ABSTRACT
Stochastic computer simulation is an important method for comparing the evolutionary patterns and processes associated with radically different intervals of time. This paper demonstrates how to simulate the evolution of complex morphologies over geological timescales of millions of generations. The simulations are used to test how various assumptions about microevolutionary parameters and processes manifest themselves on macroevolutionary timescales. Complex morphology is modelled using geometric representations of shape (e.g., landmarks or outlines), and so the procedure described here is limited to single rigid structures. The procedure is based on empirically measured phenotypic correlations, which constrain the evolutionary outcomes in biologically realistic ways. Different microevolutionary assumptions about covariances, population size, and evolutionary mode can be tested by incorporating them into the simulation parameters.
The evolution of molar tooth morphology in shrews is simulated under four different evolutionary modes: (1) randomly fluctuating selection; (2) directional selection; (3) stabilizing selection; and (4) genetic drift. Each of these modes leaves a distinctive imprint on the distribution of morphological distances, a feature that can be used to reconstruct the mode from real comparative data. A comparison of the results with real data on shrew molar diversity suggests that teeth have evolved predominantly by randomly fluctuating selection. The rate of divergence in shrew molars is greater than expected under drift, but it is neither linear nor static as expected with directional or stabilizing selection.
The evolution of morphology with randomly fluctuating selection is also simulated on a phylogenetic tree. Daughter species share derived morphologies and positions within the principal components spaces in which the simulation is run. This result suggests that phylogeny can be successfully reconstructed from multivariate morphometric data when organisms have evolved under any mode except strong stabilizing selection.
P. David Polly. School of Biological Sciences, Queen Mary, University of London, London E1 4NS, U.K.
KEY WORDS: evolutionary rates, genetic drift, molar teeth, morphological evolution, phenotypic covariances, selection, Sorex (Soricidae, Lipotyphla, Mammalia), stasis, shrew
PE Article Number: 7.2.7A
Copyright: Society of Vertebrate Paleontology December 2004
Submission: 23 August 2004 Acceptance: 27 November
2004
Whitey Hagadorn was sole Executive Editor for this article.
