Sex combs find middle ground in evolution debate.

Most scientists are familiar with the basic tenet of evolutionary biology “mutation proposes, selection disposes,” where random variation generated by mutation is thought to provide the fuel for natural selection to direct evolutionary change. If this is truly a basic tenet, then why have prominent evolutionary biologists expressed concern about the relative importance given to variation or selection in the evolutionary process? For example, West-Eberhard states “There is prejudice against variation that is non-adaptive in studies of adaptive evolution, against variation that is non-genetic in studies of genetics, and against virtually all variation in studies of the genome in the worm, (Caenorhabditis elegans), the fly (Drosophila), and the frog (Xenopus).” (ref. 1, p. 205). In contrast, Bell (ref. 2, p. xix) states “A further weakness is that when selection is described, it is often treated as one of several possible mechanisms of evolution. . . This is surely unjustifiable.” Their concern is about which of these two determines the rate and direction of evolutionary change (3–5); in other words, who’s in the evolutionary driver’s seat: variation or selection? By studying the developmental mechanisms underlying evolution of sex combs in the fruit fly Drosophila melanogaster, in PNAS, Malagón et al. provide new insight on this debate (6). According to Gould (4), this debate has a long history that precedes Darwin’s Origin of Species, occurring between biologists who argue that natural selection is the creative force controlling both the rate and direction of evolution (functionalists) vs. those who argue that it is variation that controls them (structuralists). Variation in the eyes of the functionalist is small, continuous, copious, and random with respect to the environment, allowing natural selection to gradually mold phenotypes to a functional optimum as required by the environment. Functionalists acknowledge that limited genetic variation may slow the rate of natural selection, but this does not influence its ability to direct evolution. In contrast, variation in the eyes of the structuralist is discontinuous, structured and nonrandom and dictates the rate and direction of evolution. An example of extreme structuralist arguments denying selection any role for the origin of new species is Hugo de Vries’ “mutation theory” in 1903, which argues that new species arise in a single mutational step: a saltational jump to a fully functional and novel phenotype (7). Such extreme arguments, however, have largely been discredited. Developmental constraint is a moderate version of structuralism that has been influential in modern evolutionary thought (8, 9). A developmental constraint is most often defined as “a bias on the production of variant phenotypes or a limitation on phenotypic variability caused by the structure, character, composition, or dynamics of the developmental system” (10). It can have positive (can bias variation) or negative (can limit variation) effects on evolutionary change. The relative contribution of developmental