world’s botanical heritage while
simultaneously preserving it for generations to
come.
LITERATURE CITED
Atkins, H.J., G.L.C. Bramley, and J.R. Clark. 2013. The
taxonomy of Cyrtandra (Gesneriacaeae): current
knowledge, future directions. Pp. 157–165 in: J.R.
Clark, ed. Proceedings of the World Gesneriad
Research Conference 2010. Selbyana 31: 65–253.
Atkins, H., J. Preston, and Q.C.B. Cronk. 2001. A
molecular test of Huxley’s line: Cyrtandra (Gesner-
iaceae) in Borneo and the Philippines. Biol. J. Linn.
Soc. 72:143–159.
Backer, C.A. and R.C.B. Van Den Brink. 1965.
Gesneriaceae. In Flora of Java (Spermatophytes
only). N.V. P. Noordhoff, Groningen, The Nether-
lands. 518–534.
Balke, M., I. Ribera, L. Hendrich, M.A. Miller, K.
Sagata, A. Posman, A.P. Vogler, and R. Meirer.
2009. New Guinea highland origin of a widespread
arthropod supertramp. Proc. Royal Soci. B. 276:
2359–2367.
Barret, C.H. 1998. The reproductive biology and
genetics of island plants. In Evolution on Islands.
(Ed. P.R. Grant). Oxford University Press. Oxford.
18–34.
Bramley, G.L.C. 2005. Revision of Cyrtandra Section
Dissimiles (Gesneriaceae). Blumea 50: 163–189.
Bramley, G.L.C., R.T. Pennington, R. Zakaria, S.S.
Tjitrosoedirdjo, and Q.C.B. Cronk. 2004a. Assem-
bly of tropical plant diversity on a local scale:
Cyrtandra (Gesneriaceae) on Mount Kerinci, Su-
matra. Biol. J. Linn. Soc. 81: 49–62.
Bramley, G.L.C., A. Weber, and Q.C.B. Cronk. 2004b.
The Cyrtandra (Gesneriaceae) species of Peninsular
Malaysia and Singapore. Edinburgh J. Bot. 60:
331–360.
Bramley, G.L.C. and Q.C.B. Cronk. 2003. The Cyrtan-
dra (Gesneriaceae) species of Mount Kerinci,
Sumatra. Harvard Pap. Bot. 7: 407–421.
Burtt, B.L. 1990. New and little known species of
Cyrtandra from Malaysia. Edinb. J. Bot. 47: 201–
233.
———. 2001. A survey of the genus Cyrtandra
(Gesneriaceae). Phytomorphology. Golden Jubilee
Issue: 393–404.
Burtt, B.L. and H. Wiehler. 1995. Classification of the
family Gesneriaceae. Gesneriana 1: 1–4.
Carlquist, S. 1981. Chance dispersal. Am. Sci. 69: 509–
516.
Clark, J.R. 2008. Gesneriads at the edge of the world.
Gesneriads 58: 35–38.
Clark, J.R., R.H. Ree, M.E. Alfaro, M.G. King, W.L.
Wagner and E.H. Roalson. 2008. A comparative
study in ancestral range reconstruction methods:
retracing the uncertain histories of insular lineages.
Syst. Biol. 57: 693–707.
Clark, J.R., W.L. Wagner, and E.H. Roalson. 2009.
Patterns of diversification and ancestral range
reconstruction in the southeast Asian-Pacific angio-
sperm lineage Cyrtandra (Gesneriaceae). Mol.
Phylo. Evol. 53: 982–994.
Clark, J.R. In prep. The orange-fruited clade of
Cyrtandra (Gesneriaceae) in Samoa.
Clarke, C.B. 1883. Cyrtandreae (Gesneracearum tribus).
Monogr. phan. 5: 1–303.
Selbyana selb-31-02-07.3d 3/12/13 12:49:40 182
182 SELBYANA Volume 31(2) 2013
Cowie, R.H. and B.S. Holland. 2006. Dispersal is
fundamental to biogeography and the evolution of
biodiversity on oceanic islands. J. Biogeogr. 33:
193–198.
Cronk, Q.C.B., M. Kiehn, W.L. Wagner, and J.F. Smith.
2005. Evolution of Cyrtandra (Gesneriaceae) in the
Pacific Ocean: The origin of a supertramp clade.
Amer. J. Bot. 92: 1017–1024.
Edgar, R.C. 2004. MUSCLE: multiple sequence align-
ment with high accuracy and high throughput.
Nucleic Acids Res. 32: 1792–97.
Emerson, B.C. 2002. Evolution on oceanic islands:
molecular phylogenetic approaches to understand-
ing pattern process. Mol. Ecol. 11: 951–966.
Evenhuis, N.L. and D.L. Bickel. 2005. The NSF–Fiji
terrestrial arthropod survey: overview. Bishop
Museum Occasional Papers 82: 3–25.
Fosberg, F.R. and M. Sachet. 1981. Polynesian plant
studies 6–18. In Smithsonian Contributions to
Botany, No. 47. Smithsonian Institution Press.
Washington, DC.
Gillett, G.W. 1967. The genus Cyrtandra in Fiji. Contr.
U.S. Natl. Herb. 37: 107–159.
———. 1973. The Genus Cyrtandra (Gesneriaceae) in
the South Pacific. In University of California
Publications in Botany, Volume 66. University of
California Press, Berkeley.
———. 1974. New Records for Cyrtandra (Gesneria-
ceae) in the New Hebrides. Kew Bulletin 29: 699–
709.
———. 1975. Cyrtandra (Gesneriaceae) in the Bismarck
Archipelago and Solomon Islands. Kew Bulletin 30:
371–412.
Gouy M, S. Guindon, and O. Gascuel. 2010. SeaView
version 4: a multiplatform graphical user interface
for sequence alignment and phylogenetic tree
building. Mol. Biol. Evol. 27: 221–224.
Grant, P.R. 1998. Patterns in islands and microevolution.
Pp. 1–17 in P.R. Grant ed., Evolution on Islands.
Oxford University Press, Oxford.
Hemsley, H.B. 1895. CCCCLXI. – Flora of the Solomon
Islands. Bull. Misc. Inform. Kew: 132–139.
Hillebrand, W. 1888. Flora of the Hawaiian Islands:
a description of their phanerogams and vascular
cryptogams. Lubrecht & Cramer, Monticello, N.Y.
(Facsimile ed., 1981).
Hillis, D.M. and J.J. Bull. 1993. An empirical test of
bootstrapping as a method for assessing confidence
in phylogenetic analysis. Syst. Biol. 42: 182–192.
Hilliard, O.M., B.L. Burtt, and M.H. Bokhari. 2003.
Pleuroschisma, a new section of Cyrtandra (Ges-
neriaceae) from Borneo. Gard. Bull. Sing. 55: 35–
60.
Hopper, S.D. 2000. How well do phylogenetic studies
inform the conservation of Australian plants? Aust.
J. Bot. 48: 321–328.
Huelsenbeck, J.P. and F. Ronquist. 2001. MrBayes:
Bayesian inference of phylogeny. Bioinformatics
17: 754–755.
Kiehn, M. 2001. South Pacific and Hawaiian Cyrtandra:
molecular and micromorphological studies. Malay-
an Nat. J. 55: 21–27.
Krupnick, G.A. and W.J. Kress. 2005. Plant conserva-
tion: a natural history approach. University of
Chicago Press. Chicago.
Minin, V., Z. Abdo, P. Joyce, and J. Sullivan. 2003.
Performance based selection of likelihood models
for phylogeny estimation. Syst. Biol. 52: 674–683.
Mo¨ ller, M.J. and J.L. Clark. 2013. The state of molecular
phylogenetic work in the family Gesneriaceae: a
review. Pp. 95–125 in: J.R. Clark, ed. Proceedings
of the World Gesneriad Research Conference 2010.
Selbyana 31: 65–253.
Muller-Dombois, D. and F.R. Fosberg. 1998. Vegetation
of the tropical Pacific islands. Pp. 1–733 in M.M.
Caldwell et al, eds., Ecological Studies Vol. 132.
Springer. New York.
Pence, V.C. 2002. In vitro collecting – tool for wild and
endangered species conservation. Pp. 26–29 in V.C.
Pence, J.A. Sandoval, V.M. Villalobos, and F.
Engelmann, eds. In Vitro Collecting Techniques
for Germplasm Conservation. IPGRI Technical
Bulletin No. 7. International Plant Genetic Resourc-
es Institute, Rome, Italy.
Pillion, Y., J.B. Johansen, T. Sakishima, E.H. Roalson,
D.K. Price, and E.A. Stacy. 2013. Gene discordance
in phylogenomics of recent plant radiations, an
example from Hawaiian Cyrtandra (Gesneriaceae).
Mol. Phylo. Evol 69: 293–298.
Samuel, R., W. Pinsker, and M. Kiehn. 1997. Phylogeny
of some species of Cyrtandra (Gesneriaceae)
inferred from atpB/rbcL cpDNA intergene region.
Bot. Act. 110: 503–510.
Schlag-Edler, B. and M. Kiehn. 2001. Palynology of
South Pacific Cyrtandra (Gesneriaceae) with notes
on some Hawaiiian taxa. Grana 40: 192–196.
Skog, L.E. and J.K. Boggan. 2007. World Checklist of
Gesneriaceae. Washington, DC: Dept. of Botany,
Smithsonian Institution. http://botany.si.edu/Gesneriaceae/
Checklist
Smith, A.C. 1991. Flora Vitiensis Nova: A new flora of
Fiji, Vol. 5. National Tropical Botanical Garden,
Kauai, Hawaii.
Swofford, D.L. 2002. PAUP*: Phylogenetic analysis
using parsimony, (*and other methods) version
4.0b10. Sinauer Associates, Sunderland, Mass.
Takhtajan, A.L. 1969. Flowering plants: origin and
dispersal. Oliver and Boyd. Edinburgh.
———. 1986. Floristic regions of the world. (A. Cronquist,
ed.). University of California Press. Berkeley.
Wagner, W.L., D.R. Herbst, and S.H. Sohmer. 1990.
Gesneriaceae, Cyrtandra. Pp. 735–781 in S.W. Mill,
ed. Manual of the Flowering Plants of Hawai’i.
Volume 1. University of Hawaii Press, Honolulu.
———. 1999. Manual of the flowering plants of Hawaii,
USA. Revised edition. Bishop Museum, Honolulu.
Wagner, W.L., A.J. Wagner, and D.L. Lorence. Revision
of Cyrtandra (Gesneriaceae) in the Marquesas
Islands. Phytokeys: accepted.
Wagner, W.L., J.R. Clark, and E.H. Roalson. A synopsis
of Hawaiian Cyrtandra (Gesneriaceae). In prep.
Weber, A. 2004. Gesneriaceae. Pp. 63–158 in K.
Kubitzki ed. The families of flowering plants.
Vol. 7. Dicotyledons. Lamiales.
Weber, A., J.L. Clark, and M. Mo¨ller. 2013. A new formal
classification of Gesneriaceae. Pp. 68-94 in: J.R.
Clark, ed. Proceedings of the World Gesneriad
Research Conference 2010. Selbyana 31: 65–253.
Wiehler, H. 1983. A synopsis of Neotropical Gesner-
iaceae. Selbyana 6: 1–219.
Selbyana selb-31-02-07.3d 3/12/13 12:49:40 183
CLARK ET AL.: CYRTANDRA OF THE SOLOMON ISLANDS 183
... The genus Cyrtandra is the most taxonomically challenging in the Gesneriaceae due to its large number and high proportion of poorly known and undescribed species (Burtt 2001;Atkins et al. 2013;Clark et al. 2013). Atkins et al. (2013) estimated 800 species of Cyrtandra. ...
... Large genera can be systematically addressed by following a phylogenetically informed taxonomic approach on a region-by-region basis Clark et al. 2013). This has been applied effectively by Bramley (2005) in a revision of Cyrtandra section Dissimiles in Borneo. ...
... This has been applied effectively by Bramley (2005) in a revision of Cyrtandra section Dissimiles in Borneo. The approach involves a robustly sampled phylogenetic tree wherein monophyletic clades can be characterized morphologically by one or more salient characters Clark et al. 2013). Taxa with molecular data can be assigned to the clades while taxa lacking molecular data can be tentatively assigned based on morphological similarities. ...
Article
Full-text available
Cyrtandra argentii Olivar, H.J.Atkins & Muellner sp. nov., endemic to the Philippines and named after George Argent, is herein described and illustrated. Collections associated with this new species are often confused with three other species, namely C. ferruginea Merr., C. villosissima Merr., and C. hirtigera H.J.Atkins & Cronk. Distinguishing characters including keys, updated descriptions, distribution maps, and photos of live specimens are provided to aid identification of the four species. Following the International Union for the Conservation of Nature (IUCN) criteria, C. argentii sp. nov. is considered to be Near Threatened (NT) due to its distribution in a zone susceptible to anthropogenic pressure and the lack of any formal protection.
... A sister relationship between the single known Taiwanese species, C. umbellifera, and the Pacific clade was also inferred. More recently, several studies by Clark et al. (2008Clark et al. ( , 2009Clark et al. ( , 2013 greatly increased sampling to include 70 species and sequence data for the nrITS and nrETS regions, as well as the psbA-trnH cp spacer region. Using maximum likelihood and Bayesian inference, the combined results of these analyses support several major findings: 1) Southeast Asian species are situated at the base of the tree, forming a grade of several clades, 2) C. umbellifera from Taiwan is nested within a clade of species from Southeast Asia, 3) two species from the Solomon Islands are supported as sister to the monophyletic Pacific clade, 4) the Solomon Islands, Fiji, Samoa, and the Society Islands all host species that are placed in more than one clade; in contrast the Marquesas and Hawaiian lineages are monophyletic, and 5) a greater Fiji-Samoa region appears to have played a central role in the expansion of the genus across the Pacific. ...
... Species of Aeschynanthus L. and Agalmyla Blume were included as outgroups to root the phylogeny based on the relationships inferred in Roalson and Roberts (2016). Sampling within Cyrtandra built on the earlier work of Clark et al. (2008Clark et al. ( , 2009Clark et al. ( , 2013, adding 60 species from across Southeast Asia and the Pacific including Borneo, Sulawesi, Sumatra, Philippines, Papua New Guinea, Solomon Islands, Vanuatu, Loyalty Islands, Wallis and Futuna, Fiji, Samoa, Society Islands, Austral Islands, and the Hawaiian Islands (see Table 2 for number of species sampled by region). Several samples from the Solomon Islands, Vanuatu, Fiji, and the Society Islands that have not yet been identified to species, as well as five species from Fiji that are in the process of being described (Johnson, in prep), were included based on morphological and genetic distinctness. ...
... In both analyses the majority of the backbone as well as major clades were well supported (!80 BS and !0.95 PP; Fig. 1). Species from Borneo, Sumatra, Java, Sulawesi, the Philippines, Malaysia, Papua New Guinea, Bismarck Archipelago, Solomon Islands, Vanuatu, and Australia are placed in several clades at the base of the tree (nodes 2-7; Fig. 1), and we follow Clark et al. (2013) in referring to this as the 'Southeast Asian grade.' The Southeast Asian grade includes the following: 1) a Borneo/Malaysia/Sumatra clade (node 14; 84 BS, 1.0 PP), 2) a Papua New Guinea/Solomon Islands/Vanu atu/Australia/Bismarck Archipelago clade (node 15; 100 BS; 1.0 PP), 3) a Philippines/Solomon Islands clade (node 16; 87 BS, 1.0 PP), 4) a Sulawesi/Philippines/Java clade (node 17; 100 BS, 1.0 PP), 5) a single species from Papua New Guinea (node 6; 92 BS; 1.0 PP, and 6) a Solomon Islands clade (node 18; 100 BS, 1.0 PP). ...
... As Bramley pointed out, however, the larger islands with species numbers in excess of 150 remain problematic. Clark et al. (2013) built on a regional approach and integrated phylogeny and morphology into a recent revision of the Solomon Islands by producing a phylogenetically informed revision of Cyrtandra from the islands arranged according to the four lineages present on the island. These lineages were revealed by their inclusion in a wider phylogeny of Cyrtandra in the Pacific. ...
... Crucially, integrating morphology also allowed those species from the Solomon Islands that had not yet been sequenced to be placed in the system. Clark et al. (2013) proposed that phylogenetically informed revisions of a geographic area are a proof of concept that might be useful for Cyrtandra more widely. The Solomon Islands, however, only has 24 species. ...
Article
Full-text available
Cyrtandra (Gesneriaceae), with over 800 species, is a mega-diverse genus which presents considerable taxonomic challenges due to its size. A well-sampled phylogeny of the genus across Southeast Asia has confirmed that all but one of the sections within Clarke’s 1883 genus-wide infrageneric classification are polyphyletic. It also shows that there are high levels of homoplasy in key morphological characters, although it is possible to use morphological characters to define clades in parts of the phylogenetic tree. There is some geographic structure in the phylogeny, but there is also evidence of dispersal between islands. A practical approach for tackling the taxonomy of Cyrtandra in the region, through phylogenetically informed taxonomic revisions of geographic areas, an approach which combines evidence from molecular, morphological and distribution data, is discussed. Completing our understanding of species diversity and delimitation in this genus will allow us to maximise the use of Cyrtandra as a tool for studying biogeography, speciation, diversification and conservation prioritisation in the rainforests of Southeast Asia.
... The taxa in this account are listed alphabetically and have not been assigned to the sections used by C. B. Clarke (Clarke, 1883). His sections were established in the nineteenth century and based on a fraction of the diversity of which we are now aware, and furthermore they are not widely used (Burtt, 2001;Bramley, 2003;Clark et al., 2013). ...
Article
Full-text available
Taxonomic evaluation of Cyrtandra (Gesneriaceae) from Sulawesi has resulted in the recognition of 39 species. Of these, four species are described as new here: Cyrtandra balgooyi H.J.Atkins & Karton., C. flavomaculata H.J.Atkins & Karton., C. longistamina H.J.Atkins & Karton. and C. parvicalyx H.J.Atkins & Karton. Eleven names are lectotypified and three names are placed in synonymy. Provisional conservation assessments suggest that seven of these species are Critically Endangered, six are Endangered, three are Vulnerable, two are Data Deficient and 21 are of Least Concern. Illustrations, distribution maps, and a key to the Cyrtandra species of Sulawesi are presented. Evaluasi taksonomi Cyrtandra (Gesneriaceae) dari Sulawesi menghasilkan pengenalan 39 jenis. Diantara jenis-jenis tersebut, empat jenis dideskripsikan sebagai jenis yang baru disini: Cyrtandra balgooyi H.J.Atkins & Karton., C. flavomaculata H.J.Atkins & Karton., C. longistamina H.J.Atkins & Karton. and C. parvicalyx H.J.Atkins & Karton. Sebelas nama telah dilektotipifikasikan dan tiga nama ditempatkan dalam sinonim. Penilaian konservasi sementara menyarankan bahwa tujuh dari jenisjenis ini dinyatakan Critically Endangered, enam dinyatakan Endangered, tiga dinyatakan Vulnerable, dua dinyatakan Data Deficient dan 21 sisanya dinyatakan Least Concern. Ilustrasi, peta distribusi dan kunci identifikasi untuk jenis-jenis Cyrtandra dari Sulawesi disajikan.
... We are not able to directly test this idea here given the sampling available, in part because to test for influence of geographic area on diversification using GeoSSE, there has to be at least one species that is found in both the focal area and outside the focal area. As we currently understand the diversity of Cyrtandra, only a single lineage has dispersed into the Pacific and there are no species that straddle this boundary (Atkins et al. 2001;Clark et al. 2008Clark et al. , 2009Clark et al. , 2013; however, the boundary areas separating Southeast Asia and the Pacific are the poorest explored for Cyrtandra diversity (Papua New Guinea, Solomon Islands, and vicinity). Ongoing and future work to better understand the geographical distribution of and diversification patterns of Cyrtandra may provide a compelling example of geography-driven increased diversification rates (H. ...
Article
Full-text available
Using a time-calibrated phylogenetic hypothesis including 768 Gesneriaceae species (out of ~ 3300 species) and more than 29,000 aligned bases from 26 gene regions, we test Gesneriaceae for diversification rate shifts and the possible proximal drivers of these shifts: geographic distributions, growth forms, and pollination syndromes. Bayesian Analysis of Macroevolutionary Mixtures analyses found five significant rate shifts in Beslerieae, core Nematanthus , core Columneinae, core Streptocarpus , and Pacific Cyrtandra . These rate shifts correspond with shifts in diversification rates, as inferred by Binary State Speciation and Extinction Model and Geographic State Speciation and Extinction model, associated with hummingbird pollination, epiphytism, unifoliate growth, and geographic area. Our results suggest that diversification processes are extremely variable across Gesneriaceae clades with different combinations of characters influencing diversification rates in different clades. Diversification patterns between New and Old World lineages show dramatic differences, suggesting that the processes of diversification in Gesneriaceae are very different in these two geographic regions.
Article
A taxonomic synopsis of Philippine Cyrtandra (Gesneriaceae) is presented. Following a study of 138 published names and their types, we accept 98 Cyrtandra species for the Philippine flora. Except for C. angularis, C. elatostemoides, and C. yaeyamae, all are endemic to the country. Lectotypes or neotypes are designated for all names for which this is necessary, except for six names for which we were unable to locate original material. We also validate a species name that was previously described without a Latin diagnosis (C. peninsula), synonymize three names, and provide taxonomic notes for each species. In addition, we propose two replacement names for taxa for which a legitimate name in Cyrtandra does not currently exist: C. edanoi for a Philippine species and C. siporensis for a Sumatran species. A look‐up table is provided to facilitate referencing of currently accepted names in Philippine Cyrtandra.
Article
Full-text available
A new formal classification of Gesneriaceae is proposed. It is the first detailed and overall classification of the family that is essentially based on molecular phylogenetic studies. Three subfamilies are recognized: Sanangoideae (monospecific with Sanango racemosum), Gesnerioideae and Didymocarpoideae. As to recent molecular data, Sanango/Sanangoideae (New World) is sister to Gesnerioideae + Didymocarpoideae. Its inclusion in the Gesneriaceae amends the traditional concept of the family and makes the family distinctly older. Subfam. Gesnerioideae (New World, if not stated otherwise with the tribes) is subdivided into five tribes: Titanotricheae (monospecific, East Asia), Napeantheae (monogeneric), Beslerieae (with two subtribes: Besleriinae and Anetanthinae), Coronanthereae (with three subtribes: Coronantherinae, Mitrariinae and Negriinae; southern hemisphere), and Gesnerieae [with five subtribes: Gesneriinae, Gloxiniinae, Columneinae (5the traditional Episcieae), Sphaerorrhizinae (5the traditional Sphaerorhizeae, monogeneric), and Ligeriinae (5the traditional Sinningieae)]. In the Didymocarpoideae (almost exclusively Old World, especially E and SE Asia/Malesia) two tribes are recognized: Epithemateae [with four small, but morphologically and genetically very distinctive subtribes: Loxotidinae (monogeneric with Rhynchoglossum), Monophyllaeinae, Loxoniinae and Epithematinae (monogeneric)] and Trichosporeae (the earliest name at tribal rank for the ‘‘Didymocarpoid Gesneriaceae’’). The last is subdivided into ten subtribes: Jerdoniinae (monospecific), Corallodiscinae (monogeneric), Tetraphyllinae (monogeneric), Leptoboeinae, Ramondinae (Europe), Litostigminae (monogeneric), Streptocarpinae (Africa and Madagascar), Didissandrinae, Loxocarpinae and Didymocarpinae. Didymocarpinae is the largest subtribe (ca. 30 genera and .1600 species) and still requires intensive study. It includes the most speciose genera such as Cyrtandra, Aeschynanthus, Agalmyla, Didymocarpus, Henckelia, Codonoboea, Oreocharis and Primulina and the types of the traditional tribes Didymocarpeae, Trichosporeae and Cyrtandreae.
Article
A taxonomic revision is presented for Cyrtandra in the Bismarck Archipelago and the Solomon Islands. A total of 24 species are recognized, including 17 new species, and their distributions are portrayed on four maps. Field and herbarium studies show that 8 evolutionary lines are represented in the region, 7 of these with strong affinities to New Guinea. The 7 lineages in the Bismarck Archipelago are all New Guinean, and five of these extend to the Solomon Islands. Four Solomon Islands lineages are common to Fiji.