Higher-level systematic analysis of birds: current problems and possible solutions

Avian systematics has a rich history, as evolutionary biologists have long been interested in this conspicuous and diverse group of vertebrates. Many prominent scientists, and evolutionary biologists in particular, have focused their efforts on birds. Perhaps no other group of vertebrates is so well studied. Yet, despite the attention paid to this group, much about the history of the class Aves remains controversial, both with respect to the origin of birds and the history since that origin. This puts avian systematists in a unique position, with so much information available and so many unanswered questions to pursue. The fact that avian ordinal relationships are still the center of much controversy speaks to the difficulty of the problem. While many prominent morphologists have worked on avian relationships, relatively few morphological studies have identified characters with informative variation for interordinal relationships. Molecular data offer the hope for phylogenetic information not present (or not discovered) in avian anatomy. Since the first study of avian proteins for the purposes of systematics (Sibley, 1960), several prominent molecular systematists have devoted tremendous time and resources to solving the problems of avian relationships using molecular characters (see Barrowclough, 1992 and Sheldon and Bledsoe, 1993 and references therein). So far, their efforts have not produced adequate resolution, at least not in the minds of most practicing systematists. Here, we first outline what we think we do know about higher order avian systematics and discuss some specific cases of molecular data applied to this question. Next, we consider some of the problems which may be blocking a clearer understanding of avian relationships. We then go on to offer some new directions for systematists working on this difficult group.