Blackwell Science, LtdOxford, UKZOJZoological Journal of the Linnean Society0024-4082The Lin-

nean Society of London, 2003? 2003

139?
419437
Original Article

SYSTEMATICS OF THE HEXACORALLIAM. DALY

ET AL.

Zoological Journal of the Linnean Society, 2003, 139, 419–437. With 3 figures

Systematics of the Hexacorallia (Cnidaria: Anthozoa)

MARYMEGAN DALY*, DAPHNE G. FAUTIN and VALERIE A. CAPPOLA †
Department of Ecology and Evolutionary Biology, University of Kansas and Division of Invertebrate Zoology,
University of Kansas Natural History Museum and Biodiversity Research Center, Lawrence KS 66045, USA

Received January 2003; accepted for publication June 2003

Downloaded from https://academic.oup.com/zoolinnean/article/139/3/419/2624250 by guest on 31 January 2023

The composition of and relationships among higher-level groups within the anthozoan subclass Hexacorallia
(= Zoantharia) has been controversial because independent analyses of anatomy, life history, ultrastructure, and
molecular sequences have failed to provide a consistent framework for drawing taxonomic boundaries or under-
standing phylogenetic relationships. The relationship among stony corals (order Scleractinia), sea anemones (order
Actiniaria), and corallimorpharians (order Corallimorpharia) has been particularly problematic. We synthesize exist-
ing studies and provide new anatomical and molecular evidence that bear on the question of ordinal circumscription
and relationships. We find that orders Actiniaria, Antipatharia, Ceriantharia, Corallimorpharia, Scleractinia, and
Zoanthidea are monophyletic; Corallimorpharia is most closely related to Scleractinia. We infer that many tradi-
tional diagnostic characters are shared primitive features and thus poor indicators of phylogenetic relationships.
Although the major nodes of the hexacorallian tree are well supported by multiple types of data, questions about
skeletal evolution and subordinal taxonomy remain unanswered pending denser taxonomic and character sampling.
© 2003 The Linnean Society of London, Zoological Journal of the Linnean Society, 2003, 139, 419–437.

ADDITIONAL KEYWORDS: Actiniaria – Anthozoa – Cnidaria – combined data – evolution – Scleractinia –

systematics.

INTRODUCTION etic, but its relationship to other orders is obscure

(Hyman, 1940; Wells & Hill, 1956; Berntson et al.,
Subclass Hexacorallia (= Zoantharia) of cnidarian
1999). Recent attention to the problem of circumscrip-
class Anthozoa currently contains six orders: Actini-
tion, taxonomy, and phylogeny within the Hexacoral-
aria (sea anemones), Antipatharia (black corals),
lia using molecular sequences and ultrastructural
Ceriantharia (tube anemones), Scleractinia (stony
information has only increased confusion about the
corals), Corallimorpharia (corallimorpharians), and
boundaries and histories of the various hexacorallian
Zoanthidea (zoanthids). A mosaic of traits not exclu-
orders. Hypotheses and information from various
sive to any one group diagnoses each order (Table 1).
sources have not been synthesized or reconciled, mak-
The composition of these groups, the ranks assigned to
ing it impossible to assess the state of knowledge,
them, and the hypothesized phylogenetic relation-
evaluate proposed taxonomic schemes for the Hexa-
ships among them have changed dramatically over
corallia, or explore the evolution of major features like
time (McMurrich, 1891; Stephenson, 1921; Hyman,
the scleractinian skeleton, the arrangement of mesen-
1940; Carlgren, 1944, 1949; Berntson, France & Mul-
teries, or coloniality.
lineaux, 1999). The monophyly of the two largest
We conduct simultaneous analyses using data from
orders, Actiniaria and Scleractinia, has been chal-
anatomy, biology, ultrastructure of sperm, cnidae, life
lenged (e.g. Stephenson, 1921; Schmidt, 1974; Chen
history, and molecular sequences to provide a compre-
et al., 1995); Corallimorpharia may also be polyphyl-
hensive assessment of phylogenetic relationships
etic (e.g. Duerden, 1900; Chen et al., 1995; Pires &
among hexacorallians. The assembled matrix provides
Castro, 1997). Ceriantharia is considered monophyl-
a way to explore phylogenetic relationships and char-
acter evolution, and serves as an assessment of the
state of knowledge of the various hexacorallian orders.
*Corresponding author. E-mail: dalym@ku.edu
†
Present address: Department of Communication Science and A comprehensive study of all of the relevant evidence
Disorders, Emerson College, Boston MA 02116, USA. should result in well-defined, monophyletic groups

© 2003 The Linnean Society of London, Zoological Journal of the Linnean Society, 2003, 139, 419–437 419
420 M. DALY ET AL.

Table 1. Summary of the traditional diagnostic morphological features for each hexacorallian order. The characterization
of scleractinians (and corallimorpharians) as lacking a marginal sphincter muscle is questionable; the manifestation of this
feature is actually unknown for most species in these groups

Marginal
Mesenterial sphincter Mesentery
Order Exoskeleton Habit filament Siphonoglyph muscle arrangement

Actiniaria Absent Solitary or Unilobed or None to two Endodermal, Monomorphic

clonal trilobed or more mesogloeal, or dimorphic
or none coupled pairs
Antipatharia Proteinaceous Colonial Unilobed Two None Monomorphic
coupled pairs
Ceriantharia Absent Solitary Trilobed One None Monomorphic

Downloaded from https://academic.oup.com/zoolinnean/article/139/3/419/2624250 by guest on 31 January 2023

couples
Corallimorpharia Absent Solitary or Unilobed None None?? Monomorphic
clonal coupled pairs
Scleractinia Calcareous Solitary or Unilobed None None?? Monomorphic
colonial coupled pairs
Zoanthidea Absent Solitary or Trilobed One Endodermal, Dimorphic
colonial mesogloeal, coupled pairs
or none

(Kluge, 1989; Nixon & Carpenter, 1996). Because tax- the calcareous, radial projections that flank the
onomy is increasingly expected to represent or accom- mesenteries of scleractinians (Vaughan & Wells,
modate historical relationships between taxa (e.g. de 1943; Bayer, Grasshoff & Verseveldt, 1983). Mesen-
Queiroz & Gauthier, 1992), phylogenetic and taxo- teries that extend from column wall to actinopharynx
nomic issues are explored simultaneously. We find are termed perfect (or complete); imperfect (incom-
that although the trees resulting from independent plete) mesenteries do not extend all the way to the
analyses of each type of data set differ, the phyloge- actinopharynx (Fig. 1). In some orders, mesenteries
netic signals of the morphological, 16S, 18S, and 28S are added simultaneously in series called cycles;
sequence data are congruent. We infer from the com- mesenteries of different cycles may differ in morphol-
bined molecular and morphological data that Coralli- ogy. The free edge of the mesentery is elaborated into
morpharia and Scleractinia form a clade to the a unilobed or trilobed mesenterial filament that bears
exclusion of Actiniaria. Many subordinal groups, espe- gland cells and cnidae.
cially within the Scleractinia, are para- or polyphyl- Cnidae are the defining characteristic of phylum
etic, confounding comparisons between our results Cnidaria, and are especially important in anthozoan
and hypotheses that use higher level taxonomic taxonomy (Doumenc & Van Praët, 1987; Fautin, 1988;
groups as terminal taxa. Fautin & Mariscal, 1991). The diversity of nemato-
cysts is greatest within Hydrozoa, but Anthozoa is
characterized by the greatest degree of cnidae diver-
HEXACORALLIAN MORPHOLOGY AND TAXONOMY sity, as its members have two types of cnidae in addi-
Despite the basic structural simplicity imposed by tion to nematocysts (Watson & Wood, 1988; Fautin &
the absence of organs or organ systems, the anatomy Mariscal, 1991). Spirocysts have a tubule with thread-
of hexacorallian polyps can be quite complex (e.g. like mini-tubules that entwine prey or other objects;
Fautin & Mariscal, 1991). The cylindrical polyp is these cnidae have been described as restricted to
closed on the distal end by the oral disc; the proximal Hexacorallia (e.g. Mariscal, 1974). Ptychocysts have
end may be an adhesive pedal disc, a rounded physa, an unarmed tubule that is pleated rather than heli-
or a thin sheet of tissue that lines the skeletal cup. cally folded; these cnidae are found only in cerianthar-
The oral disc is perforated by a central mouth that ians, and are used in the construction of their felt-like
leads to a tubular actinopharynx; the actinopharynx tube (Mariscal, Conklin & Bigger, 1977). The morphol-
varies in length but never extends the length of the ogy, size, and distribution of nematocysts are used to
column. The internal cavity, or coelenteron, is divided define groups, particularly within Actiniaria and Cor-
by mesenteries, sheets of tissue that extend from the allimorpharia, in which attributes of cnidae character-
column wall. The term ‘septa’, which is sometimes ize families, genera, or species (e.g. Carlgren, 1940,
used instead of ‘mesenteries’, should be reserved for 1945; Allcock, Watts & Thorpe, 1998).

© 2003 The Linnean Society of London, Zoological Journal of the Linnean Society, 2003, 139, 419–437
 SYSTEMATICS OF THE HEXACORALLIA 421

influenced by environmental and ecological factors

M (Carlgren, 1942; Veron, 1995; Chen, Willis & Miller,
A
m 1996; Miller & Babcock, 1997). Species descriptions of
B actiniarians, ceriantharians, corallimorpharians, and
zoanthideans generally include an inventory of the
C types of cnidae (= cnidom); although the cnidom is
D unknown for most species of scleractinians, it has been
detailed in a few systematic works (e.g. Carlgren,
I
1940, 1945; Schmidt, 1972, 1974; den Hartog, 1980;
II
Hidaka, 1992; Pires, 1997; Pires & Castro, 1997).
Most of the traditional diagnostic features of each
A. B. order are inapplicable to members of other hexacoral-

Downloaded from https://academic.oup.com/zoolinnean/article/139/3/419/2624250 by guest on 31 January 2023

lian groups, complicating the assessment of shared
history among hexacorallians (Lang, 1984). The cal-
careous exoskeleton defines Scleractinia, and its
variation is the basis of intra-ordinal taxonomy (e.g.
Wells, 1956); it has no homologue in the other orders.
Among the soft-bodied forms, taxonomy focuses on
exclusive features such as the mesenterial arrange-
ment of zoanthideans or cnidae and cnidae-bearing
structures of Ceriantharia (e.g. Herberts, 1987; Tiffon,
1987).
Comparison across hexacorallian orders is further
complicated by terminology. Terms describing modes
C. of asexual reproduction are not standardized (Fautin,
2002). Budding is used as a general term for sclerac-
Figure 1. Schematic cross-section through hexacorallians tinian asexual reproduction and to describe several
at the level of the actinopharynx, showing arrangement of
specialized modes of asexual reproduction in particu-
mesenteries. The radial lines represent mesenteries, the
lar scleractinians (e.g. Kramarsky-Winter & Loya,
central oval represents the actinopharynx. The filled oval
1996), corallimorpharians (e.g. Chadwick-Furman &
on each mesentery represents the retractor muscle. A, hex-
Spiegel, 2000), and zoanthideans (e.g. Ryland, 1997a).
amerously arranged, paired, coupled mesenteries typical of
Actiniaria, Scleractinia, and Corallimorpharia. Mesenter-
Although given different names, the process of longi-
ies labelled A, B, I, and II are perfect; C and D are imper- tudinal fission in actiniarians and corallimorpharians
fect. A and B, C and D, and I and II are paired; I and II are closely parallels the process of extratentacular bud-
coupled with A and B. B, hexamerously arranged, paired, ding in scleractinians (Cairns, 1988).
coupled mesentery arrangement typical of Zoanthidea. The nomenclature of cnidae is similarly problem-
Mesenteries labelled M and m are part of a dimorphic pair; atic; terms referring to a type of nematocyst with an
the macrocneme (M) is larger than microcneme (m). C, abrupt transition between the basal and distal tubule
unpaired, coupled arrangement typical of Ceriantharia. In include ‘penicillus’ (e.g. Stephenson, 1929; den Hartog,
ceriantharians, the longitudinal muscle of the mesentery is 1980), ‘p-mastigophore’ (e.g. Weill, 1934; Carlgren,
not hypertrophied into a separate retractor muscle; the 1940; Cutress, 1955; Mariscal, 1974), and ‘p-rhabdoid’
filled ovals indicate the surface of the mesentery on which (e.g. Schmidt, 1969, 1974; Pires & Castro, 1997).
the longitudinal muscle fibres run. Hexacorallian phylogeny is unresolved not simply
because of character incompatibility or terminological
imprecision, but because basic anatomical features do
Despite the apparent simplicity of these features, not circumscribe the same exclusive groups (Table 1).
their diversity has not been adequately described. For example, mesenterial filament histology links
Cappola & Fautin (2000) found that Ptychodactiaria zoanthideans and actiniarians; mesentery arrange-
belongs within Actiniaria, and concluded that the sep- ment groups actiniarians with scleractinians and cor-
aration of order Ptychodactiaria was based on miscon- allimorpharians. The boundaries and ranks of higher-
ceptions about the anatomy and histology of level hexacorallian groups have shifted with focus on
ptychodactiarians. Internal anatomy and histology different subsets of these characters. The seemingly
are virtually unknown for most scleractinians and mosaic distribution of some biological traits may
antipatharians (e.g. Moseley, 1881; Duerden, 1902b; reflect ecological plasticity or some other external
Grigg & Opresko, 1977; Lang, 1984). Anatomy and life influence, or may indicate that certain groups as cur-
history can be difficult to interpret as both can be rently constituted are heterogeneous.

© 2003 The Linnean Society of London, Zoological Journal of the Linnean Society, 2003, 139, 419–437
422 M. DALY ET AL.

ACTINIARIA, CORALLIMORPHARIA, AND SCLERACTINIA for membership in Scleractinia; all extant hexacoral-
lian species with a calcareous exoskeleton belong to
Much of the confusion about ordinal relationships cen- Scleractinia. However, the skeleton may have arisen
tres on the two largest orders, Actiniaria and Sclerac- multiple times, and may not be homologous between
tinia, and their relationship to Corallimorpharia. major clades of corals (Fautin & Lowenstein, 1994;
Members of Corallimorpharia have attributes of both Romano & Palumbi, 1996; Romano & Cairns, 2000;
Actiniaria and Scleractinia. Corallimorpharia has Stanley & Fautin, 2001). The phylogenetic value of the
been considered the sister-group to Scleractinia (e.g. calcareous skeleton has been re-evaluated in other
Duerden, 1898), a subgroup within Actiniaria (e.g. cnidarian groups: all hydrozoans with a calcareous
Stephenson, 1921), and a separate group, equal in exoskeleton were initially grouped together in Hydro-
rank to Actiniaria and Scleractinia (e.g. Carlgren, corallina, but its constituent groups Milleporina and
1942, 1949). Stylasterina are now considered more closely related

Downloaded from https://academic.oup.com/zoolinnean/article/139/3/419/2624250 by guest on 31 January 2023

A corallimorpharian polyp resembles an actiniarian to skeletonless taxa than to one another, indicating
in lacking an external or internal skeleton. The tenta- that calcareous skeletons have been lost or have
cle arrangement of a corallimorpharian is similar to evolved independently (Petersen, 1979). Similarly,
that of an anemone in the family Stichodactylidae, in Foraminifera, previously thought to include only
that multiple tentacles arise from the space between shelled amoebae, includes naked species (Pawlowski
members of a mesenterial pair (Duerden, 1898; den et al., 1999).
Hartog, 1980; Dunn, 1981); more typically in hexacor-
allians, a single tentacle arises from each intermesen-
terial space (Stephenson, 1921, 1928). Duerden (1898, STRATEGIES AND SOLUTIONS
1900) considered corallimorpharians closely related to Higher-level systematics within the morphologically
stichodactyline anemones; within his Stichodactyli- simple Hexacorallia has been re-invigorated by the
nae, he recognized similarities in tentacle arrange- advent of technologies that provide additional charac-
ment and morphology between corallimorpharian ters for phylogenetics and taxonomy. Molecular
genera such as Ricordea and actiniarian genera such sequence and protein data have been used to assess
as Stoichactis (now termed Stichodactyla) not shared phylogenetic relationships among hexacorallian sub-
with other corallimorpharians like Corynactis or Rho- classes (France et al., 1996; Berntson et al., 1999;
dactis. Stephenson (1921) noted that internal anat- Won, Rho & Song, 2001), orders (Fautin & Lowenstein,
omy of Corynactis, Rhodactis, and Ricordea, although 1994; Chen et al., 1995; Song & Won, 1997; Daly, Lip-
within the range of variation seen among actiniarians, scomb & Allard, 2002), and families (Romano &
was unlike that of Stoichactis and other stichodac- Palumbi, 1996; Chen et al., 1996; Romano & Cairns,
tyline anemones, and suggested that they be consid- 2000; Chen, Wallace & Wolstenholme, 2002). Ultra-
ered Madreporaria (= Scleractinia) of unknown structure and cnidae have been explored in the con-
affinity. Carlgren (1942) used differences in histology text of intra- and interordinal relationships (Schmidt,
to justify removing Corynactis, Rhodactis, Ricordea, 1972, 1974; Schmidt & Zissler, 1979; den Hartog,
and other corallimorpharians from Stichodactylidae, 1980; Steiner, 1993; Pires, 1997; Pires & Castro, 1997;
creating the order Corallimorpharia. Harrison & Jamieson, 1999). Because gene sequences,
With respect to internal anatomy and histology, a protein similarity, and ultrastructural features of
corallimorpharian polyp resembles a scleractinian sperm and nematocysts are potentially available for
polyp in having a unilobed mesenterial filament, lack- all hexacorallians, these kinds of data circumvent the
ing basilar muscles, and, in most species, lacking a problem of incompatibility that plague traditional
marginal sphincter muscle (den Hartog, 1980; Dunn, diagnostic features.
1982). Additionally, the cnidom of corallimorpharians Despite the promise of these new technologies,
is more like that of scleractinians: both scleractinians molecular and ultrastructural data have proven no
and corallimorpharians bear holotrichous nemato- better than anatomy or life history at resolving phy-
cysts in the tentacles (Duerden, 1898; Stephenson, logenetic or taxonomic problems within Anthozoa. For
1921; den Hartog, 1980; Pires & Castro, 1997); these example, Ceriantharia was recognized as a third
holotrichs are of a size and morphology not generally anthozoan subclass (either alone or combined with
found in Actiniaria and entirely absent in Stichodac- Antipatharia into subclass Ceriantipatharia) because
tylidae (Dunn, 1981). its members are morphologically and developmentally
Corallimorpharia and Scleractinia have been sepa- unlike other hexacorallians (Hyman, 1940; Wells &
rated primarily because a corallimorpharian does not Hill, 1956). Evidence from gene sequences (Berntson
secrete a skeleton (Hertwig, 1882a,b; Duerden, 1898; et al., 1999) and from ultrastructure (Schmidt, 1974;
Stephenson, 1921). The calcareous exoskeleton has Mariscal et al., 1977; Schmidt & Zissler, 1979) con-
been regarded as a necessary and sufficient criterion firms the distinctiveness of ceriantharians but offers

© 2003 The Linnean Society of London, Zoological Journal of the Linnean Society, 2003, 139, 419–437
 SYSTEMATICS OF THE HEXACORALLIA 423

few clues to their kinship. Sequences of nuclear large of relationships between scleractinians and soft-bod-
subunit ribosomal genes (28S) were interpreted by ied hexacorallians. However, because skeletal features
Chen et al. (1995) as supporting corallimorpharian figure prominently in hypotheses of coral evolution
polyphyly. Sequences from other nuclear (18S) and (Wells, 1956; Scrutton & Clarkson, 1991; Veron, 1995),
mitochondrial (16S) ribosomal genes have been con- and in scenarios relating scleractinians to actiniarians
strued as supporting Corallimorpharia as a grade and corallimorpharians (Hand, 1966; Fautin &
most closely related to a monophyletic Scleractinia Lowenstein, 1994; Stanley & Fautin, 2001), we include
(Berntson et al., 1999; Romano & Cairns, 2000), or as them in our analyses.
supporting a monophyletic Corallimorpharia within a
polyphyletic Scleractinia (Daly et al., 2002). There is
no way to arbitrate among these conflicting results TAXA
because no character or character system performs We gathered morphological data for hexacorallians for

Downloaded from https://academic.oup.com/zoolinnean/article/139/3/419/2624250 by guest on 31 January 2023

significantly better than any other, and no character is which two of the three most commonly sequenced
free from concern about homoplasy. Although molecu- genes (16S, 18S, and 28S) are available in GenBank
lar data are not as obviously influenced by ecological (Table 2). To minimize the effect of this restriction on
or environmental factors as some morphological char- our taxonomic sample, we concatenated sequences for
acters, gene evolution is not free from constraint (e.g. different genes from two members of a genus into a
Brown et al., 1982; Swofford et al., 1996). Further- single row of data when sequences were not available
more, the independence of each ultrastructural detail for the same species. We sequenced 16S, 18S, and 28S
or of each position within a particular region of a gene genes for a few species whose phylogenetic positions
is poorly understood. Sperm morphology may be cor- have been uncertain (Table 2). All putative hexacoral-
related with reproductive mode rather than phylog- lian orders are represented by two or more exemplars.
eny; in general, the sperm of brooding species of Although hexacorallian monophyly is well estab-
Scleractinia differs in structure and morphology lished (Schmidt, 1974; France et al., 1996; Berntson
from that of broadcast spawning species (Harrison et al., 1999), basal relationships within the subclass
& Jamieson, 1999). Finally, ultrastructural and are unclear. We used Alcyonium and Virgularia, mem-
sequence data alone are inappropriate for redefining bers of Alcyonaria, the sister-subclass of Hexacorallia,
taxonomic boundaries because these features are not as outgroups. Alcyonarians and hexacorallians differ
readily evaluated, are unknown for many species, and sufficiently in morphology that the correspondence of
are not accessible to many of the people who use structures is difficult to assess.
taxonomy.

MATRIX AND ANALYSIS

MATERIAL AND METHODS The combined matrix includes 782 characters that are
parsimony-informative for these 48 hexacorallians
MORPHOLOGICAL CHARACTERS
(Table 3): 436 sites in mitochondrial 16S rDNA, 222
All the anatomical and biological attributes in our sites in 18S rDNA, 62 sites in nuclear 28S rDNA, and
morphological data set have been used to diagnose or 55 morphological features. The anatomical, biological,
differentiate higher taxa within the Hexacorallia. and ultrastructural data were taken primarily from
Coloniality and possession of algal symbionts are com- species descriptions. We made and examined histolog-
monly used taxonomic characters within Scleractinia ical sections of many species included in this analysis
(Wells, 1956; Veron, 1995, 2000a); clonality is poten- to supplement or confirm published information. The
tially important in Actiniaria (Fautin & Smith, 1997; incompleteness of the morphological matrix (Table 3)
Pearse & Francis, 2000; Geller & Walton, 2001). reflects our understanding of hexacorallian anatomy
Although life history is typically constant within spe- and biology; this list will be amended, refined, and
cies (but see Glynn et al., 1996), life history character- broadened as data accumulate on character systems
istics like sexuality, mode of gamete release, and common to all hexacorallians.
methods of asexual reproduction may vary within gen- Our 16S rDNA alignment builds on that of Romano
era or families (e.g. Dunn, 1982; Veron, 1995; Pearse, & Cairns (2000); our 18S rDNA alignment builds on
2002). that of Daly et al. (2002). The 28S rDNA sequences
The shape of a colony, method of colony formation, were aligned using the program DAPSA (Harley,
and attributes of the individual calices are the basis of 1996), which was also used to adjust the 16S and 18S
scleractinian taxonomy (e.g. Wells, 1956; Veron, 1995, alignments to accommodate additional taxa.
2000a; Wallace, 1999). Because skeletal characteris- The combined matrix was analysed using the
tics have no homologue in the other hexacorallian ‘island-hopping’ algorithms (Nixon, 1999a) in NONA
orders, these features may seem ill-suited to analyses (Goloboff, 1995). These algorithms allow many

© 2003 The Linnean Society of London, Zoological Journal of the Linnean Society, 2003, 139, 419–437
424 M. DALY ET AL.

Table 2. Taxa included in the study, with GenBank accession numbers for sequences used. Genera marked with an aster-
isk are represented by multiple species. Higher taxonomic categories are from Wells (1956) Dunn (1982), and Cappola &
Fautin (2000). Carlgren (1949) used ‘tribe’ to refer to groups of families within Actiniaria; actiniarian tribes are thus equiv-
alent to suborders or superfamilies

Order Suborder or Tribe Family Genus 16S 18S 28S

Actiniaria Acontiaria Aiptasiidae Aiptasia AY345875 AY046885 U69684

Actiniaria Endomyaria Actiniidae Anemonia* – X53498 U69685
Actiniaria Endomyaria Actiniidae Anthopleura AF375815 Z21671 –
Actiniaria Endomyaria Actiniidae Bunodosoma* AF375814 U52974 –
Actiniaria Ptychodacteae Preactiidae Dactylanthus AY345877 AF052896 AY345873
Actiniaria Athenaria Edwardsiidae Edwardsia – AF254376 AY345870

Downloaded from https://academic.oup.com/zoolinnean/article/139/3/419/2624250 by guest on 31 January 2023

Actiniaria Acontiaria Metridiidae Metridium AF000023 U19550 –
Actiniaria Athenaria Edwardsiidae Nematostella – AF254382 AY345871
Actiniaria Endomyaria Stichodactylidae Stichodactyla* AY345874 U52977 U69687
Actiniaria Endomyaria Actiniidae Urticina U91750 – AY345872
Antipatharia – Antipathidae Antipathes – AF100943 AY026365
Antipatharia – Antipathidae Stichopathes U40286 AF052899 –
Ceriantharia – Cerianthidae Ceriantheopsis U40289 AF052898 –
Ceriantharia Cerianthidae Cerianthus U40288 AF052897 –
Corallimorpharia – Actinodiscidae Amplexidiscus AY345878 – U69692
Corallimorpharia – Corallimorphidae Corynactis* U40293 AF052895 U69689
Scleractinia Faviina Anthemiphylliidae Anthemiphyllia AF265596 – AF265652
Scleractinia Dendrophylliinae Dendrophylliidae Balanophyllia* AF265587 U52973 AF265626
Scleractinia Caryophylliina Caryophylliidae Caryophyllia AF265599 – AF265642
Scleractinia Caryophylliina Caryophylliidae Catalaphyllia L76000 – AF265637
Scleractinia Faviina Faviidae Cladocora AF265612 – AF265633
Scleractinia Caryophylliina Caryophylliidae Crispatotrochus AF265600 – AF265640
Scleractinia Faviina Meandrinidae Dichocoenia AF265607 – AF265635
Scleractinia Dendrophylliinae Dendrophylliidae Enallopsammia U40294 – AF052885
Scleractinia Caryophylliina Flabellidae Flabellum AF265582 – AF265649
Scleractinia Fungiina Fungiidae Fungia L76005 AF052884 AF265631
Scleractinia Fungiina Poritidae Goniopora* L76008 – U65515
Scleractinia Faviina Merulinidae Hydnophora L76009 – U65526
Scleractinia Dendrophylliinae Dendrophylliidae Leptopsammia AF265579 – AF265628
Scleractinia Faviina Mussidae Lobophyllia L76013 – AF265624
Scleractinia Caryophylliina Flabellidae Monomyces AF265583 – AF265651
Scleractinia Faviina Oculinidae Oculina AF265601 – AF265636
Scleractinia Caryophylliina Caryophylliidae Paracyathus AF265603 – AF265644
Scleractinia Fungiina Agaraciidae Pavona L76016 AF052883 AF263350
Scleractinia Faviina Rhizangiidae Phyllangia AF265605 AF052887 AF265641
Scleractinia Caryophylliina Flabellidae Placotrochus AF265589 – AF265650
Scleractinia Faviina Faviidae Platygyra AF265611 – AF263361
Scleractinia Caryophylliina Caryophylliidae Polycyathus AF265606 – AF265643
Scleractinia Fungiina Poritidae Porites L76020 – AF265630
Scleractinia Dendrophylliinae Dendrophylliidae Rhizopsammia – Z92907 AF265629
Scleractinia Astrocoeniina Astrocoeniidae Stephanocoenia* AF265582 – AF265623
Scleractinia Caryophylliina Caryophylliidae Thalamophyllia AF265590 – AF265638
Scleractinia Dendrophylliinae Dendrophylliidae Tubastraea L76022 Z92906 AF265625
Scleractinia Dendrophylliinae Dendrophylliidae Turbinaria AF265609 – U65513
Scleractinia Caryophylliina Caryophylliidae Vaughanella AF265595 – AF265646
Scleractinia Fungiina Fungiidae Zoopilus L76024 – AF265632
Zoanthidea Brachycnemina Sphenopidae Palythoa* AF282932 AF052892 –
Zoanthidea Macrocnemina Parazoanthidae Parazoanthus AF398921 U42453 –

© 2003 The Linnean Society of London, Zoological Journal of the Linnean Society, 2003, 139, 419–437
 SYSTEMATICS OF THE HEXACORALLIA 425

Table 3. Morphological characters used in this study. Characters are given for genera listed in Table 2; a dash indicates
that the state assignment for a particular character is unknown or inapplicable for a given taxon. Genera are grouped by
order; within each order, genera are listed alphabetically

1–10 11–20 21–30 31–40 41–50 51–56

Actiniaria
Aiptasia 2--1110100 0-0001-110 10011--011 11-100---- --01100001 100011
Anemonia 00-0110100 -001010110 1000111101 111100---- --11100001 ------
Anthopleura 2000110100 0001111110 1000111101 111100---- --11110001 010010
Bunodosoma 2000110100 0001011110 1000111101 111100---- --11110001 010010
Dactylanthus 0--00-0-01 -000011110 1010011100 01--00---- --10000001 ------
Edwardsia 0000000100 1000001111 000011-001 022100---- --01100001 000001
Metridium 2000010110 0100010110 1001101011 111100---- --01100--1 010011

Downloaded from https://academic.oup.com/zoolinnean/article/139/3/419/2624250 by guest on 31 January 2023

Nematostella 0001000100 1000001111 000011-001 022100---- --01100001 000001
Stichodactyla 2--0110111 -000111110 1000111101 111-00---- --01100001 ------
Urticina 01-0000100 -000110110 1000101100 111100---- --01-10--1 010010
Antipatharia
Antipathes 1010--0000 -00001-1-0 00000--00- 000-020--- --1---0--1 110001
Stichopathes 1010--0000 -00001-1-0 00000--00- 000-020--- --1---0--1 110001
Ceriantharia
Ceriantheopsis 0-1--010-0 -00000010- 010011-002 0--000---- --10101001 010011
Cerianthus 0-1--010-0 -00000010- 010011-002 0--000---- --10101001 010011
Corallimorpharia
Amplexidiscus 0-00100011 000002-110 10000--001 000000---- --11100001 ------
Corynactis 2000-10110 101002-110 10000--102 000000---- --11100111 011111
Scleractinia
Anthemiphyllia 0--0000100 --1002-110 10000--00- 000-01110- 42----0--1 ------
Balanophyllia 0100000100 --1002-110 10000--00- 000-01110- 5211100111 0111-1
Caryophyllia 000-000100 --1002-111 10000----- 000-01110- 0011100--1 ------
Catalaphyllia 1--01-0100 -01002-110 10000--00- 000-010000 32----0--1 ------
Cladocora 1---1-010- --1002-110 10000----- 01--011104 -2111-0--1 01111-
Crispatotrochus 0---0-010- --1002-110 10000----- 01--01110- 52----0--1 ------
Dichocoenia 1---120110 --1002-110 10000----- 01--011003 21----0--1 ------
Enallopsammia 1--0-10-00 --1002-110 10000--00- 000-010101 3-----0--1 ------
Flabellum 0---0-0100 --1002-111 10000----- 00--01100- 4011100--1 ------
Fungia 0-00--0101 2-1002-110 10000--00- 000-01110- 4211100111 001111
Goniopora 110-110100 --1002-110 10000----- 00--011-03 00111-0--1 0-----
Hydnophora 1-1-12010- --1002-110 10000----- 00--011-01 22----0--1 ------
Leptopsammia 0---0-010- --1002-110 10000----- 00--01110- 52111-0--1 ------
Lobophyllia 101-12010- --1002-111 10000----- 00--011-0- 22----0--1 ------
Monomyces 0---0-010- --1002-110 10000----- 00--011-0- 5-----0--1 ------
Oculina 1--0110100 --1002-110 10000--00- 000-011101 12----0--1 ------
Paracyathus 0---0-010- --1002-110 10000----- 00--011104 12----0--1 ------
Pavona 10-0120100 --1002-110 10000--00- 000-01--02 0--1100--1 011111
Phyllangia 1--0010100 --1002-110 10000--00- 000--11100 1211100--1 ------
Placotrochus ----0-010- --1002-110 10000----- 00--01--0- ------0--1 ------
Platygyra 1-101-0100 --1002-110 10000--00- 000-011100 22----0--1 011111
Polycyathus 1---01010- --1002-110 10000----- 00--011104 12----0--1 ------
Porites 1-00110100 --1002-110 00000--00- 000-01--01 0111100101 111101
Rhizopsammia 1--0010100 --1002-110 10000--00- 000-011000 1211100--1 ------
Tubastraea 1-10010100 --1002-110 10000--00- 000-011100 121---0101 ------
Turbinaria 100-1-0110 --1002-110 10000----- 00--011-02 1-----0--1 011111
Thalamophyllia 0---0-010- --1002-110 10000----- 00--01010- 5-----0--1 ------
Vaughanella 0---0-010- ---002-110 10000----- 00--01110- 52----0--1 ------
Zoopilus 1---12010- --1002-110 10000----- 00--01-103 02111-0--1 ------
Zoanthidea
Palythoa 1-10130100 -00000-111 12001--011 001110---- --11100001 010001
Parazoanthus 1-00130100 -00000-111 12001--101 001110---- --1--00001 100001

© 2003 The Linnean Society of London, Zoological Journal of the Linnean Society, 2003, 139, 419–437
426 M. DALY ET AL.

topologies to be sampled, and thus approximate an morphological features characterize the large clade, it
exhaustive search (Nixon, 1999b). We ran four repli- is supported by substitutions in both the 16S and the
cations of 1000 iterations, using a random seed tree 18S genes. Monophyly of the clade containing
and sampling 78 characters (10% of the matrix). Two Antipatharia, Scleractinia, and Corallimorpharia is
trees were held at each step. Character evolution was supported by 14 molecular substitutions, unilobed
explored using Winclada (Nixon, 1999a). Combined mesenterial filaments, and weak parietal muscles.
evidence trees were compared to trees supported by Scleractinia and Corallimorpharia group to the exclu-
each single data set. To examine whether any subset sion of Antipatharia based on 42 molecular substitu-
contains a phylogenetic signal that differs signifi- tions, a paired secondary cycle of mesenteries, and
cantly from that of the combined matrix, we used tests sperm in which the centriolar complex has parallel
of incongruence-length differences (Farris et al., 1994; proximal and distal centrioles linked by a centriolar
100 replicates, maximum trees held 100 000). ligament.

Downloaded from https://academic.oup.com/zoolinnean/article/139/3/419/2624250 by guest on 31 January 2023

Scleractinia contains two groups that have been
called the ‘complex’ and ‘robust’ clades (e.g. Romano &
RESULTS
Palumbi, 1996; Romano & Cairns, 2000; Chen et al.,
All 12 primary trees (L = 3501; CI = 0.62; RI = 0.64) 2002; Fig. 3). Although the names refer to skeletal
agree on the monophyly of each hexacorallian order as attributes, only molecular substitutions unambigu-
currently defined (Figs 2, 3). We found no significant ously support these clades in our analysis: the robust
difference in phylogenetic signal between the com- clade is characterized by three 16S and two 28S sub-
bined data set and the morphological, 16S, 18S, or 28S stitutions; the complex clade is united by 11 16S and
subsets. three 28S substitutions.
The subclass Hexacorallia and each of its constitu- Actiniaria contains three clades, roughly corre-
ent orders is monophyletic. Monophyly of the subclass sponding to the groups Carlgren (1949) referred to as
Hexacorallia is supported by 22 18S substitutions and the Endomyaria (Actinia, Anemonia, Anthopleura,
two morphological characters: trilobed mesenterial fil- Bunodosoma, Stichodactyla, Urticina, and the ptycho-
aments and solitariness. The most basal group within dactiarian Dactylanthus), Acontiaria (Metridium and
Hexacorallia is the Ceriantharia; members of this Aiptasia), and Edwardsiidae (Edwardsia and Nema-
order share non-retractile tentacles arranged in labial tostella) (Figs 2, 3). The endomyarian clade includes
and marginal cycles, ptychocysts, and 25 18S substi- actiniarians that have an endodermal marginal
tutions. The monophyly of the sister-group to Ceri- sphincter muscle (or no sphincter); this clade is fur-
antharia, a clade containing all other hexacorallian ther characterized by asymmetrical sperm and one
orders, is supported by nine 18S substitutions, 18S substitution. Most endomyarians included in this
microbasic p-mastigophores, and paired mesenteries. analysis are members of the largest family of actini-
Actiniaria is the only order characterized solely by arians, Actiniidae; Actiniidae is polyphyletic with
molecular data; the putative order Ptychodactiaria respect to the ptychodactiarian Dactylanthus and the
lies within Actiniaria. Zoanthidea is characterized by stichodactyline Stichodactyla. The possession of acon-
dimorphic mesentery pairs, mesogloeal canals, mesen- tia – nematocyst-packed threads borne on the edge of
teries added only in the ventrolateral exocoels, and 54 some mesenteries – is unique to the members of the
molecular substitutions. Antipatharia is characterized acontiarian clade. Members of the acontiarian clade
by a proteinaceous skeleton, non-retractile polyps, also share a mesogloeal marginal sphincter muscle,
spherically headed sperm, and eight molecular and eight 18S substitutions. The members of Edward-
substitutions. Corallimorpharia is characterized by siidae share relatively longer tentacles in the outer-
having multiple tentacles per endocoel and 16 molec- most cycle, fertile directive mesenteries, restricted
ular substitutions. Scleractinia is characterized by a retractor and parietal muscles, sperm without a
calcareous exoskeleton with a columella, macrobasic nuclear depression, and five 18S substitutions.
p-mastigophores in the filaments, and 20 molecular
substitutions. Additionally, the scleractinians in-
cluded in this study share a sequence inversion (not DISCUSSION
coded separately from the rest of the alignment) at
PHYLOGENETIC RELATIONSHIPS
position 148 in the 28S alignment; this three-base
segment, read as ‘CCT’ in scleractinians and as ‘TCC’ The proposed sister-group relationship for Zoan-
in all other hexacorallians, is flanked by conserved thidea–Antipatharia–Scleractinia–Corallimorpharia
sequences. contradicts much of the historical literature that
Based on these data, Hexacorallia is divisible into at groups Actiniaria, Corallimorpharia, and Sclerac-
least two major clades, Actiniaria and Zoanthidea–Anti- tinia. Antipatharians are poorly known, and have
patharia–Corallimorpharia–Scleractinia. Although no been presumed to be primitive hexacorallians,

© 2003 The Linnean Society of London, Zoological Journal of the Linnean Society, 2003, 139, 419–437
 SYSTEMATICS OF THE HEXACORALLIA 427

Ceriantheopsis
Ceriantharia
3 25 Cerianthus

Edwardsia
Nematostella
5 5
Aiptasia
Metridium
2 8
8 Actiniaria
Urticina
2 3 Bunodosoma
2 22 2 1
Dactylanthus
A Stichodactyla
2 1
Anemonia
* Anthopleura
B

Downloaded from https://academic.oup.com/zoolinnean/article/139/3/419/2624250 by guest on 31 January 2023

1 8
2 9 *
Palythoa
Zoanthidea
Parazoanthus
3 30 24

Antipathes
6 5 Antipatharia
Stichopathes
3 8
Amplexidiscus
Corynactis
Corallimorpharia
1 16
Pavona
C
2 13 1 Thalamophyllia
7
Stephanocoenia
Monomyces
3 2
Placotrochus
7 1
Porites
3 19 23 3 2 15 1
3 Leptopsammia
Tubastrea
10 6
3 69 5
3 2
Flabellum
Goniopora
14
Balanophyllia
3 1 Rhizopsammia
2
Anthemiphyllia
Vaughanella
Scleractinia
Caryophyllia
Crispatotochus
15 4
11 3
Turbinaria
11
Hydnophora
Lobophyllia
6
Platygyra
27 1 2 2 Catalaphyllia
Legend
Enallopsammia
1
Morphological
Dichocoenia
characters
3 14 1 Phyllangia
5
16S substitutions
Fungia
Zoopilus
18S substitutions 10 2
3 3
Oculina
28S substitutions Cladocora
1 4 2 Paracyathus
Polycyathus
1 2

Figure 2. Strict consensus of 12 equally parsimonious trees (L = 3510; CI = 0.62; RI = 0.64), with number and types of
characters unambiguously supporting each node indicated. Asterisks and lettering of branches relate to discussion of skel-
etal evolution in the text. Ordinal groups are labelled; more detailed taxonomic information is given in Table 2.

© 2003 The Linnean Society of London, Zoological Journal of the Linnean Society, 2003, 139, 419–437
428 M. DALY ET AL.

Ceriantheopsis
Ceriantharia
Cerianthus
7 8 47
Edwardsia
Edwardsiidae
Nematostella
17 20 32 33 53
Aiptasia Acontiaria
Metridium
24 29
Urticina
Bunodosoma Actiniaria
16 21 31 32 33 44
Dactylanthus
28 57
Stichodactyla Endomyaria
Anemonia
25 51 Anthopleura
14

Downloaded from https://academic.oup.com/zoolinnean/article/139/3/419/2624250 by guest on 31 January 2023

Palythoa
Parazoanthus Zoanthidea
19 45
1 22 35

Antipathes
Stichopathes
Antipatharia
8 37 52
Amplexidiscus
Corynactis Corallimorpharia
9
Pavona
Thalamophyllia
Stephanocoenia
25 33 Monomyces
Placotrochus
Complex
Porites
corals
Leptopsammia
13 54 55 Tubastrea
Flabellum
Balanophyllia
37 38 49 Goniopora
Rhizopsammia
Anthemiphyllia
Scleractinia
Vaughanella
Caryophyllia
Crispatotochus
Turbinaria
Hydnophora
Lobophyllia Robust
Platygyra
Catalaphyllia corals
3 42
Enallopsammia
42
1 5 42 Dichocoenia
Phyllangia
Fungia
Zoopilus
Oculina
Cladocora
Paracyathus
40
Polycyathus
5

Figure 3. Tree from Fig. 2, with intra-ordinal clades labelled and morphological synapomorphies optimized. Numbers
refer to Table 3. Double asterisk indicates alternative optimization of calcareous skeleton; branches along which skeleton
would have to have been lost are labelled A, B, and C; see text for further explanation.

possibly allied to ceriantharians (Hyman, 1940; (Berntson et al., 1999; Won et al., 2001) and mor-
Wells & Hill, 1956). However, the Antipatharia– phology (Won et al., 2001).
Corallimorpharia–Scleractinia clade has been found In finding that Corallimorpharia and Scleractinia
in modern phylogenetic analyses of 18S sequences are sister taxa, we concur with, for example, Duerden

© 2003 The Linnean Society of London, Zoological Journal of the Linnean Society, 2003, 139, 419–437
 SYSTEMATICS OF THE HEXACORALLIA 429

(1898), Schmidt (1974), Schmidt & Zissler (1979), den transformation series compatible with our phyloge-
Hartog (1980), Pires & Castro (1997), Romano & netic tree suggests that couples and pairs are two
Cairns (2000), Won et al. (2001), and Daly et al. (2002). separate characters, and that couples, which are
The Scleractinia–Corallimorpharia clade is corrobo- primitive for Hexacorallia, evolved before paired
rated by substitutions in two genes, two sperm ultra- mesenteries. Pairs are primitively monomorphic,
structural features, cnidae distribution, and gross with dimorphism evolving independently within
anatomy. The similarities between corallimorpharians Actiniaria (in Edwardsiidae) and in Zoanthidea. The
and stichodactyline actiniarians are parallelisms. shared similarity of paired, monomorphic couples,
Our results indicate that the morphologically which has been offered as evidence of an Actiniaria–
defined scleractinian suborders Caryophylliina, Den- Corallimorpharia–Scleractinia grouping (Hyman,
drophylliina, and Faviina (Wells, 1956) are polyphyl- 1940), is a shared primitive feature, and thus not evi-
etic. In our tree, members of the Dendrophylliina and dence of an exclusive relationship. This interpretation

Downloaded from https://academic.oup.com/zoolinnean/article/139/3/419/2624250 by guest on 31 January 2023

Caryophylliina are not closely related (in contrast to suggests that hexacorallians have a primitively bira-
Wells, 1956). The close relationship between Poritidae dial symmetry, in contrast to the radial symmetry
and Dendrophylliidae suggested by Veron et al. (1996) characteristic of many other cnidarians.
is supported by the sister-group relationship between Marginal sphincter muscles have similarly been
Goniopora and a clade containing the dendrophylliids characterized as having three alternative states:
Balanophyllia and Enallopsammia, but refuted by the absent, endodermal, or mesogloeal. Although these
close relationship between Porites and the flabellids three states accurately describe the variation in mar-
Monomyces and Placotrochus. The para- and poly- ginal musculature, this way of thinking about the
phyly of most higher level scleractinian groups may sphincter muscle does not accord with its inferred
explain the discrepancies between our interpretation evolutionary history. The evolutionary history of
of scleractinian phylogeny and those that consider sphincter muscles across Hexacorallia is complex:
only morphological characters. hypertrophy of the columnar circular muscles at the
Because the members of the Actiniaria are anatom- margin occurred independently in Zoanthidea and
ically diverse and because the order lacks a unique Actiniaria, and was reversed at least once, in the pty-
morphological diagnostic character, it has been seen chodactiarian Dactylanthus. However, the two clades
as a possibly heterogeneous assemblage of species that in Actiniaria characterized as having marginal sphinc-
cannot be accommodated in Scleractinia, Corallimor- ter muscles are sister taxa, suggesting a single origin
pharia, or Zoanthidea (Stephenson, 1921; Schmidt, for the hypertrophy into a sphincter. In Actiniaria, the
1974). Our combined analysis, like many independent presence of a marginal sphincter muscle is a synapo-
analyses of single gene data sets (e.g. France et al., morphy, with subsequent differentiation (i.e. whether
1996; Song & Won, 1997; Berntson et al., 1999; Won ectodermal or mesogloeal) characterizing less inclu-
et al., 2001; Daly et al., 2002) and like an earlier non- sive groups.
specific immunoassay (Fautin & Lowenstein, 1994), According to our results, there have been several
found a monophyletic Actiniaria. Because Actiniaria is transitions from solitariness to coloniality or clonality.
as old and diverse as its sister-group, the diversity of Clonality is not an intermediate state between soli-
Actiniaria can be explained by its phylogenetic posi- tariness and coloniality; both clonality and coloniality
tion, relative to the other hexacorallian orders, rather are best interpreted as derived from solitariness.
than by polyphyly. Clonality is common among actiniarians; within
Actiniaria, shifts between solitary and clonal mode of
life have happened several times (see Geller & Walton,
CHARACTER EVOLUTION 2001, for a discussion of this phenomenon in Antho-
Some of the morphological characters used in this pleura). In Scleractinia, a transition from solitary to
analysis are reconstructed as part of a nested series of colonial characterizes the crown robust clade; within
transformations that characterize increasingly less this clade, there are two subsequent shifts back to sol-
inclusive groups. This contradicts historical conceptu- itariness (Fungia and Paracyathus). The inferred
alizations of these features as exclusive alternative ancestral condition of the complex Scleractinia clade is
states. For example, the arrangement of mesenteries solitariness, but further optimization is ambiguous:
in members of Hexacorallia has been presented as a delayed optimization favours several independent
single character with three alternative states (Fig. 1): adoptions of coloniality, whereas accelerated optimiza-
unpaired couples; paired, monomorphic couples; and tion interprets at least one clade (Balanophyllia–
paired, dimorphic couples. These character states are Gonoipora–Rhizopsammia) as being characterized by
inadequate to describe the diversity and variability of coloniality, with Balanophyllia re-adopting solitari-
mesenterial arrangements in hexacorallians, and ness. The evolution of coloniality within scleractinians
obscure the shared similarities between states. The is congruent with Wells’ (1956) hypothesis of morpho-

© 2003 The Linnean Society of London, Zoological Journal of the Linnean Society, 2003, 139, 419–437
430 M. DALY ET AL.

genetic trends within Scleractinia, and his conclusion and 28S characters support many of the terminal
that solitary corals are ancestral to colonial forms. nodes in the Scleractinia.
Reproductive and sexual features do not unambig- Although the combined data are congruent with
uously characterize any group because of the number both the 16S and 18S subsets, the results of our com-
of taxa missing entries for these features. However, we bined analysis differ in a few respects from analyses
infer from their distribution on the tree that these considering only one of them. In analysis of only 18S
characters may be phylogenetically important. Gono- sequences, Metridium is more closely related to
chorism is the ancestral condition of hexacorallians, Edwardsia and Nematostella than to the endomyari-
including scleractinians. All brooding scleractinians ans (Daly et al., 2002), and Stichodactyla is more
are part of the crown clade within the complex corals, closely related to Anemonia or Anthopleura than to
signifying that reproduction may be related to evolu- Dactylanthus (Berntson et al., 1999; Daly et al., 2002).
tionary history as well as to ecology. Our findings disagree with earlier 16S analyses with

Downloaded from https://academic.oup.com/zoolinnean/article/139/3/419/2624250 by guest on 31 January 2023

In his surveys of anthozoan cnidae, Carlgren respect to some relationships within Scleractinia
(1940, 1945) found that holotrichous and atrichous (Fig. 2): in our trees, Turbinaria is in the robust rather
nematocysts were the types most widely distributed; than the complex clade (cf. Romano & Palumbi, 1996;
in 1940, he speculated that other types of nemato- Romano & Cairns, 2000); Romano & Cairns (2000)
cysts were derived from either holotrichs or atrichs. found that Porites and Goniopora are sister taxa,
Electron microscopy demonstrated that atrichs are whereas we interpret Porites as more closely related to
holotrichs with extremely small spines (Westfall, Placotrochus than to Goniopora; using only 16S
1965), thus simplifying Carlgren’s (1940) scheme: sequences, Romano & Cairns (2000) were unable to
holotrichs are the most primitive nematocysts. Our resolve the phylogenetic position of Placotrochus.
analysis agrees with one of the predictions of Car- Our results can also be contrasted with previous
lgren’s hypothesis: holotrichs are primitively present morphological studies. Based on his examination of
in the Hexacorallia. However, according to our inter- nematocyst ultrastructure, Schmidt (1974) inter-
pretation of nematocyst distribution, microbasic b- preted Actiniaria as paraphyletic with respect to
mastigophores are also part of the ancestral hexacor- Zoanthidea and Antipatharia. Schmidt & Zissler
allian cnidom. Diversification of nematocysts must (1979) similarly interpreted sperm ultrastructure as
have preceded the diversification of modern hexacor- supporting a paraphyletic Actiniaria. By contrast, we
allian groups. find strong support for actiniarian monophyly. Simi-
Homology of the scleractinian skeleton has been larities in cnidae morphology and distribution led
questioned on three grounds: in light of the fossil Schmidt (1974) to consider the Edwardsiidae as part
record (e.g. Wells, 1956; Veron, 1995; Oliver, 1996; of the Endomyaria; we infer that edwardsiids are
Veron et al., 1996; Stanley & Fautin, 2001), in light of distantly related to endomyarian anemones. Pires &
ecological evidence for its ephemerality (Buddemeier Castro (1997) questioned the monophyly of major
& Fautin, 1996), and in light of scleractinian and cor- scleractinian families and subfamilies based on nem-
allimorpharian relationships (e.g. Romano & Cairns, atocyst distribution; although many taxa included in
2000; Daly et al., 2002). Our results can address only their analysis are missing from ours, we concur that
the phylogenetic argument. We find that Scleractinia Caryophylliidae and Faviina are polyphyletic.
is not polyphyletic with respect to Corallimorpharia or Despite concern that morphological characters are
Actiniaria; Scleractinia is monophyletic and thus the inadequate or insufficient for recognizing groups
skeleton potentially has a single origin (Fig. 3). within Hexacorallia, many of the traditional diagnos-
tic features are recovered as synapomorphies. How-
ever, given the past emphasis on exclusive features
COMBINED ANALYSIS (e.g. Lang, 1984), the more critical test of the value of
Our examination of morphology and multiple gene morphological characters is their ability to character-
sequences includes more taxa and more detail about ize groups consisting of several orders. Two anatomi-
relationships within hexacorallian orders than any cal features (mesenterial filament morphology and
previous study of a single type of data (cf. Schmidt, paired mesenteries) and two types of nematocyst
1974; Schmidt & Zissler, 1979; Romano & Palumbi, (microbasic and macrobasic p-mastigophores) charac-
1996; Pires & Castro, 1997; Song & Won, 1997; Bernt- terize large groups within Hexacorallia. Coloniality is
son et al., 1999; Romano & Cairns, 2000; Won et al., likewise a phylogenetically significant character,
2001; Daly et al., 2002). The controversial relation- although it has a complex evolutionary history. Most
ships Chen et al. (1995) found with 28S sequences of the morphological features characterize ordinal or
alone, like a polyphyletic Actiniaria or a particularly subordinal groups, confirming that the problem with
close relationship between Edwardsia and Aiptasia, morphology is one of emphasis and focus, rather than
are not evident in this combined analysis. Both 16S of quality. The RI of the morphological characters is

© 2003 The Linnean Society of London, Zoological Journal of the Linnean Society, 2003, 139, 419–437
 SYSTEMATICS OF THE HEXACORALLIA 431

slightly lower than the RI of the whole data set (0.59 radioimmunoassay of proteins from whole animals
vs. 0.63), indicating that the morphological data are and calibrating the tree at the known divergence point
slightly less phylogenetically informative than the for two scleractinian genera. They interpreted their
molecular data. data to support Scleractinia as ancestral to Actiniaria
and Corallimorpharia. Additional sequences for Cor-
allimorpharia, and for scleractinians such as Lep-
UNRESOLVED QUESTIONS AND FUTURE DIRECTIONS topenus or Fungiacyathus, which may be closely
We interpret several classical taxonomic features as related to them (see Fautin & Lowenstein, 1994; Pires
having little or no value for reconstructing relation- & Castro, 1997), are necessary to make further
ships among hexacorallians. Trilobed mesenterial fil- progress towards addressing this problem.
aments in actiniarians and zoanthideans are either a Increased taxonomic sampling is clearly essential to
sympleisiomorphy or are derived independently in addressing other taxonomic and evolutionary ques-

Downloaded from https://academic.oup.com/zoolinnean/article/139/3/419/2624250 by guest on 31 January 2023

actiniarians and zoanthideans. Paired mesenteries of tions within Hexacorallia. The importance of a broad
equal size are likewise interpreted as a shared prim- taxonomic sample for phylogenetic reconstruction is
itive feature. The seemingly unpaired mesenteries of well documented (e.g. Hillis, 1996; Graybeal, 1998;
antipatharians may have resulted from reduction of Halanych, 1998). Because of the variability of mor-
musculature. In zoanthideans, however, the ancestral phology and biology within orders, taxonomic sample
arrangement of paired, equally sized mesenteries plays an especially important role in interpreting the
would have to undergo much more significant modifi- evolution of morphological features within Hexacoral-
cation, including a modification of the placement and lia. The Ceriantharia and Zoanthidea are each repre-
pattern of insertion for new mesenteries. sented by just two taxa, making discussion of intra-
The most parsimonious interpretation, given our ordinal relationships impossible for these groups at
tree, is that paired mesenteries arose in the common this time. The taxonomic sample of actiniarians in the
ancestor of zoanthideans, actiniarians, corallimor- combined analysis is both smaller and narrower than
pharians, and scleractinians, and that a skeleton that of some analyses that consider only 18S
arose only in Scleractinia, a septum being secreted sequences (e.g. Song & Won, 1997; Berntson et al.,
between the members of each mesenterial pair. An 1999; Daly et al., 2002). Consequently, the current
alternative explanation is that the ancestor to the interpretation of the evolution of some attributes dif-
Hexacorallia (occupying the position indicated by two fers from that of earlier studies. For example, basilar
asterisks in Fig. 2) had an exoskeleton, which was lost muscles are a synapomorphy for the clade containing
in some lineages. The loss of a skeleton could have Acontiaria and Endomyaria in this combined analysis,
happened many times, but the only non-skeletalized but are interpreted as a parallelism when more actin-
lineages of which we have knowledge are those with iarian taxa and fewer types of data are considered
extant descendants. Several lineages of skeletalized (Daly et al., 2002). Mesogloeal marginal sphincter
hexacorallians represented in the fossil record have no muscles are a synapomorphy for the Acontiaria in this
known modern representatives (Scrutton & Clarkson, analysis, but may actually characterize a clade con-
1991). Based on our tree, such a loss would have had to sisting of Acontiaria and Mesomyaria, a group of
occur minimally three times (at the branches marked actiniarians whose members lack acontia but have
A, B, and C in Fig. 2). mesogloeal marginal sphincters. Particularly critical
Hand (1966) was the first to suggest Scleractinia is to assessing the phylogenetic significance of acontia
ancestral to Actiniaria and Corallimorpharia (and per- would be the inclusion of acontiate abasilarians. Fur-
haps Zoanthidea), recognizing that paired mesenter- thermore, the selection of taxa included in this analy-
ies have no obvious function in non-skeletalized sis is ecologically biased: because of their accessibility,
hexacorallians, and are not required by size alone, as intertidal and shallow subtidal species predominate.
ceriantharians can grow to considerable size without This bias may affect the interpretation of clonality,
them. He reasoned that having a mesentery on each which is related to habitat (Francis, 1988).
side of a septum equalizes the force of contraction on Although our results (Fig. 2) support the monophyly
the two sides of the septum, preventing tearing of the of each hexacorallian order as currently defined, the
mesenteries. Paired mesenteries might also serve to limited taxon sample also affects the severity of tests
flush the interseptal spaces in animals in which the of ordinal monophyly. Members of Sideractiidae, a
basal end is inflexibly attached to the rigid skeleton. family of corallimorpharians, are essential for testing
The absence of a fossil record for non-skeletalized corallimorpharian and scleractinian monophyly, as
hexacorallians makes it difficult to establish direction- members of this group have one tentacle per
ality in the relationship between skeletalized and endocoelic and exocoelic space, like many scleractin-
non-skeletalized taxa. Fautin & Lowenstein (1994) ians and most actiniarians, rather than multiple ten-
attempted to introduce the time dimension by using a tacles, like most other corallimorpharians. Although

© 2003 The Linnean Society of London, Zoological Journal of the Linnean Society, 2003, 139, 419–437
432 M. DALY ET AL.

most scleractinians are inferred to have a single ten- morphological characters and incorporating sequences
tacle per endocoelic space (e.g. Hyman, 1940; Schmidt, for taxa from Sideractiidae, Protantheae, and Exo-
1974; Chevalier, 1987), the tentacle arrangement is coelantheae; such studies will make it possible to
unknown for most corals, and may be variable (e.g. address remaining questions about taxonomy and
Duerden, 1904). Actiniaria contains three suborders: phylogeny of Hexacorallia.
Protantheae, Endocoelantheae, and Nynanthae. Only
the largest, Nynanthae, is represented. Protantheae
has been linked to Scleractinia (Schmidt, 1974); the ACKNOWLEDGEMENTS
inclusion of members of this group would strongly test This work was supported by NSF grants DEB 99-
monophyly of Actiniaria and Scleractinia. In members 78106, DEB 95-21819 (in the programme Partner-
of Endocoelantheae, the second cycle of mesenteries ships to Enhance Expertise in Taxonomy) to D.G.F.
are inserted in the lateral endocoels. This pattern is and OCE 00-03970 (in the National Oceanic Partner-

Downloaded from https://academic.oup.com/zoolinnean/article/139/3/419/2624250 by guest on 31 January 2023

unlike that of other actiniarians, and thus illustrates ship program) to D.G.F. and R. W. Buddemeier. S. D.
variation in phylogenetically informative features Cairns and J. E. N. Veron kindly answered questions
that we have not sampled. Carlgren (1914) suggested about scleractinian taxonomy; any errors and omis-
that similar variation might occur in other hexacoral- sions are ours. D. R. Smith, T. R. White, and C. W. Cun-
lians, particularly in Scleractinia, a group in which ningham assisted with collection of sequence data.
soft-tissue anatomy has received little emphasis. Comments and suggestions from S. D. Cairns, C.
The absence of information on scleractinian soft- Hand, and two anonymous reviewers significantly
tissue morphology has three important consequences. improved this manuscript.
Although molecular data resolve groups within Scler-
actinia, at least some of these groups are supported by
only a few substitutions. Morphological data may fur- REFERENCES
ther support (or contravene) these clades. Without
Allcock LA, Watts PC, Thorpe JP. 1998. Divergence of nem-
information on scleractinians, we cannot estimate the
atocysts in two colour morphs of the intertidal beadlet anem-
degree and type of variation within each order for one Actinia equina. Journal of the Marine Biological
gross morphological features, determine the number Association of the United Kingdom 78: 821–828.
of states, or evaluate proposed character transforma- Babcock RC, Bull GD, Harrison PL, Heyward AJ, Oliver
tion series for characters that occur in all hexacoral- JK, Wallace CC, Willis BL. 1986. Synchronous spawnings
lians. The absence of information about most of 105 scleractinian coral species on the Great Barrier Reef.
scleractinians compromises the value of soft tissue Marine Biology 90: 379–394.
characters across Hexacorallia: because scleractinians Bayer FM, Grasshoff M, Verseveldt J. 1983. Illustrated tri-
are interpreted as potentially having any character lingual glossary of morphological and anatomical terms
state for many of these features, soft tissue characters applied to Octocorallia. Leiden: EJ Brill.
cannot be optimized unambiguously even for taxa for Berntson EA, France SC, Mullineaux LS. 1999. Phyloge-
which we have concrete information. netic relationships within the class Anthozoa (Phylum Cni-
The current equivalency in rank between the hexa- daria) based on nuclear 18S rDNA sequences. Molecular
corallian orders does not accurately reflect history or Phylogenetics and Evolution 13: 417–433.
relationship. Ceriantharia is the most basal group in Brown WM, Pranger EM, Wang A, Wilson AC. 1982. Mito-
Hexacorallia, and is the sister-group to the remaining chondrial DNA sequences of primates: tempo and mode of
orders; Actiniaria is the sister-group to a clade con- evolution. Journal of Molecular Evolution 18: 225–239.
taining Antipatharia, Corallimorpharia, Scleractinia, Buddemeier RW, Fautin DG. 1996. Global CO2 and evolu-
tion among the Scleractinia. Bulletin de l’Institut Océan-
and Zoanthidea. Subdividing Actiniaria into smaller,
ographique, Monaco, Spécial 14: 33–38.
monophyletic groups is not yet possible, as intra-ordi-
Cairns SD. 1982. Antarctic and subantarctic Scleractinia.
nal relationships are ambiguous. The Athenaria, an
Antarctic Research Series 34: 1–74.
intra-ordinal grouping of burrowing anemones pro-
Cairns SD. 1984. An application of phylogenetic analysis to
posed by Carlgren (1949) that includes Edwardsia and
the Scleractinia: family Fungiidae. Paleontographica Amer-
Nematostella, is polyphyletic (e.g. Berntson et al., icana 54: 49–57.
1999; Daly et al., 2002), effectively rendering Endo- Cairns SD. 1988. Asexual reproduction in solitary Sclerac-
myaria, Mesomyaria, and Acontiaria polyphyletic. tinia. Proceedings of the 6th International Coral Reef Sym-
There is no morphological character by which to rec- posium 2: 641–646.
ognize Actiniaria; although this is a problem regard- Cairns SD. 1989. A revision of the ahermatypic Scleractinia of
less of rank, a formal reclassification would entail the Philippine Islands and adjacent waters, Part 1: Fungia-
diagnosis, which is impossible at this stage. Reclassi- cyathidae, Micrabaciidae, Turbinoliinae, Guyniidae, and
fication of Actiniaria should follow more detailed, Flabellidae. Smithsonian Contributions to Zoology 486: 1–
densely sampled analyses aimed at identifying new 136.

© 2003 The Linnean Society of London, Zoological Journal of the Linnean Society, 2003, 139, 419–437
 SYSTEMATICS OF THE HEXACORALLIA 433

Cairns SD. 2000. A revision of the shallow-water azooxanthel- Dewel WC, Clark WH Jr. 1972. An ultrastructural investi-
late Scleractinia of the western Atlantic. Studies on the Nat- gation of spermiogenesis and the mature sperm in the
ural History of the Caribbean Region 75: 1–240. anthozoan Bunodosoma cavernata (Cnidaria). Journal of
Cappola VA, Fautin DG. 2000. All three species of Ptycho- Ultrastructure Research 40: 417–431.
dactiaria belong to order Actiniaria (Cnidaria: Anthozoa). Doumenc DA, Van Praët M. 1987. Ordre des Actiniaires,
Journal of the Marine Biological Association of the United ordre des Ptychodactiares, ordre des Corallimorphaires. In:
Kingdom 80: 995–1005. Doumenc D, ed. Traité de zoologie, Tome III, Fasicle 3. Paris:
Carlgren O. 1912. Ceriantharia. Danish Ingolf Expedition 5 Masson, 257–401.
(3): 1–79. Duerden JE. 1898. On the relations of certain Stichodactyli-
Carlgren O. 1914. On the genus Porponia and related genera, nae to the Madreporaria. Journal of the Linnean Society of
Scottish National Antarctic Expedition. Transactions of the London (Zoology) 26: 635–653.
Royal Society of Edinburgh 50: 49–71. Duerden JE. 1900. Jamaican Actiniaria II. Scientific Trans-
Carlgren O. 1921. Actiniaria I. Danish Ingolf Expedition 5 (9): actions of the Royal Dublin Society 7: 133–208.

Downloaded from https://academic.oup.com/zoolinnean/article/139/3/419/2624250 by guest on 31 January 2023

1–241. Duerden JE. 1902a. Report on the actinians of Porto Rico.
Carlgren O. 1940. A contribution to the knowledge of the Bulletin of the US Commission of Fish and Fisheries 20:
structure and distribution of the cnidae in the Anthozoa. 323–374.
Kungliga Fysiografiska Sällskapets Handlingar 51: 1–62. Duerden JE. 1902b. West Indian madreporarian polyps.
Carlgren O. 1942. Actiniaria II. Danish Ingolf Expedition 5 Memoirs of the National Academy of Sciences 8: 399–648.
(12): 1–92. Duerden JE. 1904. The coral Sidastrea radians and its post-
Carlgren O. 1944. Das System und die Entwicklungslinien larval development. The Carnegie Institution of Washington
der Anthozoen zugleich einige Bemerkungen über Pax’ Bear- 20: 1–130.
beitung dieser Tiergruppe in der ‘Tierwelt der Nord- und Dunn DF. 1975. Reproduction of the externally brooding sea
Ostsee’. Kungliga Fysiografiska Sällskapets i Lund Förhan- anemone Epiactis prolifera Verrill, 1869. Biological Bulletin
dlingar 14: 40–55. 148: 199–218.
Carlgren O. 1945. Further contributions to the knowledge of Dunn DF. 1981. The clownfish sea anemones: Stichodactly-
the cnidom in the Anthozoa especially in the Actiniaria. idae (Coelenterata: Actiniaria) and other sea anemones sym-
Kungliga Fysiografiska Sällskapets Handlingar 56: 3–24. biotic with pomacentrid fishes. Transactions of the American
Carlgren O. 1949. A survey of the Ptychodactiaria, Coralli- Philosophical Society 71: 3–115.
morpharia and Actiniaria. Kungliga Svenska Vetenskap- Dunn DF. 1982. Cnidaria. In: Parker SP, ed. Synopsis and
sakadamiens Handlingar, 4th Series 1: 1–121. classification of living things, Vol. I. New York: McGraw-Hill,
Chadwick-Furman NE, Spiegel M. 2000. Abundance and 669–706.
clonal replication in the tropical corallimorpharian Rhodac- Fadlallah YH. 1983. Sexual reproduction, development and
tis rhodostoma. Invertebrate Biology 119: 351–360. larval biology in scleractinian corals: a review. Coral Reefs 2:
Chadwick-Furman NE, Spiegel M, Nir I. 2000. Sexual 129–150.
reproduction in the tropical corallimorpharian Rhodactis Fadlallah YH, Karlson RH, Sebens KP. 1984. A compara-
rhodostoma. Invertebrate Biology 119: 361–369. tive study of sexual reproduction in three species of Pana-
Chen CA, Odorico DM, ten Lohuis M, Veron JEN, Miller manian zoanthids (Coelenterata: Anthozoa). Bulletin of
DJ. 1995. Systematic relationships within the Anthozoa Marine Science 35: 80–89.
using the 5′-end of the 28S rDNA. Molecular Phylogenetics Farris JS, Källersjö M, Bult C, Kluge AG. 1994. Testing sig-
and Evolution 4: 175–183. nificance of incongruence. Cladistics 10: 315–319.
Chen CA, Wallace CC, Wolstenholme J. 2002. Analysis of Fautin DG. 1988. The importance of nematocysts to actinian
the mitochondrial 12S rRNA gene supports a two-clade taxonomy. In: Hessinger DA, Lenhoff HM, eds. The biology of
hypothesis of the evolutionary history of scleractinian corals. nematocysts. New York: Academic Press, 487–500.
Molecular Phylogenetics and Evolution 23: 137–149. Fautin DG. 2002. Reproduction of Cnidaria. Canadian Jour-
Chen CA, Willis BL, Miller DJ. 1996. Systematic relation- nal of Zoology 80: 1735–1754.
ships between tropical corallimorpharians (Cnidaria: Antho- Fautin DG, Lowenstein JM. 1994. Phylogenetic relation-
zoa: Corallimorpharia): utility of the 5.8S and internal ships among scleractinians, actinians, and corallimorphari-
transcribed spacer (ITS) regions of the rRNA transcription ans (Coelenterata: Anthozoa). Proceedings of the 7th
unit. Bulletin of Marine Science 59: 196–208. International Coral Reef Symposium 2: 665–670.
Chevalier J-P. 1987. Ordre des Scléractiniares. In: Doumenc Fautin DG, Mariscal RN. 1991. Cnidaria: Anthozoa. In: Har-
D, ed. Traité de zoologie, Tome III, Fasicle 3. Paris: Masson, rison FW, Westfall JA, eds. Microscopic anatomy of inverte-
403–764. brates, Vol. 2. New York: Wiley-Liss, 267–358.
Cutress CE. 1955. An interpretation of the structure and dis- Fautin DG, Smith DR. 1997. Clonality as a taxonomic char-
tribution of cnidae in Anthozoa. Systematic Zoology 4: 120– acter of actinian species. Proceedings of the 8th International
137. Coral Reef Symposium 2: 1609–1612.
Daly M, Lipscomb DL, Allard MW. 2002. A simple test: France SC, Rosel PE, Agenbroad JE, Mullineaux LS,
evaluating explanations for the relative simplicity of the Kocher TD. 1996. DNA sequence variation of mitochondrial
Edwardsiidae (Cnidaria: Anthozoa). Evolution 56: 502–510. large-subunit rRNA provides support for a two-subclass

© 2003 The Linnean Society of London, Zoological Journal of the Linnean Society, 2003, 139, 419–437
434 M. DALY ET AL.

organization of the Anthozoa (Cnidaria). Molecular Marine ed. The cnidarians and their evolution. New York: Academic
Biology and Biotechnology 5: 15–28. Press, 135–146.
Francis L. 1988. Cloning and aggression among sea anemones Hand C, Uhlinger KR. 1992. The culture, sexual and asexual
(Coelenterata: Actiniaria) of the rocky shore. Biological Bul- reproduction, and growth of the sea anemone Nematostella
letin 174: 241–253. vectensis. Biological Bulletin 182: 169–176.
Geller JB, Walton ED. 2001. Breaking up and getting back Harley E. 1996. DAPSA. Program and Documentation. Cape-
together: evolution of symbiosis and cloning by fission in sea town, South Africa: published by the author.
anemones (genus Anthopleura). Evolution 55: 1781–1794. Harriott VJ. 1983. Reproductive ecology of four scleractinian
Gerrodette T. 1981. Dispersal of the solitary coral Balano- species at Lizard Island, Great Barrier Reef. Coral Reefs 2:
phyllia elegans by demersal planular larvae. Ecology 62: 9–18.
611–619. Harrison PL. 1985. Sexual characteristics of scleractinian
Glynn PW, Colley SB, Eakin CM, Smith DB, Cortés J, corals: systematic and evolutionary implications. Proceed-
Gassman NJ, Guzmán HM, Del Rosario JB, Feingold ings of the 5th International Coral Reef Congress 4: 337–342.

Downloaded from https://academic.oup.com/zoolinnean/article/139/3/419/2624250 by guest on 31 January 2023

JS. 1994. Reef coral reproduction in the eastern Pacific: Harrison PL. 1990. Sperm morphology and fertilization strat-
Costa Rica, Panamá, and Galápagos Islands (Ecuador). II. egies in scleractinian corals. In: Hoshi M, Yamashita O, eds.
Poritidae. Marine Biology 118: 191–208. Advances in invertebrate reproduction 5. Amsterdam:
Glynn PW, Colley SB, Gassman NJ, Black K, Cortés J, Elsevier Science, 299–304.
Maté JL. 1996. Reef coral reproduction in the eastern Harrison PL, Jamieson BGM. 1999. Cnidaria and Cteno-
Pacific: Costa Rica, Panamá, and Galápagos Islands (Ecua- phora. In: Adiyodi KG, Adiyodi RG, eds. Reproductive biol-
dor). III. Agariciidae (Pavona gigantea and Gardineroseris ogy of invertebrates, Vol. IX. Part A. New Delhi: Oxford &
planulata). Marine Biology 125: 579–601. IBH Publishing Co, 22–95.
Glynn PW, Gassman NJ, Eakin CM, Cortes J, Smith DB, Harrison PL, Wallace CC. 1990. Reproduction, dispersal,
Guzman HM. 1991. Reef coral reproduction in the eastern and recruitment of scleractinian corals. In: Dubinsky Z, ed.
Pacific: Costa Rica, Panama, and Galapagos Islands (Ecua- Coral reefs. Amsterdam: Elsevier, 133–207.
dor). I. Pocilliporidae. Marine Biology 109: 355–368. den Hartog JC. 1980. Caribbean shallow water Corallimor-
Goffredo S, Teló T, Scanabissi F. 2000. Ultrastructural pharia. Zoologische Verhandelingen Leiden 176: 3–83.
observations of the spermatogenesis of the hermaphroditic Heltzel PS, Babcock RC. 2002. Sexual reproduction, larval
solitary coral Balanophyllia europea (Anthozoa, Sclerac- development and benthic planulae of the solitary coral Mon-
tinia). Zoomorphology 119: 231–240. omyces rubrum (Scleractinia: Anthozoa). Marine Biology
Goloboff PA. 1995. NONA. Tucumon, Argentina. Published by 140: 659–667.
the author. Available at: http://www.cladistics.com. Herberts C. 1987. Ordre des Zoanthaires. In: Doumenc D, ed.
Graybeal A. 1998. Is it better to add taxa or characters to a Traité de zoologie, Tome III, Fasicle 3. Paris: Masson, 783–810.
difficult phylogenetic problem? Systematic Biology 47: 9–17. Hertwig R. 1882a. Die Actinien der Challengerexpedition.
Grigg RW, Opresko D. 1977. Order Antipatharia. In: Dev- Jena: Gustav Fischer.
aney DM, Eldridge LG, eds. Reef and shore fauna of Hawaii. Hertwig R. 1882b. Report on the Actiniaria dredged by the
Honolulu: Bishop Museum Press, 242–261. HMS Challenger during the years 1873–1876. Scientific
Haddon AC. 1898. The Actiniaria of the Torres Straits. Results of the Voyage of the HMS Challenger 6: 1–136.
Scientific Transactions of the Royal Dublin Society 6: 393– Hidaka M. 1992. Use of nematocyst morphology for taxonomy
520. of some related species of scleractinian corals. Galaxea 11:
Haddon AC, Shackleton AM. 1891a. A revision of the Brit- 21–28.
ish Actiniae. Part II: The Zoantheae. Scientific Transactions Hillis DM. 1996. Inferring complex phylogenies. Nature 383:
of the Royal Dublin Society 4: 609–672. 130–131.
Haddon AC, Shackleton AM. 1891b. Reports on the Hinsch GW, Moore JA. 1992. The structure of the reproduc-
zoological collections made in Torres Straits by Prof. A. C. tive mesenteries of the sea anemone Ceriantheopsis
Haddon, 1888–1889. Actiniae. I. Zoantheae. Scientific Trans- americanus. Invertebrate Reproduction and Development 21:
actions of the Royal Dublin Society 13: 673–701. 25–32.
Halanych KM. 1998. Lagomorphs misplaced by more charac- Holts LJ, Beauchamp KA. 1993. Sexual reproduction in
ters and fewer taxa. Systematic Biology 47: 138–146. the corallimorpharian sea anemone Corynactis californica
Hand C. 1955a. (for 1954). The sea anemones of central Cal- in a central California kelp forest. Marine Biology 116:
ifornia I. The corallimorpharian and athenarian anemones. 129–136.
Wasmann Journal of Biology 12: 345–375. Hyman L. 1940. The invertebrates, Vol. I. New York: McGraw-
Hand C. 1955b. The sea anemones of central California II. Hill.
The endomyarian and mesomyarian anemones. Wasmann Kluge AG. 1989. A concern for evidence and a phylogenetic
Journal of Biology 13: 37–99. hypothesis of relationships among Epicrates (Boidae, Ser-
Hand C. 1956. (for 1955). The sea anemones of central Cali- pentes). Systematic Zoology 38: 7–25.
fornia III. The acontiarian anemones. Wasmann Journal of Kramarsky-Winter E, Loya Y. 1996. Regeneration versus
Biology 13: 189–251. budding in fungiid corals: a trade-off. Marine Ecology
Hand C. 1966. On the evolution of the Actiniaria. In: Rees WJ, Progress Series 134: 179–185.

© 2003 The Linnean Society of London, Zoological Journal of the Linnean Society, 2003, 139, 419–437
 SYSTEMATICS OF THE HEXACORALLIA 435

Krupp DA. 1983. Sexual reproduction and early development de Queiroz K, Gauthier J. 1992. Phylogenetic taxonomy.
of the solitary coral Fungia scuteria (Anthozoa: Scleractinia). Annual Review of Ecology and Systematics 23: 449–480.
Coral Reefs 2: 159–164. Rinkevich B, Loya Y. 1979. The reproduction of the Red Sea
Lang JC. 1984. Whatever works: the variable importance of coral Stylophora pistillata. I. Gonads and planulae. Marine
skeletal and of non-skeletal characters in scleractinian tax- Ecology Progress Series 1: 133–144.
onomy. Palaentolographica Americana 54: 18–44. Romano SL, Cairns SD. 2000. Molecular phylogenetic
Manuel RL. 1981. The British Anthozoa. London: Academic hypotheses for the evolution of scleractinian corals. Bulletin
Press. of Marine Science 67: 1043–1068.
Mariscal RN. 1974. Nematocysts. In: Muscataine L, Lenhoff Romano SL, Palumbi SR. 1996. Evolution of scleractinian
HM, eds. Coelenterate biology: reviews and new perspectives. corals inferred from molecular systematics. Science 271:
New York: Academic Press, 129–178. 640–642.
Mariscal RN, Conklin EJ, Bigger CH. 1977. The ptychocyst, Ryland JS. 1997a. Budding in Acrozoanthus Saville-Kent,
a major new category of cnida used in tube construction by a 1893 (Anthozoa: Zoanthidea). In: den Hartog JC, ed. Pro-

Downloaded from https://academic.oup.com/zoolinnean/article/139/3/419/2624250 by guest on 31 January 2023

cerianthid anemone. Biological Bulletin 152: 392–405. ceedings of the 6th International Congress of Coelenterate
McMurrich JP. 1891. Phylogeny of the Actinozoa. Journal of Biology. Leiden, the Netherlands: Nationaal Natuurhis-
Morphology 5: 125–164. torisch Museum, 423–428.
Miller K, Babcock R. 1997. Conflicting morphological and Ryland JS. 1997b. Reproduction in Zoanthidea (Anthozoa:
reproductive species in the coral genus Platygyra. Biological Hexacorallia). Invertebrate Reproduction and Development
Bulletin 192: 98–110. 31: 177–188.
Moseley HN. 1881. Scleractinia of eastern Australia, part II: Ryland JS. 2000. Reproduction in British zoanthids, and an
families Faviidae, Trachyphylliidae. Scientific Results of the unusual process in Parazoanthus anguicomus. Journal of the
Voyage of the HMS Challenger 2: 1–248. Marine Biological Association of the United Kingdom 80:
Nixon KC. 1999a. Winclada (BETA), Version 0.9.9. 943–944.
Ithaca, NY: published by the author. Available at: http:// Schmidt H. 1969. Die Nesselkapseln der Aktinien und ihre
www.cladistics.com. differentialdiagnostische Bedeutung. Helgoländer Wissen-
Nixon KC. 1999b. The parsimony ratchet, a new method for schaftliche Meeresuntersuchungen 19: 284–317.
rapid parsimony analysis. Cladistics 15: 407–414. Schmidt H. 1972. Die Nesselkapseln der Anthozoen und ihre
Nixon KC, Carpenter JM. 1996. On simultaneous analysis. Bedeutung für die phylogenetische Systematik. Helgoländer
Cladistics 12: 221–241. Wissenschaftliche Meeresuntersuchungen 23: 422–458.
Oliver WA. 1996. Origins and relationships of Paleozoic coral Schmidt H. 1974. On evolution in the Anthozoa. Proceedings
groups and the origin of the Scleractinia. Paleontological of the 2nd International Coral Reef Symposium 1: 533–560.
Society Papers 1: 107–134. Schmidt H, Zissler D. 1979. Die Spermien der Anthozoen
Parker NR, Mladenov PV, Grange KR. 1997. Reproductive und ihre phylogenetische Bedeutung. Zoologica 44: 1–98.
biology of the antipatharian black coral Antipathes fiordensis Scrutton CT, Clarkson ENK. 1991. A new scleractinian-like
in Doubtful Sound, Fiordland, New Zealand. Marine Biology coral from the Ordovician of the Southern Uplands, Scot-
130: 11–22. land. Paleontology 34: 179–194.
Pawlowski J, Bolivar I, Fahrni J, de Vargas C, Shick JM. 1991. A functional biology of sea anemones. Lon-
Bowser SS. 1999. Naked foraminiferans revealed. don: Chapman & Hall.
Nature 399: 27. Song J-I, Won JH. 1997. Systematic relationship of the
Pax F. 1940. Anthozoa. H.G. Bronn’s Klassen und Ordnungen anthozoan orders based on the partial nuclear 18S rDNA
des Tierreichs 2 (3): 177–336. sequences. Korean Journal of Biological Sciences 1: 43–
Pearse VB. 2002. Prodigies of propagation: the many modes of 52.
clonal replication in boloceroidid sea anemones (Cnidaria, Stanley GD Jr, Fautin DG. 2001. The origins of modern cor-
Anthozoa, Actiniaria). Invertebrate Reproduction and Devel- als. Science 291: 1913–1914.
opment 41: 201–213. Steiner SCC. 1991. Sperm morphology of scleractinians from
Pearse VB, Francis L. 2000. Anthopleura sola, a new species, the Caribbean. Hydrobiologia 216/217: 131–135.
solitary sibling species to the aggregating sea anemone, Steiner SCC. 1993. Comparative ultrastructural studies on
A. elegantissima (Cnidaria: Anthozoa: Actiniaria: Actini- Scleractinia spermatozoa (Cnidaria, Anthozoa). Zoomorphol-
idae). Proceedings of the Biological Society of Washington ogy 113: 129–136.
113: 596–608. Steiner SCC, Cortes J. 1996. Spermatozoan ultrastructure
Petersen KW. 1979. Development of coloniality in Hydrozoa. of scleractinian corals from the eastern Pacific: Pocil-
In: Larwood G, Rosen BR, eds. Biology and systematics of loporidae and Agariciidae. Coral Reefs 15: 143–147.
colonial organisms. London: Academic Press, 105–139. Stephenson TA. 1920. On the classification of Actiniaria. Part
Pires DO. 1997. Cnidae of Scleractinia. Proceedings of the Bio- I. Forms with acontia and forms with a mesogloeal sphincter.
logical Society of Washington 110: 167–185. Quarterly Journal of Microscopical Science (New Series) 64:
Pires DO, Castro CB. 1997. Scleractinia and Corallimor- 425–574.
pharia: an analysis of cnidae affinity. Proceedings of the 8th Stephenson TA. 1921. On the classification of Actiniaria. Part
International Coral Reef Symposium 2: 1581–1586. II. Consideration of the whole group and its relationships,

© 2003 The Linnean Society of London, Zoological Journal of the Linnean Society, 2003, 139, 419–437
436 M. DALY ET AL.

with special reference to forms not treated in Part I. Quarterly Westfall JA. 1965. Nematocysts of the sea anemone Metrid-
Journal of Microscopical Science (New Series) 65: 493–576. ium. American Zoologist 5: 377–393.
Stephenson TA. 1922. On the classification of Actiniaria. Part Won JH, Rho BJ, Song JI. 2001. A phylogenetic study of the
III. Definitions connected with the forms dealt with in Part Anthozoa (phylum Cnidaria) based on morphological and
II. Quarterly Journal of Microscopical Science (New Series) molecular characters. Coral Reefs 20: 39–50.
66: 247–319.
Stephenson TA. 1928. The British sea anemones, 1. London:
The Ray Society.
Stephenson TA. 1929. On the nematocysts of sea anemones.
CHARACTER LIST
Journal of the Marine Biological Association of the United BIOLOGY
Kingdom 16: 173–201.
Stephenson TA. 1935. The British sea anemones, 2. London: 1. Polyp organization: (0) solitary; (1) colonial;
The Ray Society. (2) clonal.

Downloaded from https://academic.oup.com/zoolinnean/article/139/3/419/2624250 by guest on 31 January 2023

Stoddart JA, Black R. 1985. Cycles of gametogenesis and 2. Spawning type: (0) broadcasting; (1) brooding.
planulation in the coral Pocillopora damicornis. Marine 3. Sexuality: (0) gonochoric; (1) hermaphroditic.
Ecology Progress Series 23: 153–164. 4. Transverse fission: (0) absent; (1) present.
Swofford DL, Olsen GJ, Waddell PJ, Hillis DM. 1996. Phy- 5. Zooxanthellae: (0) absent; (1) present.
logenetic inference. In: Hillis DM, Moritz C, Mable BK, eds. 6. Budding and longitudinal fission: (0) none;
Molecular systematics. Sunderland MA: Sinaur Associates, (1) longitudinal fission (extratentacular budding);
407–514. (2) intratentacular budding; (3) stolon budding.
Tiffon Y. 1987. Ordre des Cerianthaires. In: Doumenc D, ed.
Traité de zoologie, Tome III, Fasicle 3. Paris: Masson, 211–
256. ANATOMY
Tranter PRG, Nicholson DN, Kinchington D. 1982. A
7. Tentacles: (0) marginal only; (1) marginal and
description of spawning and post-gastrula development of
the cool temperate coral, Caryophyllia smithii. Journal of the
labial.
Marine Biological Association of the United Kingdom 62: 8. Tentacles retractile: (0) no; (1) yes.
845–854. 9. Tentacle/coelenteron relationship: (0) 1 tentacle
Vaughan TW, Wells JW. 1943. Revision of the suborders, per endocoel and per exocoel; (1) 1 tentacle per
families, and genera of the Scleractinia. Special Papers of the exocoel, multiple per endocoel.
Geological Society of America 44: 1–363. 10. Tentacle length: (0) greater than or approximately
Veron JEN. 1995. Corals in space and time. Ithaca NY: Com- equal to oral disc radius; (1) less than oral disc
stock/Cornell Press. radius.
Veron JEN. 2000a. Corals of the world, 1. Townsville, Austra- 11. Relative tentacle lengths: (0) inner tentacles
lia: Australian Institute of Marine Science. longer than outer tentacles; (1) inner tentacles
Veron JEN. 2000b. Corals of the world, 2. Townsville, Aus- shorter than outer tentacles; (2) inner tentacles
tralia: Australian Institute of Marine Science. equal to outer tentacles.
Veron JEN. 2000c. Corals of the world, 3. Townsville, Austra- 12. Catch tentacles: (0) absent; (1) present.
lia: Australian Institute of Marine Science. 13. Acrospheres: (0) absent; (1) present.
Veron JEN, Odorico DM, Chen CA, Miller DJ. 1996. Reas- 14. Marginal spherules: (0) absent; (1) holotrichous.
sessing evolutionary relationships of scleractinian corals. 15. Adhesive columnar protrusions: (0) absent;
Coral Reefs 15: 1–9. (1) present.
Wallace CC. 1999. Staghorn corals of the world. Collingwood,
16. Siphonoglyph: (0) one; (1) two or more; (2) absent.
Australia: CSIRO Publishing.
17. Directives: (0) sterile; (1) fertile.
Watson GM, Wood RL. 1988. Colloquium on terminology. In:
18. Coupled mesenteries: (0) absent; (1) present.
Hessinger DA, Lenhoff HM, eds. The biology of nematocysts.
19. Paired mesenteries: (0) absent; (1) present.
New York: Academic Press, 21–23.
20. Pair morphology: (0) members same size; (1) mem-
Weill R. 1934. Contribution à l’étude des cnidaires et de leurs
nématocyctes I. Travaux de la Station Zoologique de Wime-
bers differ in size.
raux 10: 1–347. 21. Paired secondary cycle: (0) absent; (1) present.
Wells JW. 1956. Scleractinia. In: Moore RC, ed. Treatise on 22. Mesenterial addition pattern: (0) mesenteries
invertebrate paleontology, Part F, Coelenterata. Lawrence, added around circumference; (1) added ventrally;
KS: Geological Society of America/University of Kansas (2) added in ventrolateral exocoelic spaces only.
Press, F328–F444. 23. Mesenterial fusion: (0) absent; (1) present.
Wells JW, Hill D. 1956. Ceriantipatharia. In: Moore RC, ed. 24. Acontia: (0) absent; (1) present.
Treatise on invertebrate paleontology, Part F, Coelenterata. 25. Trilobed ciliated filaments: (0) absent; (1) present.
Lawrence, KS: Geological Society of America/University of 26. Gonads on mesenteries of 1st cycle: (0) absent;
Kansas Press, F165–F166. (1) present.

© 2003 The Linnean Society of London, Zoological Journal of the Linnean Society, 2003, 139, 419–437
 SYSTEMATICS OF THE HEXACORALLIA 437

27. Gonads on mesenteries of 2nd and subsequent 48. Macrobasic p-mastigophores in filaments:
cycles: (0) absent; (1) present. (0) absent; (1) present.
28. Endodermal sphincter: (0) absent; (1) present. 49. Macrobasic p-mastigophores in tentacles: (0)
29. Mesogloeal sphincter: (0) absent; (1) present. absent; (1) present.
30. Ectodermal longitudinal muscle: (0) absent; 50. Spirocysts: (0) absent; (1) present.
(1) tentacles and oral disc only; (2) whole body.
31. Basilar musculature: (0) absent; (1) present.
32. Retractor muscle: (0) weak – not forming distinct SPERM ULTRASTRUCTURE
muscle; (1) diffuse (comb-like); (2) restricted (kid-
51. Sperm head shape: (0) conical; (1) spherical.
ney shaped).
52. Sperm nuclear depression: (0) absent; (1) present.
33. Parietal muscle: (0) absent; (1) diffuse (tall, nar-
53. Sperm centriolar orientation: (0) perpendicular;
row); (2) restricted (squat, wide).
(1) parallel.

Downloaded from https://academic.oup.com/zoolinnean/article/139/3/419/2624250 by guest on 31 January 2023

34. Mesogloeal cell inclusions: (0) absent/rare;
54. Centriolar ligament: (0) absent; (1) present.
(1) present.
55. Cytoplasmic collar at sperm midpiece: (0) absent;
35. Mesogloeal lacunae: (0) absent; (1) present.
(1) present.
56. Sperm symmetry: (0) asymmetrical; (1) symmet-
SKELETAL MORPHOLOGY rical.
36. Skeleton: (0) absent; (1) mineralic; (2) proteina- Sources: Babcock et al. (1986); Cairns (1982, 1984,
ceous. 1989, 2000); Cappola & Fautin (2000); Carlgren
37. Columella: (0) absent; (1) present. (1912, 1921); Chadwick-Furman & Spiegel (2000);
38. Costae: (0) absent/weak; (1) present. Chadwick-Furman, Spiegel & Nir (2000); Dewel &
39. Colony shape: (0) encrusting; (1) erect; (2) plate- Clark (1972); Duerden (1898, 1900, 1902a,b); Dunn
like; (3) massive (boulder); (4) reptoid. (1975); Fadlallah (1983); Fadlallah, Karlson &
40. Skeletogenic tissue: (0) ectoderm; (1) mesogloea. Sebens (1984); Gerodette (1981); Glynn et al. (1991,
41. Corallum morphology: (0) cerioid; (1) plocoid; 1994, 1996); Goffredo, Teló & Scanabissi (2000);
(2) meandroid; (3) flabello-meandroid; (4) discoidal Haddon (1898); Haddon & Shackelton (1891a,b);
or cupolate; (5) trochoid or turbinate. Hand (1955a,b, 1956); Hand & Uhlinger (1992);
42. Columella: (0) styloid; (1) laminar; (2) spongy Harriot (1983); Harrison (1985, 1990); Harrison &
(trabecular). Jamieson (1999); Harrison & Wallace (1990); den
Hartog (1980); Heltzel & Babcock (2002); Hinsch &
Moore (1992); Holts & Beauchamp (1993); Krupp
CNIDAE
(1983); Manuel (1981); Parker, Mladenov & Grange
43. Holotrichs: (0) absent; (1) present. (1997); Pax (1940); Pires & Castro (1997); Rinkevich
44. Microbasic p-mastigophores: (0) absent; (1) & Loya (1979); Ryland (1997b, 2000); Schmidt (1972,
present. 1974); Schmidt & Zissler (1979); Shick (1991);
45. Microbasic b-mastigophores: (0) absent; (1) Steiner (1991, 1993); Steiner & Cortes (1996);
present. Stephenson (1920, 1921, 1922, 1928, 1935); Stod-
46. Long, thin, basitrichs: (0) absent; (1) present. dart & Black (1985); Tranter, Nicholson & Kinching-
47. Ptychocysts: (0) absent; (1) present. ton (1982); Veron (2000a,b,c); Wells (1956).

© 2003 The Linnean Society of London, Zoological Journal of the Linnean Society, 2003, 139, 419–437