Optima for Animals, 2nd ed.

Optima for Animals, 2nd ed. R. McNEILL ALEXANDER. Princeton University Press, 1996, 169 pages ISBN 0-691-02799-4 Cloth; $60.00, ISBN 0-691-027986 Paper; $24.95).

In his revised edition, Optima for Animals, R. McNeill Alexander has updated his appealing and clearly written book to include additional subjects on the topic of optimization theory and its use in the analysis and interpretation of animal form and function, growth, life history, and behavior. Alexander writes from the viewpoint that evolution is an optimizing process. He argues convincingly that much can be gained by adopting mathematical optimization models in order to explore tradeoffs in hypothesized design or behavior subjected to physical or biological constraint, or to understand why one life history strategy has evolved versus another.

The strength and attraction of the book continue to be its short, concise writing on a broad range of topics that too often can be weighed down by heavy and complex mathematical treatment, often beyond the range of the average biologist. Alexander does a very nice job of presenting the essential mathematical concepts clearly and simply at the start of the book, and referring back to the relevant equations and concepts when they are used to discuss the biological examples in the chapters that follow. Nevertheless, the sections (5.2-5.4) on optimal strategies for making queens and growing, breeding, and survival, which rely on more advanced mathematical concepts to solve the optimization problem, are likely to remain beyond most readers' grasp. The author warns that the book alone is insufficient to make one an expert on optimization theory, particularly for more sophisticated mathematical topics such as this, but the book provides a readable introduction to many of these topics, pointing the reader to more in-depth texts that can be tackled for greater understanding.

Updates to the book primarily involve additional sections on optimal foraging theory in relation to learned behavior that is applied to feeding strategies (by ducks, passerines and lions) for patchy food resources. These classes of optimization problems involve applications of the marginal value theorem (Charnov, 1976) and dynamic programming for their solution. Other changes include the addition of sections on the optimal design of herbivore digestive systems and high jump performance by human athletes based on the author's own work (in exchange for a section on burst swimming performance of fishes that has been left out of the revised edition), parental care by sunfishes, and criticisms of the optimization approach. A model for optimal territory size that was developed in the original edition is here tested by new data available for migrating hummingbirds. Otherwise, the book remains much the same as its predecessor.

The section dealing with criticisms of the optimization approach to the study of biology is a welcome addition, but it falls short of yielding a satisfactory treatment of the issues that have been raised by many who have voiced concern for this approach and the view of natural selection as an optimizing process. Alexander focuses on two criticisms raised by Gould and Lewontin (1979) in their well-known paper, "The spandrels of San Marco and the Panglossian paradigm". The first is the tendency to ignore historical constraint in the evolution of organisms, which is likely to prevent optima from ever being reached. The second concerns the tendency to seek alternative optimization models should the first one fail, at risk of ignoring that other factors or constraints may apply which have little to do with optimizing the hypothesized (perceived) function. Given the distrust and misconceptions that abound over the use of optimality and optimization in relation to evolutionary biology, a more in-depth discussion of these issues would help to strengthen the book's thesis: that natural selection is an optimizing process and that optimization methods can provide considerable insight into the biology of organisms. As an advocate of the approach, I think that much of the confusion and misguided thinking has derived from the tendency to translate optimization as a process and a technique into the viewpoint that organisms are or have evolved to be 'optimal'. Alexander does point out some of the dangers of the method and misinterpretations that may result from its misuse, but a more informed discussion of these issues would have been worth the added length. Publication of this revised edition will be certain to make these approaches and concepts available to a wide audience and, thus, help to ensure their continued use and debate within biology.

REFERENCES

Charnov, E. L. 1976. Optimal foraging: The marginal value theorem. Theor. Pop. Biol. 9:129-136.

Gould, S. J. and R. C. Lewontin. 1979. The spandrels of San Marco and the Panglossian paradigm: A critique of the adaptationist programme. Proc. Roy. Soc. B 205:581-598.

ANDREW A. BIEWENER

Department of Organismal Biology and Anatomy