Evolution in Real Time Experimental Evolution Edited by T. Garland Jr. and M. R. Rose (2009) Berkley: University of California Press. 752 pp. $45.00 (paper) ISBN 978-0- 520-24766-6. Darwin envisioned evolution by natural selection as a slow process, far too slow to be observable in a human lifetime. However, he was somewhat mistaken. Evolution can indeed be a slow and elusive process, but it can also proceed rapidly on time scales of years and decades in organisms with short generation intervals. Rapid evolution is not only observable and measurable, but it can be experimentally induced by the implementation of a selection regime. Controlled experiments designed to give rise to rapid evolution fall under the heading of ‘‘experimental evolu- tion.’’ Over the past few decades, ex- perimental evolution studies have provided some of the most satisfac- tory tests of evolutionary theory and hypotheses. The recent volume edited by Garland and Rose, Experimental Evolution, synthesizes much of this research and offers an excellent intro- duction to this growing field within evolutionary biology. The book is organized into five parts. Part I (Chapters 1–3) gives a general overview of experimental evo- lution research. Most studies can be categorized as either ‘‘artificial selec- tion’’ or ‘‘laboratory natural selection’’ experiments (see chapter by Garland and Rose). In artificial selection, indi- viduals are measured for one or more phenotypic traits and breeders for the next generation are chosen based on their phenotype. In laboratory natural selection, the environment of a freely breeding population is altered and the population adapts to the new environ- ment over multiple generations. Com- pared to other approaches to studying evolution, the unique strengths of ex- perimental evolution studies are repli- cation and control (Futuyma and Bennett). Replicating the number of populations under selection allows the investigator to assess the consis- tency of the response to selection. Replicate populations can also be compared to multiple control popula- tions to test evolutionary hypotheses. Part II (Chapters 4–8) discusses dis- tinct types of experimental evolution. Long-term experiments focusing on the process and consequences of evo- lution can be distinguished from short-term studies, which primarily examine the potential for selection to act (Travisano). Different experimen- tal systems can also be characterized as either bottom-up approaches, from molecules to organismal phenotypes, or top-down approaches, from organ- ismal phenotypes to molecules (Dykhuizen and Dean). Reverse evolu- tion studies apply selection to charac- ters that were initially selected in an opposite direction of phenotypic change (Estes and Teoto ´ nio). Evolu- tion experiments conducted in the field, such as introductions of live ani- mals, are an important point of com- parison for laboratory studies since evolution in nature can differ from evolution in a laboratory (Irschick and Reznick). Part III (Chapters 9–3) considers levels of observation in experimental evolution. Natural selection generally acts most directly on phenotypic traits at relatively high levels of bio- logical organization, such as behav- iors, components of life history, and organismal performance. These com- plex, high-level traits are composed of interacting lower-level traits, includ- ing aspects of physiology, morphol- ogy, cells, and molecules. Therefore, an evolutionary response to selection acting on high-level traits necessarily involves changes in lower-level traits (Rhodes and Kawecki; Swallow et al.). Differences among replicate populations in suites of adaptive lower-level traits can be interpreted as ‘‘multiple solutions’’ to selection (Swallow and colleagues). Complex constraints and trade-offs occur at all levels of biological organization, among high-level traits (Zera and Harshman), as well as at the lowest level of genome architecture (Rosenzweig and Sherlock). Part IV (Chapters 14–20) presents a sample of exciting applications of ex- perimental evolution, including chap- ters on the scaling relationships bet- ween morphological traits (Frankino and coworkers), speciation (Fry), and the evolution of sex (Turner and co- workers), aging (Rauser and coworkers), and altruism (Kerr). The number and diversity of studies reviewed in these chapters precludes a thorough sum- mary here. However, by way of illus- trating the power of experimental evo- lution to resolve evolutionary theories, I highlight a particularly elegant study discussed by Turner and colleagues. For more than a century, evolutionary biologists have claimed that a major advantage of sexuality over asexuality is that when there is selection, sex accelerates adaptation by increasing genetic variation. To test this idea, Goddard, Godfray, and Burt 1 com- pared the rate of adaptation by sexual and asexual (genetically engineered) yeast populations in a harsh (hot and salty) environment that imposed strong directional selection. After 100 generations, mean fitness was roughly 10% higher in the sexuals, correspond- ing to a 14% higher growth rate. In contrast, there was no detectable effect of sex among control populations maintained in a benign environment where there was little selection. Thus, experimental evolution provided a clear and direct demonstration that allelic recombination accelerates adap- tation through selection. Part V (Chapters 21 and 22) covers certain limitations of experimental evolution, particularly those inherent in laboratory experiments that intend to mimic evolution in nature (Huey and Rosenzweig). Alas, experiments can teach us a lot about evolution, but not everything. A full under- standing of evolution will be achieved only with multiple inte- grated approaches, including studies of natural populations, comparative methods, and mathematical modeling. BOOK REVIEW V V C 2010 Wiley-Liss, Inc. DOI 10.1002/evan.20271 Published online in Wiley Online Library (wileyonlinelibrary.com). Evolutionary Anthropology 19:200–201 (2010)