Content uploaded by Bert W Hoeksema
Author content
All content in this area was uploaded by Bert W Hoeksema on Oct 01, 2016
Content may be subject to copyright.
Contributions to Zoology, 85 (4) 387-422 (2016)
When forms meet genes: revision of the scleractinian genera Micromussa and Homophyllia
(Lobophylliidae) with a description of two new species and one new genus
Rober to Arrigoni1, 2, * Francesca Benzoni2, 3, 15, * Danwei Huang4, 5, Hironobu Fukami6, Chaolun Allen Chen7, 8, Michael
L. Berumen1, Mia Hoogenboom9, Damian P. Thomson10, Bert W. Hoeksema11, Ann F. Budd5, Yuna Zayasu12, Tullia
I. Te r rane o1, Yuko F. Kitano13, Andrew H. Baird14
1 Red Sea Research Center, Division of Biological and Environmental Science and Engineering, King Abdullah
University of Science and Technology, Thuwal 23955- 69 0 0, Saudi Arabia
2 Department of Biotechnologies and Biosciences, University of Milano – Bicocca, Piazza della Scienza 2, 20126,
Milan, Italy
3 UMR ENTROPIE (IRD, Université de La Réunion, CNRS), Laboratoire d’excellence-CORAIL, centre IRD de
Nouméa, New Caledonia, 101 Promenade Roger Laroque, BP A5, 98848 Noumea Cedex, New Caledonia
4 Department of Biological Sciences and Tropical Marine Science Institute, National University of Singapore,
Singapore 117543, Singapore
5 Department of Earth and Environmental Sciences, University of Iowa, Iowa City, IA, 52242, USA
6 Faculty of Agriculture, University of Miyazaki, 1-1 Gakuenkibanadai-Nishi, Miyazaki, 889-2192, Japan
7 Biodiversity Research Centre, Academia Sinica, Nangang, Taipei 115, Taiwa n
8 Institute of Oceanography, National Taiwan University, Taipei 106, Taiwan
9 College of Marine and Environmental Science and ARC Centre of Excellence for Coral Reef Studies, James Cook
University, Townsville, QLD 4811, Australia
10 CSIRO Oceans and Atmosphere, Floreat, WA 6 014, Australia
11 Naturalis Biodiversity Center, PO Box 9517, 2300 RA Leiden, The Netherlands
12 Marine Genomics Unit, Okinawa Institute of Science and Technology Graduate University, 1919-1 Tancha,
Onna-son, Okinawa, 904-0495, Japan
13 Organization for Promotion of Tenure Track, University of Miyazaki, 1-1 Gakuenkibanadai-Nishi, Miyazaki,
889-2192, Japan
14 ARC Centre of Excellence for Coral Reef Studies, James Cook University, Townsville, QL D 4811, Australia
15 E-mail: francesca.benzoni@unimib.it
* These authors contributed equally to this work
Key words: coral, evolution, phylogeny, systematics, taxonomy
Abstract
The scleractinian family Lobophylliidae is undergoing a major
taxonomic revision thanks to the combination of molecular and
morphological data. In this study, we investigate the evolution-
ary relationships and the macro- and micromorphology of six
nominal coral species belonging to two of the nine molecular
clades of the Lobophylliidae, clades A and B, and of Symphyllia
wilsoni, a lobophylliid species analyzed from a molecular point
of view for the rst time. Sequence data from mitochondrial
DNA (COI and the intergenic spacer between COI and l-rRNA),
and nuclear DNA (histone H3 and ITS region) are used to gen-
erate robust molecular phylogenies and a median-joining haplo-
type network. Molecular results are strongly in agreement with
detailed observations of gross- and ne-scale morphology of
skeletons, leading to the formal revision of the genera Micro-
mussa and Homophyllia and the description of two newly dis-
covered zooxanthellate shallow-water species, Micromussa
pacica sp. nov. Benzoni & Ar rigoni and Micromussa indiana
sp. nov. Benzoni & Arrigoni, and a new genus, Australophyllia
gen. nov. Benzoni & Arrigoni. In par ticular, Acanthastrea lord-
howensis and Montastraea multipunctata are moved into Mi-
cromussa, A. hillae is synonymized with A. bowerbanki and is
transferred to Homophyllia, and a revised diagnosis for both
genera is provided. Micromussa pacica sp. nov. is described
from the Gambier Islands with its distribution spanning New
Caledonia and eastern Australia. Despite a supercial resem-
blance with Homophyllia australis, it has distinctive macro-
and micromorphological septal features. Micromussa indiana
sp. nov., previously identied as M. amakusensis, is here de-
scribed from the Gulf of Aden and the southern Red Sea as a
distinct species that is genetically separated from M. amakuse-
nsis and is mor phologically distinct from the latter due to its
smaller corallite size and lower number of septa. Finally, mo-
lecular trees show that S. wilsoni is closely related, but molecu-
larly separated from clades A and B, and, also based on a unique
388 Arrigoni et al. – Phylogeny of Micromussa and Homophyllia
combination of corallite and sub-corallite characters, the spe-
cies is moved into Australophyllia gen. nov. These ndings con-
rm the need for using both genetic and morphological datasets
for the ongoing taxonomic revision of scleractinian corals.
Contents
Introduction ................................................................................... 388
Clade A ..................................................................................... 390
Clade B ..................................................................................... 390
Symphyllia wilsoni ...................................................................... 390
Material and methods .................................................................. 39 2
Coralsamplingandidentication ..................................... 392
DNA preparation,amplication, and sequence
analyses .................................................................................... 392
Morphological analyses ....................................................... 394
Results ............................................................................................. 394
Molecular phylogenetic and haplotype network
analyses .................................................................................... 394
Macromorphology ................................................................. 397
Micromorphology .................................................................. 39 9
Discussion ...................................................................................... 401
Clade A ..................................................................................... 401
Clade B ..................................................................................... 402
Clade J ...................................................................................... 409
Taxa outside clades A, B, and J .......................................... 410
Final remarks ........................................................................... 411
Acknowledgements ...................................................................... 412
References ...................................................................................... 412
Appendix ......................................................................................... 417
Introduction
In the last decade the increasing use of molecular tools
and novel morphological analyses have shed new light
on the evolution and systematics of scleractinian corals
(Stolarski, 2003; Fukami et al., 2004a, 2008; Wallace
et al., 2007; Budd and Stolarski, 2009, 2011; Gitten-
berger et al., 2011; Stolarski et al., 2011; Huang et al.,
2011; Kitano et al., 2014; Kitahara et al., 2016). The
integration of genetics and morphology has provided
new hypotheses about the evolutionary history of scle-
ractinian corals, and has led to a revolution in their
taxonomy at all ranks (Stolarski and Janiszewska,
20 01; Budd et al., 2010, 2012; Stolarski et al., 2011;
Benzoni et al., 2012a, b; Kitahara et al., 2012b; Huang
et al., 2014a).
The family Lobophylliidae Dai & Horng, 2009 is
an ecologically important group in tropical Indo-Pa-
cic coral reefs (Veron and Pichon, 1980; Scheer and
Pillai, 1983; Veron, 1993, 2000). Based on morpho-
logical characters, it currently comprises 12 extant
genera and 52 zooxanthellate species and is widely
distributed from the Red Sea and eastern Africa to
French Polynesia (Veron, 2000; Dai and Horng, 2009;
Budd et al., 2012; Benzoni, 2013; Hoogenboom et al.,
2015). To date, based on mitochondrial and nuclear
phylogenies, this taxon corresponds to a monophyletic
lineage consisting of nine main genus-level molecular
clades, clades A to I sensu Arrigoni et al. (2014a),
which are mostly in disagreement with the taxonomy
of lobophylliids that until recently has been based
solely on macromorphology (Arrigoni et al., 2012,
2014a). Indeed, all polytypic genera described in Ve-
ron (2000) analyzed so far, i.e. Acanthastrea Milne
Edwards & Haime, 1848, Echinophyllia Klunzinger,
1879, Lobophyllia de Blainville, 1830, Micromussa
Veron, 2000, Oxypora Saville Kent, 1871, and Sym-
phyllia Milne Edwards & Haime, 1848, are not mono-
phyletic (Arrigoni et al., 2014a). Furthermore, the phy-
logenetic position and taxonomy of two monospecic
genera, Echinomorpha Veron, 2000 and Homophyllia
Brüggemann, 1877, were still uncertain (Budd et al.,
2012) because they have previously not been investi-
gated at a molecular level, while the monotypic genus
Moseleya has already been molecularly characterized
(Huang et al., 2011).
Recent work has integrated molecular ndings with
novel macro- and micromorphological skeleton data,
leading to the identication of a new set of informative
characters useful for revising the family taxonomy and
systematics (Budd and Stolarski, 2009; Budd et al.,
2012; Arrigoni et al., 2014b, 2015). In particular, mi-
cromorphological characters such as the height, spac-
ing, and shape of septal teeth; the distribution and
shape of granules on septal faces; and the structure of
the area between teeth (Budd and Stolarski, 2009,
2011) are now known to be informative and diagnostic
for the Lobophylliidae (Budd and Stolarski, 2009;
Budd et al., 2012; Arrigoni et al., 2014b, 2015) as in
other coral families (Hoeksema, 1989; Benzoni et al.,
2007, 2012a; Gittenberger et al., 2011; Budd and Sto-
larski, 2011; Janiszewska et al., 2011, 2013, 2015; Budd
et al., 2012; Schmidt-Roach et al., 2014; Huang et al.,
2014a, b). Therefore, the family Lobophylliidae is cur-
rently undergoing a revision that started with the phy-
logenetic re-classication of the genera Australomussa
Veron, 1985, Parascolymia Wells, 1964, and Sclero-
phyllia Klunzinger, 1879, as a result of the aforemen-
tioned integrated morpho-molecular approach (Arrig-
oni et al., 2014b, 2015).
In a recent study, unexpected genetic afnities were
found between the Lobophylliidae species included in
the sister clades A and B sensu Arrigoni et al. (2014a).
389Contributions to Zoology, 85 (4) – 2016
Fig. 1. Lobophylliidae included in this
study. A) Holotype of Micromussa am-
akusensis MTQ G32485 from Japan; B)
Micromussa indiana sp. nov. (referred to
as Micromussa cf. amakusensis in text)
from the Gulf of Aden MNHN-IK-
2012-14232; C) Holotype of Acanthas-
trea lordhowensis MTQ G57483 from
Australia; D) Micromussa multipuncta-
ta (previously Montastraea) RMNH
Coel 40090 from Malaysia; E) Micro-
mussa pa cica sp. nov. (referred to as
Homophyllia cf. australis in t ext)
MNHN IK-2012-16046 from New Cale-
donia; F) Homophyllia australis IRD
HS3524 from New Caledonia; G) Homo-
phyllia bowerbanki (previously Acan-
thastrea) IRD HS3287 from New Cale-
donia; cf.H) Homophyllia bowerbanki
(previously Acanthastrea hillae) AM
MH043 from Australia; I) Australophyl-
lia wilsoni (previously Symphyllia) from
Australia J) Acanthastrea cf. hemprichii
(referred in the text as Acanthastrea cf.
hillae) UNIMIB BA115 from the Gulf of
Aden.
390 Arrigoni et al. – Phylogeny of Micromussa and Homophyllia
Fig. 2. Lobophylliidae included in this
study in situ. A) Micromussa amakusensis
from Japan; B) Micromussa indiana sp.
nov. (referred to as Micromussa cf. amaku-
sensis in text) same specimen shown in 1B;
C) Micromussa lordhowensis (previously
Acanthastrea) from Lord Howe, Australia;
D) Micromussa multipunctata (previously
Montastraea) same specimen shown in
1D; E) Micromussapacicasp. nov. (re-
ferred to as Homophyllia cf. australis in
text), same specimen shown in 1F; F)
Homophyllia australis, same specimen
shown in 1E; G) Homophyllia bowerbanki
(previously Acanthastrea), same specimen
shown in 1I; H) Homophyllia bowerbanki
(previously A. hillae) IRD HS3066 from
New Caledonia; I) Australophyllia wilsoni
from Australia; J) Acanthastrea cf. hem-
prichii (referred in the text as A. cf. hillae)
same specimen shown in 1H.
391Contributions to Zoology, 85 (4) – 2016
Clade A
Clade A is composed of Micromussa amakusensis
(Veron, 1990) (Figs 1A, 2A) and Montastraea multi-
punctata (Hodgson, 1985) (Figs 1D, 2D). The latter
was formally assigned to the Lobophylliidae but its
genetic placement represents a taxonomic issue that
calls for a formal revision because it is not related to
the type species Montastraea cavernosa (Linnaeus,
1767) (Huang et al., 2011; Arrigoni et al., 2 014 a).
Nevertheless, no detailed morphological studies have,
to date, been conducted on this species. The poorly
studied type species Micromussa amakusensis has
been recorded throughout the Indo-Pacic, from the
Gulf of Aden to Central Indo-Pacic and West Pa-
cic (Veron, 1990, 1992, 1993, 2000; Wallace et al.,
2009; Pichon et al., 2010; Arrigoni et al., 2012; Arri-
goni et al., 2014a). In the present study, material from
the Indian Ocean (Figs 1B, 2B) and Pacic Ocean
(Figs 1A, 2A) populations was examined for the rst
time through a morpho-molecular approach. In his
description of M. amakusensis from Japan, Veron
(1990) highlighted that, on the basis of macromor-
phology and in-situ appearance, this species resem-
bles Acanthastrea lordhowensis Veron & Pichon,
1982 (Figs 1C, 2C) although it has larger and less
regular corallites and more septa. Nevertheless, Ve-
ron (2000) erected the genus Micromussa to include
species previously ascribed to Acanthastrea with
corallites less than 5 mm in diameter, thus excluding
A. lordhowensis. Based on this criterion, Veron
(2000) moved Acanthastrea minuta Moll & Borel-
Best, 1984 into the genus Micromussa and also de-
scribed M. diminuta Veron, 2000 from Sri Lanka.
Although no genetic or micromorphological data is
currently available for these two species, the type
specimens were examined in this study and their tax-
onomic position discussed.
Clade B
Clade B sensu Arrigoni et al. (2014a) currently con-
tains Acanthastrea bowerbanki Milne Edwards &
Haime, 1857 (Figs 1G, 2G) and A. hillae Wells, 1955
(Figs 1H, 2H). At a molecular level, A. bowerbanki and
A. hillae are not closely related to the type species A.
echinata (Dana, 1846), which is recovered in clade E
sensu Arrigoni et al. (2014a), and the establishment of
a new genus to accommodate species in clade B is thus
tentative. The sister-relationship between the two spe-
cies is corroborated also by remarkable morphological
similarities (Chevalier, 1975; Veron and Pichon, 1980;
Veron, 1992, 2000; Wallace et al., 2009). Their coralla
are similar in growth form and mode of budding, while
their corallites are among the largest of all species cur-
rently assigned to Acanthastrea (Veron and Pichon,
1980; Veron, 2000). Furthermore, A. bowerbanki and
A. hillae show a partially overlapping geographic dis-
tribution, mainly in the Central Pacic (Veron and Pi-
chon, 1980; Veron, 1993, 2000; Wallace et al., 2009),
where they are generally uncommon throughout the
tropics but more abundant in high latitude and mar-
ginal reef localities (Veron, 1993). Acanthastrea hillae
has been also reported from the Western Indian Ocean
but these records are doubtful (Veron, 2000). For this
reason, material from Yemen assigned to this species
based on macromorphology was also included in this
study (Figs 1J, 2J).
The monotypic genus Homophyllia has a compli-
cated nomenclatural history (Vaughan and Wells,
1943; Wells, 1964; Veron and Pichon, 1980; Veron,
2000; Budd et al., 2012). Previously considered a jun-
ior synonym of Lobophyllia (Matthai, 1928; Vaughan
and Wells, 1943) and Scolymia (Veron and Pichon,
1980; Veron, 2000), it was restored by Budd et al.
(2012) following novel morphological evidence pro-
posed by Budd and Stolarski (2009). Budd et al. (2012)
showed that the micromorphology and the microstruc-
ture of the monostomatous species Homophyllia aus-
tralis (Milne Edwards & Haime, 1849) (Figs 1F, 2F)
are clearly distinguished from those of the solitary
species Scolymia lacera (Pallas, 1766) and Parascoly-
mia vitiensis (Brüggemann, 1877) (Budd and Stolar-
ski, 2009; Budd et al., 2012; Arrigoni et al., 2014b).
Nevertheless, despite the increasing amount of genetic
data concerning the Lobophylliidae (Arrigoni et al.,
2014a, 2014b, 2015, 2016), no molecular information is
available for H. australis. In this study we analyzed a
large collection of specimens encompassing the whole
range of morphological variability shown by Veron
and Pichon (1980), thus including specimens with typ-
ical morphology (Budd and Stolarski, 2009: g 2K;
Veron and Pichon, 1980: gs 408, 410, 420; Figs 1F,
2F) and other specimens with thinner and less numer-
ous septa identied as H. cf. australis (Veron and Pi-
chon, 1980: gs 412, 424; Figs 1E, 2E).
Symphyllia wilsoni
Symphyllia wilsoni Veron, 1985 (Figs 1I, 2I) is a rela-
tively poorly known species, which is distributed
from south to southwestern Australia (Veron, 2000)
392 Arrigoni et al. – Phylogeny of Micromussa and Homophyllia
and as
cribed to the genus Symphyllia by reason of its
massive coralla with large meandering valleys. How-
ever, as remarked by Veron (2000), this species pre-
sents some odd macromorphological characters, such
as small skeletal features compared to those of its con-
geners and septa resembling those of the genus Acan-
thastrea. Although several Symphyllia representatives,
including the type species, were included in molecular
analyses and found in clade I sensu Arrigoni et al.
(2014a), S. wilsoni has previously not been examined
and is therefore included in our analyses in order to
ascertain its status.
The present study aims to provide a robust molecu-
lar phylogeny reconstruction of six nominal species
included in clades A and B, and of S. wilsoni. We ana-
lyze material from different geographic localities
spanning the known range of intraspecic morpho-
logical variability and several types of sequence data:
the mitochondrial DNA regions cytochrome c oxidase
subunit I (COI) as well as the non-coding intergenic
spacer region between COI and large ribosomal RNA
subunit, and the nuclear markers ribosomal internal
transcribed spacers 1 and 2 (ITS region) and the his-
tone H3 gene. Furthermore, we investigate gross- and
ne-scale morphology of skeletons of each coral spe-
cies, including material from each type locality, except
for M. multipunctata, which was sampled from the
northernmost tip of Borneo (Waheed et al., 2 015).
Material and methods
Coralsamplingandidentication
A total of 87 coral specimens ascribed to Acanthas-
trea bowerbanki, A. echinata, A. hillae, A. lordhowen-
sis, Homophyllia australis, Micromussa amakusensis,
Montastraea multipunctata, and Symphyllia wilsoni,
were collected while SCUBA diving between 1 and 35
m depth from different localities in the Indian and Pa-
cic Ocean and analyzed from both molecular and
morphological point of view (S1). Furthermore, the
lectotype of M. diminuta and the holotype of A. minu-
ta were used in the morphological analysis. Samples of
A. bowerbanki, A. hillae, A. lordhowensis, H. austra-
lis, and S. wilsoni were sampled from their type local-
ity, Australia, as well as M. amakusensis from Japan.
Each coral sample was photographed underwater and
then collected, tagged, and approximately 1 cm2 of the
entire specimen was broken off and put in CHAOS so-
lution (not an acronym; 4 M guanidine thiocyanate,
0.1% N-lauroyl sarcosine sodium, 10 mM Tris pH 8,
0.1 M 2-mercaptoethanol) to dissolve the tissue or
xed in 95% ethanol for further molecular analyses.
The remaining corallum was immersed in sodium hy-
pochlorite for 48 hours to remove all tissues, rinsed in
freshwater and air-dried for identication and micro-
scope observations (Figs 1, 2). Specimens were identi-
ed at species level based on their morphological
structures following Milne Edwards and Haime
(1848), Wells (1964), Veron and Pichon (1980, 1982),
Hodgson (1985), Veron (1985, 1990, 2000), Wallace et
al. (2009), and Pichon et al. (2010), as well as referring
to illustrations of holotypes in their original descrip-
tions. Voucher samples were deposited at the Univer-
sity of Milano-Bicocca (UNIMIB, Milano, Italy), the
Australian Museum (AM, Sydney, Australia), the In-
stitut de Recherche pour le Développement (IRD,
Nouméa, New Caledonia), the Seto Marine Biological
Laboratory at the Kyoto University (SMBL, Kyoto, Ja-
pan), the University of Miyazaki (Miyazaki, Japan),
Naturalis Biodiversity Center (RMNH, Leiden, the
Netherlands), and the Museum of Tropical Queensland
(MTQ, Townsville, Australia).
DNA preparation,amplication, and sequence
analyses
Total genomic DNA was extracted using the DNeasy
Blood and Tissue kit (Qiagen Inc., Hilden, Germany)
from coral tissue preserved in ethanol or using a phe-
nol-chloroform-based method with a phenol extraction
buffer (100 mM Tris-Cl pH 8, 10 mM EDTA, 0.1%
SDS) from specimens conserved in CHAOS solution
(Fukami et al., 2004a; Huang et al., 2011). For all cor-
al samples, we amplied and directly sequenced one
mitochondrial marker (COI) and two nuclear markers
(histone H3 and ITS region). The COI gene, histone
H3, and ITS region were amplied using the primer
pairs MCOIF and MCOIR and the protocol proposed
by Benzoni et al. (2011), H3F and H3R (Colgan et al.,
1998), ITS4 (White et al., 1990) and A18S (Taka-
bayashi et al., 1998) and the protocol published by
Benzoni et al. (2011), respectively. Furthermore, for
species included in clade A sensu Arrigoni et al.
(2014a), IGR was amplied using MNC1F and MN-
C1R primers (Fukami et al., 2004b; Huang et al.,
2009) and the protocol described by Arrigoni et al.
(2016). All PCR products were puried with Illustra
ExoStar (GE Healthcare, Buckinghamshire, UK) and
directly sequenced in both strands using an ABI 3130xl
Genetic Analyzer (Applied Biosystems, Carlsbad, CA,
393Contributions to Zoology, 85 (4) – 2016
Fig. 3. Bayesian phylogeny reconstr uc-
tion of the family Lobophylliidae for the
analysis of the concatenated data set of
COI, histone H3, and ITS region. Num-
bers above branches indicate nodal sup-
port by means of Bayesian poster ior
probabilities (> 0.8), Maximum Likeli-
hood SH-like support (> 0.7), and Maxi-
mum Parsimony bootstrap support (>
50). Lower values of support not shown.
Clades within Lobophylliidae are col-
oured and labelled A to I according to
Arrigoni et al. (2014a). Specimens ana-
lyzed in this study are in bold.
394 Arrigoni et al. – Phylogeny of Micromussa and Homophyllia
USA). Chromatograms were manually corrected for
misreads, if necessary, and forward and reverse strands
were merged into one sequence le using CodonCode
Aligner 3.6.1 (CodonCode Corporation, Dedham,
MA, USA). In particular, chromatograms of products
obtained with ITS4 and A18S primers did not show
any intra-individual polymorphisms or double peaks,
thereby allowing direct sequencing of this region. All
newly obtained sequences were deposited in EMBL,
and accession numbers are listed in S1.
Sequence alignments were generated using the E-
INS-i option in MAFFT 7.130b (Katoh et al., 2002;
Katoh and Standley, 2013) under default parameters.
For phylogenetic analyses, sequences of COI, histone
H3, and ITS region were concatenated into one parti-
tioned dataset. Three methods, maximum parsimony
(MP), maximum likelihood (ML), and Bayesian infer-
ence (BI), were employed to reconstruct the phyloge-
netic relationships within the Lobophylliidae. For MP
analysis, tree searches were generated in PAUP*
4.0b10 (Swofford, 2003) using heuristic searches with
10000 random additions. Branch support was estimat-
ed with the bootstrap condence levels using 1000
replicates. Prior to the model-based phylogenetic anal-
yses, the best-t model of nucleotide substitution was
identied for each gene partition separately by means
of the Akaike Information Criterion calculated with
MrModeltest 2.3 (Nylander, 2004). The following sub-
stitution models were suggested: the GTR + I + G for
ITS region, the HKY + G + I for COI, and the K80 + I
for histone H3. ML topologies were calculated with
PhyML (Guindon and Gascuel, 2003) and relative sup-
port for individual clades was estimated using the Shi-
modaira and Hasegawa (SH-like) test. BI analysis was
performed employing MrBayes 3.1.2 (Huelsenbeck
and Ronquist, 2001). Two simultaneous runs of four
Markov Monte Carlo chains were conducted for 3 x
107 generations, sampling every 100 generations to en-
sure independence of the successive samples. Results
were analyzed for stationarity and convergence using
Tracer 1.6 (Rambaut and Drummond, 2009), with a
burn-in of 25% of sampled generations. Additionally,
ML phylogenetic reconstructions for each separate
COI, histone H3, and ITS region were obtained using
PhyML (Guindon and Gascuel, 2003) under the substi-
tution models proposed by MrModeltest 2.3 (Ny-
lander, 2004). The SH-like test replicates was per-
formed to assess the branch support of ML trees.
Within clade A sensu Arrigoni et al. (2014a), Network
4.6.1.2 (http://www.uxus-technology.com) was used
to construct a median-joining haplotype network
(Bandelt et al., 1999) for the IGR dataset. This method
is especially applicable to non-recombinant DNA se-
quences, such as mitochondrial DNA, and combines
all minimum spanning trees into a single network.
Alignment was converted to the Roehl format using
DnaSP (Librado and Rozas, 2009), invariable sites
were removed and sites with gaps were not considered.
Morphological analyses
Scleractinian coral skeletons of the sequenced lobo-
phylliids were analyzed both at macro- and micromor-
phological levels using light microscopy and Scanning
Electron Microscopy (SEM), respectively, in order to
nd morphological characters supporting the molecu-
lar phylogeny reconstructions. Images of coral skele-
tons were taken with a Canon G5 digital camera as well
as with a Leica M80 microscope equipped with a Leica
IC80HD camera. For scanning electron microscope
(SEM) imaging, skeleton fragments were ground to
produce a at edge, mounted on stubs using silver glue,
sputter-coated with conductive gold lm, and exam-
ined using a Vega Tescan Scanning Electron Micro-
scope at the University of Milano-Bicocca. At least ve
different corallites per species were examined at mi-
cromorphological level. For a glossary of skeletal terms
we followed Budd et al. (2012) and we also adopted
their character names, ID numbers (in brackets), and
state names. In addition to samples deposited in the in-
stitutions mentioned earlier, we analyzed specimens
and type material from the Environment Protection
Authority, Sana’a and Socotra, Yemen (EPA S), the
Muséum National d’Histoire Naturelle (MNHN, Paris,
France), the Natural History Museum (NHMK, Lon-
don, UK, formerly British Museum of Natural History,
BMNH), the Queensland Museum (QM, Brisbane,
Australia), the Marine Science Institute, University of
the Philippines (UP, Manila, the Philippines), and the
Western Australian Museum (WAM, Perth, Australia).
Results
Molecular phylogenetic and haplotype network
analyses
New sequence data of COI, histone H3, and ITS re-
gion, generated in this study from 87 coral samples
representing 11 species, were combined with published
sequences of the families Lobophylliidae, Merulini-
dae Verrill, 1865, Diploastraeidae Chevalier & Beau-
395Contributions to Zoology, 85 (4) – 2016
vais, 1987, and Montastraeidae Yabe & Sugiyama,
1941, resulting in an alignment composed of 50 nomi-
nal species. Plesiastrea versipora was selected as out-
group because of its divergence from the Lobophyllii-
dae, Merulinidae, Diploastraeidae, and Montastraei-
dae (Fukami et al., 2008; Benzoni et al., 2011; Huang
et al., 2011; Budd et al., 2012). The nal concatenated
dataset of aligned sequences of the three molecular
fragments had a total length of 1939 bp (COI: 580 bp,
histone H3: 318 bp, ITS region: 1041 bp). The ITS re-
gion was the most variable, with 191 variable sites (148
positions parsimony-informative PI), the COI gene
fragment showed 87 bp variable sites (57 positions PI),
and the histone H3 sequences featured 90 bp variable
sites (84 positions PI).
The three single gene trees did not show any sup-
ported topological conicts, although the resolution
differed notably among the three topologies (Figs S3-
S5). Nevertheless, each of the analyzed specimens be-
longed to the same molecular clade in all of the three
phylogenetic reconstructions. Bayesian, maximum
likelihood, and maximum parsimony topologies were
highly concordant and node support values were high
across the ingroup and outgroup.
The phylogram based on the concatenated (COI,
histone H3, and ITS region) molecular datasets was
broadly consistent with previously published phyloge-
ny reconstructions (Huang et al., 2011; Arrigoni et al.,
2014a, 2014b, 2015), conrming the Lobophylliidae
and Merulinidae as monophyletic taxa (Fig. 3). All of
the nine main genus-level lineages proposed by Arri-
goni et al. (2014a) for the Lobophylliidae were highly
supported by all methods of phylogeny reconstruction.
Clades A and B were sister taxa of each other. Both
were well resolved and strongly supported (Bayesian
posterior probability score Pp = 1, ML SH-like support
Ss = 1, MP bootstrapping support Bs = 99 for both
clades). A close-up of the phylogenetic relationships
among and within clades A and B is shown in Fig. S6.
Clade A contained three nominal species: the type spe-
cies of the genus Micromussa, M. amakusensis, along
with Montastraea multipunctata and A. lordhowensis.
The monophyly of the latter two species was highly
supported whereas M. amakusensis was split into two
main lineages. In particular, the colonies of M. amaku-
sensis from the type locality Japan were grouped to-
gether and sister to Montastraea multipunctata, with
the exception of the uncertain position of one sample
SMBL Cni-11051. In contrast, specimens of M. cf. am-
akusensis from Yemen formed a monophyletic lineage
with strong support (Pp = 1, Ss = 1, Bs = 95) that was
sister to the group containing M. amakusensis from Ja-
pan and Montastraea multipunctata. Within clade A,
we found a well-supported basal group (Pp = 1, Ss = 1,
Bs = 99) composed of all of the specimens identied as
Homophyllia cf. australis (Figs 1E, 2E). This lineage
was not closely related to the one including the speci-
mens of H. australis that showed the typical morphol-
ogy (Figs 1F, 2F) which formed a well-supported group
(Pp = 0.9, Ss = 0.89, Bs = 83) within clade B. Homo-
phyllia australis was instead sister to the well-support-
ed lineage (Pp = 0.95, Ss = 0.94, Bs = 90) composed of
A. bowerbanki and A. hillae. The latter two species
could be distinguished using these three molecular
markers as the average genetic distance between these
two species was 2.3 ± 0.3%, and fully overlapped with
Fig. 4. Haplotype network of clade A ob-
tained in Network 4.6.1.2 for the mito-
chondrial intergenic spacer region (IGR)
between COI and l-rRNA. The size of
circles is proportional to the frequencies
of specimens sharing the same haplo-
type. The black solid circles are indica-
tive of mutations that differentiate each
haploty pe.
396 Arrigoni et al. – Phylogeny of Micromussa and Homophyllia
the intraspecic distances within A. bowerbanki and A.
hillae that are 2.1 ± 0.3% and 2.4 ± 0.4%, respectively.
Finally, all of the analyzed colonies of A. echinata
grouped within clade E, together with the Indian Ocean
specimen of A. cf. hillae (Fig. 1J, 2J) and published se-
quences of A. rotundoora, A. subechinata, and A.
hemprichii, although the genetic boundaries at species
level within this clade remain unclear. Surprisingly,
within this family a novel clade was detected that com-
prised S. wilsoni exclusively with a very strongly sup-
ported lineage monophyly (Pp = 1, Ss = 1, Bs = 99), that
was deeply divergent from clade I, which contains the
genera Lobophyllia, Parascolymia, and all of the other
Symphyllia species analyzed so far. Symphyllia wilsoni
fell at the base of the sister clades A and B.
The nal alignment of IGR data consisted of 1608
bp, of which 52 positions were variable. Haplotype
network analysis of clade A, as inferred from the mtD-
NA IGR locus, was highly concordant with the phylog-
eny reconstruction of clade A based on COI, histone
H3, and ITS region (Fig. 4). A total of nine haplotypes
were detected and ve main clusters, corresponding to
the ve lineages found using the other markers, were
revealed. These clusters were separated by a minimum
of eight substitutions (between M. amakusensis from
Japan and A. lordhowensis) and no haplotypes were
shared between two or more clusters. In particular, we
found two closely related haplotypes specic to M.
amakusensis from Japan and differing by three base
changes, two closely related haplotypes for Montast-
raea multipunctata separated by one substitution, two
closely related haplotypes specic of A. lordhowensis
showing one mutation event, a single haplotype for all
eight specimens of H. cf. australis, and two closely re-
Tab l e 1. Macromorphology and micromorphology of the Lobophylliidae examined in this study. Explanation of characters, their ID
numbers (in brackets) and state names are from Budd et al. (2012).* = character examined on polycentric coralla; ° = a central larger
corallite may be observed; § = in series, >1/4 in uniserial corallites; Csn = number of cycle of costosepta; Sn = number of cycle of septa.
Clade A Clade B Clade J Clade E
Character Micromussa Micromussa Micromussa Micromussa Micromussa Homophyllia Homophyllia Homophyllia Australophyllia Acanthastrea
amakusensis indiana sp. nov. lordhowensis multipunctata pacicasp. nov. australis bowerbanki bowerbanki wilsoni hemprichii
(referred to as M. cf. (previously (previously (referred to as H. (previously A. (previously (previously (referred to as
amakusensis in text) Acanthastrea) Montastraea) cf. australis in text) bowerba nki ) A. hillae) Symphyllia) A. cf. hillae in text)
Intracalicular budding (1) Present Present Present Absent Present* Present* Present Present Present Present
Extracalicular budding (2) Present Present Present Present Present* Present* Present Present Present Present
Circumoral budding and associated Absent Absent Absent Absent Absent Absent Absent Absent Absent Absent
corallite polymorphism (3)
Corallite integration (40) Discrete Discrete Discrete Discrete Discrete Discrete Discrete Discrete Uniserial Discrete
Calice or valley width (7) Medium Medium Medium Small-Medium Large Large Large Large Medium Large
Number of septa (10) 4 cycles 3 - 4 cycles 3 - 4 cycles 3 - 4 cycles > 4 cycles > 4 cycles > 4 cycles > 4 cycles > 4 cycles > 4 cycles
Free septa (11) Irregular I rregular Irregular Irregular Irregular Ir regular Irregular I rregular Irregula r Irregula r
Septa spacing (per 5 mm) (12) 6 - 12 6 - 12 6 - 12 6 - 12 6 - 12 6 - 12 6 - 12 6 - 12 6 - 12 6 - 12
Relative costosepta thickness Unequal- Unequal Unequal- Unequal Unequal- Unequal Unequal- Unequal Unequal Unequal
(Cs1 and Cs2 - vs - Cs3) (13) slightly unequal slightly unequal slightly unequal slightly unequal
Corallite centre lin kage (14) Absent Absent Absent Absent Absent Absent Absent Absent Discontinuous Absent
Columella structure (15) Trabecular and Trabecular and Trabecular and Trabecular and Trabecular and Trabecular and Trabecular and Trabecular and Trabecular and Trabecular and
spongy spongy spongy spongy spongy spongy spongy spongy spongy spongy
Columella size relative to calice > 1/4 < 1/4 < 1/4 > 1/4 < 1/4 < 1/4 < 1/4 < 1/4 < 1/4§ < 1/4
width (16)
Tooth base (mid-septum) (35) Elliptical-parallel Elliptical-parallel Elliptical-parallel Elliptical-parallel Elliptical-parallel Elliptical-parallel Elliptical-parallel Elliptical-parallel Elliptical-parallel Elliptical-parallel
Tooth tips (38) Irregular Irregular Irregular Irregular Irregular Irregular Irregular I rregular I rregular Irregular
Tooth height (S1) (39) Medium Medium Medium Medium High High High High Medium Medium to high
Tooth spacing (S1) (40) Medium Medium Medium Medium Medium Wide Wide Wide Medium Medium to wide
Granules shape and distribution (43) Strong, pointed, Strong, pointed, Strong, pointed, Strong, pointed, Strong, pointed, Weak, rounded, Weak, rounded, Weak, rounded, Weak, rounded, Weak, rounded,
scattered scattered scattered scattered scattered uniformly uniformly uniformly scattered enveloped by
distributed distributed distr ibuted thickening deposits
Interarea structure (44) Smooth Smooth Smooth Smooth Smooth Smooth Smooth Smooth Smooth Smooth
Cs3 / Cs1 tooth shape (45) Equal Equal Equal Equal Equal Equal Equal Equal Equal Unequal
MacromorphologyMicromorphology
397Contributions to Zoology, 85 (4) – 2016
lated haplotypes for M. amakusensis from Yemen dif-
fering by one base changes. Notably, M. amakusensis
from Japan and M. cf. amakusensis from Yemen were
distantly related and separated by 35 - 39 substitutions.
Macromorphology
The examined lobophylliid species presented a wide
array of corallum macromorphology and corallite size
and organization. Homophyllia australis (Figs 1F, 2F,
10, 11, 12A) and H. cf. australis (Figs 1E, 2E, 8, 9,
12D) were solitary forming large and predominantly
monocentric coralla. Micromussa amakusensis (Figs
1A, 2A, 7A), M. cf. amakusensis (Figs 1B, 2B, 5, 6,
7D), A. lordhowensis (Figs 1C, 2C, S5, S6), Montast-
raea multipunctata (Figs 1D, 2D, S9, S10), A. hillae
(Figs 1H, 2H, S13A, C, E, G), A. cf. hillae (Figs 1J, 2J),
A. bowerbanki (Figs 1G, 2G, S13B, D, F, H) and S.
wilsoni (Figs 1I, 2I, 13) were colonial species forming
encrusting to massive coralla. Corallite organization
was cerioid in M. amakusensis (Fig. 1A), M. cf. am-
akusensis (Fig. 1B), and A. lordhowensis (Fig. 1C);
plocoid in Montastraea multipunctata (Fig. 1D); ceri-
oid to sub-meandroid in A. hillae (Fig. 1H) and A.
bowerbanki (Fig. 1G); and mainly meandroid in S. wil-
soni (Fig. 1I).
In all the examined species, both intracalicular and
extracalicular budding occurred (Table 1). Although
polystomatous coralla were observed in both Homo-
phyllia australis and H. cf. australis, a more pronounced
tendency to polystomatism was observed in the exam-
ined series of the latter (Figs 8C–I, 9D–H, S11). Both
intracalicular (Fig. 8C) and extracalicular (Fig. 8F–I)
modes of budding were observed in H. cf. australis. In
Tab l e 1. Macromorphology and micromorphology of the Lobophylliidae examined in this study. Explanation of characters, their ID
numbers (in brackets) and state names are from Budd et al. (2012).* = character examined on polycentric coralla; ° = a central larger
corallite may be observed; § = in series, >1/4 in uniserial corallites; Csn = number of cycle of costosepta; Sn = number of cycle of septa.
Clade A Clade B Clade J Clade E
Character Micromussa Micromussa Micromussa Micromussa Micromussa Homophyllia Homophyllia Homophyllia Australophyllia Acanthastrea
amakusensis indiana sp. nov. lordhowensis multipunctata pacicasp. nov. australis bowerbanki bowerbanki wilsoni hemprichii
(referred to as M. cf. (previously (previously (referred to as H. (previously A. (previously (previously (referred to as
amakusensis in text) Acanthastrea) Montastraea) cf. australis in text) bowerba nki) A. hillae) Symphyllia) A. cf. hillae in text)
Intracalicular budding (1) Present Present Present Absent Present* Present* Present Present Present Present
Extracalicular budding (2) Present Present Present Present Present* Present* Present Present Present Present
Circumoral budding and associated Absent Absent Absent Absent Absent Absent Absent Absent Absent Absent
corallite polymorphism (3)
Corallite integration (40) Discrete Discrete Discrete Discrete Discrete Discrete Discrete Discrete Uniserial Discrete
Calice or valley width (7) Medium Medium Medium Small-Medium Large Large Large Large Medium Large
Number of septa (10) 4 cycles 3 - 4 cycles 3 - 4 cycles 3 - 4 cycles > 4 cycles > 4 cycles > 4 cycles > 4 cycles > 4 cycles > 4 cycles
Free septa (11) Irregular I rregular Irregular Irregular Irregular Ir regular Irregular I rregular Irregula r Irregula r
Septa spacing (per 5 mm) (12) 6 - 12 6 - 12 6 - 12 6 - 12 6 - 12 6 - 12 6 - 12 6 - 12 6 - 12 6 - 12
Relative costosepta thickness Unequal- Unequal Unequal- Unequal Unequal- Unequal Unequal- Unequal Unequal Unequal
(Cs1 and Cs2 - vs - Cs3) (13) slightly unequal slightly unequal slightly unequal slightly unequal
Corallite centre lin kage (14) Absent Absent Absent Absent Absent Absent Absent Absent Discontinuous Absent
Columella structure (15) Trabecular and Trabecular and Trabecular and Trabecular and Trabecular and Trabecular and Trabecular and Trabecular and Trabecular and Trabecular and
spongy spongy spongy spongy spongy spongy spongy spongy spongy spongy
Columella size relative to calice > 1/4 < 1/4 < 1/4 > 1/4 < 1/4 < 1/4 < 1/4 < 1/4 < 1/4§ < 1/4
width (16)
Tooth base (mid-septum) (35) Elliptical-parallel Elliptical-parallel Elliptical-parallel Elliptical-parallel Elliptical-parallel Elliptical-parallel Elliptical-parallel Elliptical-parallel Elliptical-parallel Elliptical-parallel
Tooth tips (38) Irregular Irregular Irregular Irregular Irregular Irregular Irregular Irregular Ir regular Irregular
Tooth height (S1) (39) Medium Medium Medium Medium High High High High Medium Medium to high
Tooth spacing (S1) (40) Medium Medium Medium Medium Medium Wide Wide Wide Medium Medium to wide
Granules shape and distribution (43) Strong, pointed, Strong, pointed, Strong, pointed, Strong, pointed, Strong, pointed, Weak, rounded, Weak, rounded, Weak, rounded, Weak, rounded, Weak, rounded,
scattered scattered scattered scattered scattered uniformly uniformly uniformly scattered enveloped by
distributed distributed distr ibuted thickening deposits
Interarea structure (44) Smooth Smooth Smooth Smooth Smooth Smooth Smooth Smooth Smooth Smooth
Cs3 / Cs1 tooth shape (45) Equal Equal Equal Equal Equal Equal Equal Equal Equal Unequal
398 Arrigoni et al. – Phylogeny of Micromussa and Homophyllia
specimens where the former occured, adjacent centers
were linked by lamellar linkage (Fig. 8C).
Circumoral budding and associated corallite poly-
morphism was not observed in any of the examined
taxa (Table 1). However, in specimens identied as A.
bowerbanki a larger central corallite was observed
(Figs S13B, D, F, H, S14B, D, F, H) following Veron
(2000) (Table 1). Veron (2000, vol. 3, p. 26) remarked
that in colonies of this species “a central corallite is
usually conspicuous”, but he did not mention this char-
acter for A. hillae. However, a larger corallite undergo-
ing intracalicular budding was observed roughly at the
center of the holotype of this species (Fig. S13A) as
well as in the holotype of A. bowerbanki (Fig. S13B)
and in some specimens illustrated by Veron and Pi-
chon (1980: gs 441, 443).
The only species with meandroid corallite arrange-
ment was S. wilsoni (Table 1) (Figs 1I, 13A–B). Al-
though these were generally uniserial and discontinu-
ous, in some cases a biserial condition was almost at-
tained. In fact, in this species centers within a series
were linked by a thick lamellar process which in some
cases seemed to actually split the columella in two,
giving it a bilateral symmetry (Fig. 13C). Symphyllia
wilsoni was also the only species among those we ex-
amined or (to our knowledge) in the Lobophylliidae to
form monticules (Fig. 13A, E–G) resembling the hyd-
nophores typical of the merulinid genus Hydnophora
Fischer von Waldheim, 1807 and of the agariciid
Pavona varians Verrill, 1864. These monticules
formed “where sections of common wall between cor-
allites intersect and develop into conical mounds” (Ve-
ron, 2000: vol. 2, p. 346).
With reference to the ranges set by Budd et al.
(2012), calices were large (> 15 mm) in A. bowerbanki,
A. hillae, A. cf. hillae, H. australis, and H. cf. australis,
medium (8 – 15 mm) in S. wilsoni, M. amakusensis, M.
cf. amakusensis, and A. lordhowensis, and small to me-
Fig. 5. Micromussa indiana sp. nov. (re-
ferred to as Micromussa cf. amakusensis
in tex t) from the Gulf of Aden, Indian
Ocean. A) Corallum mor phology of
specimen MNHN-IK-2012-14232; B)
Detail of the same specimen as in A; C)
Within specimen variability of UNIMIB
MU183; D) EPA S C3695, specimen
from Socotra Island with some larger
corallites than the specimen in A and B;
E–F) Close ups of adjacent parts of the
specimen in C having smaller and la rger
corallites, respectively.
399Contributions to Zoology, 85 (4) – 2016
dium (3 – 9 mm) in Montastraea multipunctata (Tabl e
1). Within the examined series of specimens, the largest
corallites of H. australis (Fig. S12) reached larger di-
mensions than those of H. cf. australis (Fig. S11).
Among the species with medium sized corallites, M.
amakusensis had overall larger and more regularly
shaped corallites than M. cf. amakusensis (Figs 1A, 7A
and Figs 1B, 7D, respectively). The latter displayed a
more variable corallite diameter and outline (Fig. 5). In
fact, although some coralla can display corallites with a
larger diameter and more regular shape (Fig. 5C–D)
and resembled those of the typical M. amakusensis, in
these specimens the within corallum variability for
these characters was remarkable (Fig. 5C, E–F). Acan-
thastrea lordhowensis had larger corallites than M.
amakusensis and M. cf. amakusensis. In this species
too, corallite shape showed high variability ranging
from irregularly polygonal and elongated (Fig. S7A–B)
to a more regular outline (Fig. S7E). Such intraspecic
variability was also obvious in situ (e.g. see Fig 8C with
rounded polyps and Fig. 8G with irregularly shaped
and elongated polyps). In the plocoid Montastraea mul-
tipunctata, corallite shape was less variable, ranging
from round to oval in larger corallites although smaller
corallites can be more irregular (Figs S9, S10).
With respect to the number of septa, Acanthastrea
lordhowensis, Microsmussa cf. amakusensis, and
Montastraea multipunctata had 3 – 4 cycles of septa
while 4 cycles were found in M. amakusensis. More
than 4 cycles were observed in all the other examined
species. All examined species shared the presence of
irregular free septa and a similar spacing of septa (6 –
11 per 5 mm).
Costosepta were unequal in relative thickness be-
tween those of the rst two orders and those of the third
order in M. cf. amakusensis, Montastraea multipunc-
tata, H. australis, A. hillae, A. cf. hillae, and S. wilsoni,
while in the remainder of the examined species cos-
tosepta were unequal to slightly unequal (Table 1),
showing a certain intraspecic variability of this char-
acter. Notably, the thickness of costosepta in A. bower-
banki was only slightly unequal in the holotype (Fig.
S13B) and in some of the other examined specimens
(Fig. S13D, F, H). Similarly, this character showed an
obvious variability also in the examined specimens of
A. hillae (Fig. S13A, C, E, G). In S. wilsoni the unequal
costosepta thickness was particularly obvious especial-
ly within the series of corallites where thicker septa
reach and fuse with the lamellar process uniting the
centers (Fig. 13B–C). As a result, thicker costosepta al-
ternated with series centers in this species.
The columella was trabecular, spongy and smaller
than 1/4 of the diameter of the calice in all species
except M. amakusensis and Montastraea multipunc-
tata, where it was consistently larger (Table 1). The
dimension of the columella (relative to the calice
width) was quite variable in S. wilsoni where it was
smaller in series of corallites (Fig. 13B, D) and larger
in monocentric corallites (Fig. 13B).
Micromorphology
Despite the above-mentioned macromorphological
variability of skeletal structures (Table 1), all the ex-
amined taxa shared similar micromorphology and, in
particular, have septal teeth with an elliptical base at
mid-septum and irregular tooth tips.
Septal tooth height was medium (0.3 – 0.6 mm) in
M. amakusensis, M. cf. amakusensis, A. lordhowensis,
Montastraea multipunctata, and S. wilsoni; medium
to high in A. cf. hillae; and high (> 0.6 mm) in H. aus-
tralis, H. cf. australis, A. hillae, and A. bowerbanki
(Table 1) (Fig. 14). Both Wells (1955), in his original
description of A. hillae, and later Chevalier (1975) re-
marked that in this species septal teeth tend “to be
relatively small near tops of walls and increasing
greatly in height toward the columella, especially in
the case of the longer septa but decreasing again at the
inner ends of the septa” (Wells, 1955) (Fig. S13A). We
observed the same variation in septal teeth height in
the septa of the rst and second cycle in all examined
specimens of A. hillae (e.g. Fig. S13C, E, G). Although
the same was not observed in the holotype of A. bow-
erbanki, which have thinner septa than the holotype of
A. hillae, this peculiar within-septum tooth height
variation was observed in several specimens of this
species (Fig. 5D, F, H), although it was less obvious in
specimens with thinner septa (Fig. 5H). The same pat-
tern in septal teeth height variation was consistently
observed in all examined specimens of H. australis
(Fig. 10B–D, F–H), but not in those of H. cf. australis
(Figs 8, 10E–F), in which septal teeth of the rst two
cycles decreased in height from the wall to the colu-
mella. In A. lordhowensis, although septal teeth were
of uniform height within septa, a certain degree of in-
traspecic variation of this character was observed
with some specimens having more developed teeth
(Fig. S7C, F) than others (Fig. S7D–E).
Although septal tooth size varied between cycles of
septa (being larger in those of lower orders and smaller
in those of higher orders) their shape was consistently
uniform within the same species in all cases except A.
400 Arrigoni et al. – Phylogeny of Micromussa and Homophyllia
cf. hillae (Table 1). Tooth spacing was wide (> 1.0
mm) in H. australis, A. hillae, and A. bowerbanki
(Fig. 14L–Q) and medium (0.3 – 1.0 mm) in all the
other species. Inter-area structure was smooth in all
species (Table 1). Septal side granulation was weak,
rounded, and uniformly distributed in H. australis, A.
hillae, and A. bowerbanki (Fig. 14P–R); weak, round-
ed, and scattered in S. wilsoni (Fig. 14S); weak,
rounded and enveloped by thickening deposits in A.
cf. hillae (Fig. 14T); strong, pointed and scattered in
M. amakusensis, M. cf. amakusensis, A. lordhowen-
sis, Montastraea multipunctata, and H. cf. australis
(Fig. 14F–J) (Table 1).
A peculiar micromorphology was observed in S.
wilsoni where columellae in series of corallites were
separated, as described above, by ridges formed by the
fusion of thicker septa with the lamellar process unit-
ing centers. SEM observations had revealed the pres-
ence of clusters of granules arranged over this lamellar
process in a saddle-like fashion (Fig. 13H–J).
Fig. 6. Micromussa indiana sp. nov. (re-
ferred to as Micromussa cf. amakusensis
in text) in situ. A) Specimen UNIMIB
MU215 showing a beige colouration
with red polyp outline at Al Mukallah;
B) Colony with retracted polyps having a
green, light grey and red colouration at
Hyllanyia Island, Bir Ali; C) Burum; D)
Al Badi Island, Kamaran Islands, South-
ern Red Sea; E) Colony with irregularly
shaped polyps at Habban Island, Aden;
F) Colony with a green brown morph at
Di Hamri, Socotra Island, this colony
was found at the same site as the one in
G; G) Di Hamri, Socotra Island white
red morph sympatric with F; H) Hyl-
lanyia Island, Bir Ali.
401Contributions to Zoology, 85 (4) – 2016
Discussion
As a result of the evaluation above, a number of formal
taxonomic actions are undertaken hereafter (Fig. 15),
including the description of two new species, Micro-
mussapacica sp. nov. (so far Homophyllia cf. austra-
lis) and Micromussa indiana sp. nov. (so far Micromus-
sa cf. amakusensis), and one new genus, Australophyl-
lia gen. nov., to accommodate S. wilsoni. Acanthastrea
hillae is considered a junior synonym of Acanthastrea
bowerbanki and this species is formally moved into
Homophyllia. Finally, Montastraea multipunctata and
A. lordhowensis are moved to the genus Micromussa,
whereas M. minuta is moved to Acanthastrea and M.
diminuta to Goniopora. For detailed descriptions and
systematic account we refer to the Appendix. The ge-
nus Micromussa is characterized by medium tooth
spacing (0.3 – 1.0 mm) and by strong, pointed and scat-
tered septal side granulation, whereas Homophyllia
shows wide tooth spacing (> 1.0 mm) and weak, round-
ed, and uniformly distributed septal side granulation
(Fig. 14). Conversely, Australophyllia shows medium
tooth spacing (0.3 – 1.0 mm), weak, rounded, and scat-
tered septal side granulation, and monticules.
In this study we explored the molecular phylogeny
and proposed a new taxonomy for seven nominal scler-
actinian coral species ascribed to the Lobophylliidae
(Fig. 15), corroborating the molecular data with the
evaluation of the gross- and ne-scale skeleton mor-
phology. The analyzed specimens were recovered in
three main genus-level clades based on the multi-locus
phylogeny reconstruction (Fig. 3) and, in particular, M.
amakusensis, M. indiana sp. nov., M.pacica, M. lord-
howensis, and M. multipunctata belonged to clade A
sensu Arrigoni et al. 2014a); H. australis and H. bower-
banki belonged to clade B sensu Arrigoni et al. (2014a);
A. wilsoni was self-standing in a new clade. Despite
these species sharing some morphological features, we
showed that several macro- and micromorphological
characters were diagnostic for the denition of these
three genus-level lineages (Table 1). Moreover, mor-
phological data supported molecular ndings, reveal-
ing the presence of two distinct entities within H. aus-
tralis (Figs 1F, 2F, 3, 9-12, S11, S12, Table 1) as well as
M. amakusensis (Figs 1E, 2E, 3, 4-7, Table 1).
Clade A
The results presented in this study substantially in-
crease the known species and macromorphological
diversity of Micromussa. The genus is now composed
of ve species all of which are investigated in the pre-
sent study, i.e. M. amakusensis, M. indiana sp. nov., M.
lordhowensis, M. multipunctata, and M. pacica sp.
nov. (Fig. 15).
The case of M. indiana sp. nov. is a remarkable ex-
ample of how the presumed morphological variability
of a single species, M. amakusensis, over a large geo-
graphic distribution range can actually hide multiple
identities. Indeed, Micromussa cf. amakusensis speci-
mens from the Indian Ocean (Yemen) (Figs 1B, 5, 7A-
C) and the typical material from the Pacic Ocean
(Japan) (Figs 1A, 7D-F) were both recovered in clade
A albeit in two distinct and well supported clades (Fig.
3). Although they look remarkably similar in situ,
showing both a bright colouration (Figs 2B, 6 and Fig.
2A; Veron, 2000, vol. 3: 10–11, gs 1–5, respectively),
and present a similar medium calice size, up to 4 cy-
cles of septa, and a similar septal spacing and micro-
morphology (Table 1), they can be distinguished on the
basis of several macromorphological characters (Fig.
7). In the typical M. amakusensis, septa of the rst
three cycles are of equal thickness and height (Fig. 7A-
B) and those of the third are longer than ¾ of the rst
two and in most cases almost reach the columella.
Conversely, in M. indiana sp. nov. septal length varies
from slightly unequal to unequal (Fig. 5), with septa of
the third (often incomplete) cycle being shorter than ½
of those of the rst two and, overall, given the more
irregular corallite outline, the second to fourth cycles
can be more or less complete depending on the coral-
lite (Figs 5E-F, 7D). Finally, M. indiana sp. nov. has a
smaller columella composed of less threads (Fig. 7F)
than M. amakusensis (Fig. 7C). A new species of Mi-
cromussa is therefore formally described hereafter to
accommodate M. indiana sp. nov. Therefore, the previ-
ous lack of direct comparison of skeletal morphology
from Indian Ocean and Pacic Ocean material (Veron,
2000; Pichon et al., 2010) has perpetrated this error
and underestimated the increasingly clear peculiarities
of the Indian Ocean coral fauna (Arrigoni et al., 2012;
Obura, 2012; Reijnen et al., 2014). However, once type
material, specimens collected from type localities, and
a large reference collection from Yemen were com-
pared, the macromorphological differences between
these species became obvious. Recent works have re-
vealed several other cases of deep genetic divergence
between Indian and Pacic populations in some spe-
cies ascribed to other families, such as Blastomussa
merleti (Arrigoni et al., 2012), Coelastrea aspera and
C. palauensis (Huang et al., 2014b), Favites halicora
(Ar r igoni et al., 2012), Goniopora somaliensis (Kitano
402 Arrigoni et al. – Phylogeny of Micromussa and Homophyllia
et al., 2014), Pocillopora spp. (Pinzon et al., 2013), and
Stylophora pistillata (Stefani et al., 2011; Keshawmur-
thy et al., 2011; Flot et al., 2011). Such evidence strong-
ly argues against the concept of “geographic subspe-
cies” proposed by Veron (1995) in order to explain the
wide variety of geographic variations in some nominal
species living both in the Indian and Pacic Ocean.
Micromussa lordhowensis represents one of the
various cases of Acanthastrea mis-assignment high-
lighted by Arrigoni et al. (2014a, 2015). As shown by
these authors, Acanthastrea, as interpreted until Veron
(2000), was the most polyphyletic genus in the family
Lobophylliidae based on mitochondrial and nuclear
phylogeny reconstructions. In the present study, M.
lordhowensis was studied for the rst time from a phy-
logenetic perspective and transferred to Micromussa
as it was found to be unrelated to the genus type, A.
echinata (Fig. 3), but closely related to the other spe-
cies in Micromussa. Furthermore, morphological
analyses conrmed that M. lordhowensis displays the
septal size, shape, and granulation typical of all Micro-
mussa species rather than the smoother septal sides
ornamentation of Acanthastrea (see also Arrigoni et
al., 2015). In the original description of M. multipunc-
tata, Hodgson (1985) stated that, despite some charac-
teristics shared with the other species of Montastraea
(now exclusively a monospecic Atlantic genus, see
Budd et al. (2012)), M. multipunctata is unusual on the
basis of growth form, polyp shape, and notably septal
dentations. Indeed this species is also different from
all the others species examined in this study due to its
plocoid corallite organization. Nevertheless, the mo-
lecular results presented in this study show that M. mul-
tipunctata clearly belongs to the lineage composed of
the other four Micromussa species (Fig. 3). Moreover,
this species shares a similar septal teeth micromor-
phology with the other Micromussa species having, for
example, the same type of strong septal sides and tips
granulation (Fig. 12).
The other new species of Micromussa described in
this study, M. pacica sp. nov., represents a different
case altogether. This solitary species has been con-
fused for a long time with the largely sympatric Homo-
phyllia australis (Veron and Pichon, 1980). Super-
cially, these two species are indeed impressively simi-
lar, especially in the eld, although a closer observa-
tion of the skeletal features allowed separating them
effectively, a distinction fully conrmed by the mo-
lecular results.
Clade B
The genus Homophyllia was resurrected by Budd et al.
(2012) following recent morphological observations on
H. australis and Parascolymia vitiensis (Budd and Sto-
larski, 2009). The authors demonstrated that these two
Pacic species are clearly unrelated based on the septa
granulation, the area between teeth, and the thickening
deposits (Budd and Stolarski, 2009; Budd et al., 2012).
The phylogeny reconstruction proposed in this study
conrms that H. australis belongs to the Lobophyllii-
dae and that it is not related to P. vitiensis (Fig. 3), thus
Fig. 7. Micromussa amakusensis (A–C)
and Micromussa indiana sp. nov. (re-
ferred to as Micromussa cf. amakusensis
in text) (D–F) compared, Scanning
Electron Microscopy images. A–D) top
views of calices; B–E) side view of the
septa in two adjacent calices; C–F) top
view of the columella. Arabic numerals
at the outer end of the septa in A indicate
the cycle number (from 1 to 4).
403Contributions to Zoology, 85 (4) – 2016
supporting the taxonomic changes proposed by Budd et
al. (2012). The combination of morpho-molecular data
solves a long-standing taxonomic riddle concerning the
placement of these two species (Matthai, 1928;
Vaughan and Wells, 1943; Wells, 1964; Veron and Pi-
chon, 1980; Veron, 2000) and represents an excellent
example of concordance between genetics and novel
morphological characters (Stolarski and Roniewicz,
2001; Budd et al., 2 010).
Two variable, though consistently different, morphs
of the solitary species Homophyllia australis were dis-
tinguished in the examined material and the molecular
analyses supported our initial separation of the large
set of specimens included in our analyses in two dif-
ferent clades, namely the typical H. australis in clade
B and M. pacica s p. n o v. in clade A (Fig. 3). With
reference to the characters dened by Budd et al.
(2012), both morphs have large calices (Figs 8, 10,
12A, D), more than 4 cycles of septa with those of the
fth cycle or higher free (Fig. 12B, E), similar septal
spacing and a trabecular and spongy columella (Figs 8,
10) of similar size relative to calice width. However,
calices in H. australis can attain larger dimensions
(Fig. S12) and reach up to 6 cycles of septa while in
Fig. 8. Micromussapacicasp. nov. (re-
ferred to as Homophyllia cf. australis in
tex t) (Specimens in A–F included in the
phylogeny reconstruction in Fig. 3). A)
Top and B) Side view of MNHN IK-
2012-16046; C) IRD HS3359 polystoma-
tous specimen, image shows detail of the
linkage (dashed line); D) MNHN IK-
2012-16045 polystomatous specimen; E)
Side view of IRD HS3327; F) Top view
of MNHN IK-2012-16043; G–I) Show
top, side and lateral corallite view, re-
spectively, of MNHN IK-2012-16044 a
specimen with two calices from the
Gambier Islands. White arrows indicates
a budding calice.
404 Arrigoni et al. – Phylogeny of Micromussa and Homophyllia
M.
pacicasp. nov. the largest calices are smaller and
a maximum of 5 cycles of septa were observed (Fig.
S11). A more pronounced tendency to polystomatism
was observed in M.pacicasp. nov. than in the typical
H. australis. Moreover, in H. australis septa are mark-
edly unequal in thickness and height in all examined
specimens (Figs 10, 12A-C, S12) while in M.pacica
sp. nov. septa are slightly unequal (Figs 8, 12D-F, S11)
with only some cases of unequal septa observed (Figs
9A, S11), and while in the former septal teeth height
increases from mid-septum towards the columella in
the rst two cycles, in the latter this variation is not
observed. Finally, the medium septal tooth spacing
and the strong, pointed and scattered septal granula-
tion, typical of all the species recovered in clade A
with Micromussa clearly distinguish M. pacica sp.
nov. from the typical H. australis having large spacing
and weak, rounded, uniformly distributed granulation.
The two forms co-occur in the eld (e.g. in New Cal-
edonia and in Western Australia) and their similar
Fig. 9. Micromussa pacicasp. nov. (re-
ferred to as Homophyllia cf. australis in
tex t) in situ. A) Brown and white coral-
lum with unequal septa from the lagoon
pinnacles north of Mangareva Island,
Gambier Islands, French Polynesia
(23°04.12’S, 134°55.83’W, 05/07/2011);
B) IRD HS3202, green and white polyp;
C) IRD HS3527, white and dark red pol-
yp, same as in Fig. S7; D) MNHN IK-
2012-16043; E) IRD HS3359; F) IRD
HS3327; G) IRD HS3483; H) Polys-
tomatous colony from Taravai Island,
Gambier Islands, French Polynesia
(23°08.72’S; 135°03.09’W, 08/07/2011).
White arrows point at budding polyps.
405Contributions to Zoology, 85 (4) – 2016
macromorphology is likely to have misled previous
authors in detecting more subtle differences at the sep-
tal level. For example, Veron and Pichon (1980) includ-
ed specimens of M.pacicasp. nov. (e.g. g. 428) and
of H. australis (e.g. g. 430) in the same series.
Doubts on the identication of Acanthastrea-like
specimens with large polyps as Acanthastrea hillae or
H. bowerbanki can be found in the literature. Cheva-
lier (1975) having examined the type material of both
species remarked “Acanthastrea hillae parait très
proche, peut-être même identique à A. bowerbanki”
(A. hillae seems very close, maybe even identical to A.
bowerbanki). However, he also noted the differences
in thickness of septa and development of septal teeth
that we have also observed in our examination of the
holotypes of both species despite their similar corallite
size (Fig. S13A, B). Veron (2000) indicated the pres-
ence of a central larger corallite as a diagnostic char-
acter for H. bowerbanki but he does not mention this
for A. hillae. However, the holotypes of both species
Fig. 10. Top (A, C, E, G) and side (B, D,
F, H) views of Homophyllia australis
(specimens included in the phylogeny
reconstruction in Fig. 3. A) IRD HS3441,
top view; B) Side view of the same coral-
lum as in A; C) IRD HS3470, top view;
D) Side view of the same corallum as in
C; E) IRD HS3545, top view; F) Side
view of the same corallum as in E; G)
IRD HS3524, top view; H) side view of
the same corallum as in G. Black arrows
point to the large septal teeth of the rst
cycle of septa.
406 Arrigoni et al. – Phylogeny of Micromussa and Homophyllia
have a central larger corallite undergoing intracalicu-
lar budding (Fig. S13A, B). Conversely, Veron and Pi-
chon (1980) in their treatment of A. hillae, noted a “su-
percial resemblance between some coralla of this
species and Moseleya latistellata which also has a
central corallite, similar type of budding and corallites
of similar size and shape”. They further separated both
species on the basis of the different septal teeth and
presence of paliform lobes. In the present study we ex-
amined a large series of specimens spanning the range
of the morphological variability that separates the
holotypes of H. hillae and A. bowerbanki, thus includ-
ing specimens with thicker (Fig. S13C) and thinner
(Fig. S13D) septa, as well as specimens with more ob-
vious larger central corallites (Fig. S13D) and lack
thereof (Fig. S13E). All these were recovered in the
same well-supported clade in clade B (Fig. 3) and no
differences in any of the morphological characters
considered (other than relative costosepta thickness)
were observed. We conclude that in absence of evi-
Fig. 11. Homophyllia australis in situ. A)
IRD HS3544; B) IRD HS3311, same as
in Figs 9 A–C; C) IRD HS3545, same as
in Figs 5 E–F, S6; D) IRD HS3441, same
as in Figs 5 A–B, S6; E) IRD HS3470,
same as in Figs 5 C–D, S6; F) IRD
HS3525, same as in Fig. S6; G) IRD
HS3447, same as in Fig. S6; H) IRD
HS3526, polystomatous also shown in
Fig. S6.
407Contributions to Zoology, 85 (4) – 2016
dence of morphological and molecular characters that
allow the separation of the two nominal species, H.
bowerbanki is a senior synonym of A. hillae, and
therefeore has preference. Furthermore, based on the
placement of this species in clade B together with H.
australis, rather than in clade E with the type species
of Acanthastrea, and the consistent micromorphologi-
cal afnities between the two species (Table 1), we for-
mally move A. hillae into the genus Homophyllia as a
junior synonym of H. bowerbanki.
As a result of the morphological and molecular in-
vestigations presented in this study, the once monospe-
cic genus Homophyllia now includes two species, H.
australis, the predominantly solitary type species, and
the colonial H. bowerbanki (Fig. 15). Although previ-
ous authors noted the pronounced septal sides granula-
tion of both species (Veron and Pichon, 1980), the
strikingly different corallum shape likely kept them
from considering this skeletal feature as phylogeneti-
cally informative, thus solitary and colonial species
were maintained in different genera. A similar situa-
tion has also been recently reported for another lobo-
phylliid genus, Sclerophyllia, originally consisting of a
monostomatous species (Klunzinger, 1879). The genus
was recently revised based on molecular and morpho-
logical evidence and is currently composed of a soli-
tary and a colonial species (Arrigoni et al., 2015). This
result is not entirely unexpected, because in the scler-
actinian family Fungiidae, monostomatous and polys-
tomatous species were traditionally also classied in
different genera despite close phylogenetic relations
(Hoeksema 1989, 1991, 1993). Molecular analyses have
resulted in a new classication of this family, in which
this distinction does not matter anymore, such as
among species of the genera Cycloseris and Pleuractis
(Gittenberger et al., 2011; Benzoni et al., 2012).
Despite supercial macromorphological similarities
between H. bowerbanki and Acanthastrea species,
mainly the size and arrangement of corallites and the
number of septa (Veron and Pichon, 1980), consistent
Fig. 12. Homophyllia australis (IRD HS3311; A–C) and Micromussa pacicasp. nov. (referred to as Homophyllia cf. australis in t ext)
(D-F) compared, Scanning Electron Microscopy images. A–D) top views of calices; B–E) enlarged side view of the calice; C–F) side
view of septa. Arabic numerals at the outer end of the septa in B and E indicate the cycle number (from 1 to 5). White arrows point at
the larger septal tooth present at the inner end of septa in typical Homophyllia australis.
408 Arrigoni et al. – Phylogeny of Micromussa and Homophyllia
differences in septal tooth micromorphology were evi-
denced between these species (Budd and Stolarski,
2009; Arrigoni et al., 2015). Furthermore, a deep ge-
netic divergence separates the clade including H. bow-
erbanki from the lineage leading to the species of Acan-
thastrea (Fig. 3; Arrigoni et al., 2014a, 2 014b, 2015).
Indeed, some large specimens of Acanthastrea, like the
colonies of A. hemprichii included in molecular and
morphological analyses of the present study (Figs 1J,
2J), can look similar to H. bowerbanki and A. hillae
(Figs 1G-H, 2G-H) and were therefore preliminarily
identied as such in the eld. However, none of the
specimens from the Indian Ocean identied as A. hillae
in the present study actually belong to A. hillae. Thus, it
is possible that the supposed presence of this species in
the Indian Ocean (Veron, 2000) is actually derived
from erroneous identications of A. hemprichii. Ve r o n
(2000, vol. 3, p. 28) himself reports that “records from
the western Indian Ocean are doubtful”. If this is the
case, the geographic distribution of the genus Homo-
phyllia would be restricted to the western Pacic, en-
compassing tropical, sub-tropical, and temperate condi-
tions. Moreover, the two species of Homophyllia are
predominantly sub-tropical, being uncommon within
Fig. 13. Australophyllia wilsoni (previ-
ously Symphyllia) AM WIL1. A) mean-
dering valleys and a hydnophoroid for-
mation (white arrow) which can be found
in this species; B) Full sized corallite on
the left hand side, side by side with a val-
ley in which the centers have an unusual
morphology and thicker septa (black ar-
rows) seem to separate the columellae,
which are also almost split in two; C) A
close up of the columellae shown in B;
D) SEM of the columella sitting deep in
the valleys; E) SEM top view of a hydno-
phoroid formation; F) SEM side view of
the same hydnophoroid formation as in
E; G) Close up of F showing septal side
granulation; H) In the foreground a colu-
mella and in the background behind it
the thicker septa separating ad jacent
columellae indicated by the black arrows
in B and C; I) A detail of the peculiar
structure forming saddle-shaped struc-
ture on the two adjoining inner ends of
the thicker septa separating adjacent
columellae; J) Granulation of the saddle-
shaped structure shown in I.
409Contributions to Zoology, 85 (4) – 2016
their range but relatively frequent in sub-tropical locali-
ties, such as Japan, New Caledonia, and south-western
Australia (Veron and Marsh, 1988; Veron, 1993, 2000;
Wallac e et al., 2009). For example, in Australia they are
rare on the Great Barrier Reef but are relatively com-
mon south to Moreton Bay (Veron and Pichon, 1980;
Veron and Marsh, 1988; Wallace et al., 2009).
Clade J
The most unexpected result of the present study is the
recovery of Australophyllia wilsoni, formerly assigned
to Symphyllia, as a distinct lineage within the Lobo-
phylliidae. The multi-locus phylogeny reconstruction
and each of the three single gene topologies are con-
cordant in supporting this unique assignment al-
though the best resolution is obtained using the con-
catenated data set (Figs 3, S3-S6). Considering the
concatenated COI-histone H3-ITS region data set, the
interclade genetic distances between A. wilsoni and
the other eight clades go from the smallest values with
clade A (3.2 ± 0.4%) and clade B (3 ± 0.4%) to the
largest one with clade G (9.6 ± 0.7%) (S2), while all of
the other distances vary between 5.6 and 6.2. These
Fig. 14. Scanning Electron Microscopy images of the septal teeth (A–E, K–O) and granules (F–J, P–T) of the examined Lobophylliidae
recovered in clade A (A–J), clade B (K–M, P–R), clade J (N, S), clade E (O, T) (see Fig. 3 for clades and colour codes). A, F) Micromussa
amak us en sis; B, G) Micromussa indiana sp. nov. (referred to as Micromussa cf. amakusensis in text); C, H) Micromussa lordhowensis
(previously Acanthastrea) AM 1642; D, I) Micromussa multipunctata (previously Montastraea) RMNH Coel 40090; E, J) Micromussa
pacicasp. nov. (referred to as Homophyllia cf. australis in text);K, P) Homophyllia australis IRD HS3311; L, Q) Homophyllia bower-
banki (previously Acanthastrea) AM 4629; M, R) Homophyllia bowerbanki (previously A. hillae) AM MH019; N, S) Australophyllia
wilsoni (previously Symphyllia) AM WIL1; O, T) Acanthastrea cf. hemprichii (referred in the text as A. cf. hillae) UNIMIB BA115.
410 Arrigoni et al. – Phylogeny of Micromussa and Homophyllia
distances completely overlap with the pairwise inter-
clade distances for the other clades, thus conrming the
genetic distinctiveness of A. wilsoni within its family.
In the original description of A. wilsoni, Veron (1985)
placed the species in the genus Symphyllia considering
the massive or sub-massive attened colony and a gen-
eral resemblance of the meandroid corallite arrange-
ment to that of this genus, although corallites are small-
er than those of any other Symphyllia species (Veron,
2000). Despite a supercial appearance of the macro-
morphology of the colony to some merulinds, such as
Plat ygyra and Oulophyllia, Veron (1985) included A.
wilsoni within the Mussidae (now an exclusively Atlan-
tic taxon, see Budd et al. (2012)) because of the size of
septal dentations and the thick and eshy aspect of liv-
ing polyps. Our molecular analyses demonstrate that
the species belongs to the Lobophylliidae but that it is
not closely related to any of the known extant lobophyl-
liid genera, showing a sister relationship with the group
composed by Micromussa and Homophyllia (Fig. 3).
These genetic ndings are also supported by a combi-
nation of several macro and micromorphological char-
acters illustrating the uniqueness of A. wilsoni among
the other taxa examined in the present study (Table 1),
and among the lobophylliids in general, also due to the
presence of monticules and the morphology of the colu-
mella in series of calices.
Another interesting feature of A. wilsoni is repre-
sented by its geographic distribution. The species is
restricted to the temperate waters of south-west Aus-
tralia, recorded from Shark Bay to Geographe Bay
along the coasts of Western Australia and thence east
to Bremer Bay in south Australia (Veron, 1985, 1993,
2000; Veron and Marsh, 1988). It is usually found in
shallow water on kelp-dominated coastal exposed rock
surfaces (Veron, 1985, 1993; Veron and Marsh, 1988).
This peculiar distribution range mostly overlaps that
of another distinctive species, Coscinaraea marshae,
and it is also similar to that of the south-eastern Aus-
tralian species Coscinaraea mcneilli, otherwise it is
unlike that of any other known extant coral species
(Veron and Pichon, 1980; Veron and Marsh, 1988; Ve-
ron, 1993, 2000).
Taxa outside clades A, B, and J
For Goniopora diminuta and Acanthastrea minuta, no
molecular and micromorphological data are available
and, to our knowledge, few specimens are deposited at
museums. We examined the lectoype of the former
species and the holotype of the latter one (Fig. S15).
The present macromorphological observations al-
lowed us to clarify that neither species actually be-
longs to Micromussa.
The lectotype of G. diminuta (MTQ G 55851, Ve-
ron, 2002, gs 235–237, Fig. S15A, B, D) is in fact a
specimen of the poritid genus Goniopora, most likely
a G. somaliensis. The corallites in this specimen show
the typical septal structure and fusion of the genus Go-
niopora Blainville, 1830, a columella made of a single
Fig. 15. Recent changes in classication
of the taxa analyzed in this study. Black
lines indicate no changes, dashed black
lines indicate movements of species
among genera, light blue lines indicate
synonymies, and grey lines indicate to
which taxon the known material of the
new taxa described in this study (high-
lighted in grey) had been previously as-
signed to.
411Contributions to Zoology, 85 (4) – 2016
vertical process and the presence of ve pali around it
(Fig. S15B). Moreover, the observation of the skeletal
structure in longitudinal section shows the typical po-
rous mesh of trabeculae and synapticulae observed in
poritiids (Fig. S15D). The size of the calices (3 – 4
mm) and the pattern of septal fusion resemble those
observed in the holotype of Goniopora somaliensis
Vaughan, 1907 (Fig. S15C). In situ images of this spe-
cies in Veron (2000, vol. 39, gs 2–4) show, as also
remarked by the author, a lack of the “thick eshy tis-
sue over the skeleton” typical of the lobophylliids.
However, as no image showing the appearance of the
holotype before collection was published it cannot be
ascertained if these images actually correspond to the
same specimen. On the basis of the skeletal morphol-
ogy described above, this species is, therefore in the
present revision not included in the genus Micromussa
but in the genus Goniopora.
The holotype of Acanthastrea minuta (Moll &
Best, 1984, RMNH Coel. 15275, Fig. S15E, F) shows
all the characters of the genus Acanthastrea, to which
it is restored, rather than those of Micromussa and thus
it is not included in Micromussa in the present study.
This specimen has a well-developed spinose coenos-
teum, less than six septa per 5 mm in terms of septal
spacing, and less than six teeth per septum (Fig. S15F).
Veron (2000) moved this species to the genus Micro-
mussa likely due to the small diameter of the calices.
However, the image of the skeleton published by Veron
(2000, vol. 3: 8) shows a specimen with a different
morphology from that of the holotype with very une-
qual costosepta in relative thickness. On the basis of
these observations the species is returned to the genus
Acanthastrea.
The material from the Indian Ocean identied as
Acanthastrea cf. hillae on the basis of macromor-
phology (Figs 1H, 2H) was recovered in clade E to-
gether with all the Acanthastrea species, including
the genus type species (Fig. 3). Despite a similar
macromorphology to A. hillae and H. bowerbanki for
all the characters examined in this study, including
the size of calices and the number of septa (Table 1),
the SEM observations allowed us to ascertain that
septal granulation in the examined material is weak,
rounded, enveloped by thickening deposits and, thus,
different from that of all the species recovered in
clades A and B (Fig. 14O, T). This suggests that
Acanthastrea specimens with large calices may in-
deed have been previously misidentied as A. hillae
as in the case of the specimen shown by Veron & Pi-
chon (1980, gs 442 and 444). As no specimens of A.
hillae have been collected in the Indian Ocean so far,
either for this study or in the examined museum col-
lections (namely at the MTQ, MNHN, or RMNH),
the records for this species outside the Pacic Ocean
are considered dubious and might be referred as A. cf.
hemprichii until further morpho-molecular studies
will clarify their status.
Final remarks
A growing volume of work dealing with taxonomy
and systematics of scleractinian corals have demon-
strated that the use of both genetic and morphological
analyses is an indispensable approach to understand
and clarify the evolution of these invertebrates (Ben-
zoni et al., 2007, 2010, 2012a, 2012b; Budd et al.,
2010, 2012; Gittenberger et al., 2011; Stolarski et al.,
2011; Kitaha ra et al., 2012a, 2012b; Schmidt-Roach et
al., 2014; Huang et al., 2014b; Arrigoni et al., 2014c;
Ter r a n e o et al., 2014). In this study we combine ro-
bust molecular analyses based on multiple DNA re-
gions with detailed observations of morphology at
colony, corallite, and sub-corallite scales. The present
work increases the current knowledge of taxonomy
and biodiversity of the family Lobophylliidae and,
more generally, highlights the importance of analyz-
ing both genetics and morphology. The inclusion of
as many species as possible and from different local-
ities, as done here for M. amakusensis, in a molecular
phylogenetic framework is a necessary step towards a
comprehensive reconstruction of coral evolutionary
history as an increasing number of previously un-
studied taxa exhibit unexpected phylogenetic place-
ments that have been misunderstood and ignored by
traditional systematics (Fukami et al., 2004a, 2008;
Kita hara et al., 2010, 2012a, 2012b; Stolarski et al.,
2011; Huang et al., 2011, 2014b; Arrigoni et al.,
2014a; Kitano et al., 2014). We strongly encourage
the examination of specimens from multiple locali-
ties, ideally including the entire geographic distribu-
tion range of a species and above all the type locality,
in order to dene the intraspecic morphological
variation and evaluate the possible presence of cryp-
tic or previously overlooked species. Finally, we rec-
ommend also the inclusion of microstructural data to
the presented molecular and micromorphological ob-
servations in future works as they were demonstrated
to be useful and diagnostic in previous systematic
revisions (Budd et al., 2012; Kitahara et al., 2012b;
Huang et al., 2014a: Arrigoni et al., 2014c; Janisze-
wska et al., 2 015 ).
412 Arrigoni et al. – Phylogeny of Micromussa and Homophyllia
Acknowledgements
New Caledonia data and specimens were obtained during the
CORALCAL4 (http://dx.doi.org/10.17600/12100060), BI BE -
LOT (http://dx.doi.org/10.17600/14003700), CORALCAL5
(http://dx.doi.org/10.17600/15004300), and CHEST (http://dx.
doi.org/10.17600/15004500) expeditions on the RV Alis. We are
grateful to the Captain and the crew for their valuable help dur-
ing the campaigns and their knowledge of the Neocaledonian
reefs during eldwork. The participation of F.B. and B.W.H. was
possible thanks to the kind help and support of C. Payri and B.
Dreyfus. We are also grateful to J.L. Menou, J. Butscher, G.
Lasne, E. Folcher, and A. Arnaud. We are grateful to E. Karsenti
(EMBL), É. Bourgois (Tara Expeditions), and the OCEANS
Consortium for sampling in Gambier Island during the Tara
Oceans expedition. We thank the comm itment of the following
people and sponsors who made this singular expedition possible:
CNRS, EMBL, Genoscope/CEA, VIB, Stazione Zoologica An-
ton Dohrn, UNIMIB, ANR (projects POSEIDON/ANR-09-
BLAN-0348, BIOMARKS/ANR-08-BDVA-003, PRO-
METHEUS/ANR-09-GENM-031, and TARA-GIRUS/ ANR-
09-PCS-GENM-218), EU FP7 (MicroB3/No.287589), FWO,
BIO5, Biosphere 2, agnès b., the Veolia Environment Foundation,
Region Bretagne, World Courier, Illumina, Cap L’Orient, the
EDF Foundation EDF Diversiterre, FRB, the Prince Albert II de
Monaco Foundation, Etienne Bourgois, and the Tara schooner
along with its captain and crew. Tara Oceans would not exist
without continuous support from 23 institutes (http://oceans.tara-
expeditions.org). This article is contribution number 44 of the
Tara Oceans Expedition 2009-2012. We are grateful to E. Du-
trieux (CREOCEAN), C.H. Chaineau (Total SA), R. Hirst, and
M. AbdulAziz for allowing and supporting research in Yemen.
We thank M. Pichon, S. Basheen, M.A. Ahmad, A. Suliman, F.N.
Saeed, C. Riva, S. Montano, and A. Caragnano for their help in
various aspects of eld work in Yemen. T.H. Synclair-Taylor is
acknowledged for (the best) in situ images of Homophyllia from
Lord Howe Island, Aust ralia. We are grateful to J. Moore for pro -
viding images of type material from the USNM. Coral collecting
by B.W.H. was done during the Tun Mustapha Park Expedition
2012, which was jointly-organized by WWF Malaysia, Univer-
siti Malaysia Sabah, Sabah Parks (Malaysia) and Naturalis Biodi-
versity Center, with special thanks to Z. Waheed who arranged
the research permits. Coral collecting by A.H.B. was funded by
the ARC Centre of Excellence for Coral Reef Studies. We are
grateful to P. Gentile for technica l help with SEM and P. Gal li for
laboratory support at UN IMIB. R.A. gratefully acknowledges
the National Science Council of Taiwan for Summer Program in
Taiwan 2013. This work is supported by King Abdullah Univer-
sity of Science and Technology’s Red Sea Research Center (CCF.
grant FCC/1/1973-07 to M.L.B.).
References
Arrigoni R, Berumen ML, Terraneo TI, Caragnano A, Bouw-
meester J, Benzoni F. 2015. Forgotten in the taxonomic lit-
erature: resurrection of the scleractinian coral genus Sclero-
phyllia (Scleractinia, Lobophylliidae) from the Arabian
Peninsula and its phylogenetic relationships. Systematics
and Biodiversity 49 : 140 -163.
Arrigoni R, Kitano YF, Stolarski J, Hoeksema BW, Fukami H,
Stefani F, Galli P, Montano S, Castoldi E, Benzoni F. 2014c.
A phylogeny reconstruction of the Dendrophylliidae (Cnida-
ria, Scleractinia) based on molecular and micromorphogical
criteria, and its ecological implications. Zoologica Scripta
43: 661- 688.
Arrigoni R, Richards ZT, Chen CA, Baird AH, Benzoni F.
2014b. Phylogenetic relationships and taxonomy of the coral
genera Australomussa and Parascolymia (Scleractinia, Lo-
bophylliidae). Contributions to Zoology 8 3: 19 5-215.
Arrigoni R, Stefani F, Pichon M, Galli P, Benzoni F. 2012.
Molecular phylogeny of the Robust clade (Faviidae, Mussi-
dae, Merulinidae, and Pectiniidae): An Indian Ocean per-
spective. Molecular Phylogenetics and Evolution 65: 183 -
193.
Arrigoni R, Terraneo TI, Galli P, Benzoni F. 2014a. Lobophyl-
liidae (Cnidaria, Scleractinia) reshufed: pervasive non-
monophyly at genus level. Molecular Phylogenetics and
Evolution 73: 60-64.
Arrigoni R, Vacherie B, Benzoni F, Barbe V. 2016. The com-
plete mitochondrial genome of Acanthastrea maxima (Cni-
daria, Scleractinia, Lobophylliidae). Mitochondrial DNA
27: 927-928.
Bandelt HJ, Forster P, Röhl A. 1999. Median-joining networks
for inferring intraspecic phylogenies. Molecular Biology
and Evolution 16: 37- 48.
Benzoni F. 2013. Echinophyllia tarae sp. n. (Cnidaria, Antho-
zoa, Scleractinia), a new reef coral species from the Gambier
Islands, French Polynesia. ZooKeys 318: 59 -7 9.
Benzoni F, Arrigoni R, Stefani F, Pichon M. 2011. Phylogeny of
the coral genus Plesiastrea (Cnidaria, Scleractinia). Contri-
butions to Zoology 8 0: 231-2 49.
Benzoni F, Arrigoni R, Stefani F, Stolarski J. 2012a. Systemat-
ics of the coral genus Craterastrea (Cnidaria, Anthozoa,
Scleractinia) and description of a new family through com-
bined morphological and molecular analyses. Systematics
and Biodiversity 10: 417-433.
Benzoni F, Arrigoni R, Stefani F, Reijnen BT, Montano S,
Hoeksema BW. 2012b. Phylogenetic position and taxonomy
of Cycloseris explanulata and C. wellsi (Scleractinia: Fun-
giidae): lost mushroom corals nd their way home. Contri-
butions to Zoology 81: 125 -14 6.
Benzoni F, Arrigoni R, Waheed Z, Stefani, F, Hoeksema BW.
2014. Phylogenetic relationships and revision of the genus
Blastomussa (Cnidaria: Anthozoa: Scleractinia) with de-
scription of a new species. TheRaf esBullettinofZoology
62: 358-378
Benzoni F, Stefani F, Pichon M, Galli P. 2010. The name game:
morpho-molecular species bounda ries in the genus Psam-
mocora (Cnidaria, Scleractinia). Zoological Journal of the
Linnean Societ y 160: 421- 456.
Benzoni F, Stefani F, Stolarski J, Pichon M, Mitta G, Galli P.
2007. Debating phylogenetic relationships of the scleractin-
ian Psammocora: molecular and morphological evidences.
Contributions to Zoology 76: 35-54.
Budd AF, Stolarski J. 2009. Searching for new morphological
characters in the systematics of scleractinian reef corals:
comparison of septal teeth and granules between Atlantic
and Pacic Mussidae. Acta Zoologica 90: 142-165.
Budd AF, Fukami H, Smith N, Knowlton N. 2012. Taxonomic
classication of the reef coral family Mussidae (Cnidaria:
413Contributions to Zoology, 85 (4) – 2016
Anthozoa: Scleractinia). Zoological Journal of the Linnean
Society 16 6: 4 65-52 9.
Budd AF, Romano SL, Smith ND, Barbeitos MS. 2010. Re-
thinking the phylogeny of Scleractinian corals: a review of
morphological and molecular data. Integrative and Com-
parative Biolog y 50: 411-427.
Budd AF, Stolarski J. 2011. Corallite wall and septal micro-
structure in scleractinian reef corals: compar ison of molecu-
lar clades within the family Faviidae. Journal of Morphol-
ogy 272: 66-88.
Chevalier JP. 1975. Les Scléractiniaires de la Mélanésie Fran-
çaise (Nouvelle-Calédonie, Iles Chestereld, Iles Loyauté,
Nouvelles Hébrides). Expédition Francaise Sur les Récifs Co-
ralliens de la Nouvelle-Calédonie, Deuxieme Partie 7: 1-407.
Colgan DJ, McLauchlan A, Wilson GDF, Livingston SP, Edge-
combe GD, Macaranas J, Gray MR. 1998. Histone H3 and
U2 snRNA DNA sequences and arthropod molecula r evolu-
tion. Australia n Journal of Zoology 46: 419 -4 37.
Dai CF, Horng S. 2009. Scleractinia fauna of Taiwan II. The
robust group. Taipei: National Taiwan University.
Flot JF, Blanchot J, Charpy L, Cruaud C, Licuanan WY, Naka-
no Y, Payri C, Tillier S. 2011. Incongruence between morpho
-
types and genetically delim ited species in the coral genus
Stylophora: phenotypic plasticity, morphological conver-
gence, morphological stasis or interspecic hybridization?
BMC Ecology 11: 2 2 .
Fukami H, Budd AF, Levitan DR, Jara J, Kersanach R, Knowl-
ton N. 2004b. Geographic differences in species boundaries
among members of the Montastraea annularis complex
based on molecular and morphological markers. Evolution
58 : 32 4 -33 7.
Fukami H, Budd AF, Paulay G, Sole-Cava A, Chen CA, Iwao,
K, Knowlton N. 2004a. Conventional taxonomy obscures
deep divergence between Pacic and Atlantic corals. Nature
427: 832-835.
Fukami H, Chen CA, Budd AF, Collins A, Wallace C, Chuang
YY, Chen C, Dai CF, Iwao K, Sheppard C, Knowlton N.
2008. Mitochondrial and nuclear genes suggest that stony
corals are monophyletic but most families of stony corals
are not (Order Scleractinia, Class Anthozoa, Phylum
Cnidari a). PLoS ONE 3: e3222.
Gittenberger A, Reijnen BT, Hoeksema BW. 2011. A molecu-
larly based phylogeny reconstruction of mushroom corals
(Scleractinia: Fungiidae) with taxonomic consequences and
evolutionary implications for life history traits. Contribu-
tions to Zoology 8 0: 107-132 .
Guindon S, Gascuel O. 2003. A simple, fast, and accurate algo-
rithm to estimate large phylogenies by maximum likeli-
hood. Systematic Biology 52: 69 6 -70 4.
Hodgson G, Carpenter K. 1995. Scleractinian corals of Kuwait.
PacicScience 49: 227-246.
Hoeksema BW. 1989. Taxonomy, phylogeny and biogeography
of mushroom corals (Scleractinia: Fungiidae). Zoologische
Verhandelingen 254: 1-29 5.
Hoeksema BW. 1991. Evolution of body size in mushroom cor-
als (Scleractinia: Fungiidae) and its ecomorphological con-
sequences. Netherlands Journal of Zoology 41: 12 2-139.
Ho e ks em a BW. 1993. Historical biogeography of Fungia (Pleu-
ractis) spp. (Scleractinia: Fungiidae), including a new spe-
cies from the Seychelles). Zoologische Mededelingen 67:
639- 654.
Hoogenboom MO, Frank GE, Blowes SA, Chase TJ, Zawada
KJA, Dornelas M. 2015. Disparity between projected geo-
graphic ranges of rare species: a case study of Echinomor-
pha nishihirai (Scleract inia). Marine Biodiversit y Records
8: e147.
Huang D, Benzoni F, Fukami H, Knowlton N, Smith ND, Budd
AF. 2014a. Taxonomic classication of the reef coral families
Merulinidae, Montastraeidae, and Diploastraeidae (Cnida-
ria: Anthozoa: Scleractinia). Zoological Journal of the Lin-
nean Society 171: 2 77-355.
Huang D, Benzoni F, Arrigoni R, Baird AH, Berumen ML,
Bouwmeester J, Chou LM, Fukami H, Licuanan WY, Lovell
ER, Meier R, Todd PA, Budd AF. 2014b. Towards a phyloge-
netic classication of reef corals: the Indo-Pacic genera
Merulina, Goniastrea and Scapophyllia (Scleractinia, Me-
ru l i nid ae). Zoologica Scripta 43: 531-548.
Huang D, Licuanan WY, Baird AH, Fukami H. 2011. Cleaning
up the “Bigmessidae”: molecular phylogeny of scleractinian
corals from Faviidae, Merulinidae, Pectiniidae, and Trachy-
phylliidae. BMC Evoutionary Biology 11 : 37.
Huang D, Meier R, Todd PA, Chou LM. 2009. More evidence
for pervasive paraphyly in scleractinian corals: systematic
study of Southeast Asian Faviidae (Cnidaria; Scleractinia)
based on molecular and morphological data. Molecular
Phylogenetics and Evolution 50: 10 2-116.
Huelsenbeck JP, Ronquist F. 2001. MRBAYES: Bayesian infer-
ence of phylogenetic trees. Bioinformatics 17: 754-755.
Janiszewska K, Jaroszewcz J, Stolarski J. 2013. Skeletal ontog-
eny in basal scleractinian micrabaciid morals. Journal of
Morphology 274: 243-257.
Janiszewska K, Stolarski J, Benzerara K, Meibom A, Mazur M,
Kitahara MV, Cairns SD. 2011. A unique skeletal micro-
structure of the deep-sea micrabaciid scleractinian corals.
Journal of Morphology 231: 191-203.
Janiszewska K, Stolarski J, Kitahara MV, Neuser ND, Mazur
M. 2015. Microstructural disparity between basal m icra-
baciids and other Scleractinia: new evidence from Neogene
Stephanophyllia. Lethaia. do i: 10.1111 / le t.12119
Katoh K, Standley DM. 2013. MA F FT multiple sequence align-
ment software version 7: improvements in performance and
usability. Molecular Biology and Evolution 30: 772 -780.
Katoh K, Misawa K, Kuma K, Miyata T. 2002. MAFFT: a nov-
el method for rapid multiple sequence alignment based on
fast Fourier tra nsform. Nucleic Acids Resourches 30: 3059-
3066.
Keshavmurthy S, Yang SY, Alamaru A, Chuang YY, Pichon
M, Obura DO, Fontana S, De Palmas S, Stefani F, Benzoni
F, Mac-Donald A, Noreen AME, Chen C, Wallace CC,
Pillay R, Denis V, Amri AY, Reimer JD, Mezaki T, Shep-
pard C, Loya Y, Abelson A, Mohammed MS, Baker AC,
Mostafavi PG, Suharsono BA, Chen CA. 2013. DNA bar-
coding reveals the coral “laboratory-rat”, Stylophora pis-
tillata encompasses multiple identities. ScienticReports
3: 1520.
Kitahara MV, Cairns SD, Stolarski J, Miller DJ. 2012a. Delto-
cyathiidae, an early-diverging family of Robust corals (An-
thozoa, Scleractinia). Zoologica Scripta 42: 201-212.
Kitahara MV, Cairns SD, Stolarski J, Blair D, Miller DJ. 2010.
A comprehensive phylogenetic analysis of the Scleractinia
(Cnidaria, Anthozoa) based on mitochondr ial CO1 sequence
data. PLoS ONE 5: e11490.
414 Arrigoni et al. – Phylogeny of Micromussa and Homophyllia
Kitahara MV, Stolarski J, Cairns SD, Benzoni F, Stake JL,
Mi ller DJ. 2012b. The rst mo der n solitary Agar iciidae (An -
thozoa, Scleractinia) revealed by molecular and microstruc-
tural analysis. Invertebrate Systematics 26: 303-315.
Kitahara, MV, Fukami H, Benzoni F, Huang D. 2016. The new
systematics of Scleractinia: integrating molecular and mor-
phological evidence. In: Goffredo S, Dubinsky Z. (eds.), The
Cnidaria, Past, Present and Future: The World of Medusa
and Her Sisters. Springer Netherlands, Dordrecht.
Kitano YF, Benzoni F, Arrigoni R, Shirayama Y, Wallace CC,
Fukami H. 2014. A phylogeny of the family Poritidae
(Cnidaria, Scleractinia) based on molecular and morpho-
logical analyses. PLoS ONE 9: e98406.
Librado P, Rozas J. 2009. DnaSP v5: A software for comprehen-
sive analysis of DNA polymorphism data. Bioinformatics
25: 1451-1452 .
Matthai G. 1928. A monograph of the recent meandroid As-
traeidae. Catalogue of Madreporarian Corals British Mu-
seum (Natural Histor y) 7: 1-28 8.
Milne-Edwards M, Haime J. 1848. Recherches sur les polypi-
ers; 4eme mémoire. Monographie des Astréides. Annales
des Sciences Naturelles 10: 209-320.
Nylander JAA. 2004. MrModeltest v2. Uppsala: Evolutionary
Biology Centre, Uppsala University.
Obura DO. 2012. The diversity and biogeography of Western
Indian Ocean reef-building corals. PLoS ONE 7: e45013.
Pichon M, Benzoni F, Chaineu CH, Dutriex E. 2010. Field
Guide to the hard corals of the southern coast of Yemen.
Paris: Biotope Par thenope.
Pinzόn JH, Sampayo E, Cox E, Chauka LJ, Chen CA, Voolstra
CR, LaJeunnesse TC. 2013. Blind to morphology: genetics
identies several widespread ecologically common species
and few endemics among Indo-Pacic cauliower corals
(Pocillopora, Scleractinia). Journal of Biogeography 40:
1595-1608.
Rambaut A, Drummond AJ. 2009. Tracer: MCMC Trace Anal-
ys is To ol. Version 1.5. Available via http://beast.bio.ed.ac.uk
Reijnen BT, McFadden CS, Hermanlimianto YT, van Ofwegen
LP. 2014. A molecular and mor phological exploration of the
generic boundaries in the family Melithaeidae (Coelentera-
ta: Octocorallia) and its taxonomic consequences. Molecu-
lar Phylogenetics and Evolution 70: 383-401.
Scheer G, Pillai CSG. 1983. Report on the stony corals from the
Red Sea. Zoologica 133: 1-198.
Schmidt-Roach S, Miller KJ, Lundgren P, Andreakis N. 2014.
With eyes wide open: a revision of species within and close-
ly related to the Pocillopora damicornis species complex
(Scleractinia; Pocilloporidae) using morphology and genet-
ics. Zoology Journal of the Linnean Society 170: 1-33.
Stefani F, Benzoni F, Yang SY, Pichon M, Galli P, Chen CA.
2011. Comparison of morphological and genetic analyses
reveals cryptic divergence and morphological plasticity in
Stylophora (Cnidaria, Scleractinia). Coral Reefs 30: 1033-
1049.
Stolarski J. 2003. 3-Dimensional micro- and nanostructural
characteristics of the scleractinian corals skeleton: a biocal-
cication proxy. Acta Palaeontologica Polonica 48: 497-
530.
Stolarski J, Roniewicz E. 2001. Towards a new synthesis of evo-
lutionary relationships and classication of Scleractinia.
Journal of Paleontology 75: 1090 -1108.
Swofford DL. 2003. PAUP*. Phylogenetic Analysis Using Par-
simony (*and other methods). Version 4. Sunderland: Sin-
auer Associates.
Takabayashi M, Carter DA, Loh WKT, Hoegh-Guldberg O.
1998. A coral-specic primer for PCR amplication of the
internal transcribed spacer region in ribosomal DNA. Mo-
lecular Ecology 7: 925 -931.
Terraneo TI, Berumen ML, Ar rigoni R, Waheed Z, Bouw-
meester J, Caragnano A, Stefani F, Benzoni F. 2014. Pachy-
seris inattesa sp. n. (Cnidaria, Anthozoa, Scleractinia): a
new reef coral species from the Red Sea and its phylogenetic
relationships. Zookeys 433: 1-30.
Vaughan TW, Wells JW. 1943. Revision of the Sub-orders,
Families and Genera of the Scleractinia. Geological Societ y
of America Special Papers 44: 1-363.
Veron JEN. 1985. New Scleractinia from Australian coral
reefs. Records of the Western Australian Museum 12: 147-
183.
Veron JEN. 1992. Hermatypic corals of Japan. Australian Insti-
tute of Marine Science Monograph Series 9: 1-23 4.
Veron JEN. 1993. A biogeographic database of hermatypic
cor
als. Australian Institute of Marine Science Monograph
Series 10: 1-433.
Veron JEN. 1995. Corals in space and time: the biogeography
and evolution of the Scleractinia. New York: Cornell Uni-
versity P ress.
Veron JEN. 2000. Corals of the World. Townsville: Australian
Institute of Marine Science.
Veron JEN, Marsh LM. 1988. Hermatypic corals of Western
Australia: records and annotated species list. Records of the
Western Australian Museum 29: 1-136.
Veron JEN, Pichon M. 1980. Scleractinia of Eastern Australia,
III: Families Agariciidae, Siderastreidae, Fungiidae, Oculi-
nidae, Merulinidae, Mussidae, Pectiniidae, Caryophylliidae,
Dendrophylliidae. Australia n Institute of Marine Science
Monograph Series 4: 1-433.
Waheed Z, van Mil HGJ, Syed Hussein MA, Jumin R, Golam
Ahad B, Hoeksema BW. 2015. Coral reefs at the norther n-
most tip of Borneo: an assessment of Scleractinian species
richness patterns and benthic reef assemblages. PLoS ONE
10: e 0146006.
Wallace CC, Done BJ, Muir PR. 2012. Revision and catalogue
of worldwide staghorn corals Acropora and Isopora (Scler-
actinia: Acroporidae) in the Museum of Tropical Queens-
land. Memoirs of the Queensland Museum 57: 1-255.
Wallace CC, Chen CA, Fukami H, Muir PR. 2007. Recognition
of separate genera within Acropora based on new morpho-
logical, reproductive and genetic evidence from Acropora
togianensis, and elevation of the subgenus Isopora Studer,
1878 to genus (Scleractinia: Astrocoeniidae; Acroporidae).
Coral Reefs 26: 231-239.
Wallace CC, Fellegara I, Muir PR, Harrison PL. 2009. The
scleractinian corals of Moreton Bay, eastern Australia: high
latitude, marginal assemblages with increasing species
rich ness. Memoirs of the Queensland Museum - Nature 54:
1-118.
Wells JW. 1955. Recent and subfossil corals of Moreton Bay,
Queensland. University of Queensland Papers Department
of Geology 4: 1-18.
Wells JW. 1964. The recent solitary mussid scleractinian corals.
Zoologische Mededelingen Leiden 39: 375 -384.
415Contributions to Zoology, 85 (4) – 2016
White TJ, Bruns T, Lee S, Taylor J. 1990. Amplication and
direct sequencing of fungal r ibosomal RNA genes for phylo-
genetics. Pp. 315-322 in: Innis MA, Gelfand DH, Sninsky JJ,
White TJ., eds, PCR Protocols: A Guide to Methods and
Application. San Diego: Academic Press.
Received: 17 August 2015
Revised and accepted: 25 January 2016
Published online: 30 September 2016
Editor: R.W.M. van Soest
Online Supplementary Information
S1. List of the material examined in this study from a molecular point of view. For each specimen we list code,
identication, sampling locality, collector, and molecular markers used for the genetic analyses.
S2. Pairwise comparisons of genetic distance within and between clades of the family Lobophylliidae. Standard
deviations listed on the upper right hand portions for each set of comparisons.
S3. ML tree based on mitochondrial COI dataset. Node values are ML SH-like support (> 0.7).
S4. ML tree based on nuclear histone H3 dataset. Node values are ML SH-like support (> 0.7).
S5. ML tree based on ITS region dataset. Node values are ML SH-like support (> 0.7).
S6. Close-up of phylogenetic relationships among and within clades A and B from the phylogenetic tree reported in
Fig. 3 and based on the combinaed COI, H3, and ITS region dataset.
S7. Coralla of Micromussa lordhowensis (previously Acanthastrea) examined in this study showing the range of
intra-specic macromorphological variability of corallite shape, septal fusion and thickness. A) Holotype, MTQ
G57483; B) AM 5098*; C) AM 5050*; D) AM AU055; E) AM AU008; F) AM 1642*. * = specimens included in the
phylogeny reconstruction in Fig. 3 (clade A).
S8. Intra-specic variability in situ of specimens of Micromussa lordhowensis (previously Acanthastrea) examined
in this study. A) Two adjacent colonies with different colouration; B) A colony with brown and white rounded polyps
and brown peristome; C) A colony with green and white rounded and irregular polyps and white peristome; D) A
colony with brown and white rounded and irregular polyps and green peristome; E) A colony with green and white
elongated polyps; F) Detail of a colony with grey and white irregular polyps and grey peristome; G) A colony with
brown and white elongated and irregular polyps and red peristome; H) A colony with green and white irregularly
shaped polyps and wide red peristome.
S9. Coralla of Micromussa multipunctata (previously Montastraea) examined in this study. A) RMNH Coel 24241;
B) RMNH Coel 40077*; C) RMNH Coel 40078* (same colony as in Fig. S11C); D) RMNH Coel 40099*. * = speci-
mens
included in the phylogeny reconstruction in Fig. 3 (clade A).
S10. Intra-specic variability in situ of specimens of Micromussa multipunctata (previously Montastraea) exam-
ined in this study. A) A colony from Malaysia with “a chocolate-brown coenosarc colouration” described as the
most typical by Hodgson (1985); B) A large colony from Malaysia with bright red and white polyps; C) Same colo-
ny as in Fig. 1C with orange retracted polyps; D) RMNH Coel 40078 from Malaysia, same colony as in Fig. S10C
with bright red polyps.
416 Arrigoni et al. – Phylogeny of Micromussa and Homophyllia
S11. Coralla of some of the specimens of Micromussapacica sp. nov. (referred to as Homophyllia cf. australis in
text) examined in this study showing the intra-specic macromorphological variability of this species. Specimen
code close to the images. All specimens are included in the phylogeny reconstruction in Fig. 3 (clade A).
S12. Coralla of some of the specimens of Homophyllia australis examined in this study showing the intra-specic
macromorphological variability of this species. Specimen code close to the images. All specimens are included in
the phylogeny reconstruction in Fig. 3 (clade B).
S13. Coralla of Homophyllia bowerbanki (previously Acanthastrea bowerbanki (B, D, F, H) and Acanthastrea
hillae (A, C, E, G)) examined in this study showing the overlapping range of macromorphological variability be-
tween the two nominal species. A) Holotype of Acanthastrea hillae, UQF 179; B) Holotype of Acanthastrea bow-
erbanki, MNHN850; C) IRD HS3531*; D) IRD HS3285*; E) AM MH043*; F) IRD HS3446*; G) AM AU184 (left
hand side corallum) growing adjacent to Homophyllia australis (single larger corallite on the right); H) AM AU249,
same colony in Fig. 10H. White arrows point at the two columellae of the bicentric calice central to the corallum in
both holotypes. Black arrows point at the prominent septal teeth of the rst cycle of septa towards the columella.* =
specimens included in the phylogeny reconstruction in Fig. 3 (clade B).
S14. Intra-specic variability in situ of specimens of Homophyllia bowerbanki (previously Acanthastrea bower-
banki and Acanthastrea hillae) examined in this study. A) AM AU183, mottled green and red colony growing adja-
cent to Homophyllia australis (on the right); B) IRD HS3286, dark brown and white mottled colouration with a
prominent central polyp; C) Orange and green colony at Noddy Island, Admiralty Group, Australia (31°30.060’S;
159°3.823’E, 16/12/2015); D) Brown and white coloured colony with a prominent central polyp at Lord Howe Island,
Australia (31°30.695’S; 159°3.426’E, 16/12/2014); E) Beige and white colony with irregularly shaped polyps at Lord
Howe Island, Australia (31°32.996’S; 159°3.796’E, 17/12/2014); F) AM AU174, green, brown and white colony with
large polyps roughly arranged around a central rounder one, Noddy Island, Admiralty Group, Australia (17/12/2015);
G) A large colony with brown polyps and white peristome at Balls Pyramid, Lord Howe Island, Australia
(31°45.402’S, 159°14.253’E, 18/12/2014) (Picture by T.H. Sinclair-Taylor); H) AM AU249, large bright orange colo-
ny with a prominent central polyp (Picture by T.H. Sinclair-Taylor).
S15. Lectoype of Micromussa diminuta Veron, 2000 (A–B, D), holotypes of Goniopora somaliensis (C) and Acan-
thastrea minuta Moll & Borel-Best, 1984 (E–F). A) MTQ G55851, whole specimen; B) Corallites showing the
typical septal fusion and the pali of the genus Goniopora; C) USNM 21989, detail of the corallites; D) Longitudinal
section of the corallites of the same specimen as in A and B showing the porous nature of septa and corallite walls
resulting from the fusion of trabeculae and synapticulae typically observed in Goniopora; E) RMNH Coel 15275;
F) Close up of a corallite showing the typical shape of costoseptal teeth of the genus Acanthastrea. White arrows
point at the single process forming the columella which is surrounded by pali.
417Contributions to Zoology, 85 (4) – 2016
Appendix
Systematics
Family Lobophylliidae Dai & Horng, 2009
Genus Micromussa Veron, 2000
Type species: Acanthastrea amakusensis Veron, 1990
Revised diagnosis. Solitary or colonial with encrusting
to submassive and massive coralla. Budding intracali-
cular and extracalicular. Corallites monomorphic.
Corallite integration discrete. Monticules absent. Cal-
ice width small-medium, medium (4 – 15 mm) and
large (> 15 mm). Septa in 3 – 4 cycles (24 – 36 septa)
or more of four cycles (≥ 48 septa). Free septa irregu-
lar. Between 6 and 11 septa per 5 mm. Costosepta
slightly unequal to unequal in relative thickness. Colu-
mellae trabecular and spongy, > or < 1/4 of calice
width. At midcalice tooth base elliptical-parallel and
tooth tip irregular. Septal teeth height medium (0.3 –
0.6 mm) but can be high (> 0.6 mm) in M.pacica sp.
nov. Tooth space medium (0.3 – 1.0 mm). Interarea
smooth. Granules strong, pointed, scattered on septal
face. Tooth shape equal between rst and third order
septa. Tooth size equal between midcalice and inner
end in the rst two cycles of septa.
Remarks. The main characters differentiating Mi-
cromussa from Homophyllia are the shape and distri-
bution of granules (Table 1) which are strong, pointed,
and scattered in the former genus and weak, rounded,
uniformly distributed in the latter.
Species included.
Micromussa amakusensis (Veron, 1990) (Figs 1A,
2A, 7A–C, 14A, F)
Acanthastrea amakusensis Veron, 1990 p. 137, gs
42– 4 4, 82
Micromussa amakusensis (Veron, 1990) Veron (2000)
vol. 3, pp. 10–11, gs 1–4; Turak and Devantier 2011 p.
164 and gs therein
Acanthastrea lordhowensis Veron & Pichon, 1982;
Turak and Devantier 2011 p. 166 and gs therein
Type material. The holotype (MTQ G32485, Fig. 1A)
from Amakusa Island, Japan is deposited at the MTQ.
Examined material. Japan – MTQ G32485 (holo-
type), Amakusa Island, 10m, 1988, coll. J.E.N. Veron;
SMBL Cni-11051, Amakusa Island, 3.3m, 07/10/2009,
coll. Y. Zayasu; SMBL Cni-11060, Amakusa Island,
3.9m, 07/10/2009, coll. Y. Zayasu; SMBL Cni-11049,
Amakusa Island, 3.6m, 07/10/2009, coll. Y. Zayasu;
SMBL Cni-11055, 4.5m 07/10/2009, coll. Y. Zayasu;
SMBL Cni-11046, Amakusa Island, 2.0m 07/10/2009,
coll. Y. Zayasu.
Distribution. Japan and the Coral Triangle. Previous
records of this species in the Indian Ocean (Veron,
2000) refer to Micromussa indiana sp. n o v.
Micromussa lordhowensis (Veron & Pichon, 1982)
comb. nov. (Figs 1C, 2C, 14C, H, S5, S6)
Acanthastrea sp. Veron & Pichon, 1980 pp. 264–266,
gs 455–456
Acanthastrea lordhowensis Veron & Pichon, 1982 p.
138; Veron (2000) vol. 3, pp. 14–15, gs 1–6; Wallace
et al. (2009) gs 7D, 58A–F
Micromussa amakusensis (Veron, 1990) Wallace et al.
(2009) gs 7F, 59A–B
Type material. The holotype G57483 (Figs 1C, S8A)
from Lord Howe Island, Australia is deposited at the
MT Q.
Examined material. Australia – (coll. A.H. Baird):
AM 1596, AM 1597, Solitary Islands, Sandon Reef,
0.5m, 30/09/2012; AM 1598, Solitary Islands, San-
don Reef, 0.5m, 30/09/2012; AM 1642, Solitary Is-
lands, Sandon Reef, 0.5m, 02/10/2012; AM 5019, AM
5023, AM 5038, Solitary Islands, Bubble Cave, 12m,
08/07/2014; AM 5050, Solitary Islands, Bubble Cave,
12m, 08/07/2014; AM 5063, AM 5079, AM 5085,
AM 5098, AM 5099G, AM 5099R, Solitary Islands,
Anemone Bay, 12m, 08/07/2014; AM MH042, Lord
Howe Island, Admiralty Group, 20/03/2013, 14m,
coll. M. Hoogenboom; MTQ (Moreton Bay collec-
tion, coll. C.C. Wallace, I. Fellegara, P. Muir):
G56540, Moreton Bay, Peel Island, 1m, 01/12/2001;
G6628, Moreton Bay; MTQ (AIMS monograph coral
collection, coll. M. Pichon and J.E.N. Veron): G57483
(holotype), Lord Howe Island, 1m; G58513, Lord
Howe Island; BM(NH).1983.9.212; (coll F. Benzoni):
AM AU008, Mutton Bird Bay (30°18.237’S;
153°9.041’E), 06/12/2014; AM AU009, Mutton Bird
Bay (30°18.237’S; 153°9.041’E), 06/12/2014; AM
AU014, Minnie Waters pools (29°46.732’S; 153°
18.199’E), 07/12/2014; AM AU044, SW Solitary,
08/12/21014; AM AU048, SW Solitary, 08/12/21014;
AM AU055, Split Island (30°14.494’S; 153°10.780’E),
08/12/21014; AM AU060, Split Island (30°14.494’S;
153°10.780’E), 08/12/21014; AM AU120, N Solitary
Bubble (29° 55.644’S153° 23.359’E), 10/12/2014; AM
418 Arrigoni et al. – Phylogeny of Micromussa and Homophyllia
AU121, N Solitary Bubble (29°55.644’S; 153°
23.359’E), 10/12/2014.
Distribution. According to Veron (2000) Western
Australia and the coral triangle. Previous records of
this species in the Indian Ocean (Veron, 2000) refer to
Micromussa indiana sp. nov.
Micromussa multipunctata (Hodgson, 1985) comb.
nov. (Figs 1D, 2D, 14D, I, S7, S8)
Montastrea multipunctata Hodgson, 1985, gs 1–8, 9;
Veron (2000) vol. 3, p. 221, gs 4–7
Micromussa minuta (Moll & Borel-Best, 1984) Veron
(2000) in part, vol. 3, p. 8, skeleton picture
Type material. Four syntypes (UP C-783, UP C-786,
UP C-787, UP C-788) from Tambuli Reef, Mactan Is-
land, Cebu, Philippines are deposited at the UP.
Examined material. Malaysia, North Sabah –
RMNH (coll. B.W. Hoeksema): Coel. 40077, station
TMP20, Banggi Outer NE Reef (07°22.54´N,
117°22.25´E), 8m, 14/09/2012; Coel. 40078, station
TMP20, Banggi Outer NE Reef (07°22.54´N,
117°22.25´E), 9m, 14/09/2012; Coel. 40099, station
TMP19, Outer Latoan Patch (07°22.54´N,
117°22.25´E), 8m, 14/09/2012; Philippines – RMNH
Coel. 24241 Mactan Island, Cebu, 1-5m, 01/05/1981;
UP P1L02161, Talim Point, Batangas, Philippines,
6.6m, 22/08/2009, coll. D. Huang.
Distribution. From the Coral Triangle to the northern
central Pacic (Veron, 2000).
Micromussa indiana Benzoni & Arrigoni sp. nov.
(Figs 1B, 2B, 5, 6, 7D–F, 14B, G)
Micromussa amakusensis (Veron, 1990) Claereboudt
(2006) pp. 226–227 and gures therein; Pichon et al.
(2010) pp. 232–233, gs 1–4
Type material. The holotype (MNHN IK-2012-14232,
Figs 1B, 2B, 5A–B) from Al Mukallah, Yemen is de-
posited at MNHN.
Examined material. Yemen, Red Sea – UNIMIB
(coll. F. Benzoni): KA099, Tiqfash Island, Kamaran
Islands (15°42.057’N; 42°23.230’E), 30/09/2009;
KA113, Tiqfash Island, Kamaran Islands (15°42.033’N;
42°23.144’E), 30/09/2009; KA119, Tiqfash Island,
Kamaran Islands (15°42.033’N; 42°23.144’E),
30/09/2009; Yemen, Gulf of Aden – MNHN-
IK-2012-14232, Al Mukallah (14°30.801’N;
49°10.338’E), 20/03/2007, coll. F. Benzoni and M. Pi-
chon; UNIMIB (coll. F. Benzoni): AD069, Aden
(12°45.267’N; 44°54.983’E), 10/03/2009; BA072, Bir
Ali (13°59.180’N; 48°15.692’E), 19/11/2008; BA117,
Bir Ali (13°55.648’N; 48°23.234’E), 22/11/2008;
BU001, Burum (14°18.480’N; 48°57.899’E),
22/03/2007; FP, Al Mukallah (14°30.801’N;
49°10.338’E), 20/03/2007; MU183, Al Mukallah
(14°30.801’N; 49°10.338’E), 20/03/2008; MU184, Al
Mukallah (14°30.801’N; 49°10.338’E), 20/03/2008;
MU186, Al Mukallah (14°30.801’N; 49°10.338’E),
20/03/2008; MU215, Al Mukallah (14°31.046’N;
49°10.285’E), 21/03/2008; Yemen, Socotra Island –
EPA S C3695, 01/03/1999; UNIMIB SO071, Di Hamri
(12°40.518’N; 54°11.394’E), 14/03/2010, coll. F. Ben-
zoni; MTQ G57461 Ras Bidou; Kenya – RMNH Coel.
17290 Watamu Marine National Reserve, Mayungu,
3m, 06/04/1983, coll. H. Moll.
Etymology. The name refers to the currently known
geographic distribution of this species, restricted to
the Indian Ocean.
Description. Colonial with encrusting coralla. Bud-
ding intracalicular and extracalicular. Corallites ir-
regularly polygonal in outline. Calice width 5 – 10
mm, diameter variable between calices of the same
corallum. Septa in 3 – 4 cycles (24 – 36 septa), those of
the rst two cycles (the second sometimes incomplete)
reach the columella, those of the other cycles, often
incomplete, are free. Costosepta are unequal in rela-
tive thickness (Fig. 5B). Columellae trabecular and
spongy < 1/4 of calice width (Fig. 5C). Septal teeth
height medium (0.3 – 0.6 mm). Interarea smooth.
Granules strong, pointed, scattered on septal face.
Holotype. The specimen is an irregularly shaped
fragment of a colony, it measures 12 x 8.5 x 2 cm and
has an encrusting growth form (Fig. 5A). The coral-
lites are irregularly polygonal in outline and their larg-
est diameter ranges from 0.5 to1 cm (Figs 1B, 5B). Up
to 4 cycles of septa are visible in the largest corallites,
while only the rst three are present in smaller coral-
lites, sometimes the second and the third ones incom-
plete (Fig. 5B). Septa of the rst 2 cycles reach the
columella. Columellae are trabecular and measure
from 1 to 2.5 mm in diameter. In vivo, the colony had
a greenish colour with bright red peristome (Fig. 2B).
Remarks. Within the genus, this species is distin-
guished from M. amakusensis which has slightly larg-
er corallites, 4 cycles of more regularly arranged septa
the rst 3 of which reach a larger columella (Fig. 7A)
(> 1/4 of calice width) made of more numerous pro-
cesses (Fig. 7A, C). Septal teeth are more regularly
spaced along the septum in M. amakusensis (Fig. 7B)
419Contributions to Zoology, 85 (4) – 2016
than in M. indiana (Fig. 7E). The latter has smaller
corallites and less numerous septa than M. lordhowen-
sis and is readily told apart from M. multipunctata
which is plocoid, and from M.pacicasp. n ov. which
has much larger and predominantly solitary coralla
with more numerous septa (Table 1).
Colour. The peristome has a different colour from
the rest of the polyp all the combinations of bright red,
orange, brown, green, grey and white have been ob-
served (Fig. 6).
Ecology. This species is found between 2 and 10 m
depth in protected environments where it grows on
hard substrates often in proximity of pockets of sedi-
ment. The thin colonies are often infested by poly-
chaetes, boring bivalves, cirripeds and gall crabs (e.g.
the holotype in Fig. 5A).
Distribution. Southern Red Sea, North-western
Gulf of Aden, Socotra Island, Oman, Kenya.
Micromussa pacica Benzoni & Arrigoni sp. nov.
(Figs 1E, 2E, 8, 9, 12D–F, 14E, J, S9)
Scolymia australis Veron & Pichon 1980 in part, gs
425–426, 428–429; Veron, 2000 in part, vol. 3 pp. 70–
71, gs 1, 6
Type material. The holotype (MNHN IK-2012-16043,
Figs 8F, 9D, S11) and a paratype (MNHN IK-2012-
16044, Fig. 8G–I), both from Taravai Island, Gambier
Islands, French Polynesia are deposited at MNHN.
Examined material. Australia – RMNH Coel.
16306, Erith Island, Murray Pass, May 1974; AM
AU133, Lord Howe Island (31°31.076’S; 159°3.929’E),
14/12/2014, coll. F. Benzoni; AM AU141, Lord Howe
Island (31°33.224’S; 159°4.480’E), coll. F. Benzoni,
15/12/2014; New Caledonia – MNHN IK-2012-16045,
Loyalty Islands, Ouvéa Island, Bagaat (20°37.448’S;
166°16.252’E), 21/02/14, coll. F. Benzoni, BIBELOT
Expedition; MNHN IK-2012-16046, Loyalty Islands,
Ouvéa Island (20°25.426’S; 166°29.098’E), 22/02/14,
coll. F. Benzoni, BIBELOT Expedition; IRD HS3202,
Cap Begat (21°21.242’S; 165°52.142’E), 19/04/2012,
coll. F. Benzoni; IRD (BIBELOT Expedition, Loyalty
Islands, coll. F. Benzoni): HS3327, Loyalty Islands,
Maré Island, south of Cap Machan (21°24,967’S;
167°49,152’E), 15/02/14; HS3359, Maré Island, Tadine
Bay (21°35.283’S; 167°51.258’E), 15/02/14; HS3471,
Lifou Island, Cap Ainé Martin (20°46.857’S;
167°02.272’E), 20/02/14; HS3527, Astrolabe Reef,
Oues Reef (19°52.194’S; 165°33.375’E), 24/02/2014;
HS3528, Astrolabe Reef, Oues Reef (19°52.194’S;
165°33.375’E), 24/02/2014; French Polynesia, Gam-
bier Islands – MNHN IK-2012-14249, coll. M. Seurat,
15/09/1902; MNHN (Tara Oceans Expedition, coll. F.
Benzoni): IK-2012-16043, Taravai Island (23°08.72’S;
135°03.09’W), 08/07/2011; IK-2012-16044, Taravai Is-
land (23°08.72’S; 135°03.09’W), 08/07/2011; UNIMIB
(Tara Oceans Expedition, coll. F. Benzoni): GA130,
NE Lagoon pinnacles (23°04.12’S; 134°55.83’W),
05/07/2011; GA150, Taravai Island (23°08.72’S;
135°03.09’W), 08/07/2011; GA186, Tekava Island
(23°10.13’S; 134°51.51’W), 10/07/11; French Polyne-
sia, Austral Islands – MTQ (International collection):
G64026, Rapa Island, Tarakoi; G64051, Rapa Island,
Baie Aurei; G64052, Rapa Island, Baie Aurei.
Etymology. The name refers to the geographic distri-
bution of this species, restricted to the central and
western Pacic Ocean.
Description. Solitary (Fig. 8A–B) to polystomatous
(Fig. 8C) or forming coralla up to 3 corallites (Fig.
8D–I). Budding intracalicular (Fig. 8C) and extracali-
cular (Fig. 8F–I). Corallites round in outline. Calice
width 15 – 25 mm. Septa in 4 – 5 cycles (≥ 48 septa),
those of the rst three cycles reach the columella,
those of the fourth fuse with those of the third just
before reaching the columella, and those of the fth or
more are free and often incomplete (Fig. 8D–E). Cos-
tosepta slightly unequal to unequal in relative thick-
ness (Fig. 8D). Columellae trabecular and spongy, <
1/4 of calice width (Fig. 8D–F). Septal teeth height
medium (0.3 – 0.6 mm) but can be high (> 0.6 mm) in
some specimens. Tooth space medium (0.3–1 mm). In-
terarea smooth. Granules strong, pointed, scattered on
septa l face.
Holotype. The specimen includes two corallites
growing on a slab of dead tabular Acropora (Figs 8F,
S9). The largest corallite is rounded in shape and it
measures 2.3 cm in diameter, the smaller one, budding
from the former extracalicularly, is oval in shape and is
0.9 cm in diameter (shown by the arrow in Fig. 8F).
The living colony included a third even smaller coral-
lite (Fig. 9D) which was sampled for molecular analy-
ses. The largest corallite presents 5 cycles of septa, the
fourth one incomplete. Septa of the rst 3 cycles reach
the columella which is trabecular and 3.5 mm in diam-
et er. In vivo, the colony had bright red peristome (Fig.
9D).
Remarks. Within the genus, this species is readily
distinguished from all the others by the size of corallites
and number of corallites and septa. The macromorphol-
ogy and the in situ appearance of this species are closer
to those of H. australis, with similar coral
lum shape
420 Arrigoni et al. – Phylogeny of Micromussa and Homophyllia
and number of septa, than to any other congener with-
in Micromussa. However, H. australis has overall larg-
er corallites and wider septal tooth spacing. Septal
teeth are between 2 to 4 mm apart in H. australis (Fig.
12B–C) and between 0.5 and 2.5 mm apart in in M.
pacica(Fig. 12E–F). In fact, the character which per-
haps is most useful for telling the two species apart is
the increase in tooth size between midcalice and inner
end in the rst two cycles of septa observed in H. aus-
tralis (arrows in Fig. 10B–C) but not in the new spe-
cies. In particular, the tallest septa on primary septa in
H. australis can reach up to 4 mm while (e.g. Fig. 10D)
while they reach an observed maximum of 2.5 mm in
M. pacica(e.g. Fig. 8E).
Colour. The peristome has a different colour from
the rest of the polyp all the combinations of bright red,
orange, brown, green, grey and white have been ob-
served (Fig. 9). In situ, this species and H. australis
can co-occur (for example, in New Caledonia and Aus-
tralia) and are found in the same habitats. They can be
told apart by the more uniform within polyp coloura-
tion in the former (Fig. 9), and the more frequent and
abundant presence of white areas in the peristome and,
especially, over the septa in the latter (Fig. 11).
Ecology. This species is found between 10 and 40 m
depth in protected environments with high sedimenta-
tion as well as in the deeper parts of outer reef slopes.
The holotype and other specimens were growing on
hard substrate coming from coral rubble consisting of
fragments of dead tabular Acropora (Fig. S11).
Distribution. French Polynesia (Gambier and Aus-
tral Islands), New Caledonia (Grand Terre and Loyalty
Islands), and eastern Australia.
Remarks. Two specimens of this species were sam-
pled in the Gambier Islands in 1902 by M. Seurat, then
director of the Biological Reseach Laboratory in
Rikitea (1902-1904), and deposited in the MNHN gen-
eral collection (IK-2012-14249). They had been previ-
ously identied as Mussa costata Dana, 1846 and Sco-
lymia ? by different authors.
Genus Homophyllia Brüggemann, 1877
Type species: Homophyllia australis (Milne Edwards
& Haime, 1849)
Revised diagnosis. Solitary or colonial with encrusting
to submassive and massive coralla. Budding mainly
intracalicular, extracalicular budding is also observed.
Corallites monomorphic although central larger and
rounder corallites are often observed in colonial cor-
alla. Corallite integration discrete. Monticules absent.
Calice width large (> 15 mm). Septa in more than four
cycles (≥ 48 septa). Free septa irregular. Between 6 and
11 septa per 5 mm. Costosepta unequal in relative
thickness. Columellae trabecular and spongy, < 1/4 of
calice width. At midcalice tooth base elliptical-paral-
lel and tooth tip irregular. Septal teeth height high (>
0.6 mm). Tooth space wide (> 1 mm). Interarea smooth.
Granules weak, rounded, uniformly distributed on
septal face. Tooth shape equal between rst and third
order septa. Tooth size increases between midcalice
and inner end in the rst two cycles of septa.
Remarks. The main characters differentiating
Homophyllia from Micromussa are the shape and dis-
tribution of granules (Table 1) which are weak, round-
ed, uniformly distributed in the former genus and
strong, pointed, and scattered in the latter.
Species included.
Homophyllia australis (Milne Edwards & Haime,
1849) (Figs 1F, 2F, 10, 11, 122A–C, 14K, P, S10)
Caryophyllia australis Milne Edwards & Haime, 1849
p. 320, pl. 8 g. 2
Scolymia cf. vitiensis (Bruggemann, 1877) Veron &
Pichon (1980) g. 412
Scolymia australis (Milne Edwards & Haime, 1849)
Veron & Pichon (1980) in part, gs 427, 429–431; Ve-
ron (2000) in part vol. 3, pp. 70–71, gs 2–5
Scolymia (= Homophyllia) australis (Milne Edwards
& Haime, 1849) Budd & Stolarski (2009) gs 2K, 4K,
5A, 6K, 7E, 7K, 11H
Type material. Two syntypes NHMUK 1840.11.30.77
and NHMUK 1840.11.30.79) from South Australia.
Original designation Brüggemann, 1877: 310.
Examined material. Australia – AM 4631, Lord
Howe Island, 19/03/2013, 1m, coll. A.H. Baird; AM
AU176, Noddy Island, Admiralty Group (31°30.060’S;
159°3.823’E), 16/12/2014, coll. F. Benzoni; AM AU177,
Noddy Island, Admiralty Group (31°30.060’S;
159°3.823’E), 16/12/2014, coll. F. Benzoni; AM AU178
Noddy Island, Admiralty Group (31°30.060’S;
159°3.823’E), 16/12/2014, coll. F. Benzoni; New Cal-
edonia – MNHN IK-2012-14248, coll. J.P. Chevalier;
IRD HS3311, coll. F. Benzoni; IRD (BIBELOT Expe-
dition, Loyalty Islands, coll. F. Benzoni): HS3441,
Luengoni, Lifou Island (21°01.798’S; 167°24.710’E),
18/02/2014; HS3447, Doking Bay, Lifou Island
(20°42.813’S; 167°09.284’E), 19/02/2014; HS3469,
Cap Ainé Martin, Lifou Island (20°46.857’S;
167°02.272’E), 20/02/2014; HS3470, Cap Ainé Mar-
tin, Lifou Island (20°46.857’S; 167°02.272’E),
421Contributions to Zoology, 85 (4) – 2016
20/02/2014; HS3524, Astrolabe Reef (19°52.194’S;
165°33.375’E), 24/02/2014; HS3525, Astrolabe Reef
(19°52.194’S; 165°33.375’E), 24/02/2014; HS3526, As-
trolabe Reef (19°52.194’S; 165°33.375’E), 24/02/2014;
IRD (BIBELOT Expedition, Grande Terre, coll. F.
Benzoni): HS3544, Port Bouquet, Nenii Reef
(21°40.193’S; 166°24.373’E), 25/02/2014; HS3545,
Port Bouquet, Nenii Reef (21°40.193’S; 166°24.373’E),
25/02/2014; IRD (CORALCAL5 Expedition, Ile des
Pins, coll. F. Benzoni): HS3591, Ilôt Ndié (22°31.560’S;
167°12.151’E), 29/09/2015; IRD (CHEST Expedition,
Chesterelds and Bellona, coll. F. Benzoni): HS4177,
Ilot Loop (19°53’.763’S; 158°27.996’E), 19/11/2015.
Distribution. Western Pacic, North-west to South-
west Australia, New Caledonia.
Remarks. Veron and Pichon (1980) record this spe-
cies also “from the western Pacic east to the Marshall
Islands and Fiji”. We did not examine material from
this region and cannot ascertain if the mentioned re-
cord actually refers to this species or to Micromussa
pacicasp. nov., which has likely been confused with
it in the literature so far (Veron and Pichon, 1980).
Homophyllia bowerbanki (Milne Edwards & Haime,
1857) comb. nov. (Figs 1G–H, 2G–H, 14L–M, Q–R,
S13, S14)
Acanthastrea bowerbanki Milne Edwards & Haime,
1857 p. 503, pl. D6, g. 1; Veron (2000) vol. 3 p. 26,
gs 1–3; Wallace et al. (2009) g. 56
Acanthastrea cf. bowerbanki (Milne Edwards &
Haime, 1857) Veron & Pichon (1980) gs 449–454
Acanthastrea hillae Wells, 1955 pl. 2, gs 2– 3; Chev-
alier (1975) pl. XXX, g. 2; Veron & Pichon (1980) gs
440–441, 443, 445–447; Veron (2000) vol. 3 pp. 28–
29, gs 1–5; not Claereboudt (2006) pp. 214–215 and
gures therein; not Dai and Horng (2009) p. 80 and
gures therein; Wallace et al. (2009) g. 57
Acanthastrea cf. hillae Wells, 1955 Chevalier (1975)
pl. XXX, g. 4 (Holotype of A. bowerbanki)
Type material. The holotype (scle850, Fig. S13A) from
Australia is deposited at the MNHN..
Examined material. Australia – AM 4629, Lord
Howe Island, 1m, 19/03/2013, coll. A.H. Baird; AM
MH019, Lord Howe Island, Admiralty Group,
19/03/2013, 14m, coll. M. Hoogenboom; AM MH043,
Lord Howe Island, Admiralty Group, 20/03/2013, 14m,
coll. M. Hoogenboom; AM MH046, Lord Howe Is-
land, Admiralty Group, 14m, 20/03/2013, coll. M.
Hoogenboom; MTQ (Moreton Bay collection, coll.
C.C. Wallace, I. Fellegara, P. Muir): G58484, Strad-
broke Island, 8m, 22/05/2005; G58486, Goat Island,
2m, 23/05/2005; G56536, Moreton Bay, Peel Island,
01/02/2002; G55343, Moreton Bay, Goat Island, 2m,
23/02/2005; MTQ (AIMS monograph coral collection,
Coll. M. Pichon and J.E.N. Veron): G43111, Byron Bay,
1m; G43110, Dewar Island, 10m; G58034, Lord Howe
Island, 20m; G58033, Lord Howe Island, 20m; G58032,
Lord Howe Island, 1m; G58035, Heron Island, 10m;
G58036, Lord Howe Island, 1m; University of Miyaza-
ki AuB167, Moreton Bay, 20/07/2007, 7m, coll. H. Fu-
kami; New Caledonia – IRD (CORALCAL4 Expedi-
tion, coll. F. Benzoni): ): HS3066, Yandé Island
(20°03.620’S; 163°47.312’E), 12/04/2012; HS3163,
Moneo (21°03.354’S; 165°35.297’E), 18/04/2012;
HS3169, Cap Bocage (21°10.578’S; 165°36.202’E),
18/04/2012; HS3225, Bogota Reef (21°25.352’S;
165°00.207’E), 21/04/2012; HS3285, Canal Woodin
(22°23.800’S; 166°46.440’E), 25/04/2012; HS3286,
Canal Woodin (22°23.800’S; 166°46.440’E),
25/04/2012; HS3287, Canal Woodin (22°23.800’S;
166°46.440’E), 25/04/2012; HS3288, Canal Woodin
(22°23.800’S; 166°46.440’E), 25/04/2012; HS3298,
Recif Kué (22°38.638’S; 166°36.292’E), 26/04/2012;
IRD (BIBELOT Expedition, Loyalty Islands, coll. F.
Benzoni): HS3438, Luengoni (21°01.798’S;
167°24.710’E), 18/02/2014; HS3501, Beautemps-Beau-
pré, Motu One (20°22.141’S; 166°07.115’E),
23/02/2014; HS3531, Nenii Reef (21°40.193’S;
166°24.373’E), 25/02/2014; HS3525, Astrolabe Reef
(19°52.194’S; 165°33.375’E), 24/02/2014; HS3526, As-
trolabe Reef (19°52.194’S; 165°33.375’E), 24/02/2014;
HS3446, Cap Escarpé, Lifou Island (20°41.039’S;
167°13.596’E), 19/02/2014; HS3489, Ouvéa Island
(20°26.663’S; 166°24.016’E), 22/02/2014; IRD
(CHEST Expedition, Chesterelds and Bellona, coll.
F. Benzoni): HS3983, Ilot Reynard (19°12.788’;
158°56.806’E), 10/11/2015. Japan – SMBL Cni-10422,
Nishidomari, Kochi Prefecture, 3.0m, 13/01/2006,
coll. H. Fukami; SMBL Cni-10113, Yokonami, Kochi
Prefecture, 8.4m, 22/10/2008, coll.Y. Zayasu; SSH12,
Shirahama, Wakayama Prefecture, 3.0m, 05/2005 coll.
H. Fukami.
Distribution. Western Pacic with high latitudinal
distribution in the north and in the south, North and
Eastern Australia, New Caledonia. Previous records of
A. hillae in the Indian Ocean (Veron, 2000) refer to
Acanthastrea hemprichii.
Genus Australophyllia Benzoni & Arrigoni gen. nov.
Type species: Australophyllia wilsoni (Veron, 1985)
comb. nov.
422 Arrigoni et al. – Phylogeny of Micromussa and Homophyllia
Etymology. The name refers to the peculiar and nar-
row geographic distribution of this genus.
Diagnosis. Colonial with submassive to massive
coralla. Budding intracalicular and extracalicular.
Corallites monomorphic. Corallite uniserial and dis-
crete. Monticules present. Calice width medium (4 –
15 mm). Septa in more than four cycles (≥ 48 septa).
Free septa irregular. Between 6 and11 septa per 5 mm.
Costosepta unequal in relative thickness. Adjacent
corallite centers discontinuous with lamellar linkage.
Columellae trabecular and spongy, < 1/4 of calice
width in series although it can be larger in monocen-
tric corallites. At midcalice tooth base elliptical-paral-
lel and tooth tip irregular. Septal teeth height medium
(0.3 – 0.6 mm). Tooth space medium (0.3 – 1 mm).
Interarea smooth. Granules weak, rounded, scattered
on septal face. Tooth shape equal between rst and
third order septa. Tooth size equal between midcalice
and inner end in the rst two cycles of septa.
Remarks. The main characters differentiating Aus-
tralophyllia from the other genera in this study are the
uniserial corallites, which are discrete in all the other
examined taxa, the discontinuous corallite centre link-
age as opposed to the absence of linkage in the other
taxa. Moreover, in Australophyllia and the granules on
septal sides are weak, rounded, scattered while they
are strong, pointed, and scattered in Micromussa,
weak, rounded, uniformly distributed in Homophyllia,
and weak, rounded, enveloped by thickening deposits
in Acanthastrea (Table 1).
Species included (monotypic genus).
Australophyllia wilsoni (Veron, 1985) comb. nov.
(Figs 1I, 2I, 13, 14N–S)
Symphyllia wilsoni (Veron, 1985) pp. 169-170, gs 18-
22; Veron, 2000, vol. 3, p. 53, gs 2–4
Type material. The holotype (Z910) from Rat Island,
Houtman Abrolhos Islands, Western Australia is de-
posited at the WAM. Two paratypes (Z911, Z912) are
deposited at the WAM.
Examined material. Western Australia – (coll.
D.P. Thomson): AM WIL1, Hall Bank, 09/04/2013,
9m; AM WIL2, Hall Bank, 09/04/2013, 9m; AM
WIL3, Hall Bank, 09/04/2013, 9m; AM WIL4, AM
WIL5; RMNH Coel. 22399, West Australia, Abrolhos
Islands, 2m, 03/04/1976; WAM 168-84 (holotype), Rat
Island, Houtman Abrolhos Islands, Western Australia,
8m, 1983, coll. J.E.N. Veron; paratype (WAM 169-84),
Port Denison, Western Australia, 9m, 1982, J.E.N. Ve-
ron; paratype (WAM 170-84), Port Denison, Western
Australia, 12m, 1982, coll. J.E.N. Veron.
Distribution. This species is restricted to South-
western Australia and thrives in temperate conditions.
Genus Acanthastrea Milne Edwards & Haime,