lpetrich

For those not familiar with this term, this is a proposed superphylum that includes the molting animals, those that crawl out of their old outer skin as they grow.

It is divided into two main subgroups:

Panarthropoda (arthropods, onychophorans (look like caterpillars), and tardigrades)
Cycloneuralia (from how their brains surround their throats; nematodes, nematomorphs, kinorhynchs, priapulids, loriciferans (the latter three are obscure tiny ocean-floor worms))

Alongside of it is a big group called the Lophotrochozoa, which includes mollusks, annelids, brachiopods, bryozoans, flatworms, and several other phyla; and branching before those two groups is the long-recognized group Deuterostomia, which includes echinoderms and chordates.

Ecdysozoa was announced in (1) and has become widely accepted; the paper (2) reports that the Ecdysozoa grouping also shows up in some additional ribosomal-RNA molecules not used in the first study, and the paper (3) finds evidence of this grouping in Hox-gene patterns.

Now for some recent counterevidence. Or is it really counterevidence? The paper (4) by Blair, Ikeo, Gojobori, and Hedges reports on a test of Ecdysozoa using three completely-sequenced genomes, the only ones so far in the animal kingdom, it must be said. The genomes are from our species, the fruit fly Drosophila melanogaster, and the nematode Caenorhabditis elegans -- the latter two are the ecdysozoans here.

The team compared 100 genes with similar sequences from each, and found that the nematode branch was longer than the human and fly branches -- and longer in many of the genes tested. Thus they concluded that nematodes are more distant.

However, there are some difficulties with their work.

Rates of molecular evolution are known to be variable, even when gene function appears to be constant over time. One possible cause is variable gene-repair and gene-proofreading efficiency. So if Caenorhabditis elegans is a "fast-clock species", to use a common term in the molecular-evolution literature, this work would only rediscover that; some have, in fact, concluded that C. elegans is a fast-clock species by doing comparisons among more species, though with only a few genes.

And the lack of broader comparisons is, IMO, a major flaw of that Blair et al. paper. One can only distinguish different rates of evolution by including some additional species in the analysis, and they could have used some other species for which a complete genome is available: baker's yeast Saccharomyces cerevisiae and the flowering plant Arabidopsis thaliana. Using these species would be "rooting" the tree with the help of some "outgroups".

However, these are somewhat distant outgroups, and closer ones may be more worth using -- free-living animals with relatively simple anatomy like cnidarians (jellyfish, sea anemones, hydras, etc.) or poriferans (sea sponges). However, there are currently no genome-sequencing efforts for any species in these groups.


References:

(1) Nature 1997 May 29;387(6632):489-93
(2) Molecular Biology and Evolution 2002 Mar;19(3):289-301
(3) Nature 1999 Jun 24;399(6738):772-6
(4) BMC Evol Biol 2002 Apr 8;2(1):7

Here's a publicly-available site with lots of biomedical abstracts: <a href="http://www.ncbi.nlm.nih.gov:80/entrez/query.fcgi?CMD=&DB=PubMed" target="_blank">NIH PubMed</a>