Experimental Evolution: Concepts, Methods, and Applications of Selection Experiments

2. The Importance of Experimental Studies in Evolutionary Biology

15 2 THE IMPORTANCE OF EXPERIMENTAL STUDIES IN EVOLUTIONARY BIOLOGY Douglas J. Futuyma and Albert F. Bennett METHODS OF STUDYING EVOLUTION Experimental Evolution Studies on Natural Populations Comparative Methods TWO EXAMPLES OF THE CONTRIBUTION OF EXPERIMENTAL STUDIES TO UNDERSTANDING EVOLUTION Adaptive Gain and Correlated Loss Studying the Adaptive Role of Genetic Drift Fitting the Experiment to the Question CONCLUSION Experimental Evolution: Concepts, Methods, and Applications of Selection Experiments, edited by Theodore Garland, Jr., and Michael R. Rose. Copyright © by the Regents of the University of California. All rights of reproduction in any form reserved. METHODS OF STUDYING EVOLUTION Experimental studies complement the several other major approaches to analyzing evolutionary processes. Each approach has both advantages and limitations. We first describe the advantages of experimental evolution, and then we contrast it with other approaches. EXPERIMENTAL EVOLUTION The essence of experimental evolution is conceptually quite simple. For many generations , a series of replicated populations is exposed to a novel environment, while a parallel series of populations is maintained within the ancestral environment, thereby serving as experimental controls. The environmental novelty may involve alteration of any aspects of the abiotic, biotic, or demographic condition of the ancestral population. Usually only a single environmental variable is altered to keep the experiment as simple as possible . It is, however, feasible to change multiple environmental factors simultaneously. The novel experimental environment provides new selective conditions and hence promotes evolution. New genetic variants (produced through recombination, mutation, or other processes) may be differentially advantaged or disabled in the altered conditions, producing differential reproduction and increase of the favored genotypes. Several different methods of environmental alteration are used in experimental evolutionary studies, including artificial truncation, culling, and laboratory natural selection (Rose et al. 1990; Garland 2003; Rose and Garland this volume). The distinctions among and advantages and limitations of each of these methods are discussed more fully elsewhere in this volume (especially Huey and Rosenzweig). They are all similar, however, in creating conditions favoring evolutionary alteration of the condition of the ancestral population. After a sufficient number of generations, the novel experimental populations may be compared directly with the controls (or in some cases directly to their own ancestors), and any variety of a priori hypotheses concerning evolution may thereby be tested. The longer the antecedents of the ancestral population have been maintained in the ancestral (control) condition, the more likely any differences observed within the novel experimental populations will be specific to the novel environmental alteration (see also Travisano this volume). What is particularly advantageous about this approach to evolutionary studies? Its special strengths lie in the essence of any experiment: replication and control. By replicating the number of populations exposed to the novel environment, an investigator can, in effect, repeat the opportunity for evolutionary change and determine if the outcome has consistency. In the metaphor of S. J. Gould (1989), the tape of life and evolution can be replayed as often as desired to determine the similarity of its resulting products. Evolutionary lineages of natural populations are a series of unique events, and it is never possible to determine if things could have turned out differently or what factors were responsible for the observed differentiation. In experimental evolution, however, the 16 • I N T R O D U C T I O N T O E X P E R I M E N T A L E V O L U T I O N diversity of responses can be examined directly. Through replicated measurements among experimental populations, this diversity can be analyzed statistically. Further, the novel experimental populations can as a group be statistically compared with the group of control populations (or, in some situations, with the ancestral population itself), with the number of degrees of freedom determined by the number of independent replicates of each. The statistical significance of any differences between the experimental and control groups can be evaluated to test evolutionary hypotheses. Experimental evolution has certain additional attractive features. Unlike studies of historical evolution in the natural world, there is never any uncertainty about ancestral condition: the experimenters know the condition and composition of the ancestral population because they chose and observed it. Also in contrast to historical evolution, experimental evolution can define and control the degree of environmental change, limiting it to a single factor or any desired combination of factors. So many environmental aspects change simultaneously in the natural world that it is difficult to know to...