Many studies in the early 1980s and afterwards (Alberch 1980, Goodwin 1994, Newman and Müller 2000) have argued that, even if natural selection cannot always act on how the phenotypes are produced (Mayr 1982), understanding developmental dynamics is important because it determines which kind of morphological variation can appear from genetic variation (question 1). This is normally stated as development constraining (Maynard Smith et al. 1985) or biasing morphological evolution. These concepts are devised to contrast with the previous view in which any small morphological variation is possible. However, all morphological variation is produced by development and because of development. This means that even the neo-Darwinian assumptions on variation logically require development to work in a specific way (as described in the previous section and in Salazar-Ciudad 2006a). However, the question is not whether development constrains evolution (or whether there are developmental constraints) because development is always there. In fact, contrary to what has been proposed, developmental constraints cannot be tested (Beldade et al. 2002; see Salazar-Ciudad 2006a for a discussion). The question is, instead, how different kinds of development affect evolution. This question should be addressed by looking at which variation each kind of development can produce (and not which imaginable variation they cannot produce): these are called here the variational properties of a type of development (Salazar-Ciudad et al. 2003).
Another central concept in the criticisms to the neo-Darwinian approach to morphological evolution is novelty. Many definitions consider novelty (Müller and Wagner 1991) as phenotypic change that does not fit into the scheme of small quantitative changes in existing traits (for example, the appearance of new traits). It is often proposed that novelty occurs rarely, while more gradual quantitative changes occur often. Novelty research often points out that the qualitative aspects of morphological variation may be related to developmental dynamics. However, since neo-Darwinism does not consider development, nor any intrinsic patterns of morphological variation appearing from it, the distinction between rare qualitative changes and common quantitative changes may not be the best way to describe morphological variation. If what is known about the morphological variation possible by development is taken into account (Salazar-Ciudad 2006b), then there are several kinds of possible morphologies, each of them with its specific multivariate pattern of variation. In a sense, the concept of novelty is useful only as far as the gradual totipotent view of variation is accepted.
Was this article helpful?