Principles of Development and Evolution
So far, I have outlined the general properties of phenotypes, shown how they relate to development, and presented a model of adaptive evolution based on established principles of development and genetics. Now, using this general framework, I can summarize how developmental plasticity facilitates evolution. Jacob (1977) characterized evolution as “tinkering.” It shuffles and recombines what is already there. Frazzetta (1975), in another felicitous comparison with machines, wrote that evolution manages “the gradual improvement of a machine while it is running” (p. 20). Both of these qualities are possible due to characteristics of phenotypes that are not shared with most machines. Tinkering works because the phenotype is made of recombinable modular components that can be turned off and on in different conditions and can function in more than one context, what Gerhart and Kirshner (1997; Kirschner and Gerhart, 1998) call “weak linkage” to any particular use. Improvement without disruption of function works because of the remarkable active flexibility, and redundancy, in the development of parts. As a result of these two qualities—modularity and plasticity—an organism has the unmachinelike ability to respond to a new situation or to a new gene with the production of a new trait, and then to multiply, through reproduction, the ability to produce this trait. Differential reproduction starts the cycle of variation, selection, and cross-generational change that we call evolution—the most unmachinelike process of all. Many reasons have been given to believe that evolutionary change is difficult and even resisted in a well-adapted population (see chapter 1). The evolution of a novel specialization requires that a single lineage persist while undergoing extensive change. The conditions sometimes mentioned as favoring directional evolution, such as strong competition, very different or changing environments, small founder populations, or very long periods of time (see Mayr, 1982b), also favor population extinction. The idea of developmental cohesiveness, outlined in chapter 1, led to the further belief that major developmental change early in ontogeny would be disruptive. The cohesiveness theme persists even though it long has been clear that innovation does not occur exclusively by terminal addition (see chapter 1).