STAYTON, C. TRISTAN; Bucknell University: Is convergence surprising? A simulation study using a Brownian motion model of evolution among quantitative characters

Convergence, the independent evolution of similar phenotypes among distantly related species, is often taken as evidence of adaptation in response to similar selective pressures. However, some degree of convergence is expected given a large enough sampling of species; this is true even if evolution in quantitative character space is random. The amount of convergence expected for a given data set under such random evolutionary processes is usually unknown, as simulation studies of convergence are rare. Thus, researchers currently lack an appropriate null model for evaluating the significance of convergent patterns, or even a baseline for determining whether their study system might show a �surprising� (i.e., greater than expected) amount of convergence. In this exploratory study, I develop an operational metric of convergence, and explore the average amount of convergence that is produced in randomly generated (Brownian motion model) phylogenies. Monte Carlo simulations are used to assess the effects of phylogeny size and balance, as well as number of variables and variance structure within the data set, on the expected amount of convergence. In comparing these randomly-generated data sets to a real data set consisting of shape measurements on a sample of lizard skulls (including a case of convergence between among herbivorous lizards), I find that large amounts of convergence, equal to those observed in empirical data sets, can be generated by random evolution in character space.