Searching for Deep Variation in the Model Systems of Evo-devo

Evolution by natural selection requires variations that (a) exhibit fitness differences, and (b) are heritable. We use these criteria to examine what McDonald called a “great Darwinian paradox,” namely, that there is little evidence for relevant variation at “those regulatory loci set deep with the control network” (1983, 92-3) of metazoan ontogenies. Currently, the absence of heritable variation in what Wimsatt calls the “deeply-entrenched” features of ontogeny has led many investigators to argue that profound temporal asymmetries exist in evolutionary history; in short, things were different in the pre-Cambrian. Yet it is unclear how ontogenetic architectures early in metazoan history could have differed fundamentally from present-day systems. We have commenced a survey of the literature of the principal model systems of evo-devo, as well as lesser-known systems, looking for evidence of heritable mutations in such features as cleavage patterns, axis specification, or modes of gastrulation. The continued absence of such evidence suggests that it may be time to re-examine basic assumptions. THE PROBLEM A. There are striking differences in the early development in animals, even within classes and orders. B. Assuming that these animals are descended from a common ancestor, these divergences suggest that early development evolves relatively easily. C. Evolution by natural selection requires heritable variation. D. Heritable variations in early development are not observed.* E. Therefore, we may be justified in questioning basic assumptions.