Assembling the spiralian tree of life

The advent of numerical methods for analysing phylogenetic relationships, along with the study of morphology and molecular data, have driven our understanding of animal relationships for the past three decades. Within the protostome branch of the animal tree of life these data have sufficed to establish two major clades—Ecdysozoa, a clade of animals that all moult, and Spiralia (often called Lophotrochozoa), a clade whose most recent com- mon ancestor had spiral cleavage. In this chapter we outline the current knowledge of protostome relationships and discuss future perspectives and strategies to increase our understanding of rela- tionships within the main spiralian clades. Novel approaches to coding morphological characters are a pressing concern, best dealt with by scoring real observations on species selected as terminals. Methodological issues, such as the treatment of inapplicable characters and the coding of absences, may require novel algorithmic developments. Taxon sampling is another pressing issue, as ter- minals within phyla should include enough spe- cies to represent their span of anatomical disparity. Furthermore, key fossil taxa that can contribute novel character state combinations, such as the so-called ‘stem-group lophotrochozoans’, should not be neglected. In the molecular forum, expressed sequence tag (EST)-based phylogenomics is play- ing an increasingly important role in elucidating animal relationships. Large-scale sequencing has recently exploded for Spiralia, and phylogenomic data are lacking from only a few phyla, including the three most recently discovered animal phyla (Cycliophora, Loricifera, and Micrognathozoa).
While the relationships between many groups now find strong support, others require additional infor- mation to be positioned with confidence. Novel morphological observations and phylogenomic data will be critical to resolving these remaining questions. Recent EST-based analyses underpin a new taxonomic proposal, Kryptrochozoa (the least inclusive clade containing the Brachiopoda and Nemertea).


6.1 Introduction

The protostomes consist of Chaetognatha, a relatively small group of uncertain affinity, and two megadiverse clades—Ecdysozoa and Spiralia (the latter also sometimes referred to as Lophotrochozoa). Spiralia, which includes many kinds of worms, flatworms, molluscs, and related animal groups, comprises a greater number of animal phyla than any other non- overlapping metazoan clade. Specifically, these are Annelida (subsuming several former phyla: Echiura, Pogonophora, Sipuncula, Vestimentifera, and perhaps Myzostomida), Brachiopoda, Bryozoa, Cycliophora, Entoprocta, Gastrotricha, Gnathostomulida, Micrognathozoa, Mollusca, Nemertea, Phoronida, Platyhelminthes, and Rotifera (including the former phylum Acanthocephala) (Giribet, 2002, 2008; Halanych, 2004; Matus et al.,
2006b; Dunn et al., 2008). This amount of phyletic diversity within Spiralia adds up to about half of the traditional extant animal phyla and, in terms of species numbers, includes the second largest phy- lum (Mollusca) as well as the two phyla with some


52

S P IR A L I A N IN T E RRE L A T I ON S HIP S 53



of the largest body plan disparity (Mollusca and Annelida). Although the relationships among these phyla have remained controversial—Ecdysozoa versus Articulata issues aside (e.g. see reviews in Giribet, 2003; Scholtz, 2002)—recent phylogenomic analyses have shed light on the subject. The goals of this chapter are to review the relationships of the spiralian phyla and the techniques for studying such relationships, to establish a current working framework for the main divisions within the clade, and to formalize the supraphyletic classification of Spiralia following current phylogenetic views.


6.1.1 Protostome groups and affinities

Developmental characters such as the fate of the blastopore—which often becomes the adult mouth (see Chapter 4)—and the mode of formation of the mesoderm are typically cited features for support- ing a protostome clade (Nielsen, 2001). Depending on composition, the protostomes have sometimes been considered a paraphyletic assemblage of worm-like animals characterized by the pres- ence of a dorsal (or circumesophageal) brain con- nected to a ventral longitudinal nerve cord, often paired. The proposal that acoels and nemertoder- matids are basal bilaterian animals (Ruiz-Trillo et al., 1999, 2002; Jondelius et al., 2002) rather than Platyhelminthes, renders the traditionally formu- lated Protostomia (Nielsen, 2001) paraphyletic, since the deuterostomes are then closer to the remaining protostomes. Neither nemertodermatids nor acoels have a ventral centralized nerve cord (e.g. Raikova et al., 2004a,b). From this evidence, and the cur- rent phylogenetic framework for metazoans, acoels and nemertodermatids are not considered part of Protostomia in the following discussion, render- ing Protostomia monophyletic given our current understanding of metazoan phylogeny.
Ecdysozoa is currently recognized as monophy- letic in most analyses. Some recent genome-wide analyses have questioned the validity of the clade (Blair et al., 2002; Dopazo et al., 2004; Philip et al.,
2005; Rogozin et al., 2007b; Wolf et al., 2004)—a result that now appears to be due to poor taxon sampling (Philippe et al., 2005a; Dunn et al., 2008). Ecdysozoa is discussed in detail by Telford and colleagues in Chapter 8 of this volume (see also
Telford et al., 2008). The name Spiralia was first coined by Schleip (1929) because of the stereotyp- ical spiral development that occurs only within this clade (Nielsen, 2001; Maslakova et al., 2004a). Spiralia contains all animals with spiral devel- opment. This character, like many others within Metazoa, shows homoplasy, apparently in the form of secondary reduction—but never as convergence outside the clade. This indicates that any animal with spiral development is an unambiguous mem- ber of the clade Spiralia, although the absence of this type of development does not discriminate against its membership.
In one of the first uses of phylogenetic nomen- clature, Lophotrochozoa was defined by Halanych et al., (1995) as a node-based name, defined as the
‘last common ancestor of the three traditional lophophorate taxa, the mollusks, and the annelids, and all of the descendants of that common ances- tor’. This referred to a subgroup of Spiralia, per- haps being synonymous with Trochozoa, the exact scope of which depends on the placement of lopho- phorates. Aguinaldo et al., (1997) later emended Lophotrochozoa by listing a set of taxa that they viewed as being part of this taxon, namely ‘anne- lids, molluscs, rotifers, phoronids, brachiopods, bryozoans, platyhelminths and related phyla’, thus including all or almost all non-ecdysozoan proto- stomes. This delineation approach to a taxon is in conflict with the original phylogenetic definition of Lophotrochozoa and has resulted in considerable confusion in the literature ever since. The clade names Lophotrochozoa and Spiralia have tended to be used as synonyms, based on Aguinaldo et al., (1997), while in other cases the original phylogen- etic definition of Lophotrochozoa is used and it is shown as a subtaxon of Spiralia. We prefer to employ the name Lophotrochozoa in the spirit of how it was originally defined, though Halanych et al., (1995) were not clear as to whether or not they included Entoprocta as part of Bryozoa, as proposed by Nielsen (2001). There is some recent evidence that Bryozoa should include Entoprocta (Hausdorf et al., 2007), in which case this would be moot. One primary reason why neither Lophotrochozoa nor Spiralia has stabilized in usage is that the rela- tionships among the specifying taxa defining Lophotrochozoa (lophophorate taxa, molluscs, and

54 AN I M AL EV O L UTI O N



annelids) have yet to be resolved. This is particu- larly problematic with regard to the lophophorate taxa. The first phylogenomic studies that include lophophorates indicate that Lophotrochozoa is nested within a clade of animals with spiral cleav- age (Helmkampf et al., 2008a,b) or perhaps equiva- lent to Spiralia (Figure 6.1) (Dunn et al., 2008). In the case of synonymy, Lophotrochozoa is the jun- ior synonym, since Spiralia has precedence. We, like many other current authors (e.g. Hausdorf et al., 2007; Henry et al., 2007; von Döhren and Bartolomaeus, 2007; Helmkampf et al., 2008a), therefore use Spiralia rather than Lophotrochozoa, both based on the fact that it has precedence and because it refers to a clear synapomorphy.
Spiralia has been suggested to comprise two putative clades, Platyzoa (Cavalier Smith, 1998) and Trochozoa (Roule, 1891). Trochozoa is pre- ferred (see Rouse 1999; Giribet et al., 2000) over the more recently coined Eutrochozoa (Ghiselin,
1988) used by some authors (Eernisse et al., 1992; Valentine, 2004). Whether or not Platyzoa is a clade (e.g. Zrzavý, 2003; Glenner et al., 2004; Todaro et al.,
2006; Dunn et al., 2008) remains unclear, as many
analyses provide low support for the exact pos- ition of the ‘platyzoan’ phyla, although most tend to group them in a clade or in a grade giving rise to Trochozoa.
The identification by Dunn et al., (2008) of a core set of stable taxa whose relationships are well supported provides a more detailed pic- ture of Platyzoa. The only stable taxon putatively assigned to Platyzoa was Platyhelminthes (Dunn et al., 2008, their Figure 2), which was found to be sister to Trochozoa with strong support in analyses restricted to stable taxa. All other platyzoans were unstable in these analyses (Dunn et al., 2008, their Figure 1 and Supplement), and their position could not be resolved with confidence. All platyzoan taxa investigated to date have relatively long branches, which has led some authors to suspect that support for the group is a systematic error (Telford, 2008). The well-supported position of Platyhelminthes as sister to Trochozoa (which cannot be a result of long branch attraction since Trochozoa does not contain taxa with long branches) may serve as an anchor that is spuriously attracting other long-branch taxa whose placement does not have








Ecdysozoa




Protostomia Platyzoa
Spiralia



Trochozoa


Kryptrochozoa
Deuterostomia Kinorhyncha Loricifera Priapulida Nematoda Nematomorpha Tardigrada Onychophora Arthropoda Chaetognatha Bryozoa Cycliophora Platyhelminthes Gnathostomulida Gastrotricha Micrognathozoa Rotifera Entoprocta Mollusca Annelida Phoronida Brachiopoda Nemertea
















Figure 6.1 Hypothesis of protostome relationships mostly based on recent phylogenomic analysis and morphology.
This tree has not been generated by a consensus or other numerical technique. Protostomes are divided into two sister clades, Ecdysozoa and Spiralia, the latter divided into Platyzoa and Trochozoa. Phyla with dashed lines
lack phylogenomic data and are placed on the tree based mostly on morphology. The thickness of internal branches reflects support. Kryptrochozoa is a new supraphyletic taxon proposed here.

S P IR A L I A N IN T E RRE L A T I ON S HIP S 55



strong signal. The resolution of this problem (including tests of this specific hypothesis) will require a two-pronged strategy of greatly improved taxon sampling within putative platyzoan groups and detailed investigations of the inference of their position designed specifically to identify systematic error.
An unequivocal apomorphy for Platyzoa (Figure 6.2a,b and Plate 2) is hard to delineate mor- phologically. The putative clade contains a series of acoelomate or pseudocoelomate animals—no coelomates belong to this group—some with spe- cial types of jaws formed of cuticularised rods, or Gnathifera (Gnathostomulida, Micrognathozoa, and Rotifera) (Kristensen and Funch, 2000; Sørensen,
2003; see Figure 6.2b). Platyzoa also includes other types of flatworms, including Platyhelminthes (Figure 6.2a) and Gastrotricha (but see Zrzavý,
2003, for an alternative position of gastrotrichs as sister to Ecdysozoa). With the exception of polyclad flatworms, and the parasitic acanthocephalan rotifers, platyzoans are strict direct developers. The possible membership of Cycliophora within Platyzoa remains a contentious issue (see Giribet et al., 2004).
Trochozoa (Figure 6.2d–f; see also 6.4a–d) con- tains those groups with a typical trochophore larva, namely Annelida (Figure 6.2g), Mollusca (Figure 6.2f), and Entoprocta (Figure 6.2e), and also includes some lophophorates. Recently, some nemertean larvae have been interpreted as modi- fied trochophores, with a vestigial prototroch (Maslakova et al., 2004b). Trochophores do not occur outside the clade, but several ingroup members do not develop through a trochophore. Such is the case, for instance, for Brachiopoda and Phoronida (see Figure 6.4a,c). Even within groups where a trochophore is widespread there are clear cases of it being lost (e.g. clitellate annelids and cephalopod molluscs).
The position of the chaetognaths (Figure 6.2h), also known as ‘arrow worms’, has been debated for a long time, but recent studies based on morph- ology (Harzsch and Müller 2007) and phylogenom- ics (Marlétaz et al., 2006; Matus et al., 2006b; Dunn et al. 2008) strongly suggest that they are early diverging relatives of protostome taxa despite having deuterostome-like development. Beyond
this, though, there is little clarity as to their exact position. Depending on character and taxon sam- pling, they have been placed as sister to or within Spiralia (Matus et al., 2006b; Dunn et al., 2008, Helmkampf et al., 2008b), within Ecdysozoa (Matus et al., 2006b; Helmkampf et al., 2008a,b), or as sister to Spiralia + Ecdysozoa (Marlétaz et al., 2006; Matus et al., 2006b; Dunn et al., 2008). The nervous system of chaetognaths has recently been found to be similar to that of other protostomes in having a typical circumoral arrangement of the anterior CNS (Harzsch and Müller, 2007). Resolving the placement of Chaetognatha is critical to the recon- struction of some of the most basic developmental characters, including cleavage mode and the fate of the blastopore.
Cycliophoran affinities with entoprocts
(Figure 6.2e) are still under consideration (Giribet
et al., 2004), and one of the cycliophoran larval
forms, the chordoid larva, has been interpreted as
a modified trochophore (Funch, 1996). Trochozoa
includes, in general, larger protostomes than
Platyzoa. Many trochozoans are coelomates with
large body cavities and metanephridia-based
excretory organs in the adult phase. Some, how-
ever, are functionally acoelomate, with protone-
phridial excretory systems as adults (Nemertea), or
acoelomates, and therefore with protonephridia as
excretory organs.


6.1.2 Problematica

One of the most problematic protostome groups, not just in terms of its phylogenetic placement, is the symbiotic group Myzostomida (Eeckhaut and Lanterbecq, 2005). From its original description, Myzostoma cirriferum (Figure 6.2c) was considered a trematode (Leuckart, 1827). Since then, myzos- tomids have been associated with crustaceans, tardigrades, pentastomids, and with polychaete annelids. Jägersten (1940) grouped Myzostomida and Annelida (as two separate classes) into a protostome group called Chaetophora. Based on the shared unusual ultrastructure of the sperm in myzostomes and acanthocephalan rotifers, with a pulling (instead of pushing) flagellum, both groups were classified into the phylum Procoelomata (Mattei and Marchand, 1987). However, due to the


Figure 6.2 Examples of some spiralian taxa: (a), (b) Platyzoa; (c), (d), (h) uncertain; (e)–(g) Trochozoa. (a) The free-living platyhelminth Hoploplana californica. (b) An undescribed species of seisonid rotifer Paraseison taken from its crustacean host Nebalia. (c) A myzostomid Myzostoma cirriferum taken from its crinoid host. (d) Several zooids of a bryozoan colony. (e) Anterior end of entoproct Pedicellina sp.
(f) Dorsal view of the sacoglossan mollusc Thuridella picta. (g) A syllid polychaete annelid brooding embryos on its dorsum. (h) A benthic spadellid chaetognath, Spadella, All photographs by G. W. Rouse. (See also Plate 2.)

S P IR A L I A N IN T E RRE L A T I ON S HIP S 57



presence of segmentation, parapodia-like structures with chaetae and aciculae, and an apparent trocho- phore larva, many modern authors consider myzos- tomes as polychaete annelids (e.g. Brusca and Brusca, 2003; Nielsen, 2001; Rouse and Fauchald,
1997; Rouse and Pleijel, 2001; but see Haszprunar,
1996b). Molecular phylogenetic studies of myzosto-
mids have suggested rather contradictory hypoth-
eses, with some analyses suggesting a relationship
to rotifers (including acanthocephalans) and cyclio-
phorans (Zrzavý et al., 2001)—a somehow expanded
‘Procoelomata’ of Mattei and Marchand (1987)—or
as closer to Platyhelminthes than any other spiral-
ians (Eeckhaut et al., 2000). Both cases presuppose
a platyzoan affinity of myzostomes, as opposed to
their more traditional trochozoan kinship. However,
a recent molecular analysis using mitochondrial
genome data and multiple nuclear genes suggests
that myzostomes should be placed back among the
annelids (Bleidorn et al., 2007). This debate is prob-
ably not yet settled, as EST data do not currently
support an annelid affinity for myzostomids (Dunn
et al., 2008).
The position of several other spiralian clades
has remained elusive even in recent phylog-
enomic analyses using large numbers of genes
(Dunn et al., 2008). Such is the case for bryozoans,
entoprocts, gastrotrichs, gnathostomulids, and
rotifers. Despite most of these taxa forming part
of Platyzoa, their position was unstable across
analytical methods or model selection, and nodal
support for their exact position was not conclusive.
Results of those analyses also differed with respect
to the positions of bryozoans (Figure 6.2d) and
entoprocts (Figure 6.2e) compared with another
recent phylogenomic analysis (Hausdorf et al.,
2007), which suggested a sister-group relationship
between these two phyla, although with a much
smaller taxon sampling. A reason for the instabil-
ity found in Dunn et al., (2008), at least for some
of these phyla, was a poor EST representation due
to shallow library examination (e.g. only c. 1000
clones sequenced, as in the case of myzostomids)
or poor library quality (as in the case of ento-
procts). This leaves a clear strategy for improving
our understanding of the relationships of these
phyla—sequencing additional clones or develop-
ing additional libraries.
6.1.3 Working framework for spiralian relationships

All the previous work, and especially the recent resolution obtained based on EST data for broad taxon sampling, has contributed to a roadmap for resolving metazoan relationships (e.g. com- pare the resolution of the spiralian trees in the recent reviews by Giribet et al., 2007, or Giribet,
2008, with the phylogenomic results of Dunn et al.,
2008). Focusing on the spiralians—the topic of this
chapter—previous work has delimited two puta-
tive main clades and their core membership. The
affinities of bryozoans, chaetognaths, cycliopho-
rans, and myzostomids may still be debatable,
but it seems that the remaining protostome phyla
(following the definition provided above) either
belong to Platyzoa (Gastrotricha, Gnathostomulida,
Platyhelminthes, Rotifera) or Trochozoa (Annelida,
Brachiopoda, Entoprocta, Mollusca, Nemertea,
Phoronida) (Figures 6.1, 6.2, and 6.4). Interestingly,
only Trochozoa includes members with chaetae
(Annelida, Brachiopoda) (Hausen, 2005; Lüter, 2000).
Other chaeta-like structures in chitons (Leise and
Cloney, 1982), juvenile octopods (Brocco et al., 1974),
and the gizzard-teeth of a bryozoan (Gordon, 1975)
are often considered to be convergent (Hausen,
2005), but could instead be plesiomorphic for tro-
chozoans. Resolution of the placement of myzosto-
mids, with their unequivocal chaetae (Lanterbecq
et al., 2008), may provide further clarification as to
how this feature has evolved.
Relationships among the phyla that constitute
Platyzoa are not well established (e.g. Giribet et al.,
2004; Dunn et al., 2008), although good apomorphies
exist for one of its subclades, Gnathifera (Kristensen
and Funch, 2000; Sørensen 2001, 2003). The internal
phylogenies of several of these phyla are well
understood as total evidence and multilocus ana-
lyses have been published for Gnathostomulida
(Sørensen et al., 2006), Platyhelminthes (Figure 6.2a)
(Littlewood et al., 1999), Rotifera (Figure 6.2b)
(Sørensen and Giribet, 2006), and Gastrotricha
(Zrzavý, 2003).
Many trochozoan internal relationships have been
recently resolved with strong support. Monophyly
of annelids (Figure 6.2g) and its membership have
been corroborated only in recent phylogenomic

58 AN I M AL EV O L UTI O N



analyses (Hausdorf et al., 2007; Dunn et al., 2008). There now seems to be a consensus about the affinities of the former phyla Echiura, Sipuncula, and Pogonophora/Vestimentifera as highly modi- fied annelid subtaxa (McHugh, 1997; Hessling and Westheide, 2002; Hausdorf et al., 2007; Struck et al., 2007; Dunn et al., 2008). In spite of this pro- gress, resolution of relationships within Annelida, especially as to the placement of the ‘root’ of the annelid tree, is far from agreed upon. Molecular analyses (e.g. Bleidorn et al., 2003; Rousset et al.,
2004, 2007; Colgan et al., 2006; Struck et al., 2007) radically contrast with the most comprehensive analyses of annelid relationships based on morph- ology (Rouse and Fauchald, 1995, 1997; Rouse and Pleijel, 2001; see a recent review in Rouse and Pleijel, 2007). Without doubt, an important leap is needed in the number of data to be incorporated into annelid phylogenetic studies.
Internal relationships of the other large trocho- zoan phylum, Mollusca (Figure 6.2f), do not present a much brighter picture. Relationships based on morphology (e.g. Salvini-Plawen and Steiner, 1996; Haszprunar, 2000) and molecules (e.g. Passamaneck et al., 2004; Giribet et al., 2006) are still at odds for relationships within this group, and molecular analyses have traditionally had trouble recover- ing molluscan monophyly until the incorporation of phylogenomic data (Hausdorf et al., 2007; Dunn et al., 2008). Only a recent multilocus analysis of molluscan relationships was able to recover mono- phyly, although with low clade support (Giribet et al., 2006), and the internal relationships among the molluscan classes were for the most part not resolved.
Internal sipunculan phylogeny has been add- ressed in recent times, based on both morphology and molecular analyses (e.g. Maxmen et al., 2003; Schulze et al., 2005, 2007), and it is now becoming clear that Sipuncula is affiliated with Annelida (Struck et al., 2007; Dunn et al., 2008), though their closest annelid relatives have yet to be established. The phylogeny of Brachiopoda has also been assessed in a range of analyses (Carlson, 1995; Cohen et al., 1998; Cohen, 2000; Cohen and Weydmann,
2005; the latter two also including several phoro- nid species), and some authors have proposed that Phoronida may be a subgroup of Brachiopoda
(Cohen, 2000; Cohen and Weydmann, 2005; but see Bourlat et al., 2008; Helmkampf et al., 2008a). Although several higher-level morphological and molecular analyses exist within the phylum Nemertea (e.g. Sundberg et al., 2001; Thollesson and Norenburg,
2003), results still depend on few markers and are not integrated with morphological analyses. Little synthetic phylogenetic work has been published on the phylogeny of entoprocts or bryozoans.


6.1.4 Controversial fossils: molluscs, annelids, brachiopods, or
stem-group Spiralia?

The Palaeozoic fossil record is rich in protostome taxa (Budd and Jensen, 2000; Valentine, 2004), and in general it is thought that most metazoan phyla were already present in the Cambrian. Recent dis- coveries of Lower Cambrian sipunculans (Huang et al., 2004) and chaetognaths (Szaniawski, 2005; Vannier et al., 2007) reduce the number of animal phyla missing from the Palaeozoic record. The most conspicuous phylogenetic gap in the Palaeozoic record is for Platyzoa (Figure 6.1), no members of which have yet been found (Giribet, 2008).
Although considerable advances have recently been made in understanding the morphology and phylogenetic context of several potentially pivotal spiralian fossils, disagreement about the interpret- ation of structures relative to extant phyla have left the picture clouded. Especially relevant fossils are the sclerotome-bearing Wiwaxia (Butterfield,
1990; Conway Morris, 1985; Eibye-Jacobsen, 2004) (Figure 6.3a and Plate 3), Orthrozanclus (Conway Morris and Caron, 2007) (Figure 6.3c), and Halkieria (Conway Morris and Peel, 1995; Vinther and Nielsen, 2005) (Figure 6.3b), and the unarmoured Odontogriphus (Caron et al., 2006) (Figure 6.3e). Much of the controversy about these fossils—which are at least uniformly recognized as spiralians but then variably assigned to either Annelida, Mollusca, or Brachiopoda—is encapsulated in a debate over whether a clearly homologous feeding apparatus in Odontogriphus and Wiwaxia is (Scheltema et al.,
2003; Caron et al., 2006, 2007), or is not (Butterfield,
2006, 2008), a radula, and whether these animals
are (Butterfield, 2006), or are not (Eibye-Jacobsen,
2004; Caron et al., 2007), segmented. r

Figure 6.3 Exceptionally preserved Palaeozoic spiralian fossils. (a) Wiwaxia corrugata (Middle Cambrian, photo courtesy of Jean-Bernard Caron). (b) Halkieria evangelista, sclerites (sc) anterior shell (as) and posterior shell (ps) (Lower Cambrian, photo
courtesy of Jakob Vinther). (c) Orthrozanclus reburrus, anterior shell (as), sclerites (sc) (Middle Cambrian, photo courtesy of Jean-Bernard Caron). (d) Acaenoplax hayae, dorsal shell plates (dsv), spines (sp) (Silurian, digital reconstruction courtesy of Mark Sutton).
(e) Odontogriphus omalus, radula (r) and ctenidia (ct) (Middle Cambrian, photo courtesy of Jean-Bernard Caron). (See also Plate 3.)




Wiwaxia and Halkieria have long been associated, based on similar sclerite morphology and left– right sclerite zones (Bengston and Conway Morris,
1984), and some recent studies unite them (with Orthrozanclus and several other fossil ‘coelosclerito- phorans’) in a putative clade, Halwaxiida (Conway Morris and Caron, 2007). Explicit cladistic analyses have resolved the halwaxiids as a clade (Conway Morris and Caron, 2007; Sigwart and Sutton, 2007) or a grade (Vinther et al., 2008) in the mollusc stem group. Mollusc affinities for Halkieria have been
advanced based on similarities to chitons in par- ticular (Vinther and Nielsen, 2005). Arguments for a halkieriid origin of brachiopods (Holmer et al.,
2002) have been weakened by the discovery that the supposed intermediate tannuolinids are not in fact scleritome-bearing but are sessile, bivalved organisms with brachiopod-like ultrastructure (Holmer et al., 2008).
Halwaxiid monophyly is disputed by Butterfield (2006), largely on the basis of Wiwaxia sharing putative autapomorphies of Annelida, especially

60 AN I M AL EV O L UTI O N



microvillar setae that are histologically identical to annelid chaetae, and specifically similar to the flattened notochaetae of chrysopetalid polycha- etes (Butterfield, 1990). Others have accepted the homology of these chaetae (as do we, as a primary homology statement), but excluded Wiwaxia from Annelida based on its lack of parapodia and seg- mentation (Eibye-Jacobsen, 2004). Butterfield (2006) attempted to retain Wiwaxia in Annelida by arguing that a dorsal chaetal scleritome is sufficient to iden- tify Wiwaxia as an annelid. We are unconvinced, because although the relevant chaetae of the fossil polychaetes (Canadia) that Butterfield (2006) refers to are notochaetal and hence dorsal, Wiwaxia has no evidence for parapodia or segmentation whatso- ever. Butterfield’s (2006) evocation of a special style of creeping that transforms segmentation/parapo- dia/neuropodia in Wiwaxia beyond recognition is decidedly ad hoc. Likewise, assuming that Wiwaxia is segmented because Odontogriphus is supposed to be segmented (Butterfield, 2006), despite a lack of convincing evidence from the fossils (Caron et al.,
2007), is unnecessary. The structural similarity of Wiwaxia and halkieriid sclerites (Bengston, 2006) and their sclerite zones (Conway Morris and Peel,
1995) is not so easily dismissed, and relegating their similarity to convergence or inheritance from the spiralian stem lineage is uncompelling.
A recurring theme in the spiralian fossil record is the variable assignment of certain fossils to either Annelida or Mollusca (indeed these character con- junctions in fossils are an argument in favour of spiralian monophyly). For example, the Silurian Acaeonoplax (Figure 6.3d) has generally been accepted as an aplacophoran mollusc (Sutton et al.,
2001b, 2004; Sigwart and Sutton, 2007), an interpret- ation with which we concur, not least based on its calcareous spicules and seriated shell plates. Note that the mollusc interpretation had been challenged by Steiner and Salvini-Plawen (2001) who argue for annelid features being present in the fossils. The discovery of soft-part preservation associated with the typical shell plates permits a confident assign- ment of the long-problematic Ordovician–Permian Machaeridia to Annelida (Vinther et al., 2008), though this finding contradicts the recent place- ment of machaeridians within Mollusca (Sigwart and Sutton, 2007). Machaeridians throw up an
unexpected character combination in Spiralia— calcareous shell plates (with marginal growth) in animals that have parapodial chaetae. Such unique character combinations underscore the utility of including fossil terminals in cladistic analyses of morphology. All of the fossil spiralians discussed above are coded as terminals in our matrix for the Assembling the Protostome Tree of Life project (authors’ work in progress).
In the primary reference phylogeny of Dunn et al., (2008, their Figure 2) a clade labelled Clade C was proposed, including a set of spiralian animals primitively with a trochophore larva. Clade C was not considered a synonym of Trochozoa because the most current phylogenetic hypotheses (Hausdorf et al., 2007; Dunn et al., 2008, their Figure 2) did not include Entoprocta, or Entoprocta was considered the sister group to Clade C (Dunn et al., 2008, their Figure 1). Entoprocta have spiral development and a trochophore larva, and therefore, until its exact position is resolved, we prefer not to name Clade C formally. The monophyly of Clade C is consistent with a homology between chaetae in annelids and brachiopods and spicules in molluscs. Both kinds of structure are epidermal extracellular formations whose secretory cells develop into a cup or a fol- licle with microvilli at their base. This homology has been anticipated by palaeontologists (Conway Morris and Peel, 1995, their Figure 50) who have proposed a common origin of mollusc spicules and annelid/brachiopod chaetae as modifications of sclerites as developed in the scleritome of various Cambrian fossil taxa that have subsequently been assigned to the Halwaxiida (Conway Morris and Caron, 2007). The hollow sclerites of the Cambrian fossils (sharing a suite of morphological details encompassed under the ‘coelosclerite’ concept of Bengston 2006) are variably organic (e.g. Wiwaxia) or aragonitic (e.g. Halkieria). The character delimi- tation of a mollusc/halwaxiid sclerite applied by Vinther et al., (2008), i.e. an ectodermal elem- ent secreted to a finite size by a basal epithelium, applies to chaetae as well.


6.2 Novel approaches in morphology

Morphology and development, including early
cleavage patterns, have played fundamental roles

S P IR A L I A N IN T E RRE L A T I ON S HIP S 61



in shaping our understanding of animal relation- ships. A major shift within the past two decades towards using molecular evidence has resulted in some major rearrangements in the tree of ani- mal life. Morphological analysis still has its role in continuing to decipher animal relationships and in interpreting the results derived from molecular studies. Animal morphological analyses under- went a first revolution with the advent of cladistic techniques, which led to the proposal of numerous phylogenetic hypotheses based on numerical (par- simony-based or maximum-likelihood) analyses of explicit data matrices (e.g. Eernisse et al., 1992; Nielsen et al., 1996; Zrzavý et al., 1998; Sørensen et al., 2000; Peterson and Eernisse, 2001; Jenner and Scholtz, 2005). Morphological matrices are not exempt from arbitrary decisions in inclusion (or exclusion) of characters, along with their definition and the identification of character states. Common problems with these previous approaches to meta- zoan morphological data matrices are the uncrit- ical recycling of characters (see Jenner, 2001) and the assignment of homology to absences of a given character state (Jenner, 2002) (see Jenner, 2004a, for a general discussion). Another major problem is the decision of what character state is assigned to higher (supraspecific) taxa, as all metazoan mor- phological phylogenies published so far rely on coding. This could lead to the arbitrary choice of character states—often in a hypothesis-driven manner (Jenner, 2001). An alternative to some of these problems is to code real observations for a selected number of species instead of supraspecific taxa (Yeates, 1995; Prendini, 2001) in the same fash- ion that species are used for molecular analyses. This solution is not only appealing from an oper- ational perspective, but also philosophically, since it allows for a stricter test of monophyly than pre- vious strategies.
No metazoan-wide morphological matrix has yet been produced using species as terminals, although we are currently working on such a matrix (G. Edgecombe et al., work in progress). This strategy is not without difficulties. It requires two principal conditions: incorporating multiple species per phylum—ideally a collection of species that represent the phyletic morphological disparity and requiring careful species choice; and coding
observations for many species. Certain characters, especially those of development and ultrastructure are unobserved (or not described) for many termi- nals and filling the matrix requires a considerable amount of work. Filling all cells with observations for a metazoan matrix of hundreds of taxa and hundreds of characters requires substantial effort by a much larger group of researchers than the team assembled. It is therefore necessary to perfect collaborative software that allows data matrices to be updated by a group of authors over the web (see, for example, http://morphobank.geongrid.org/ or http://www.mesquiteproject.org/) and requires what we call ‘coding parties’, where experts in taxa or characters meet periodically to discuss charac- ters, character states, and specific coding of taxa.
Defining the characters and their states remains difficult (e.g. Jenner, 2004a,b). Hence some research- ers believe that such efforts should be done expli- citly and discussed by the scientific community in large, developing specific ontologies (e.g. Ramírez et al., 2007; see also https://www.morphdbase.de/). For example, characters that have played fundamen- tal roles in shaping animal relationships through time, such as the fate of the blastopore, segmenta- tion, or the origins of body cavities, are still poorly understood and their states need further research and discussion. Problems with assigning a hom- ologous state to the lack of a feature have received little attention in the literature. The same is true for inapplicable characters and their specific treatment by computer algorithms (e.g. Pleijel, 1995; Lee and Bryant, 1999; Strong and Lipscomb, 1999).
One possible solution to these issues is the add- ition of as many states as there are non-homologous absences, although this would incur the necessity of adding complex Sankoff characters with more than the 10 states allowed by some software imple- mentations. Another is to have characters treated in a hierarchical and integrative way, such that less inclusive characters are not considered in an analysis until the more general characters that encompass them are. These less inclusive charac- ters are only applied in parts of the tree that are applicable. Another possibility would be the use of a dynamic approach to morphology (Schulmeister and Wheeler, 2004; Ramírez, 2007), analogous to the direct optimization of molecular characters.

62 AN I M AL EV O L UTI O N



Although this option is still in its infancy, it could deal with the problem of absences and inapplicable character states in a completely different way.
The impact of fossils in numerical analyses of morphology (e.g. Cobbett et al., 2007), or when combined with molecules (e.g. Wheeler et al., 2004) should not be underestimated when attempting to reconstruct spiralian relationships, especially due to the presence of so many fossils that show intermediate morphologies of extant animals (see above). It is therefore imperative to incorporate data from key, exceptionally preserved fossils into these data matrices if we truly want to understand the evolution of spiralian animals, even though not all clades are equally represented in the fossil record.


6.3 Phylogenomic approaches

Since the publication of the earliest metazoan ana- lyses based on molecular data (Field et al., 1988; Lake, 1990), molecular phylogenetics has revolution- ized our understanding of metazoan relationships. Novel concepts, now widely accepted by the com- munity, such as Ecdysozoa (Aguinaldo et al., 1997), Spiralia (Halanych et al., 1995), and the more con- troversial Platyzoa (Giribet et al., 2000) are rooted on molecular analyses of ribosomal RNA sequence data. The specific position of many ‘odd’ taxa have also benefited from molecular techniques. Salient examples are those of acoels (Ruiz-Trillo et al., 1999), nemertodermatids (Jondelius et al. 2002), xenotur- bellids (Bourlat et al., 2003), nemerteans (Turbeville et al., 1992), and gnathostomulids (Giribet et al.,
2000). However, the positions of other phyla such as loriciferans, micrognathozoans, or cycliopho- rans are not well resolved in molecular analyses (Giribet et al., 2004; Park et al., 2006). Most of these studies were based on one or at most a few targeted molecular markers.
New strategies that consider molecular data from many genes rather than just a few have emerged within the past few years, and are col- lectively designated with the catch-all label of
‘phylogenomics’ (Delsuc et al., 2005). The first such analyses culled a subset of widespread genes from the few complete eukaryotic genomes to tackle metazoan relationships and test hypotheses such
as Ecdysozoa or Coelomata (Blair et al., 2002;
Dopazo et al., 2004; Philip et al., 2005; Wolf et al.,
2004). These studies, although broad in the num-
ber of genetic data included, suffer from one of the
most crucial phylogenetic biases—deficient taxon
sampling. Not surprisingly, analyses including
a broader taxon sampling followed immediately,
focusing not only on whole-genome approaches,
but taking advantage of EST projects (Philippe and
Telford, 2006). Such studies have provided insights
into the relationships of several protostome phyla
(Hausdorf et al., 2007; Roeding et al., 2007; Dunn
et al., 2008), allowing resolution of long-standing
questions such as the sister-group relationships
of Arthropoda–Onychophora (Roeding et al., 2007;
Dunn et al., 2008), or the overall topology among
the trochozoan phyla (Dunn et al., 2008) (see
Figure 6.1).
Several current research groups, especially those
funded under the US National Science Foundation
AToL and the German Deep Phylogeny pro-
grammes have focused on closing the gap in
missing protostome diversity using targeted EST
studies (see Hausdorf et al., 2007; Roeding et al.,
2007; Dunn et al., 2008). These novel data already
encompass nearly all animal phyla, and just a few
phyla of small-sized animals are missing (cur-
rently there are EST/genomic data missing for
Cycliophora, Loricifera, Micrognathozoa, and
Nemertodermatida). Continued improvements in
sequencing technologies and computational tools
will soon make it possible for far more taxa within
critical groups to be incorporated, and will make
phylogenomic approaches more cost-effective than
traditional directed-PCR approaches for a greater
number of problems. Phylogenomic approaches
will also become truly genomic with the final tran-
sition to systematics labs sequencing complete
genomes on a routine basis (ESTs are just a stop-
gap until that time).


6.4 Conclusions: the future of spiralian phylogeny

Much recent progress has been made on the phyl- ogeny of Spiralia. There is now strong support for the existence and many internal relationships of Trochozoa, though its complete composition

S P IR A L I A N IN T E RRE L A T I ON S HIP S 63



remains uncertain due to persistent problems placing Bryozoa and Entoprocta (Dunn et al., 2008). Multiple sources of evidence place the remaining spiralian taxa in the clade Platyzoa, though there are lingering questions as to whether or not support for this clade is due to systematic error. Resolving these two questions will be important priorities for moving forward. However, key questions such as the affinities of bryozoans and cycliophorans, or the exact position of gastrotrichs, myzostomids, and micrognathozoans—to mention just a few— lack a convincing answer. Two qualitative changes taking place in the study of animal relationships may contribute towards an even more resolved pic- ture of the spiralian tree. First is the fresh study of animal morphology and development, translated into a data matrix where observations (instead of inferences) and species (instead of supraspecific taxa)—including fossils—are coded. Second is the widespread use of phylogenomic techniques, now beginning to span a much greater swath of spiralian diversity. Once hundreds (or thousands) of genes become available for a wide sampling of protostome species, relationships may finally be established with great support. We will then be
able to proceed to the even more fundamental task of attempting to explain the origins of morpho- logical disparity and taxonomic diversity.

6.5 A new taxonomic proposal

Given recent progress in resolving several nodes of the spiralian tree of life, we formalize a taxonomic proposal derived from the analyses recently pub- lished by Dunn et al. (2008):

Kryptrochozoa Dunn, Edgecombe, Giribet, Hejnol,
Martindale, Rouse new taxon.
• Definition: the least inclusive clade containing the Brachiopoda and Nemertea (Figure 6.4 and Plate 4).
• Intention of the name: this name is intended to refer to a clade comprised of Nemertea and Brachiopoda.
• Etymology: compounding the Greek kryptos (hidden) and Trochozoa with reference to the modi- fication of the trochophore larvae (Figure 6.4).
• Reference phylogeny: in the primary reference phylogeny (Dunn et al., 2008; Figure 6.2) this clade was labelled Clade A.




(a)
(b)










200 mm 80 mm


(c) (d)
Figure 6.4 Examples of taxa and larval forms in Kryptotrochozoa, a new subtaxon of Trochozoa. (a) An actinotroch larva of an
unidentified phoronid species. (b) Fluorescently labelled pilidium larva of the nemertean Cerebatulus lacteus (photograph by Patricia
Lee and Dave Matus). (c) Anterior end of phoronid brachiopod Phoronis hippocrepia (photograph by G. W. Rouse). (d) Dorsal view of the nemertean Micrura sp. (photograph by G. W. Rouse). (See also Plate 4.)

64 AN I M AL EV O L UTI O N



• Discussion: the clade includes two phyla (note Phoronida is accepted here as part of Brachiopoda, following Cohen, 2000, and Cohen and Weydmann,
2005; though see Helmkampf et al., 2008a, and Bourlat et al., 2008) that have modified trocho- phores, in the case of some heteronemerteans (Maslakova et al., 2004a,b), or larvae that do not show evident homologies to trochophore larvae in other nemerteans and brachiopods (including pho- ronids). Examples of these are the pilidium larva of some derived Nemertea (Figure 6.4b) and the acti- notroch larvae found in Phoronida (Figure 6.4a).
• Remarks: this clade was proposed in the phyl- ogenomic analysis of Dunn et al. (2008). This clade has never been proposed previously, though it has since been recovered in other molecular ana- lyses (Bourlat et al., 2008; Helmkampf et al., 2008b). There are no obvious morphological apomorphies. Bootstrap support for Kryptrochozoa was much lower for independent analyses of non-ribosomal (14%) and ribosomal genes (15%) than it was in the combined 150-gene analyses (Dunn et al., 2008, their Supplementary Figure 10a). This may help explain why it has not been recovered in previous analyses; support requires that many genes be ana- lysed in combination, and Dunn et al. (2008) is the only phylogenomic analysis to date to include the relevant taxa.
6.6 Acknowledgements

We wish to acknowledge Tim Littlewood and Max Telford for organizing the Linnean Tercentenary Symposium on Animal Evolution at the Royal Society. The Royal Society and the Novartis Foundation made the symposium possible. Ron Jenner, an anonymous reviewer, and the editors provided insightful comments that improved earl- ier versions of this article. Errors and biases remain our responsibility. All the collaborators on the protostome AToL project (Jessica Baker, Noemí Guil, Reinhardt Møbjerg Kristensen, Dave Matus, Akiko Okusu, Joey Pakes, Elaine Seaver, Martin Sørensen, Ward Wheeler, Katrine Worsaae), without whom the research could have not been conducted, provided stimulating discussions that have led to the writing of this chapter. Illustrations of fossils were kindly provided by Jean-Bernard Caron, Mark Sutton, and Jakob Vinther, and thanks to Patricia Lee and Dave Matus for the labelled pilidium image. This mater- ial is based upon work supported by the National Science Foundation AToL programme under grant nos 0334932, 0531757, 0531558. GG was recipient of a fellowship from the Ministerio de Educación y Ciencia (Spain) for a sabbatical stay at the Centre d’Estudis Avançats de Blanes during the writing of this chapter.