Measures of stratigraphic fit to phylogeny and their sensitivity to tree size, tree shape, and scale

Abstract

Measures of stratigraphic fit to phylogeny are analyzed to test how they are affected by the shape and size of the phylogenetic trees and by the number of stratigraphic intervals encompassed. Monte Carlo randomizations are used to investigate the sensitivity of three commonly used measures (SCI, GER and MSM*) approximating their distribution of possible values under certain conditions. All are shown to vary in different ways as parameters are varied, although MSM* seems to be the most invariant in the analyzed parameter space. These results suggest that the raw metrics should not be used for comparing the fit of different taxonomic groups or competing phylogenetic trees of the same group that differ in tree size or balance. Tree balance also affects the distributions used in significance tests based on randomization and therefore their results should not be interpreted in terms of the amount of conflict implied by a phylogenetic tree. © The Willi Hennig Society 2004.