Molecular Phylogenetics and Evolution 49 (2008) 23–31

Contents lists available at ScienceDirect

Molecular Phylogenetics and Evolution

journal homepage: www.elsevier.com/locate/ympev

Testing the new animal phylogeny: A phylum level molecular analysis

of the animal kingdom
Sarah J. Bourlat a,*,1, Claus Nielsen b, Andrew D. Economou a, Maximilian J. Telford a
a
Department of Biology, University College London, Darwin Building, Gower Street, London WC1E 6BT, UK
b
Zoological Museum, Invertebrate Department, Universitetsparken 15, DK-2100 Copenhagen, Denmark

a r t i c l e i n f o a b s t r a c t

Article history: The new animal phylogeny inferred from ribosomal genes some years ago has prompted a number of rad-
Received 14 August 2007 ical rearrangements of the traditional, morphology based metazoan tree. The two main bilaterian clades,
Revised 27 May 2008 Deuterostomia and Protostomia, find strong support, but the protostomes consist of two sister groups,
Accepted 10 July 2008
Ecdysozoa and Lophotrochozoa, not seen in morphology based trees. Although widely accepted, not all
Available online 18 July 2008
recent molecular phylogenetic analyses have supported the tripartite structure of the new animal phy-
logeny. Furthermore, even if the small ribosomal subunit (SSU) based phylogeny is correct, there is a frus-
Keywords:
trating lack of resolution of relationships between the phyla that make up the three clades of this tree. To
Metazoan phylogeny
Ecdysozoa
address this issue, we have assembled a dataset including a large number of aligned sequence positions
Lophotrochozoa as well as a broad sampling of metazoan phyla.
Taxon sampling Our dataset consists of sequence data from ribosomal and mitochondrial genes combined with new
data from protein coding genes (5139 amino acid and 3524 nucleotide positions in total) from 37 repre-
sentative taxa sampled across the Metazoa. Our data show strong support for the basic structure of the
new animal phylogeny as well as for the Mandibulata including Myriapoda. We also provide some reso-
lution within the Lophotrochozoa, where we confirm support for a monophyletic clade of Echiura, Sipun-
cula and Annelida and surprising evidence of a close relationship between Brachiopoda and Nemertea.
Ó 2008 Elsevier Inc. All rights reserved.

1. Introduction The revolutionary paper of Aguinaldo et al. (1997) recovered

monophyletic clades of protostomes and deuterostomes, and di-
The interrelationships among living phyla of metazoans have vided the protostomes into ecdysozoans and lophotrochozoans.
been the subject of controversy for a century, and represent a chal- The Ecdysozoa—named for the shared characteristic moulting of
lenge in both morphological and molecular terms. The morphology the cuticle (ecdysis)—unites the arthropods with the pseudocoe-
based view of metazoan phylogeny implicit or explicit in zoology lomate nematodes, priapulids and other worms bearing an intro-
textbooks arranges metazoan phyla into acoelomates, pseudocoel- vert. The Lophotrochozoa, originally identified by Halanych
omates and coelomates in accordance with a gradual increase in (Halanych et al., 1995) unites the annelids and molluscs and a
complexity of body plans. However, an alternative scheme dividing number of other protostome phyla, most of which possess one or
the bilaterian animals into protostomes and deuterostomes was other of the two characteristics that give the clade its name: a cil-
established by Grobben (1908, 1910) and has had many adherents iated feeding structure called a lophophore and a trochophore type
over the years, e.g. Kaestner (1954/55), Remane et al. (1976), Niel- larva. In traditional phylogenies, a clade called the Articulata uni-
sen (1985, 2001), and Storch and Welsch (2004). The first molecu- ted the annelids and arthropods on the basis of shared segmenta-
lar phylogenetic analyses of animal phylogeny date back 20 years tion, but the molecular analyses separate these two groups and
(Field et al., 1988) and since this time our understanding of animal place Arthropoda in the Ecdysozoa and Annelida in the
relationships has undergone a series of revolutions (Halanych, Lophotrochozoa.
2004). Almost all of these changes in our understanding of how Another radical aspect of this so-called ‘‘new animal phylog-
the animals evolved derive from analyses of a single gene: the eny” is the relocation of several phyla that had been thought
small subunit ribosomal RNA gene (SSU rRNA). to represent intermediate grades of complexity and early
branches (Platyhelminthes, Nemertea and Nematoda) amongst
the coelomate groups at the crown of the tree (Aguinaldo and
* Corresponding author. Fax: +46 8 5195 4125. Lake, 1998). Analyses of SSU rRNA also moved phyla that had
E-mail address: [email protected] (S.J. Bourlat).
1 been linked to the deuterostomes, such as chaetognaths and
Present address: Department of Invertebrate Zoology, Swedish Museum of
Natural History, Box 50007, SE-104 05 Stockholm, Sweden. lophophorates, into the protostomes. Additional support for the

1055-7903/$ - see front matter Ó 2008 Elsevier Inc. All rights reserved.
doi:10.1016/j.ympev.2008.07.008
24 S.J. Bourlat et al. / Molecular Phylogenetics and Evolution 49 (2008) 23–31

Ecdysozoa/Lophotrochozoa split has been obtained from the 2. Results and discussion
analysis of Hox genes (de Rosa et al., 1999), horseradish peroxi-
dase (HRP) antibody staining (Haase et al., 2001), large subunit The tree obtained with our dataset of concatenated nuclear,
ribosomal RNA (LSU) (Mallatt and Winchell, 2002), myosin heavy ribosomal and mitochondrial genes is in agreement with the basic
chain (Ruiz-Trillo et al., 2002), and sodium/potassium ATPase structure of the ‘new animal phylogeny’ and supports the Ecdyso-
(Anderson et al., 2004). Resolution within the Lophotrochozoa zoa/Lophotrochozoa hypothesis (Fig. 1). We find that the Coelo-
and Ecdysozoa remains limited. mata hypothesis, which groups the animals with a mesodermally
In contrast to the support for the new animal phylogeny, a lined body cavity to the exclusion of the pseudocoelomate nema-
series of studies focussing on large datasets and few taxa todes and the acoelomate flatworms is very strongly rejected using
(reflecting the availability of whole genomes from a small num- the Bayes factor test (2 loge(B10) = 749.56) (Bayes factors represent
ber of model organisms (Blair et al., 2002; Wolf et al., 2004)) re- the ratio of the model likelihoods of the topologies of the two mod-
sulted in support for a monophyletic clade of coelomate els under consideration and values of 2 loge(B10) (two times the dif-
animals—‘Coelomata’: essentially a return to more traditional ference between the harmonic means of the two models) > 10 are
gradist theories. It seems likely, however, that this result is considered strong evidence to support one model over another).
due to phylogenetic errors stemming from the fast evolving Cae- Our tree (Fig. 1) supports a monophyletic clade of protostomes
norhabditis elegans (Copley et al., 2004; Irimia et al., 2007; (Bootstrap Value (BV) 96, Bayesian Posterior Probability (BPP)
Philippe et al., 2005). 1.00), monophyletic deuterostomes (92/1.00) and confirms the
Perhaps the most comprehensive study of metazoan relation- Ecdysozoa/Lophotrochozoa split (70/1.00 and 96/1.00, respec-
ships to date is of Philippe et al., with 146 genes from 35 species tively).
(Philippe et al., 2005), in which the problem of long branch
attraction is addressed by removing fast evolving taxa from the 2.1. Ecdysozoa
analysis, and using a better model of sequence evolution, result-
ing in the placement of nematodes and platyhelminthes in the The clade Ecdysozoa, originally described by Aguinaldo et al. on
Ecdysozoa and Lophotrochozoa, respectively. Some authors the basis of SSU DNA, groups the animals that share a cuticle shed
claimed that metazoan relationships cannot be fully resolved be- by moulting (or ecdysis) (Aguinaldo et al., 1997). The ecdysozoan
cause they ‘represent a closely spaced series of cladogenetic clade is recovered in our tree (Fig. 1, 70/1.00) and consists of the
events’, reflecting the Cambrian explosion (Rokas et al., 2005). nematodes + nematomorphs at the base (73/1.00) and then the pri-
However, using strategies to reduce the non-phylogenetic signal, apulids as a sister group to all arthropods (77/1.00).
such as increasing the number of species, replacing fast evolving
species by a slowly evolving one, and using a better model of se- 2.2. Arthropoda: Mandibulata or Paradoxopoda?
quence evolution (Baurain et al., 2007) results in improved reso-
lution in the metazoan tree. In addition, small sets of genes can Within the Arthropoda, our tree (Fig. 1) groups Crustacea
be just as effective given increased attention to taxon sampling with Hexapoda (93/1.00) in the Pancrustacea. This relationship
(Hedtke et al., 2006). had been observed previously and confirms existing molecular
In the present study, we use molecular data from the large phylogenies based on LSU and SSU (Mallatt et al., 2004), and
and small ribosomal subunits (3524 reliably aligned nucleotide studies based on mitochondrial genomes (Boore et al., 1998;
positions), from whole mitochondrial genomes (2048 reliably Hwang et al., 2001). In a more recent study, hexapods have been
aligned amino acids) and from 8 nuclear protein coding genes shown to be, in effect, terrestrial crustaceans (Regier et al.,
(3090 reliably aligned amino acids). To cover the broad range 2005). More controversially, our tree supports the Mandibulata
of divergences in our tree (deep splits at the level of the Precam- (88/1.00), the Pancrustacea + Myriapoda grouping together, with
brian) as well as more recent divergences (towards the tips of Chelicerata as an outgroup, rather than the Paradoxopoda (Myr-
the tree), we have used genes with different levels of conserva- iapoda with Chelicerata). Mandibulata, the monophyletic group-
tion across the metazoan phyla, and which overlap in their res- ing of all arthropods with mandibles (insects, crustaceans and
olution (see Section 4). myriapods) to the exclusion of the chelicerates has been sup-
We have sampled 168 species across the Metazoa and com- ported in molecular and morphological studies (Snodgrass,
bined them into 37 higher order taxa, each representing a phy- 1938; Giribet et al., 2001, 2005; Edgecombe et al., 2003). How-
lum or a class of the animal kingdom. Within the Ecdysozoa we ever, the Paradoxopoda grouping has been recovered in numer-
include priapulids, nematomorphs, nematodes, pycnogonids, ous molecular studies based on SSU and LSU (Mallatt et al.,
arachnids, chilopods, diplopods, xiphosurans, crustaceans and in- 2004; Mallatt and Giribet, 2006), mitochondrial genes (Hwang
sects. Within the Lophotrochozoa, we have sampled polychaete et al., 2001; Negrisolo et al., 2004) and protein coding genes
and clitellate annelids, sipunculids, echiurans, bivalve, gastropod, (Pisani et al., 2004). There are as yet no derived morphological
polyplacophoran, and cephalopod molluscs, nemerteans, phoro- features that unite the Paradoxopoda (Hwang et al., 2001; Mall-
nids and brachiopods as well as several deuterostomes. Our att et al., 2004; Pisani et al., 2004).
aim was first, to test the Lophotrochozoa/Ecdysozoa hypothesis The relative support for the Mandibulata versus Paradoxopoda
and second, to provide more resolution within these groups. To hypothesis was tested using Bayes factors. Bayes factor tests on
address the problem of long branch attraction, we use the slow- the dataset reject the Paradoxopoda tree topology
est evolving taxa from our dataset and we root the tree using (2 loge(B10) = 60.64).
the closest possible outgroups (Porifera, Hydrozoa and
Anthozoa).
2.3. Arthropoda: Chelicerata
Data were analyzed using Bayesian analysis of concatenated
translated protein sequences and the rRNA DNA sequences. To
In our analysis, we find that the horseshoe crabs (Xiphosura)
make the datasets more complete, we have, where necessary,
cluster with the arachnids (95/1.00), in agreement with other
pooled data from several species into a composite concatenated se-
molecular and morphological studies (Regier et al., 2005; Wheeler
quence representative of a given monophyletic group (see Table 1);
and Hayashi, 1998; Giribet et al., 2001), and that there is weak sup-
since we aim to obtain resolution at the level of the ‘phyla’ we con-
port for placing pycnogonids as basal chelicerates (57/0.97).
sider this to be acceptable for this type of analysis.
 S.J. Bourlat et al. / Molecular Phylogenetics and Evolution 49 (2008) 23–31 25

Table 1 Table 1 (continued)

Species used in concatenation
Higher order Lineage Species used in data % Missing data
Higher order Lineage Species used in data % Missing data taxon concatenation and gaps
taxon concatenation and gaps Crustaceans Phylum Arthropoda Artemia sp. 6.8
Xenoturbella Phylum Xenoturbellida Xenoturbella bocki 45.8 Subphylum Crustacea Triops longicaudatus
Triops cancriformis
Asteroids Phylum Echinodermata Asterias forbesii 13.8
Daphnia pulex
Asterias amurensis
Dilocarcinus pagei
Asterina miniata
Ostracoda sp.
Asterina pectinifera
Lepas anserifera
Asterias rubens
Carcinus maenas
Astropecten
Argulus sp.
brasiliensis
Acanthocyclops vernalis
Echinoids Phylum Echinodermata Encope michelini 6.4 Harbansus
Dendraster excentricus paucichelatus
Echinus esculentus Skogsbergia lerneri
Strongylocentrotus
Thysanurans Phylum Arthropoda Class Thermobia domestica 18.3
purpuratus
Insecta Ctenolepisma lineata
Eucidaris tribuloides
Ctenolepisma
Anthocidaris crassispina
longicaudata
Lytechinus variegatus
Arbacia punctulata Anopheles Phylum Arthropoda Class Anopheles albimanus 14.8
Arbacia lixula Insecta Anopheles gambiae
Anopheles
Hemichordates Phylum Hemichordata Saccoglossus 14.8
quadrimaculatus
Class Enteropneusta bromophenolosus
Saccoglossus sp. Drosophila Phylum Arthropoda Class Drosophila 0.5
Harrimania Insecta melanogaster
planktophilus Apis Phylum Arthropoda Class Apis mellifera 16.7
Balanoglossus carnosus Insecta
Saccoglossus
kowalevskii Clade III Phylum Nematoda Ascaris lumbricoides 21.2
Ptychodera flava nematodes Ascaris suum
Brugia malayi
Homo Phylum Chordata Homo sapiens 0 Onchocerca volvulus
Mus Phylum Chordata Mus musculus 0.2 Caenorhabditis Phylum Nematoda Caenorhabditis elegans 0.5
Teleosts Phylum Chordata Danio rerio 13.9 Nematomorphs Phylum Nematomorpha Paragordius varius 35.1
Oncorhynchus mykiss Gordius aquaticus
Chordodes morgani
Arachnids Phylum Arthropoda Tetranychus urticae 4.2
Subphylum Chelicerata Rhipicephalus Priapulids Phylum Priapulida Priapulus caudatus 13.3
appendiculatus Halicryptus spinulosus
Tegenaria gigantea Polychaetes Phylum Annelida Class Phyllodoce sp. 15.7
Phormictopus sp. Polychaeta Arenicola marina
Aphonopelma chalcodes Nereis macrydi
Loxosceles reclusa Sabella pavonina
Neacarus texanus Ophelina sp.
Heptathela Marenzelleria viridis
hangzhouensis Nereis succinea
Habronattus Nereis limbata
oregonensis Platynereis dumerilii
Aphonopelma sp.
Mastigoproctus Echiurans Phylum Echiura Urechis caupo 22.4
giganteus Echiurus echiurus
Listriolobus pelodes
Xiphosurans Phylum Arthropoda Limulus polyphemus 13.2
Subphylum Chelicerata Carcinoscorpius Clitellates Phylum Annelida Class Allolobophora sp. 17.8
rotundicauda Clitellata Lumbricus terrestris
Hirudo medicinalis
Pycnogonids Phylum Arthropoda Anoplodactylus 21.1 Eisenia fetida
Subphylum Chelicerata lentus Helobdella stagnalis
Tanystylum orbiculare Lumbricus terrestris
Endeis laevis
Colossendeis sp. Sipunculids Phylum Sipuncula Phascolion strombus 29.2
Callipallene sp. Phascolopsis gouldii
Nymphon gracile Bivalves Phylum Mollusca Class Mytilus edulis 7.1
Chilopods Phylum Arthropoda Lithobius sp. 8.2 Bivalvia Crassostrea gigas
Subphylum Myriapoda Strigamia maritima Pinctada fucata
Scolopendra Ostrea edulis
polymorpha Mytilus
Scutigera coleoptrata galloprovincialis
Scolopocryptops Macoma nasuta
sexspinosus Lampsilis cardium
Tuoba laticeps Nucula proxima
Thereuonema sp. Gastropods Phylum Mollusca Class Marisa sp. 11.2
Diplopods Phylum Arthropoda Polyxenus fasciculatus 6.7 Gastropoda Ilyanassa obsoleta
Subphylum Myriapoda Diplopoda sp. Tectura testudinalis
Archispirostreptus Lottia austrodigitalis
gigas Batillus cornutus
Oxidus gracilus Tegula brunnea
Narceus annularus Philine aperta
Narceus americanus Biomphalaria sp.
Orthoporus sp. Lottia digitalis
Thyropygus sp. Deroceras reticulatum
(continued on next page)
26 S.J. Bourlat et al. / Molecular Phylogenetics and Evolution 49 (2008) 23–31

Table 1 (continued) 2.4. Nematoida and Priapulida

Higher order Lineage Species used in data % Missing data
taxon concatenation and gaps Our analysis suggests that the nematomorphs (or horsehair
Polyplacophorans Phylum Mollusca Class Leptochiton sp. 16.3 worms), nematode-like parasites whose larvae are parasitic in
Polyplacophora Chaetopleura apiculata arthropods, are a sister group to the nematodes (73/1.00). This
Lepidochitona hartwegi
Katharina tunicata
grouping, named the Nematoida, is supported (although weakly)
in combined analyses of LSU and SSU (Mallatt et al., 2004; Mallatt
Cephalopods Phylum Mollusca Class Loligo pealei 27.7
Cephalopoda Octopus cyanea and Giribet, 2006), and in morphological studies. Nematodes and
Octopus rubescens nematomorphs share a number of synapomorphies, including the
Loligo bleekeri
structures of the body wall, the cuticle and the ectodermal nerve
Octopus vulgaris
cords (Nielsen, 2001).
Nemerteans Phylum Nemertea Cerebratulus sp. 44.2
Lineus longissimus
Our dataset recovers the priapulids as sistergroup of the arthro-
Lineus ruber pods (77/1.00). This placement disagrees with other molecular
Amphiporus sp. analyses based on SSU and LSU (Mallatt and Giribet, 2006; Mallatt
Brachiopods Phylum Brachiopoda Terebratalia transversa 29.2 et al., 2004), which position the Nematoida closer to the arthro-
Order Articulata Terebratulina retusa
pods than the priapulids.
Phoronids Phylum Phoronida Phoronis psammophila 27.1
Morphological studies, on the other hand unite priapulids, nem-
Phoronis
vancouverensis atodes and nematomorphs (together with kinorhychs, loriciferans
Phoronopsis viridis and gastrotrichs) into a monophyletic Cycloneuralia (Schmidt-
Bryozoans Phylum Bryozoa Tubulipora sp. 27.6 Rhaesa, 1998) or Introverta (Nielsen, 2001). The name Cycloneura-
Watersipora lia refers to a unique collar-shaped, peripharyngeal brain present
subtorquata
Class Gymnolaemata Bugula turrita
in gastrotrichs, nematodes, priapulids, kinorhynchs and loricifer-
Flustrellidra hispida ans (Nielsen, 2001). These phyla also share an inversible anterior
Urochordates Phylum Chordata Ciona intestinalis 17.7 end (the introvert) and, with the exception of the gastrotrichs,
Subphylum Urochordata Ascidia sp. the presence of a moulted cuticle and absence of locomotory cilia.
Ciona savignyi
We tested the alternative ‘Introverta’ hypothesis, (the mono-
Poriferans Phylum Porifera Leucosolenia sp. 16.1 phyletic grouping of priapulids, nematodes and nematomorphs,
Geodia neptuni
Tethya actinia based on the presence of an introvert), using Bayes factors and find
Axinella corrugata that this tree topology is rejected as significantly less well sup-
Scypha sp. (Sycon sp.) ported than one grouping the priapulids as sister group to the
Leucosolenia sp.
arthropods (2 loge(B10) = 46.18).
Anthozoans Phylum Cnidaria Class Urticina eques 22.3
Anthozoa Anemonia erythraea
Metridium senile 2.5. Lophotrochozoa
Antipathes galapagensis
Hydrozoans Phylum Cnidaria Class Hydra sp. 50.7 The Lophotrochozoa is a clade originally identified by SSU data
Hydrozoa Hydra vulgaris (Halanych et al., 1995) and includes, amongst several other phyla,
Hydractinia echinata
Hydra circumcincta the annelids, the molluscs, the phoronids, the brachiopods and the
Hydra littoralis bryozoans. Our tree (Fig. 1) supports the lophotrochozoan assem-
blage with a BV 96 and BPP 1.00. The relationships within the Lopho-
trochozoa are still not well understood, possibly due to poor taxon
Early studies, such as those of Snodgrass (Snodgrass, 1938) had sampling in multigene analyses. Other factors might also have con-
already established a morphological relationship between arach- tributed to lack of resolution, such as rapid radiation or multiple sub-
nids, xiphosurans and also the sea spiders (pycnogonids). However, stitutions, (as seen within the Mollusca) (Winnepenninckx et al.,
some studies based on molecules and morphology combined 1996), rate heterogeneity (Passamaneck et al., 2004) and saturation
(Giribet et al., 2001) placed the pycnogonids at the base of all ex- problems. As an indication of the lack of resolution in SSU analyses,
tant arthropods. The phylogenetic position of the pycnogonids the ribosomal genes alone do not resolve the various classes of mol-
leads to debate concerning the origin of arthropod head append- luscs or of the annelids as monophyletic groups (Aguinaldo et al.,
ages. The debate centres around the suggested homology of pycno- 1997; Giribet, 2002; Giribet et al., 2000; Halanych, 1998). A recent
gonid chelifores to the protocerebral ‘great appendages’ of certain study of lophotrochozoan phylogeny using LSU and SSU combined
Cambrian stem-group arthropods rather than to the deuteocere- recovers the nemerteans, annelids and molluscs as monophyletic
bral chelicerae of spiders (Maxmen et al., 2005). If true, this would groups but suggests that the Lophophorata (brachiopods, phoronids
support the idea of the pycnogonids as a basal branch of the arthro- and bryozoans) are polyphyletic (Passamaneck and Halanych, 2006),
pods (Budd and Telford, 2005). We tested the alternative hypothe- as previously found in the first study of lophophorates based on com-
sis that pycnogonids are basal to all other arthropods using Bayes plete SSU DNA (Halanych et al., 1995).
factors, and we find that this tree topology cannot be rejected (2 lo-
ge(B10) = 6.84), showing that the placement of pycnogonids is not 2.6. Mollusca: mono- or polyphyletic?
strongly resolved using our data.
The sistergroup relationship between pycnogonids and cheli- Our tree does not resolve the Mollusca as monophyletic (Fig. 1).
cerates, while poorly supported, is in agreement with some previ- We tested the alternative hypothesis that Mollusca are monophy-
ous molecular studies (Mallatt et al., 2004; Regier et al., 2005; letic but this alternative tree topology is rejected as significantly
Mallatt and Giribet, 2006), and with morphological analyses (Dun- less well supported using our dataset (2 loge(B10) = 188.84). Mono-
lop and Arango, 2005; Waloszek and Dunlop, 2002). This result phyletic Mollusca have been recovered from analyses of LSU and
suggests that pycnogonid chelifores may indeed be homologous SSU combined (Passamaneck and Halanych, 2006).
to spider chelicerae as traditionally thought and subsequent stud- Running the analyses excluding the mitochondrial data partition
ies of Hox gene expression boundaries seem to confirm this (Jager results in a monophyletic Mollusca (data not shown), putting into
et al., 2006). question the choice of mitochondrial genomes used in resolving
 S.J. Bourlat et al. / Molecular Phylogenetics and Evolution 49 (2008) 23–31 27

the Mollusca in our analysis. The tree constructed using mitochon- contradicted by molecular studies which show that all the loph-
drial genomes only (Appendix 3c) shows that the bivalves and gas- ophorate groups are protostomes (de Rosa et al., 1999; Halanych
tropods have long branches, which might have an effect on the et al., 1995). In other studies, based on SSU and morphology, bra-
accuracy of tree reconstruction in the combined dataset. In addition, chiopods and phoronids are found to be sister groups within the
the bivalve Mytilus galloprovincialis is known to have an unusual protostomes (Halanych et al., 1995; Peterson and Eernisse, 2001),
doubly uniparental mode of mitochondrial inheritance (Mizi et al., or the phoronids an ingroup of the brachiopods (Cohen, 2000; Co-
2005), which may result in mitochondrial recombination in that spe- hen et al., 1998). A recent study based on SSU and LSU combined
cies. Ribosomal data alone also do not resolve the Mollusca as mono- suggests that the Lophophorata are polyphyletic, and that the bra-
phyletic (Appendix 3b), a result which can be attributed to rate chiopods themselves are polyphyletic (Halanych et al., 1995;
heterogeneity in molluscan SSU and LSU, especially among the ceph- Passamaneck and Halanych, 2006).
alopods (Passamaneck et al., 2004). In our tree based on nuclear data Our tree positions the phoronids as a sister group to the poly-
only, we find monophyletic Mollusca. The polyplacophorans are ba- placophorans (70/1.00). We find that the brachiopods, nemerteans
sal to the bivalves, gastropods and cephalopods (Appendix 3a), in and bryozoans form a monophyletic clade.
support of the Conchifera hypothesis, an assemblage which groups We tested the hypothesis that the phoronids and brachiopods
the gastropods, cephalopods, bivalves, scaphopods and monopla- belong within the deuterostomes using Bayes factor tests, and find
cophorans based on the presence of a shell with a periostracum, (to- that this tree topology is very strongly rejected using the standard
gether secreted by a complex shell gland), to the exclusion of the criterion (2 loge(B10) = 1274.8).
polyplacophorans, which have eight shells (Nielsen, 2001).
2.9. Nemerteans
2.7. Annelida includes echiurans and sipunculids
Our results surprisingly show the nemerteans as the sister
Our data (Fig. 1) support a monophyletic clade of annelids, with group of the articulate brachiopods (88/1.00) (inarticulates were
high support values (BV 100, BPP 1.00). This clade includes the seg- not included in this study). The nemertean + brachiopod associ-
mented polychaetes and clitellates (oligochaetes + leeches) as well ation is an unexpected result, but our analyses lack platyhel-
as the echiurans (spoon worms) and the sipunculids (peanut minth sequences, and this may have had some influence on
worms), unsegmented marine worms which have probably second- this topology. The nemerteans have traditionally been associ-
arily lost their segmentation. The branching order of these phyla is ated with the flatworms, with which they share some morpho-
not well resolved in our tree (Fig. 1), and our taxon sampling does logical and embryological similarities. The strongest nemertean/
not permit us to see if the polychaetes are the paraphyletic stem- flatworm synapomorphy was long thought to be the lack of a
group of the other three lineages. Previous molecular studies have body cavity, but there is a debate as to whether the nemerteans
already suggested that the echiurans and sipunculids are derived are really acoelomate (Jenner, 2004; Turbeville, 2002). Similari-
annelids (Halanych et al., 2002; McHugh, 1997; Struck et al., 2007; ties in the larval ciliary bands of the pilidium larva of some
Bleidorn et al., 2006), and that Clitellata may also be included within nemertines and the Müller’s larva of some flatworms have been
the polychaete annelids (Struck et al., 2007; McHugh, 2005). Tradi- pointed out by Nielsen (2001). SSU data support the position of
tionally, the polychaetes (mainly marine worms) and the clitellates nemertines within a protostome coelomate clade (Turbeville
(the terrestrial oligochaetes and leeches) have been regarded as sep- et al., 1992). Recent LSU data also suggested an association of
arate monophyletic groups. In agreement with these results, the the nemertines with the brachiopods and phoronids, the nemer-
mitochondrial gene order of the sipunculid Phascolopsis gouldii is tean Tubulanus branching with the Brachiopoda in the LSU tree
very similar to that of the oligochaete Lumbricus terrestris (Boore of Passamaneck and Halanych (Passamaneck and Halanych,
and Staton, 2002). According to ribosomal data, the echiurans are re- 2006). However, the brachiopods are polyphyletic in this tree,
lated to capitellid polychaetes, and should be regarded as secondar- some grouping with the annelids and others as basal to other
ily unsegmented polychaetes (Bleidorn et al., 2003). The observation lophotrochozoan phyla (Passamaneck and Halanych, 2006).
of a segmented ventral nervous system in juvenile echiurans simi- There is also some evidence for a brachiopod-nemertean associ-
larly supports the inclusion of the echiurans in the annelids ation from their mitochondrial gene orders. The ND2-CO1 gene
(Hessling and Westheide, 2002). Embryology of the sipunculids, par- boundary seems to be shared between nemerteans, brachiopods
ticularly the existence of a so-called molluscan cross, has been inter- and spiralian taxa (Turbeville, 2002). However, it is difficult to
preted as indicative of molluscan affinities, but this has now been point out any morphological synapomorphy between brachio-
shown to be a poor character, which has been based on subjective pods and spiralian taxa including nemerteans.
interpretations of blastomere patterns (Jenner, 2003; Maslakova et
al., 2004). Our data clearly support the monophyletic origin of poly- 2.10. Deuterostomes
chaete annelids, clitellates, echiurans and sipunculids, suggesting
that the former morphological analyses of these groups may have The deuterostomes have traditionally been distinguished from
underestimated their morphological diversity, and implies that seg- the protostomes based on a number of morphological features
mentation is likely to have been lost in the lineages leading to the such as the fate of the blastopore, the origin of the mesoderm
echiurans and sipunculids. and the radial cleavage pattern of the embryo. The deuterostome
grouping has remained largely unchanged in molecular studies,
2.8. Lophophorates comprising echinoderms, hemichordates, urochordates, cephalo-
chordates, vertebrates, and the new phylum Xenoturbellida (Bour-
The lophophorates (brachiopods, phoronids and bryozoans) lat et al., 2003). We find high support for the deuterostome
have traditionally been classified together based on similarities grouping in our dataset (92/1.00), with the exception of the uro-
of the ciliated tentacles, called the lophophore, and the putative chordates (see below).
archimery of the body, i.e. the possession of a prosome, a meso- Within the deuterostomes, our tree supports the Ambulacraria
some with the lophophore, and a metasome, each with one or a (echinoderms + hemichordates) (75/1.00). The Ambulacraria
pair of coelomic sacs (Hyman, 1959). Some authors have regarded grouping has been recovered in other molecular studies based on
the phoronids and brachiopods as basal deuterostomes, based on SSU (Halanych, 1995) and on LSU and SSU combined (Furlong and
embryological characters (Nielsen, 2001), but this idea has been Holland, 2002; Winchell et al., 2002). In terms of synapomorphies,
28 S.J. Bourlat et al. / Molecular Phylogenetics and Evolution 49 (2008) 23–31

Fig. 1. Bayesian inference protein tree of the animal kingdom based on concatenated nuclear, ribosomal and mitochondrial genes (support values: BI posterior probability/
bootstrap).

the echinoderms and hemichordates share a dipleurula larva (Niel- possibly be explained by their fast rate of evolutionary change
sen, 2001). attracting them to the base of the tree. This result has also been
Our data confirm the phylogenetic position of Xenoturbella as a seen in previous studies (Telford et al., 2003). We tested the alter-
sister group to the Ambulacraria (52/0.82), and the Ambulacraria/ native hypothesis that the urochordates belong to the deuterosto-
Xenoturbella clade as a sister group to the vertebrates, as previously mes using Bayes factors, and we find that this topology is rejected
shown in studies based on SSU, CO1 and CO2 genes (Bourlat et al., as significantly less well supported using the standard criterion
2003), whole mitochondrial genome and phylogenomic data (2 loge(B10) = 56.39).
(Bourlat et al., 2006).
3. Conclusions
2.11. Urochordates
To address the problems of both stochastic and systematic
Our tree shows the urochordates at the base of the Bilateria. errors in reconstructing the phylogeny of the Metazoa, we have
This is presumably an artefact of long branch attraction, and can assembled a dataset from 22 nuclear and mitochondrial genes.
 S.J. Bourlat et al. / Molecular Phylogenetics and Evolution 49 (2008) 23–31 29

We have used the traditional PCR approach to generate novel data the EST analysis of Dunn et al. clearly supports Pycnogon-
from 3 nuclear loci to add to the previously available datasets to ida + Chelicerata, Cycloneuralia, as well as the more plausible
which we have added new sequences from several taxa. To maxi- monophyletic Mollusca and Urochordata as the sister group to
mise the completeness of each operational taxonomic unit (OTU), Chordata. These discrepancies highlight the strength of the EST
we merged the data from a total of 168 species into 37 composite approach and demonstrate that better resolution can be achieved
sequences representing almost all major animal lineages biasing with larger datasets and better gene selection strategies that
the selection towards the more slowly evolving exemplars. The avoid paralogy (Dunn et al., 2008). There are a number of other
structure of our tree is in general agreement with previous efforts ways in which resolution in the Metazoan tree can be improved,
and, while obviously not entirely independent, gives further sup- the most important of which are the inclusion of data from addi-
port to the tripartite structure of the new animal phylogeny con- tional taxa in order to reveal homoplasy, and the development of
sisting of deuterostomes, lophotrochozoans and ecdysozoans. We methods to deal with systematic error such as unequal or un-
also draw attention to a number of relationships that, being con- even rates of evolution and compositional bias (Phillipe and
troversial, are of particular note, (i) the monophyly of the Mandib- Telford, 2006).
ulata (i.e. Myriapoda grouping with Crustacea and Hexapoda
rather than with Chelicerata), (ii) the monophyletic group of anne-
lids, echiurans and sipunculids, (iii) the basal position of Nemato- 4. Materials and methods
ida within Ecdysozoa and the sistergroup relationship between
Priapulida and Euarthropoda, (iv) the relationship between Bra- 4.1. Degenerate primer design
chiopoda and Nemertea.
Most of these clades, although not all found in previous molec- To find orthologs across the metazoan taxa, protein coding
ular phylogenies based on smaller data sets do make sense of the genes from Saccharomyces cervisiae, Caenorrhabditis elegans,
morphological features of the animals in question. The Mandibula- Drosophila melanogaster and Human were searched using the Basic
ta share a common head morphology which includes but is not re- Local Alignment Search Tool (BLAST) (http://www.ncbi.nlm.nih.
stricted to the common presence of a mandible (Edgecombe et al., gov/BLAST/) against each other and in all possible combinations.
2003). The Pycnogonida and Chelicerata share chelate first append- Genes were chosen according to percentage similarity, alignable length
ages which have been confirmed as positionally homologous by re- and the availability of conserved regions for primer design. Primers
cent studies of Hox gene expression (Jager et al., 2006). The Echiura were designed using codehop (http://blocks.fhcrc.org/blocks/
and Sipuncula have been shown to have a very similar mitochon- codehop.html). The chosen genes were dyskerin (a centromere/
drial gene order to the Annelida and even to have an annelid-like microtubule binding protein of 60–70% similarity across the yeast,
metameric nervous system in the larvae (Hessling and Westheide, fly, worm and human genomes), vacuolar ATP synthase subunit
2002; Wanninger et al., 2005). (with 70–80% percentage similarity) and carnitine palmytoyltrans-
We have compared the support for each of these relationships ferase (30–40% similarity). In addition, other genes were chosen
with alternative topologies using Bayes factors and in almost all according to their prior use in other studies and availability in
cases (apart from the position of the pycnogonids as chelicerates GenBank, such as enolase, glyceraldehyde 3-phosphate dehydroge-
rather than basal arthropods, 2 loge(B10) = 6.84) we have been nase, elongation factor 1-a, sodium/potassium ATPase and RNA
able to reject the alternative topology as significantly well sup- polymerase II. Others such as myosin were avoided deliberately
ported. While this may seem encouraging, we believe that the because they have a large number of paralogs. Mitochondrial gen-
current criterion for the use of Bayes factors is not stringent omes and SSU and LSU ribosomal DNA sequences were downloaded
enough. from GenBank through NCBI (http://www.ncbi.nlm.nih.gov/).
We do not put great faith in the relative positions of Priapulida
and Nematoida on our tree as the reverse pattern has also been de- 4.2. Specimen collection
scribed but highlight it in order to emphasise the lack of resolution
here. The most surprising of the clades we highlight is the nemer- Marine invertebrates were collected from Millport Marine Sta-
tean plus brachiopod clade. While highly supported and seemingly tion, UK, Kristineberg Marine Research Station, Sweden or Friday
robust, it seems problematic to accept the sistergroup relationship Harbour Laboratories, Washington, USA, and identified to species
between these two very different clades. The nemerteans are clas- level by Claus Nielsen. Arthropod specimens were kindly donated
sic spiralians with a trochophore larva apparently homologous to by colleagues.
that of annelids and molluscs (Maslakova et al., 2004) while bra-
chiopods are radially cleaving and there has been no suggestion 4.3. RT-PCR, cloning, sequencing
that their larva could be a derived trochophore (Nielsen, 2005).
Another concern is that we do not find the Brachiopoda and Pho- Total RNA was prepared using the RNeasy minikit (Qiagen).
ronida as sistergroups suggesting the phylogenetic signal might cDNA for RT-PCR was prepared using Expand Reverse Transcrip-
not be as reliable as one might hope. This said, it is uncontentious tase (Roche) and random hexanucleotide primers (Roche). Poly-
that the non-spiralian brachiopods are relatively close to a number merase chain reaction was carried out using degenerate primers
of spiralian groups and so, unless shown to be outside of a spiralian (see Appendix 1) and Taq polymerase (Roche) at the following tem-
clade, an explanation of their different pattern of embryogenesis peratures: 1 cycle: 94 °C, 2 min; 30 cycles 94 °C, 30 s; 50 °C, 60 s;
will always be required. We note additionally, that the phyloge- 72 °C 90 s; 1 cycle 72 °C, 10 min. These non-stringent annealing
nomic analysis of Dunn et al. based on 150 genes from 77 taxa conditions fit all primer pairs in most species, but meant that mul-
(Dunn et al., 2008) published during the reviewing process of our tiple bands were obtained on the gel.
manuscript also supports a sister group relationship of nemerteans The right sized PCR products were gel purified using QIAquick
and brachiopods, referred to as ‘Clade A’, giving further support for gel extraction kit (Qiagen) and cloned using pGEM-T vector (Pro-
this association. mega). Clones were sequenced using T7 (50 -TAATACGACTCACTA
The problems that remain in our analysis include portions of TAGGG-30 ) and SP6 (50 -GATTTAGGTGACACTATAG-30 ) primers and
the tree that are plausible yet weakly supported (Pycnogon- BigDye Terminator v3.1 (Applied Biosystems). Sequences were
ida + Chelicerata, Priapulida + Euarthropoda) or implausible analyzed and contigs made using Sequencher (Gene Codes Corpo-
(non-monophyletic Mollusca and basal Urochordata). In contrast, ration) and Lasergene (DNASTAR).
30 S.J. Bourlat et al. / Molecular Phylogenetics and Evolution 49 (2008) 23–31

4.4. Alignment analysis was constrained to contain alternative potential mono-

phyletic groups. The Bayes factor (B10) represents the ratio of the
Protein coding nucleotide sequences were translated according model likelihoods of the two topologies under consideration
to the universal genetic code for nuclear protein or their taxon spe- (Nylander et al., 2004; Wiens et al., 2005). Values of 2 loge(B10)
cific genetic code (for mitochondrial proteins) using the software were calculated (i.e., two times the difference between the har-
translatorX, and the amino acids were aligned using Muscle (Edgar, monic means of the post burn-in log-likelihoods of the two mod-
2004) as the alignment tool. The amino acid alignment was edited els) and values > 10 are considered to be very strong evidence
by eye in MacClade (version 4.07 (Maddison and Maddison, 2000)). favouring one model over the other (Kass and Raftery, 1995).
The original nucleotides were finally aligned according to the ami-
no acid alignment again using translatorX (Telford, unpublished). Acknowledgments
Unreliably aligned positions were excluded from the analyses
using MacClade version 4.06. S.J.B. was funded by the Wellcome Trust and by the BBSRC. We
thank Cassandra Extavour, Ben Hanelt, Heather Wilson and John S.
4.5. Data concatenation Young for kindly providing specimens. We also wish to thank Julia
Llewellyn-Hughes and Claire Griffin at the Natural History Mu-
In order to reduce the proportion of missing data in the concat- seum (London) sequencing facility. We would like to thank two
enated alignment, sequences from several species were combined anonymous reviewers for improving the manuscript.
into a composite higher-level taxon (Class or Phylum) and re-
named accordingly (e.g. Asteroids, Echinoids, Arachnids, Gastro- Appendix A. Supplementary data
pods). In cases where more than one species representing the
particular higher-level taxon was represented for a given gene, Supplementary data associated with this article can be found, in
the available sequences (e.g. several different mollusc sequences) the online version, at doi:10.1016/j.ympev.2008.07.008.
were ranked according to their average distance from all other se-
quences in the alignment (see Table 1 for constituent species for
References
each higher-level taxon). A composite sequence was then built
up using as much of the top ranked sequences (shortest average Aguinaldo, A.M.A., Lake, J.A., 1998. Evolution of the multicellular animals. Am. Zool.
distance) as was present. A Perl script was written to automate this 38, 878–887.
procedure. In the case of Xenoturbella bocki, Homo sapiens, Mus Aguinaldo, A.A., Turbeville, J.M., Linford, L.S., Rivera, M.C., Garey, J.R., Raff, R.A., Lake,
J.A., 1997. Evidence for a clade of nematodes, arthropods and other moulting
musculus, Drosophila melanogaster, Apis mellifera, and Caenorhabdi- animals. Nature 387, 489–493.
tis elegans we used sequence data from those species only. Anderson, F.E., Cordoba, A.J., Thollesson, M., 2004. Bilaterian phylogeny based on
analyses of a region of the sodium-potassium ATPase beta-subunit gene. J. Mol.
Evol. 58, 252–268.
4.6. Phylogenetic analyses
Baurain, D., Brinkmann, H., Philippe, H., 2007. Lack of resolution in the animal
phylogeny: closely spaced cladogeneses or undetected systematic errors? Mol.
For Bayesian analyses of amino acid data (Huelsenbeck and Biol. Evol. 24, 6–9.
Ronquist, 2001) the data were divided into three partitions: ribo- Blair, J.E., Ikeo, K., Gojobori, T., Hedges, S.B., 2002. The evolutionary position of
nematodes. BMC Evol. Biol. 2, 7.
somal RNA, mitochondrial protein and nuclear protein. For nucleo- Bleidorn, C., Podsiadlowski, L., Bartolomaeus, T., 2006. The complete mitochondrial
tides the GTR + G5 was used. MtREV substitution matrix genome of the orbiniid polychaete Orbinia latreillii (Annelida, Orbiniidae)—a
(mitochondrial sequences) or WAG substitution matrix (nuclear novel gene order for Annelida and implications for annelid phylogeny. Gene
370, 96–103.
genes) were fixed and likelihood settings were set to rates = gam- Bleidorn, C., Vogt, L., Bartolomaeus, T., 2003. New insights into polychaete
ma, ngammacat = 5. Two independent runs with two chains were phylogeny (Annelida) inferred by 18S rDNA sequences. Mol. Phylogenet. Evol.
run for 5,000,000 generations and sampled every 100th generation. 29, 279–288.
Boore, J.L., Staton, J.L., 2002. The mitochondrial genome of the Sipunculid
A majority rule consensus tree was constructed from trees sampled Phascolopsis gouldii supports its association with Annelida rather than
after the likelihood estimates had plateaued. MrBayes, version Mollusca. Mol. Biol. Evol. 19, 127–137.
3.1.2 was used (Huelsenbeck and Ronquist, 2001). Boore, J.L., Lavrov, D.V., Brown, W.M., 1998. Gene translocation links insects and
crustaceans. Nature 392, 667–668.
Bourlat, S.J., Juliusdottir, T., Lowe, C.J., Freeman, R., Aronowicz, J., Kirschner, M.,
4.7. Non-parametric bootstrapping Lander, E.S., Thorndyke, M., Nakano, H., Kohn, A.B., Heyland, A., Moroz, L.L.,
Copley, R.R., Telford, M.J., 2006. Deuterostome phylogeny reveals monophyletic
chordates and the new phylum Xenoturbellida. Nature 444, 85–88.
As the mixture of evolutionary models we used (separate mod-
Bourlat, S.J., Nielsen, C., Lockyer, A.E., Littlewood, D.T., Telford, M.J., 2003.
els for nuclear and mitochondrial amino acids and rRNA nucleo- Xenoturbella is a deuterostome that eats molluscs. Nature 424, 925–928.
tides) are not available in likelihood packages with which we Budd, G.E., Telford, M.J., 2005. Evolution: along came a sea spider. Nature 437,
could run a standard non-parametric bootstrap, we constructed 1099–1102.
Cohen, B.L., 2000. Monophyly of brachiopods and phoronids: reconciliation of
100 boostrapped datasets followed by 100 Bayesian analyses. Each molecular evidence with Linnaean classification (the subphylum
of the three data types was bootstrapped 100 times and then the Phoroniformea nov). P. Roy. Soc. Lond. B Biol. 267, 225–231.
separate bootstrap datasets concatenated. The same models were Cohen, B.L., Gawthrop, A., Cavalier-Smith, T., 1998. Molecular phylogeny of
brachiopods and phoronids based on nuclear-encoded small subunit
used as for the main analysis in Bayesian analyses of each of the ribosomal RNA gene sequences. Philos. Trans. R. Soc. B 353, 2039–2061.
100 concatenated datasets. The MCMC was run for 500,000 gener- Copley, R.R., Aloy, P., Russell, R.B., Telford, M.J., 2004. Systematic searches for
ations. The first 80% of trees were discarded before the Bayesian molecular synapomorphies in model metazoan genomes give some support for
Ecdysozoa after accounting for the idiosyncrasies of Caenorhabditis elegans.
consensus tree was estimated. We used the consensus tree from Evol. Dev. 6, 164–169.
each MrBayes run to produce the final bootstrap consensus. The de Rosa, R., Grenier, J.K., Andreeva, T., Cook, C.E., Adoutte, A., Akam, M., Carroll, S.B.,
PHYLIP program consense (Felsenstein, 1989) was used to con- Balavoine, G., 1999. Hox genes in brachiopods and priapulids and protostome
evolution. Nature 399, 772–776.
struct an overall consensus from the 100 Bayesian consensus trees. Dunlop, J.A., Arango, C.P., 2005. Pycnogonid affinities: a review. J. Zool. Syst. Evol.
Res. 43, 8–21.
4.8. Bayes factor tests Dunn, C.D., Hejnol, A., Matus, D.Q., Pang, K., Browne, W.E., Smith, S.A., Seaver, E.,
Rouse, G.W., Obst, M., Edgecombe, G.D., Sorensen, M.V., Haddock, S.H.D.,
Schmidt-Rhaesa, A., Okosu, A., Kristensen, R.M., Wheeler, W.C., Martindale,
MrBayes was also used to calculate Bayes factors to compare M.Q., Giribet, G., 2008. Broad phylogenomic sampling improves resolution of
the optimal topology with that of topologies in which the Bayesian the animal tree of life. Nature 452, 665–780.
 S.J. Bourlat et al. / Molecular Phylogenetics and Evolution 49 (2008) 23–31 31

Edgar, R.C., 2004. MUSCLE: multiple sequence alignment with high accuracy and Maxmen, A., Browne, W.E., Martindale, M.Q., Giribet, G., 2005. Neuroanatomy of sea
high throughput. Nucleic Acids Res. 32, 1792–1797. spiders implies an appendicular origin of the protocerebral segment. Nature
Edgecombe, G.D., Richter, S., Wilson, G.D.F., 2003. The mandibular gnathal edges: 437, 1144–1148.
Homologous structures throughout Mandibulata? Afr. Invert. 44, 115–135. McHugh, D., 2005. Molecular systematics of polychaetes (Annelida). Hyrobiologia,
Felsenstein, J., 1989. PHYLIP-phylogeny inference package. Cladistics 5, 164–166. 309–318.
Field, K.G., Olsen, G.J., Lane, D.J., Giovannoni, S.J., Ghiselin, M.T., Raff, E.C., Pace, N.R., McHugh, D., 1997. Molecular evidence that echiurans and pogonophorans are
Raff, R.A., 1988. Molecular phylogeny of the animal kingdom. Science 239, 748– derived annelids. Proc. Natl. Acad. Sci. USA 94, 8006–8009.
753. Mizi, A., Zouros, E., Moschonas, N., Rodakis, G.C., 2005. The complete maternal and
Furlong, R.F., Holland, P.W., 2002. Bayesian phylogenetic analysis supports paternal mitochondrial genomes of the Mediterranean mussel Mytilus
monophyly of ambulacraria and of cyclostomes. Zoolog. Sci. 19, 593–599. galloprovincialis: implications for the doubly uniparental inheritance mode of
Giribet, G., Richter, S., Edgecombe, G.D., Wheeler, W.C., 2005. The position of mtDNA. Mol. Biol. Evol. 22, 952–967.
crustaceans within the Arthropoda—evidence from nine molecular loci and Negrisolo, E., Minelli, A., Valle, G., 2004. The mitochondrial genome of the house
morphology. In: Koenemann, S., Jenner, R.A. (Eds.), Crustacean Issues 16: centipede scutigera and the monophyly versus paraphyly of myriapods. Mol.
Crustacea and Arthropod Relationships. Festschrift for Frederick R. Schram. Biol. Evol. 21, 770–780.
Taylor & Francis, Boca Raton, pp. 307–352. Nielsen, C., 2005. Trochophora larvae: cell-lineages, ciliary bands and body regions.
Giribet, G., 2002. Current advances in the phylogenetic reconstruction of metazoan 2. Other groups and general discussion. J. Exp. Zool. (Mol. Dev. Evol.) 304B, 401–
evolution. A new paradigm for the Cambrian explosion? Mol. Phylogenet. Evol. 447.
24, 345–357. Nielsen, C., 2001. Animal Evolution, Interrelationships of the Living Phyla, second
Giribet, G., Edgecombe, G.D., Wheeler, W.C., 2001. Arthropod phylogeny based on ed. Oxford University Press, Oxford, England.
eight molecular loci and morphology. Nature 413, 157–161. Nielsen, C., 1985. Animal phylogeny in the light of the trochaea theory. Biol. J. Linn.
Giribet, G., Distel, D.L., Polz, M., Sterrer, W., Wheeler, W.C., 2000. Triploblastic Soc. 25, 243–299.
relationships with emphasis on the acoelomates and the position of Nylander, J.A.A., Ronquist, F., Huelsenbeck, J.P., Nieves-Aldrey, J.L., 2004. Bayesian
Gnathostomulida, Cycliophora, Plathelminthes, and Chaetognatha: a phylogenetic analysis of combined data. Syst. Biol. 53, 47–67.
combined approach of 18S rDNA sequences and morphology. Syst. Biol. 49, Passamaneck, Y., Halanych, K.M., 2006. Lophotrochozoan phylogeny assessed with
539–562. LSU and SSU data: evidence of lophophorate polyphyly. Mol. Phylogenet. Evol.
Grobben, K., 1910. Claus-Grobben: Lehrbuch der Zoologie, second ed. Elvert’sche 40, 20–28.
Verlagsbuchhandlung, Marburg. Passamaneck, Y.J., Schander, C., Halanych, K.M., 2004. Investigation of molluscan
Grobben, K., 1908. Die systematische Einteilung des Tierreichs. Verhandlungen der. phylogeny using large-subunit and small-subunit nuclear rRNA sequences. Mol.
Zoologisch-Botanischen Gesellschaft in Wien 58, 491–511. Phylogenet. Evol. 32, 25–38.
Haase, A., Stern, M., Wachtler, K., Bicker, G., 2001. A tissue-specific marker of Peterson, K.J., Eernisse, D.J., 2001. Animal phylogeny and the ancestry of bilaterians:
Ecdysozoa. Dev. Genes Evol. 211, 428–433. inferences from morphology and 18S rDNA gene sequences. Evol. Dev. 3, 170–
Halanych, K.M., 2004. The new view of animal phylogeny. Annu. Rev. Ecol. Evol. 205.
Syst. 35, 229–256. Phillipe, H., Telford, M.J., 2006. Large-scale sequencing and the new animal
Halanych, K.M., Dahlgren, T.G., McHugh, D., 2002. Unsegmented annelids? Possible phylogeny. TREE 21, 614–620.
origins of four lophotrochozoan worm taxa. Int. Comp. Biol. 42, 678–684. Philippe, H., Lartillot, N., Brinkmann, H., 2005. Multigene analyses of bilaterian
Halanych, K.M., 1998. Considerations for reconstructing metazoan history: Signal, animals corroborate the monophyly of Ecdysozoa, Lophotrochozoa, and
resolution, and hypothesis testing. Am. Zool. 38, 929–941. Protostomia. Mol. Biol. Evol. 22, 1246–1253.
Halanych, K.M., 1995. The phylogenetic position of the pterobranch hemichordates Pisani, D., Poling, L.L., Lyons-Weiler, M., Hedges, S.B., 2004. The colonization of land
based on 18S rDNA sequence data. Mol. Phylogenet. Evol. 4, 72–76. by animals: molecular phylogeny and divergence times among arthropods.
Halanych, K.M., Bacheller, J.D., Aguinaldo, A.M., Liva, S.M., Hillis, D.M., Lake, J.A., BMC Biol. 2, 1.
1995. Evidence from 18S ribosomal DNA that the lophophorates are protostome Regier, J.C., Shultz, J.W., Kambic, R.E., 2005. Pancrustacean phylogeny: hexapods are
animals. Science 267, 1641–1643. terrestrial crustaceans and maxillopods are not monophyletic. Proc. Biol. Sci.
Hedtke, S.H., Townsend, T.M., Hillis, D.M., 2006. Resolution of phylogenetic conflict 272, 395–401.
in large data sets by increased taxon sampling. Syst. Biol. 55, 522–529. Remane, A., Storch, V., Welsch, U., 1976. Systematische Zoologie. Fischer, Jena.
Hessling, R., Westheide, W., 2002. Are Echiura derived from a segmented ancestor? Rokas, A., Kruger, D., Carroll, S.B., 2005. Animal evolution and the molecular
Immunohistochemical analysis of the nervous system of developmental stages signature of radiations compressed in time. Science 310, 1933–1938.
of Bonellia. J. Morphol. 252, 100–113. Ruiz-Trillo, I., Paps, J., Loukota, M., Ribera, C., Jondelius, U., Baguñá, J., Riutort, M.,
Huelsenbeck, J.P., Ronquist, F., 2001. MRBAYES: Bayesian inference of phylogeny. 2002. A phylogenetic analysis of myosin heavy chain type II sequences
Bioinformatics 17, 754–755. corroborates that Acoela and Nemertodermatida are basal bilaterians. Proc.
Hwang, U.W., Friedrich, M., Tautz, D., Park, C.J., Kim, W., 2001. Mitochondrial Natl. Acad. Sci. USA 99, 11246–11251.
protein phylogeny joins myriapods with chelicerates. Nature 413, 154–157. Schmidt-Rhaesa, A., 1998. Phylogenetic relationships of the nematomorpha—a
Hyman, L.H., 1959. The invertebrates. Smaller Coelomate Groups, vol. 5. McGraw- discussion of current hypotheses. Zool. Anz. 236, 203–216.
Hill, New York. Snodgrass, R.E., 1938. Evolution of the Annelida, Onychophora, and Arthropoda.
Irimia, M., Maeso, I., Penny, D., Garcia-Fernandez, J., Roy, S.W., 2007. Rare coding Smithsonian Miscellaneous Collections 97 (6), 1–159.
sequence changes are consistent with ecdysozoa, not Coelomata. Mol. Biol. Evol. Storch, V., Welsch, U., 2004. Lehrbuch der Zoologie, sixth ed. Spektrum, Heidelberg.
24, 1604–1607. Struck, T.H., Schult, N., Kusen, T., Hickman, E., Bleidorn, C., McHugh, D., Halanych,
Jager, M., Murienne, J., Clabaut, C., Deutsch, J., Le Guyader, H., Manuel, M., 2006. K.M., 2007. Annelid phylogeny and the status of Sipuncula and Echiura. BMC
Homology of arthropod anterior appendages revealed by Hox gene expression Evol. Biol. 7, 57.
in a sea spider. Nature 441, 506–508. Telford, M.J., Lockyer, A.E., Cartwright-Finch, C., Littlewood, D.T., 2003. Combined
Jenner, R.A., 2004. Towards a phylogeny of the Metazoa: evaluating alternative large and small subunit ribosomal RNA phylogenies support a basal position of
phylogenetic positions of Platyhelminthes, Nemertea, and Gnathostomulida, the acoelomorph flatworms. Proc. Biol. Sci. 270, 1077–1083.
with a critical reappraisal of cladistic characters. Contrib. Zool. 73, 3–163. Turbeville, J.M., 2002. Progress in nemertean biology: development and phylogeny.
Jenner, R.A., 2003. Unleashing the force of cladistics? Metazoan phylogenetics and Integr. Comp. Biol. 42, 692–703.
hypothesis testing. Integr. Comp. Biol. 43, 207–218. Turbeville, J.M., Field, K.G., Raff, R.A., 1992. Phylogenetic position of phylum
Kaestner, A., 1954/55. Lehrbuch der Speziellen Zoologie, Teil 1., 1 Halbband. Gustav Nemertini, inferred from 18S rRNA sequences: molecular data as a test of
Fischer, Jena. morphological character homology. Mol. Biol. Evol. 9, 235–249.
Kass, R.E., Raftery, A.E., 1995. Bayes Factors. J. Am. Stat. Assoc. 90, 773–795. Waloszek, D., Dunlop, J.A., 2002. A larval sea spider (Arthropoda: Pycnogonida) from
Maddison, D.R., Maddison, W.P., 2000. MacClade 4: Analysis of Phylogeny and the Upper Cambrian ‘Orsten’ of Sweden, and the phylogenetic position of
Character Evolution. Version 4.0. Sinauer Associates, Sunderland, pycnogonids. Palaeontology 45, 421–446.
Massachusetts. Wanninger, A., Koop, D., Bromham, L., Noonan, E., Degnan, B.M., 2005. Nervous and
Mallatt, J., Giribet, G., 2006. Further use of nearly complete, 28S and 18S rRNA genes muscle system development in Phascolion strombus (Sipuncula). Dev. Genes
to classify Ecdysozoa: 37 more arthropods and a kinorhynch. Mol. Phylogenet. Evol. 215, 509–518.
Evol. 40, 772–794. Wheeler, W.C., Hayashi, C.Y., 1998. The phylogeny of the extant chelicerate orders.
Mallatt, J.M., Garey, J.R., Shultz, J.W., 2004. Ecdysozoan phylogeny and Bayesian Cladistics Int. J. Willi Hennig Soc. 14, 173–192.
inference: first use of nearly complete 28S and 18S rRNA gene sequences Wiens, J.J., Fetzner, J.W., Parkinson, C.L., Reeder, T.W., 2005. Hylid frog phylogeny
to classify the arthropods and their kin. Mol. Phylogenet. Evol. 31, 178– and sampling strategies for speciose clades. Syst. Biol. 54, 778–807.
191. Winchell, C.J., Sullivan, J., Cameron, C.B., Swalla, B.J., Mallatt, J., 2002. Evaluating
Mallatt, J., Winchell, C.J., 2002. Testing the new animal phylogeny: first use of hypotheses of deuterostome phylogeny and chordate evolution with new LSU
combined large-subunit and small-subunit rRNA gene sequences to classify the and SSU ribosomal DNA data. Mol. Biol. Evol. 19, 762–776.
protostomes. Mol. Biol. Evol. 19, 289–301. Winnepenninckx, B., Backeljau, T., De Wachter, R., 1996. Investigation of molluscan
Maslakova, S.A., Martindale, M.Q., Norenburg, J.L., 2004. Fundamental properties of phylogeny on the basis of 18S rRNA sequences. Mol. Biol. Evol. 13, 1306–1317.
the spiralian developmental program are displayed by the basal nemertean Wolf, Y.I., Rogozin, I.B., Koonin, E.V., 2004. Coelomata and Not Ecdysozoa: evidence
Carinoma tremaphoros (Palaeonemertea, Nemertea). Dev. Biol. 267, 342–360. from genome-wide phylogenetic analysis. Genome Res. 14, 29–36.