Content uploaded by Marco F Duretto
Author content
All content in this area was uploaded by Marco F Duretto on Aug 06, 2020
Content may be subject to copyright.
Boronia (Rutaceae) is polyphyletic: Reinstating Cyanothamnus
and the problems associated with inappropriately defined outgroups
Marco F. Duretto,
1
Margaret M. Heslewood
1
& Michael J. Bayly
2
1National Herbarium of New South Wales, Royal Botanic Gardens & Domain Trust, Mrs Macquaries Rd, Sydney, New South Wales
2000, Australia
2School of BioSciences, The University of Melbourne, Victoria 3010, Australia
Address for correspondence: Marco F. Duretto marco.duretto@rbgsyd.nsw.gov.au
DOI https://doi.org/10.1002/tax.12242
Abstract The monophyly of Boronia (Rutaceae) was tested using 134 accessions of 120 species belonging to 39 genera from sub-
family Amyridoideae. Taxa included representatives of all eight sections of Boronia plus species of most genera in the two main
clades related to Boronia that had been identified by earlier studies. These samples included a good representation of genera from both
rainforest and sclerophyllous biomes. Maximum parsimony and Bayesian inference analyses were performed using three plastid
markers (psbA-trnH,trnL-trnF,rbcL) and two nuclear ribosomal markers (ITS, ETS). Separate analyses of plastid and nuclear
sequences using either maximum parsimony or Bayesian inference analyses recovered similar topologies. Apart from Boronia, the
broad generic relationships of previous analyses were largely supported. Boronia is polyphyletic with section Cyanothamnus being
more closely related to a large clade containing genera found in rainforest, including Melicope,Acronychia and their relatives. The
remaining seven sections of Boronia formed a strongly supported and isolated group. Boronia sensu stricto is sister to a clade contain-
ing the Cyanothamnus-Melicope-Acronychia clade plus a clade containing Euodia,Zieria and other small genera found in rainforest
or sclerophyllous communities. Issues with circumscriptions of ingroups and outgroups for previous analyses of Boronia and the
complex relationship between Australasian genera found in rainforest and sclerophyllous communities are both discussed. Cya-
nothamnus is reinstated at generic level. Appropriate nomenclatural changes are made to transfer all currently recognised series, spe-
cies, subspecies and varieties of Boronia sect. Cyanothamnus to the genus Cyanothamnus.
Keywords Australasia; biome shift; Malesia; molecular phylogeny; outgroup selection; polyphyletic genera
Supporting Information may be found online in the Supporting Information section at the end of the article.
■INTRODUCTION
Boronia Sm. (Rutaceae) is a genus of 157 species found in
Australia and New Caledonia (Duretto & al., 2013;
R.L. Barrett & al., 2015; Bayly & al., 2015). Most are shrubs,
though some are herbs, and a few can become small trees. Boro-
nia is easy to recognise: the plants are scleromorphic, have
opposite-decussate leaves (rarely in whorls of three), bisexual,
4-merous, diplostemonous flowers (5-merous in one subspe-
cies), free petals, carpels that are fused only along the style,
and explosively dehiscent fruits. These features are common
elsewhere in Rutaceae, though not usually in this combination,
and defining morphological apomorphies for the genus are
apparently lacking. The genus is classified into eight sections;
seven, Alatae Duretto, Algidae Duretto, Boronia,Cyanotham-
nus (Lindl.) F.Muell., Imbricatae Engl., Pedunculatae (Benth.)
Duretto & Bayly and Va l va t a e (Benth.) Engl., are endemic to
Australia, and one, Boronella (Baill.) Duretto & Bayly, is
confined to Grande Terre, New Caledonia (Bayly & al., 2015).
In recent phylogenetic studies based on molecular data,
Boronia has been shown to be taxonomically isolated and
not closely related to any other genus traditionally placed in
the tribe Boronieae (see further discussion on the tribe below).
Boronia is sister to a large clade of genera found mainly
in Australasia (Australia, New Guinea, New Caledonia,
New Zealand, and nearby islands) and Malesia (Indonesia,
Malaysia, Philippines) in both rainforest (e.g., Melicope J.R.
Forst. & G.Forst., Euodia Gaertn.) and sclerophyllous commu-
nities (viz. Neobyrnesia J.A.Armstr., Zieria Sm.) (see Groppo
& al., 2008, 2012; Bayly & al., 2013: fig. 3, clade D; Appelhans
&Wen,2020).Melicope is more widespread than other genera
in this clade and is found from Madagascar, through India and
Australasia, and to the islands of the eastern Pacific (Kubitzki
& al., 2011; Appelhans & al., 2014a,b, 2017, 2018). Ivodea
Capuron, from Madagascar and the Comoros Archipelago,
was thought to also belong in this clade (see Bayly
& al., 2013; Rabarimanarivo & al., 2015) though recent ana-
lyses using molecular data indicate it is more closely related
to the African and Malagasy genus Vepris Comm. ex A.Juss.
(Appelhans & Wen, 2020). The Boronia-Melicope clade is sis-
ter to another Australasian clade containing the remainder of
the generatraditionally placed in tribe Boronieae that are found
in sclerophyllous communities (e.g., Correa Andrews, Pheba-
lium Vent., Philotheca Rudge) plus genera found in rainforest
Article history: Received: 13 Jan 2019 | returned for (first) revision: 2 May 2019 | (last) revision received: 20 Jan 2020 | accepted: 21 Feb 2020
Associate Editor: Jessica M. Prebble | © 2020 International Association for Plant Taxonomy
Version of Record 481
TAXON 69 (3) •June 2020: 481–499 Duretto & al. •Boronia is polyphyletic
SYSTEMATICS AND PHYLOGENY
and/or the New Caledonian maquis (viz. Halfordia F.Muell.,
Myrtopsis O.Hoffm., Neoschmidia T.G.Hartley) (Groppo
& al., 2008, 2012; Bayly & al., 2013: clade C). These clades
(together = clade B of Bayly & al., 2013) are sister to an Aus-
tralasian clade containing Flindersia R.Br. and Geijera Schott
and their relatives, which are largely confined to rainforest
(Groppo & al., 2008, 2012; Bayly & al., 2013: clade E), as well
as Pitavia Molina from Chile, South America (Groppo
& al., 2012).
Past cladistic analyses involving Boronia fall into three
main groups: (1) subfamilial studies that included many gen-
era but understandably only a few representatives from each
genus (Scott & al., 2000; Groppo & al., 2008, 2012; Bayly
& al., 2013; Appelhans & al., 2014a, 2018; Appelhans
& Wen, 2020); (2) analyses of a good representation of the
genus Boronia but with a small number of outgroups
(Weston & al., 1984; Shan & al., 2006; Bayly & al., 2015;
Duretto & al., in prep.); or (3) studies investigating species
groups (Burgman, 1985) or a specific section of Boronia
(Duretto & Ladiges, 1999; Duretto, 2003).
In previous subfamilial studies, Boronia was represented
by only one species in Scott & al. (2000; from section Peduncu-
latae), Groppo & al. (2008, 2012; section Boronia), Appelhans
& al. (2014a, 2018; section Boronella), and Appelhans &
Wen (2020; section Boronia) and four species in Bayly & al.
(2013; two species from section Boronella, one species from
section Va l va t a e ,andB. scabra Lindl., which is currently placed
incertae sedis). These studies did not include a representative
sample of the diversity found in Boronia, with sections Alatae,
Algidae,Cyanothamnus and Imbricatae never being included
in a study of this type. An unsaid assumption was that all sec-
tions, plus the taxa placed incertae sedis, constitute a monophy-
letic group. Only Bayly & al. (2013) included representatives
from more than one section but still did not cover the phyloge-
netic diversity of the genus as all sampled species are placed in
the clade sister to section Cyanothamnus (see Weston
& al., 1984; Bayly & al., 2015). Boronia was not included in
other broader molecular phylogenies of Rutaceae (e.g., Chase
& al., 1999; Poon & al., 2007; Morton & Telmer, 2014). These
subfamilial and familial studies varied considerably in what
genera were included, especially those found in the
Australasian-Malesian region, with Bayly & al. (2013) having
the most complete sampling, followed by Groppo & al.
(2008, 2012).
In the generic cladistic analyses of Boronia, whether based
on morphological (Weston & al., 1984) or molecular data (Shan
& al., 2006; Bayly & al., 2015), an underlying assumption has
been that the Boronia-Boronella Baill. clade was monophyletic.
In these analyses a small number of outgroups were chosen to
root the tree: an outgroup that represented the “rest of Boro-
nieae”by Weston & al. (1984); Geleznowia verrucosa Turcz.
by Shan & al. (2006) as it was in the same tribe as Boronia;Neo-
byrnesia and Zieria by Bayly & al. (2015) as representatives of
the large clade sister to Boronia (Bayly & al., 2013).
In no analysis was both the full diversity of Boronia and
the full diversity of related genera sampled. In effect, the
monophyly of Boronia has not been tested by previous phylo-
genetic studies.
The composition of Boronia and its relationship to
other genera is nomenclaturally important as it is the type of
tribe Boronieae, which has been shown to be polyphyletic
(see Scott & al., 2000; Groppo & al., 2008, 2012; Bayly
& al., 2013; Appelhans & al., 2014a; Appelhans & Wen 2020).
Traditionally, Boronieae is placed in subfamily Rutoideae (see
Engler, 1931; Kubitzki & al., 2011), which has also been
shown to be polyphyletic and to be closely associated with
subfamilies Flindersioideae and Toddalioideae (Gadek
& al., 1996; Chase & al., 1999; Scott & al., 2000; Poon
& al., 2007; Groppo & al., 2008, 2012; Bayly & al., 2013;
Morton & Telmer, 2014; Appelhans & Wen 2020). The uncer-
tain status of these subfamilies and tribes was highlighted by
Kubitzki & al. (2011), who adopted an informal classification
for Rutoideae and used “alliances”such as the “Boronia alli-
ance”, acknowledging these were possibly unnatural groups
of uncertain rank.
Groppo & al. (2012) provided an excellent review of the
history of the subfamily classification of Rutaceae and dis-
cussed various options for revising the classification, discuss-
ing the need for more detailed analyses that include various
genera missing from published works. They opted for a con-
servative approach and proposed that Rutaceae have just the
two subfamilies, Cneoroideae and Rutoideae (which includes
Aurantioideae, Flindersioideae, traditional Rutoideae, and
Toddalioideae). Morton & Telmer (2014), who did not
include crucial genera (e.g., Cneoridium,Haplophyllum)in
their study and did not address the issues raised by Groppo
& al. (2012) regarding adopting a narrow circumscription of
Rutoideae, published a new subfamilial classification for
Rutaceae where subfamily Rutoideae was narrowly circum-
scribed and most genera of traditional Rutoideae, including
all Australasian-Malesian genera, were placed in subfamily
Amyridoideae. There were no tribal classifications or descrip-
tions in the classification systems described by Groppo
& al. (2012) or Morton & Telmer (2014). Additional work is
obviously required on the subfamilial classification of Ruta-
ceae especially in this group. For the sake of easy communica-
tion in this paper, we will adopt the concept of subfamily
Amyridoideae proposed by Morton & Telmer (2014) as it is
the last name used for this group. Recent molecular data
though clearly indicates that Boronia, and the remainder of
tribe Boronieae and many other genera found in rainforest,
belongs in a large predominantly Australasian-Malesian clade
that is sister to a clade containing Flindersia and Geijera and
their relatives (see Bayly & al., 2013: fig. 3, clades A and E,
respectively).
In this paper we aim to test the monophyly of Boronia
by including representatives of all sections and most series
of the genus, plus a large number of genera represent-
ing the Australasian-Malesian clades identified by Groppo
& al. (2008), Bayly & al. (2013: clade A), Appelhans & al.
(2014a, 2018) and Morton & Telmer (2014) in which Boronia
is placed.
482 Version of Record
Duretto & al. •Boronia is polyphyletic TAXON 69 (3) •June 2020: 481–499
The taxonomic issues in Boronia sect. Boronia sensu lato,
which includes the four species that are placed as incertae
sedis as identified by Bayly & al. (2015), are being addressed
in a parallel paper (Duretto & al., in prep.).
■MATERIALS AND METHODS
Taxon sampling. —This study is part of a broader project
that is investigating the systematics of a large clade of mainly
Australasian genera found in rainforest and sclerophyllous com-
munities that corresponds to clade A of Bayly & al. (2013). We
have constructed a database of c. 500 taxa as part of separate
investigations into a range of species-rich groups, e.g., Boronia,
the Phebalium-Philotheca group and major clades containing
genera found in rainforest. Preliminary analyses of this entire
dataset revealed the polyphyly of Boronia. The data presented
here are a subset of this dataset to address only the issues relevant
to the circumscription of the genus Boronia.Analysesofthe
remaining clades are forthcoming though some of the data
have been presented in the treatment of Acronychia J.R.Forst.
& G.Forst. by Holzmeyer & al. (2015).
Our dataset comprised 134 accessions of 120 species
belonging to 39 genera from subfamily Amyridoideae; most
are newly extracted specimens, and these are supplemented
with samples from previously published studies. The ingroup
corresponds to clade B and the outgroup to its sister clade,
clade E, of Bayly & al. (2013: fig. 3). A broad ingroup was
selected to ensure that we were able to test the monophyly of
Boronia with confidence. The taxa sampled are listed in Appen-
dix 1 along with their voucher details and GenBank accession
numbers for all sequences.
For the ingroup, 66 accessions of 58 species of Boronia
were sampled from all eight recognised sections (the numbers
in brackets in the following paragraphs indicate: the number
of samples/the number of species sampled/the number of spe-
cies in that taxon or placed incertae sedis): sect. Alatae
(1/1/1); sect. Algidae (2/2/3); sect. Boronella (3/2/5); sect. Bo-
ronia (20/20/43); sect. Cyanothamnus (22/15/23); sect. Im-
bricatae (1/1/1); sect. Pedunculatae (3/3/11); sect. Val v a t a e
(12/12/66); incertae sedis (2/2/4). Sampling included represen-
tatives of most of the five subsections and nine series of
section Valvatae, and all six series of section Cyanothamnus;
these two are the only sections of Boronia to have formal infra-
sectional classifications. Species selected for the remaining
sections were chosen to cover the morphological variation seen
in those sections. There are four species placed incertae sedis in
Boronia (i.e., not assigned to a section; Bayly & al., 2015) of
which two, B. inornata Turcz. and B. scabra, are included in
this analysis. The remaining two, B. ovata Lindl. and B. humi-
fusa Paul G.Wilson, are thought to be closely related to B. sca-
bra (Wilson, 1971, 1998; Bayly & al., 2015).
The ingroup also included 58 accessions of 52 species from
30 genera that are a representative sample of the genera found in
both rainforest and sclerophyllous communities from both clades
C(Phebalium-Philotheca group) and D (Boronia-Melicope
group) of Bayly & al. (2013: fig. 3, = clade B). Clade C is here
represented by Asterolasia F.Muell. (1/1/19), Correa (1/1/11),
Crowea Sm. (1/1/3), Diplolaena R.Br. (1/1/15), Eriostemon
Sm. (1/1/2), Halfordia (2/1/3), Leionema (F.Muell.) Paul G.Wil-
son (1/1/28), Myrtopsis (1/1/9), Nematolepis Turcz. (1/1/7),
Neoschmidia (1/1/2), Phebalium (1/1/28) and Philotheca
(2/2/53) plus three genera not included in Bayly & al. (2013),
Microcybe Turcz. (1/1/4), Muiriantha C.A.Gardner (1/1/1) and
Rhadinothamnus Paul G.Wilson (1/1/3). Chorilaena Endl.
(closely related to Rhadinothamnus and Nematolepis), Drum-
mondita Harv. and Geleznowia Turcz. (both closely related to
Philotheca) were included in Bayly & al. (2013) but not in this
analysis. Clade D of Bayly & al. (2013), in addition to Boronia,
is here represented by Acronychia (2/2/48), Brombya F.Muell.
(1/1/2), Comptonella Baker f. (3/2/8), Dutaillyea Baill. (1/1/2),
Euodia (4/4/6), Medicosma Hook.f. (4/4/25), Melicope (includ-
ing Platydesma H.Mann; 10/8/c. 235), Neobyrnesia (2/1/1),
Picrella Baill. (1/1/3), Pitaviaster T.G.Hartley (1/1/1), Sarcome-
licope Engl. (1/1/9), Zieria (7/7/62), as well as Maclurodendron
T.G.Hartley (1/1/6), Perryodendron T.G.Hartley (2/1/1) and
Tetra c t o m i a Hook.f. (1/1/6). The last three genera were not
includedinBayly&al.’s (2013) analysis but are part of this group
based on morphological (Hartley, 1979, 1997, 2001a,b; Bayly
& al., 2013) and molecular (Holzmeyer & al., 2015; Appelhans
& al., 2014a) evidence. The other genus that probably belongs
with the ingroup, Dutailliopsis T.G.Hartley (see Bayly & al.,
2013), was not included as material was unavailable.
The outgroup included single accessions of 10 species
from 8 of the 11 genera that make up the Flindersia-Geijera
clade (clade E of Bayly & al., 2013; see also Groppo
& al., 2012): Acradenia Kippist (1/1/2), Bosistoa F.Muell.
ex Benth. (1/1/4), Bouchardatia Baill. (1/1/1), Coatesia
F.Muell. (1/1/1), Dinosperma T.G.Hartley (1/1/4), Geijera
(2/2/10), Flindersia (2/2/17) and Pentaceras Hook.f. (1/1/1).
The three genera not sampled are Crossosperma T.G.Hartley,
Lunasia Blanco and Pitavia.
DNA extraction, PCR, sequencing, alignment. —Leaf
samples were taken from frozen silica-dried specimens or
from herbarium sheets. The plant material was disrupted dry
in a TissueLyser II (Qiagen, Valencia, California, U.S.A.)
using tungsten beads and total genomic DNA extracted using
the Qiagen DNeasy Plant Mini Kit following the manufac-
turer’s instructions. Five DNA regions were sequenced: two
nuclear regions, the external (ETS) and internal transcribed
spacers (ITS) of the ribosomal DNA repeats; and three plastid
regions, the psbA-trnH intergenic spacer (psbA-trnH), the
trnL-trnF region (including the trnL intron and trnL-trnF
intergenic spacer) and for a subset of 29 taxa the rbcL gene.
Sequence data for the rbcL gene was available primarily for
outgroup taxa but also for a broad range of genera from the
ingroup. We included the rbcL data, though missing from
the majority of taxa, in the hope it would test support for the
major clades (see discussion on missing data and tree con-
struction in Johnson & al., 2011). The following primers were
used for PCR amplification and sequencing: ETS, myrtF
(Lucas & al., 2007) and ETS-18S (Wright & al., 2001); ITS,
Version of Record 483
TAXON 69 (3) •June 2020: 481–499 Duretto & al. •Boronia is polyphyletic
18SF and 26SR (Prince, 2010), or ITS5 and ITS4 (White
& al., 1990), the former primer pair proved less likely to
co-amplify fungal contaminants in extracts from herbarium
material; psbA-trnH, psbAF (Sang & al., 1997) and trnH2
(Tate & Simpson, 2003); trnL-trnF region, primers c and f
(Taberlet & al., 1991); rbcL, RUTrbcL1F and rbcL1343R
(Bayly & al., 2013).
All PCR reactions were carried out in 25 μl volumes con-
taining 200 μM of each primer, 200 μMofeachdNTP,
0.004% BSA, 2–2.5 mmol MgCl
2
and 1 U Ta q DNA polymer-
ase. ITS and trnL-trnF amplifications used Promega GoTaq
DNA Polymerase (Promega, Madison, Wisconsin, U.S.A.),
while amplifications for ETS, rbcL and psbA-trnH utilised
Immolase DNA polymerase (Bioline, Luckenwalde, Germany)
and a hot start PCR (with an initial cycle of 10 min at 95C).
PCR reactions were subjected to 40 cycles as follows: denatur-
ation for 30 s at 94C; annealing for 30 s at 50C–58C; and
extension for 1 min at 72C, with a final extension for 4 min
at 72C. The annealing temperature for ETS, psbA-trnH and
trnL-trnF was 53C, ITS (primers 18SF and 26SR) 58Cor
(primers ITS5 and ITS4) 55CandrbcL 50C. Double-
stranded PCR templates were purified, and sequencing per-
formed by Macrogen (Seoul, South Korea).
Consensus sequences were assembled using ABI software
Sequence Navigator v.1.0.1 and aligned by eye in PAUP*v.4.
b10 (Swofford, 2002). In aligning sequences, gaps were posi-
tioned to maximize conformity to known indel types such as
simple and inverted duplications of adjacent sequences
(Levinson & Gutman, 1987; Golenberg & al., 1993). Overlap-
ping indels of different lengths, and insertions of the same
length but bearing different relationships to surrounding
sequence, were treated as having independent origins, while
indels of the same length and position and showing minor dif-
ferences in nucleotide sequence were scored as the same state
(Simmons & Ochoterena, 2000). Potentially informative indels
were scored as additional presence/absence characters and
appended to the database. Gaps were treated as missing data
in the phylogenetic analyses. Coding sequences of the rbcL
gene were translated in MacClade v.4.08a (Maddison &
Maddison, 2000) to check for internal stop codons. The full
data matrix,including indel characters, is depositedin TreeBase
(Accession number 25687; see http://treebase.org/). Several
small regions that could not be unambiguously aligned and a
homoplastic inversion in psbA-trnH (highly incongruent with
other characters) were excluded from all analyses.
Phylogenetic analyses. —Separate analyses using max-
imum parsimony or Bayesian inference were run using either
individual loci, the concatenated chloroplast or nuclear loci
and, after assessment for congruence, the combined chloro-
plast and nuclear sequences.
Heuristic searches of the combined or partitioned datasets
were conducted in PAUP*v.4b10 (Swofford, 2002) using tree
bisection reconnection branch-swapping to recover all equally
most-parsimonious (MP) trees. One thousand replicates of
random taxon addition searching were conducted in order to
detect multiple islands of trees, with subsequent use of the
“condense”option to delete duplicate trees. Multistate charac-
ters were treated as polymorphisms and swapping performed
on best trees. When searching exhausted computer memory
(for chloroplast analyses), restricted searching was employed
saving only 100 trees per replicate. Branch supports were cal-
culated using jackknife (JK) rather than bootstrap resampling,
following the recommendations of Simmons & Freuden-
stein (2011). Jackknife analyses utilised faststep searching in
which each replicate was performed using random-sequence
addition and no branch swapping, 10,000 replicates and the
percentage of characters deleted in each replicate set at one-
third. Jackknife values >50% were interpreted as weak sup-
port for clades; >75%–89% moderate support; 90%–99%
strong support and 100% was considered robust.
The MP phylogenies generated were compared to those
obtained using the Markov chain Monte Carlo (MCMC)
method implemented in MrBayes v.3.2.2 (Ronquist & al.,
2012). The most appropriate nucleotide substitution models to
apply were determined using the Akaike’s information criterion
in MrModeltest v.2.3 (Nylander, 2004), with data partitioned
into the five regions indicated above and excluding the
appended scored indels. All regions fit general time-reversible
likelihood (GTR) substitution models (nst = 6), either with
gamma distribution of rate variation among sites (GTR + Γ
model; trnL-trnF) or also with a proportion of invariant sites
(GTR + Γ+ I model; ETS, ITS, psbA-trnH,rbcL).
Bayesian posterior probabilities (PP) were estimated
using three independent runs of 10 million generations using
four chains with tree sampling every 1000 generations. All
parameters were set to be unlinked and with rates variable
between partitions, with all other priors for the analysis set flat
(i.e., as Dirichlet priors). Runs were assessed as sufficient
when checked for convergence with Tracer v.1.6 (Rambaut
& al., 2014) and when the standard deviation of split frequen-
cies reached 0.001. Trees generated prior to the four Markov
chains reaching stationarity (burn-in ~25%) were discarded,
and remaining trees were used to construct a 50% majority-
rule consensus tree, with nodes assigned posterior probabili-
ties of 0.95–1.00 considered supported. Clades with 100%
JK and PP of 1.00 were considered fully supported. Bayesian
analyses were also conducted including indels from all regions
combined as an extra partition. For these analyses, the indels
were binary encoded, and we applied a default two-state Mar-
kov model with gamma distribution of rates and coding set to
variable (as there were no invariant sites). State freqpr was set
to fixed (empirical) to reflect only having two states. Inclusion
of indels resulted in moderate improvements in branch sup-
ports, so final analyses included them as additional characters.
■RESULTS
After exclusion of 299 bp of ambiguous sequence
regions, the analysed 134 accessions and 120 taxon dataset
comprised 5518 bp, including 1540 parsimony-informative
(PI; 193 being scored indels) and 598 variable but
484 Version of Record
Duretto & al. •Boronia is polyphyletic TAXON 69 (3) •June 2020: 481–499
parsimony-uninformative (PU) characters. The plastid portion
comprised 3908 bp, of which 744 were informative including
110 scored indels: psbA-trnH 341 PI (61 indels); rbcL:67PI
(no indels); trnL-trnF: 336 PI (49 indels). The nuclear portion
comprised 1610 bp, of which 796 were informative including
83 scored indels: ETS 388 PI (35 indels); ITS 408 PI
(48 indels). Separate analyses of the nuclear or chloroplast
sequences identified no phylogenetic conflict in branching
patterns amongst the major clades (suppl. Fig. S1, S2). Apart
from the placement of Tetractomia and the lack of support for
clade 1 in the both analyses of nuclear sequence data (see
below), no substantial changes in structure were seen, the only
differences were in support values for clades and resolution.
On this basis, our final analyses presented here represent the
combined molecular data. Likewise, parsimony and Bayesian
analyses showed a high level of congruence, and Figs. 1–3
illustrate both jackknife clade support values >50% and poste-
rior probabilities imposed on the major clades of the Bayesian
majority-rule consensus tree.
The analysis of the combined dataset produced 108 equally
most parsimonious trees in a single island of length 8312 steps
and consistency index (CI) 0.39. The low CI value is primarily
driven by the nuclear data. In single-locus analyses, consistency
indices for both nuclear loci were low (0.32 ETS,0.30 ITS) and
substantially higher in all plastid loci (0.53 psbA-trnH,0.66
trnL-trnF, 0.74 rbcL). Alignment of the nuclear loci was pre-
dominantly configured to deal with 1–2 base indels, with high
possibly saturating substitutions, while in the chloroplast loci
insertions were predominantly longer repeats of adjacent
sequence and ranged from 1 to 64 bp in length, with one dele-
tion of 460 bp. A 21-bp inversion in psbA-trnH was highly
homoplastic; that region was excluded from all analyses, and
presence/absence of the inversion was scored as an indel.
Analysis. —Our analyses recovered the main clades of
subfamily Amyridoideae identified by previous analyses,
e.g., Groppo & al. (2008, 2012), Bayly & al. (2013), Appel-
hans & al. (2014a), and Morton & Telmer (2014) (Fig. 1).
The backbone of the tree and all major lineages formed mod-
erately to fully supported clades. The ingroup clades 2 (the
Acronychia-Melicope-Boronia clade; includes clades 3–7),
3, 6 and 7, and the outgroup clade (the Flindersia-Geijera
clade) were fully supported (1.00 PP, 100% JK), clade 5 (the
Euodia-Zieria clade) was strongly supported (1.00 PP, 96%
JK), and clade 1 (the Phebalium-Philotheca clade; 1.00 PP,
89% JK) had moderate support. Each of the eight sections of
Boronia sensu Bayly & al. (2015) were also fully supported
or, for the monotypic sections, markedly divergent (Figs. 2, 3).
Boronia is demonstrably polyphyletic with section Cya-
nothamnus (clade 6) supported as a member of clade 4 (1.00
PP, 84% JK) and separated from a fully supported clade con-
taining all the remaining species of Boronia (clade 3). This
result was mirrored in the separate analyses of the nuclear and
chloroplast sequences (suppl. Figs. S1, S2). Within clade
4, section Cyanothamnus is placed in a moderately supported
clade where it is the unsupported sister (0.87 PP) to the Acrony-
chia-Melicope clade (clade 7). Clades 6 and 7 together are sister
to an unsupported clade containing Tetractomia and clade
5, which consists of the disparate elements including the Euo-
dia clade, Zieria,Neobyrnesia and Perryodendron.Clades
5–7withTetractomia make up clade 4, which is sister to Boro-
nia less section Cyanothamnus (clade 3).
The outgroup contains genera that are characterised by
(3–)4–5(–6)-merous flowers, and leaves that can be either
opposite-decussate or alternate. Unlike other genera in Aus-
tralasia, the association of 4-merous flowers with opposite-
decussate leaves does not always hold, e.g., Bosistoa is char-
acterised by 5-merous flowers and opposite-decussate leaves.
Most outgroup genera are restricted to rainforest, and for both
Flindersia and Geijera this is true for most species though a
few species of each are found in sclerophyllous communities.
Clade 1 (the Phebalium-Philotheca clade; 16 spp./15 gen-
era sampled to represent 203 spp./18 genera) contains genera
characterised by 5-merous flowers and alternate leaves. It
includes genera from the Australasian region that are found in
sclerophyllous communities (e.g., Phebalium,Philotheca); the
New Caledonian genera found in maquis (Myrtopsis,Neoschmi-
dia –which differ from other taxa in having 4- or 5-merous
flowers and opposite-decussate leaves); and the Australasian
Halfordia (Bayly & al., 2016; found in maquis, woodland and
rainforest). Also in this clade is the Australian genus Correa,
which is characterised by having 4-merous flowers and
opposite-decussate leaves and is found largely in sclerophyllous
communities. This clade corresponds to clade C of Bayly
& al. (2013) and is sister to clade 2. This clade was not supported
in analyses of nuclear sequence data alone where its components
formed a polytomy with clade 2 (suppl. Fig. S1).
Clade 2 (the Melicope-Cyanothamnus-Euodia-Boronia
clade) contains genera characterised by 4-merous flowers (one
subspecies in Boronia has 5-merous flowers) and opposite-
decussate, rarely whorled, leaves. Of its included clades, clade
3(Boronia less section Cyanothamnus) is sister to clade
4(theMelicope-Cyanothamnus-Euodia clade: includes Tet ra -
ctomia, and clades 5, 6, 7). Clades 6 and 7 areweakly supported
as sisters to each other. Bouchardatia and Dinosperma of the
outgroup and Correa of clade 1 are also characterised by having
4-merous flowers and opposite-decussate leaves.
Clade 3 (the Boronia sensu stricto clade; 43 spp. sampled
representing 134 spp.) contains only Boronia less section
Cyanothamnus (viz. 7 sections plus B. inornata and B. scabra)
(Fig. 2). Results are consistent with the classification of
both Boronella and Pedunculatae at sectional level and the
placement of both B. inornata and B. scabra as incertae sedis
(see Bayly & al., 2015). The structure of the main Boronia
clade is similar to that presented by Bayly & al. (2015), less
section Cyanothamnus, and will not be discussed further
here as it is the subject of another paper (Duretto & al.,
in prep.).
Clade 4 (the Melicope-Cyanothamnus-Euodia clade) cor-
responds to clade D of Bayly & al. (2013) and contains three
strongly supported clades (5–7) and Tetractomia (not included
in Bayly & al., 2013) (Fig. 1). The relationships between these
four clades ([Tetractomia, 5], [6, 7]) though resolved, are not
Version of Record 485
TAXON 69 (3) •June 2020: 481–499 Duretto & al. •Boronia is polyphyletic
Boronia sect. Cyanothamnus
Fig. 3
Boronia less Cyanothamnus
Fig. 2
Perryodendron parviflorum 1
Neobyrnesia suberosa 1
Zieria arborescens
Zieria oreocena
Zieria southwellii
Zieria veronicea
Zieria fraseri
Zieria minutiflora
Zieria smithii
Tetractomia tetrandra
Myrtopsis myrtoidea
Asterolasia drummondii
Correa baeuerlenii
Leionema ambiens
Diplolaena microcephala
Microcybe multiflora
Phebalium canaliculatum
Nematolepis phebalioides
Rhadinothamnus euphemiae
Crowea angustifolia
Eriostemon australasius
Muiriantha hassellii
Philotheca spicata
Philotheca virgata
Halfordia kendack
Neoschmidia pallida
Euodia pubifolia
0.02
Melicope vitiflora 1
Neobyrnesia suberosa 2
Perryodendron parviflorum 2
Melicope vitiflora 2
Euodia hylandii
Euodia hortensis
Euodia montana
Pitaviaster haplophyllus
Brombya platynema
Medicosma glandulosa
Medicosma fareana
Medicosma riparia
Medicosma cunninghamii
Sarcomelicope follicularis
Melicope ternata
Picrella ignambiensis
Dutaillyea sp.
Comptonella microcarpa
Comptonella sessilifoliola
Melicope obscura
Melicope elleryana
Melicope stellulata
Melicope spathulata
Melicope rostrata
Melicope inopinata
Maclurodendron sp.
Acronychia chooreechillum
Acronychia baeuerlenii
Coatesia paniculata
Geijera parviflora
Geijera salicifolia
Flindersia acuminata
Flindersia dissosperma
Dinosperma stipitatum
Pentaceras australe
Acradenia frankliniae
Bosistoa transversa
Bouchardatia neurococca
Clade 1
Clade 2
Clade 3
Clade 4
Clade 5
Clade 6
Clade 7
RF
SC
SC
SC
SC
SC
RF
RF
RF
RF
1/100
1/84
0.87/-
1/100
1/96
1/100
1/100
1/100
1/100
1/100
1/100
1/100
1/89
1/-
Outgroup
1/100 1/100
1/99
1/80
1/98
1/-
1/100
1/91
1/91
0.99/53
1/100
1/100
1/79
1/100
0.99/-
1/100
-/60
1/100
1/100
0.98/82
1/96
1/100
1/100
1/100
1/100
1/-
1/100
1/87
-/62
1/99
1/82
1/-
1/-
1/100
1/60
0.97/- 0.99/69
1/100
0.99/70
Fig. 1. Bayesian estimate of the phylogeny of Australasian members of subfamily Amyridoideae (Rutaceae) based on the combined analysis of two
nuclear and three plastid markers (ITS, ETS, psbA-trnH,trnL-trnF and rbcL) for 120 species in 39 genera. Values at nodes indicate posterior prob-
ability (PP) >0.95/jackknife (JK) clade support values >50%. Coding for habitat type: RF = rainforest, SC = sclerophyll. Tree structure for Boronia
sensu stricto (clade 3) given in Fig. 2, and for Boronia section Cyanothamnus (clade 6) given in Fig. 3.
486 Version of Record
Duretto & al. •Boronia is polyphyletic TAXON 69 (3) •June 2020: 481–499
strongly supported. Tetractomia was placed sister to the
Melicope-Acronychia clade (here clade 7) in Appelhans
& al. (2014a) with good support and in our full parsimony
analysis with 74% jackknife support (not shown) and in ana-
lyses of nuclear sequence data alone (1.00 PP, 94% JK; see
suppl. Fig. S1). Clade 4 contains Australasian-Malesian gen-
era found in rainforest, the widespread Melicope, and three
taxa found in sclerophyllous communities, Zieria,Neobyrne-
sia and Boronia sect. Cyanothamnus.
Clade 5 (the Euodia-Zieria clade; 16 spp./7 genera sam-
pled representing 74 spp./7 genera) is a tritomy of a poorly
supported clade containing the monotypic genera Neobyrne-
sia and Perryodendron (both with long branches), a monophy-
letic and fully supported Zieria, and another clade containing
Brombya,Pitaviaster,Melicope vitiflora (F.Muell.) T.G.Hart-
ley and a polyphyletic Euodia (Fig. 1). This last clade was not
retrieved in the parsimony analysis. Some confidence can be
placed on the isolated positions of both Neobyrnesia and
B. pancheri
0.007
1/100
B. parvifolia 1
B. parvifolia 2
B. scabra
B. crenulata
B. gracilipes
B. heterophylla
B. citriodora
B. pilosa
B. latipinna
B. galbraithiae
B. imlayensis
B. muelleri
B. microphylla
B. safrolifera
B. pinnata
B. thujona
B. floribunda
B. rivularis
B. deanei
B. serrulata
B. rhomboidea
B. falcifolia
B. filifolia
B. alulata
B. excelsa
B. repanda
B. bowmanii
B. squamipetala
B. lanceolata
B. granitica
B. ledifolia
B. keysii
B. rosmarinifolia
B. ternata
B. lanuginosa
B. cymosa
B. denticulata
B. spathulata
B. parviflora
B. alata
B. inornata
B. algida
B. edwardsii
sect. BORONELLA
sect. BORONIA sect. VALVATAE
sect. PEDUNCULATAE
sect. ALATAE
sect. ALGIDAE
sect. IMBRICATAE
insertae sedis
insertae sedis
1/100
1/100
1/100
1/100
0.9/68
1/100
1/100
1/100
0.88/-
1/-
1/- 1/100
0.93/64
1/100
1/82
1/80
1/95
0.99/-
1/98
1/89
1/81
1/87
1/80
1/98
1/98
1/100
1/100
0.84/83
0.99/72
0.98/50
1/100
1/100
1/99
1/97
1/100
CLADE 3
Fig. 2. Bayesian estimate of the phylogeny of Boronia less section Cyanothamnus (clade 3) based on the combined analysis two nuclear and three
plastid markers (ITS, ETS, psbA-trnH,trnL-trnF and rbcL). Values at nodes indicate posterior probability (PP)/jackknife (JK) clade support values
>50%. Sections of Boronia in upper case.
Version of Record 487
TAXON 69 (3) •June 2020: 481–499 Duretto & al. •Boronia is polyphyletic
Perryodendron as each is represented by two accessions.
Zieria and Neobyrnesia do not have a sister relationship as
expected (see Armstrong & Powell, 1980; Armstrong, 2002;
Bayly & al., 2013, 2015; R.A. Barrett & al. 2015, 2018), but
the placement of Neobyrnesia as sister to Perryodendron in
our analyses is unsupported (0.68 PP, <50% JK). This rela-
tionship between the genera of clade 5 has not been postulated
before now. Clade 5 is similar to the Euodia clade of Appel-
hans & al. (2014a) with the addition of Zieria and Neobyrne-
sia. In Bayly & al. (2013), clade 5 was not resolved with the
Euodia clade and a Zieria-Neobyrnesia clade being each part
of a tritomy with the Melicope-Acronychia clade. Melicope
vitiflora,Perryodendron and Tetractomia were absent from
Bayly & al.’s (2013) analysis.
Clade 6 (the Boronia sect. Cyanothamnus clade; 15 spp.
sampled representing 23 spp.) is a strongly supported clade con-
taining all sampled species of Boronia sect. Cyanothamnus
(Fig. 3). Of the six series of the section, the three represented
by single species, Coerulescentes Duretto (B. coerulescens
F.Muell.), Fabianoides Duretto (B. fabianoides (Diels) Paul
G.Wilson) and Polygalifoliae Duretto (B. polygalifolia Sm.),
all have long branch lengths suggesting substantial isolation.
Two of the three series represented by multiple taxa, Penicilla-
tae Duretto (B. penicillata Benth., B. baeckeacea F.Muell.)
and Defoliatae Duretto (B. tenuis (Lindl.) Benth., B. defoliata
F.Muell.), form strongly supported clades. Series Cyanotham-
nus (remaining eight species) is polyphyletic with respect to
series Polygalifoliae and series Penicillatae. Boronia ramosa
(Lindl.)Benth. is isolated and sister to the remainder of the Cya-
nothamnus-Polygalifoliae-Penicillatae clade. Boronia rigens
Cheel is sister to a clade containing B. anethifolia A.Cunn. ex
Endl. and B. occidentalis Duretto, and this clade is sister to
series Penicillatae. This last clade is sister to a clade containing
B. anemonifolia A.Cunn., B. nana Hook., B. inconspicua Benth.
ser. CYANOTHAMNUS
B. anemonifolia subsp. anemonifolia 1
B. anemonifolia subsp. anemonifolia 2
B. anemonifolia subsp. variabilis
B. inflexa subsp. inflexa
B. nana var. nana
B. nana var. hyssopifolia
B. inconspicua
B. anethifolia 1
B. anethifolia 2
B. occidentalis 2
B. occidentalis 1
B. rigens
B. baeckeacea subsp. patula
B. penicillata
B. polygalifolia 1
B. polygalifolia 2
B. ramosa
B. coerulescens subsp. coerulescens 1
B. coerulescens subsp. coerulescens 2
B. defoliata
B. tenuis
B. fabianoides subsp. rosea
ser. CYANOTHAMNUS
ser. CYANOTHAMNUS
ser. PENICILLATAE
ser. POLYGALIFOLIAE
ser. COERULESCENTES
ser. DEFOLIATAE
ser. FABIANOIDES
1/100
1/100
1/66
0.99/72
0.99/-
1/100
1/100
1/100
1/100
0.62/71
1/96 1/100
1/94
1/100
1/100
1/100
1/74
0.005
1/100
1/100
CLADE 6
Fig. 3. Bayesian estimate of the phylogeny of Boronia section Cyanothamnus (clade 6) based on the combined analysis of two nuclear and three
plastid markers (ITS, ETS, psbA-trnH,trnL-trnF and rbcL). Values at nodes indicate posterior probability (PP)/jackknife (JK) clade support values
>50%. Series of Boronia section Cyanothamnus in upper case.
488 Version of Record
Duretto & al. •Boronia is polyphyletic TAXON 69 (3) •June 2020: 481–499
and B. inflexa Duretto. The strong sister relationship between
B. tenuis and B. defoliata (both series Defoliatae) supports the
classification based on morphological data described by Duretto
(2013) and Duretto & al. (2013) in contradiction of the RAPD
data analyses of Shan & al. (2006) where B. tenuis was sister
to the remainder of Boronia.
Clade 7 (the Melicope-Acronychia clade; 19 spp./8 genera
sampled representing c. 340 spp./9 genera) is a clade of genera
found in rainforest and includes a polyphyletic Melicope,
Acronychia with Maclurodendron,Comptonella,Dutaillyea,
Medicosma,Picrella, and Sarcomelicope (Fig. 1). The inter-
nal structure of this clade is moderately to fully supported.
This clade matches the Acronychia-Melicope clade of Appel-
hans & al. (2014a), albeit with the removal of Tetractomia (see
clade 4 above), and an unlabelled clade within clade D of
Bayly & al. (2013).
■DISCUSSION
Outgroup and ingroup circumscription in earlier ana-
lyses of Boronia.—The analyses presented here do not dis-
agree with previous cladistic analyses of Boronia (Weston
& al., 1984; Shan & al., 2006; Bayly & al., 2015) in regard
to the placement of section Cyanothamnus topologically.
These previous analyses found section Cyanothamnus to be
sister to the remainder of the genus. None of these analyses
included a suitable range of related taxa for both the ingroup
or the outgroup. In effect, the trees were inappropriately
rooted as it was assumed Boronia (with Boronella Baill.,
and section Cyanothamnus) was monophyletic. For example,
in Bayly & al. (2015), given the results presented here, it
would have been more appropriate to root the tree so that
section Cyanothamnus was sister to the clade containing
Zieria and Neobyrnesia (the assumed outgroup) and this clade
sister to the remainder of Boronia including Boronella. The
outgroup in the analysis presented by Weston & al. (1984)
was a polyphyletic amalgamation of many genera from tribe
Boronieae (here: clades 5, in part, and 1) and again omitted
most of the relevant genera that are more closely related to
Boronia, that is Australasian-Malesian taxa found in rainfor-
est (here: clades 5, in part, 7, and Tetractomia). What occurred
in both these analyses was that the outgroup was too narrowly
defined, and that part of the ingroup was more closely related
to the outgroup (or parts of the outgroup) than it was to the
remainder of the ingroup. Shan & al.’s (2006) analysis used
an appropriate outgroup (from clade 1), but both the ingroup
and outgroup were severely under-represented with no taxa
from clades 5 or 7. Curiously, Cyanothamnus was paraphy-
letic in this analysis, with B. tenuis sister to the remainder of
Boronia, a result not corroborated here (see Results, Fig. 3).
The placement of Boronia sect. Cyanothamnus in the
analysis presented here was unexpected (from morpholog-
ical data and existing paradigms) and highlights the need
to include the full diversity of the taxa being studied, as well
as their relatives, and confirm the assumed ingroup is
monophyletic and so more closely related to themselves than
any are to the assumed outgroup (see, for example, discussion
in Wilberg, 2015 and references therein). The past errors are
not unreasonable given the knowledge and assumed relation-
ships at the time of those analyses. The analysis presented here
further highlights the critical need to test current classification
systems in Rutaceae using a very broad range of taxa (see also
Introduction).
Infrageneric classification of Boronia and the place-
ment of Cyanothamnus.—Boronia is demonstrably polyphy-
letic with section Cyanothamnus (clade 6) isolated from the
remaining seven sections of Boronia, which together form a
well-supported, monophyletic group (clade 3) (Figs. 1, 2). Cya-
nothamnus is embedded in a well-supported Australasian-
Malesian clade (clade 4) that contains mainly genera that are
found in the rainforest but also two genera from sclerophyllous
communities, Zieria and Neobyrnesia. These last two genera of
clade 4 are resolved as at least two independent lineages,
although relationships between them are not well supported.
The structure within Boronia sensu stricto, with seven
sections, is similar to that presented by Bayly & al. (2015),
which is not surprising as the same subset of taxa were
included (we have added only B. galbraithiae Albr.) (Fig. 2).
All seven sections are strongly supported, or, for the mono-
typic sections, markedly divergent, and the two taxa placed
incertae sedis by Bayly & al. (2015), B. scabra and B. inor-
nata, remain isolated. We will not discuss the classification
of these taxa in more detail here as the classification of Boro-
nia sect. Boronia sensu lato is the subject of another paper
(Duretto & al., in prep.).
Boronia, including section Cyanothamnus, cannot be
retained as a monophyletic genus for two compelling reasons.
Firstly, to do so would require all 17 genera in clade 2 to be
combined into one large genus (c. 580 spp.) that would present
a great diversity of morphological forms and would be impos-
sible to define. Clade 2 is united by the presence of opposite
leaves and 4-merous flowers though in Rutaceae neither con-
dition nor their combination is unique to the clade (see also
Results). Also, several species in the clade do not have both
traits. Secondly, Acronychia,Euodia and Melicope all nomen-
claturally predate Boronia by 23 years, and so to retain all spe-
cies of Boronia in the one genus would require transferring all
taxa of Boronia to one of these genera, probably Melicope.
Acronychia,Euodia and Melicope are all acknowledged not
to be monophyletic (see Harbaugh & al., 2009; Appelhans
& al., 2014a,b, 2017, 2018; Holzmeyer & al., 2015), and a
broader or narrower circumscription of one or all of these gen-
era has been discussed by those authors but only in the context
of genera found in rainforest. The results presented here con-
firm their findings that M. vitiflora should be removed from
Melicope and that further work is required on both the Euodia
and Acronychia-Melicope clades of Appelhans & al. (2014a)
to better define generic limits.
That said, the situation for Boronia is clear: a narrower
circumscription of the genus is required. For stable and sensi-
ble taxonomy, Cyanothamnus Lindl. will be reinstated as a
Version of Record 489
TAXON 69 (3) •June 2020: 481–499 Duretto & al. •Boronia is polyphyletic
genus, after 170 years of being placed in synonymy. All
accepted taxa in Boronia sect. Cyanothamnus are here trans-
ferred to this genus (see Taxonomy below). Cyanothamnus
has only been treated as generically distinct from Boronia by
Lindley (1839), who described the genus, and Bartling (1848).
All infrageneric classification systems published for Boronia
have included Cyanothamnus as an infrageneric taxon (see
review in Bayly & al., 2015). Cyanothamnus can be distin-
guished from Boronia by having petals that are usually prom-
inently glandular and with an inflexed tip (versus not
prominently glandular, a straight tip or with a subterminal
apiculum on the abaxial surface) and prominently rugose,
minutely verrucose seeds (versus smooth, except in B. cymosa
Endl., and often also minutely verrucose).
Boronia sensu stricto (less section Cyanothamnus). —
Boronia sensu stricto (called Boronia from here on) is relatively
easy to define: scleromorphic shrubs with opposite-decussate
leaves (rarely in whorls of three, section Boronella), bisexual,
diplostemonous, 4-merous flowers (1 subspecies 5-merous),
petals without an inflexed hook, and dehiscent fruits. Weston
& al. (1984) identified the lack of an inflexed hook on the petal
as a possible synapomorphy for Boronia less Cyanothamnus.
The inflexed hook is a widespread feature in Australasian gen-
era of subfamily Amyridoideae and may be an underlying plei-
siomorphy. In Boronia, the petal tip can be either straight
(sections Alatae,Algidae,Boronia [few species], Imbricatae,
Pedunculatae,Valvatae), or with a subterminal apiculum on
the abaxial surface (sections Boronella,Boronia sensu lato
[not all species]). The petal tip not being inflexed is not unique
to Boronia in subfamily Amyridoideae (Weston & al., 1984).
The petal tip in the Melicope-Cyanothamnus-Euodia-Boronia
clade (clade 2) has three character states: 1, an inflexed tip;
2, straight; and 3, with a subterminal apiculum on the abaxial
surface. Any of these states could theoretically have evolved
from any ofthe others, and sothe possible defining synapomor-
phy for Boronia, the lack of an inflexed tip, is actually two dif-
ferent character states that may or may not have each evolved
once and in sequence (e.g., inflexed to straight to having a sub-
terminal apiculum abaxially).
Boronia has strong support from molecular data, in fact it
is one of the better-supported clades in this analysis, and it is
interesting that morphologically it is very difficult to find
robust apomorphies for the genus, which is probably not sur-
prising given the large number of species. From an identifica-
tion point of view it is easy to define the genus on what
probably are plesiomorphic states. Rutaceae, despite being very
diverse, have very simple flowers, leaves and arrangements of
these organs, and not a great variety of fruit and seed types
within major clades. Most characters are highly homoplastic:
e.g., leaves being simple, unifoliolate, trifoliolate or pinnate;
petals being valvate or imbricate; the presence or absence of
stellate hairs. See discussions on fruit types in specific clades
by Bayly & al. (2013) and Holzmeyer & al. (2015), for
example.
Boronia is very diverse morphologically when compared
to other genera in Rutaceae, and a case could be put forward
to further divide the genus into several genera, based on the sec-
tions, that can, for the most part, be easily defined using mor-
phological apomorphies. Further work on the circumscription
and relationships of the sections, and the species currently
placed incertae sedis, is needed before that approach is adopted.
Infrageneric classification of the genus Cyanothamnus.
—[Note: series and species names used in Figures are those
used under Boronia, and those used in this section are those
used under Cyanothamnus.] The species sampled for Cya-
nothamnus in this study are representative of the diversity
found in the genus. All six series were sampled: series Cya-
nothamnus (8 of 12 spp. including both species found in
Western Australia, including the type, C. ramosus Lindl.),
series Polygalifolii (Duretto) Duretto & Heslewood (1 of 1),
series Coerulescentes (Duretto) Duretto & Heslewood (1 of
1), series Penicillati (Duretto) Duretto & Heslewood (2 of
3), series Fabianoides (Duretto) Duretto & Heslewood (1 of
2), and series Defoliati (Duretto) Duretto & Heslewood (2 of
4). Apart from series Cyanothamnus, all series were recovered
as monophyletic and strongly supported (Fig. 3; note series
names in the figure are those used under Boronia). Series Cya-
nothamnus is polyphyletic with series Polygalifolii and series
Penicillati embedded in it. The four species of series Cya-
nothamnus that were not included in this study are from inland
areas of subtropical and tropical Queensland, and we consider
them closely related to the widespread C. occidentalis
(Duretto) Duretto & Heslewood (see Duretto, 2003).
A number of options are available concerning the classifica-
tion of Cyanothamnus including: (1) retain the current classifi-
cation acknowledging the series that contains the type is
probably polyphyletic and requiring further research; (2) synony-
mise series Polygalifolii and series Penicillati under series Cya-
nothamnus; (3) divide series Cyanothamnus into three or four
series so series Polygalifolii and series Penicillati are retained;
(4) make series Cyanothamnus monotypic (C. ramosus)and
move the other 11 species into series Penicillati; or (5) abandon
the classification system described by Duretto (2013) and Dur-
etto & al. (2013) and not transfer the series to Cyanothamnus
and list them as synonyms of the genus instead.
Options 2 and 4 are not desirable as both series Polygalifolii
and series Penicillati are well-defined taxa, each with distinctive
morphological synapomorphies (see Duretto, 2013; Duretto
& al., 2013). Including these two series in series Cyanothamnus
would make that series very difficult to define morphologically,
as would expanding series Penicillati to include an additional
11 species from series Cyanothamnus.Option5wouldberetro-
grade as all but one of the series are strongly supported and so
should be recognised. Option 3, though ideal, requires signifi-
cant research that should include additional or more rapidly
evolving DNA regions than those used in this and previous stud-
ies, and would detract from the main purpose of this paper, test-
ing the monophyly of Boronia. To that end we choose the
conservative option 1 (see Taxonomy below).
Evolution of sclerophyllous and rainforest elements in
the Australian flora. —Bayly & al. (2013) discuss the diver-
gence between lineages of Australasian Rutaceae inhabiting
490 Version of Record
Duretto & al. •Boronia is polyphyletic TAXON 69 (3) •June 2020: 481–499
rainforest and sclerophyllous communities. They considered it
reasonable to assume that the Australasian Rutoideae (= Amyri-
doideae) were ancestrally denizens of rainforest. They identi-
fied three main groups from sclerophyllous communities: the
Philotheca-Phebalium clade, Boronia sensu stricto (they did
not include representatives of Cyanothamnus in their study)
and Zieria with Neobyrnesia. The results presented here
further complicate the situation by adding another large and
taxonomically isolated genus, Cyanothamnus, that is confined
to sclerophyllous communities. Also, the presumed close rela-
tionship between Zieria (62 spp.; eastern Australia, New
Caledonia) and Neobyrnesia (1 sp.; N Northern Territory) is
not recovered, and the relationships of these two genera with
the two lineages from clade 5 that are found in rainforest,
Perryodendron (1 sp.; Moluccas, New Guinea, New Britain)
and the Euodia clade (c. 13 spp.: E Australia, New Guinea,
and east to Samoa and Niue), are unresolved, with long branch
lengths suggesting extended periods of isolation.
Australasian Rutaceae are diverse in both rainforest and
sclerophyllous communities at the generic level, with most
genera endemic to Australia and/or New Caledonia. Most
genera are restricted to either rainforest or sclerophyllous
communities. Notable exceptions are Flindersia and Geijera
of the outgroup, where most species are restricted to rainfor-
ests though in each genus there are a small number of species
restricted to sclerophyllous communities. The same occurs for
some New Caledonian genera found mostly in rainforest (see
Bayly & al., 2013). Likewise, some genera found mainly in
sclerophyllous communities, e.g., Boronia,Correa,Nemato-
lepis and Zieria, have a few species found in wetter communi-
ties, usually wet sclerophyll or alpine habitats, though Correa
lawrenceana Hook., Nematolepis squamea (Labill.) Paul G.
Wilson and Zieria arborescens Sims can be found in or along
the edge of rainforest. Genera found in sclerophyllous com-
munities are almost exclusively Australian and tend to be
more species rich than rainforest genera (apart from the wide-
spread and species-rich Melicope-Acronychia clade) by an
order of magnitude.
The major clades, apart from clade 5, each contain taxa
almost exclusively restricted to one biome (Fig. 1): the out-
group (rainforest; exceptions in Flindersia and Giejera); the
Phebalium-Philotheca clade (clade 1; sclerophyllous commu-
nities, Halfordia also in rainforest; few species of Correa and
Nematolepis found in wetter communities); Boronia sensu
stricto (clade 3; sclerophyllous communities, few species
found in wetter communities); Tetractomia (rainforest); Euo-
dia-Zieria clade (clade 5: Zieria and Neobyrnesia, sclerophyl-
lous communities; remaining genera from rainforest);
Cyanothamnus (clade 6; sclerophyllous communities); and
the Melicope-Acronychia clade (clade 7: rainforest). Species
in sclerophyllous communities are usually shrubby, and those
in rainforests, including most of species from genera normally
found in sclerophyllous communities, are small to large trees.
In clade 2 there are four lineages found in sclerophyllous
communities (Boronia sensu stricto, Cyanothamnus,Zieria,
Neobyrnesia) and four that are confined to rainforests (the
Melicope-Acronychia clade, Tetractomia,theEuodia clade,
Perryodendron). This clade is sister to a clade largely found
in sclerophyllous communities (clade 1) and these sister to
the outgroup, which is largely confined to rainforest.
Biome transition/shift is a rare event (see Crisp & al.,
2009) and is usually discussed as a shift from wet warm com-
munities to dry and/or temperate communities (see discus-
sion in Donoghue & Edwards, 2014). In Australia it is well
documented that during the Cenozoic the continent dried, the
rainforests contracted dramatically and dry- and fire-adapted
communities diversified and dominated the landscape, with
many genera having significant radiations in this habitat (see
discussions in Crisp & al., 2004; Crayn & al., 2006; Bayly
& al., 2013; and references therein, for example). A notable,
and at the time unexpected, relatively recent discovery was
that the dry-adapted Tremandraceae were embedded in the
Elaeocarpaceae, which are largely confined to rainforest
(Crayn & al., 2006; and references cited therein). Here we pre-
sent another unexpected and independent rainforest/sclero-
phyll biome shift represented by the genus Cyanothamnus,
which is confined to scleromorphic communities, being sister,
though without strong support, to the large clade of genera
found in rainforests (clade 7).
Obviously, multiple biome shifts have occurred. When the
biomes (rainforest versus sclerophyllous communities) are
plotted on the tree in Fig. 1 and a biome shift is treated as a
character state change (changing from rainforest to sclerophyl-
lous or vice versa), it is equivocal what the plesiomorphic char-
acter state would be for the larger clades (e.g., the main clade,
clades 2 or 5, or the Cyanothamnus-Melicope clade). It is
equally parsimonious for biome shifts to have occurred in
either direction, or a combination thereof. For example, if we
assume the plesiomorphic state for the tree is occupying rain-
forest (the outgroup and branches further out from here [see
Bayly & al., 2013], are largely genera confined to rainforest)
then it is equally parsimonious for the plesiomorphic state for
clade 2 to be either rainforest or scleromorphic. There could
be two steps for rainforest to scleromorphic (to clade 1, and
to clade 3); or a single step to scleromorphic before clade
1 diverges andthen multiple reversalsback to rainforest further
up the tree. Results presented here indicate that in Australasia,
biome shifts between rainforest and sclerophyllous communi-
ties have occurred multiple times and not conclusively in one
direction, from rainforest to sclerophyllous communities. A
clearer picture of relationships (e.g., better molecular support
and resolution in the tree) would assist with understanding
the history of biome shifts and of morphological/ecological
evolution
■CONCLUSION
Boronia as it is currently circumscribed is polyphyletic,
and a narrower and monophyletic circumscription with seven
sections is presented here. Boronia sensu stricto is an isolated
genus taxonomically being sister to a large clade containing
Version of Record 491
TAXON 69 (3) •June 2020: 481–499 Duretto & al. •Boronia is polyphyletic
18 genera found in both rainforest and sclerophyllous commu-
nities. Cyanothamnus is more closely related to the Melicope-
Acronychia group of genera that are found in rainforests from
Madagascar to the islands of the eastern Pacific, though most
genera and species are confined to the Australasian-Malesian
region. The genus Cyanothamnus is reinstated and is endemic
to Australia. The evolution of rainforest and sclerophyllous
elements in Australasian Rutaceae is complicated with more
than a few lineages in each ecosystem, suggesting that habitat
shifts possibly occurred in both directions and certainly multi-
ple times.
■TAXONOMY
Generic descriptions are provided for Cyanothamnus and
a narrower circumscription of Boronia. For descriptions of all
other taxa see Duretto & al. (2013, and references therein).
Boronia Sm., Tracts Nat. Hist.: 288, t. 4–7. 1798 –Type
(designated by Wilson in Nuytsia 12: 121. 1998):
B. pinnata Sm.
Perennial herbs or shrubs, rarely small trees; glabrous or
with simple and/or stellate hairs. Leaves opposite, decussate,
rarely in whorls of 3 (see section Boronella), simple or im-
paripinnate. Flowers axillary or terminal, solitary or in cymes
or pseudo-umbels or panicles, bisexual, 4-merous, rarely
5-merous (B. scabra). Sepals free, open, imbricate or valvate,
persistent or caducous. Petals free, imbricate or valvate, not
obviously glandular; tip straight or with a subterminal apicu-
lum on the abaxial surface; 1- or 3-veined at base; caducous
or persistent. Stamens 8, rarely 4 of them caducous (B. parvi-
flora Sm.), 8 or 4 fertile (see section Boronia); filaments
inwardly curved, semiterete, glabrous or hairy, usually verru-
cose towards apex; anthers introrse, apiculate or not, con-
nective usually inconspicuous or cream coloured. Disc
prominent, usually entire, rarely lobed (see section Boronia).
Carpels 4; ovaries free though united at apex on adaxial mar-
gin by the solitary style. Fruit of 1–4 basally connate follicles
(cocci), dehiscing explosively ventrally with separating, elas-
tic endocarp. Seed: sclerotesta smooth or minutely tubercu-
late, rarely prominently rugose (B. cymosa), glossy or dull.
An Australian and New Caledonian genus of 133 species
classified into 7 sections. For keys and descriptions, see Duretto
& al. (2013), Bayly & al. (2015) and R.L. Barrett & al. (2015).
Cyanothamnus Lindl., Sketch Veg. Swan R.: 18. 1839 ≡Boro-
nia ser. Cyaneae Benth., Fl. Austral. 1: 309, 319. 1863 ≡
Boronia sect. Cyanothamnus (Lindl.) F.Muell., Fragm. 9:
113. 1875 ≡Boronia sect. Cyaneae (Benth.) De Wild., Icon.
Horti. Then. 2: 67. 1901, nom. illeg. –Type (designated by
Wilson in Nuytsia 12: 144. 1998): C. ramosus Lindl.
Perennial herbs or shrubs; glabrous or with simple or rarely
stellate hairs (ser. Penicillati). Leaves opposite decussate, sim-
ple, imparipinnate, bipinnate or tripinnate. Flowers axillary or
rarely terminal on short shoots (C. coerulescens (F.Muell.)
Duretto & Heslewood), or in terminal cymes (C. defoliatus
(F.Muell.) Duretto & Heslewood), solitary or in cymes, bisex-
ual, 4-merous. Sepals free, open or imbricate, persistent. Petals
free, imbricate, usually prominently glandular, 1-veined at base
with spreading lateral veins; tip inflexed; caducous or persis-
tent. Stamens 8, all fertile; filaments inwardly curved, linear
to elliptic (rarely slender terete), verrucose towards apex (rarely
smooth), glabrous or hairy; connective usually prominent and
dark coloured, rarely inconspicuous or cream coloured; anthers
scarcely or prominently white-apiculate, inflexed or erect,
rarely apiculum absent. Disc prominent, entire. Carpels 4; ova-
ries free though united at apex on adaxial marginby the solitary
style. Fruit of 1–4 basally connate follicles (cocci), dehiscing
explosively ventrally with separating, elastic endocarp. Seed:
sclerotesta prominently rugose, minutely verrucose, dull, often
with a glaucous white deposit.
An Australian genus of 23 species found in all states and
mainland territories except the Northern Territory. Eleven spe-
cies are confined to eastern Australia,including Tasmania, and
another 11 to south-western Western Australia. Cyanothamnus
coerulescens is widespread and is found from south-western
Western Australia to western Victoria and south-western New
South Wales. The genus has six series: three, Defoliati,Fabi a -
noides and Penicillati, are confined to south-western Western
Australia, one, Polygalifolii, is confined to eastern Australia,
and the two remaining are found in both areas. Series Cya-
nothamnus appears to be polyphyletic (see discussion above),
and further research is required to resolve the matter.
Keys and descriptions of the series, species, subspecies
and varieties can be found under Boronia sect. Cyanothamnus
in the Flora of Australia (Duretto & al., 2013).
Species, subspecies and varieties transferred to
Cyanothamnus
Only four species have previously been described under Cya-
nothamnus: two of these names, C. ramosus and C. tenuis Lindl.,
can be used. Cyanothamnus anethifolius Bartl. is the basionym of
B. ramosa subsp. anethifolia (Bartl.) Paul G.Wilson, and a new
combination for this subspecies is made below. Cyanothamnus
tridactylites Bartl. (described in 1848) is predated by the conspe-
cific B. anemonifolia A.Cunn. (described in 1825) for which a
new combination is made below. Boronia anethifolia A.Cunn.
ex Endl. requires a new name under Cyanothamnus as the epithet
is already used by a taxon based on another type, C. anethifolius
Bartl. (see C. ramosus subsp. anethifolius).
The new name and combinations are listed in alphabetical
order by what the taxon was called under Boronia in Duretto
& al. (2013), which is in most cases the basionym. The new com-
binations for the series are given after the species, and the species
compositions for each of the series are given with that series.
For detailed information on synonymy and types refer to
Duretto & al. (2013) in general as well as Wilson (1971, 1998)
for south-western Western Australian taxa, Neish & Dur-
etto (2000) for C. anemonifolius (A.Cunn.) Duretto & Hesle-
wood and C. nanus (Hook.) Duretto & Heslewood, and
492 Version of Record
Duretto & al. •Boronia is polyphyletic TAXON 69 (3) •June 2020: 481–499
Duretto (2003) for taxa found in eastern Australia and
Tasmania.
Cyanothamnus acanthocladus (Paul G.Wilson) Duretto &
Heslewood, comb. nov. ≡Boronia acanthoclada Paul
G.Wilson in Nuytsia 12(1): 146–148. 1998.
Cyanothamnus anemonifolius (A.Cunn.) Duretto & Hesle-
wood, comb. nov. ≡Boronia anemonifolia A.Cunn. in
Field, Geogr. Mem. New South Wales: 330. 1825.
Cyanothamnus anemonifolius subsp. aurifodinus (P.G.
Neish) Duretto & Heslewood, comb. nov. ≡Boronia ane-
monifolia subsp. aurifodina P.G.Neish in Muelleria 14:
9. 2000.
Cyanothamnus anemonifolius subsp. variabilis (Hook.) Dur-
etto & Heslewood, comb. nov. ≡Boronia variabilis Hook.
in J. Bot. (Hooker) 1: 255. 1834 ≡B. anemonifolia subsp.
variabilis (Hook.) P.G.Neish in Muelleria 14: 11. 2000.
Cyanothamnus anemonifolius subsp. wadbilligensis (P.G.
Neish) Duretto & Heslewood, comb. nov. ≡Boronia ane-
monifolia subsp. wadbilligensis P.G.Neish in Muelleria
14: 10. 2000.
Cyanothamnus quadrangulus Duretto & Heslewood, nom.
nov. ≡Boronia anethifolia A.Cunn. ex Endl. in Endlicher
& al., Enum. Pl.: 16. 1837.
The transfer of B. anethifolia A.Cunn. ex Endl. to Cya-
nothamnus would result in a later homonym due to the existence
of C. anethifolius Bartl. (see C. ramosus subsp. anethifolius).
No other name is available for this species, and so a new name
is proposed. The specific epithet chosen is derived from the
Latin quadri- (four) and angulus (angle) and refers to the dis-
tinctly four-sided stems of this species.
Cyanothamnus baeckeaceus (F.Muell.) Duretto & Hesle-
wood, comb. nov. ≡Boronia baeckeacea F.Muell.,
Fragm. 4: 28. 1863.
Cyanothamnus baeckeaceus subsp. patulus (Paul G.Wilson)
Duretto & Heslewood, comb. nov. ≡Boronia baeckeacea
subsp. patula Paul G.Wilson in Nuytsia 12: 148. 1998.
Cyanothamnus bipinnatus (Lindl.) Duretto & Heslewood,
comb. nov. ≡Boronia bipinnata Lindl. in Mitchell,
J. Exped. Trop. Australia: 225. 1848.
Cyanothamnus bussellianus (F.Muell.) Duretto & Hesle-
wood, comb. nov. ≡Boronia busselliana F.Muell.,
Fragm. 9: 113. 1875.
Cyanothamnus coerulescens (F.Muell.) Duretto & Hesle-
wood, comb. nov. ≡Boronia coerulescens F.Muell. in
Trans. Philos. Soc. Victoria 1: 11. 1854.
Cyanothamnus coerulescens subsp. spicatus (Paul G.Wilson)
Duretto & Heslewood, comb. nov. ≡Boronia coerulescens
subsp. spicata Paul G.Wilson in Nuytsia 1: 200. 1971.
Cyanothamnus coerulescens subsp. spinescens (Benth.)
Duretto & Heslewood, comb. nov. ≡B. spinescens
Benth., Fl. Austral. 1: 319. 1863 ≡B. coerulescens subsp.
spinescens (Benth.) Paul G.Wilson in Nuytsia: 200. 1971.
Cyanothamnus defoliatus (F.Muell.) Duretto & Heslewood,
comb. nov. ≡Boronia defoliata F.Muell., Fragm. 9:
113. 1875.
Cyanothamnus fabianoides (Diels) Duretto & Heslewood,
comb. nov. ≡Eriostemon fabianoides Diels in Bot. Jahrb.
Syst. 35: 322, t. 39K, L. 1904 ≡Boronia fabianoides
(Diels) Paul G.Wilson in Nuytsia 1: 119. 1970.
Cyanothamnus fabianoides subsp. roseus (Paul G.Wilson)
Duretto & Heslewood, comb. nov. ≡Boronia fabianoides
subsp. rosea Paul G.Wilson in Nuytsia 12: 150. 1998.
Cyanothamnus inconspicuus (Benth.) Duretto & Heslewood,
comb. nov. ≡Boronia inconspicua Benth., Fl. Austral. 1:
313. 1863.
Cyanothamnus inflexus (Duretto) Duretto & Heslewood,
comb. nov. ≡Boronia inflexa Duretto in Muelleria 17:
40. 2003.
Cyanothamnus inflexus subsp. grandiflorus (Duretto) Dur-
etto & Heslewood, comb. nov. ≡Boronia inflexa subsp.
grandiflora Duretto in Muelleria 17: 43. 2003.
Cyanothamnus inflexus subsp. montiazurus (Duretto) Dur-
etto & Heslewood, comb. nov. ≡Boronia inflexa subsp.
montiazura Duretto in Muelleria 17: 42. 2003.
Cyanothamnus inflexus subsp. torringtonensis (Duretto)
Duretto & Heslewood, comb. nov. ≡Boronia inflexa
subsp. torringtonensis Duretto in Muelleria 17: 44. 2003.
Cyanothamnus montimulliganensis (Duretto) Duretto &
Heslewood, comb. nov. ≡Boronia montimulliganensis
Duretto in Muelleria 17: 29. 2003.
Cyanothamnus nanus (Hook.) Duretto & Heslewood, comb.
nov. ≡Boronia nana Hook., Icon. Pl. 3: t. 270. 1840.
Cyanothamnus nanus var. hyssopifolius (Melville) Duretto
& Heslewood, comb. nov. ≡Boronia nana var. hyssopifo-
lia Melville in Kew Bull. 9: 463. 1954.
Cyanothamnus nanus var. pubescens (Benth.) Duretto
& Heslewood, comb. nov. ≡Boronia polygalifolia var.
(?) pubescens Benth., Fl. Austral. 1: 321. 1863 ≡B. nana
Version of Record 493
TAXON 69 (3) •June 2020: 481–499 Duretto & al. •Boronia is polyphyletic
var. pubescens (Benth.) J.H.Willis in Victorian Naturalist
73: 192. 1957.
Cyanothamnus occidentalis (Duretto) Duretto & Heslewood,
comb. nov. ≡Boronia occidentalis Duretto in Muelleria
17: 36. 2003.
Cyanothamnus penicillatus (Benth.) Duretto & Heslewood,
comb. nov. ≡Boronia penicillata Benth., Fl. Austral. 1:
322. 1863.
Cyanothamnus polygalifolius (Sm.) Duretto & Heslewood,
comb. nov. ≡Boronia polygalifolia Sm., Tracts Nat.
Hist.: 297, t. 7. 1798.
Cyanothamnus ramosus subsp. anethifolius (Bartl.) Duretto
& Heslewood, comb. nov. ≡Cyanothamnus anethifolius
Bartl. in Lehmann, Pl. Preiss. 1: 170. 1845 ≡Boronia
ramosa subsp. anethifolia (Bartl.) Paul G.Wilson in
Nuytsia 1: 201. 1971.
Cyanothamnus ramosus subsp. lesueuranus (Paul G.Wilson)
Duretto & Heslewood, comb. nov. ≡Boronia ramosa
subsp. lesueurana Paul G.Wilson in Nuytsia 12:
151. 1998.
Cyanothamnus rigens (Cheel) Duretto & Heslewood, comb.
nov. ≡Boronia rigens Cheel in J. Proc. Roy. Soc. New
South Wales 62: 297. 1929.
Cyanothamnus subsessilis (Benth.) Duretto & Heslewood,
comb. nov. ≡Boronia subsessilis Benth., Fl. Austral. 1:
322. 1863.
Cyanothamnus warangensis (Duretto) Duretto & Heslewood,
comb. nov. ≡Boronia warangensis Duretto in Muelleria
17: 31. 2003.
Cyanothamnus westringioides (Paul G.Wilson) Duretto &
Heslewood, comb. nov. ≡Boronia westringioides Paul
G.Wilson in Nuytsia 12: 152. 1998.
Cyanothamnus yarrowmerensis (Duretto) Duretto & Hesle-
wood, comb. nov. ≡Boronia yarrowmerensis Duretto in
Muelleria 17: 32. 2003.
Cyanothamnus: infrageneric classification
Included under each series are the species, subspecies and
varieties placed in those series.
Cyanothamnus ser. Coerulescentes (Duretto) Duretto &
Heslewood, comb. nov. ≡Boronia ser. Coerulescentes
Duretto, Fl. Australia 26: 582. 2013.
Monotypic, occurring in south-western Western Australia,
South Australia, Victoria, and New South Wales: Cyanothamnus
coerulescens,C. coerulescens subsp. coerulescens,C. coerules-
cens subsp. spicatus,C. coerulescens subsp. spinescens.
Cyanothamnus Lindl. ser. Cyanothamnus
A series of 12 species; occurring in all Australian states and
the Australian Capital Territory though it is absent from the
Northern Territory: Cyanothamnus anemonifolius,C. anemoni-
folius subsp. anemonifolius,C. anemonifolius subsp. aurifodi-
nus,C. anemonifolius subsp. variabilis,C. anemonifolius
subsp. wadbilligensis,C. bipinnatus,C. inconspicuus,
C. inflexus,C. inflexus subsp. grandiflorus,C. inflexus subsp.
inflexus,C. inflexus subsp. montiazurus,C. inflexus subsp. tor-
ringtonensis,C. montimulliganensis,C. nanus,C. nanus var.
hyssopifolius,C. nanus var. nanus,C. nanus var. pubescens,
C. occidentalis,C. quadrangulus,C. ramosus,C. ramosus
subsp. anethifolius,C. ramosus subsp. lesueuranus,C. ramosus
subsp. ramosus,C. rigens,C. warangensis,C. yarrowmerensis.
Cyanothamnus ser. Defoliati (Duretto) Duretto & Heslewood,
comb. nov. ≡Boronia ser. Defoliatae Duretto, Fl.
Australia 26: 583. 2013.
A series of four species endemic to south-western West-
ern Australia: Cyanothamnus bussellianus,C. defoliatus,
C. subsessilis,C. tenuis.
Cyanothamnus ser. Fabianoides (Duretto) Duretto & Hesle-
wood, comb. nov. ≡Boronia ser. Fabianoides Duretto,
Fl. Australia 26: 583. 2013.
A series of two species in south-western Western Australia:
Cyanothamnus acanthocladus,C. fabianoides,C. fabianoides
subsp. fabianoides,C. fabianoides subsp. roseus.
Cyanothamnus ser. Penicillati (Duretto) Duretto & Hesle-
wood, comb. nov. ≡Boronia ser. Penicillatae Duretto,
Fl. Australia 26: 583. 2013.
A series of three species from south-western Western
Australia: Cyanothamnus penicillatus,C. westringioides,
C. baeckeaceus,C. baeckeaceus subsp. baeckeaceus,
C. baeckeaceus subsp. patulus.
Cyanothamnus ser. Polygalifolii (Duretto) Duretto & Hesle-
wood, comb. nov. ≡Boronia ser. Polygalifoliae Duretto,
Fl. Australia 26: 582. 2013.
Monotypic, occurring in Queensland and New South
Wales: Cyanothamnus polygalifolius.
■AUTHORS CONTRIBUTIONS
MFD and MHH developed the scientific framework of this study
that fits into the framework of a larger study by MFD, MJB and
MHH. MFD, MHH and MJB wrote the article. MFD, MJB and MHH
collected material. MHH completed the laboratory work. MHH and
MFD performed the molecular study and the data analyses. MFD com-
pleted the nomenclature. —MFD, https://orcid.org/0000-0003-1013-
4291; MHH, https://orcid.org/0000-0003-0100-8023; MJB, https://
orcid.org/0000-0001-6836-5493
494 Version of Record
Duretto & al. •Boronia is polyphyletic TAXON 69 (3) •June 2020: 481–499
■ACKNOWLEDGEMENTS
We would like to thank Andrew Ford (CSIRO), Paul Forster
(BRI), Andrew Orme (NSW), Matt Renner (NSW), Susan Rutherford
(NSW) and Andy Young for supplying plant material for our study;
the Directors of CANB, MEL, MELU, NSW and PERTH for access
to their herbaria and the loan of material, and the Director of the
Australian National Botanic Garden for permission to sample the liv-
ing collection; the Hermon Slade Foundation (Grant HSF13/6) for
providing funding for a separate project, on the Phebalium and
Philotheca group of genera (Rutaceae), that generated many of the
sequences used in this project; Peter Wilson for assistance with Latin
and nomenclature; and Phil Garnock-Jones (Victoria Univ., NZ) for
critically reading the manuscript and providing excellent suggestions
that improved it.
■LITERATURE CITED
Appelhans, M.S. & Wen, J. 2020. Phylogenetic placement of Ivodea
and biogeographic affinities of Malagasy Rutaceae. Pl. Syst. Evol.
306: 7. https://doi.org/10.1007/s00606-020-01633-3
Appelhans, M.S., Wen, J. & Wagner, W.L. 2014a. A molecular phy-
logeny of Acronychia,Euodia,Melicope and relatives (Rutaceae)
reveals polyphyletic genera and key innovations for species rich-
ness. Molec. Phylogen. Evol. 79: 54–68. https://doi.org/10.1016/
j.ympev.2014.06.014
Appelhans, M.S., Wen, J., Wood, K.R., Allan, G.J., Zimmer, E.A. &
Wagner, W.L. 2014b. Molecular phylogenetic analysis of Hawai-
ian Rutaceae (Melicope,Platydesma and Zanthoxylum) and their
different colonisation patterns. Bot. J. Linn. Soc. 174: 425–448.
https://doi.org/10.1111/boj.12123
Appelhans, M.S., Wood, K.R. & Wagner, W.L. 2017. Reduction of
the Hawaiian genus Platydesma into Melicope section Pelea
(Rutaceae) and notes on the monophyly of the section. PhytoKeys
91: 125–137. https://doi.org/10.3897/phytokeys.91.21363
Appelhans, M.S., Wen, J., Duretto, M.F., Crayn,D. & Wagner, W.L.
2018. Historical biogeography of Melicope (Rutaceae) and its
close relatives with a special emphasis on Pacific dispersals.
J. Syst. Evol. 56: 576–599. https://doi.org/10.1111/jse.12299
Armstrong, J.A. 2002. The genus Zieria (Rutaceae): A systematic and
evolutionary study. Austral. Syst. Bot. 15: 277–463. https://doi.
org/10.1071/SB00040
Armstrong, J.A. & Powell, J.M. 1980. Neobyrnesia (Rutaceae): A
new genus endemic to Northern Australia. Telopea 1: 399–408.
https://doi.org/10.7751/telopea19803602
Barrett, R.A., Bayly, M.J., Duretto, M.F., Forster, P.I., Ladiges, P.Y.
& Cantrill, D.J. 2015 A chloroplast phylogeny of Zieria
(Rutaceae) in Australia and New Caledonia shows widespread
incongruence with species-level taxonomy. Austral. Syst. Bot.
27: 427–449. https://doi.org/10.1071/SB14033
Barrett, R.A., Bayly, M.J., Duretto, M.F., Forster, P.I., Ladiges, P.Y.
& Cantrill, D.J. 2018. Phylogenetic analysis of Zieria (Rutaceae)
in Australia and New Caledonia based on nuclear ribosomal DNA
reveals species polyphyly, divergent paralogues and incongruence
with chloroplast DNA. Austral. Syst. Bot. 31: 16–47. https://doi.
org/10.1071/SB16034
Barrett, R.L., Barrett, M.D. & Duretto, M.F. 2015. Four new species
of Boronia (Rutaceae) from the Kimberley region of Western
Australia. Nuytsia 26: 89–109.
Bartling, F.T. 1848. Diosmeae. Pp. 226–228 in: Lehmann, C. (ed.),
Plantae Preissianae, vol. 2. Hamburgi: Sumptibus Meissneri.
Bayly, M.J., Holmes, G.D., Forster, P.I., Cantrill, D.J. & Ladiges, P.Y.
2013 Major clades of Australasian Rutoideae (Rutaceae) based on
rbcL and atpB sequences. PLoS ONE 8: e72493. https://doi.org/
10.1371/journal.pone.0072493
Bayly, M.J., Duretto, M.F., Holmes, G.D., Forster, P.I., Cantrill, D.J.
& Ladiges, P.Y. 2015. Transfer of the New Caledonian genus
Boronella to Boronia (Rutaceae) based on analyses of cpDNA
and nrDNA. Austral. Syst. Bot. 28: 111–123. https://doi.org/10.
1071/SB15008
Bayly, M.J., Holmes, G.D., Forster, P.I., Cantrill, D.J., Mun-
zinger, J. & Ladiges, P.Y. 2016 Phylogeny, classification and
biogeography of Halfordia (Rutaceae) in Australia and New Cal-
edonia. Pl. Syst. Evol. 302: 1457–1470. https://doi.org/10.1007/
s00606-016-1344-0
Burgman, M.A. 1985. Cladistics, phenetics and biogeography of
populations of Boronia inornata Turcz. (Rutaceae) and the Euca-
lyptus diptera Andrews (Myrtaceae) species complexes in West-
ern Australia. Austral. J. Bot. 33: 419–431. https://doi.org/10.
1071/BT9850419
Chase, M.W., Morton, C.M. & Kallunki, J.A. 1999. Phylogenetic
relationships of Rutaceae: A cladistic analysis of the subfamilies
using evidence from rbcL and atpB sequence variations. Amer.
J. Bot. 86: 1191–1199. https://doi.org/10.2307/2656983
Crayn, D.M., Rossetto, M. & Maynard, D.J. 2006. Molecular phy-
logeny and dating reveals an Oligo-Miocene radiation of dry-
adapted shrubs (former Tremandraceae) from rainforest tree pro-
genitors (Elaeocarpaceae) in Australia. Amer. J. Bot. 93:
1328–1342. https://doi.org/10.3732/ajb.93.9.1328
Crisp, M., Cook, L. & Steane, D. 2004. Radiation of the Australian
flora: What can comparisons of molecular phylogenies across
multiple taxa tell us about the evolution of diversity in present-
day communities? Philos. Trans., Ser. B 359: 1551–1571.
https://doi.org/10.1098/rstb.2004.1528
Crisp, M.D., Arroyo, M.T.K., Cook, L.G., Gandolfo, M.A.,
Jordan, G.J., McGlone, M.S., Weston, P.H., Westoby, M.,
Wilf, P. & Linder, P.H. 2009. Phylogenetic biome conservatism
on a global scale. Nature 458: 754–758. https://doi.org/10.1038/
nature07764
Donoghue, M.J. & Edwards, E.J. 2014. Biome shifts and niche evo-
lution in plants. Annual Rev. Ecol. Evol. Syst. 45: 547–572.
https://doi.org/10.1146/annurev-ecolsys-120213-091905
Duretto, M.F. 2003. Notes on Boronia (Rutaceae) in eastern and north-
ern Australia. Muelleria 17: 19–135.
Duretto, M.F. 2013. Boronia (Rutaceae). Pp. 582–583 in: Wilson, A.
(ed.), Flora of Australia, vol. 26, Meliaceae, Rutaceae and Zygo-
phyllaceae. Melbourne: CSIRO Publishing.
Duretto, M.F. & Ladiges, P.Y. 1999. A cladistic analysis of Boronia
section Valvatae (Rutaceae). Austral. Syst. Bot. 11: 635–636.
https://doi.org/10.1071/SB97040
Duretto, M.F., Wilson, P.G. & Ladiges, P.Y. 2013. Boronia.
Pp. 12–282 in: Wilson, A. (ed.), Flora of Australia, vol. 26, Me-
liaceae, Rutaceae and Zygophyllaceae. Melbourne: CSIRO
Publishing.
Engler, H.G.A. 1931. Rutaceae. Pp. 187–359 in: Engler, H.G.A.
& Prantl, K. (eds.), Die natürlichen Pflanzenfamilien, ed. 2,
vol. 19a. Leipzig: Engelmann.
Gadek, P.A., Fernando, E.S., Quinn, C.J., Hoot, S.B., Terrazas, T.,
Sheahan, M.C & Chase, M.W. 1996. Sapindales: Molecular
delimitation and infraordinal groups. Amer. J. Bot. 83: 802–811.
https://doi.org/10.1002/j.1537-2197.1996.tb12769.x
Golenberg, E.M., Clegg, M.T., Durbin, M.L., Doebley, J. & Ma, D.P.
1993. Evolution of a noncoding region of the chloroplast genome.
Molec. Phylogen. Evol. 2: 52–64. https://doi.org/10.1006/mpev.
1993.1006
Groppo, M., Pirani, J.R.R., Salatino, M.L.F., Blanco, S.R. &
Kallunki, J.A. 2008. Phylogeny of Rutaceae based on two non-
coding regions from cpDNA. Amer. J. Bot. 95: 985–1005.
https://doi.org/10.3732/ajb.2007313
Groppo, M., Kallunki, J.A., Pirani, J.R.R. & Antonelli, A. 2012. Chil-
ean Pitavia more closely related to Oceania and Old World Rutaceae
than Neotropical groups: Evidence from two cpDNA noncoding
Version of Record 495
TAXON 69 (3) •June 2020: 481–499 Duretto & al. •Boronia is polyphyletic
regions, with a new subfamilial classification of the family. PhytoKeys
19: 9–29. https://doi.org/10.3897/phytokeys.19.3912
Harbaugh, D.T., Wagner, W.L., Allan, G.J. & Zimmer, E.A. 2009.
The Hawaiian Archipelago is a stepping stone for dispersal in
the Pacific: An example from the plant genus Melicope
(Rutaceae). J. Biogeogr. 36: 230–241. https://doi.org/10.1111/j.
1365-2699.2008.02008
Hartley, T.G. 1979. A revision of the genus Tet ra c t o m i a (Rutaceae).
J. Arnold Arbor. 60: 127–153. https://doi.org/10.5962/bhl.part.12825
Hartley, T.G. 1997. Five new rain forest genera of Australasian Ruta-
ceae. Adansonia 19: 189–212.
Hartley, T.G. 2001a. Morphology and biogeography in Australasian-
Malesian Rutaceae. Malayan Nat. J. 55: 197–219.
Hartley, T.G. 2001b. On the taxonomy and biogeography of Euodia
and Melicope (Rutaceae). Allertonia 8(1): 1–328.
Holzmeyer, L., Duretto, M., Crayn, D., Hörandl, E., Heslewood, M.,
Jayanthan, J. & Appelhans, M.S. 2015. Phylogeny of Acrony-
chia (Rutaceae) and first insights into its historical biogeography
and the evolution of fruit characters. PLoS ONE 10(8):
e0136296. https://doi.org/10.1371/journal.pone.0136296
Johnson, K.A., Holland, B.R., Heslewood, M.M. & Crayn, D.M.
2011. Supermatrices, supertrees and serendipitous scaffolding:
Inferring a well-resolved, genus-level phylogeny of Styphelioi-
deae (Ericaceae) despite missing data. Molec. Biol. Evol. 62:
146–158. https://doi.org/10.1016/j.ympev.2011.09.011
Kubitzki, K., Kallunki, J.A., Duretto, M. & Wilson, P.G. 2011.
Rutaceae. Pp. 276–356 in: Kubitzki, K. & Bayer, C. (eds.), The
families and genera of vascular plants, vol. 10. Berlin: Springer.
https://doi.org/10.1007/978-3-642-14397-7_16
Levinson, G. & Gutman, G.A. 1987. Slipped-strand mispairing: A major
mechanism for DNA sequence evolution. Molec. Biol. Evol. 4:
203–221. https://doi.org/10.1093/oxfordjournals.molbev.a040442
Lindley, J. 1839. A sketch of the vegetation of the Swan River Colony.
Pp. i–lvii, pl. 1–9 in: Appendix to the first twenty-three volumes of
Edwards’s Botanical Register. London: James Ridgeway. https://
doi.org/10.5962/bhl.title.6590
Lucas, E.J., Harris, S.A., Mazine, F.F., Belsham, S.R., Nic
Lughadha, E.M., Telford, A., Gasson, P.E. & Chase, M.W.
2007. Suprageneric phylogenetics of Myrteae, the generically
richest tribe in Myrtaceae (Myrtales). Taxon 56: 1105–1128.
https://doi.org/10.2307/25065906
Maddison, W.P. & Maddison, D.R. 2000. MacClade 4: Analysis of
phylogeny and character evolution. Sunderland, MA: Sinauer.
Morton, C.M. & Telmer, C. 2014. New subfamily classification for
the Rutaceae. Ann. Missouri Bot. Gard. 99: 620–641. https://doi.
org/10.3417/2010034
Neish, P.G. & Duretto, M.F. 2000. The taxonomy of Boronia anemo-
nifolia and B. rigens (Boronia sect. Cyanothamnus, Rutaceae).
Muelleria 14: 3–16.
Nylander, J.A.A. 2004. MrModeltest, version 2.3. Program distributed
by the author. Evolutionary Biology Centre, Uppsala University.
https://github.com/nylander/MrModeltest2
Poon, W.-S., Shaw, P.-C., Simmons, M.P. & But, P.P.-H. 2007. Con-
gruence of molecular, morphological, and biochemical profiles in
Rutaceae: A cladistic analysis of the subfamilies Rutoideae and
Toddalioideae. Syst. Bot. 32: 837–846. https://doi.org/10.1600/
036364407783390692
Prince, L.M. 2010. Phylogenetic relationships and species delimitation
in Canna (Cannaceae). Pp. 307–331 in: Seberg, O., Petersen, G.,
Barfod, A.S. & Davis, J.I. (eds.), Diversity, phylogeny, and evolu-
tion in the monocotyledons: Proceedings of the Fourth Inter-
national Conference on the Comparative Biology of the
Monocotyledons and the Fifth International Symposium on Grass
Systematics and Evolution. Århus: Aarhus University Press.
Rabarimanarivo, M.N., Rakotonirina, N.H., Phillipson, P.B.,
Lowry, P.P., II, Labat, J.-N. & Pignal, M. 2015. Révision du
genre Ivodea Capuron (Rutaceae), endémique de Madagascar et
de l’archipel des Comores. Adansonia, sér. 3, 37(1): 63–102.
https://doi.org/10.5252/a2015n1a6
Rambaut, A., Suchard, M.A., Xie, D. & Drummond, A.J. 2014.
Tracer, version 1.6. Program available from http://beast.bio.ed.
ac.uk/Tracer
Ronquist, F., Teslenko, M., Van der Mark, P., Ayres, D.L.,
Darling, A., Höhna, S., Larget, B., Liu, L., Suchard, M.A. &
Huelsenbeck, J.P. 2012. MrBayes 3.2: Efficient Bayesian
phylogenetic inference and model choice across a large model
space. Syst. Biol. 61: 539–542. https://doi.org/10.1093/sysbio/
sys029
Sang,T.,Crawford,D.&Stuessy,T.1997. Chloroplast DNA phylogeny,
reticulate evolution, and biogeography of Pae onia (Paeoniaceae).
Amer. J. Bot. 84: 1120–1136. https://doi.org/10.2307/2446155
Scott, K., McIntyre, C. & Playford, J. 2000. Molecular analyses sug-
gest a need for a significant rearrangement of Rutaceae subfam-
ilies and a minor reassessment of species relationships within
Flindersia.Pl. Syst. Evol. 223: 15–27. https://doi.org/10.1007/
BF00985324
Shan, F., Yan, G. & Plummer, J.A. 2006. Basic chromosome number
in Boronia (Rutaceae) —Competing hypotheses examined. Aus-
tral. J. Bot. 54: 681–689. https://doi.org/10.1071/BT05050
Simmons, M.P. & Freudenstein, J.V. 2011. Spurious 99% bootstrap
and jackknife support for unsupported clades. Molec. Phylogen.
Evol. 61: 177–191. https://doi.org/10.1016/j.ympev.2011.06.003
Simmons, M.P. & Ochoterena, H. 2000. Gaps as characters in
sequence-based phylogenetic analyses. Syst. Biol. 49: 369–381.
https://doi.org/10.1093/sysbio/49.2.369
Swofford, D.L. 2002. PAUP*: Phylogenetic analysis using parsimony
(*and other methods), version 4.0b10. Sunderland, Massachusetts:
Sinauer.
Taberlet, P., Gielly, L., Pautou, G. & Bouvet, J. 1991. Universal
primers for amplification of three non-coding regions of chloro-
plast DNA. Pl. Molec. Biol. 17: 1105–1109. https://doi.org/10.
1007/BF00037152
Tate, J.A. & Simpson, B.B. 2003. Paraphyly of Tarasa (Malvaceae)
and diverse origins of the polyploid species. Syst. Bot. 28:
723–737. https://doi.org/10.1043/02-64.1
Weston, P., Carolin, R. & Armstrong, J. 1984. A cladistic analysis of
Boronia Sm. and Boronella Baill. (Rutaceae). Austral. J. Bot. 32:
187–203. https://doi.org/10.1071/BT9840187
White, T.J., Bruns, T., Lee, S. & Taylor, J. 1990. Amplification and
direct sequencing of fungal ribosomal RNA genes for phyloge-
netics. Pp. 315–322 in: Innis, M., Gelfand, D., Sninsky, J.
& White, T. (eds.), PCR protocols: A guide to methods and appli-
cations. San Diego: Academic Press. https://doi.org/10.1016/
B978-0-12-372180-8.50042-1
Wilberg, E.W. 2015. What’s in an outgroup? The impact of outgroup
choice on the phylogenetic position of Thalattosuchia (Croco-
dylomorpha) and the origin of Crocodyliformes. Syst. Biol. 64:
621–637. https://doi.org/10.1093/sysbio/syv020
Wilson, P.G. 1971. Taxonomic notes on the family Rutaceae, princi-
pally of Western Australia. Nuytsia 1: 197–207.
Wilson, P.G. 1998. New names and new taxa in the genus Boronia
(Rutaceae) from Western Australia, with notes on seed characters.
Nuytsia 12: 119–154.
Wright, S., Yong, C.G., Wichman, S.R., Dawson, J.W. &
Gardner, R.C. 2001. Stepping stones to Hawaii: A transequator-
ial dispersal pathway for Metrosideros (Myrtaceae) inferred from
nrDNA (ITS+ETS). J. Biogeogr. 28: 769–774. https://doi.org/10.
1046/j.1365-2699.2001.00605.x
496 Version of Record
Duretto & al. •Boronia is polyphyletic TAXON 69 (3) •June 2020: 481–499
Appendix 1. Voucher details and GenBank accession numbers of taxa sampled. Names follow new taxonomy.
Taxon name + taxonomic authority, Country: largest political subdivision/locality, Principal collector + number (Herbarium and accession number), ETS, ITS,
trnL-trnF,psbA-trnH,rbcL GenBank accession numbers. An asterix (*) indicates a new sequence. A dash (–) indicates missing data. Abbreviations are:
ABGMA = Australian Botanic Garden Mount Annan, ACT = Australian Capital Territory, ANBG = Australian National Botanic Garden, cult. = cultivated,
FR = Forest Reserve, NP = National Park, NSW = New South Wales, Qld = Queensland, RBGS = Royal Botanic Gardens Sydney, SF = State Forest, Vic. =
Victoria, WA = Western Australia.
Acradenia frankliniae Kippist, Australia: NSW/cult. ABGMA (ex Woodbank Nursery), M.F. Duretto 3091 (NSW 1058547), MN082803*, MN082849*,
MN082987*, MN082893*, MN083034*;Acronychia baeuerlenii T.G.Hartley, Australia: NSW/Nightcap Range, M. Rossetto s.n. (SCU17, ABNIG),
LN849223, LN849139, LN849180, LN849162, –;Acronychia chooreechillum (F.M.Bailey) C.T.White, Australia: Qld/Bellenden Ker Range - Mt Bartle Frere,
I.R. Telford 11393 (CANB - CBG9102594), MN082804*, MN082850*, MN082988*, MN082894*,–;Asterolasia drummondii Paul G.Wilson, Australia:
WA/Yandin Hill, B.J. Mole 330 (NSW 1003538), MN082805*, MN082851*, MN082989*, MN082895*, MN083035*;Boronia alata Sm., Australia:
WA/Leeuwin-Naturaliste NP, M.J. Bayly 1955 (MEL 2383602), KP867734, KP867656, KP867809, MN082896*,–;Boronia algida F.Muell., Australia:
Vic./Mount Buffalo, M.J. Bayly 1958 (MEL 2383605), KP867742, KP867677, KP867779, MN082897*,–;Boronia alulata Sol. ex Benth., Australia:
Qld/33 km along Middle Peak track to Shelburne Bay, P.I. Forster 33637 (BRI AQ743487), KP867724, KP867696,KP867793, MN082898*,–;Boronia bow-
manii F.Muell., Australia: Qld/track to Shelburne Bay Homestead, P.I. Forster 33638 (BRI AQ743488), KP867723, KP867692, KP867760, MN082899*,–;
Boronia citriodora Gunn ex Hook.f. subsp. citriodora, Australia: Tas./Cradle Mountain NP, M.F. Duretto 2244 (HO, NSW 1057657), KP867709,
KP867697, KP867806, –,–;Boronia crenulata Sm., Australia: WA/between Denmark and Nornalup, M.J. Bayly 1957 (MEL 2383610), KP867740,
KP867660, KP867788, MN082900*,–;Boronia cymosa Endl., Australia: WA/southeast of Eneabba, M.J. Bayly 1906 (MEL 2383604), KP867729,
KP867684, KP867771, MN082901*,–;Boronia deanei Maiden & Betche, Australia: ACT/cult. ANBG (ex NSW, near top of Fitzroy Falls), M.J. Bayly
2001 (MELU 105858), KP867733, KP867704, KP867795, MN082902*,–;Boronia denticulata Sm., Australia: WA/Duke of Orleans Bay, M.J. Bayly 1944
(MEL 2383606), KP867725, KP867658, KP867772, MN082903*,–;Boronia edwardsii Benth., Australia: SA/Mount Scrub Road, M.J. Bayly 1974 (MEL
2383596), KP867744, KP867694, KP867786, MN082904*,–;Boronia excelsa Duretto, Australia: Qld/Mount Windsor NP, P.I. Forster 34665 (BRI
AQ745462), KP867752, KP867659, KP867810, MN082905*,–;Boronia falcifolia A.Cunn. ex Endl., Australia: Qld/Great Sandy NP, P.I. Forster 34199
(BRI AQ0743521), KP867754, KP867678, KP867801, MN082906*,–;Boronia filifolia F.Muell., Australia: SA/Cox Scrub, M.J. Bayly 1977 (MEL
2383598), KP867722, KP867654, KP867767, MN082907*,–;Boronia floribunda Sieber ex Rchb., Australia: NSW/Hornsby Heights, M.F. Duretto 3007
(NSW 1005407), KP867730, KP867671, KP867811, MN082908*,–;Boronia galbraithiae Albr., Australia: Vic./Biragolong SF - Mt Difficulty, M.J. Bayly
2031 (MELU), –, MN082852*, MN082990*, MN082909*,–;Boronia gracilipes F.Muell., Australia: ACT/ cult. ANBG (ex WA, Valley of the Giants, near
Nornalup), M.J. Bayly 2003 (MELU 105859), KP867735, KP867669, KP867774, MN082910*,–;Boronia granitica Maiden & Betche, Australia:
Qld/Passchendaele SF, M.T. Mathieson 259 (BRI AQ745449), KP867731, KP867680, KP867764, MN082911*,–;Boronia heterophylla F.Muell.,
Australia: ACT/cult. ANBG (ex Kuranga Native Nursery), M.J. Bayly 2004 (MELU 105860), KP867736, KP867657, KP867775, MN082912*,–;Boronia
imlayensis Duretto, Australia: ACT/cult. ANBG (ex NSW, Mount Imlay), M.J. Bayly 2005 (MELU 105861), KP867728, KP867652, KP867776,
MN082913*,–;Boronia inornata Turcz., Australia: WA/Lake King - Ravensthorpe road, M.J. Bayly 1947 (MEL 2383608), KP867719, KP867688,
KP867773, MN082914*,–;Boronia keysii Domin, Australia: Qld/Great Sandy NP, M.T. Mathieson 281 (BRI AQ746001), KP867713, KP867689,
KP867770, MN082915*,–;Boronia lanceolata F.Muell., Australia: NT/Nitmiluk NP - Edith Falls, M.F. Duretto 532 (MEL 2040294, BRI AQ536036),
KP867732, KP867679, KP867790, MN082916*,–;Boronia lanuginosa Endl., Australia: NT/Nitmiluk NP - Edith Falls, M.F. Duretto 1243 (MELU),
KP867715, KP867691, KP867785, –,–;Boronia latipinna J.H.Willis, Australia: Vic./Wonderland Range, M.J. Bayly 1983 (MEL 2383593), KP867749,
KP867666, KP867794, MN082917*,–;Boronia ledifolia (Vent.) DC., Australia: NSW/Ku-ring-gai Chase NP, M.F. Duretto 3006 (NSW 1005406),
KP867758, KP867700, KP867781, MN082918*,–;Boronia microphylla Sieber ex Rchb., Australia: NSW/Newnes SF, P. Hind 6706 (NSW 406248),
KP867755, KP867683, KP867787, MN082919*,–;Boronia muelleri (Benth.) Cheel, Australia: Vic./Bunyip State Park, M.J. Bayly 1968 (MELU 120564a),
–, KP867665, KP867803, MN082920*,–;Boronia pancheri (Baill.) Duretto & Bayly, New Caledonia: Province Sud/Pernod Creek, M.J. Bayly 2046 (MEL
2383623, BRI, NOU), KP867739, KP867682, KP867784, MN082921*, JN987078; Boronia parviflora Sm., Australia: Tas./Tasman Peninsula, M.F. Duretto
2238 (HO 561573), KP867727, KP867651, KP867780, –,–;Boronia parvifolia (Baker f.) Duretto & Bayly 1, New Caledonia: Province Sud/Pernod Creek, M.J.
Bayly 2047 (MEL 2383624, BRI, NOU), KP867741, KP867673, KP867762, MN082922*, JN987077; Boronia parvifolia (Baker f.) Duretto & Bayly 2,New
Caledonia: Province Nord/Mt Kaala, M.J. Bayly 2115 (MEL 2383638, BRI, NOU), KP867750, KP867662, KP867783, MN082923*,–;Boronia pilosa Labill.
subsp. pilosa, Australia: Tas./Chimney Pot Hill Road, M.F. Duretto 2126 (HO 549436), KP867746, KP867648, KP867766, MN082924*,–;Boronia pinnata
Sm., Australia: NSW/Ku-ring-gai Chase NP, M.F. Duretto 3005 (NSW 1006081), KP867720, KP867672, KP867798, MN082925*,–;Boronia repanda (F.
Muell. ex Maiden & Betche) Maiden & Betche, Australia: Qld/Kurrajong Lane, M.T. Mathieson 201 (BRI AQ745476), KP867721, KP867649, KP867807,
MN082926*,–;Boronia rhomboidea Hook., Australia: Tas./Cradle Mountain NP, M.F. Duretto 2245 (HO, NSW 1057658), KP867708, KP867695,
KP867797, MN082927*,–;Boronia rivularis C.T.White, Australia: Qld/Great Sandy NP, M.T. Mathieson 279 (BRI AQ745999), –, KP867687, KP867789,
MN082928*,–;Boronia rosmarinifolia A.Cunn. ex Endl., Australia: Qld/Mt Bilewilum, P.I. Forster 34191 (BRI AQ743519), KP867751, KP867685,
KP867763, MN082929*,–;Boronia safrolifera Cheel, Australia: NSW/cult. RBGS (ex NSW, Crowdy Bay NP), M. Viler 10 (NSW 787933), KP867747,
KP867653, KP867778, MN082930*,–;Boronia scabra Lindl. subsp. scabra, Australia: WA/near Young River, M.J. Bayly 1946 (MEL 2383611),
KP867737, KP867663, KP867799, MN082931*, JN987079; Boronia serrulata Sm., Australia: NSW/cult. RBGS (ex NSW, Royal NP), A.N. Rodd 5623
(NSW 196291), KP867759, KP867693, KP867804, MN082932*,–;Boronia spathulata Lindl., Australia: WA/north of Duke of Orleans Bay, M.J. Bayly
1945 (MEL 2383607), KP867757, KP867681, KP867768, MN082933*,–;Boronia squamipetala Duretto, Australia: Qld/Heathlands Ranger Base - Telegraph
Line (south), P.I. Forster 33759 (BRI AQ743438), KP867706, KP867699, KP867805, MN082934*,–;Boronia ternata Endl. var. ternata, Australia: WA/Boor-
abbin NP, M.J. Bayly 1931 (MEL 2383603), KP867726, KP867701, KP867777, MN082935*, JN987080; Boronia thujona A.R.Penfold & M.B.Welch,
Australia: ACT/cult. ANBG (ex NSW, Budderoo NP), M.J. Bayly 2007 (MELU 105862), KP867717, KP867661, KP867802, MN082936*,–;Bosistoa trans-
versa J.F.Bailey & C.T.White, Australia: ACT/cult. ANBG (ex Qld, Mitchell Logging Area in SF 256), P. Beesley 933A (CANB CBG8604257), MN082806*,
MN082853*, MN082991*, MN082937*,–;Bouchardatia neurococca (F.Muell.) Baill., Australia: Qld/Northern edge of Bukali Scrub - about 9 km due NNE
of Monto, A.B. Pollock 2605 (NSW 822881), MN082807*, MN082854*, MN082992*, MN082938*,–;Brombya platynema F.Muell., Australia: Qld/end of
Shell Pocket road - El Arish, A. Ford 4819 (L), –, HG971315, HG971163, HG971034, –;Coatesia paniculata F.Muell., Australia: Qld/Mt Wooroolin - 4 km
WNW of Kingaroy, I.R. Telford 11035 (CANB CBG9102208), MN082808*, MN082855*, MN082993*, MN082939*, MN083036*;Comptonella microcarpa
(Perkins) T.G.Hartley, New Caledonia, Munzinger 679 (MO), HG971472, HG971318, HG971286/HG971274, HG971035, –;Comptonella microcarpa
(Perkins) T.G.Hartley, New Caledonia, M.J. Bayly 2089 (MEL 2383666A), –,–,–,–, JN987087; Comptonella sessilifoliola (Guillaumin) T.G.Hartley, New
Caledonia, McPherson 18023 (MO), HG971475, HG971322, HG971276/HG971288, HG971038, –;Correa baeuerlenii F.Muell., Australia: NSW/cult. RBGS
(ex NSW, Mumbulla Creek), J.A. Armstrong 1289 (NSW4217532), MN082809*, MN082856*, MN082994*, MN082940*,–;Crowea angustifolia var. platy-
phylla Benth., Australia: WA/Valley of the Giants - near Nornalup, B.J. Mole 486 (NSW 1003687), MN082810*, MN082857*, MN082995*, MN082941*,
MN083037*;Cyanothamnus anemonifolius (A.Cunn.) Duretto & Heslewood subsp. anemonifolius 1, Australia: Vic./Brisbane Ranges, M.J. Bayly 1965
(MELU 105863), KP867712, KP867667, KP867769, –,–;Cyanothamnus anemonifolius (A.Cunn.) Duretto & Heslewood subsp. anemonifolius 2,
Australia: NSW/Wadbilliga NP, S. Rutherford 157 (NSW 971602), MN082812*, MN082859*, MN082997*, MN082943*, MN083038*;Cyanothamnus ane-
monifolius subsp. variabilis (Hook.) Duretto & Heslewood, Australia: Tas./Lobster Falls, M.F. Duretto 3501 (NSW 1005459), MN082811*, MN082858*,
Version of Record 497
TAXON 69 (3) •June 2020: 481–499 Duretto & al. •Boronia is polyphyletic
Appendix 1. Continued.
MN082996*, MN082942*,–;Cyanothamnus baeckeaceus subsp. patulus (Paul G.Wilson) Duretto & Heslewood, Australia: WA/Yilgarn Ranges, E.D. Adams
2/0807 (PERTH 7802293), MN082813*, MN082860*, MN082998*, MN082944*,–;Cyanothamnus coerulescens (F.Muell.) Duretto & Heslewood subsp.
coerulescens 1, Australia: SA/Cox Scrub, M.J. Bayly 1976 (MEL 2383599), KP867743, KP867650, KP867796, MN082945*,–;Cyanothamnus coerulescens
(F.Muell.) Duretto & Heslewood subsp. coerulescens 2, Australia: SA/Dukes Hwy, D.M. Crayn 1038 (NSW 757952), MN082814*, MN082861*, MN082999*,
MN082946*,–;Cyanothamnus defoliatus (F.Muell.) Duretto & Heslewood, Australia: WA/Busselton, A. Young 33 (NSW 1006034), MN082815*,
MN082862*, MN083000*, MN082947*,–;Cyanothamnus fabianoides subsp. roseus (Paul G.Wilson) Duretto & Heslewood, Australia: WA/Tarin Rock
Grain Silo, W.A. Thompson 1377 (PERTH 8359091), MN082816*, MN082863*, MN083001*, MN082948*,–;Cyanothamnus inconspicuus (Benth.) Duretto
& Heslewood, Australia: WA/Yilgarn Ranges, W.A. Thompson 2286 (PERTH 8299668), MN082817*, MN082864*, MN083002*, MN082949*,–;Cya-
nothamnus inflexus (Duretto) Duretto & Heslewood subsp. inflexus, Australia: NSW/Gibraltar Range NP - Surveyor’s Creek Track, R. Johnstone 2716
(NSW 799539), MN082818*, MN082865*, MN083003*, MN082950*,–;Cyanothamnus nanus var. hyssopifolius (Melville) Duretto & Heslewood,
Australia: ACT/cult. ANBG (ex Vic., 12 km N of Bonang), R.O. Makinson 1003 (CANB CBG9107027), MN082819*, MN082866*, MN083004*,
MN082951*,–;Cyanothamnus nanus var. pubescens (Benth.) Duretto & Heslewood, Australia: Vic./Grampians, M.J. Bayly 1989 (MEL 2383591),
KP867745, KP867686, KP867791, MN082952*,–;Cyanothamnus occidentalis (Duretto) Duretto & Heslewood 1, Australia: Qld/Wondul Range NP, A. Orme
1064 (NSW 848488), MN082820*, MN082867*, MN083005*, MN082953*,–;Cyanothamnus occidentalis (Duretto) Duretto & Heslewood 2, Australia:
NSW/Wollar Rd - southwest of Merriwa, C.J. Quinn & N. Lam s.n. (UNSW 23915), KP867714, KP867702, KP867761, MN082954*,–;Cyanothamnus peni-
cillatus (Benth.) Duretto & Heslewood, Australia: WA/Tarin Rock Grain Silo, G. Byrne 1899 (PERTH 7804962), MN082821*, MN082868*, MN083006*,
MN082955*,–;Cyanothamnus polygalifolius (Sm.) Duretto & Heslewood 1, Australia: Qld/Girraween NP, M.T. Mathieson 1263 (BRI AQ0813615),
KP867705, KP867664, KP867808, MN082956*,–;Cyanothamnus polygalifolius (Sm.) Duretto & Heslewood 2, Australia: NSW/cult. ABGMA (ex NSW,
Kattang NR, Camden Head, Perpendicular Point), M.F. Duretto 3051 (NSW 1057660), MN082822*, MN082869*, MN083007*, MN082957*,
MN083039*;Cyanothamnus quadrangulus Duretto & Heslewood 1, Australia: Qld/Mount Maroon, S.P. Phillips 1994 (BRI AQ748158), KP867753,
KP867674, KP867782, MN082958*,–;Cyanothamnus quadrangulus Duretto & Heslewood 2, Australia: Qld/Girraween NP, M.T. Mathieson 383 (BRI
AQ755777), KP867738, KP867670, KP867792, MN082959*,–;Cyanothamnus ramosus Lindl., Australia: WA/22 km east of Eneabba, M.J. Bayly 1911
(MEL 2383601), KP867756, KP867668, KP867800, MN082960*,–;Cyanothamnus rigens (Cheel) Duretto & Heslewood, Australia: NSW/Katoomba,
M.F. Duretto 3069 (NSW 1005418), MN082823*, MN082870*, MN083008*, MN082961*,–;Cyanothamnus tenuis Lindl., Australia: WA, OAK 3RC2
(PERTH 07544596), MN082824*, MN082871*, MN083009*, MN082962*,–;Dinosperma stipitatum (C.T.White & W.D.Francis) T.G.Hartley, Australia:
Qld/SF Reserve 605 - O’Leary Creek crossing, A. Ford 3212 (NSW 673501), MN082825*, MN082872*, MN083010*, MN082963*, MN083040*;Diplolaena
microcephala Bartl., Australia: WA/Sandford Rock NR - 10.5 km by road NE of Westonia, G.T. Chandler 2194 (CANB 602361), MN082826*, MN082873*,
MN083011*, MN082964*,–;Dutaillyea sp., New Caledonia, Munzinger 790 (MO), HG971477, HG971324, HG971277, HG971040, –;Eriostemon austra-
lasius Pers., Australia: NSW/Ku-ring-gai Chase NP, M.F. Duretto 3000 (NSW 1005402), MN082827*, MN082874*, MN083012*, MN082965*, MN083041*;
Euodia hortensis J.R.Forst. & G.Forst., Singapore: cult. Singapore Botanic Gardens, M. Appelhans 398 (US), HG971478, HG971325, HG971168, HG971041,
–;Euodia hylandii T.G.Hartley, Australia: Qld, P.I. Forster 25754 (L), HG971479, HG971326, HG971169, HG971042, –;Euodia montana T.G.Hartley, Papua
New Guinea, James 381 (LAE, BISH, GOET), HG971480, HG971327, HG971170, HG971043, –;Euodia pubifolia T.G.Hartley, Australia: Qld/cult. (ex Qld,
Daintree NP, Noah Creek), Sankowsky 1711 (QRS 121993.1), HG971481, EU493186, EU493243, EU493205, –;Flindersia acuminata C.T.White, Australia:
NSW/cult. RBGS (ex Qld, Windsor Tableland), M.F. Duretto 2999 (NSW 4117113), MN082828*, MN082875*, MN083013*, MN082966*, MN083042*;
Flindersia dissosperma (F.Muell.) Domin, Australia: Qld/4.2 km from Rolleston - Comet road towards Springsure, A.R. Bean 20744 (NSW 728423),
MN082829*, MN082876*, MN083014*, MN082967*,–;Geijera parviflora Lindl., Australia: NSW/cult. ABGMA (ex Qld, Leander Station, northwest of
Longreach), E. Clark s.n. (NSW 4133808), MN082830*, MN082877*, MN083015*, MN082968*, MN083043*;Geijera salicifolia Schott, Australia:
NSW/Central Coast, G. Williams s.n. (UNSW 23768), MN082831*, MN082878*, MN083016*, MN082969*, MN083044*;Halfordia kendack (Montrouz.)
Guillaumin, Australia: Qld/Wilson’s Peak, P.I. Forster 30512 (NSW 920002), MN082832*, MN082879*, MN083017*, MN082970*,–;Halfordia kendack
(Montrouz.) Guillaumin, New Caledonia, M.J. Bayly 2066 (MEL 2383625A), –,–,–,–, JN87112; Leionema ambiens (F.Muell.) Paul G.Wilson, Australia:
NSW/ cult. ABGMA (ex Qld, Darling Downs, near Wallangarra), M.F. Duretto 3039 (NSW 1058543), MN082833*, MN082880*, MN083018*,
MN082971*, MN083045*;Maclurodendron sp., Malaysia: Sabah, John 145743 (L), HG971483, HG971329, HG971289, –,–;Medicosma cunninghamii
(Hook.) Hook.f., Australia: NSW/Myocum, R. Johnstone 3243 (NSW 970747), MN082834*, MN082881*, MN083019*, MN082972*,–;Medicosma fareana
(F.Muell.) T.G.Hartley, Australia: NSW/cult. RBGS (ex Qld, Noah Creek), K. Hill 2095 (NSW 200414), MN082835*, MN082882*, MN083020*,
MN082973*, MN083046*;Medicosma glandulosa T.G.Hartley, Australia: Qld/Cedar Bay NP - Mt Finnigan summit, P.I. Forster 25045 (L), HG971484,
HG971330, HG971172, HG971045, –;Medicosma riparia (P.Royen) T.G.Hartley, Australia: Qld/Moreton Telegraph Station, P.I. Forster 32536 (NSW
812557), MN082836*, MN082883*, MN083021*, MN082974*,–;Melicope elleryana (F.Muell.) T.G.Hartley, Australia: Qld, Lorence 6602 (PTBG),
HG002551, EU493184, EU493241, EU493203, –;Melicope elleryana (F.Muell.) T.G.Hartley, Australia: Qld, P.I. Forster 34003, (MEL 2324754A), –,–,–,
–, JN987118; Melicope inopinata J.Florence, French Polynesia: Marquesas Islands/Nuku Hiva, Meyer 887 (not provided with data), HG002564, EU493176,
EU493233, EU493195, –;Melicope obscura (Cordem.) T.G.Hartley, La Réunion, Larsen OBS1 (C) HG971547, HG971395, HG971225, HG971096, –;Meli-
cope rostrata (Hillebr.) Appelhans, K.R.Wood & W.L.Wagner, U.S.A.: Hawaiian Islands/Kaua‘i, Wood 8223 (PTBG), HG002647, EU493181, EU493238,
EU493200, –;Melicope spathulata A.Gray, U.S.A.: Hawaiian Islands/Kaua‘i, Wood 14213 (PTBG), HG002650, HG002508, HG002939, HG002752, –;Meli-
cope stellulata T.G.Hartley, Papua New Guinea, M. Appelhans 427 (LAE, US), HG971582, HG971429, HG971255, HG971127, –;Melicope ternata J.R.Forst.
& G.Forst., Germany: cult. Göttingen Botanic Garden, M. Appelhans 487 (GOET), HG971585, HG971432, HG971258, HG971130, –;Melicope vitiflora
(F.Muell.) T.G.Hartley 1, Papua New Guinea, M. Appelhans 433 (LAE, US), HG971592, HG971439, HG971265, HG971134, –;Melicope vitiflora (F.Muell.)
T.G.Hartley 2, Australia: Qld/Mt Lewis FR, P.I. Forster 29363 (L) HG002633, HG002492, HG002851/HG002927, HG002741, –;Microcybe multiflora Turcz.
subsp. multiflora, Australia: WA/c. 4 km W of Southern Cross, B.J. Mole 386 (NSW 1003597), MN082837*, MN082884*, MN083022*, MN082975*,–;
Muiriantha hassellii (F.Muell.) C.A.Gardner, Australia: WA/Stirling Range, B.J. Mole 471 (MEL 1003670), MN082838*, AY631911, MN083023*,
MN082976*, MN083047*;Myrtopsis myrtoidea (Baill.) Guillaumin, New Caledonia, McPherson 18026 (MO), –, HG971441, HG971283, HG971136, –;
Nematolepis phebalioides Turcz., Australia: WA/Hopetoun, B.J. Mole 436 (NSW 1003641), MN082839*, AY631910, MN083024*, MN082977*,
MN083048*;Neobyrnesia suberosa J.A.Armstr. 1, Australia: NT/Kakadu NP, R. Mueller s.n. (CANB CBG8316286), –, EU281855, EU281921, –,–;Neobyr-
nesia suberosa J.A.Armstr. 2, Australia: NT/Kakadu NP, M.J. Bayly 1904 (MEL2383567), KP867718, KP867655, KP188925, –, JN987125; Neoschmidia pal-
lida (Schltr.) T.G.Hartley, New Caledonia: Provincer Sud/Mt Dore, M.F. Duretto 1405 (HO 561944), MN082840*, MN082885*, MN083025*, MN082978*,
MN083049*;Pentaceras australe (F.Muell.) Benth., Australia: NSW/cult. ABGMA (ex NSW, Uki), L.J. Forlonge s.n. (NSW 4051097), MN082841*,
MN082886*, MN083026*, MN082979*, MN083050*;Perryodendron parviflorum (C.T.White) T.G.Hartley 1, Papua New Guinea, Pullen 7313
(US) HG971595, HG971443, HG971267, HG971138, –;Perryodendron parviflorum (C.T.White) T.G.Hartley 2, Indonesia: Irian Jaya/N of Ayawasi - Birds
Head Peninsula, A.M. Polak 1297 (CANB 532580), MN082842*, MN082887*, MN083027*, MN082980*,–;Phebalium canaliculatum (F.Muell. & Tate)
J.H.Willis, Australia: WA, M.J. Bayly 2457 (MELU), MN082843*, MN082888*, MN083028*, MN082981*, MN083051*;Philotheca spicata (A.Rich.) Paul
G.Wilson, Australia: WA/Coorow–Greenhead Rd, B.J. Mole 339 (NSW 1003544), MN082844*, MN082889*, MN083029*, MN082982*, MN083052*;Phi-
lotheca virgata (Hook.f.) Paul G.Wilson, Australia: Tas./Freycinet NP, M.F. Duretto 3519 (NSW 1005478), MN082845*, MN082890*, MN083030*,
MN082983*,–;Picrella ignambiensis (Guillaumin) T.G.Hartley & Mabb., New Caledonia, McPherson 19132 (MO), HG971599, HG971446, HG971284/
HG971302, HG971142, –;Pitaviaster haplophyllus (F.Muell.) T.G.Hartley, Australia: Qld/end of Shell Pocket road - El Arish., A. Ford 4821 (L),
498 Version of Record
Duretto & al. •Boronia is polyphyletic TAXON 69 (3) •June 2020: 481–499
Appendix 1. Continued.
HG971600, HG971447, HG971270, HG971143, –;Rhadinothamnus euphemiae (F.Muell.) Paul G.Wilson, Australia: WA/Cape Le Grand NP - Summit of Mt
Le Grand, B.J. Mole 424 (NSW 1003630), MN082846*, AY631912, MN083031*, MN082984*, MN083053*;Sarcomelicope follicularis T.G.Hartley, New
Caledonia, Munzinger 668 (MO), HG971601, HG971448, HG971303, HG971144, –;Tetractomia tetrandra (Roxb.) Merr., Borneo, Beaman 8917 (L),
HG971602, HG971449, HG971271, HG971145, –;Zieria arborescens Sims subsp. arborescens, Australia: Vic./Noojee SF, M.J. Bayly 1868 (MELU
120871), KP867748, KP867675, KP188949, –, JN987143; Zieria fraseri subsp. robusta Duretto & P.I.Forst., Australia: Qld/Carnarvon NP (Salvator Rosa),
M.B. Thomas 3739 (BRI AQ830008), KP867710, KP867698, KP188932, –,–;Zieria minutiflora Domin subsp. minutiflora, Australia: Qld/Glasshouse Moun-
tains NP - Mount Coonowrin, P.I. Forster 34313 (BRI AQ862319), KP867711, KP867690, KP188919, –,–;Zieria oreocena J.A.Armstr., Australia: Vic./Gram-
pians NP, M.J. Bayly 1982 (MEL 2383595), KP867716, KP867703, KP188904, –,–;Zieria smithii Jacks., Australia: NSW/Border Ranges NP - Bar Mountain,
M.A.M. Renner 6760 (NSW 876659), MN082847*, MN082891*, MN083032*, MN082985*,–;Zieria southwellii J.A.Armstr., Australia: NSW/Gibraltar
Range NP, M.A.M. Renner 6730 (NSW 876703), MN082848*, MN082892*, MN083033*, MN082986*,–;Zieria veronicea (F.Muell.) Benth., Australia:
SA/Cox Scrub, M.J. Bayly 1980 (MEL 2383600), KP867707, KP867676, KP188929, –,–.
Version of Record 499
TAXON 69 (3) •June 2020: 481–499 Duretto & al. •Boronia is polyphyletic
