Volume 80, No. 1 THE QUARTERLY REVIEW OF BIOLOGY March 2005 THE EVOLUTION OF PARENTAL CARE Mart R. Gross Department of Zoology, University of Toronto Toronto, Ontario M5S 3G5 Canada e-mail: [email protected] abstract Our understanding of parental care behavior can be significantly advanced through the applica- tion of Williams’s Principle, which states that reproduction has not only a benefit but also a cost to lifetime fitness. My laboratory has formalized Williams’s Principle into the relative value theorem and found that its application to fishes, the taxa with the most diverse patterns of parental care, can help to explain which sex provides care and how much. In fishes, it is often the male that provides parental care, not because the male obtains greater benefits from this care, but probably because he pays fewer costs. Fish dynamically adjust their investment into parental care according to the number of offspring in their brood, past investment, genetic relatedness, and alternative mating opportunities, all of which affect the value of current offspring relative to potential future offspring. These results may also help us understand the joy and the challenges of parental care in humans. costs of reproduction, and a refinement of Introduction Lack’s principle” (1966b). His explicit rec- ARENTAL CARE is an excellent exam- ognition of the cost of reproduction and P ple of a behavior that would seem to ben- gene-based fitness helped lay a foundation efit the species. It promotes the survival and for current life history theory (e.g., Stearns well-being of the next generation at a cost to 1992; Roff 2002). The significance of these the resources of the current generation, and concepts to scientists is reflected in how cita- tion of these works continues to increase (Fig- the apparently sacrificial acts of individual ure 1). Eventually, their importance may con- parents compel our admiration. Human pop- tribute to reshaping human social, political, ulations have entwined parental behavior and religious rules surrounding parental into social, political, legal, and religious rules, care. and modern governments are often elected, I became interested in parental care behav- or not, based on attitudes toward parenting ior as a graduate student in the 1970s; for sev- and the family. Despite, or perhaps because eral decades, my students, colleagues, and I of, this ardent ideology of parental care in worked to understand its evolutionary prop- human society, we remain largely ignorant of erties. Our model organisms have been the the biological basis for why we invest into rais- fishes, the vertebrate group providing the ing our offspring. greatest diversity of patterns in parental care. In the 1960s, at a time when parental care In this article, I wish to show the utility of Wil- was widely viewed among biologists to be an liams’s Principle in answering two major ques- adaptation that benefits the species, George tions about parental care: (1) “who cares?” Williams authored Adaptation and Natural (i.e., which sex is favored by natural selection Selection: A Critique of Some Current Evolutionary to show parental care for the young?); and Thought (1966a), and “Natural Selection, the (2) “how much?” (i.e., what factors enter into 37 38 THE QUARTERLY REVIEW OF BIOLOGY Volume 80 Figure 1. Citation Frequency Over Time of Natural Selection, the Costs of Reproduction, and a Refinement of Lack’s Principle (Williams 1966b) (a) Annual number of citations, and (b) cumulative number of citations, recorded by the Institute for Sci- entific Information (ISI) Web of Knowledge, from 1966 through 2003. decision rules about the quantity of care to origins of care, while the question of the provide?). The question of which sex exhibits amount of care helps us to understand its care helps us to understand the evolutionary dynamic adjustment in quality. As most biol- March 2005 SYMPOSIUM IN HONOR OF GEORGE C WILLIAMS 39 ogists know today, parental care behavior is tively determine lifetime reproductive suc- not an adaptation to benefit the species; cess. A further consideration is that the total rather, parental care has evolved because it effort that can be expended by an organism will be allocated into (1 ס maximizes the selfish genetic interests of the over its lifetime (E parent. What is of interest to biologists today somatic effort (SE), which maintains the par- are the myriad life histories and sophisticated ent’s own viability and growth, and reproduc- calculations that animals and plants have tive effort (RE), which produces descendants. evolved to invest in the survival of their own It follows that lifetime reproductive success is genes into the future. definable as success in the present (as a func- tion of RE) and success in the future (as a .(F(SE ם (P(RE ס Williams’s Principle function of SE), or LRS Williams’s Principle was named as such Williams’s Principle is the recognition that when my postdoctoral fellow, Craig Sargent, investment into the present comes at the cost who had recently obtained his PhD in George of investment into the future. Natural selec- Williams’s laboratory at the State University tion will therefore favor behaviors that maxi- of New York, Stony Brook, and I realized that mize lifetime reproductive success subject to Williams’s concept of life history tradeoffs this constraint. In other words, reproduction and a cost of reproduction (Williams 1966a, has a cost (i.e., investment made into current 1966b) was central to bringing rigor into the progeny is traded off with investment made study of parental care, which at the time was into future progeny), and natural selection a relatively nonquantitative topic emerging will optimize the allocation of investment into from ethology (Gross and Sargent 1985; Sar- the present relative to the future. From this, gent and Gross 1985, 1986). We were both the relative value theorem, in which animals interested in using fishes to study parental invest according to the value of their current care and mating systems. Robert Trivers brood relative to their own expected future (1972) was implicitly using Williams’s con- reproduction (Coleman et al. 1985; Sargent cepts and significantly advancing the field. At and Gross 1985), can be derived. Continued the University of Michigan, Bobbi Low was investment into present progeny to increase employing Williams’s tradeoff theory to offspring survivorship and fertility should be understand optimal mating and parental weighed against investments into expected effort in relation to environmental uncer- future progeny through increased adult sur- tainty (Low 1978). Others were also inter- vivorship and breeding success. The optimal ested in the notion of tradeoffs for under- investment of RE occurs where the rates of standing parental care (e.g., Pressley 1981; return on investment into present and future Carlisle 1982; Fagerstron 1982). reproduction are equal, as this will maximize Modern life history theory recognizes that lifetime reproductive success (Figure 2). Nat- natural selection favors the evolution of ural selection does not maximize reproduc- behavior that will maximize lifetime repro- tive success at any one time; it maximizes life- ductive success (LRS; the number of copies time reproductive success by optimizing of genes that an individual leaves to future effort across time. These ideas are now well- generations across its entire lifespan). Imag- summarized in parental care (e.g., Clutton- ine a parent that has just produced a brood Brock 1991) and general life history texts of offspring. Its remaining lifetime reproduc- (e.g., Stearns 1992; Roff 2002). tive success can be divided into two compo- nents: that which is obtained through the Who Cares? present brood (P); and that which is obtained There are significant taxonomic patterns through all future broods (F). The number in which one sex or the other will show care of current offspring, their genetic relatedness for the young (e.g., Reynolds et al. 2002). to the parent, and their survival—as well as Among vertebrate species, for instance, mam- the probability of future broods and their mals exhibit predominantly female-only care numbers, relatedness, and survival—cumula- (found in about 90% of families) and a smat- 40 THE QUARTERLY REVIEW OF BIOLOGY Volume 80 evolved all forms of parental care, they are especially noteworthy for their male invest- ment. There has been considerable discussion about why fishes exhibit these alternative care states (e.g., Balshine and Earn 1998). In par- ticular, why is there so much male-only care? The best explanation seems to come from the application of Williams’s Principle: the con- sideration of both the costs and benefits of investing in parental care for a present brood relative to investment into future broods (Gross and Sargent 1985; Sargent and Gross 1986, 1993; Alcock 2001). In fish, female reproductive success (as measured by egg number) typically increases at an accelerating rate with adult body size, at least more so than it does in males. Male reproductive success tends to increase at a diminishing rate. Now imagine a mutant gene for showing parental care, and consider both the benefit to the parent’s present reproductive success and the cost to its future reproductive success. Since most fish display indeterminate growth based on energy, the cost of present investment at Figure 2. Graphical Representation of the expense of future investment is likely to Williams’s Principle. be exhibited in a feature such as body size; The bottom panel is a hypothetical diagram of how thus, a fish that does not show care for its reproductive success in the present and in the future young will be larger in the future than one may vary with investment into reproductive effort that does. Because of the different relation- (RE) and somatic effort (SE), respectively.