Developmental Constraints and Evolution: A Perspective from the Mountain Lake Conference on Development and Evolution

Developmental constraints (defined as biases on the production of variant phenotypes or limitations on phenotypic variability caused by the structure, character, composition, or dynamics of the developmental system) undoubtedly play a significant role in evolution. Yet there is little agreement on their importance as compared with selection, drift, and other such factors in shaping evolutionary history. This review distinguishes between "universal" and "local" constraints; it deals primarily with the latter, which apply to a limited range of taxa. Such constraints, typically, can be broken even within the taxa to which they apply, though with varying degrees of difficulty. The origin of constraints is discussed, five distinctive of constraint being explicitly considered. Three means of identifying constraints are set forth, as well as four means of distinguishing developmental from selective constraints. None of the latter (use of a priori adaptive predictions, direct measurement of selection, direct measurement of heritable variation, and use of the comparative method) is foolproof. In the final section, three larger issues regarding the role of developmental constraints in evolution are discussed: the extent to which evolutionary stasis can be explained in developmental terms, the extend to which evolutionary trends and patterns might be a consequence of developmental constraints, and the extent to which various genetic and developmental mechanisms have evolved in virtue of the need of lineages to manifest evolutionary plasticity (or adaptability) if they are to survive. Although no definitive conclusions are reached on these larger issues, we bring recent advances in developmental biology, evolutionary theory, and (to a limited extent) molecular biology to bear on them.