Uncertainty in the reconstruction of ancestral character states and limitations on the use of phylogenetic comparative methods.

Phylogenetic approaches to the study of comparative data have a long history in the field of animal behaviour (e.g. Lorenz 1941; Greene & Burghardt 1978; Gittleman 1981). However, since the publication in 1991 of Brooks & McLennan’s Phylogeny, Ecology and Behavior and Harvey & Pagel’s The Comparative Method in Evolutionary Biology, there has been a huge increase in the use of phylogenetic methods in comparative studies. Indeed, a comparative study would be unpublishable in most journals today if it did not incorporate phylogenetic methods of analysis, or at least a discussion of phylogenetic relationships. Welcome as this development is, the time has come for a critical reappraisal of phylogenetic methods for the analysis of comparative data (Ricklefs & Starck 1996; Price 1997; Cunningham et al. 1998). A variety of methods have now been published and one might wonder how much an analysis is affected by the choice of method. More generally, given that the use of phylogenetic methods now appears to be standard practice, one might also wonder whether situations exist in which phylogenetic methods are unnecessary or even inappropriate.