Although there are three large phyla of segmented animals, most of what we know about the evolution and development of a segmented body plan comes from just two of these: the arthropods and the chordates. The third “segmented phylum,” the annelids, is usually thought of as the poor cousin, and its representation in the evo-devo world has been comparatively low.
Recent work has served to strengthen what annelid researchers have always known. Segmentation in Annelida is interesting not only because this is “the third segmented phylum,” but also because it raises many exciting questions in its own right. Struck et al. (2007)—a multiresearcher consortium belonging to the NSF-funded “WormNet”—present the most thorough and best-analyzed molecular phylogeny of annelids to date. According to their analysis, the segmented annelids harbor within their ranks two groups, previously awarded phylum status, that have lost all signs of external segmentation. The Echiura (spoon worms) and the Sipuncula (peanut worms) are two groups of unsegmented burrowing marine worms that have been suspected of having annelid affinities for several years (McHugh 1997; Boore and Staton 2002; Halanych et al. 2002; Hessling and Westheide 2002). The Echiura show a segmental development of the central nervous system (Hessling and Westheide 2002), but there is no similar evidence for the Sipuncula. Their inclusion, deep within annelid phylogeny, strongly suggests that they have independently lost segmentation.
Bleidorn (2007), one of the authors of the new molecular phylogeny, argues that character loss, in general, is a central factor to be considered in any phylogenetic reconstruction of annelids. Character loss is known to be a significant confounding element in morphological cladistic analysis (Porter and Crandall 2003; Jenner 2004), and its prevalence in annelids should come as no surprise. What might come as a surprise is that it is segmentation that is lost (and twice at that!). Segmentation is usually thought of as such a fundamental feature of body plan organization, that it is used as a defining character at the phylum level. The fact that within annelids, segmentation is evolutionarily labile suggests that there is something very different happening with segmentation there, relative to arthropods or chordates. One would be hard pressed to find any example of segmentation loss in either of the two other phyla (a notable exception being the parasitic crustacean Sacculina that loses all traces of segmentation for part of its life cycle—Hoeg and Lutzen 1995).
There is ongoing work on segmentation in annelids that is starting to yield interesting results, though this is still limited mainly to a small number of model species (Shimizu et al. 2001; Prud'homme et al. 2003; Fischer and Dorresteijn 2004; Weisblat et al. 2004; Seaver et al. 2005; Seaver and Kaneshige 2006). Those working on the emerging model annelid Platynereis dumerilii are touting their favorite animal as being a generalized bilaterian and having an “ancestral body plan,” and therefore being best suited for studies on the evolution of animal body plans at the highest taxonomic levels (Tessmar-Raible and Arendt 2003; Fischer and Dorresteijn 2004). Regardless of whether this statement is true (or not: Jenner 2006) it ignores the more exciting aspects of studying annelid segmentation. There is clearly much more data on the diversity of segmentation patterns and mechanisms waiting to be mined from this fascinating group: perhaps much more than in their two “richer” cousins.
