Many cladistic analyses of animal phylogeny have been published by authors arguing that their results are well supported. Comparison of these analyses indicates that there can be as yet no general consensus about the evolution of the animal phyla. We show that the various cladistic studies published to date differ significantly in methods of character selection, character coding, scoring and weighting, ground-pattern reconstructions, and taxa selection. These methodological differences are seldom made explicit, which hinders comparison of different studies and makes it impossible to assess a particular phylogeny outside its own scope. The effects of these methodological differences must be considered before we can hope to reach a morphological reference framework needed for effective comparison and combination with the evidence obtained from molecular and developmental genetic studies.

One way to build larger, more comprehensive phylogenies is to combine the vast amount of phylogenetic information already available. We review the two main strategies for accomplishing this (combining raw data versus combining trees), but employ a relatively new variant of the latter: supertree construction. The utility of one supertree technique, matrix representation using parsimony analysis (MRP), is demonstrated by deriving a complete phylogeny for all 271 extant species of the Carnivora from 177 literature sources. Beyond providing a ‘consensus’ estimate of carnivore phylogeny, the tree also indicates taxa for which the relationships remain controversial (e.g. the red panda; within canids, felids, and hyaenids) or have not been studied in any great detail (e.g. herpestids, viverrids, and intrageneric relationships in the procyonids). Times of divergence throughout the tree were also estimated from 74 literature sources based on both fossil and molecular data. We use the phylogeny to show that some lineages within the Mustelinae and Canidae contain significantly more species than expected for their age, illustrating the tree's utility for studies of macroevolution. It will also provide a useful foundation for comparative and conservational studies involving the carnivores.

Tunicates are primitive chordates that develop a transient ‘tail’ in the larval stage that is generally interpreted as a rudimentary version of the vertebrate trunk. Not all tunicates have tails, however. The groups that lack them, salps and pyrosomes, instead have a trunk-like reproductive stolon located approximately where the tail would otherwise be. In salps, files of blastozooids are formed along the sides of the stolon. The tail and caudal trunk in more advanced chordates could have evolved from a stolon of this type, an idea referred to here as the ‘stolon hypothesis’. This means the vertebrate body could be a composite structure, since there is the potential for each somite to incorporate elements originally derived from a complete functional zooid. If indeed this has occurred, it should be reflected in some fashion in gene expression patterns in the vertebrate trunk. Selected morphological and molecular data are reviewed to show that they provide some circumstantial support for the stolon hypothesis. The case would be stronger if it could be demonstrated that salps and/or pyrosomes are ancestral to other tunicates. The molecular phylogenies so far available generally support the idea of a pelagic ancestor, but offer only limited guidance as to which of the surviving pelagic groups most closely resembles it. The principal testable prediction of the stolon hypothesis is that head structures (or their homologues) should be duplicated in series in the trunk in advanced chordates, and vice versa, i.e. trunk structures should occur in the head. The distribution of both rhabdomeric photoreceptors and nephridia in amphioxus conform with this prediction. Equally striking is the involvement of the Pax2 gene in the development of both the inner ear and nephric ducts in vertebrates. The stolon hypothesis would explain this as a consequence of the common origin of otic capsules and excretory ducts from atrial rudiments: from the paired rudiments of the parent oozooid in the case of the otic capsule (these express Pax2 according to recent ascidian data), and from tubular rudiments in the stolon in the case of the excretory ducts.