Cladistic Analysis Reveals Brainless Urbilateria

Abstract

sister groups by any stretch of the imagination. In a very elegant paper, Moroz [2009] reviews the literature, subjects the data to a correct cladistic analysis, and concludes that centralized nervous systems have evolved independently, perhaps as many as seven times. In a particularly striking example, he demonstrates that complex brains have evolved at least three times in molluscans alone. His analysis indicates that Urbilateria did not possess a tripartite brain, or any brain at all. Instead, these animals probably possessed an uncentralized nerve net. It is possible, however, that parts of this nerve net were already concentrated into dorsal, lateral and ventral elements, related to sensory, feeding and locomotory functions, respectively. Moroz appears to pose an evolutionary paradox, however. How can there be homologous genes and molecular mechanisms determining brains that are not homologous? He suggests that modular type molecular mechanisms arose early in metazoan evolution and may be involved in determining many different, complex cellular phenotypes. Hemichordates appear to be a clear example, where many evolutionarily conserved homeodomainrelated transcription factors and morphogens determine the main body axes system Both insects and mammals possess homologous genes and molecular mechanisms that are involved in global body patterning, including the patterning of the central nervous system [reviewed in Hirth and Reichert, 2007]. This has led to the hypothesis that the common ancestor of arthropods and chordates (Urbilateria) possessed a tripartite brain, in which Otx was expressed in the anterior part of the developing brain, and Hox genes in the posterior part, with a Pax -expressing domain positioned between them [Hirth and Reichert, 2007]. Although this hypothesis has been widely cited and has entered textbooks as dogma, it has never been credible, as it is based on an erroneous two-taxon paradigm, in which arthropods and mammals are interpreted as immediate sister groups. Although some strict cladists insist that two-taxon analyses are never valid, this is clearly not the case. When highly similar features are recognized in two immediate sister groups, a cladistic analysis does dictate that these features be interpreted as homologous. Exceptions must certainly exist (due to parallelism, for example), but at present there is no cladistic methodology for recognizing them. In any case, the important point here is not that arthropods and mammals comprise only two taxa, it is that they are not immediate Published online: September 30, 2010