Leigh_Conflict_1999

Leigh_Conflict_1999 - l4 - REEVE AND KELLEFI .' that have...

Info iconThis preview shows pages 1–9. Sign up to view the full content.

View Full Document Right Arrow Icon
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 2
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 4
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 6
Background image of page 7

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 8
Background image of page 9
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: l4 - REEVE AND KELLEFI .' that have difi'erent biological characteristics, the latter characteristics all hav- ing been fixed by within-population selection.) Under this intriguing view, lineages with relatively high repulsive and low attractive forces (ie, those in which lower-level vehicles are less likely to form higher-level vehicles) are more likely to become extinct, leading to a long-term lineage selection for clades that exhibit wen-elaborated, high-level vehicles. BI. Similarly, how do ath'active, repulsive. and centrifugal forces interact to shape the properties of high-level vehicles like animal societies? Can re- - pulsive forces sometimes nullify each other within high—level vehicles (as in : policing against selfishness) and thus leave an imprint on the characteristics of the high-level vehicle (such as increased efficiency resulting from greater intemal cooperation)? Are there atu-active forces (perhaps originating only _' afier the initial creation of the high-level vehicle) that would nullify all or f some of the original repulsive forces and thus leave an imprint on the chap . acteristics of the high-level vehicle? These topics are addressed in chapters 8—11. For example, Keller and:- Reeve (chap. 8) discuss how policing and bribing can promote innagroup - cooperation within animal societies by in effect weakening repulsive force: or strengthening attractive forces. Similarly, Maynard Smith (chap. 10) in-_ vestigates the conditions favoring the emergence and enforcement of social. contract strategies to punish selfish behaviors in human societies. Finally. the evolution of co-adapted traits in obligater mutualistic species (cg. figs and; their associates; Herre, chap. 11} provide yet another example of forces that arose or strengthened after the initial creation of a higher-level vehicle from the mutualistic pair of organisms. that is, following the evolu-‘i' tion of complete reproductive interdependence. Cl. Perhaps the most unexplored question concerns how interactions be- tween lower-level vehicles might affect the interactions between intermedi .3 ate-level vehicles and thus affect the properties of the highest-level For example, Keller and Reeve (chap. 3) describe one of Reeve's (1' “it: t;- hypotheses for the absence of nepotism within insect societies. lnn'agenomiil selection on parentally imprinted alleles involved in kin recognition [1 " level vehicles) might favor sabotaging of the potential nepotism ’ r' _ machinery of individuals {intermediate-level vehicles), leading to the lack if nepotism and thus increased cooperation within hymenopterart soci ' 453‘ (highest-level vehicles). -- " Acknowledgments We thank E. Leigh for useful comments on the manuscript. We were * I-I'i' ported by grants from the Swiss and [1.8. NSF. Levels of Selection, Potential Conflicts, and Their Resolution: The Role of the "Common Good ” Egbert Giles Leigh, Jr. Adaptation is shaped by the competitive process of natural selection (Darwin 1859). Genes are the units whose “self—interest" drives natural selectio. In other words, nothing nonrandom happens in the evolution of a species unless it “serves the self-interest"—causes the differential reproduction—of some gene (Dawkins 1976; Bourke and Franks 1995). Yet no one of these genes, ultimate unit of self-interest though it might be, can do a thing outside the context provided by the rest of its genome and the organism for which that genome is appropriate. Alone, a gene is as useless as a fragment of a com- puter program without the rest of the program, a ootnputer suited to run the program, and an operator capable of using the program and the machine. In short. the units of competitive self-interest that make up a genome are utterly interdependent. How did the competitive process of natural selection shape so intricate a munialism‘? Ecological communities are structured to a large extent by competition: Competition among individuals for food or space, and competition between consumers and their potential prey over who uses the resources in these prey‘s bodies (Hutchinson 1959; Paine 1966). Competition among plants for llght, water, and nutrients. and between consumers and their intended prey. is particularly intense in tropical forest (Robinson 1985; Richards 1996). Yet tropical forest is not onlyr a climax of competition but an apex of mutualism. Plants depend on fungi for the uptake of nutrients [Allen 1991) and on ani- Irttals for pollination of their flourers. dispersal of their seeds, some- Slmeshfor burial of then- seeds out of the reach of insect pests [Corner 1964; and}: e 1:989; Forget l991). These mutualisms make possible the diversity The uxunance of n'opica] forest {Corner 1964; Regal 19'r'1r'; Crepet 1934). neegsconsntute an extraordinary web of interdependence. A tree species that rm Whagoptrs to bury Its seeds needs other tree species to keep the agoutis p01]- cn 1t itself 15 not fru1ttng (Forget 1994]. The durian whose flowers are d Iltta‘ted by bats needs mangroves to keep these bats in nectar when the “Han s forest has few plants in flower (Lee 1980). Although ecological 16 - LEIGH communities are theaters of competition, their species depend on each other in many ways. No species can survive outside an ecosystem that provides it food, shelter, even the air that it breathes. How has competition among spe- cies brought forth the interdependence tltat characterizes ecological conunu- nities? _ Such questions are avatars of a fundamental problem in ethology (Moynihan 1998}. All vertebrates, and many other animals. depend on some conspecific for help of some sort, at least in conceiving offspring. Yet mates are poten- tial competitors (Lessells, chap. 5). Who can forget Fabre’s (1989, vol. 1. pp. I HHS—1106) account of the headless trunk of a male mantis, still continuing to impregnate the female that has already eaten its head and is now chewing - down on its thorax? In view of such possibilities as this, what keeps compe- _ tition from destroying the common good that could he created by cooperat- ing? The problem of which social mechanisms maintain cooperation among - potential competitors (Moynihan 1998), and how this cooperation evolved tau begin with (Hamilton 1964a), is most acute for animals that live in groups - {Keller and Reeve, chap. 3; Kitchen and Packer, chap. 9}. Fellow members of a group depend on each other for the advantages tltey derive from group life: increased safety from predators or competitors. benefits of mutual assis- tance such as grooming for ectoparasites, and the like. Yet fellow group members are also each other‘s closest competitors for food. mates, and other -= resources. What keeps competition among a group‘s members from oven" whelming their common interest in their group‘s effectiveness and annihilat- *- ing the common good of their cooperation? Ethology is now a marginal subject: The more fashionable of its practitioners have hastened to label themselves sociobiologists or behavioral ecologists. Yet this “fundamental problem of ethology" is the unifying theme of this book. Meteover, we” wrote this book because we think that the recurrence of this problem III_ various levels of biological organization is one of the grand unifying them of biology and anthropology. Avatars of the Ethologist’s Problem This fundamental problem of ethology is parallelled at other levels of bio-:- logical and social organization. not least in human societies (Maynard Smiflll chap. 10)- Perhaps it is no accident that a clear formulation and discussion - this type of problem is already given in Aristotle’s Politics (Barnes 1934, 1986—2129). Here. Aristotle is concerned with how best to achieve barman}- between the good of a city-state and the enlightened self-interest of its hahitants. Aristotle observed. “In all arts and stint-ices the end is a. good, 1 the greatest good and in the highest degree a good in the most authoritali ROLE OF THE “COMMON GOOD" - II" [Science] of all—this is the political science of which the good is justice. in other words, the common interest" (Politics 1232b, 14—16; p. 2035 in Barnes 1934}. Aristotle considered that a city-state‘s organization (constitu- tion) was more likely to persist if it clearly served the common good of its inhabitants: otherwise. it would be more liable to overthrow by conspiracy or revolution. Indeed, the common good turns out to playr a cmcial role in all avatars of our problem, for a mutualism will evolve only if it serves the common interest of all participants (Leigh 1991}. An obvious. yet mysterious. avatar of the ethologist's problem concerns species in ecosystems. A crucial feature of ecosystems is the variety of inter- dependence among species whose members are all competing for resources needed to survive and reproduce. Ecosystems can be viewed as functional entities. with producers. transformers, decomposers. etc. We speak of eco- systems being injured by human disturbance, as if ecosystems are organized to fulfill functions. Aristotle (Physics [99b4, p. 340 in Barnes 1984) re- marked that in systems organized to function, the abnormal and the disrupted are usually less functional. Indeed, only if a system is organized to fulfill a function can we speak of it as being impaired by change (Fisher 1958, pp. 41—44}. Such talk leaves much unanswered. What is the “function” of an ecosystem? Ecosystems are not units of selection, organized to reproduce themselves. It makes more sense to view ecosystems, like human societies as commonwealths in whose integrity member species share a common stake, The mystery lies in the nature of this common interest, and in whether [and If so. how] this common interest affects the evolution of the ecosys- tem‘s species (cf. D. S. Wilson 1930). Unlike nation-states. ecosystems, and many animal societies the harmony of cellular organization and developmental process seems so absolute that it :t'as long taken for granted. Yet, just as dysfunctions induced by gene muta- tons provide essential clues to the mechanisms of gene action and develop- 2;:rtla1'process, so other dysfunctions—segregation-distotters, cytoplasmic we ratio mutants, cancer, the like—reveal that organisrnic harmonies once took- for granted ongmated from cooperation among relatively inde- 2:“??l entities (Hurst et a]. 1996', Michod. chap. 4; Pontiankowski, chap. méjélr eggplptpznof these conflicts reveals a past evolutionary breakthrough, a whenby mu a? artsnton (Maynard Smith and Szathmtiry 1995), more integgralptirlfolghfltwfltli mdependcm‘ cooperating PHI-flies com into ammogist‘s bl a en became the central untts of selection. The Gem” _ pro em thus has fundamental parallels in developmental and at biology. Indeed, it now a th ' - ppears at even molecular biologists would do we" to air where and how the fundamental I - problem of ethology l D thetr work. as the following examples show. It ates Same . . . . favor N alleles spread by biasing meiotic segregation-ratios in their own - evertheless, at most chromosomal loci, meiosis is one of the fairest 18 - LEIGH lotteries known to art or nature in that a gamete of an individual herein- zygous at a given locus has an equal chance of receiving either allele from that locus. Where meiosis is fair, an allele can spread only if it benefits t; carriers {and therefore the genomes these individuals carry). Accordingly fair meiosis represents the common interest of the genome as a whole '-_m_ [9?1}. What factors preserve the honesty of meiosis and render segregati_ distortion so rare? . In species whose members inherit organelles only from the mother, cytoplasmic (organeliari mutants cause female-biased sex ratios (Hurst 19933 .- Where both egg and spenrt contribute organelles to a zygote, conflicts r.- tween maternal organelles and their paternal counterparts may threaten it}. zygote (Eberhard 1980}. Such conflicts of interest between organe . and their host cells—traces of a time when organelles were indepe u- __ organisms that had somehow entered host cells (Margulis 1993}—-raise I question. How is harmony between cells and their organelles normally I: :n' tained (Eberhard 1980}? ' Cancers represent a conflict between an individual and certain of its o- :5; Yet the human body is considered the very archetype of harmonious - 7. of the whole by its parts (1 Corinthians 12:20—25). How are conflicts ts“ tween a multicellular organism and its constituent cells avoided or um: mized (Boss 198?}? A These examples raise several general issues. 1|What advantages did original entities derive by joining in groups? How did natural selection 7 __ force the common interest of a group‘s members in their group‘s we : :—' What circumstances would lend this selection such power that the identity 5 the individuals involved is almost lost in that of their group? Commume of interest: Its Origin and Preservation ADVANTAGES or GROUP LIFE. AND SYMBIOSIS Joining others offers two kinds of advantage. The more familiar, which e: plains the origin of most animal societies, is safety in numbers: more eyes I' share the watch for predators, more teeth and claws to help defend ' ' against competitors (cf. Kitchen and Packer, chap. 9}. The other advantage: ,. the complementation ofdifferent functions or, if you will, a mutually I v ; cial division of labor, such as corals and their zooxanthellae gain from ' i ' biosis, and plants and their mycorrhizae, pollinators, and seed-dis rt gain from their partnerships (cf. Douglas 1994). The genes of a {an I-I-T share a common interest in each other’s presence because each gene i.- grarns a process that benefits the carrier on whose reproductive success '. depend (Seathmary. chap. 3}. The community of interest between a cell :Ii' ROLE OF THE "COMMON GOOD” - 13 its organelles like that among the cells of a metazoan. is founded upon the advantageous way the functions of the different parts of the organism com. plement each other. PROTECTING THE COMMON INTEREST A mutualism, however, is simply a reciprocal exploitation (Herre, chap. 11}. One or more partners to a mutualism might benefit, at least in the short term, by parasitizing the others, so that the relationship no longer serves the com- mon good of the parties concerned. What factors can preserve the common grind against such threats? I will outline several of the relevant processes in the context of the major transitions to which they presumably gave rise. SELF—SUFFICIENT COMMON INTEREST Mixed bird flocks—a characteristic feature of lowland tropical forest (Moynihan l96t'Zl—remind us that common interest sometimes suffices of itself to maintain a mutualism (Maynard Smith 1991a; Leigh and Rowell 1995). In a mixed flock, a pair of each of several nuclear species, sometimes accompanied by their young, forage together in a jointly defended territory (Moynihan 19W; Gradwoh] and Greenberg 1980'). Such a group often prog- resses over a regular beat. As it does so, a bird on a smaller territory may join when the flock enters it. Some birds with larger territories may move from flock to flock {Gradwohl and Greenberg 1980; Mann 1935) The ad- vantages pf flocking are more eyes to watch for predators and the prospect of coordinated defense (mobbing) against smaller predators (Moyrrihan. 1962, [9?9; Willis 19".!2, pp. 135—1470. Social relations within mixed flocks are sometimescomplex (Moynihan 1962), and there appears to have been :flm;coevo]utron among member species of some flocks to facilitate flock- “:Et arlc'tavror (Moynlhan 1963)._Nonetheless, mixed flocks are assemblages mem betooéomosely knit to function as units of selection. No mutual enforce- Pmsenr yon _ e natural tendency of each bird to exclude conspecific strangers es LlllS mutualism. In sum, it mixed—species bird flock sim 1 ex- Pl‘eBsaes the mutual benefit of safety in numbers. p y ' spfig'gpesgpltltatlrng genes to be carried from one individual to another’s off- bemeén m I reproduction creates opportunities for a variety of conflicts: [993]}; or aes (Lessells, chap. 5}. between a mother and its fetus (Haig older young (Godfray, cha . 6) and am til (we below}. Nonemel p . _ orig e genes of a genome mmmon interest of “:55, sexual reproduction “usually reflects the unenforced cit-Eu could unajdfll (%ll?glflsdlifil;ssl;l producrng more varied offspring than key of htinterest can maintain more elaborate mutualisms. In a mon- munkey mat c;s;hm:?:$a;?a defense against predators is essential, a group member by farlrng to play its part, 20 - LElGH endangers its own life {Leigh and Rowcll 1995). To begin with. it has dimin-r '_ ished the group on which it depends for its own safety. Moreover. insofar as predators return to groups that they have already raided with success. this monkey may have hastened its own death. Under such circumstances. the. ' advantage of an individual and the good of its group coincide rather closely,-_ at least if such deadbeat monkeys cannot escape the results of their misdo-JI ings by migrating to other groups. This proviso reminds us that social orga-._ nitration plays an essential role in aligning an individual’s advantage with the. good of its group. SELECTION AHONG GROUPS AND THE COMMON ENTEREST or A onoue‘s MEMBERS r' The circumstances that allow selection among groups to override selecti- within groups are quite restricted. Sometimes, however, groups are organi a so that selection among groups enforces the common interest of their - ' bers. The effect of selection among individuals within a group on a me ' characteristic is the intensity of selection on that characteristic (the re sion of log fitness on that characteristic’s magnitude) times the vari ' : available for selection. as measured by the group‘s heritable {additive netic} variance in this characteristic [Price 1910}. Likewise. the effective of selection among groups is the intensity of selection among groups ' the heritable variance among group means (Price 19372}. Selection . groups is usually less effective than selection within groups because vari among groups is usually smaller than variance within groups. Exchange migrants among groups and the joining of emigrants from two or more I---_. ent groups to found a new group both reduce variance among. relative E variance within. groups. Moreover, groups often live longer than their rm:- ponent individuals. making selection among groups slower. Selection = n "-3 groups can contend with an opposing selection within groups on equal :2. lug; that is. a given percentage increase in the average group’s total number offspring groups can counterbalance the same percentage increase in lif' II" reproductive success of the average individual, only if the groups are I nized in such a way that (1) less than one migrant is exchanged per groups per group lifetime; [2] a new group nearly always has a single - :I L" group where all its founders were born; and (3) there are more groups II I individuals per group {Leigh 1933. 1991). If we treat organelles of a particular kind within a cell as a group. satisfy conditions that make selection among groups decisive [leigh 1 : Organelles do not migrate from one cell to another. A mitotically - u n . '5- cell obtains its organelles from its “parent.” Finally. zygotes almost a] = inherit organelles of a given kind from one parent [Eberhard 1980). might organelles have acquired the social characteristics rendering them ceptible to group selection? The tendency of mitochondria to defend ROLE OF THE "common coon" - 21 egg against nonspecifics invading with sperm {Eberhard H.380] suggests that when protorrutochondria originally invaded their ancestral hosts, presumably as parasites (Margulis 1993). they benefited more by caring for their current has: than by increasing their transmission to other hosts at their current host's expense. Thrs circumstance reflects the benefits of the complementa- tion of functions between protomitochondria and their hosts (Blackstone [995}. Were it not for this fundamental community of interest between proto- Drganelles and their hosts. the circumstances pcn‘nitting group selection on mitochondria would never have evolved. The territorial behavior of organ- elles. prompted by interests in common with their hosts. set the stage for a group selection In their host cells‘ interest. lOnce this stage was set. the effectiveness of this group selection was sealed by selection on their hosts for anisogarny. which facilitates uniparental transmission of organelles to zygotes {Cosmides and Tooby 1981; Hurst 1995. 19963). The fate of organ- elles was further identified with that of their hosts by the transfer of certain organelle: genes to the nuclear genome (Trench 1991}. I Selection among groups also must have played a crucial role in the evolu- [10" of multicellnlar organisms. Metazoans. like vascular plants. all descend from sexually reproducing ancestors- Ancestral multicellular organisms pre- sumably arose as aggregates developed mitotically from sexually produced zygotes. Therefore, variation within aggregates was lower than variation among aggregates—provided that cells could not migrate from one aggre- gatelto another—so that selection among aggregates swamped selection wrthrn aggregates (Maynard Smith and Seathmary 1995). Both sexual repro- duction and the rarity of invasion of self by nonself were crucial to the evolution of multicellular organisms. Transforming these aggregates into EEEEtne individuals. however, required the evolution of features suppress-mg Lowgpits between organisms and their cells. or limiting their consequences. aggre arig somatic mutation rate and limiting the total number of cells per “9ng h: restrict genetic variation within aggregates [Michod I99Ta}. Buss Stages of sejzpugpelifrdl 1n many‘metazoans. maternal conn-ol of early mg": cc” lines a d m I sequestration of the germ line hmit damage from can walls kee n at in plants, which cannot sequester. germ lines, rigid however, halde Euzfpell-lmes from spreading. Selection among aggregates. A son of Election :ritrve for such refinements to evolve. at the very origin of fife Egg groups may also have played an essential role RNA OCl'iljst and Hogeweg 1991]. “In the beginnin ” Sequences capable of both serv' ' ' g. appear to have funcfi mg as catalysts and bemg replicated oned as protogenes (Maynard Smith and Seathrnary 1995). Becau ‘ had to be so proofreading enzymes were yet to appear. these sequences short enough so that selecti ' ' mummafi ’ ' on could keep replication errors from each Hg (Ergen 1992. Szathrnary, chap. 3). Thanks to the small size of on - B: an array of dlfi'erent RNA quasi-species with complementary func- 22 I LEIGH Lions were needed to form a self-replicating network. This network was first imagined to be a hypercyct'e. in which A enabled B’s production. B enabled - Cs, and so forth up to the final product. say F. which catalyzed A's produc- tion {Eigen et a]. 1981). Seathmary (chap. 3) now thinks such metabolic networks were unlikely to be hypercycles; he thinks of them as miniature ecosystems of mutually dependent catalysts. The compartmenting of hyper- -' cycles or metabolic networks into prorocells allowed a selection among protocells to enforce the connnon interest of a network‘s constituent proto- . genes or quasi—species in the good of their network (Frank I996; Szathmary. I chap. 3}. KIN SELECTION It sometimes pays genes to program their carriers to sacrifice reproductive output in order to help a relative if this assistance enhances the reprodu .17. of this related carrier. Such sacrifice is favored if the relative’s gain in repro-1 duction. times the correlation between the genotypes of the relative and 1:3" original carrier. exceeds the sacrifice in the carrier’s reproduction (Hamiltori, 1964a; West Eberhard 191'}; Bourke and Franks 1995}. Students of kin .~ lection love to cite Haldane‘s supposed willingness to lay down his life It save two {full} brothers. four half sisters, or eight first cousins {presumahl the ages and future reproductive prospects of each of these relati a". were comparable to Haldane’s own). Indeed, selection among groups can a - viewed as a form of kin selection. for selection among groups is efficaci .. only if the intraclass correlation among the genomes of fellow members of F:- group is high enough so that an organism benefits by helping other mem of its group in contests or competition with outsiders (Crow and Acid I932]. Kin selection played a crucial role in the evolution of insect soci'r (West-Eberhard 19'Ir'5. IQTS; Keller and Reeve. chap. 8]. Where there safety in numbers. or benefit from reusing old nest cells or building I ' ones upon the old. wasps may benefit by nesting in groups. Group 11 ' among related individuals of formerly solitary species was the first step the evolution of social wasps, and probably the first step in the evolution - all types of social hymenoptera (West-Eberhard 1973}. Nesting in y I I I enhances competition among group members. Winners may eat losers‘ e :-‘- or prevent losers from laying. Dangers of nesting alone may be such ' ever, that if the loser is related to the winner. the loser may do best I" The characteristic cycle of instincts in solitary wasps—egg development I - 7‘ nest building. followed by provisioning the egg and larva once the ovary tween winner and Inserts) [West-Eberhard 198?}. If a subordinate’s ne laid egg is eaten. or if a subordinate has resorbed an egg she was not allo ' ' HOLE at: THE "common coon" - 23 w lay, the empty ovary may enhance that subordinate‘s instinct for provi- sioning eggs or larvae. which can be easily diverted to the young of the domjnanr, neglected because the dominant mother is attending to aggressive interactions and nest defense, attention that also protects the nest from poten- tial predators. The more essential it is to nest in groups. the greater the Pmporfion of the reproductive output dominants can monopolize without provoking losers to leave {Reeve and Nonacs 1992; Keller and Reeve 1994b. chap 8). The dominant must also be clearly stronger than subordinates. to avoid me prospect of subordinates fighting to the death to take over the nest. other processes. however, must come into play before kin selection can fa- vor the evolution of truly complex insect societies. As we shall see below. both ecological constraints and social organization must be arranged so as to suppress the possibility of appreciable reproduction by workers before a clear-cut. complete division of labor can become consistent with the com- mon interest of an insect society's members. MUTUAL ENFORCEMENT A group’s members can enforce the common good by punishing members who violate it (Trivers 191'1). Axelrod (1934) modeled the feasibility of mutual enforcement by the game of “iterated prisoner's dilemma." Consider a group of two. At each play, a member can cooperate or defect. If both cooperate. each earns three points; if both defect, each earns one point; if one cooperates and the other defects. the defector earns five points. and the cooperator, none. Here, we have a potential “tragedy of the commons" [Hardin 1968}. Whatever the opponent does. defecting earns more at any one play; yet, if the participants play each other repeatedly. they share a connnon interest in continual cooperation. Cooperation is best enforced by a strategy related to tit for rat: Cooperate on the first play. and at play It. do as the opponent did at play a — l (Axelrod and Hamilton 1981). In an error-prone world, "win-stay. lose—shift" is more reliable. Here, the rule is. if the nth play earned at least three points. repeat it the next time. otherwise shift (Nowak and Sigmund 1993). Such strategies fail, however. if players can easily Change partners. To ensure cooperation. individuals must be penalized for Chansng partners. Hamlets. simultaneously hermaphroditic coral reef fish, avoid expending half their reproductive effort on male functions by trading eggs for each Other to fertilize. Trading eggs avoids the need to produce excess sperm or to fight for mates {Fischer 1981). Hamlets pair off at spawning time. Members Of a Pint exchange sex roles in successive spawns [hence the egg—trading). If“ the exchanges continue. each fish offers more eggs for its partner to hertlllzc. as if it were becoming more confident of its partner’s good faith. If. “we‘v'flr. a fish tries to play the cheaper male role for two successive 24 - LElGH ends, and the cheater is penalized by the time required I spawns, cooperation to find and inspire confidence in a new mate (Fischer 1983). In Monera—bacteria and bluegreen algae—genes are arranged in a single circular chromosome. which expresses the common interest of its genes in each other’s presence. This community of interest, however, is not unlirrtitedfi Genes for bacteriophage it are sometimes part of the bacterial. chr0rnosome,_ where they are well-behaved, but sometimes they leave the chromosome and , multiply explosively as independent virus particles, killing their host cell in ' order to infect others (Watson et al. 1987, pp. 51Tf’f]. A gene’s ability to- spread independently to other cells undermines its community of interest with the rest of its genome, just as the community of interest among a: group‘s members would crumble if each could easily move to other groups; A gene can create a conflict with its genome by becoming a “segregation- ' distorter," that is, biasing meiosis in its own favor, to spread itself its population. The conflict becomes manifest when segregation dist ‘ '- spreads an allele that harms its carriers [Lyttle 191?). How can the f' of meiosis be enforced? Alleles on different chromosomes segregate ' . pendently of the distorter”. None of these alleles can “ride the distorter" coattails.“ If the distorter inflicts a phenotypic defect on its carriers, all u -- alleles will suffer from it. At any of these unlinked loci, selection fa u':_ mutants suppressing the distorter, for they spare some of their deacon : __'j from the distorter’s defect that they would otherwise have inherited (Prout- al. 1973]. In this sense, a genome's genes have a common interest in meiosis (Leigh 1911, 1991). Selection for such suppressors is effective tle 1919}. Suppression of successive distorters appears to have el' II most of the possible means for biasing meiosis: How else are we to stand the notorious difficulty of selecting for changed sex ratios '1 , species with chromosomal sex determination (Maynard Smith 1918', ‘ liams l9't'9}? Nonetheless, other levels of selection must have been invol {I in the spread of honest meiosis. Surely those lineages whose species ' - " less susceptible to segregation distortion lasted longer and radia successfully (cf. Nunney, chap- 12). _ Truly complex insect societies can evolve only when subordinates -= I I'i benefit by producing young of their own. This condition ensures that a 4" 1 ordinate‘s only hope of spreading its genes is to help its dominant re ==- '{ reproduce [West—Eberhard 1975). Colonies of insects whose su ' seldom or never try to reproduce on their own, such as honeybees, array =3 and leafcutters, are marvels of self-organization. where each worker forms its appointed tasks automa 'cally. without any trace of compulsion even direction by the queen (E. 0. Wilson 1930: Franks 1939; Seeley l ' ; Honeybee queens create a situation in which workers make it unprofi 3" for each other to reproduce by mating with up to 20 males and. mixing ':. ROLE OF THE "COMMON GOOD" I 25 sperm thoroughly. Thus. most of a worker’s colleagues are half sisters. A worker is more closely related to its mother’s eggs than to a half sister‘s, so it eats eggs latd by half sisters. Mutual policing by workers makes it point- less for thern to lay eggs and thus creates a cotmnon interest among them in helping their queen {Ramieks and 'v’isscher 1989). The common interest of chimpanzees, Pan troglodytes, in their group’s welfare leads them to enforce the rudiments of morality {de Waal 1996). Chimpanzees can recognize each other and have the ability to imagine how they would act If in the circumstances of another individual, a self- awareness most clearly evidenced by the dcceptions they practice on each other {de Waal 1996}. These abilities enable chimps to do unto others as the others have done unto them, and to expect others to do to them as they have done to those others (do 1llllaal 1996. p. 136}. Chimps have a clear sense of grati- tude for. and willingness to repay, those who have done them favors a desire for retribution toward stingy troop members that do not share food and a desire to exact revenge if wronged, a desire whose excesses the troop dominant 15 expected to restrain [de Waal 1996, p. 161). These attributes suggest that chimpanzees have the rudiments of a sense of justice. The mutual enforcement of morality among chimpanzees is manifested in various ways. At the Arnhem Zoo, a whole troop of captive chimps under- took to chase and thrash two adolescent females who delayed the feeding of the-troop for two hours by refusing to return when the troop was called back to Its shelter one evening [no ape is fed until all retum). They learned The next evening they were the first to retum (de Waal 1996, p. 89} (Shim- panzees also expect their dominants to be fair, and to protect the In :1;ng troop, a young male who, with the help of an older colleague. fights-EH e troop s donunant,‘ showed improper bias when he interceded in femaié fe Invartably Sided with his colleague and with a few high-ranking mm coarltcnds. Because this dominant was clearly not living up to expecta— afle; Hisltlltgns of females prevented him from intervening in fights there- was .acce c- der colleague. who settled fights in a fair and restrained manner, Who fun ptc as med1ator instead. Thus, the chimpanzee group has a say in mower cttons as mediator and how he does it (de Waal 1996, p. 130). In fgumsSli‘lrgzlrpéhve group, females restrained the dominant male from taking [mg of his fvenge on a young male whom he had discovered mating with wdmme a Ton“: ffimales (dc. 1996, p. 9'1}. More generally, chimps been in a finmizlehifate reconcrltattons between troop members who have hles for mfitionccmapllpgfl [13205]. Indeed, even though there are squab- Drganizafifln of limit group Mam 1:11;: :fcommon interest tit-the hierarchical resources, keeping the peaée, and orgmgpggwtpgk ft: cpordtnattng access to tors or competing groups (de Waal 1996, p. 133).gI up 5 response to mada- 26 - LEIGH - Understanding Genetic Conflicts Clarifies the Study of Evolution GENETIC CONFLICTS, NATURAL SELECTION, AND THE Brorocv TEACHER One of the Scandals of biology teaching is how the directing role of natural- selection in adaptive evolution is usually argued. The student is told that“ evolution is driven by. or consists of. changes in allele frequencies in popu- . lations (some would find even this an unnecessarily contentious remarkfi Then the lecturer enumerates the possib ' ' quency: sampling error (genetic drift), mutation, migration, and natural so: lection [differential replication), and concludes that, because the only one mg: these four that can lead to adaptation is natural selection, natural selection ii the cause of adaptive evolution. This logic leads to the extraordinary stance whereby “To buttress the theory of natural selection the same ' . stances of ‘adaptation’ [and many more] are used, as in an earlier but distant age testified to the wisdom of the Creator and revealed to s' piety the immediate finger of God" [Thompson 1942, p. 963'). An exercise in. “logic” has transformed a mechanistic hypothesis into a dens ex machine. The analysis of genetic conflicts. however, enables one to look for fr a: prints of the decisive role of natural selection in adaptive evolution. 7 selection among groups to achieve a major evolutionary transition, such =7 the evolution of eukaryotes or the transformation of multicellular aggteg into true metazoan individuals, certain conditions must be met {Leigh 1 _ 1991}, Migration among groups must be annulled. Either each group must i"- founded by migrants from a single parent group (as in the uniparental I =II'r-_ - mission of organelles of a given lrind to a zygote, very probably the ': 5 condition for all sexual eukaryotes; cf. Hurst 1995), or there must be 1-H", other means to ensure that among-group exceeds within—group genetic v ’ 9 ance [such as the sexual production of zygotes, which then develop uIII-I tically into organisms of many, genetically identical cells). The means :.. which conflicts between cells and their organelles (Eberhard 1980) or H -_ flicts between individuals and their cells (Buss 1987', Maynard Smith :- ": Szathmary 1995) are suppressed or minimized provide nonnatalrable ‘ prints of the decisive role of natural selection in these n-ansitions. evolutionary transition has left traces of the genetic conflicts which that cans by which these conflicts were su - ' For that reason, Maynard Smith and Szathrnary’s study of the major I tions in evolution represents the first book in which evolutionary ' testifies to evolutionary mechanism. Their book shows how to remove "-5 argument for the directive role of annual selection in macroevolution I I; ' alytic logic to that of empirical observation. ROLE OF THE "COMMON GOOD" - 2? HIERARCHIES AND EVOLVABILITY The greatest stumbling block for laymen (and many biologists) trying to undersmnd the theory of evolution by natural selection is seeing how natural selection of random mutations could lead to the complexity and precision of adaprarion characteristic of living organisms. Quite distinguished minds find this idea an oxymoron (Polanyi 1953', Gilson l9?l; Fabre 1989). What char- acteristics of organisms are responsible for “evolvability”? What features allow living things to evolve by natural selection of random mutations? One essential feature is modular organimtion, which allows mutation or selec- tion to affect one feature of an organism without interfering with the others [Wagner and Altenberg 1996; Gerhart and Kirschner 199?). THE V'IRTUES 0F MODULAR ORGANIZATION "f or adaptive evolution. He imagined a system whose actual state was specified by a point X = x1, x2, . . .x n in an Edi—rnensronal space, whose optimal state was specified by a different point _ 0.. 02. . . on, and whose fitness was a monotonically decreasing func- tion of the Euclidean distance |[X —0||, the square root of the sum (11-0 )2 + {13—02}- + . . , + [x,,— ")2. Now imagine a mutation a randdm change that shifts the system’s state from X to X + r. The probability that X + r is more fit than X is the proportion of the sphere of radius ||r|| about X enclosed wrthin the sphere of radius llX—DII about 0. This probability is the :naller the'larger ||r|i relative to IIX-OII. For a given ratio of these variates. mtinppilz-aaility of improvement is smaller the larger n, that is, the more “on philih ef characteristic affected by the mutation. The modular organiza- essehfial 01 genes themselves and the characteristics they affect, plays an one feat“:- e tribemaking adaptive evolution possible, for modularity allows p‘ 23'} wa to changed without changing anything else {Lcwontin 1973, flow] :E'elecgner 1996: Wagner and‘Altenberg 1996]. Indeed, episodic direc- the “:51 0f ltrilon on one characteristic combined with stabilizing selection on mauve: indc phenotype could favor transforming that characteristic into a guns to :0 ependent module (Wagner 1996). The principle here is analo- 30 that meipér s {199]} argument that societies must be changed piecemeal, Moreover tleft of each change can be assessed with minimum ambiguity. can cause [file 3 capacity for accommodating a variety of phenotypic insults ganism w ad‘ugl‘lI-JUS, independently programmed characteristics of an or- changein onejof thln' extraordinarily appropriate ways to a major genetic story of a goai fi on number. Maynard-Smith (1958, pp. 39—280] tells the Useless by a InLariat studied by E. J. Slljpet', whose front feet were rendered “my Eda r tion. This goat walked bipedally, which led to an extraordi- P We series of rearrangements in its skeleton and musculature. This 28 ' LEIGH capacity for accommodation among modules raises the question of what ‘ sorts of evolutionary jumps are possible when the environment is permissive enough. HIERARCHIES, LEVELS OF SELECTION, AND EvDLVABlLlTY Students of hierarchies like to tell the parable of the watchmaiters (Leigh I and Rowell 1995; Seeley 1995; Wagner 1995]. One constructed his watches , in a modular manner, funnng subassemblies of ten basic parts apiece, as». semblies of ten subassemblies apiece, and so forth until he had finished his watch. The other dispensed with subassemblics. Both watchmalters were.- subject to frequent interruptions. When interrupted, the one only had to start over on his current subassembly, whereas the other had to start from scratch; .. Naturally, only the modular watchmaker finished any watches. A more apt-3 posite story might be the organization of genetic algorithms, computer pro; ; grams simulating natural selection of random mutation to solve complex: optimization problems. The most workable of these programs successively evolve partial solutions serving as building blocks. which can then be ooma't' bined to generate the final solution (Wagner 1995). -' Evolution seems to have followed similar paths. Time and again. major? evolutionary transitions occurred when larger wholes formed from smith.- “ready-made" parts that had already been tested by natural selection (Ma, nard Smith and Saat‘tunéry 1995). Thus, organisms are modules of n u m of modules . . . and so on toward Pascal’s infinitely small. Even genes ' ' ' to be combinations of domains IUD—3&0 nucleotides long (Eigen 1992, 22), which is about as big a gene as can be readily optimized by an. selection of chance mutations (Eigen 1992, p. 30). Eigen believes that -_'. acquired their essential features when they were much more mutable u -n' now—before they cohered into chromosomes. T Modules that are themselves living creatures, or at least units of se - ' ' 7' have distinct advantages. When complementation of functions u no: mutualisrn between cells and their organelles, organelles played the role '7 self-designing macromutations for their host, macromutations with u r r a. u-l beyond the wildest dreams of a Goldschmidt (1940) or a lovu'up (1916). analogous capacity for accommodaan applies to animals in societies. 77 previous section mentioned West—Eberhard's (198T) description of how in". complex and beautiful division of labor is built on the varied reactions ', solitary individuals to different environmental conditions. This saine v: -- '5' ity for accorrunodation made it possible for Smythe (1991) to create in 7 generation a breed of social pacas from what is naturally a fiercely "- species whose adults live in pairs, by suitable adjustments in the rearing ?_ the newborn and very young. This capacity for self—design in the interests . their group. among parts that have not yet ceased to be units of selection l; ROLE OF THE "COMMON GOOD" - 2‘3 emlufion) in [heir own fight’ mm“ have Plalr'l'fi a crucial role in many evolu— tionary transitions, LEvELs or SELECTION AND TRUTH, BEAUTY, AND GOODNESS Some theologians, like laid (1933, pp. 60-63), and some biologists, like T. H_ Huxley {1894) and George Williams [1989), place truth. beauty, and goodness utterly beyond the reach of natural selection. I agree readin enough with D'Arcy Thompson’s (1942, p. i3} remark, ‘Consciousness is not explained. to my comprehension by all the nerve-paths and neurones of the physiologist; nor do I ask of physics how goodness shines in one man‘s face and evtl betrays itself in another." Nonetheless, our capacity to distin- guish right from wrong, appreciate beauty, and know truly are related to how natural selection affects social beings. A previous section discussed de Waal's (1996} evidence for protomorality among chimpanzees, and its foun- dation on the troop’s sense of justice, which, in good Aristotelean fashion. serves the troop’s common interest. Whence comes a sense of beauty is an odder issue. Sexual reproduction creates opportunity for members of one sex to compete for matings with the other. One consequence of such sexual selection is the evolution by males of characteristics that attract females: The peacock‘s tail, the blazing colors and thnarg-ipus plumes of birds of paradise, and the like (Darwin [359, p. 39). n s assertion that male birds competed by appealing to the aesthetic sense of females of their species caused some offense. Nevertheless, human beings prize the colors. shapes, and sounds that many animals use to attract mates, and the colors. shapes, and scents by which flowers attract pollina- tors. Moreover. it is often essential for females to judge the beauty of males anght, for a male’s beauty is often a good index of his health and suitability as a mate I Hamilton and 21:]: 1932; Saino et a]. 199?}. do not need complete knowledge, but what they do know, they food and m truly enough to avoid predators and other dangers and to find of a mind Cpres (Lorenz 191'3]. Darwm’s words of doubt about the efficacy Jam 1933 esggnded from a monkey‘s. oft quoted by anti-Darwinians (cf. hm l. truss the point. Even we may not know “things as they pmdifl [hair imust know enough about some objects in our environment to actions on thempaxfil; on us, or on each other. and to judge the impact of our the animals tho; e same rs true, to a lesser extent, of other animals. Even reveal their prey eithiriptrc must generalize about the features that guise their head and 13 51-3513; measures by which cryptrc insects dis- [RDbinson 1935} Th 3. I enses against just this power to generalize . e capacrty to predlct and generalize must be equally so - LEIGH: important in social life. Even a chimpanzee‘s deceptions depend on its ahi]... ity to predict how best to deceive its intended dupe. which appears to suppose an ability to imagine how it would respond to its acts were it in 1;qu dupe's place {lolly 1991). Concluding Remarks Adaptive evolution presupposes modular organization. indeed, a precise rm; derstanding of the nature and history of the modular organization of It ' things is needed to assess their potential for adaptive evolution and It.:t'-' reveal that living things are organized to facilitate their evolution by us | -1' selection of “random” mutations (Leigh 1931']. -r The most objective mark of evolutionary progress is the series of evol I tionary transitions where parts combined to form larger. more effecti 1 wholes (Maynard Smith l988}. Each such transition involved potential that between different levels of selection. These conflicts, and the ways I: are resolved. comprise one of the grand unifying themes of biology. _- Parts join to form larger wholes only if there is a genuine community interest among the parts. and if circumstances allow the enforcement of m common interest. In the major transitions of evolution1 community of n a; est plays the same crucial role as in Aristotle’s Politics. The traces of the means by which conflicts betwoen levels of selection :-.;. resolved in favor of the higher level represent unmistakable footprints of I- i; decisive role of natural selection in macroevolution. These traces are '_ stances where evolutionary history testifies to evolutionary mechanism. - _ Finally, evolutionary studies of social animals suggests that truth, r : and goodness are not totally beyond the reach of evolutionary biology. The First Replicators Eors Szathma'ry The replicator concept of Dawkins (1976) has turned out to be extremely useful in analyzing evolutionary questions. Here I follow the definition of Hun (1980}. who emphasized that replicators must pass on their structure [El-gay intact. Although selection acts on them directly. interactors [such as organisms) do not qualify as replicators because their structures are not cop- ied. I shall come back to this important conceptual issue at the end of this chapter because organisms usually qualify as reproducers. My primary interest here lies in the origin of the earliest units of evolu— tion. Entities qualify as units of evolution if they meet the following criteria (Maynard Smith 198?}: l. Multiplication. Entities should give rise to more entities of the satne kind. 2. Heredity. Like begets like; A-type entities produce A-type entities; B-type entitles produce B-type entities. etc. 3. r Variabilinu Heredity is not exact; occasionally A type objects give rise to A type objects (it may be that A“ = B). If objects of different types have a hereditary difference in their fecundity audior survival, the population undergoes evolution by natural selection. To explam the origin of life, we need to explain the origin of heredity in terms of chemistry. Heredity merely means that like begets like. This in tflu'n, requires variation: multiplication of an entity that can exist in only one bzrrrrllaflgi not constitute heredity and could not form the basis of evolution luuon re ec‘t‘ron. [argue that mere heredity is not enough. Ongoing evo- can exisgurres unlimited heredtty," that is. the existence of replicators that bemw h Indan indefinttely large number of forms. Although. as I outline co _. ere rty with a small number of possible types can exist without Dying, 1!; seems very probable that unlimited heredity requires tern late Wfing of replicators with a modular structure. P em) :{rlipejrrment relevant to the origins of replicators (and life in gen- chemjst He f e [put more than a hundred years ago by Butlerov. a Russian hows “mar fit; that, tf formaldehyde is kept in a reaction vessel for a few me “fomosc crately alkaline conditions, sugars readily form. Nowadays Sims of sugarsteactron appears to be a_ formidable network of interconver— Caimsvsmjm afinépng them ribose, winch is a building block of RNA (e.g., a suffidcm mm alker 197M]. Even more interesting is the fact that. given exponential kin ttitnt of formaldehyde. the accumulation of sugars follows I is the s“ e cs. indicating that something is replicating in the solution. grits that replicate: The “hard core” of the reaction is the cyclic, ...
View Full Document

This document was uploaded on 08/24/2011.

Page1 / 9

Leigh_Conflict_1999 - l4 - REEVE AND KELLEFI .' that have...

This preview shows document pages 1 - 9. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online