Bradley C. Livezey
Section of Birds
Carnegie Museum of Natural History
Phone: 412.622.5591, email: livezeyb@CarnegieMNH.org

Avian systematics remains diverse in approach but deficient in empirical investigation of higher-order relationships. DNA hybridization studies of birds, formerly considered to hold great promise for phylogenetic inference, have been criticized on a number of theoretical and practical levels. At present, most molecular systematists now employ direct sequencing of DNA, typically mitochondrial cytochrome b and nuclear genes. Sequence-based studies—showing great variation in taxonomic representation, scale of analysis, success in compiling data comparable across ingroup taxa, and competence in phylogenetic methodology—include members of several taxonomic orders. In the few cases where comparable analyses have been attempted, there is considerable congruence between morphological and molecular reconstructions of phylogeny, although variation among molecular studies is surprisingly high.

Many ornithologists, however, have been convinced that the utility of comparative morphology for the reconstruction of avian phylogeny had been eclipsed by molecular studies. This situation is surprising in light of the published track record of morphological and molecular cladistics in ornithology during the last ten years, a fair assessment of which would place the former quite favorably with the latter. In reality, workers from both morphological and molecular perspectives are participants in the same revolution in systematics—phylogenetic analysis—as an alternative to the intuitive classifications of the past.

A comprehensive morphological analysis is needed to reassess the higher-order placements of a number of critical and problematic, fossil and modern avian groups. In collaboration with R. L. Zusi of the Smithsonian Institution, I recently completed a ten-year, phylogenetic assessment of the major taxonomic groups of birds based on a cladistic analysis of comparative anatomy, including all modern nonpasseriform families and a number of fossil groups (e.g., Archaeopteryx, Hesperornis, and newly discovered taxa from China and Spain), to be rooted using non-avian theropod dinosaurs. The anatomical comparisons will include myology, osteology of the skull, trunk, and limbs, as well as selected characters of the trachea and integument. The total number of taxa to be included number approximately 175, and almost 3,000 informative characters were used. Data are compiled in Bulletin of Carnegie Museum of Natural History 37: 1-544 (with CD), and the analysis in Zoological Journal of Linnean Society 149: 1-94.

Figure 3: Pencil illustration of the skull of a Magellanic Flightless Steamer-Duck (Tachyeres pteneres), indicative of the structural details to be found in the avian skull. Ongoing analysis indicates that almost 1,000 characters informative for the reconstruction of avian relationships can be taken from the skull alone (including conformation of sutures, foramina, and individual bones), with hundreds more to be compiled from the postcranial skeleton, musculature, and integument.