Invertebrate Problematica: kinds, causes, and solutions

11.1 Progress and remaining controversy

The field of high-level metazoan phylogenetics is moving extremely fast. Estimates of a consensus phylogeny for the Metazoa continue to change, particularly as ever-larger data sets begin to accumulate. Notable among the newer studies are phylogenomic analyses (Hausdorf et al., 2007; Roeding et al., 2007; Brinkmann and Philippe,
2008; Dunn et al., 2008; Helmkampf et al., 2008a,b;
Lartillot and Philippe, 2008; Struck and Fisse,
2008), the results of which variously strengthen
previous points of consensus (e.g. the dichotomy
of Ecdysozoa and Lophotrochozoa), introduce
new points of controversy (e.g. Ctenophora as
sister group to all other metazoans), and leave
other phylogenetic problems unresolved (e.g. the
phylogenetic position of Ectoprocta). Through the
application of increasingly sophisticated models
of evolution to unparalleled quantities of data for
larger numbers of taxa, these analyses underscore
the value of the guidelines summarized in Jenner
and Littlewood (2008) for continuing progress
in our understanding of metazoan phylogeny.
Nevertheless, as we discuss below, these increas-
ingly comprehensive phylogenetic studies should
not be uncritically accepted as being free from
underlying flaws. Whereas phylogenetic analyses
of relatively small data sets were chiefly marred
by stochastic or sampling errors, analyses of lar-
ger data sets are subject to increasingly serious
interpretational difficulties as systematic errors
become visible.
Equally notable are new studies describing the detailed morphology or development of living and extinct taxa such as those by Maas et al. (2007) and Stach et al. (2008). Such studies shed light on steps involved in the evolution of body plans, and add- itionally provide new and independent evidence with which to evaluate molecular estimates of phylogenetic relationships. Stach et al.’s (2008) cell lineage analysis of the appendicularian Oikopleura dioica, for example, adds significantly to the debate about the phylogenetic position of appendicular- ians, which even with the addition of genomic information are labelled as an unstable rogue taxon (Brinkmann and Philippe, 2008).
Finally, synoptic perspectives, in which diverse sources of evidence have been compiled and syn- thesized, offer the most recent attempts to recon- struct the details of the evolution of animal body plans within the framework of the latest phylog- enies (see, for example, Schmidt-Rhaesa, 2007, Sperling et al., 2007, and Nielsen, 2008). The trees themselves are merely the first necessary step in our quest to understand metazoan evolution.
This chapter is modified from Jenner and Littlewood (2008), and although the general argu- ments of that paper are summarized here, we adopt a more taxon-focused perspective. We examine recent progress in high-level animal phylogen- etics with specific attention to the invertebrate Problematica, i.e. those taxa that are particularly difficult to position in the animal tree of life.
In recent years great strides have been made in solving the phylogenetic positions of several classical Problematica, such as Xenoturbella bocki


107

108 AN I M AL EV O L UTI O N



and Buddenbrockia plumatellae (Bourlat et al., 2006; Jiménez-Guri et al., 2007), principally by means of molecular phylogenetic analyses. However, new studies have also identified unexpected Problematica of a new kind, such as Acoela or Ctenophora (Brinkmann and Philippe, 2008; Dunn et al., 2008). Classical Problematica were problematic chiefly as a result of the lack of data (Haszprunar et al., 1991). In contrast, the new Problematica are problematic despite, or as a result of, the accumula- tion of large molecular data sets. Either phylogen- etic methods are not able to deal with systematic errors inherent in large data sets, leading to rogue taxa that are very difficult to place (Acoela), or the large amounts of new data suggest a phylogen- etic position that is unprecedented (Ctenophora), and which necessitates a fundamental rethinking of body plan evolution. Strikingly, as Figure 11.1 shows, roughly half of the ‘phylum-level’ taxa in the Metazoa can be labelled as Problematica on the basis of current evidence.
We review the methodological and biological causes of Problematica in the context of high-level metazoan phylogeny, and provide possible strat- egies for dealing with them. We discuss fossil and extant Problematica from the perspectives of mor- phological and molecular phylogenetics. A sum- mary of attempts to grapple with Problematica provides insights into the relative abilities of dif- ferent kinds of data and phylogenetic methods to deal with some of the most challenging problems in all of systematics.


11.2 Problematica—causes and recognition criteria

Problematica confront phylogeneticists with all the problems that can beset phylogenetic ana- lysis. Problematica arise when we lack unam- biguous phylogenetic signals that can relate them to other taxa. In many cases, such as the classical Problematica (Haszprunar et al., 1991), this was simply the result of not (yet) having enough know- ledge of a taxon. This is also the case for many fossil Problematica with unfavourable preserva- tion. However, as large phylogenomic data sets become increasingly common, it has become clear
that even a large amount of data are no automatic solution to resolving interrelationships. In certain cases, the wealth of data can even be the cause of new problems, as phylogenetic methods fall vic- tim to systematic errors that were undetectable in smaller data sets; here the source of the problem is in estimating interrelationships that leaves taxa in ambiguous positions.
We distinguish three main categories of reason for either the absence of sufficient phylogenetic signal or its obfuscation by other signals: (1) not enough phylogenetic signal has evolved; (2) the phylogenetic signal is lost through extinction; (3) the phylogenetic signal is lost or obscured by evolution of a non-phylogenetic signal.
In the first category, if lineage splitting events suc- ceed each other rapidly, there may not be enough time for distinctive features to evolve that can be used to group descendant species. Although the length of the fuse of the Cambrian explosion is still debated, this has long been considered a distinct possibility for the divergence of the animal phyla.
In the second category, extinction may exacer- bate the problem of inferring clades on the basis of homoplasy, or erase phylogenetic signal altogether if the organisms are not discovered. For example, reconstruction of the panarthropod stem group revealed that the subventral mouth shared by extant arthropods and onychophorans has evolved convergently (Eriksson and Budd, 2000). As is well known, fossils can contribute important phylogen- etic signal (Cobbett et al., 2007), and in view of the considerable differences between the body plans of extant phyla, extinction must have removed sub- stantial amounts of morphological phylogenetic signal that can only be retrieved by the study of fossils.
The third category groups several causes related to evolutionary change that can erode or obscure phylogenetic signal with the same effects for phylogenetic analysis as extinction of taxa, even when all relevant taxa are included in the ana- lysis. This is especially important when inferring phylogenies with short stems and long terminal branches (Rokas and Carroll, 2006), features com- mon to estimates of metazoan phylogeny. Firstly, if newly evolved lineages have not yet evolved

P R OB LE MA TI C A 109




NCBI - GenBank
Genomes


No.

nucleotides

prots mt

species core ESTs GSS

Choanoflagellata
140
14152
106036
60757
18477
1
Calcarea 1000 787 0 86 0
Demospongiae 9000 2492 8 1702 18 1
Hexactinellida 10000 71 0 0 0 104 1
Homoscleromorpha ~30 32 11176 0 64 2
Cnidaria 9000 176992 437394 2 56258 29 1 1
Polypodium hydriforme 1 5 0 0 0 0
Myxozoa 1300 677 765 0 30 0
Placozoa 1 350 58 0 217 4 1
Ctenophora 150 112 24292 0 65 0
Acoela 300 152 2974 0 56 0
Nemertodermatida 20 32 0 0 14 0
Orthonectida (Mesozoa) 24 2 0 0 0 0
Rhombozoa (Mesozoa) 78 55 0 0 33 0
Echinodermata 7000 505937 350163 86326 51609 20 1
Hemichordata 106 200 202190 0 182 2 1
Xenoturbellida 2 26 2137 0 60 1
Cephalochordata 29 82337 335040 66720 1509 8 1
Tunicata 2566 84497 1253519 1898 4720 6 1 1
Vertebrata (non-human) 58389 20287276 17701135 7996959 1163890 885 56 17
Vertebrata (human) 1 3413921 8137747 1212854 453753 2 2 >10
Chaetognatha 100 460 1227 0 345 2
Arthropoda 1100200 3898620 3544588 791280 451420 170 21 20
Onychophora 165 238 0 0 211 1
Pentastomida 100 9 0 0 26 1
Loricifera 22 1 0 0 0 0
Priapulida 18 64 2281 0 79 1 1
Kinorhyncha 150 10 0 0 1 0
Tardigrada 980 851 5235 1063 135 0 1
Nematoda >25000 362528 1022639 683724 126769 26 3 20
Nematomorpha 320 32 0 0 10 0
Gastrotricha 450 68 0 0 3 0
Myzostomida 170 123 0 0 50 0
Lobatocerebromorpha 1 0 0 0 0 0
Bryozoa 4500 1276 0 2 702 2
Entoprocta 150 60 0 0 100 2
Rotifera 1800 1702 3219 1 1537 2
Acanthocephala 1000 440 0 0 287 1
Micrognathozoa 1 5 0 0 2 0
Gnathostomulida 80 75 0 0 25 0
Cycliophora 2 340 0 0 277 0
Platyhelminthes 55000 216602 448555 62384 17505 24 4
Brachiopoda 335 383 0 0 320 3 3
Phoronida 20 90 0 0 81 0
Nemertea 7500 576 0 0 314 0
Mollusca 70000 260920 659140 5661 33600 41 1 2
Sipunculida 320 155 0 0 127 0
Annelida 15000 9397 310246 0 4618 6 2 6
Approx. 60,000 species represented on GenBank total number of estimated species > 1.38 million

Figure 11.1 A conservative consensus estimate of metazoan phylogeny based on the information in Table 11.1. It shows indications of estimated number of known species and, from the NCBI (GenBank) data bases, the number of nucleotide sequences (core), the number of nucleotides from large-scale expressed sequence tag (EST) or genome (GSS) projects, the number of protein (prots) sequences, the number of mitochondrial genomes (mt), and the number of completed and on-going genome projects (as of mid-2008).

110 AN I M AL EV O L UTI O N



complete intrinsic isolating mechanisms, exten- sive introgressive hybridization may occur, even of morphologically distinct species (Wiens et al.,
2006). Although extensive gene exchange between morphologically distinct species is considered rare (Coyne and Orr, 2004), this could scramble any ori- ginal phylogenetic signal (Clarke et al., 1996; Chan and Levin, 2005); it has recently been suggested to be a possible reason why even vast numbers of genome data may not be able to resolve high-level phylogenetic relationships (Hallström and Janke,
2008). Causes in this category also relate to the power of natural selection or shared internal con- straints to produce extensive convergent evolution, and parallelisms (non-random non-phylogenetic signal) that may lead to the false inference of mono- phyletic taxa. This can be an important problem for both morphological and molecular phylogen- etic analyses (Waegele and Mayer, 2007). Here we should distinguish between stochastic (sampling) error and systematic error. Small data sets can be prone to stochastic error as chance similarities (random noise) can incorrectly group unrelated taxa. Increasing the amount of data helps to avoid stochastic error, but can introduce the far more ser- ious problem of systematic error.
Systematic errors are tree reconstruction arte- facts that result from the inability of a method to deal with biases in a data set that can conflict with or obscure phylogenetic signal. Systematic error may result from, for example, base or amino acid compositional biases between taxa, differences in evolutionary rates between taxa or regions of the sequences, and shifts of position-specific evolu- tionary rates (heterotachy). As expertly discussed in a series of papers by Philippe and co-workers (Philippe et al., 2005b; Brinkmann and Philippe,
2008; Lartillot and Philippe, 2008), these phenomena can cause strongly non-random, non-phylogenetic signals that can mislead phylogenetic analyses. The difficulty of trying to disentangle phylogenetic and non-phylogenetic signals is potently illustrated by the continuing debate about the validity of either Coelomata or Ecdysozoa using large data sets for a small sample of taxa (Rogozin et al., 2007b; Roy and Irimia, 2008a). The different results reported by different authors reflect how well their methods are able to deal with systematic error.
The above causes can affect phylogenetic ana- lyses of both fossil and extant taxa at any taxonomic level and independent of the type of evidence used. Difficulties generally become greater with increasing age of the divergence events we attempt to reconstruct, and all causes mentioned have probably confounded attempts to place particu- lar Problematica in the tree of the Metazoa. In the following sections we pay more detailed attention to specific causes that are of relevance for certain Problematica.
Several criteria can be used to recognize Problematica: (1) the number of alternative sister- group hypotheses; (2) the phylogenetic spread and hierarchical range of alternative sister-group hypotheses; (3) controversial homology assess- ments; (4) absence of phylogenetically informative characters; and (5) assessment of molecular data quality.
The first two criteria are straightforward for recognizing Problematica when comparing dif- ferent phylogenetic analyses, either by differ- ent workers or based on different treatments of the same data set. Classic Problematica, such as Chaetognatha, Ectoprocta, and Pogonophora, have long exhibited both a large number of alternative sister group hypotheses, and a large phylogenetic spread among these alternatives (covering both Protostomia and Deuterostomia). The phylogen- etic spread of alternative hypotheses is positively related to the hierarchical depth across which the alternatives may be distributed. For example, the placement of Pentastomida is problematic only within the Panarthropoda, with a position either within Crustacea or in the arthropod stem group as the two main contending hypotheses (Waloszek et al., 2005b). In contrast, the fossil vetulicolians are problematic on a much larger scale, across a wide phylogenetic spread (Bilateria), and a large hierarchical depth (ranging from being attributed to a separate ‘phylum-level’ clade, to belonging to a subtaxon of Tunicata) (Aldridge et al., 2007). Vetulicolians also illustrate the challenges of hom- ologizing imperfectly preserved and poorly under- stood features of fossils with key characters in extant taxa, with each decision strongly affecting the resulting phylogenetic hypothesis (Aldridge et al., 2007; Swalla and Smith, 2008). Other taxa

P R OB LE MA TI C A 111



are problematic because of the lack of, or insuffi- cient study of, informative characters. Myxozoa, for example, are very likely to be derived cnidar- ians (Jiménez-Guri et al., 2007) that share so few characters with their closest non-parasitic relatives that most textbooks did not even include them in the Metazoa until very recently. Lacking detailed knowledge may also cause Problematica to be excluded from phylogenetic discussions. Species such as Jennaria pulchra, the lobatocerebrids, Xenoturbella bocki (until recently), Buddenbrockia, and myxozoans, but also myzostomids and pen- tastomids, are frequently excluded from morpho- logical phylogenetic analyses. This is not because their phylogenetic position is so well understood. Finally, Problematica can be provisionally iden- tified by the tell-tale signs of systematic errors in molecular data sets, such as mutational saturation of sequences, compositional biases in nucleotides or amino acids, and different evolutionary rates between taxa (Philippe et al., 2005b; Waegele and Mayer, 2007; Brinkmann and Philippe, 2008). When such features are not properly dealt with they can cause tree reconstruction artefacts.


11.3 Fossil Problematica

11.3.1 The vagaries of preservation, typological thinking, and model choice

All the difficulties that beset phylogenetic ana- lyses of extant taxa also play a role in the system- atization of fossils. With fossils, however, several additional factors can cause problems, of which we think three are of particular importance. First, preservational artefacts can lead to formidable problems of interpretation. Although the major- ity of fossils can be related to extant body plans without much difficulty, ‘unusual objects do occur in rocks’ (Yochelson, 1991, p. 288). Problematica are particularly common from the fossil record of the late Neoproterozoic and earliest Phanerozoic (c. 575–500 million years ago) and it is especially these forms that may provide unique clues to the origin and diversification of early animal body plans. Yet many important taxa found in this time interval defy unambiguous interpret- ation because of the limits of preservation, and
taphonomic changes of the organism and sur- rounding sediment. This is clearly illustrated in recent debates over the putative Precambrian ani- mal Vernanimalcula (a coelomate bilaterian?), the Cambrian animal Odontogriphus (segmented?), the oldest putative metazoan eggs and embryos (ani- mals or bacteria?), and in the continuing debate about the Ediacaran biota (Dzik, 2003; Fedonkin,
2003; Bengtson and Budd, 2004; Chen et al., 2004; Narbonne, 2005; Butterfield, 2006; Caron et al., 2006; Bailey et al., 2007a; Donoghue, 2007).
Budd and Jensen (2000) nominated typological thinking as another factor that may hinder the phylogenetic systematization of fossils, especially in the context of extant taxa. By a misguided emphasis on differences, fossils have automatically been labelled Problematica if their body plan did not exactly conform to that of a living phylum (see also Briggs et al., 1992). Such reasoning is incom- patible with established phylogenetic logic, but it is nevertheless prevalent (Jenner, 2006a).
A third factor that inescapably affects thinking about fossil Problematica is that fossils are predom- inantly interpreted in the light of our knowledge of living species. Consequently, disagreements about the phylogenetic placement of fossil Problematica frequently hinge upon the use of different living species as models for interpretation, as illustrated by the vetulicolians (Aldridge et al., 2007). Related to this is that phylogenetic analyses of fossils may be strongly dependent upon a very small number of informative features that can be homologized between fossils and extant taxa. Consequently, the interpretation of these features can have a very strong effect on phylogenetic conclusions, whether that seems justified or not (for vetulicolians see Swalla and Smith, 2008).


11.3.2 Solving fossil Problematica: stem groups, new fossils, new techniques

Yochelson (1991, p. 289) remarked that he could only offer ‘a few platitudes’ about how ‘to do’ fossil Problematica. We hope the following sug- gestions are helpful. In essence, fossils should be treated like any other living taxon. Attempts to systematize fossils will lead to the establishment of stem groups (Conway Morris, 2000; Budd and

112 AN I M AL EV O L UTI O N



Jensen, 2000; Budd, 2001b, 2003). Although differ- ences between fossils and extant taxa should not be ignored, they should not be interpreted typo- logically as evidence against affinities (Budd and Jensen, 2000; Jenner, 2006). Putative stem-group taxa are expected to exhibit some, but not all, of the diagnostic characters of crown groups, and by creating paraphyletic series of stem taxa we can illustrate the orderly sequential evolution of body plans. This may not be easy of course. If crucial information is not preserved, a fossil may not be reliably placed. Specifically, the lack of a diagnos- tic crown-group character state in a fossil, due to taphonomy, could bias a phylogenetic analysis by erroneously placing the fossil in the stem group, a problem that might be widespread (Donoghue, and Purnell, 2009). In such cases, unless new fossils are found or new techniques reveal new information, ambiguity will endure.
The main reason why fossil Problematica occur frequently in the late Neoproterozoic and early Phanerozoic is extinction. These fossils document the early evolution of animal body plans. The older fossils are, the more they are expected to fall outside the limits of extant body plans (Budd,
2003; Valentine, 2004). Unless body plan evolu- tion takes large leaps, failure to systematize fossil Problematica is chiefly the result of not (yet) know- ing related taxa that can bridge their morphology with those of the crown group. Hence, most pro- gress is made with fossil Problematica when new specimens are found. Better-preserved fossils and forms with novel character combinations address the problems of taxon and character matrix com- pleteness, allowing unknowns to be substituted with characters. Nevertheless, this approach relies on much fieldwork and a great deal of luck.
Palaeontological and analytical techniques are constantly being developed that present ways of discerning new characters, or of better resolving existing ones, and of handling existing data. For example, the three-dimensional reconstruction of fossil forms from thin serial sections has achieved remarkable levels of resolution, thanks to refine- ments in microscopy and computer rendering. This has provided valuable phylogenetic infor- mation for a diversity of taxa, ranging across the Bilateria (Sutton et al., 2001a,b, 2005a,b,c; Thomson
et al., 2003). X-ray tomographic microscopy and Raman spectroscopy combined with confocal laser scanning microscopy have also yielded images and insights into the biomolecular nature of fossils with unrivalled resolution (Schopf and Kudryavtsev, 2005; Donoghue et al., 2006a; Chen et al., 2007).
Other advances will come from improvements in methods of phylogeny reconstruction. Model- based methods of analysis have proven their worth with molecular data, particularly in dealing with long-branch problems in phylogenetic reconstruc- tion. Such methods, although still in their infancy, are now available for the analysis of morphological and fossil data as well (Lewis, 2001). This prom- ises the chance to include incomplete taxa, such as fossil Problematica, with morphological and even molecular data from extant taxa using maximum likelihood or Bayesian techniques (Wiens, 2005), while at the same time parsimony-based meth- ods are refined to be able to deal efficiently with large amounts of diverse phylogenetic evidence (Wheeler et al., 2006).


11.4 Extant invertebrate Problematica

11.4.1 An apparent paradox: a weak molecular signal and large amounts of morphological evolution

It is not surprising that Problematica are encoun- tered when metazoan phylogeny is analysed on the basis of extant taxa alone. First, any compari- son between two extant species belonging to dif- ferent phyla has to bridge in the order of 1 billion years of independent evolution. This is ample time to erase signs of ancestry, either through extensive modification or loss of characters, and for conver- gent evolution to obscure phylogenetic signal. It may thus be unsurprising that sessile taxa (ecto- procts, brachiopods, phoronids), very small (pos- sibly miniaturized) taxa (tardigrades, placozoans, Lobatocerebrum), and parasitic taxa (pentastomids, myxozoans) have been particularly prominent Problematica. Another consequence is that molecu- lar phylogenies of the Metazoa bear the typical sig- nature of short stems and long terminal branches, providing ample opportunity for long branch

P R OB LE MA TI C A 113



attraction (Waegele and Mayer, 2007). This has been a problem for the placement of several taxa, ranging from myxozoans to acoels (Philippe et al.,
2007). Second, the major metazoan lineages may have radiated very rapidly, potentially allowing for very little phylogenetic signal to evolve. Although it remains disputed whether lack of resolution is a convincing signature of closely spaced cladogen- etic events (Giribet, 2002; Rokas et al., 2005; Rokas and Carroll, 2006; Baurain et al., 2007; Whitfield and Lockhart, 2007), if current molecular clock esti- mates of metazoan divergence times are approxi- mately accurate (Peterson et al., 2004, 2005, 2008), the fact remains that the major metazoan lineages diverged over a time span that is significantly shorter than the subsequent independent history of modern phyla (including their stem groups). The appearance in the fossil record of a variety of crown phyla with their distinctive body plans as early as the Cambrian (Budd, 2003; Valentine,
2004) implies that important morphological traces of ancestry were probably already erased early in metazoan history.
Intriguingly, the relative branch lengths of morphological metazoan phylogenies seemingly contradict the absence of sufficient phylogenetic signal. These typically show a much smaller dis- crepancy between the length of stems and terminal branches, or even the opposite pattern of relatively longer stems and shorter tips (Zrzavý et al., 1998,
2001; Nielsen, 2001; Peterson and Eernisse, 2001; Brusca and Brusca, 2003; Zrzavý, 2003). Large amounts of body plan evolution are commonly inferred along almost all stems. This raises interest- ing issues about the relationship between genetic and phenotypic evolution that we cannot address here. What is pertinent though is the large amount of body plan evolution inferred across a relatively small number of speciation events. For example, depending on the precise topology of the tree, pos- sibly just six or seven nodes separate the body plan of the last common ancestor shared by (at least some) sponges and the remaining animals, and the last common ancestor of the chordates! Unless half a dozen speciation events are really all that is required to evolve from a sponge-grade organiza- tion to that of a protochordate, we must be missing something. That something is fossils.
Recent studies of the fossil record have yielded important insights that may help explain why extant Problematica are to be expected. First, Wagner (P. J. Wagner, 2000, 2001; Wagner et al.,
2006) drew the important conclusion that during evolutionary history taxa tend to exhaust their character state spaces. This means that, as clades age, homoplasies increase in frequency. Not sur- prisingly, homoplasies are common between the major lineages of animals (Valentine, 2004). Our estimates of homoplasy based on morphological phylogenetic studies are likely underestimates, giving a widespread problem of character coding (Jenner, 2004b).
Distressingly, P. J. Wagner (2001) noted that the inclusion of fossils into a phylogenetic analysis of extant species could reveal a significant amount of previously hidden character change along branches subtending extant taxa. This positive correlation between the amount of character change that is discovered and the number of taxa included is well known by molecular systematists, and is known as the node density effect. However, its effect for morphological phylogenetics and inference of body plan evolution has barely been acknowl- edged (Jenner and Wills, 2007). Hence, the inclu- sion of even incomplete fossil taxa has the potential to reveal that synapomorphies of extant taxa may in fact be homoplasies or symplesiomorphies, and their inclusion can improve accuracy of the phylo- genetic relationships inferred between living taxa (Wiens, 2005). The reconstruction of stem groups is crucial for a complete picture of body plan evolu- tion, and there is ample evidence that phylogenetic inferences based on extant taxa can be misled; for arthropod examples see Budd (2001b) and Eriksson and Budd (2000). The amount of character evolution that is missed by a focus on extant taxa is increas- ingly illustrated by studies showing that rates of morphological character change may be highest early in the history of a clade, which may go hand in hand both with the general early establishment of morphological disparity in the history of large clades and indications that morphological trans- formations had larger step sizes early in a clade’s history (Valentine, 2004; Ruta et al., 2006; Erwin,
2007). In combination, these insights suggest that by focusing on living taxa only we are missing a

114 AN I M AL EV O L UTI O N



lot of character evolution, the recognition of which is crucial to prevent clades being based on homo- plasies or symplesiomorphies.


11.4.2 From the unequal eye to morphological cladistics

To see all things with equal eye is not within our power: humans, and especially human narrators, always look upon the world with an unequal eye.
O’Hara (1992, p. 140)

Before computers came to the assistance of phylo- genetic analysis, Problematica were an inescapable by-product of phylogenetic inference. Without the help of a computer it is impossible to achieve a balanced and unbiased evaluation of large num- bers of comparative data for more than a few taxa. Emphasis on different aspects of available evidence as well as the lack of a uniform phylogenetic meth- odology fostered disagreement between workers. Consequently, from the beginning of our discipline one researcher’s central insights were not uncom- monly labelled another’s ‘fata morgana’ [mirage] (Hubrecht, 1887, p. 641), and the coordinating theme of one school of zoological thought would deserve to be ‘dead and buried’ in the opinion of proponents of another (Hyman, 1959, p. 750).
The widespread adoption of cladistic reasoning in the second half of the 20th century increased the promise of reaching a general consensus on metazoan phylogeny. Yet, without the help of computers, progress was slow as the amount of conflicting evidence allowed many mutually exclusive conclusions. The computer-assisted morphological cladistic analyses of metazoan phylogeny published over the last decade greatly advanced the objectivity, explicitness, and test- ability of phylogenetic hypotheses. In this period the field progressed significantly beyond the trad- itional textbook trees (Adoutte et al., 2000), but perhaps the most important insight of this era of fruitful debate was discovering exactly how prob- lematic many taxa and clades actually were. As reviewed elsewhere (Jenner, 2004a,b), differences in the construction of data matrices, including different strategies of character selection, char- acter coding and scoring, and taxon selection,
resulted in many incompatible phylogenies. Taxa such as Chaetognatha and Ectoprocta behave like phylogenetic renegades, residing in as many dif- ferent clades as there are studies, and although other aspects of the phylogenetic backbone seemed more secure (monophyly of Protostomia, Spiralia), total agreement between analyses is absent. Evidently, the phylogenetic signal residing in morphology needs to be supplemented with molecular evidence.


11.4.3 Old Problematica solved and new
Problematica revealed

A new phylogenetic synthesis for the Metazoa (Halanych, 2004) (Figure 11.1) has emerged largely on the basis of molecular evidence. The backbone of this phylogeny is based on nuclear ribosomal sequences (18S and 28S rDNA), and despite chal- lenges (Rogozin et al., 2007b) its major aspects are confirmed by increasingly sophisticated phylog- enomic analyses based on larger amounts of data, and employing improved model-based analytical methods (Philippe et al., 2005b; Baurain et al., 2007; Brinkmann and Philippe 2008; Irimia et al., 2007; Hausdorf et al., 2007; Roeding et al., 2007; Dunn et al., 2008; Helmkampf et al., 2008a,b). These stud- ies have done a fine job in solving some of the classical Problematica. For example, the enigmatic myxozoans and Polypodium have now been firmly placed within Cnidaria, the pogonophorans and vestimentiferans are now placed within Annelida, Xenoturbella is firmly placed as the sister group to Ambulacraria, and chaetognaths and the lopho- phorates are now definitely excluded from the Deuterostomia. The reliable placement of these taxa reveals why they were problematic before. They are all highly modified taxa and they have either lost complexity, or evolved an otherwise unique body plan.
However, in the case of Chaetognatha, for example, the ‘solution’ is not yet complete (Table 11.1), as their exact phylogenetic position remains uncertain. Our finding that over half of the major metazoan lineages listed in Table 11.1 can be classified as Problematica is quite remark- able. We classify as category I Problematica those taxa for which there is still no consensus about

P R OB LE MA TI C A 115



either their broad phylogenetic neighbourhood, let alone their precise position, or for which a pre- cise understanding of their phylogenetic position is of particular importance for understanding major transitions in the evolution of animal body plans. Category II Problematica are those for which we have some idea about their general phylogen- etic neighbourhood, but we are still far removed from reliably placing them. Knowing the precise position of these taxa will aid our understanding of body plan evolution mostly within the confines of relatively smaller clades, principally within the Lophotrochozoa. Only 21 out of 45 lineages in Table 11.1 can reasonably be labelled as non- Problematica, and six of these fall within the three
‘phyla’ Porifera, Cnidaria, and Annelida.
Probably the most important reason for the con-
tinued existence of Problematica is systematic error,
despite the fact that most of them have now been
included in at least one phylogenomic analysis.
Even though the limited overlap in genes between
published phylogenomic analyses (see supplemen-
tary information in Dunn et al., 2008) may lead one
to suspect sampling artefacts, systematic error is
the inescapable explanation of several discrepan-
cies noted between different analyses. In studies
such as those of Lartillot and Philippe (2008), Dunn
et al. (2008), Helmkampf et al. (2008b), and Struck
and Fisse (2008), conspicuous differences between
analyses of the same data set with different meth-
ods indicate sensitivity to systematic errors. The fit
between the chosen method/evolutionary model
and the data set is crucial, but one shoe does not
necessarily fit all. For example, even though the
CAT mixture model has been promoted as a super-
ior model (Brinkmann and Philippe, 2008; Lartillot
and Phillippe, 2008) for phylogenomics, particu-
larly in the fight against long branch attraction,
application to the data set of Struck and Fisse (2008)
generated some likely nonsense results, such as the
position of Syndermata within the Ecdysozoa (simi-
lar to the results of Helmkampf et al., 2008b, and
Marlétaz et al., 2008, when they applied the CAT
model). Dunn et al. (2008) noted that the position of
Tardigrada is strongly model dependent for their
data set. Clearly, data quality and model fit need to
be assessed for each data set individually. Examples
of taxa for which the lack of current consensus
is likely to be in part due to systematic error are Tardigrada, Acoela, Myzostomida, Bryozoa, Syndermata, Gastrotricha, Platyhelminthes, and Gnathostomulida. These taxa are unstable in phyl- ogenomic analyses, and are often fast evolving for the sampled markers. We think that the clade unit- ing Myzostomida, Acoela, and Gnathostomulida in Dunn et al. (2008) is emblematic for this problem. If this clade goes, anything goes.
Systematic error may also be the reason why phy- logenomic analyses may or may not support mono- phyly of Deuterostomia depending on the choice of evolutionary model (Lartillot and Philippe, 2008; Marlétaz et al. 2008). This illustrates that the over- all relationships between larger clades may also be very uncertain, which holds true in particular for the topology within Lophotrochozoa.
Finally, new Problematica can be revealed by new discoveries or reinterpretations of estab- lished taxa on the basis of both molecular and morphological evidence. For example, detailed morphological study and preliminary molecular phylogenetic analysis of the interstitial worm-like genus Diurodrilus strongly suggests that it does not fall within the polychaetes, as previously assumed, but may instead represent an independent lin- eage of animals potentially related to gnathifer- ans (Worsaae and Kristensen, 2003; Worsaae and Rouse, 2008). This shows both the importance of the continued surveying of under-explored habi- tats for new groups of organisms, and the import- ance of properly integrating new discoveries in a phylogenetic framework so as to illuminate the evolution of animal body plans.


11.4.4 Guidelines for future progress in metazoan phylogeny

A large literature exists on troubleshooting molecular systematics. Some excellent recent reviews include: Gribaldo and Philippe (2002), Sanderson and Shaffer (2002), Delsuc et al. (2005), Philippe et al. (2005b), Boore (2006), Philippe and Telford (2006), Rokas and Carroll (2006), Wiens (2006), and Whitfield and Lockhart (2007). We extract a number of guidelines that we feel need to be kept in mind to ensure continued progress in understanding.

Table 11.1 A list of all major ‘phylum-level’ metazoan taxa with notes on Problematica. Problematic taxa for which either very little is known (e.g. Lobatocerebrum sp., Planctosphaera pelagica), or which are likely to be incertae cedis on lower taxonomic levels (Aeolosomatidae within annelids) are not included. Taxa are classified into categories I–IV, based on a consideration of the degree of controversy surrounding their phylogenetic placement, and their importance for understanding body plan evolution. Category I includes true modern Problematica for which there exists greatest uncertainty about their phylogenetic position and/or those for which an understanding of their true position is crucial for understanding major transitions in the evolution of body plans. Category II groups Problematica for which there is still serious uncertainty about their phylogenetic position, but for which the alternative hypotheses are more restricted in either number or phylogenetic depth. Categories III and IV group non-Problematica. Category III groups taxa for which the phylogenetic neighbourhood seems secure, and for which future efforts should primarily focus on positioning
the taxa either close to one or a few, or within other ‘phyla’. Category IV houses taxa for which their sister-group relationships now seem established. Please note that we did not try to achieve a comprehensive listing of alternative sister-group hypotheses. We restricted ourselves principally to recent phylogenomic and molecular phylogenetic analyses. Including a full consideration of available morphological and combined evidence analyses would have collapsed our classification into one bucket of Problematica that included all listed taxa. The table should be taken as a tool to facilitate discussion about the focus of future work, and as a framework for comparison with morphology-based studies. Note that cases of congruence between different phylogenomic analyses, in particular those of Dunn et al. (2008) and those of H. Philippe and colleagues, can be interpreted as providing largely independent support for phylogenetic hypotheses given the limited overlap between the genes upon which these analyses are based.

Cat. Taxon Alternative sister groups Remarks Recent references
I Demospongiae Hexactinellida, Calcarea (Homoscleromorpha Eumetazoa) Borchiellini et al. (2004) reported the intriguing finding that the demosponges are only monophyletic (Demospongiae sensu stricto) when the homoscleromorphs are excluded. Newer analyses have upheld the separate status of the homoscleromorphs, but the first molecular phylogenetic analysis of hexactinellid sponges indicates that these fall within a paraphyletic Demospongiae (Dohrmann
et al., 2008). The position of Demospongiae sensu stricto + Hexactinellida with respect to homoscleromorphs and calcareans remains unresolved, as is the question of the monophyly of sponges (see below) Borchiellini et al. (2004), Erpenbeck and Wörheide (2007), Sperling et al. (2007), Dohrmann et al. (2008)
I Calcarea Eumetazoa, Homoscleromorpha, Homoscleromorpha + Eumetazoa Poriferan paraphyly, with calcareans and non-poriferan metazoans most closely related to each other, was reported repeatedly in papers from the late 1990s. However, newer and more comprehensive phylogenetic analyses now paint a different picture. Either Calcarea are most closely related to homoscleromorphs (previously considered to be demosponges, see below), a hypothesis supported
by nuclear and mitochondrial ribosomal sequences (Dohrmann et al. 2008) as well as an unpublished phylogenomic analysis (R. Derelle, unpublished doctoral thesis, 2007), or calcareans are the sister group to Homoscleromorpha + Eumetazoa (Sperling et al., 2007) Erpenbeck and Wörheide (2007), Sperling et al. (2007), Dohrmann et al. (2008)
I Homoscleromorpha Eumetazoa, Calcarea Paraphyly of sponges, in particular with homoscleromorphs as the sister group to the remaining metazoans, has recently been used as an interpretative framework for understanding the earliest steps of metazoan body plan evolution (Sperling et al. 2007; Nielsen 2008). However, although sponge paraphyly is supported by some studies (Sperling et al , 2007), it is not supported by others (Dohrmann et al., 2008). Interestingly, an unpublished phylogenomic analysis that includes
representatives of calcareans, demosponges, homoscleromorphs, and hexactinellids supports poriferan monophyly (R. Derelle, unpublished doctoral thesis, 2007; M. Manuel, personal communication). If confirmed, this largely removes the rationale for using sponge body plans for understanding the origin of eumetazoan body plans Borchiellini et al. (2004), Derelle (2007), Erpenbeck and Wörheide (2007), Sperling et al. (2007); Dohrmann et al. (2008)

I Placozoa Cnidaria, Cnidaria + Bilateria, Myxozoa + Bilateria, Bilateria, Porifera + Cnidaria
The phylogenetic position of Placozoa remains profoundly puzzling, with morphology, mitochondrial genomes, nuclear ribosomal sequences, or combined morphological and molecular evidence providing no consensus whatsoever. We eagerly await their first inclusion in a phylogenomic analysis. Considering the fact that placozoans represent the morphologically simplest animal ‘phylum’ it is of great interest
to see if they are primitively simple or secondarily simplified
Eernisse and Peterson (2004), Glenner et al. (2004), Wallberg et al. (2004), Dellaporta et al. (2006), Cartwright and Collins (2007), da Silva et al. (2007), Ruiz-Trillo et al. (2008)


I Ctenophora All other metazoans, Planulozoa (Cnidaria, Placozoa, Myxozoa, Bilateria), Porifera

The phylogenetic position of the Ctenophora is one of the biggest problems at the base of the Metazoa. Specifically, molecular sequence data consistently place the ctenophores outside a clade including cnidarians and bilaterians, whereas interpretations of morphological and embryological data instead suggest a closer relationship between ctenophores and bilaterians. With the exception of the unpublished PhD thesis of R. Derelle (2007) ctenophores were first included in the phylogenomic
analysis of Dunn et al. (2008). Surprisingly, this placed them as a sister group to all remaining metazoans, in agreement with the analyses in Derelle’s thesis. If confirmed, this either implies that sponges and placozoans have become greatly simplified, or that comb jellies have convergently evolved an astonishing amount of developmental and morphological complexity, shared with
eumetazoans

Eernisse and Peterson (2004), Cartwright and Collins (2007), Derelle (2007), Dunn et al. (2008), Nielsen (2008)


I Acoela Nemertodermatida
+ Nephrozoa, Nemertodermatida, various clades of deuterostomes, Gnathostomulida, Protostomia

Although acoels (and nemertodermatids) are placed at the base of the Bilateria when nuclear ribosomal and protein-coding genes are considered, they are unstable in more data-rich phylogenomic analyses, placing them either in (or sister to) Deuterostomia, or in (or sister to) Protostomia. Hox cluster data are ambiguous at the moment. Interestingly, data on the presence of miRNAs seems to support a
placement of acoels at the base of the Bilateria. Given that so far miRNA data seem to be remarkably free of homoplasy, and the investigated acoels possess only a subset of miRNAs shared between protostomes and deuterostomes, this is strong support for their exclusion from Nephrozoa, the clade of bilaterians characterized by possession of complex organs such as nephridia. However, this result is apparently contradicted by phylogenomic analyses that place acoels in various positions higher in
the tree, which would imply they are secondarily simplified. Thus, on the balance of current evidence it is impossible to place them with any confidence. Considering their morphological simplicity and the lack of a biphasic life cycle, proper placement of the acoels and nemertodermatids will have important consequences for character optimization, and thus our understanding of the evolution of complex
morphology and life cycles

Sempere et al. (2006, 2007), Brinkmann and Philippe (2008), Philippe et al. (2007), Wallberg et al. (2007), Baguñà et al. (2008), Deutsch (2008),
Dunn et al. (2008)


I Nemertodermatida Nephrozoa, Acoela The relationship of nemertodermatids and acoels on the basis of ribosomal sequence data remained uncertain. Although both were positioned basal to the remaining bilaterians (Nephrozoa), the monophyly of Acoelomorpha remained uncertain. The most recent molecular phylogenetic analyses
(Wallberg et al., 2007; Baguñà et al., 2008) support the status of acoels and nemertodermatids as independent lineages. However, in view of the fact that nemertodermatids have not yet been included in phylogenomic analyses, and the unstable position of the acoels in such analyses,
we cannot yet ascertain the precise positions for these two taxa

Wallberg et al. (2007), Baguñà
et al. (2008)

Table 11.1 (Continued.)


Cat. Taxon Alternative sister groups
Remarks Recent references


I Tardigrada Nematoida, Onychophora, Onychophora + Arthropoda
Ribosomal sequence data have suggested the possibility that tardigrades and onychophorans are sister taxa (Mallatt and Giribet, 2006). However, the phylogenetic position of tardigrades in more recent phylogenomic studies has been more difficult to determine due to differences between the studies in taxon sampling. In analyses that exclude Nematomorpha and or Onychophora, such as Brinkmann and Philippe (2008), Roeding et al. (2007), Helmkampf et al. (2008b), and Lartillot and Philippe (2008), tardigrades are sister group to Nematoda. The study of Dunn et al. (2008) shows that the phylogenetic position of tardigrades is very sensitive to the choice of molecular evolutionary model, so that currently a choice is
not possible. However, when both tardigrades and onychophorans are included, phylogenomic evidence suggests unequivocally that onychophorans are more closely related to arthropods than are tardigrades. Placement of tardigrades separate from onychophorans and arthropods could imply their independent
evolution of limbs, which would be an astonishing case of convergent evolution
Mallatt and Giribet (2006), Brinkmann and Philippe (2008), Roeding et al. (2007),
Dunn et al. (2008), Helmkampf et al. (2008b), Lartillot and Philippe (2008)


I Ectoprocta Entoprocta, Platyzoa = (Platyhelminthes, Acoela, Gastrotricha, Myzostomida, Gnathifera), Platyhelminthes, all
other lophotrochozoans, Myzostomida
Bryozoans remain a true Problematicum, as neither morphological evidence nor available molecular analyses can agree on their monophyly, or their phylogenetic position. Dunn et al. (2008) identified them as an unstable taxon, for which increased species sampling is necessary for fully resolving their position. It is remarkable that the phylogeny of Dunn et al. (2008) has two main clades within
Lophotrochozoa, to which the sessile entoprocts and ectoprocts, respectively, are sister taxa. Interestingly, the coelomate ectoprocts are the sister group to the clade of acoelomate groups, whereas the acoelomate entoprocts are sister to the clade of coelomate lophotrochozoans. Outgroup comparison with the coelomate chaetognaths would indicate that the acoelomate condition has evolved independently in
entoprocts and the other acoelomate lophotrochozoans
Eernisse and Peterson (2004), Halanych (2004), Passamaneck and Halanych (2006), Hausdorf et al. (2007), Dunn et al.
(2008)


I Gastrotricha Platyhelminthes, all other nephrozoans, Micrognathozoa, Cycliophora, Rotifera
Gastrotricha is another problematic taxon that has been labelled as unstable in phylogenomic analyses (Dunn et al., 2008). Dunn et al. (2008) support a sister-group hypothesis with Platyhelminthes. However, other analyses based on a smaller number of data (but with better taxon sampling or including morphology) suggest other possibilities. Morphology and molecules appear to conflict with each other, as cuticle characters group gastrotrichs with the ecdysozoans, and sequence data place them in Lophotrochozoa
Eernisse and Peterson (2004), Halanych (2004), Todaro et al. (2006), Dunn et al. (2008)


I Chaetognatha Lophotrochozoa, Protostomia, Onychophora, Priapulida

Phylogenomic analyses of this classic Problematicum have not yet reached a consensus on whether arrow worms are a sister group to Lophotrochozoa (Matus et al., 2006b; Dunn et al., 2008), or Protostomia (Marlétaz et al., 2006, 2008; Brinkmann and Philippe, 2008; Lartillot and Philippe, 2008), and their position can be sensitive to method of analysis (Helmkampf et al. 2008b). Note that Helmkampf et al. (2008a) united chaetognaths with priapulids within the Ecdysozoa. A position of the chaetognaths as sister group to a major clade(s) of bilaterians will have major consequences for how we reconstruct the evolution of a host of organ systems, given the chaetognaths’ unique mix of what are traditionally perceived to be characters of distinct clades

Eernisse and Peterson (2004), Halanych (2004), Marlétaz
et al. (2006, 2008), Matus
et al. (2006b), Hausdorf et al. (2007), Dunn et al. (2008), Helmkampf et al. (2008a,b), Lartillot and Philippe (2008)

I Myzostomida Acoela + Gnathostomulida, within Ectoprocta, within Annelida
A genuine Problematicum, myzostomids possess morphological and embryological characters that seem to unite them to various phyla, from annelids to rotifers. However, uncritical treatment of this evidence has compromised morphological and combined evidence phylogenetic analyses (Jenner, 2003). Taxon sampling is a crucial parameter for resolving their position using molecular data. In Dunn et al. (2008) myzostomids are grouped with acoels and gnathostomulids in the Lophotrochozoa. Although this position far removed from the annelids finds apparent support in some previous molecular phylogenetic analyses as well (Giribet et al., 2004; Passamaneck and Halanych, 2006), Bleidorn et al. (2007) found that long branch attraction probably affected the results of these studies. It is notable that molecular phylogenetic studies that include a greater sample of annelids (Hall et al., 2004; Colgan et al., 2006; Rousset et al., 2007) consistently unite myzostomids with annelids. Hence, on the balance of current evidence, their position remains uncertain. If myzostomids are not annelids, the amount of convergent evolution of morphological and developmental details shared with particular annelid taxa will be astonishing
Jenner (2003), Giribet et al. (2004), Hall et al. (2004), Colgan et al. (2006), Passamaneck and Halanych (2006), Bleidorn et al. (2007), Roussett et al. (2007), Dunn
et al. (2008),


II Cnidaria Porifera, Placozoa (Myxozoa Bilateria), Placozoa, Bilateria

The two chief alternative hypotheses that are based on phylogenomic analyses cannot decide whether cnidarians are sister group to bilaterians or poriferans. However, it should be kept in mind that phylogenomic analyses do not yet include placozoans. Nevertheless, irrespective of which of these alternatives will turn
out to be correct, the shared morphological and developmental complexity of cnidarians and bilaterians is
unlikely to be convergent

Halanych (2004), da Silva et al. (2007), Dunn et al. (2008), Ruiz-Trillo
et al. (2008)


II Rhombozoa (Dicyemida and Heterocyemida)
Triploblasts or lophotrochozoans
Although they are likely to be triploblasts, the phylogenetic positions of both rhombozoans and orthonectids remain essentially unknown. Nevertheless, being highly specialized parasites, we expect their body plans to be highly modified. A recent phylogenetic analysis of dicyemid Pax6 and Zic genes supported a bilaterian affinity of dicyemids (Aruga et al., 2007)
Zrzavý (2001), Halanych
(2004), Aruga et al. (2007)


II Orthonectida Triploblasts or lophotrochozoans

As with the rhombozoans, the phylogenetic position of the orthonectids remains entirely unresolved on the basis of both scanty molecular and morphological evidence (Slyusarev and Kristensen, 2003; Halanych, 2004)

Slyusarev and Kristensen
(2003), Halanych (2004)


II Mollusca Annelida, Annelida + Platyhelminthes, Annelida +
Sipunculida + Phoronida
+ Brachiopoda + Nemertea, Annelida + Sipunculida, Nemertea, Nemertea + Sipunculida
+ Annelida, a diverse set of lophotrochozoan phyla, Entoprocta

Although our table lists a larger number of alternative sister-group hypotheses for the Mollusca than for any other taxon, this is in part due to differences in taxon sampling between different analyses, which artificially inflates the number of alternative sister-group hypotheses to some extent. Focusing on just those phylogenomic analyses with the broadest sampling of taxa (Helmkampf et al., 2008a,b; Dunn et al., 2008), we can conclude that although the exact sister group of the Mollusca remains elusive, it is at least part of
a lophotrochozoan clade including Annelida, Sipunculida, Nemertea, Phoronida, and Brachiopoda. It is noteworthy that on the basis of new morphological evidence Haszprunar and Wanninger (2008) recently proposed that a sister-group relationship between Mollusca and Entoprocta ‘is currently among the best documented interrelationships of two metazoan phyla’. Strikingly, no molecular support for this hypothesis exists

Eernisse and Peterson (2004), Passamaneck and Halanych (2006), Hausdorf
et al. (2007), Wanninger et al. (2007), Dunn et al. (2008), Haszprunar and Wanninger (2008), Helmkampf et al. (2008a,b), Lartillot and Philippe (2008)


Table 11.1 (Continued.)


Cat. Tax on Alternative sister groups
Remarks Recent references


II Annelida (including the former pogonophorans and vestimentiferans)
Mollusca, Phoronida + Brachiopoda + Nemertea, Platyhelminthes, Mollusca + Nemertea
The most broadly sampled phylogenomic analyses available (Dunn et al., 2008; Helmkampf et al., 2008a,b) have not convincingly resolved the position of annelids. Whereas Dunn et al. (2008) support a relationship of annelids with phoronids, brachiopods, and nemerteans, Helmkampf et al. (2008a,b) instead favour a relationship with phoronids and ectoprocts. The other phylogenomic analyses either suggest a close relationship to Mollusca and Nemertea, or Platyhelminthes, but these studies have more restrictive taxon sampling that does not allow all hypotheses to be tested
Brinkmann and Philippe (2008), Dunn et al. (2008), Helmkampf et al. (2008a,b), Lartillot and Philippe (2008), Struck and Fisse (2008)


II Nemertea Brachiopoda,

As discussed in Jenner (2004b), available morphological and combined evidence analyses support the

Eernisse and Peterson (2004),

Brachiopoda + Phoronida, Nemertea as part of a clade including the Neotrochozoa (Mollusca, Annelida, Sipunculida, Echiura).
Jenner (2004b), Dunn et al.

Neotrochozoa, Mollusca, Sipunculida + Annelida
Molecular sequence data in isolation, however, provide a less clear picture, partly as a result of differences in taxon sampling between studies. Intriguingly, the most comprehensive phylogenomic analyses to date (Dunn et al. 2008; Helmkampf et al. 2008b) support a sister-group relationship between nemerteans and brachiopods + phoronids, a relationship foreshadowed in some analyses based on ribosomal gene sequences (Glenner et al., 2004; Todaro et al., 2006). The phylogenomic analyses of Helmkampf et al. (2008a) and
Struck and Fisse (2008), however, group nemerteans with various neotrochozoans, to the exclusion of
brachiopods and phoronids
(2008), Helmkampf et al. (2008a,b), Struck and Fisse (2008)


II Platyhelminthes Gastrotricha, Annelida, Neotrochozoa + Brachiopoda + Phoronida
+ Nemertea, Ectoprocta, Syndermata, Mollusca + Annelida + Sipunculida, Neotrochozoa +
Brachiopoda + Phoronida
Morphological phylogenetic analyses have failed to identify a sister group of Platyhelminthes (Jenner, 2004b). Differences in taxon sampling and different results due to the application of different reconstruction methods on the same data set make it difficult to evaluate the merits of molecular and phylogenomic analyses.
The sister-group relationship with annelids (Brinkmann and Philippe, 2008; Lartillot and Philippe, 2008) is probably an artefact of insufficient taxon sampling, although the analysis of Todaro et al. (2006) based on
18S sequences and a broad taxon sampling also supports this hypothesis. A sister-group relationship of platyhelminths to a clade of neotrochozoans (plus brachiopods, phoronids, and nemerteans if these taxa are
included) is supported by Helmkampf et al. (2008a) (plus Ectoprocta), Hausdorf et al. (2007), Marlétaz et al.
Eernisse and Peterson (2004), Jenner (2004b), Passamaneck and Halanych (2006), Todaro et al. (2006), Hausdorf et al. (2007), Baguñà et al. (2008), Brinkmann and Philippe (2008), Dunn et al. (2008), Helmkampf et al. (2008a,b),

+ Nemertea + Ectoprocta, (2008) (plus Ectoprocta and Entoprocta), and Baguñà et al. (2008). Helmkampf et al. (2008b) provide
Marlétaz et al. (2008), Struck

other lophotrochozoans
uncertain support for platyhelminths as sister group to the remaining lophotrochozoans. However, a closer relationship to non-coelomate lophotrochozoans, especially syndermates and gastrotrichs (when included) is found in other analyses (Passamaneck and Halanych, 2006; Hausdorf et al., 2007; Dunn et al., 2008; Helmkampf et al., 2008b; Marlétaz et al., 2008; Struck and Fisse 2008). Consequently, on the basis of current evidence it is still impossible to nominate a reliable sister group to Platyhelminthes within Lophotrochozoa. Very provisionally one may conclude on the basis of the most comprehensive analyses that Platyhelminthes is a part of one of two main clades within the Lophotrochozoa, which in turn is sister to
a clade containing Neotrochozoa, and Nemertea, Phoronida, and Brachiopoda when these are included
and Fisse (2008), Lartillot and
Philippe (2008)




II Entoprocta Neotrochozoa + Brachiopoda + Phoronida + Nemertea, Cycliophora, Ectoprocta, Mollusca
The phylogenetic position of Entoprocta remains problematic. Morphological phylogenetic analyses suggest a variety of different sister taxa ranging from ectoprocts to lobatocerebromorphans. Notably, recent studies by Haszprunar and Wanninger (2008) and Wanninger et al. (2007) strengthen a nexus of similarities
between entoproct creeping larvae and a variety of adult and larval molluscan features. These similarities have been meant to strongly imply a sister-group relationship between entoprocts and molluscs. However, no molecular phylogenetic support for this hypothesis is available. The most comprehensive phylogenomic analysis available (Dunn et al., 2008) supports a sister-group relationship of entoprocts to the coelomate lophotrochozoans (Neotrochozoa, Nemertea, Brachiopoda, Phoronida). In contrast, the phylogenomic studies of Hausdorf et al. (2007) and Helmkampf et al. (2008b) found support for a monophyletic Bryozoa, with entoprocts and ectoprocts as sister taxa. The phylogenomic analysis of Marlétaz et al. (2008) finds some support for this hypothesis as well, although the result is dependent on the model of sequence evolution used. Combined molecular and morphological analyses, such as Glenner et al. (2004) and
Eernisse and Peterson (2004), show a closer relationship between entoprocts and Cycliophora (and possibly
Syndermata)
Eernisse and Peterson (2004), Hausdorf et al. (2007), Wanninger et al. (2007), Dunn et al. (2008), Haszprunar and Wanninger (2008), Helmkampf
et al. (2008b)


II Syndermata (Rotifera and Acanthocephala)
Myzostomida + Acoela
+ Gnathostomulida, Gnathostomulida, Gnathostomulida
+ Micrognathozoa, Platyhelminthes, Lophotrochozoa, Micrognathozoa, within Ecdysozoa, Ectoprocta
Morphological evidence strongly favours a relationship of syndermates to gnathostomulids and Micrognathozoa. However, robust molecular evidence that unites these taxa (Gnathifera) to the exclusion of others is not currently available. Previous molecular or combined evidence analyses suggest a variety of
possible sister-group relationships. In recent phylogenomic studies, such as Dunn et al. (2008), Helmkampf et al. (2008b), and Marlétaz et al. (2008), Rotifera are very unstable (grouping alternatively within Lophotrochozoa or Ecdysozoa), and different molecular phylogenetic analyses support different
sister-group hypotheses
Eernisse and Peterson (2004), Halanych (2004), Hausdorf
et al. (2007), Passamaneck and Halanych (2006),
Todaro et al. (2006), Baguñà et al. (2008), Dunn et al. (2008), Helmkampf et al. (2008b), Marlétaz et al. (2008), Struck and
Fisse (2008)


II Micrognathozoa Rotifera, within Gnathifera, Cycliophora, Cycliophora + Gnathostomulida, Entoprocta

Morphological evidence firmly unites Limnognathia maerski with syndermates and gnathostomulids. However, molecular phylogenetic analyses are at the moment less conclusive (Giribet et al., 2004;
Todaro et al., 2006), supporting either a relationship with syndermates, gnathostomulids, cycliophorans, or entoprocts. They have not yet been included in phylogenomic studies

Giribet et al. (2004), Halanych (2004), Todaro et al. (2006)


II Gnathostomulida Acoela, Rotifera, within Gnathifera, Gastrotricha + Rotifera
+ Micrognathozoa +
Cycliophora

Morphological evidence strongly unites gnathostomulids with syndermates and Micrognathozoa. Labelled
as an unstable taxon in the phylogenomic analysis of Dunn et al. (2008), gnathostomulids have not yet been placed reliably in molecular analyses. We suspect that their placement as sister group to the acoels in
Dunn et al. (2008) is a systematic error due to long branch attraction

Eernisse and Peterson (2004), Halanych (2004), Todaro
et al. (2006), Dunn et al.
(2008)


Table 11.1 (Continued.)


Cat. Tax on Alternative sister groups
Remarks Recent references



II Cycliophora Entoprocta, Syndermata, Rotifera + Micrognathozoa, Micrognathozoa
The phylogenetic position of Cycliophora on the basis of both morphological and molecular evidence remains uncertain. They have not yet been included in a phylogenomic analysis
Eernisse and Peterson (2004), Giribet et al. (2004), Halanych (2004), Passamaneck and Halanych (2006), Todaro
et al. (2006)


III Hexactinellida Demospongiae sensu stricto, Demospongiae (Calcarea Eumetazoa), within Demospongiae

Although there is some 18S rDNA support for Hexactinellida representing the sister group to all other metazoans, recent molecular phylogenetic analyses based on several loci (Dohrmann et al., 2008) instead support the nesting of glass sponges within demosponges. An unpublished phylogenomic analysis based on more data, but fewer taxa (R. Derelle, doctoral thesis, 2007; M. Manuel pers. comm.) supports a
sister-group relationship of hexactinellids and Demospongiae sensu stricto (excluding homoscleromorphs)

Nichols (2005), Derelle (2007), Erpenbeck and Wörheide (2007), Dohrmann
et al. (2008)

III Echiura Annelida Available evidence now reliably places echiurans inside the annelids as possible sister group to Capitellidae Rouse and Pleijel (2007)

III Sipunculida Annelida Available molecular evidence suggests that sipunculids are sister group to, or part of the Annelida Eernisse and Peterson (2004), Hausdorf et al. (2007), Rouse and Pleijel (2007), Dunn
et al. (2008), Struck and
Fisse (2008)


III Phoronida Inarticulate brachiopods, Brachiopoda, Brachiopoda +
Nemertea
Although it is beyond doubt that phoronids are closely related to brachiopods, it is at the moment unclear whether phoronids fall within brachiopods as sister group to inarticulates (Cohen and Weydmann 2005), or whether they are separate lineages. Yet, in contrast, Helmkampf et al. (2008a) found a sister-group relationship between the single phoronid species and a species of ectoproct in their analyses, with brachiopods being more distantly related. This is in turn contradicted by Dunn et al. (2008) and Helmkampf et al. (2008b) which found support for brachiopods to be the sister group of phoronids,
but in Dunn et al. (2008) the position of phoronids is sensitive to method of phylogenetic analysis
Halanych (2004), Cohen and
Weydmann (2005), Dunn et al. (2008), Helmkampf et al. (2008a)


III Brachiopoda Phoronida, Nemertea,

Intriguingly, besides support for a connection to phoronids, the phylogenomic study by Dunn et al. (2008)

Passamaneck and Halanych

Ectoprocta + Phoronida + found some support for a close relationship between brachiopods and nemerteans, a relationship also
(2006), Todaro et al. (2006),

Nemertea + Mollusca +
Annelida
suggested in some previous analyses of ribosomal gene sequences (Passamaneck and Halanych 2006; Todaro et al. 2006). It should be noted that the phylogenetic position of brachiopods may change depending on the method of analysis for a given data set. This is obvious, for example, in the
studies of Passamaneck and Halanych (2006) and Dunn et al. (2008)
Dunn et al. (2008)

IV Myxozoa Falls within Cnidaria The status of myxozoans as parasitic and highly modified cnidarians is now robustly supported by a phylogenomic analysis. Previous suggestions of myxozoans being the sister group to Bilateria were the result of long branch attraction, but more studies are needed to establish with confidence whether myxozoans fall within Cnidaria or are sister group to Cnidaria
Jiménez-Guri et al. (2007), Evans et al. (2008)


IV Polypodium hydriforme

Sister to or
part of Hydrozoa

Combined 18S and 28S rDNA support the position of Polypodium within Cnidaria. However, due to
long branch attraction problems affecting the analyses, the precise relationship between myxozoans and
Polypodium remains unclear

Evans et al. (2008)


IV Vertebrata Urochordata In a remarkable reversal of received opinion a new interpretation of morphological evidence and a phylogenomic analysis yielded support for the sister-group relationship between Tunicata (Urochordata)
and Vertebrata (Craniata) (Ruppert, 2005; Delsuc et al., 2006). This previously heterodox hypothesis is now based on largely independent phylogenomic support (Dunn et al. 2008; Lartillot and Philippe 2008;
Putnam et al., 2008), and has rapidly gained general approval

Ruppert (2005), Delsuc et al. (2006), Dunn et al. (2008), Lartillot and Philippe (2008), Putnam et al. (2008), Swalla and Smith (2008)

IV Urochordata Vertebrata See under ‘Vertebrata’



IV Cephalochordata Urochordata +
Vertebrata
The monophyly of chordates is beyond doubt, and the sister-group hypothesis between cephalochordates and a clade of Urochordata and Vertebrata is now robustly supported
Ruppert (2005), Dunn et al. (2008), Lartillot and Philippe (2008), Putnam et al. (2008), Swalla and Smith (2008)

IV Echinodermata Hemichordata The sister-group relationship between echinoderms and hemichordates is robustly supported Ruppert (2005), Dunn et al. (2008), Lartillot and Philippe (2008), Swalla and Smith (2008)

IV Hemichordata Echinodermata See under ‘Echinodermata’



IV Xenoturbellida Echinodermata +
Hemichordata
In contrast to the phylogenomic analysis of Philippe et al. (2007), which suggested that Xenoturbella
was possibly the sister group of Acoela, more recent analyses support Xenoturbella as the sister group to
Ambulacraria (Echinodermata Hemichordata), together named Xenambulacraria (Bourlat et al., 2006)
Bourlat et al. (2006), Philippe et al. (2007), Dunn et al. (2008), Lartillot and Philippe (2008), Swalla and Smith (2008)


IV Nematoda Nematomorpha The emerging consensus on the sister-group relationship between nematodes and nematomorphs is now
also supported by phylogenomic analyses (Dunn et al., 2008; T. Juliusdottir, R. Jenner, M. Telford, R. Copley, unpublished data)

Eernisse and Peterson (2004), Halanych (2004), Mallatt and Giribet (2006), Dunn et al. (2008)






Table 11.1 (Continued.)


Cat. Tax on Alternative sister groups
Remarks Recent references

IV Nematomorpha Nematoda See under “Nematoda’


IV Priapulida Loricifera or
Kinorhyncha
As long as loriciferans are not yet included in phylogenomic analyses, the sister-group relationship between Priapulida and Kinorhyncha in such studies should be interpreted with caution. Morphological evidence allows no conclusive resolution, with either Priapulida or Kinorhyncha as the sister group to Loricifera. The phylogenomic analysis of Dunn et al. (2008) included kinorhynchs, which are resolved
as the sister group of priapulids
Eernisse and Peterson (2004), Halanych (2004), Dunn
et al. (2008)


IV Kinorhyncha Loricifera or
Priapulida
See under ‘Priapulida’


IV Loricifera Kinorhyncha or
Priapulida
See under ‘Priapulida’


IV Onychophora Arthropoda, Chelicerata

Although the close relationship between velvet worms and arthropods is uncontested, it is currently unclear exactly how they relate to each other on the basis of phylogenomic analyses. A sister-group relationship to either Arthropoda (Roeding et al. 2007; Dunn et al. 2008) or, surprisingly, Chelicerata (Roeding et al. 2007; Marlétaz et al. (2008) is supported. The latter hypothesis is also supported by a recent phylogenetic
analysis of neuroanatomical characters (Strausfeld et al. 2006), although most other morphological evidence has traditionally been interpreted as evidence for onychophorans and arthropods as separate
lineages

Strausfeld et al. (2006), Roeding et al. (2007), Dunn et al. (2008)

IV Arthropoda Onychophora See under ‘Onychophora’


IV Pogonophora and
Vestimentifera
Within Annelida The pogonophorans and vestimentiferans are now confidently placed within the polychaetes Rouse and Pleijel (2007), Rousset et al. (2007)

P R OB LE MA TI C A 125



Several factors need to be balanced to produce a good phylogenetic analysis: number of taxa, number of characters, quality of data, and qual- ity of analytical models. The interaction between these variables determines whether the results of a phylogenetic analysis are informative and reli- able, or suffer from stochastic or systematic error. Stochastic error arises as chance correspondences overwhelm true phylogenetic signal when there are not enough informative data. Systematic error results when reconstruction methods are inaccur- ate and are unable to deal with bias in the raw data, which can have several causes (Philippe and Telford, 2006). The common problem of long branch attraction (Anderson and Swofford, 2004; Waegele and Mayer, 2007) can be both a stochastic or a systematic error.
In trying to avoid stochastic error by increasing the number of characters in a data set, systematic errors may become increasingly prominent due to insufficient taxon sampling, uneven amounts of data across taxa, and a failure to detect non-phy- logenetic signals in the data. So far the molecular data generated for different phyla is wildly uneven (Figure 11.1) because of the bias towards key taxa that are important as model organisms, or organ- isms of biomedical or economic importance, or simply because they are the easiest to collect. To avoid systematic error it is therefore important to strive for a better balance in the number of taxa and characters (Philippe and Telford, 2006):

1. Avoiding stochastic error:
• Increase the number of characters. In molecu- lar systematics this is the main rationale for doing phylogenomics, based on large numbers of data generated through genome projects, EST projects, or large-scale projects targeting particular genes with degenerate primers (Delsuc et al., 2005; Philippe et al., 2005a; Philippe and Telford, 2006; Baurain et al., 2007). However, workers should be aware that uncritically concatenating information from different genes may cause systematic errors (Bapteste et al., 2008; Hartmann and Vision, 2008).

2. Avoiding systematic error:
• Develop better models of sequence evolu- tion that can deal with problematic data to prevent
systematic error (Delsuc et al., 2005; Philippe et al.,
2005b; Rokas and Carroll, 2006; Baurain et al.,
2007).
• Move towards less homoplastic characters such as rare genomic changes (Boore, 2006; Rokas and Carroll, 2006).
• Sample more taxa, including at least several species representing a high-level metazoan taxon, which may do more to prevent systematic error than aiming to have whole-genome sequences for fewer taxa (Hillis et al., 2003).
• Recognize and remove problematic data, such as fast-evolving taxa or characters, or characters the evolution of which violates phylogenetic model assumptions (Lecointre and Deleporte, 2004; Delsuc et al., 2005; Philippe et al., 2005b).

3. Other considerations:
• Care should be taken not to be misled by gene duplication (paralogy), causing gene trees to diverge from the species tree.
• Be aware that heuristic analyses can get caught in local optima, with different methods showing different degrees of sensitivity to this (Brinkmann and Philippe, 2008).
• To maximize the power to test the phylogen- etic position of a particular taxon, try to include at least all the taxa that have previously been pro- posed to be its closest relatives.
• If practical, reconstruct a phylogenetic scaf- fold based on a restricted number of taxa scored for many characters. Additional taxa can then be added sequentially on the basis of smaller numbers of characters (Wiens, 2006). The addition of incom- pletely known taxa can boost accuracy and confi- dence. To prevent systematic error it may be better to add a smaller number of characters scored for many taxa, rather than many characters for fewer taxa.
• If there is not enough phylogenetic signal,
focus on characters with higher rates of evolution.
• Assess data quality as a standard part of any
phylogenetic analysis (Brinkmann and Philippe,
2008; Waegele and Mayer, 2007).
• Exploit combined evidence analyses, where
possible including fossil data (Giribet, 2002;

126 AN I M AL EV O L UTI O N



Eernisse and Peterson, 2004), while recognizing the interpretational difficulties associated with combin- ing molecular exemplar species and inferred mor- phological ground patterns.
• Sample different genes that evolve at differ- ent rates to be able to resolve different regions of the tree (Sanderson and Shaffer, 2002; Glenner et al.,
2004; Philippe and Telford, 2006).
• Boost the number of descriptive and com- parative morphological studies to revise outdated received wisdom, and provide more data crucial for the inference of body plan evolution (Nielsen, 2001; Jenner, 2006b). Papers such as those by Wanninger et al. (2007) and Stach et al. (2008) are valuable in their contribution to phylogenetic debate and our understanding of body plan evolution.
• Adopt an experimental approach (sensitivity analysis) to phylogenetic analysis to see how results change depending on different assumptions.
• Re-evaluate contentious morphological evi- dence in the light of independent molecular phy- logenies, especially to detect cases of unrecognized character loss (Jenner, 2004c).
• Carefully construct morphological data sets to
maximize testing power (Jenner, 2004a).
• Adopt standardized methods for the presen- tation, annotation, and analysis of molecular data. To this end Leebens-Mack et al. (2006) have called for a standard for reporting on phylogenies, the MIAPA (minimum information about a phylogen- etic analysis), in which each component of a phylo- genetic analysis (alignment procedures, alignment,
sequences, voucher specimens, methods and parameters used, etc.) is outlined using universally accepted criteria. This will facilitate better evaluation and comparison of results of different analyses.

In summary, the recognition of Problematica reveals more than the sum of their missing or ambiguous parts. In avoiding fragmentary fos- sils or extant organisms with combinations of chimaeric or autapomorphic features, and by excluding long branching taxa or heavily biased nucleotide and protein sequences from molecular analyses, we may bring near completeness to data matrices and greater stability to our phylogenetic analyses, but probably at the expense of accuracy and an understanding of the full evolutionary pic- ture. Problematica reveal themselves as supremely important; for without their inclusion and accurate placement, other relationships are liable to change. In understanding how to deal with Problematica we understand the limits of systematics and our ability to have faith in our reconstructions of the tree of life.


11.5 Acknowledgements

RAJ gratefully acknowledges the UK Biotechnology and Biological Sciences Research Council for finan- cial support. RAJ thanks Dr M. Obst for his stimulat- ing thoughts on metazoan evolution shared during the scientific collecting expedition Pandalina V. We thank Drs Michaël Manuel and Katrine Worsaae for sharing unpublished information about their research.