Derek Roff University of California, Riverside Sinauer Associates, Inc. 2002 PREFACE An obvious feature of the natural world is that it is made up of an enormous variety of different life histories. Even different populations of the same species can show considerable variation in life history patterns and parameters. This book describes how evolutionary biologists have attempted to account for this variation. My aim is to provide a framework within which such analyses can be understood and demonstrate by example the enormous explanatory and predictive power of present life history theory. I have kept mathematical derivations to a minimum and placed material that is useful but which can be skipped over in "boxes" within the body of the text. The analysis of life history variation is necessarily a quantitative endeavor and its theoretical foundation requires a mathematical structure. Therefore, the "minimal" amount of mathematical detail is still substantial. Nevertheless there is plenty of room for the "on-mathematical" biologist. What is important is to understand the assumptions underlying the various models and hence their potential limitations. CONTENTS Preface vii Chapter 1: Overview 1 Chapter 2: A Framework for Analysis 13 Chapter 3: Trade-offs 93 Chapter 4: Evolution in Constant Environments 151 Chapter 5: Evolution in Stochastic Environments 287 Chapter 6: Evolution in Predictable Environments 359 Chapter 7: Topics for Future Study 459 References 467 Index 517 EXCERPT FROM OVERVIEW One of the most obvious features of the living world is that organisms differ enormously in their life histories. Such traits as the age at maturity, adult size, mortality rate, and age-specific fecundity rate all show wide variation. For example, the flatfish Hippoglossoides platessoides occurs on both sides of the Atlantic, but even though the populations are taxonomically the same species, the difference in life histories is profound. On the Grand Banks of Newfoundland, H. platessoides grows more slowly than off the coast of Scotland, reaches a much larger size (60 cm vs 25 cm), matures later (14 yrs vs 3 yrs), lives longer (30 yrs vs 6 yrs), and has a lower age-specific fecundity (Figure 1.1). This book is based on the proposition that such variation is explicable under the assumption that life history variation is primarily the result of natural selection and represents adaptation. A critical tool for the study of the pattern and evolution of life histories is mathematical analysis. This statement is a self-evident fact to most population biologists. However, the use of mathematics in ecological investigations has had a much rockier road than its use in genetical analysis, and its general acceptance as an important tool dates only from the 1960s. (See Kingsland 1985 for an excellent historical survey of the rise of mathematical approaches in ecology from the work of Lotka in the 1920s to the studies of MacArthur up to 1970.) The importance of the mathematical approach to the understanding of genetic variation is amply illustrated by Provine's review of the history of population genetics (Provine 1971) and by the biographies of three of the most influential geneticists of this century: Fisher (Box 1978), Haldane (Clark 1984), and Wright (Provine 1986). Even by the latter half of the 1940s, mathematical thinking had still not made a significant impact on ecological theory. Allee et al. (1949, p. 271) observed that "theoretical population ecology has not advanced to a great degree in terms of its impact on ecological thinking." An early antipathy to the use of mathematical analysis may account in part for the delay in the merging of the ecological and evolutionary perspectives in what is now commonly known as life history analysis. An influential factor encouraging the use of mathematical investigation into life history variation was Lamont Cole's 1954 paper "The population consequences of life history phenomena," which set out one of the basic mathematical frameworks by which the consequences of variation in life history traits can be analyzed. Cole's paper ushered in an era of research predicated on the integration of mathematics and biology in the study of the evolution of life history patterns. In his review, Cole asked how changes in demographic attributes, such as the age at first reproduction, influenced the rate of increase of a population. Except for citations of its historical importance, Cole's paper gained widespread notice because of an apparent paradox with respect to the value of semelparity versus iteroparity: "For an annual species, the absolute gain in intrinsic population growth which could be achieved by changing to the perennial reproductive habit would be exactly equivalent to adding one individual to the average litter size" (Cole 1954, p. 118' the italics are Cole's). With the aid of hindsight, the resolution to this paradox is simple to the point of being trivial (see Chapter 4), but its importance lay in drawing attention to the value of mathematical analysis of life history phenomena. Cole's paper enunciated two important principles that are the basis of life history analysis: The birth rate, the death rate, and the age composition of the population, as well as its ability to grow, are consequences of the life-history features of the individual organisms. These population phenomena may be related in numerous ways to the ability of the species to survive in a changing physical environment or in competition with other species. Hence it is to be expected that natural selection will be influential in shaping life-history patterns to correspond to efficient populations. Thus natural selection is seen as maximizing some quantity, here termed "efficient populations," but elsewhere in the paper identified as the rate of population growth. This is not to be taken as indicating that Cole favored the idea of group selection. The tenor of his paper makes it clear that his use of population can be understood in modern terms to be equivalent to genotype. Thus Cole is making the point that selection favors those genotypes that have the highest rates of increase. The second important principle put forward by Cole is that natural selection favors those patterns of birth, death, and reproduction that maximize the rate of increase. This observation was certainly not unique to Cole and can be traced back to Fisher (1930) and in verbal form to Darwin and Wallace. Andrewartha and Birch (1954) emphasized the importance of the potential for increase, devoting a whole chapter to the concept in their book The Distribution and Abundance of Animals. Birch later stressed the relationship between the genotype and its rate of increase, r: "Natural selection will tend to maximize r for the environment in which the species lives, for any mutation or gene combination which increases the chance of genotypes possessing them contributing more individuals to the next generation (that is, of increasing r) will be selected over genotypes contributing fewer of their kind to successive generations" (Birch 1960, p. 10). Mathematical modeling has been, and continues to be, an important component of the analysis of life history variation (Stearns 1976, 1977, 1992; Parker and Maynard Smith 1990; Roff 1992a; Charnov 1993, McNamara 1993; Charlesworth 1994; Houston and McNamara 1999; Fox et al. 2001). The purpose of model construction is to address a particular aspect of the real world, ranging from a very detailed analysis of a very specific circumstance, to an assessment of a general proposition. All models necessarily are simplifications of reality. To ensure that the results are robust to the assumptions, Levins (1966, p. 423) recommended the use of several different models incorporating different assumptions: "Then, if these models, despite their different assumptions, lead to similar results we have what we can call a robust theorem which is relatively free of the details of the model. Hence our truth is the intersection of independent lies." The analysis of life history evolution has developed along two lines, one concentrating on following evolutionary change via the underlying genetical determination of the traits, and another that focuses on the phenotype and, as indicated above, assumes that there is some phenotypic measure of fitness that is maximized. In recent years the barrier between these two approaches has been breaking down as more and more researchers appreciate the value of combining them. WHERE TO ORDER Sinauer Associates, Inc. 23 Plumtree Road Sunderland, MA 01375-0407 Telephone: (413) 549-4300 Fax: (413)-549-1118 E-mail: orders@sinauer.com Website: www.sinauer.com PRICE: $52.95 ISBN: 0-87893-756-0
URL: http://www.primate.wisc.edu/pin/review/life.html
Page last modified:
May 2, 2002
Maintained by the WPRC Library
Return to Review Copies Received
Return to PIN Home Page