The Assumptions and Challenges of Ancestral State Reconstructions

Ancestral state reconstruction is at the center of a tremendous resurgence in inter-est in comparative methods in evolution-ary biology. Also known as character opti-mization or character mapping, this set of methods is responsible for getting many nonsystematists to think about phylogeny and evolutionary history. This recent ood of interest can be traced to a series of im-portant papers (e.g., Felsenstein, 1985; Cod-dington, 1988; Donoghue, 1989), two inu-ential books (Brooks and McLennan, 1991; Harvey and Pagel, 1991), and the computer program MacClade (Maddison and Mad-dison, 1992). But although ancestral recon-structions have been embraced by both sys-tematists and evolutionary biologists, the reliability and assumptions of these meth-ods need continued examination. An in-creasing number of authors have discussed the assumptions and challenges of charac