Aeolosoma major

Rouse, Greg W. & Pleijel, Fredrik, 2007, Annelida *, Zootaxa 1668, pp. 245-264 : 250-255

Major polychaete taxa

Scolecida. The name Scolecida, derived from skolex (Greek), meaning worm, was applied with reference to an annelid group by Rouse & Fauchald (1997), though it has historically been used with reference to parasitic non-annelids. Taxa included were Arenicolidae , Capitellidae ( Fig. 1 C inset), Cossuridae , Maldanidae ( Fig. 2 E), Opheliidae ( Fig. 2 D), Orbiniidae ( Fig. 2 C) and Paraonidae . The name Scolecida is derived from Scoleciformia, a name introduced by Benham (1896) for a similar group of taxa. Only two apomorphies support the clade Scolecida in Rouse & Fauchald (1997), the presence of parapodia with similar rami and the possession of two or more pairs of pygidial cirri, and both of these are homoplastic. In many ways, this group represents the simple-bodied forms of polychaetes and it is likely that further analysis will show that it is not monophyletic. In fact to date no molecular sequence analyses have recovered any assemblage like Scolecida and it usually appears as a polyphyletic assemblage (e.g., Struck et al. 2007).

Palpata. Palpata was a new name coined by Rouse & Fauchald (1997) and such a group of polychaetes has never been formulated before. Virtually all non-Scolecida polychaetes, except a few incertae sedis taxa, were placed into Palpata. The name is based on an apomorphy for the group, the presence of palps. Palps can be divided into two structurally different groups, grooved “feeding” palps and ventral, “sensory” palps ( Orrhage 1980). Feeding palps usually have ciliated paths, often located in a longitudinal groove, giving each palp a Ushaped cross-section. Ventral sensory palps are morphologically more uniform than grooved palps. In most cases they are tapering or digitiform and relatively short, compared to grooved palps. Given the fact that the presence of palps and a limited peristomium were the only synapomorphies supporting this taxon in Rouse & Fauchald (1997), the validity of Palpata requires further investigation and it has yet to be recovered as a monophyletic group in any broad molecular analyses (see below). Rouse & Fauchald (1997) proposed that Palpata contains two major taxa, Aciculata and Canalipalpata .

Aciculata. Aciculata was a new name erected by Rouse & Fauchald (1997) for one of the most the strongly supported clades in their analyses. The name refers to one of the apomorphies for the group; the presence of a particular chaetal type called aciculae. Aciculae are stout chaetae that differ from others in that much or all of the chaeta remains internalized in the parapodium. Numerous other features such as the presence of ventral sensory palps, prostomial antennae, dorsal cirri, ventral cirri, one pair of pygidial cirri, and segmental organs in most segments were proposed to be apomorphic for Aciculata. Aciculata is divided here into the three major clades Amphinomida , Eunicida and Phyllodocida , and also with a few taxa as incertae sedis. Aciculata was recovered as a clade by Struck et al. (2007), if Orbiniidae (usually placed in Scolecida, although some members actually have chaetae that can be interpreted as aciculae) was included.

Phyllodocida . The name Phyllodocida was coined by Dales (1962) and was subsequently used by other workers, such as Fauchald (1977) and Pettibone (1982), for essentially the same group of polychaetes. Prior to Dales (1962), the group had never been formulated as a monophyletic taxon. Rouse & Fauchald (1997) found strong support for the monophyly of this assemblage, indicated by the ventral position of sensory palps, the presence of anterior enlarged cirri, the loss of dorsolateral folds, the presence of an axial muscular proboscis, and the presence of compound chaetae with a single ligament. A major difference in the formulation of Phyllodocida by Rouse & Fauchald (1997), and that of previous workers, was that they placed Myzostomida ( Fig. 1 D, E) (as Myzostomatidae ) in the group (but see Eeckhaut et al., 2000). Rouse & Fauchald (1997) did not further subdivide Phyllodocida , since their analyses resulted in dramatically different topologies for various Phyllodocida taxa. Based on the results of Pleijel & Dahlgren (1998), two names were used by Rouse & Pleijel (2001) for clades within Phyllodocida ; Aphroditiformia (or scale-worms) and Nereidiformia. The remaining taxa in Phyllodocida are treated as “unplaced”. There are numerous family-ranked taxa in Phyllodocida with representatives shown here of Phyllodocidae ( Fig. 3 A), Sphaerodoridae ( Fig. 3 B), Nereididae ( Fig. 3 C), Syllidae ( Fig. 3 D), Chrysopetalidae ( Fig. 4 A), Nephtyidae ( Fig. 4 B), and Polynoidae ( Fig. 4 E).

Eunicida . The name Eunicida was first used by Dales (1962), but encompassed a well-delineated group known prior to that as Eunicea, Euniceformia, Eunicimorpha or simply Eunicidae . This grouping included polychaetes with a ventral muscularized pharynx with ventral mandibles and dorsal maxillae (and a few that had lost jaws) and is very likely a monophyletic assemblage. Another putative apomorphy for the clade would appear to be a peristomium forming a two rings, although this feature is not present in all the subgroups. Rouse & Fauchald (1997) expanded the traditionally delineated Eunicida to include Amphinomidae and Euphrosinidae although this is not accepted here as the relationships between them is uncertain. Eunicida , following Rouse & Pleijel (2001), includes Dorvilleidae ( Fig. 4 C), Eunicidae , Hartmaniellidae , Histriobdellidae , Lumbrineridae , Oenonidae and Onuphidae . A recent comprehensive molecular phylogeny of Eunicida was undertaken by Struck et al. (2006) who found a topology where a labidognath clade (i.e., a “ Eunicidae ”/ Onuphidae / Lumbrineridae clade) was significantly rejected, a result that goes against most traditional view of the relationships within the group. Prior to this Struck et al. (2005), in a molecular study suggested that Dinophilidae ( Fig. 1 A)—a group long associated with Eunicida , either as a family or as part of Dorvilleidae ( Eibye-Jacobsen & Kristensen 1994) —does not belong in Eunicida .

Amphinomida . Amphinomida contains Amphinomidae ( Fig. 4 D) and Euphrosinidae and has never been the subject of a detailed modern phylogenetic revision. A taxon, Amphinomae, equivalent to Amphinomida , dates back to Savigny (1822). Amphinomida was first used as a taxon name by Dales (1962), but names such as Amphinomorpha have also been applied. Although Amphinomida is well supported by a number of apomorphies, including calcareous chaetae and nuchal organs forming a caruncle, both their position within Aciculata and the interrelationships within the group are uncertain. Rouse & Fauchald (1997) in their classification included Amphinomidae and Euphrosinidae in Eunicida , whereas Rouse & Pleijel (2001) instead treated them as a separate taxon, Amphinomida .

Aciculata unplaced. Rouse & Fauchald (1997) placed Aberranta Hartman, 1965 (as Aberrantidae ), Nerillidae ( Fig. 1 H) and Spinther Johnston, 1845 (as Spintheridae ) as parts of Aciculata, but did not include them in any other sub-taxon of the group. These three taxa have never been related to each other and are not to be regarded as closely related by being grouped together here. Of these taxa, only Spinther actually has aciculae, but all three are likely to have sister groups with different members of Aciculata. Worsaae et al. (2005) assessed the phylogenetic position of Nerillidae and Aberranta using morphology and 18S rDNA, but failed to provide any strong conclusions as to the placement for either group.

Canalipalpata . Canalipalpata (referring to the presence of grooved palps) is a name that was first used by Rouse & Fauchald (1997) and is a massive group of polychaetes that encompasses around half the number of described species. The only apomorphy for Canalipalpata is the presence of grooved palps and so the clade must be regarded as weakly supported. The groove along each palp is longitudinal and ciliated and, in contrast to those of Aciculata, function as feeding structures. Rouse & Fauchald (1997) identified three major clades within the Canalipalpata ( Sabellida , Spionida and Terebellida ) and a number of taxa that were regarded as incertae sedis within the group.

Sabellida . The apomorphy for Sabellida , as formulated by Rouse & Fauchald (1997), was the fusion of the prostomium with the peristomium, hence it is weakly supported. This name has been used to contain Sabellariidae ( Fig. 5 I), Sabellidae ( Fig. 5 C) and Serpulidae ( Fitzhugh, 1989) and these three were included by Rouse & Fauchald (1997), plus two other taxa. A dramatic shift compared with traditional systematics was that Rouse & Fauchald (1997) placed Siboglinidae ( Fig. 5 A), (formerly outside polychaetes as Pogonophora and Vestimentifera), as part of Sabellida . They also included Oweniidae ( Fig. 5 B), a taxon that has a “chequered” systematic history as a polychaete. It should be noted that in further analyses by Rouse (1999; 2000), Oweniidae did not group with the remaining Sabellida , whereas Chaetopteridae ( Fig. 5 F) (included here in as part of Spionida ) did. A recent combined analysis of molecular and morphological data ( Rousset et al. 2004) also did not recover the Sabellida as formulated here but did suggest Oweniidae and Siboglinidae to be closely related, as did Struck et al. (2007). Clearly further study is required. In addition, similarities between Sabellariidae ( Fig. 5 I) and Pectinariidae (in Terebellida ) also deserve further study ( Rouse & Pleijel 2001). Sabellariidae have previously been considered as part of Terebellida (e.g., Fauchald 1977).

Terebellida (Cirratuliformia, Terebelliformia) . The name Terebellida was first used by Dales (1962) and included Ampharetidae , Pectinariidae and Terebellidae , all polychaetes having multiple grooved palps. Prior to this, the names Terebellomorpha or Terebelliformia had often been used for essentially the same grouping of taxa. Terebellida was expanded by Rouse & Fauchald (1997) to include a clade in which most have a single pair of palps (e.g., Acrocirridae , Cirratulidae , Flabelligeridae ), but recent studies have not recovered this relationship (see below). Rouse & Fauchald (1997) identified several synapomorphies for this overall grouping, namely the presence of a first segment with no chaetae, a gular membrane, and a heart body. The clade within Terebellida that has taxa with a single pair of palps (with exceptions such as some Cirratulidae with numerous palps and Ctenodrilinae and Fauveliopsidae with none) was referred to as Cirratuliformia in Rouse & Pleijel (2001), a name that has been used previously for a somewhat similar grouping (e.g., Fauchald 1977). It contains Acrocirridae , Cirratulidae , Fauveliopsidae , Flabelligeridae , and Sternaspis Otto, 1821 . The clade comprised of Alvinellidae , Ampharetidae , Pectinariidae , Terebellidae and Trichobranchidae is referred to here as Terebelliformia, also a fairly “traditional” formulation.

Spionida . The name Spionida was first used by Dales (1962) to contain Spionidae and a number of similar groups, as well as taxa such as Paraonidae and Sabellariidae . These groups are not considered closely related today, and Spionida in Rouse & Pleijel (2001) contained Apistobranchus Levinsen, 1883 , Chaetopteridae , Magelonidae , Heterospio Ehlers, 1875 , Poecilochaetus Claparède in Ehlers, 1875, Trochochaeta Levinsen, 1883 , and Uncispionidae , though the latter four taxa should reasonably be all treated as members of Spionidae ( Blake & Arnofsky 1999) . The positions of Chaetopteridae and Magelonidae deserve further investigation. The synapomorphies for Spionida listed by Rouse & Fauchald (1997) were the presence of a pair of peristomial grooved palps, nuchal organs forming posterior projections, and anterior excretory nephridia and posterior segmental organs for gamete release.

Canalipalpata unplaced. Rouse & Fauchald (1997) placed Polygordiidae and Protodrilida (as Protodriloididae , Protodrilidae and Saccocirridae ) as part of Canalipalpata , but did not place them within any other subtaxon. In their complete cladistic analyses, these taxa formed a clade that was either associated with taxa that belong within Canalipalpata, Scolecida or were part of a basal polytomy of Polychaeta, so their decision was arbitrary. However, there is good evidence to support Protodrilida being placed in Canalipalpata ( Purschke & Jouin 1988) , perhaps near Spionida . On the other hand, placement of Polygordiidae in Canalipalpata must be regarded as suspect. Rouse & Fauchald (1997) made scoring errors in regard to Polygordiidae , the most important being that the animals cannot be regarded as having grooved palps. It was suggested by Rouse & Pleijel (2001) that investigation of a sister group relationship with, or within, Opheliidae may be worthwhile, and this is actually an old idea (e.g., Giard 1880).

Annelida unplaced. Rouse & Fauchald (1997) included Aeolosomatidae ( Fig. 1 B), Potamodrilus Lastockin, 1935 (as Potamodrilidae ), Parergodrilidae and Psammodrilidae in their complete taxon set analyses. In one analysis, these taxa either fell as part of a large basal polytomy of polychaetes, or Aeolosomatidae , Potamodrilus and Parergodrilidae formed a clade with Ctenodrilinae , and Psammodrilidae grouped with Capitellidae ( Fig. 1 C inset), Arenicolidae and Maldanidae ( Fig. 2 E). In another analysis, Aeolosomatidae , Potamodrilus and Parergodrilidae either formed a grade with respect to Polygordiidae ( Fig. 1 F), Protodrilidae ( Fig. 1 G) and Protodriloididae , or they were a basal clade of polychaetes. Psammodrilidae either were a basal polychaete group, or were sister group to a large clade that was mainly comprised of taxa with grooved palps. Thus, it was not possible to place these taxa with any confidence. Prior to the study by Rouse & Fauchald (1997), Aeolosomatidae ( Fig. 1 B) and Potamodrilus were usually considered in relation to Clitellata, either as a member of the group ( Bunke 1967), or more recently as the sister group to Clitellata ( Bunke 1985). This was then rejected by Bunke (1986), who also could not relate them to any group of polychaetes, thus leaving the two taxa “isolated”. The most recent analysis is from a molecular perspective ( Struck et al. 2002), and this study found no support for a relationship of Aeolosomatidae with Clitellata or with any other particular annelid group. Fauchald (1977) grouped Parergodrilidae with Ctenodrilidae , but did not justify this decision. Otherwise, the group has also been treated as “isolated” within polychaetes. A recent molecular study by Jördens et al. (2004) indicated that Parergodrilidae may be closely related to orbiniids; the position of the terrestrial polychaete Hrabeiella Pizl & Chalupsky, 1984 , however, was inconclusive. Psammodrilidae have been treated as a singular group of polychaetes since they were first discovered by Swedmark (1952). Rouse & Fauchald (1997) suggested that a relationship for Psammodrilidae with Arenicolidae and Maldanidae should be assessed; this was proposed by Meyer & Bartolomaeus (1996; 1997), but the present anatomical evidence is weak.

Figures

FIGURE 2. — Clitellata and Scolecida. A. Lumbricus terrestris (Clitellata, Lumbricidae) from USA, dorsal view of entire specimen. — B. Macrobdella decora Say, 1824 (Clitellata, Hirudinidae) from USA, dorsal view of entire specimen. — C. Scoloplos arimiger (O. F. Müller, 1776) (Scolecida, Orbiniidae) from Sweden, whole specimen. — D. Euzonus mucronatus (Treadwell, 1914) (Scolecida, Opheliidae) from California, whole specimen. — E. Clymenella sp. (Scolecida, Orbiniidae) from Belize. All photos G. Rouse, except 2 B. courtesy of Mark Siddall.