Coral Reefs (2001) 20: 39±50
DOI 10.1007/s003380000132
R EP O RT
J.H. Won á B.J. Rho á J.I. Song
A phylogenetic study of the Anthozoa (phylum Cnidaria) based
on morphological and molecular characters
Received: 16 March 2000 / Accepted: 21 October 2000 / Published online: 2 March 2001
Ó Springer-Verlag 2001
Abstract The phylogenetic relationships within the Anthozoa were re-evaluated based on 41 morphological
characters and nuclear sequences of 18S ribosomal
DNA (29 anthozoans as ingroups and 3 hydrozoans as
outgroups). The parsimony trees derived from the
morphological data did not coincide closely with the
molecular data, and the presence of several polytomies
at some nodes of the trees resulted in ambiguities among
the systematic relationships. On the other hand, the
combined analysis using total evidence presents a more
resolved and highly supported topology, as is indicated
by higher bootstrap values and decay indices than either
analysis alone. However, strict and semi-strict consensus
trees derived from taxonomic congruence show a poorer
resolution for the phylogeny of Anthozoa. The trees
constructed from the molecular data, using neighborjoining and maximum-likelihood methods, are nearly
congruent with the result from the total evidence. Based
on these results, Anthozoa is divided into three subclasses: Alcyonaria, Zoantharia, and Ceriantipatharia.
The Ceriantipatharia now includes only one order, Ceriantharia, since the order Antipatharia is more closely
related to orders within the Zoantharia. The Alcyonaria
is a monophyletic group, in which the order Pennatulacea is basal, and orders Alcyonacea and Telestacea
branch later. The order Gorgonacea is divided into two
suborders, Holaxonia and Scleraxonia. Bellonela is more
related to order Stolonifera, forming a monophyletic
group. In Zoantharia, the order Zoanthinaria is basal,
and the remaining taxa are divided into two clades: one
includes the order Actiniaria and the other includes orders Antipatharia, Corallimorpharia, and Scleractinia.
The latter two orders form a monophyletic group. This
J.H. Won á B.J. Rho á J.I. Song (&)
Department of Biological Science,
Ewha Womans University,
Seoul 120±750, Korea
E-mail: [email protected]
Tel.: +82-02-32772364; Fax: +82-02-32772385
study presents a di€erent phylogeny of actiniarians from
the earlier hypothesis of scleractinian ancestry.
Keywords Phylogeny á Anthozoa á Morphology á
18S rDNA sequences
Introduction
The Anthozoa is a well-de®ned class, with several unique
features distinct from other cnidarians, and is divided
into 15 orders in three subclasses (Table 1); however, the
evolutionary relationships of subclasses and orders
within the Anthozoa have long remained equivocal.
Wells and Hill (1956) proposed three subclasses in the
Anthozoa based on the arrangement of mesenteries:
subclass Ceriantipatharia, which branches o€ ®rst, and
then the two remaining subclasses, Alcyonaria and
Zoantharia (Fig. 1A). Schmidt (1974) later divided the
Anthozoa into two subclasses: Alcyonaria and Zoantharia, based mainly on the composition of nematocysts
(Fig. 1B). In subclass Zoantharia, he suggested that the
order Ceriantharia, which was placed in subclass Ceriantipatharia together with order Antipatharia by Wells
and Hill (1956), is the most primitive, and Antipatharia
has a common ancestor with the order Zoanthinaria.
Evolutionary studies of the class Anthozoa using molecular systematics have been applied to some taxa of
Zoantharia (McCommas 1991; Fautin and Lowenstein
1992; Garthwaite et al. 1994; Romano and Palumbi
1996; Veron et al. 1996). Molecular systematics provides
an added measure to establish the interrelationship
among anthozoans and to resolve the diculties of
morphological classi®cation of the anthozoans.
In this study, we present analyses of phylogenetic
relationships between taxa of Anthozoa based on 18S
ribosomal DNA (rDNA) sequence and morphological
characters. Parsimony analysis for morphological data
and parsimony, neighbor-joining (NJ), and maximum
likelihood (ML) methods for molecular data were
applied, as well as a combined approach using the two
data sets.
40
Molecular data analysis
Materials and methods
Complete 18S rDNA sequences were determined for 16 anthozoans, including nine alcyonarians, six zoantharians, and one ceriantipatharian, using two hydrozoans as outgroups (Table 2).
Previously determined 18S rDNA sequences from 13 anthozoans
and one hydrozoan were retrieved from GenBank.
DNA was isolated from ethanol-preserved tissue by using a
modi®cation of the standard procedure (Sambrook et al. 1989).
Anthozoans, with skeleton, were submerged for 2 or 3 days in
Table 2 Taxonomic position of species used in this study
Taxonomic position
Table 1 Classical classi®cation scheme of anthozoans according to
Dunn (1982)
Class
Subclass
Order
Anthozoa
Alcyonaria
Protoalcyonaria
Stoloniferaa
Telestaceaa
Gastraxonacea
Gorgonaceaa
Alcyonaceaa
Helioporacea
Pennatulaceaa
Actiniariaa
Corallimorphariaa
Scleractiniaa
Zoanthinariaa
Ptychodactinaria
Antipathariaa
Cerianthariaa
Zoantharia
Ceriantipatharia
a
Anthozoan orders used in this study
Fig. 1 Phylogenetic relationships among Anthozoa: A based on
mesenteric arrangements (Wells and Hill 1956) and B based on
nematocysts (Schmidt 1974)
Phylum Cnidaria
Class Hydrozoa
Order Hydroida
Suborder Thecata
Family Sertulariidae
1. Selaginopsis cornigera
Suborder Athecata
Family Corynidae
2. Coryne pusillaa
Family Solanderiidae
3. Solanderia secundaa
Class Anthozoa
Subclass Alcyonaria
Order Stolonifera
Family Clavulariidae
4. Clavularia mikadoa
Order Telestacea
Family Pseudocladochonidae
5. Pseudocladochonus hicksonia
Order Gorgonacea
Suborder Holaxonia
Family Acanthogorgiidae
6. Acalycigorgia inermisa
7. Acalycigorgia irregularisa
Family Paramuriceidae
8. Calicogorgia granulosa
Family Plexauridae
9. Euplexaura crassa
Suborder Scleraxonia
Family Melithaeidae
10. Acabaria habereria
11. Acabaria formosaa
Order Alcyonacea
Family Alcyoniidae
12. Alcyonium gracillimum
13. Bellonella rigida
14. Bellonella unicolora
Family Nephtheidae
15. Dendronephthya putteria
Order Pennatulacea
Family Pennatulidae
Family Virgulariidae
17. Virgularia junceaa
Subclass zoantharia
Order Zoanthinaria
Family Parazoanthidae
18. Parazoanthus axinellae
Order Actiniaria
Family Actiniidae
19. Anemonia sulcata
20. Actinia equina
21. Anthopleura kurogane
22. Paracondylactis hertwigia
23. Epiactis japonica
Family Haliplanellidae
24. Haliplanella luciaa
Family Isophelliidae
25. Flosmaris mutsuensis
Order Corallimorpharia
Family Corallimorphidae
26. Corynactis sp.a
GenBank accession
Z92899
AJ133558
AJ133506
AJ133543
AJ133544
AJ133545
AJ133546
Z92900
Z92901
AJ133547
AJ133548
Z92902
Z49195
AJ133549
AJ133550
AJ133551
U42453
X53498
AJ133552
Z21671
AJ133553
Z92904
AJ133554
Z92905
AJ133559
41
Table 2 (continued)
Taxonomic position
Order Scleractinia
Suborder Caryophylliina
Family Flabellidae
27. Javania insignisa
Suborder Dendrophylliina
Family Dendrophylliidae
28. Tubastraea aurea
29. Tubastraea coccineaa
30. Rhizopsammia minuta mutsuensis
Subclass Ceriantipatharia
Order Antipatharia
Family Antipathidae
31. Antipathes lata
Order Ceriantharia
Family Cerianthidae
32. Cerianthus ®liformisa
a
Analysis of combined data
GenBank accession
Combined analysis of molecular and morphological data was
conducted by taxonomic congruence and total evidence. Incongruence indices IMF (Michevich and Farris 1981) and IM (Swo€ord
1991) between the two data sets were calculated.
AJ133555
Results
Z92906
AJ133556
Z92907
Z92908
AJ133557
18S rDNA sequences are newly reported in this study
0.5 M EDTA before extraction. A pair of primers was selected for
the ampli®cation of 18S rDNA (5¢-CCTGGTTGATCCTGCCAG-3¢, 5¢-TAATGATCCTTCCGCAGGTT-3¢). Polymerase
chain reactions (PCR) were performed using the following protocol: 1 cycle at 94 °C (5 min), 52 °C (2 min), 72 °C (10 min); 35
cycles at 94 °C (1 min), 52 °C (2 min), 72 °C (3 min); and 1 cycle
at 94 °C (1 min), 52 °C (2 min), 72 °C (10 min). PCR products
were puri®ed using a quick-spin PCR puri®cation kit (QIAGEN)
and cloned into pT7 Blue T-vector (QIAGEN). In addition to the
T7 promoter primer and U-19mer reverse primer, 11 internal
primers were used. The sequences of the primers are as follows
[the number indicates the position of each primer's 5¢ base on the
Placopecten magellanicus 18S rDNA (Rice 1990)]: 32, 5¢-ACCTTGTTACGACTTTT A-3¢; 162, 5¢-ACGGGCGGTGTGTAC3¢; 362, 5¢-TCTAAGGGCATCACA-3¢; 477, 5¢-TCTCGTTCGTTATCG-3¢; 530, 5¢-CCATGCACCACCACCC-3¢; 657, 5¢CCGTCAATTCCTTTAAGTTT-3¢; 873, 5¢-CCAAGAATTTCACC-3¢; 1028, 5¢-TAATTTTTTCAAAGT-3¢; 1207, 5¢-GAATTACCGCGGCTG-3¢; 1423, 5¢-ATTCCCCGTCACCCG-3¢;
1631, 5¢-ACCTCTAGAATTACC-3¢. Sequencing was conducted
using the Taqtrack kit (Promega). Sequences were aligned using
CLUSTAL W (Tompson et al. 1994) and highly divergent regions
that could not be reliably aligned were excluded from the analysis.
Phylogenetic analysis was performed using the parsimony,
neighbor-joining (NJ), and maximum likelihood (ML) methods.
In the parsimony method, the heuristic search option in PAUP
3.1.1 (Swo€ord 1993) was used. We used the NEIGHBOR and
DNAML programs in PHYLIP 3.54 (Felsenstein 1994) for the NJ
and ML methods. For the NJ method, we calculated the distance
matrix using the DNADIST with the Kimura 2-parameter distance model; in the ML method, the transition/transversion ratio
was set at 2.
Morphological data analysis
Morphological characters, including external and internal characters, were selected for all specimens for which we had sequence data. In total, 41 characters were gathered from the
literature (Appendix 1); some of them overlapped with those of
Schmidt (1974) and Bridge et al. (1995). When a character was
not applicable to a certain taxon, we symbolized it with N and
treated it as missing data in the analysis (see Table 3). We
analysed the data matrix using PAUP; all multistate characters
were treated as unordered and all characters were equally
weighted.
Morphological data
Parsimony analysis of the morphological data produced
52 trees (77 steps) with a consistency index (CI) of 0.805
and a retention index (RI) of 0.939. The strict consensus
tree had four polytomies (Fig. 2). The bootstrap values
were not high, except for nodes joining the two pennatulaceans and the four scleractinians which showed
values of 99 and 91%, respectively. When only informative data were used, 52 trees (72 steps) resulted, with a
CI of 0.792 and an RI of 0.939. The topology of the
consensus tree was the same as with the former analysis.
We estimated that the ®ve additional steps were caused
by four autapomorphic characters and one ambiguous
character through the plot of characters in PAUP. We
were able to determine that 12 homoplasious characters
produced the extra steps in the tree.
Molecular data
The G+C ratio for all specimens averaged 46.31%
(44.2±49.3%), less than that of crustaceans (49.9±50%;
Kim et al. 1992, 1993) and molluscs (51.5%; Bargues
and Coma 1997). The complete sequences of 18S rDNA
aligned by CLUSTAL W contained 1,895 nucleotides, of
which 43 nucleotides (positions 1770±1812) were excluded due to some uncertainties in alignment. To examine the pattern of variable sites, we cut 20 nucleotides
in order from the 5¢-end of gene and number(s) of
variable sites in each 20 nucleotides was counted (data
are not shown). The examination of secondary structure
of 18S rRNA of Anemone sulcata (Hendricks et al. 1990)
indicated that the interior region of the structure was
more conserved, suggesting that there may be a closer
relation of the region with the protein or function of
ribosome (Spirin 1986). The most variable region was
1,800th, which coincides with the excluded segment.
Only 266 informative sites were used in the parsimony method. The result produced 153 trees (845 steps)
with a CI of 0.504 and an RI of 0.749. These values are
lower than that of the morphological tree. The topology
of the strict consensus tree (Fig. 3) is quite similar but
not exactly congruent with that of the morphological
tree, and polytomies make the interpretation of the tree
dicult. However, ten nodes coincide between the two
trees and are well supported by high bootstrap values
and decay indices. For instance, the alcyonarian clade
was supported by a bootstrap of 93% and displayed a
decay index of ³3.0. This clade was also present in the
42
Fig. 2 Strict consensus of mostparsimonious trees using morphological data, based on 52
trees (77 steps) with CI=0.805
and RI=0.939. Characters
changing on each branch are
listed below or to right of a
branch. Numbers at nodes are
percentages of 100 bootstrap
replicates that support the
branch; only those over 50%
are presented. Solid squares
Unambiguous changes of characters (synapomorphy); open
squares ambiguous changes
(homoplasy); open circles autapomorphy]
morphological tree, where it was supported by a bootstrap of 73%.
The NJ method with Kimura's distance model produced a topology more similar with parsimony analyses
of morphological data than with the molecular data
(Fig. 4). Cerianthus ®liformis has a common ancestor
with zoantharians, but in that node, the bootstrap
value is lower than 50% while most nodes of the tree
are relatively well supported. In Alcyonaria, order
Pennatulacea is basal and two clades, soft corals and
other alcyonarians, diverged later. In Zoantharia,
Parazoanthus axinellae in the order Zoanthinaria
probably diverged ®rst. The other zoantharians were
divided into two groups. However, the tree produced
using the NJ method when compared to that produced
from parsimony analysis of the morphological data
shows that Actiniaria and Gorgonacea are not monophyletic.
The maximum-likelihood analysis produced a tree
that was slightly di€erent from the NJ tree (Fig. 5).
Cerianthus ®liformis branched ®rst before the other
anthozoans were di€erentiated, and all species of the
order Actiniaria formed a monophyletic group. Scleraxonia rather than Holaxonia showed a common ancestor with the clade containing Stolonifera and
Bellonella. Order Antipatharia is closer to Actiniaria
than to Scleractinia.
Combined data
Two combined approaches, namely taxonomic congruence and the total evidence, were conducted. The result
of the total evidence showed an improved tree (Fig. 6).
CI and RI values of this tree were higher than those
calculated for the molecular data. In general, bootstrap
43
Fig. 3 Strict consensus of mostparsimonious trees using 18S
rDNA data, based on 153 trees
(845 steps) with CI=0.504 and
RI=0.749, excluding uninformative characters. Numbers
above branches indicate bootstrap values and numbers in
parentheses are decay values
values were also elevated. However, the strict and semistrict consensus trees from taxonomic congruence
showed a poorer resolution for the phylogenetic analysis
of anthozoans because of the presence of many polytomies (Fig. 7). The result of the total evidence also corresponded to those of the NJ and ML methods.
Incongruence indices, IMF and IM, of the two data sets
were as low as 0.011 and 0.048, respectively.
Discussion
Anthozoans are so varied in their morphology that
important characters used to identify one taxon may
be insigni®cant or lacking in another taxon. For example, spicules, an essential character in alcyonarians,
are not present in zoantharians, while internal features
and nematocysts, important characters in zoantharians,
are rarely shown in alcyonarians. Most of the fossil
records of anthozoans are restricted to hard corals,
and their diversity often causes some diculties in
identi®cation. Dendrobranchia, with its spiny axis, has
been assigned to the order Antipatharia, but in recent
years it was placed in the alcyonarian order Gorgonacea because of its eight pinnate tentacles (Opresko
and Bayer 1991).
Schmidt (1974) and Bridge et al. (1995) introduced
and used cladistics to study anthozoans. The former
used 29 morphological characters, including nematocysts, sperm, and other internal characters, to compare
taxa in Zoantharia but did not treat Alcyonaria in detail.
The latter analyzed the evolution of Cnidaria using 29
morphological characters, but only 5 anthozoans were
included, which was insucient to show the higher relationship within the anthozoans. In this study, 9 characters from previous studies [character nos. 2, 3, 4, 26,
44
Fig. 4 Neighbor-joining tree
from 18S rDNA sequences.
Numbers at nodes are percentage of 1,000 bootstrap replicates that support the branch.
Lengths of branch are proportional to scale given in nucleotide substitutions per site
and 30 in Appendix 1, cited from Schmidt (1974), and
character nos. 20, 25, 27, 28, and 30, from Bridge et al.
(1995)] and 32 additional characters were used. More
external characters appeared ambiguously in the parsimony tree than internal characters.
Molecular techniques have been used for phylogenetic study, and recently combined analysis using various data has also been used. We conducted a combined
approach using 18S rDNA sequences and morphological data. The total evidence showed an improved
topology for the phylogeny of Anthozoa compared to
those of each data set analysed separately, but taxonomic congruence made a worse tree for many polytomies. The superiority of the total evidence approach had
been reported in other studies (Kluge 1989; Eernisse and
Kluge 1993; Omland 1994; Turbeville et al. 1994; Vrana
et al. 1994; Lafay et al. 1995; Patrick et al. 1998). Incongruence indices between molecular and morphological data in this study were relatively low compared with
other studies (Wyss et al. 1987; Kluge 1989; Omland
1994).
Wells and Hill (1956) proposed the new subclass
Ceriantipatharia based on a resemblance between antipatharian polyps and the cerinula larvae of Ceriantharia and because both orders have six protocnemes in
couples, of which the ®rst-formed central transverse one
is fertile. Order Antipatharia was included in the order
Gorgonacea, due to characteristics of its axis and dendroid colony, but later it was placed in the most differentiated zoantharian order because its nematocysts,
sperm, and the element of its skeleton are more similar
to that of zoantharians (Opresko 1972). In this study,
45
Fig. 5 Maximum-likelihood
tree from 18S rDNA sequences.
Numbers at nodes are percentage of 100 bootstrap replicates
that support the branch; only
those over 50% are presented
the order Antipatharia is more closely related to the
zoantharian order Scleractinia than to the order Ceriantharia. Two orders, Antipatharia and Ceriantharia,
have some characters in common but are also di€erent
in many respects. Ceriantharians are always solitary,
occurring on soft bottoms, and possess two types of
tentacles and no hard skeleton. Antipatharian colonies
are attached on hard bottoms with a chitinous axis and
have six tentacles. Moreover, Ceriantharia have many
pairs of complete mesenteries that are bilaterally added
on the ventral side, whereas Antipatharia have six, ten,
or twelve biradial ones. For these reasons, the two orders are too di€erent to unite as a subclass. On the
other hand, the orders Antipatharia and Scleractinia
have a mesenteric ®lament with no ciliated tract in
common.
Results from the total evidence and ML method
indicate that Ceriantharia evolved independently from
a common ancestor of anthozoans, whereas the NJ
method suggests that it shared a common ancestor with
zoantharians. However, Ceriantharia was always basal
among anthozoans, whether grouped with zoantharians
46
Fig. 6 Consensus tree from
morphological and 18S rDNA
sequence data. Strict consensus
of three minimum-length trees
is shown. Trees are 922 steps
long (CI=0.524, RI=0.771).
Solid squares Unambiguous
changes of morphological
characters
or not. All methods of analysis unite alcyonarians as a
monophyletic group, in which the order Pennatulacea
diverged ®rst, with the two orders Alcyonacea and
Telestacea diverging in sequence. There are few studies
of alcyonarian evolution because the orders have such
distinctive features. In general, the simple order Stolonifera is considered the most primitive and the order
Pennatulacea, which lives on soft bottoms like the
Ceriantharia, is regarded as more derived. Therefore,
our results present a contrary view to the existing
classi®cation.
Grassho€ (1984) regarded the class Anthozoa as the
most primitive group among Cnidaria, and his view has
been upheld by recent molecular studies (Hori and
Satow 1991; Bridge et al. 1992, 1995). According to
Grassho€, anthozoans, having only the polyp stage, are
close to the beginning of cnidarian evolution, and
movable solitary polyps are less derived than sessile
colonies, which explains the position of orders Ceriantharia and Pennatulacea in this study.
In this study, Gorgonacea appears to be monophyletic or paraphyletic, but it always contains two clades,
suborders Holaxonia and Scleraxonia, both of which
have an axis of gorgonin. The order Alcyonacea is not
monophyletic because Bellonella always clusters with
the order Stolonifera. The genus Bellonella, like the
Stolonifera, has no thickened coenenchyme, and the
anthocodia are retracted in the calyx. This study suggests that Alcyonacea, except Pennatulacea, di€erentiated in a short time and that the distances between them
are small.
Among the Zoantharia, the order Zoanthinaria is the
most primitive and the remaining orders form two
clades, in which three orders, Antipatharia, Corallimorpharia, and Scleractinia, form one clade and the
Actiniarians form the other. The order Zoanthinaria
47
Fig. 7 Combined trees of strict
and semi-strict consensus trees
of morphological and 18S
rDNA sequence data
has a pair of directive mesenteries, whereas the other
zoantharians have two pairs or no complete directive
mesenteries. Hand (1966) examined the evolutionary
relationships among zoantharians and suggested that
actiniarians diverged from scleractinians, and the order
Corallimorpharia, with intermediate characters, was
clustered with the order Actiniaria. His hypothesis was
supported by Fautin and Lowenstein (1992) using a
radioimmunoassay. However, Chen et al. (1995) presented the Scleractinia as a monophyletic group, similar
to the results in this study. The common ancestor of the
two orders, Scleractinia and Corallimorpharia, might be
inferred from the absence of directive mesenteries and
siphonoglyphs.
Acknowledgments This study was supported by the academic research fund of the Ministry of Education, Republic of Korea
(BSRI-94-98-4421). We are deeply grateful for this generous assistance and also to Dr. S.D. Cairns of the Department of Invertebrate Zoology, National Museum of Natural History (NMNH),
Smithsonian Institution, for critically reading this manuscript.
Appendix 1: Morphological characters used
in the phylogenetic analysis
1. Chitinous exoskeleton: absent (0); coenosarc encased in exoskeleton but hydranths and gonozooids
lack exoskeleton (1); coenosarc, hydranths, and
gonozooids encased in exoskeleton (2)
2. Gut formation: epiderm invaginates at the oral end
as an actinopharynx (0); gut lined with endoderm
only (1)
3. Gonad: derived from ectoderm (0); derived from
endoderm (1)
4. Mesentery in polyp: absent (0); present (1)
5. Cnidocil: absent (0); present (1)
6. Colony types: stolonate (0); simple branched (1);
encrusting (2); digitiform (3); arborescent (4); polyps
on leaves of rachis (5)
7. Peduncle or physa: absent (0); present (1)
8. Dimorphism: absent (0); present (1)
48
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
Siphonozooids: absent (0); present (1)
Calyx of polyp: absent (0); present (1)
Verrucae: absent (0); present (1)
Acrorhagi: absent (0); present (1)
Pedal disk: absent (0); weak (1); strong (2)
Arrangement of tentacles: scattered (0); one cycle
(1); more than two cycles (2)
Type of tentacles: one type (0); two types ± long
marginal and short labial (1)
Shape of tentacles: ®liform (0); pinnate (1); capitate
(2)
Number of tentacles: 6 (0); 8 (1); more than 9 (2)
Number of perfect mesenteries: 8 (0); 10 (1); more
than 12 (2)
Types of mesentery: only perfect mesenteries (0);
perfect mesenteries and imperfect mesenteries not
divisible (1); perfect mesenteries and imperfect
mesenteries divisible (2)
Mesenteric ®lament: consisting of 2 strips (0); consisting of 3 strips (1); consisting of just 1 strip (2)
Ciliated tract on mesenteric ®lament: absent (0);
present (1)
Directive mesentery: absent (0); one pair (1); two
pairs (2)
Mesentery formation: only primary mesenteries (0);
new septa arise in the exocoel to either side of the
ventral directive (1); from ventral (2); between
directive and transverse septa (3); anywhere around
the circumference (4)
24. Symmetry: biradial symmetry (0); bilateral symmetry (1)
25. Pairing of mesentery: paired (0); coupled (1)
26. Siphonoglyph: absent (0); one (1); one or two (2)
27. Sphincter muscle: absent (0); endodermal (1); mesodermal (2)
28. Acontia: absent (0); present (1)
29. Thick coenenchymal mass: absent (0); present (1)
30. Spirocyst: absent (0); present (1)
31. Ptychocyst: absent (0); present (1)
32. p-mastigophore: absent (0); present (1)
33. Calcareous exoskeleton: absent (0); present (1)
34. Axial skeleton: absent (0); stout and round axis (1);
thorny axis (2)
35. Axis with central cord: absent (0); present (1)
36. Stem canal around axis: absent (0); present (1)
37. Calcareous spicule: absent (0); sculptured spicules
(1); minute oval bodies and three-¯anged rods (2);
spindles and smooth rod with median projections
(3)
38. Large sclerites: spicule separated (0); some spicule
fused to form large sclerites (1)
39. Spicules in tentacle: absent (0); present (1)
40. Gorgonin: absent (0); present (1)
41. Columella: absent (0); present (1)
Note: (1) scores of character states are given in parentheses; (2) the data matrix from which the morphological
tree was derived is shown in Table 3.
Table 3 Morphological data matrix. Matrix includes both binary and multiple state characters. Character states assigned were 0, 1, 2, 3, 4,
5, and N (in applicable). Forty one columns correspond to character numbers in Appendix 1. All characters were treated as unordered in
the parsimony analysis
Class
Species
Characters
Multiple state
Hydrozoa
Anthozoa
Selaginopsis cornigera
Coryne pusilla
Solanderia secunda
Clavularia mikado
Pseudocladochonus hicksoni
Acalycigorgia inermis
Acalycigorgia irregularis
Calicogorgia granulosa
Euplexaura crassa
Acabaria habereri
Acabaria formosa
Alcyonium gracillimum
Bellonella rigida
Bellonella unicolor
Dendronephthya putteri
Leioptilus ®mbriatus
Virgularia juncea
Parazoanthus axinellae
Anemonia sulcata
Actinia equina
Anthopleura kurogane
Paracondylactis hertwigi
Epiactis japonica
Binary
1
1234567890
2
1234567890
3
1234567890
4
1234567890
1
2100110100
1100100100
1100140100
0011020001
0011010001
0011040001
0011040001
0011040001
0011040000
0011040001
0011040001
0011040000
0011030001
0011030001
0011040000
0011051110
0011051110
0011020000
00110N0000
00110N0000
00110N0000
00110N0000
00110N0000
0001002NNN
0000022NNN
0000022NNN
0001011000
0001011000
0001011000
0001011000
0001011000
0001011000
0001011000
0001011000
0001011000
0001011000
0001011000
0001011000
0001011000
0001011000
0002002221
0122002211
0122002211
1122002211
0122002211
1022002211
NNNNNN0000
NNNNNN0000
NNNNNN0000
1101110000
1101110000
1101110000
1101110000
1101110000
1101110000
1101110000
1101110000
1101110010
1101110000
1101110000
1101110010
1101110000
1101110000
1211011001
1240021001
1240021001
1240021001
1240021001
1240021001
0000NN0NN0
0000NN0NN0
0000NN0NN0
0000NN1010
0000NN1110
0001101011
0001101011
0001101011
0001111011
0001003011
0001003011
0000NN1010
0000NN1010
0000NN1010
0000NN1010
0000NN2000
0001000NN0
0100NN0NN0
0100NN0NN0
0100NN0NN0
0100NN0NN0
0100NN0NN0
0100NN0NN0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
49
Table 3 (continued)
Class
Species
Characters
Multiple state
Haliplanella lucia
Flosomaris mutsuensis
Corynactis sp.
Javania insignis
Tubastraea aurea
Tubastraea coccinea
Rhizopsammia minuta mutsuensis
Antipathes lata
Cerianthus ®liformis
Binary
1
1234567890
2
1234567890
3
1234567890
4
1234567890
1
00110N0000
00110N0000
00110N0000
00110N0000
0011020000
0011020000
0011020000
0011040000
00110N1000
0022002211
0022002221
0012022212
0002002212
0002002212
0002002212
0002002212
0001000102
0002102201
1240020101
1240022101
0040001001
0040000001
0040000001
0040000001
0040000001
0230120001
1121110001
0100NN0NN0
0100NN0NN0
0100NN0NN0
0110NN0NN0
0110NN0NN0
0110NN0NN0
0110NN0NN0
0102000NN0
10000N0NN0
0
0
0
0
1
1
1
0
0
References
Bargues MD, Coma SM (1997) Phylogenetic analysis of lymnaeid snails based on 18S sequences. Mol Biol Evol 14(5):
569±57
Bridge D, Cunningham CW, Schierwaer B, Desalle R, Buss LW
(1992) Class-level relationships in the phylum Cnidaria: evidenced from mitochondrial genome structure. Proc Natl Acad
Sci USA 89:8750±8753
Bridge D, Cunningham CW, Schierwaer B, Desalle R, Buss LW
(1995) Class-level relationships in the phylum Cnidaria: molecular and morphological evidence. Mol Biol Evol 12(4):679±
689
Chen CA, Odorico DM, Lohuis MT, Veron JEN, Miller DJ (1995)
Systematic relationships within the Anthozoa (Cnidaria: Anthozoa) using the 5¢-end of the 28S rDNA. Mol Phylogenet
Evol 4(2):175±183
Dunn DF (1982) Cnidaria. In: Parker SP (ed) Synopsis and classi®cation of living organisms. McGraw-Hill, New York, pp 669±
706
Eernisse DJ, Kluge AG (1993) Taxonomic congruence versus
total evidence, and aminote phylogeny inferred from fossils, molecules, and morphology. Mol Biol Evol
10(6):1170±1195
Fautin DG, Lowenstein JM (1992) Phylogenetic relationships
among scleractinians, actinians, and corallimorpharians (Coelenterata Anthozoa). In: Proc 7th Int Coral Reef Symp Publ 2,
pp 665±670
Felsenstein J (1994) PHYLIP (phylogenetic inference package),
3.54. University of Washington, Seattle
Garthwaite RL, Potts DC, Veron JEN, Done TJ (1994) Electrophoretic identi®cation of poritid species (Anthozoa: Scleractinia). Coral Reefs 13:49±56
Grassho€ M (1984) Cnidarian phylogeny ± a biochemical approach. Palaeontogr Am 54:127±135
Hand C (1966) On the evolution of the Actiniaria. In: Rees WJ (ed)
The Cnidaria and their evolution. Academic Press, London,
pp 135±146
Hendricks L, Van de Peer Y, Van Herck M, Neefs JM, De Wachter
R (1990) The 18S ribosomal RNA sequences of the sea anemone Anemonia sulcata and its evolutionary position among
other eukaryotes. FEBS Lett 269(2):445±449
Hori H, Satow Y (1991) Dead-end evolution of the Cnidaria as
deduced from 5S ribosomal RNA sequences. Hydrobiologia
216/217:505±508
Kim W, Min GS, Kim SH (1992) The 18S ribosomal RNA gene of
a crustacean decapod Oedignathus inermis: a comparison with
Artemia salina gene. Nucleic Acids Res 21(15):3583
Kim W, Yoon SM, Kim J (1993) The 18S ribosomal RNA gene of
a crustacean branchiopod Bosmina longirostris: comparison
with another branchiopod Artemia salina. Nucleic Acids Res
20(17):4658
Kluge AG (1989) A concern for evidence and a phylogenetic hypothesis of relationships among Epicrates (Boidae, Serpentes).
Syst Zool 38(1):7±25
Lafay B, Smith AB, Christen R (1995) A combined morphological
and molecular approach to the phylogeny of asteroids (Asteroidea: Echinodermata). Syst Biol 44(2):190±208
McCommas SA (1991) Relationships within the family Actiniidae
(Cnidaria, Actiniaria) based on molecular characters. Hydrobiologia 216/217:509±512
Michevich MF, Farris JS (1981) The implications of congruence in
Mendia. Syst Zool 30:351±370
Omland KE (1994) Character congruence between a molecular and
a morphological phylogeny for dabbling ducks (ANAS). Syst
Biol 43(3):369±386
Opresko DM (1972) Redescriptions and reevaluations of the antipatharians described by L.F. de PourtaleÁs. Bull Mar Sci
22(4):950±959
Opresko DM, Bayer FM (1991) Rediscovery of the enigmatic
coelenterate Dendrobrachia (Octocorallia, Gorgonacea) with
descriptions of two new species. Trans R Soc S Aust 115(1):1±19
Patrick MO, Jonathan BC, Margaret GK (1998) Phylogeny of the
Drosophila saltans species group based on combined analysis of
nuclear and mitochondrial DNA sequence. Mol Biol Evol
15(6):656±664
Rice EL (1990) Nucleotide sequence of the 18S ribosomal RNA
gene from the Atlantic sea scallop Placopecten magellanicus
(Gmelin, 1791). Nucleic Acids Res 18:5551
Romano SL, Palumbi SR (1996) Evolution of scleractinian corals
inferred from molecular systematics. Science 271:640±642
Sambrook J, Fritsch E, Maniatis T (1989) Molecular cloning a
laboratory manual, 2nd edn. Cold Spring Harbor Laboratory
Press, New York
Schmidt H (1974) On evolution in the anthozoan. In: Proc 2nd Int
Coral Reef Symp, Brisbane, Publ 1, pp 533±560
Spirin AS (1986) Ribosome structure and protein biosynthesis.
Benjamin/Cummings, Menlo Park
Swo€ord DL (1991) When are phylogeny estimates from molecular
and morphological data incongruent? In: Miyamoto MM,
Cracraft J (eds) Phylogenetic analysis of DNA sequences.
Oxford University Press, New York, pp 295±333
Swo€ord DL (1993) PAUP (phylogenetic analysis using parsimony), ver. 3.1.1. Illinois Natural History Survey, Chicago
Tompson JD, Higgins DG, Gibson TJ (1994) CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-speci®c gap penalties
and weight matrix choice. Nucleic Acids Res 22:4673±4680
Turbeville JM, Schulz JR, Ra€ RA (1994) Deuterostome phylogeny and the sister group of the chordates: evidence from
molecules and morphology. Mol Biol Evol 1(4):648±655
50
Veron JEN, Odorico DM, Chen CA, Miller DJ (1996) Reassessing
evolutionary relationships of scleractinian corals. Coral Reefs
15:1±9
Vrana PB, Milinkovitch MC, Powell JR, Wheeler WC (1994)
Higher level relationships of the Arctoid Carnivora based on
sequence data and ``total evidence''. Mol Phylogenet Evol
3(1):47±58
Wells JW, Hill D (1956) Coelenterata. In: Moore RC (ed) Treatise
on invertebrate paleontology; part F. University of Kansas
Press, Lawrence, pp F1±F166
Wyss AR, Novacek MJ, McKenna MC (1987) Amino acid
sequence versus morphological data and the interordinal
relationships of mammals. Mol Biol Evol 4:99±116