Life History Evolution. Derek A. Roff. Sinauer Associates, Sunderland, MA, 2001. 527 pp., illus. $54.95 (ISBN 0878937560 paper).
Derek Roff is one of the foremost authorities in life history, with a scholarly output that includes both theoretical and empirical work in population ecology and evolution. This is his second book about life history (the first was The Evolution of Life Histories, Chapman and Hall, 1992). In addition, he published Evolutionary Quantitative Genetics (Chapman and Hall, 1997), inspired in part by the quantitative genetics chapter from his 1992 book. Roff's latest contribution to our understanding of life history is Life History Evolution. This book is an impressive synthesis of the theory and empirical work of the last several decades, and it is an important addition to our understanding of the evolution of the diverse developmental and reproductive patterns found in the natural world.
As in his earlier book on life history evolution, Roff begins the present book with a short overview of quantitative genetics, which includes the derivation of some central models. He then moves into the heart of the book, starting with a discussion of biological tradeoffs. This general discussion of tradeoffs is followed by three chapters on life history evolution, one each for constant, stochastic, and predictable environments. With an exceedingly broad approach involving models and concepts from often disconnected regions of population biology, Roff makes many insightful connections, such as the relevance of game theory models of frequency-dependent selection in life history evolution. His concluding chapter, rather than being a review of topics covered (he concludes each chapter with a review), is a list of 20 topics for future research that he views as particularly important for advancing our understanding of the evolution of life histories.
In each chapter, Roff begins with a presentation of the genetic and optimization models relevant to the central topics. Following the presentation of the principal models is a careful presentation of the empirical research bearing on either the assumptions or the predictions of the models. Presentation of the literature and the links between the theory and the empirical data (when such exist) is the strongest part of the book. Especially enlightening were the occasional discussions of experimental methods that ought to be followed, with examples of studies that fell into or avoided the pitfalls. The recognition and discussion of problems with controls in brood manipulations (e.g., pp. 134–138) is an example of Roff at his best. Strikingly, he shows no taxonomic bias in the data chosen; they come from field and laboratory studies of a wide range of organisms, both botanical and zoological.
My main criticisms of the book concern the detail of the mathematical presentations and the editing. First, although the author states that his aim is to keep the mathematical derivations to a minimum, a very large number of derivations are presented. Deriving so many models that are available to the reader in their original papers, rather than presenting just the assumptions and predictions of each, often overwhelms the “nonmathematical” reader (to use the author's own term from the preface) with mathematical detail. One simple addition could have made the book easier to read: inclusion of a list of symbols and their use (symbols are not listed in the index) at the end of the book or at the end of each chapter. Interestingly, the sixth chapter does include tables of symbols for many of the central models. The failure to include such tables for the bulk of the book drove this reader to try to construct one, and in the process I found that some symbols have multiple meanings. This is trivial when the meaning is explicitly stated, and understandable because Roff was sticking as closely as possible to the original notations. However, symbol meanings are not always explicitly stated with each use, which leads to problems in the presentation of some models.
The book also suffers from poor organization and editing. The structure of the central three chapters risks losing the attention of less dedicated readers. Of the 465 pages of text, the central three chapters make up 66 percent. The fourth chapter alone has 136 pages, including many long mathematical derivations; it covers diverse topics such as the generation and maintenance of genetic variation, evolution of the individual components of life history, evolution of tradeoffs among life history components, and evolution of differences among species. Admittedly, there is an aesthetic appeal to a series of chapters entitled “Evolution in Constant Environments,” “Evolution in Stochastic Environments,” and “Evolution in Predictable Environments” (chapters 4, 5, and 6, respectively). However, subdivision of these chapters (particularly the fourth) would have made reading a less daunting task. There are also a large number of editorial errors both in the text and in the mathematical equations. Although many of the errors are trivial, some of them interfere with understanding and, in a few cases, result in erroneous or incomprehensible presentation of data or research protocols.
These problems may reflect the small number of readers before publication: Only three readers are acknowledged in the preface, and presumably all were colleagues familiar with the subjects covered. Having a larger number of readers—perhaps including some “nonmathematical” biologists—might have resulted in a text more approachable to a wider range of biologists. I suspect that the detailed presentation of a large number of models and empirical research may limit this book's readership to students of life history and may reduce the number of readers who work through the book in its entirety. This is a pity, because the discussions of the assumptions, conclusions, and empirical support for the various models would benefit a much broader range of population biologists, students, and professionals.
Department of Biology, University of Vermont, Burlington, VT 05405