PUNCTUATED EQUILIBRIUM MEETS MOSAIC EVOLUTION: WHAT HAPPENS WHEN WE UNPACK TRAITS AND TRENDS?

While much of the last two decades of discussion on punctuated equilibrium has focused on processes that could influence patterns of morphological evolution, renewed research has highlighted the patterns themselves, using either comparisons of phenotypic variance with branch lengths or numbers of nodes on phylogenetic trees, or the application of statistical analysis and/or model selection criteria to distinguish static, gradual or random walk patterns within sequences of fossil populations. What these studies have in common is that the quantitative assessment of morphological change is mostly based on single traits, either size or shape. In many cases, where more than one trait was measured from a sequence, each trait was treated separately in the assessment of evolutionary change, implicitly assuming that single traits are adequate representations of species-level change. Here we examine a large dataset of 647 trait sequences across 159 fossil species lineages, using the same protocol of AIC model selection criteria on each sequence to test the validity of this assumption. We find that different evolutionary modes characterize different traits within most of the species lineages we analyzed. Moreover, the likelihood of conflicting patterns of evolutionary modes amongst traits increases as total number of traits increases. Even in lineages where the overall trend in morphology is dominated by a certain mode, there will be some traits that show other modes of change. More relevant to the issue of single traits as adequate representations of species-level change is how to determine the general tendency. Scientists will always be extrapolating from a small sample of available morphological (or genetic, or other) traits in order to characterize evolutionary modes of species lineages, despite the apparent ubiquity of mosaic evolution and the accompanying pitfalls our study reveals. A more nuanced understanding of evolutionary mode thus demands strong support from the convergence of evidential inferences; not just whether single traits adequately represent species’ evolutionary modes, but rather the consilience of evolutionary modes of traits and of total variation of species, of biological modularity and multilevel selection, and the seeming pervasiveness of mosaic patterns of morphological evolution.