Discussion of Phylogenetic Relationships
Figure 1: Summary of phylogenetic hypothesis in Rouse and Pleijel (2001). Alternative arrangements are discussed in the Discussion of Phylogenetic Relationships. The position of Clitellata/Oligochaeta (includes earthworms and leeches) and Echiura is yet to be resolved.
The most recent comprehensive systematization of polychaetes, that proposed by Rouse and Fauchald (1997) from their cladistic parsimony analyses, has been used here (Fig. 1) with some alterations (see Rouse and Pleijel, 2001). Allowing for the likely errors in the placement of many taxa, and the fact that there were conflicting results included in the original analyses, the most fundamental problem inherent in the systematization used here may be that of the placement of the root for any tree of Annelida.
Basal annelids, according to Rouse and Fauchald (1997), are taxa such as Clitellata/Oligochaeta and simple-bodied forms within Scolecida. Their trees also excludes Echiura from Annelida. This result was based on outgroup choices such as Mollusca and Sipuncula, and may well be misleading. Alternative hypotheses are therefore worth outlining, though they do not follow normal cladistic practice. Storch (1968), following a detailed study on the musculature of Annelida, proposed that scale-worms, a diverse clade within Phyllodocida, are representative of the plesiomorphic condition for Annelida. He suggested that there was a radiation from this group, but that Chrysopetalidae were most closely related to scale-worms. The implication of his hypothesis is that Phyllodocida represents a paraphyletic group, from which all other polychaete taxa arise. Westheide (1997) (and see Westheide et al., 1999) suggests that the basic (i.e., plesiomorphic) 'body plan' of Annelida comprises features meaning that the root of the Annelida tree would be placed with taxa from Aciculata. This would either result in a paraphyletic Phyllodocida or Amphinomida, depending on which taxon is used as the root (Fig. 4).
From a molecular perspective, McHugh (1997) found Clitellata, Pogonophora (= Siboglinidae) and Echiura nested among various polychaetes using sequence data from elongation factor 1 alpha in a parsimony analysis (Fig. 4). The placement by McHugh (1997) of Pogonophora as a polychaete group was congruent with Rouse and Fauchald’s (1997) results, but those of Clitellata and Echiura were markedly different. The conflict between this molecular sequence data and the morphological results could be caused by several factors. One possibility is that Clitellata/Oligochaeta and Echiura have lost a number of morphological features that would help identify their sister group among polychaetes. Further morphological study, combined with sequence data, may uncover these 'losses' (see Hessling and Westheide, 2002). However, the molecular sequence data sets assembled to date have been marked by both a limited number of taxa and characters. An exception is Brown et al. (1999), where DNA sequence data from three nuclear genes across a wide taxonomic diversity of annelids was subject to parsimony analysis. They recovered some morphological groupings such as Cirratulidae, Terebellidae and Eunicida, but did not show a monophyletic Phyllodocida or Aciculata, nor did they find any parts of these taxa to be basal groups of Annelida. However, some expected groupings were not recovered. Martin (2001) found that the placement of Clitellata/Oligochaeta among polychaetes could not be resolved, and he could not recover a monophyletic Annelida owing to the placement of taxa such as Mollusca and Sipuncula. None of the major taxa used here, such as Palpata, Aciculata, Phyllodocida, Canalipalpata, Sabellida or Terebellida were recovered in Martin's (2001) analysis. Also less diverse taxa such as Nereididae, Spionidae and Aphroditiformia were not recovered.
Does the fossil record help?
In a review of the fossil record of annelids Rouse and Pleijel (2001) suggested that the oldest unequivocal fossil polychaetes, such as Canadia from the Cambrian, belong within Phyllodocida. Subsequent fossil polychaetes that can be confidently placed outside Phyllodocida do not appear until the Carboniferous. No other fossil polychaetes from the Cambrian can be unequivocally assigned to extant polychaete taxa. There are several likely appearances from the Ordovician, including Serpulidae, Spionidae and the radiation of Eunicida. Ensuing appearances suggest that by the end of the Carboniferous most major polychaetes lineages had appeared. The exception appears to be Scolecida, with the earliest known fossils being the dubious Archarenicola (Arenicolidae) from the Triassic, and one assignable to Paraonidae from the Cretaceous. With the rooting option employed in Figure 1, it appears that some of the earliest appearing fossil polychaetes belong to derived clades (e.g., Eunicida and Phyllodocida). This could be interpreted in two ways: (1) the root placement in Figure 1 is wrong, and so Aciculata, comprised of Amphinomida, Eunicida and Phyllodocida, may in fact represent a paraphyletic 'stem' group for the rest of polychaetes; (2) a number of major polychaete clades had already evolved in, or before, the 'Cambrian explosion', but fossils have not yet been found. The third possibility is that the overall topology used in Figure 1 may be profoundly incorrect. If we accept that the basic topology shown in Figure 1 is correct, but do not root the tree, then a diagram as shown in Figure 5 is the result. This may represent the most conservative representation of our understanding of annelid relationships.
No one has provided updates yet.