Historical Patterns in the Study of Avian Social Behavior ’

The Condor 96232-243 0 The CooperOrnithological Society 1994

HISTORICAL PERSPECTIVES

HISTORICAL PATTERNS IN THE STUDY OF AVIAN SOCIAL BEHAVIOR ’

JERRAM L. BROWN Department of BiologicalSciences, State Universityof New York, Albany, NY 12222

The songs,calls, plumagesand other behaviorsofbirds ably for many disciplines. I call this the fission-fusion typically reflect their social relations. The study of so- pattern of history. cial behavior, therefore, is basic to our understanding of the most conspicuousand endearingfeatures of birds. THE “MODERN SYNTHESIS” Our knowledge of these behaviors has expanded con- A well-known example of the fission-fusionpattern is siderably in the hundred-plus years since Darwin, but the fission of evolutionary biology in the early part of his books on the origin of species,sexual selection, and this century into the specialtiesof population genetics, the expression of emotions still provide the base to systematics,and paleontology,followed by their fusion which most modem studiescan be traced(Darwin 1859, as the “modem synthesis,” as it was termed by Huxley 1871, 1872). This review provides a historical per- (1942). This interpretation of evolution reaffirmed the spectiveon someofthe ways by which natural selection Darwinian view of evolution by selectionof small vari- in the broadest sense has been thought to influence ations among individuals, and it strongly influenced social behavior in birds. The first section of this paper virtually all of the senior evolutionary biologists (in- reviewsthe major trendsand eventsfrom 1859to 1992. cluding bird behaviorists)of today. Its major agentsof In the secondsection I look more closely at one aspect influence were synthetic books in genetics(Dobzhan- of socialbehavior and analyzethe logicalprocesses that sky 1937) systematics(Mayr 1942) and paleontology led to scientific progressin a period that is familiar to (Simoson 1944). This period of taking stock, mainlv me as a participant, namely from 1952-1992. I then in the 1940s. has been -acclaimedas a-major epoch in contrast the two perspectives. evolutionary biology (Mayr and Provine 1980, Mayr I have tried to be neutral and objective, but because 1993); but no synthesisor consensuslasts forever, and I have been involved in some of the issuesand con- many changeshave taken placethat resultin a different troversies during the period I make no claim to have and more controversial evolutionary biology today been completely successful.I draw heavily on my own (Antonovics 1987). experiencesbecause I know them best; it could hardly Perhaps the most important legacy of this period for be otherwise.The completely neutral participant in this social behavior was the consensus that was reached on field, in my opinion, does not exist. the subject of speciation and the origin of behavioral isolating mechanisms. This view was fully described THE FISSION-FUSION PATTERN by Mayr (I 963) for evolutionists generally. Its last ap- In Darwin’s era disciplines that we now recognize as pearance as a major part of a behavior text may have distinctly different from each other, such as ethology been as a chapter in Brown (1975). Closely related to and ecology,had not split off from the vaguelydefined the study of speciation is the study of geographical area ofnatural history, and other disciplineswere much variation in behavior, with avian vocal dialects being lessspecialized (Benson 1988). The recognitionofecol- a good example. Sociobiologists and behavioral ecology and ethology in the 1950s neuro-ethology in the ogists in the 1980s largely ignored the relationship be- 1960s behavioral ecologyin the 1970sand molecular tween behavior and speciation, but the modem focus ecology in the 1990s illustratesthe continuing fission- on sexual selection has revived some interest in the ing process.At the same time that fission by special- subject (West-Eberhard 1983). ization was progressing,isolation among the fissioning THE MACRO-EVOLUTIONARY PERSPECTIVE OF fields increasedand bridgesbetween the emergingspe- CLASSICAL ETHOLOGY cialties became rarer. When bridges did occur, however, they were likely to be important and influential. Beginning with Whitman’s (1899, 1919) comparative The overall pattern is one of fission by specialization studiesof speciesof Columbidae and Heinroth’s (19 11) followed by selective fusion and cross-fertilization. studies of various birds in zoos, ornithologists were Similar conclusions have been reached for ecology impressed by behaviors that lacked conspicuous sus- (Moore 1920, Brooksand McLennan 1991) and prob- ceptibility to environmental influence during development (“innate”). In the same period, ornithologists began experiments that would elucidate the learning of behaviors such as vocalizations and social preferences (e.g., imprinting) in some species. These studies ’ Received 25 August 1993. Accepted 13 October confirmed the early impressions that some behavior 1993. patterns were useful in classification because they re-

12321 HISTORICAL PERSPECTIVES 233

vealed phylogenetic relationships. Thus behavior be- that characterized ethology then ignored the brain, came an aid to systematics and the construction of which was the source of the behavioral phenomena the phylogenies (Brooks and McLennan 199 I), Lorenz’s ethologists were trying to study. To begin to fill this (194 1) study of the Anatidae being a classic example. gap I attempted to bring together what was then known Although popular in the immediate post-Lorenzian era about “motivation” of displays with what was known (Mayr 1958) when as a graduate student I indulged in about their neural basis, in the process introducing the it myselfin a trivial way (Brown 1959). this application term neuro-ethology (Brown and Hunsperger 1963). In of social behavior to taxonomy gradually declined in the next decade I tried to develop this new field using the 1960s and 1970s to near extinction. Systematics birds (Brown 1969b, 1969~; Ingle and Crews 1985). looked elsewhere for inspiration and controversy, My own efforts were not influential (Brown 197 I, 1973) namely to molecular biology. but exciting papers would later emerge by Nottebohm We have now reversed this relationship. Molecular and colleagues (reviewed in Konishi 1985) and by oth- systematics has become an exciting new tool to study ers on invertebrate nervous systems. The study of “in- the evolution of behavior, instead of behavior being stinct” and social behavior in birds and other animals used to study systematics. This is an example of fusion became neuro-ethological, illustratingagain the pattern across a bridge between two separate specializations. of fission followed by fusion. The construciion of a phylogeny based on presumably neutral genetic variation is now used to studv the aae THE POPULATION BIOLOGY OF BEHAVIOR , L and branching patterns of avian behaviors, such as The second major gap in ethology at the time, as re- broodparasitism (Lanyon 1992) helping(Edwardsand flected in Tinbergen’s books (1951, 1953b). was the Naeem 1993) and foraging specializations (Richman lack of a population biology of behavior, which de- and Price 1992). veloped rapidly later in the 1960s (see below). No doubt Ethology in its early days guided by Lorenz was a Tinbergen and his students were early contributors in Germanic discipline. One had to learn German to read this area (Dawkins, M. S. et al. 1990) but in my opin- the famous Kumpan paper (Lorenz 1937, 1970) and ion the principal impetus in this area among American most of the important literature was in the German students, such as Orians, Fretwell and myself, came language. To become an ethologist one had to visit from population biology, not from ethology. Germany, as did Margaret Morse Nice, author of a In 1975, three books appeared that emphasized a landmark behavioral study of the Song Sparrow (Mel- new perspective among textbooks on behavior, that of ospiza melodia;Nice 1943) in 1938. All this changed natural selection. Previous texts had emphasized the when Tinbergen, a multilingual Dutch ethologist, ac- developmental and physiological “Mechanisms of An- cepted a chair at Oxford, England and began an im- imal Behavior” (Marler and Hamilton 1966) or had portant series of publications in English. For Americans, attempted a “synthesis of ethology and comparative it happened just in time; for his subsequent books, psychology”(Hinde 1966, 1970). In contrast, the “cen- “The Study of Instinct” and “The Social Life of Ani- tral unifying theme of biologicalevolution ” character- mals,” were extremely influential for ornithologists in izes Brown (I 975:~~) who wrote that “the central con- the 1950s leading to establishment ofanimal behavior cepts in this book are concerned with populations.” with an avian emphasis in the curriculum of many Similarly, although the primary message of “Socio- American universities. (It was not taught at Cornell biology” (Wilson 1975) was interpreted by social sci- University, where I was a student, until about 1955 or entists to mean an emphasis on nature rather than at Berkeley until I957.) A history of the study ofanimal nurture (Barlow 1989, 199 1). for behaviorists it was a behavior in North America has been written by Dews- heavy dose of population biology. Alcock (1975 and bury (1989). Tinbergen, besides being one of the kind- later editions) has also emphasized natural selection est and most likable persons I have ever met, was an rather than physiological or developmental mecha- expert field naturalist and loved to work on the social nisms. The 1970s were a period of major synthesis in behavior of gulls (Tinbergen 1953a, 1959). Tinbergen. which bridges between behavior study and population Lorenz, andion Frisch were honored by a Nobel P%ze biology were built, at some cost to the study of mech- in 1973 (Marler and Griffin 1973). Festschrifts for Lo- anisms and development. renz (Schleidt 1988) and Tinbergen (Baerends et al. Although these three books had some influence of 1976; Dawkins, M. S. et al. 1990) and brief autobi- their own, they clearly reflected a trend that was already ographies (Dewsbury 1985) provide further details on substantial. Even without these textbooks the trend their lives and scientific contributions. The history of toward fusion of population ecology, genetics, and be- ethology has been described by Klopfer and Hailman havior was booming and would have continued. The (1967). Thorne (1979). Beer (1963) and Barlow (1989. unprecedented front-page promotion of “Sociobiolo- i 99 I):‘Some’early papers in kthology are reprinted in gy” in the New York Times (May 28, 1975) beforethe the collection of Burghardt (1985). book had even been seen by most “sociobiologists” While ethologists deserved their honors, I was per- and the Sunday-supplement treatment by the Times sonally disappointed at the time by two great gaps in (Oct. 12, 1975) made this scholarly book a media event the ethology ofthose years. As a naive graduate student and consequently a political issue (see Barlow I99 1). I chose as a subject for my Ph.D. research a topic along Nevertheless, anthropology had already absorbed les- the lines of students of Tinbergen, namely the analysis sons from the fusion of ethology and ecology (Tiger of the “motivation” of the displays and -vocalizations 1969, Tiger and Fox I97 I). The animal nature of “the of an avian species (Brown 1964b). This allowed me naked ape” had intrigued the common man (Lorenz to realize that the analysis accordingto conflict ofdrives 1963, Morris 1967) well before Wilson’s tome. 234 HISTORICAL PERSPECTIVES

What then were the origins of these trends, if we thologists that was based partly on misunderstanding cannot attribute them to textbook writers? I would of the concept, as shown by Dawkins (1979) and partly nominate two sources, the population ecology of David on old-fashioned conservatism. It would take three de- Lack and the population genetics of W. D. Hamilton. cades for facts and reason to overcome this resistance. Certainly these writers influenced many ornithologists Returning to the theme of fission and fusion, I sug- for decades. Two continuing themes in Lack’s writings gest that the principal origins of what we now know as were the ecology of populations of birds (Lack 1954b, sociobiology were on the one hand from the tremen- 1966) and the evolution of clutch size (Lack 1947, dous impetus that Hamilton’s theory gave to the study 1948a, 1948b, 1954a, 1968) by individual rather than of social insects and sociality in general and, on the “group” selection (an idea with a long history before other hand, from what I have termed the comparative Lack-see Burbridge 1992). Although Lack had al- ecology qf social systems that developed in the 1960s ready convinced most ornithological readers of the su- among ornithologists. This fusion of population ge- premacy of individual selection, the opposing view fa- netics and the work of avian field ecologists in the voring population- or deme-level selection was tradition of David Lack provided at least part of the presented conspicuously by Wynne-Edwards (I 962). wave of interest that resulted in the syntheses of 1975 The theme was not original with Wynne-Edwards (see (Alcock 1975, Brown 1975, Wilson 1975) and certainly Kalela 1954) but he published a lavish and ponderous was a strong stimulus to the study of avian social be- elaboration of it. Squid-like this book propelled the havior in the 1970s. population biology of social behavior backward in a cloud of black ink. BEHAVIORAL ECOLOGY EMERGES Reaction was strong and immediate. Although some Add optimal foraging theory to sociobiology and you sophisticates chose to ignore the crude reasoning of have behavioral ecology-or close to it. Optimal for- Wynne-Edwards because it was so obviously wrong as aging theory arose from theories of niche exploitation a general explanation, others attempted to refute his written by the messiah of American ecology and one arguments (Wiens 1966). Still others felt that this was of his prominent students (MacArthur and Pianka 1966) the time to elaborate the ways by which social behavior and by J. M. Emlen (1966). MacArthur’s followers were could be influenced by individual selection. For ex- influential, and they quickly expanded this approach ample, territoriality was Wynne-Edwards’ prime ex- into a flourishing field. The-early period was reviewed ample of a population-limiting behavior, but Brown bv Pvke. Pulliam and Chamov (1977) at the time and (1964a) presented a general model that showed how by Schoener (1987) retrospectively. Although it may various kinds of territorial behavior could evolve by be argued that foraging is usually not considered to be individual selection, without the need for population- social behavior, ideas from optimal foraging theory level selection. Similarly, Orians (1969) advanced a constituted the best developed applications of opti- graphical model to explain the evolution of mating mality methods to behavior, and these methods came systems based on individual selection. Similar argu- to be applied also to conventional social behavior (Car- ments on behalf of individual selection were made by ace 1979a, 1979b). Caraco showed how the transition Crook (1965) for a variety of avian social systems. between flocking and territoriality in juncos could be Many of the papers of this era were concerned with predicted on the basis offoragingand predation hazard. spacing behavior and its effects on populations (Brown Caraco and Wolfs (1975) paper on optimal group size and Orians 1970). Together these led to a comparative in lions spawned a series of papers on optimal group ecology of social systems, still a popular field today size in birds (reviewed in Pulliam and Caraco 1984. (Lott 199 1). An even more global approach was taken and Mange1 and Clarke 1988) including one on group- by Williams (1966) who attempted to outline how territorial defense in birds (Brown 1982b). Optimality adaptations in general could evolve by individual se- theory has been highly developed for the social insects lection. These works and many others established a (Oster and Wilson 1978) and it was even applied to consensus position on the side of individual selection. helping behavior in birds (Brown and Pimm 1985); but They created an atmosphere that was hostile to any in general it has been more useful for foraging than for mechanism that seemed to differ from old-fashioned social behavior. Perhaps this was because foraging be- individual selection. This conservative view proved to havior is more susceptible to simple models and ex- be an impediment to the acceptance of some exciting periments. new ideas. The combination of optimal foraging theory and the The most influential new idea in the realm of selec- population biology ofbehavior into behavioral ecology tion thinking was the theory ofinclusive fitness launched has proven to be more viable than its separate parts. by Hamilton (1963, 1964). To prevent this kind of The field first became visible with two books by Peter selection from being confused with “group selection,” Klopfer (1962, 1970). Its first journal, BehavioralEcol- Maynard Smith (1964) coined the term “kin selec- ogy and Sociobiology,began in 1976; and its second, tion,” including both direct and indirect components BehavioralEcology,in 1990. The field attracted its own of inclusive fitness (as they were later recognized by “official textbook writers” (Krebs and Davies 1978, Brown 1980). While it was not spelled out explicitly, 1984, 1987, 199 1) and founded its own societv in 1986 confusion of group and kin selection was implicit in in Albany, NY. Other recent surveys include Morse some of the early literature attacking the use of inclu- (1980) and Siblv and Smith (1985). The social behavior sive fitness theory as part of an explanation of helping of birds received much attention in all these books, behavior in birds. Whether for this reason or for others, especially since most of the authors and editors had such as recoiling from the term altruism, inclusive- worked on birds to some extent. fitness thinking met a hostile reception among omi- Territorial behaviorand resources.Because of its rel- HISTORICAL PERSPECTIVES 235 evance to population stability, a subject that had been 1992a, 1992b; Davies 1992). This topic will be covered considered at length by both Lack (1954b) and Wvnne- in detail in a later essay in this series. Edwards (1962),-the territorial behavior of birds re- Sexual selection.A major change in the way orni- ceived considerable attention in the 1960s (reviewed thologists view sexual selection has occurred in recent by Brown 1964d, Klomp 1972, Watson and Moss 1970). years. An anecdote from my own experience illustrates With the focus of interest turned to cost/benefit the- it. In 1975 I gave modest space in a chapter on sexual ories (Brown 1964a), however, attention shifted to the selection to Fisher’s (1930) ideas and raised the pos- energetic consequences of territorial behavior (Car- sibility of the importance of female choice in birds. In penter and MacMillen 1976). An influential study of 1978 at a meeting of behavioral ecologists at Ann Ar- costs and benefits ofterritorial behavior in nectar-feed- bor this passage was cited in a negative way; for the ing birds in Africa (Gill and Wolf 1975, Pyke 1979) dogma then among ornithologists was that sexual se- was the culmination of a productive research program lection in birds was caused only by aggressive com- on nectar-feeding birds in North America (e.g., Wolf petition among males, which could be easily seen (as and Hainsworth 197 1). These studies nencrallv agreed in LeCroy’s, 198 1, explanation of sexual selection in . I with the expectations raised by cost/benefit theories birds of paradise), and not at all by competition among (Brown 1964a). males through attraction and persuasion of females. Studies on the transition between territorial and Note that in sexual selection, competition is always flocking behavior in the non-breeding season were pur- between the members of one sex (i.e., “intrasexual” sued in a variety of species (reviewed by Milinski and and never between members of different sexes (“inter- Parker 199 1, Davies and Houston 198 1, Davies 1983). sexual”). The jargon that distinguishes between these Both experimental and optimality methods were em- concepts as inter- vs. intra-sexual selection is, there- ployed. Many predictions of a cost/benefit nature were fore, nonsensical (Brown 1983b). Its continued usage tested and the theory was further elaborated with re- depends on the sheeplike nature of some authors who spect to avian social behavior. use the terms simply because some others do so, re- The idealfree distributionand resources.When ter- gardless of their real meaning. ritories are compressible, increasing density of breeders The landmark study in this area and one that killed may depress reproductive success. This depression the old dogma with a single blow was a carefully con- lowers the value of a territory in a good habitat so that trolled field experiment on the Long-tailed Widow a newly arriving bird might have better success by (Steganuraparadisaea; Andersson 1982a). The data breeding in a poorer habitat with lower density than a allowed the hypothesis of aggressive competition to be better habitat with a higher density. This trade-off be- rejected, leaving competition by persuasion and at- tween habitat quality and density was first formalized traction as the remaining alternative. This empirical into an optimality model and tested with data by Brown study together with Zahavi’s (1975) reshuffling of Fish- (I 969a). A similar trade-off, although complicated by er’s and J. M. Emlen’s (1973:5 I) ideas into the “hand- additional factors, is inherent for females in the model icap” theory stimulated much work on mate choice in oforians (1969). A year later the density-habitat trade- birds. off was again formahzed using very elementary algebra Accepting that female choice had to be taken seri- by Fretwell and Lucas (1970). who described what thev ously for birds, modelers turned their attention to the named the ideal free distribution (IFD) ofcompetitors. reasons why females preferred particular traits under Fretwell, an ornithologist, promoted the global appli- conditions where resources were not at stake. Did fe- cability ofthis concept in a book (1972). Ornithologists males prefer males with exaggerated signals because at first paid little attention to the IFD but after a con- they identified genes that would make sons superior at vincing experimental demonstration ofit in fishes (Mi- attracting females, or did females prefer such males linski 1979) it became a popular topic for modelers because they identified genes that would make both and lab tests in behavioral ecology (reviewed in Milin- sons and daughters more viable (Andersson 1982b, ski and Parker 1991). A study of “ideal free ducks” 1986) or both? Models of the former situation allowed demonstrated the tradeoff in free-living birds (Harper the initial stage of the male trait “before” selection to 1982). Its relevance to community ecology has been be entirely neutral (Lande 1980, Kirkpatrick 1982). developed at length by Rosenzweig’s group using hum- Fisher (19 15, 1930) however, thought that the process mingbirds among their test animals (reviewed in Ro- would begin with traits that were correlated with gen- senzweig 199 1). eral good condition and that females would choose on Mating systems.As cost/benefit modeling became the basis of male condition. In other words, he com- popular in the 1960s attention turned to mating sys- bined the two processes. Many authors have chosen to tems. Early work done on North American Icteridae present these theories as alternatives, thus polarizing (Orians 1961) led to a graphical polygyny threshold the field and tending to delay compromises. It seems model (Verner and Willson 1966, Orians 1969). In it possible to me that many sexual signals in birds identify a decisive role was assigned to evaluation by females males whose progeny will be both more viable and of territory quality and other conditions that affect fe- more sexually attractive. There should be a continual male reproductive success. The classification of Emlen tendency for selection to carry condition-sensitive traits and Oring (1977) in contrast, named and explained “too far,” with their “dishonest” character only being avian mating systems on the basis of the male’s be- selected against after some delay. havioral response to various kinds of environment. More recently, research on avian mating systems has THE LOGIC OF SCIENCE HISTORY focused on the role of sexual conflict and the variability Selected parts of the history of any field of science can ofmating systems (Davies 1985; Hatchwell and Davies be viewed as a progression in which events lead logi- 236 HISTORICAL PERSPECTIVES tally to new hypotheses,which lead to new predictions, of inclusive fitnesstheory was a mixture of disbelief of new tests, and new discoveriesin a cycle that tends to my observations, skepticism of any theoretical inter- continue indefinitely. I do not think that the origin of pretation, and hostility based mainly, in my opinion, a major theory can be explained easily in this way, but on misunderstandingand conservatism.I experienced after such a theory has appeared many of its testable them all personally. Misunderstandingof inclusive fit- predictions and tests do follow logically from the the- nesstheory in the 1960sand 1970swas rampant. May- ory. Several aspectsof avian social behavior could be nard Smith (1964) attempted to separatekin selection treated in this way, but space permits only one. The from the stigmaof “group selection” usingthe placenta history of study of helpers-at-the-nest well illustrates as an example of a kin-selected trait. Later Dawkins this approach. (1979) felt compelled to describe “Twelve misunderstandingsof inclusive fitness theory.” DISCOVERY AND RECOGNITION OF In my own earliest theorizing about helping I at- HELPERS-AT-THE-NEST tempted to combine ecologicalthinking from the study The term helpers-at-the-nest was coined by Skutch of avian social systemswith the new insightsof Ham- (1935) who first recognizedthe inherent interest of the ilton (Brown 1969a).The principal vehicle for this syn- subject and drew attention to it in a series of original thesis was my three-phasetheory, which viewed eco- observationsand reviews (1953, 1959, 1961, 1987). A logical factors, such as shortageof females or suitable few earlier workers had noticed the phenomenon(My- territories, as “setting the stage” for helping by causing ers 1915) without drawing much attention to it, and delayed breeding and delayed dispersal(Brown 1974). some of Skutch’s contemporariesmay well have made This theory was expanded and updated by Orians et their observationswhile unawareof Skutch’s work (e.g., al. (1977) Koenig and Pitelka (1981) and Emlen (1982). Ritter 1938, Yamashina 1938) but it was Skutch’s The common denominator was that ecologicalfactors early work that initiated historicalcontinuity of interest favoring delayed dispersalwere viewed as “setting the in the subject. Helping attracted little interest at first, stage” for helping. however, althougha few specieswith helperswere sub- Despite these efforts to integrate the ecologicaland sequently studied without awarenessof inclusive ftt- genetic aspects of the subject, polarization occurred. nesstheory (e.g., Davis 1942, Rowley 1965, Yamashi- Ecology and genetics came to be viewed by some as na 1938, Brown 1963). Thus, although helping was alternative explanationsof helping. This interpretation known, it was essentiallyignored in spite of Skutchand arose from the argument that nonbreeding Scrub Jays a few others for over 35 years. stay home not for the indirect fitness advantages of helping but becausethey could not find suitable ter- THE ROLE OF THEORY ritories (Woolfenden and Fitznatrick 1978). The DOS- A nice demonstration of the importance of theory in sibility that both factors operated sequentiallyand to- stimulating research is the effect of inclusive-fitness gether, which I had proposedearlier, was forgotten for theory on the study of helping. My own work on help- awhile. The resultingpolarization of the field based on ing dates from the day in 1963 when I first read Ham- such misunderstandingswas extensive; and its effects ilton’s (1963) original paper on the evolution of altru- are still felt. I have analyzed the detailed origins of this ism in the biologylibrary at the University of Rochester. situation elsewhere (1978b, 1987a, 1987b). Polariza- Surely,I thoughtthen, many ornithologistswill rush to tions are common in other fields too and often arise test some of the predictions about helping that could from oversimplified dichotomies, for example, over be generated using Hamilton’s rule. I was wrong. Al- innate vs. learned behavior, sympatric vs. allopatric though a non-ornithologist saw the connection (Wil- speciation, punctuated equilibria vs. gradual change. hams 1966) it was not until I presented quantitative The literature of this period cannot be fully understood observations on helping in the Mexican Jay (Aphdo- without appreciating the influence of polarization. c~nla ultramarina) in the context of altruism seven years later (Brown 1970) that widespread interest in EVALUATION OF COSTS AND BENEFITS: RULES AND CURRENCIES the relevanceof Hamilton’s rule to helpingwas aroused. Theories not requiring kinship were also developed It was not at first realized that Hamilton’s rule was as alternatives (Trivers 197 1, Axelrod and Hamilton incorrect for plugging in data from field studies on 1981); and these too were considered in relation to helping. Hamilton’s rule was unsuitablebecause it was helping (Brown 1975, 1978a; Brown and Brown 1980; phrased in terms of fitness of members of the same Caraco and Brown 1986: Liaon and Ligon 1978,1983; generation rather than reproductive successof mem- Ligon 1983). It is difficult toprove, butthere is general bers of different generations.Although not widely ap- agreementthat the exponential increasein the number preciated,a turning point in studiesofdelayed breeding of publications on helping that is illustrated in Figure and helpingcame when a simple model was formulated 1.1 of Brown (198713)was due to interest generatedby that made possible a reasonably correct accounting inclusive fitnesstheory and its alternatives. Regardless procedure for ecological costs and benefits estimated of which theory one favors, if any, it seemsfair to say acrossgenerations from natural populations.This pro- that the study of avian helping was theory-driven in cedure was based on what I have called the offspring the period 1970-l 990. In this sensethe studyofhelping rule (Brown 1987b) to distinguish it from Hamilton’s mav be said to have advanced logically.- _ rule. Before I publishedthe offspringrule I wrote to Ham- THE DANGER OF DICHOTOMIES: POLARIZATION ilton suggestingthat his rule was wrong for field studies. The first reaction by many in the early 1970s to my He verykindlywrote back and mathematically derived position that it was usefulto view helpingin the context my version for field studies, the offspring rule, from HISTORICAL PERSPECTIVES 231

his version. Both rules were correct; Hamilton’s rule parental age. This reservation stimulated tests by means was designed for fitness effects in the same generation of helper-removal experiments. These showed that the principally among siblings, which facilitated theoreti- helper-associated increment to breeding success ex- cal models; the offspring rule was designed for effects pected under inclusive-fitness theory persisted even on lifetime reproductive success across two genera- under controlled conditions (Brown et al. 1982) at least tions, especially from offspring-helpers to parent-re- in the more typical cases. Similar and more robust cipients. He urged me to publish the offspring rule results with a passerine bird were obtained much later separately (pers. comm.), but 1 opted to publish it as with the Scrub Jay (Aphelocoma coerulescens)in Flor- an appendix to my forthcoming book (Brown 1975) ida (Mumme 1992). Recent field experiments on other where it was soon noticed by one perceptive reader, S. questions have also provided decisive tests of theory Emlen. (Pruett-Jones and Lewis 1990. Komdeur 1992). Mathematical theorists were rightly unimpressed with the simple algebra that led to the offspring rule (they FUTURE EFFECTS never mention it); nevertheless, the offspring rule was As more data became available it was possible to ex- not published until I2 years after Hamilton’s rule so amine future effects on fitness. The logic of habitat- it was not trivial. With Hamilton’s idea now expressed saturation arguments (Selander 1964; Brown 1969d, in a form suitable for field studies, empiricists began 1974; Woolfenden and Fitzpatrick 1978; Emlen, S. T., to use it. It was soon applied to answer the question 1978) predicted that a future direct benefit of staying of whether or not delays in breeding were justified by home as a nonbreeder could be the acquisition of a increments to indirect fitness through helping (Brown breeding position at home or nearby. Thus, I cannot 1978a; Emlen, S. T., 1978). Both authors agreed that agree with authors who claim that this possibility was delays in order to help were not justified by the gain overlooked. In their advocacy of future direct effects, in indirect fitness in several species, a conclusion that however, these authors overlooked future indirect ef- was consistent with the ecological theories of delayed fects. The possibility of future indirect effects was first breeding held by Selander (1964). Brown (1969d). raised by Brown (I 980) and the first estimates of them Woolfenden and- Fitzpatrick ‘( 1978) Koenig ‘and Pi: from field data were made by Reyer (1984) and Ra- telka (I 98 I), Emlen (1982)and most later authors. This benold (1985). Later papers further formalized the cal- result suggested that helping was not altruistic but was culation of future indirect effects and raised some con- a strategy of “last resort” (Brown 1969d) or one of troversy about accounting methods (Mumme et al. “making the best of a bad situation” (Marzluff and 1989; Creel 1990a, 1990b). Thus, a more complete and Balda 1990). At least two later studies, however, made accurate account of costs and benefits was stimulated a good case for helping being altruistic in some cases by the theoretical concept of future indirect fitness. (Reyer 1984; Emlen, S T., and Wrege 1989). DIRECT BENEFITS OF HELPING THE ROLE OF LONG-TERM STUDIES By causing so much attention to be given to indirect It has been implied and even stated that Brown and kin selection Hamilton also caused reactions that em- Emlen’s result is a consequence of long-term studies phasized its alternative, namely direct selection. This (Stacey and Koenig 1990, Koenig et al. 1992). This first appeared in the form of the concept of reciprocal view is incorrect. None of the data selected by Brown altruism, which was based on a simple application of and Emlen came from studies that at the time had been game theory embodied in the Prisoner’s Dilemma game carried out more than five years (e.g., Rowley 1965, (Trivers 1971). Cooperation or mutualism based on Parry 1973, Woolfenden 1975). The required data can direct benefit can take two forms, one that is dependent be gathered in only one reasonably typical year (e.g., on the behavior of the potential cooperator, as in the Brown and Brown I98 I) or in five years (Emlen, S. T., Prisoner’s Dilemma (in general called score-keeping and Wrege 1989). mutualism) and one that is not (byproduct mutualism; Long-term studies are valuable for many reasons, Brown 1983a). The first case to be analyzed in this way but the major conceptual and empirical advances in was the sharing of feedings of mixed broods by two or this field came well before the first long-term study was more sets of parents (Brown and Brown 1980, Caraco published (Woolfenden and Fitzpatrick 1984) or the and Brown 1986). A clear rejection ofeither hypothesis studies were based on experiments. Long-term studies was difficult, but the authors favored byproduct mu- provided a firmer base for the early findings and ulti- tualism. mately led to other discoveries (Stacey and Koenig Another attempt to explain helping without indirect 1990). selection by invoking Trivers was described by Ligon and Ligon (1978, 1983) to explain helpina bv non- FIELD EXPERIMENTS breeding Green Woodhoopoes (Phoe&&s ~urpu- More important than long-term studies for results that rem) and helping generally (Ligon 1983). The wood- are relatively free of alternative explanations are field hoopoe case has been interpreted differently (Brown experiments. The data utilized by Brown and Emlen 1987b), and Ligon’s general argument has found little were subject to a criticism that had been raised by Lack support (Emlen, S. T. 199 1). (I 968) on theoretical grounds and by Brown and Balda (1977) and Gaston (1978) on the basis of their field FLEXIBLE STRATEGIES AND HELPING data. These authors pointed out that when groups with Behavioral ecologists often use the adaptationist pro- more helpers outproduced groups with fewer helpers gram (Gould and Lewontin 1979) as it should be used this result could have been caused by correlated dif- (Brown 1982a), to generate hypotheses for testing. Their ferences in territory quality or other factors, such as predictions, however, are typically limited to consid- 238 HISTORICAL PERSPECTIVES erations ofcost and benefit. These alone are insufficient textbooks of Krebs and Davies. Quite unnecessarily to establish that the behavior of interest has resulted they have simply assumed the operation of natural from natural selection (Endler 1986). Thus, although selection in a vague way and neglected nearly all genetic cost/benefit studies may test predictions of selectionist “details.” I believe that it is better to emphasize the models, even if the results agree with the predictions null hypothesis that selection is not directly responsi- they can often be caused by factors other than natural ble. This position stimulates us to try to reject the null selection acting in the manner of the model. Textbooks hypothesis by studying whether or not selection is im- of behavioral ecology (Krebs and Davies 1978, 1984, portantly involved, using the formal methods reviewed 1987, 199 1) typically assume natural selection rather by Endler (I 986). The failure of behavioral ecologists than trying to prove it. With this attitude by behavioral in the past to deal explicitly with the details ofselection ecologists it is not surprising that Endler (1986) lists has strengthened the hand of those who claim that the no examples of behavioral phenotypes that have been diversity in social behavior among species of insects established by reasonable criteria to be caused by nat- and birds is due entirely to phenotypic plasticity in a ural selection. One can think of a few candidate ex- range of environments (West-Eberhard 1987, 1989: amples, but in any case this approach has not been Jamieson and Craig 1987; Jamieson 199 1). When ge- fashionable and has been little used. netic markers are available in studies of mate choice It was only a matter of time, therefore, before be- they reveal inadequacies in the currently popular the- haviorists began to realize that many seemingly adap- ories of the evolution of mate choice (a theme that is tive behavioral differences between species or popu- too long to review here). lations could also be caused environmentally, i.e., If selection is involved, then our attention shifts to without genetic change. Helping behavior was a prime the question of what developmental and physiological subject for this “ne? approach. As usual, entomol- mechanisms mediate between the genes and the be- ogists led the wav. West-Eberhard (1987. 1989), de- havior. In the case of helping, when is kin recognition veloped the idea of flexible strategies to explain how learned and what cues are used? Is there evidence that simply changing the environment could be enough to selection has elaborated the kin-recognition cues in induce greater sociality; and Sakagami and Maeta’s colonial species more than non-colonial species, as im- (1987) elegant experiments on bees provided evidence plied by Brown (1987c)? Have endocrine mechanisms that greatly strengthened the argument. been specially selected in species with helpers? What Similar arguments were developed for helping be- are the hormonal correlates of feeding nestlings in havior in birds. The idea that helping might be a “neu- breeders and non-breeders of species with and without tral trait” was first raised by Woolfenden and Fitzpat- non-breeding helpers (Vleck et al. 199 I)? Are such cor- rick (Woolfenden and Fitzpatrick 1984). In response I relates heritable? argued that the ability of helpers in colonial species to Finally, if physical condition and health are impor- recognize kin with high consistency (while helpers in tant in competition for resources, perhaps the immune noncolonial species did not) argued in favor of adap- system is involved to a greater extent than is currently tation because it was not explained by the “unselected appreciated. Much evidence suggests that mammals hypothesis”(Brown 1987~). This exchange was ignored s&%etimes choose mates using &nes of the major his by most readers, but the same ideas were picked up tocompatibility complex (Brown and Eklund 1994). soon by other authors. The neutral-trait idea surfaced Perhaps birds do too by some unknown mechanism. as the unselected hypothesis (Jamieson and Craig 1987; Jamieson 1989, 199 1) and my counter-argument was DISCUSSION developed also by others (Emlen, S. T., et al. 1991). The first section of this review presents a broad view Although the flexible strategy approach has been used that identifies major trends and patterns and gives credit in the past to explain plural breeding in communally mainly to conventional and frequently cited sources. breeding birds (Brown and Brown 1980, 1990) it was In contrast, the second section adopts a different ap- not presented in a controversial way and so did not proach. namely one that emphasizes the logical pro- attract attention. Jamieson’s papers have now drawn gression of the subject. In the broad perspective, major much attention to the problem, and it should receive synthetic works are cited to document trends and pat- more attention. terns; but in the focused view of a participant, credit goes less to reviews and oft-cited papers, which char- FUTURE TRENDS acterize the later development of a field, and more to Looking beyond the current concentration of studies works that made a difference in the early history. Some- on mate choice and the use of phylogenetic trees based times the relevant passages were short. Often they were on molecular data to study the phylogeny of behavior, presented quietly, without fanfare. Not uncommonly, I can see at least one neglected area that deserves at- only one or a few people were initially influenced by tention. The study of social behavior needs to move these passages. Thus, the papers credited in the second from cost/benefit thinking to what can be called selec- section are often not the most frequently cited papers tionist thinking (Chamov 1982) and beyond that to or those written by authors with a high profile. They genetically based selectionist thinking. It is not enough are influential not in the sense that they were popular to use the adaptationist program and optimality meth- or widely read but in the sense of leading logically to ods; we must think in terms of how selection actually future steps. I hope I have shown that the two ap- moves a phenotype from one stage to another; and we proaches used in this paper, one based on widely cited must do this in a realistic genetic framework. Game “landmarks” and the other based on cause and effect, theory is not enough. can be quite different. The present gap in this area is reflected in the various After writing this article I came across Ernst Mayr’s HISTORICAL PERSPECTIVES 239

“How to Write History of Biology” (Mayr 1982). In BROWN, J. L. 1963. Social organization and behavior defense of my frankly subjective treatment above I can of the Mexican Jay. Condor 65: 126-l 53. do no better than to quote a passage from it: BRO\KN, J. L. 1964a. The evolution of diversity in Subjectivity enters at every stage of history writing, avian territorial systems. Wilson Bull. 76: 160-l 69. especially when one is seeking explanations and asks BROWN, J. L. 1964b. The integration of agonistic be- why, as is necessary in problematic history. One can- havior in the Steller’s Jay Cyunocittastelleri(Gme- not arrive at explanations without using one’s per- lin). Univ. Calif. Publ. Zool. 60:223-328. sonal judgment, and this is inevitably subjective. A BROWN, J. L. 1969a. The buffer effect and produc- subjective treatment is usually far more stimulating tivity in tit populations. Am. Nat. 103:347-354. than a coldly objective one because it has a greater BROWN, J. L. 1969b. The control of avian vocaliza- heuristic value. tion by the central nervous system, p. 79-96. In R. A. Hinde led.1. Bird vocalizations. Cambridge I am extremely grateful to the following for gener- Univ. Press, Cambridge, England. ously but critically commenting on this paper: G. W. Barlow, D. A. Dewsbury, and J. P. Hailman. I thank BROWN, J. L. 1969~. Neuro-ethological approaches the National Science Foundation and the National In- to the study of emotional behavior: stereotypy and stitute of Mental Health for a series of grants that sup- variability. Ann. N. Y. Acad. Sci. 159:1084-1095. ported my various research activities during the period BROWN, J. L. 1969d. Territorial behavior and pop- covered in this review. ulation regulation in birds. Wilson Bull. 8 1:293- 329. BROWN, J. L. 1970. Cooperative breeding and altru- LITERATURE CITED istic behavior in the Mexican Jay, Aphelocoma ALCOCK, J. 1975. Animal behavior. An evolutionary ultramarina. Anim. Behav. 18:366-378. approach, 1st ed. Sinauer Associates, Inc., Sun- BROWN, J. L. I97 I. An exploratory study of vocal- derland, MA. ization areas in the brain of the Redwinged Black- ANDERSSON,M. 1982a. Female choice selects for ex- bird (Agelaiusphoeniceus).Behaviour 39:93-127. treme tail length in a widowbird. Nature 299:8 l8- BROWN. J. L. 1973. Behavior elicited bv electrical 820. stimulation of the brain of the Steller’s Jay. Con- ANDERSSON, M. 1982b. Sexual selection, natural se- dor 75: 1-16. lection and quality advertisement. Biol. J. Linn. BROWN, J. L. 1974. Alternate routes to sociality in sot. 17:375-393. jays-with a theory for the evolution of altruism ANDERSSON, M. 1986. Evolution of condition-de- and communal breeding. Amer. Zool. 14:63-80. pendent sex ornaments and mating preferences: BROWN. J. L. 1975. The evolution of behavior. Nor- sexual selection based on viability differences. ton, New York. Evolution 40:804-8 16. BRO\KN, J. L. 1978a. Avian communal breeding sys- ANTONOVICS, J. 1987. The evolutionary dys-synthe- tems. Ann. Rev. Ecol. System. 9: 123-l 55. sis: which bottles for which wine? Am. Nat. 129: BROWN, J. L. 1978b. Avian heirs of territory. Bio- 321-331. science 281750-752. AXELROD, R., AND W. D. HAMILTON. I98 I. The evo- BROWN. J. L. 1980. Fitness in complex avian social lution of cooperation. Science 2 I I: 1390-l 396. systems, p. 115-128. In H. Mark1 led.], Evolution BAERENDS,G., C. BEER,AND A. MANNING [EDS.]. 1976. of social behavior: hypotheses and empirical tests. Function and evolution in behaviour. Essays in Verlaa Chemie GmbH. Weinheim. honor of Professor Niko Tinbergen, F.R.S. Oxford BROWN, J.-L. 1982a. The adaptationist program. Sci- Univ. Press, Oxford, England. ence 217:885-886. BARLOW, G. W. 1989. Has sociobiology killed ethol- BROWN, J. L. 1982b. Optimal group size in territorial ogy or revitalized it?, p. l-45. In P.P.G. Bateson animals. J. Theoret. Biol. 95:793-810. and P. H. Klopfer [eds.], Perspectives in ethology, BROWN, J. L. 1983a. Cooperation-a biologist’s di- Vol. 8. Plenum Press, New York. lemma, p. l-37. In J. S. Rosenblatt, R. A. Hinde, BARLOW, G. W. 1991. Nature-nurture and the de- C. Beer, and M. Busnel [eds.], Advances in the bates surrounding ethology and sociobiology. study of behavior. Academic Press, New York. Amer. Zool. 3 1:286-296. BROWN, J. L. 1983b. Intersexual selection. Nature BEER, C. G. 1963. Ethology-the zoologist’s ap- 3021472. proach. Tuatara I I: 170-I 77. (and 12: 16-39) BROWN, J. L. 1987a. Book review: the Florida Scrub BENSON, K. R. 1988. From museum research to lab- Jay: demography of a cooperative-breeding bird. oratory research: the transformation of natural Auk 104:350-352. history into academic biology, p. 49-83. In R. BROWN, J. L. 1987b. Helping and communal breed- Rainger, K. R. Benson, and J. Mainenschein [eds.], ing in birds: ecology and evolution. Princeton Univ. The American development of biology. Univ. of Press, Princeton, NJ. Pennsylvania Press, Philadelphia. BRO\KN, J. L. 1987~. Testing inclusive fitness theory BROOKS, D. R., AND D. A. MCLENNAN. 199 I. Phy- with social birds, p. 103-I 14. In Y. Ito, J. L. Brown, logeny, ecology, and behavior. A research program and J. Kikkawa [eds.], Animal societies: theories in comparative biology. Univ. of Chicago Press, and facts. Japan Sci. Sot. Press, Tokyo. Chicago. BROWN, J. L., AND E. R. BROWN. I98 1. Kin selection BROWN, J. L. 1959. Method ofhead scratching in the and individual selection in babblers, p. 244-256. Wrentit and other species. Condor 61:53. In R. D. Alexander and D. W. Tinkle [eds.], Nat- 240 HISTORICAL PERSPECTIVES

ural selection and social behavior: recent results DARWIN,C. 1872. The expressionof the emotions in and new theory. Chiron Press,New York. man and animals. (1965 edition.) Univ. of Chi- BROWN,J. L., ANDE. B. BROWN. 1990. Mexican Jays: cago Press, Chicago. uncooperativebreeding, p. 268-288. In P. B. Sta- DAVIES,N. B. 1983. Time allocation between terri- cey and W. D. Koenig [eds.],Cooperative breeding tories and flocks and owner-satellite conflict in in birds: long-term studiesof ecology and behav- foragingPied Wagtails,Motacilla a&a. J. An. Ecol. ior. Cambridge Univ. Press, Cambridge. 521621-634. BROWN,J. L., ANDR. W. HUNSPERGER.1963. Neuro- DAVIES,N. B. 1985. Cooperation and conflict among ethology and the motivation of agonistic behav- Dunnocks, Prunella modularis, in a variable mat- iour. Anim. Behav. 11:439-448. ing system. Anim. Behav. 33:628-648. BROWN,J. L., ANDG. H. ORIANS. 1970. Spacingpat- DAVIES,N. B. 1992. Dunnock behaviour and social terns in mobile animals. Ann. Rev. Ecol. System. evolution. Oxford Univ. Press,Oxford, England. 11239-262. DAVIES,N. B., AND A. I. HOUSTON. 198I. Owners BROWN,J. L., ANDR. P. BALDA. 1977. The relation- and satellites:the economics of territory defence ship of habitat quality to group size in Hall’s bab- in the Pied Waatail. MotaciNa alba. J. Anim. Ecol. bler (Pomatostomus halli). Condor 791312-320. 50:157-180. - BROWN, J. L., AND E. R. BROWN. 1980. Reciprocal DAVIS, D. E. 1942. The phylogeny of social nesting aid-giving in a communal bird. Zeit. Tier. 53:3 13- in the Crotophaginae. Quart. Rev. Biol. 17:115- 324. 134. BROWN,J. L., E. R. BROWN,S. D. BROWN,AND D. D. DAWIUNS,M. S., T. R. HALLIDAY,AND R. DAWK~NS Dow. 1982. Helpers: effectsof experimental re- [EDS.]. 1990. The Tinbergen legacy. Chapman moval on reproductive success.Science 2 15:42l- and Hall, New York. 422. DAWKINS,R. 1979. Twelve misunderstandingsof kin BROWN,J. L.. ANDA. EKLUND. 1994. Kin recognition selection. Zeit. Tier. 5 I: 184-200. and the maior histocompatibihtv complex: an in- DEWSBURY,D. A. [ED.]. 1985. Leaders in the study tegrative review. Am. Nat. 143:i7@-196. of animal behavior. Bucknell Univ. Press, Lew- BROWN.J. L.. AND S. L. PIMM. 1985. The oriain of isburg, PA. helping: the role of variability in reproductive po- DEWSBURY,D. A. 1989. A brief history of the study tential. J. Theoret. Biol. I 12:465-477. ofanimal behavior in North America. Persp.Ethol. BURBRIDGE,D. 1992. Lackinga solution. Nature 355: 8:85-122. 118. DOBZHANSKY,T. 1937. Genetics and the origin of BURGHARDT,G. M. [ED.]. 1985. Foundationsofcom- species,1 st ed. Columbia Univ. Press,New York. parative ethology, 1st ed. Van Nostrand Reinhold, EDWARDS,S. V., AND S. NAEEM. 1993. The phylo- New York. genetic component of cooperative breeding in CARACO,T. 1979a. Time budgetingand group size: perching birds. Am. Nat. 14:754-789. a test of theory. Ecology 60:618-627. EMLEN,J. M. 1966. The role of time and energy in CARACO,T. 1979b. Time budgetingand group size: food ureference. Am. Nat. 100:61 l-6 17. a theory. Ecology 60:6 11-617. EMLEN,J: M. 1973. Ecology: an evolutionary ap- CARACO,T., ANDJ. L. BROWN. 1986. A game between proach. Addison-Wesley, Reading, MA. communal breeders:when is food-sharing stable? EMLEN,S. T. 1978. Cooperative breeding, p. 245- J. Theoret. Biol. 118:379-393. 28 1. In J. R. Krebs and N. B. Davies [eds.], Be- CARACO,T., AND L. L. WOLF. 1975. Ecologicalde- havioural ecology: an evolutionary approach. terminants of group sizes of foraging lions. Am. Blackwell Scientific, Oxford, England. Nat. 109:343-352. EMLEN,S. T. 1982. The evolution of helping. I. An CARPENTER.F. L.. AND R. E. MACMILLEN. 1976. ecological constraints model. Am. Nat. 119:29- Threshold model of feeding territoriality and test with a Hawaiian Honeycreeper.Science 194:639- 39. 642. EMLEN,S. T. 1991. Evolution ofcooperative breeding CHARNOV,E. 1982. The theory of sex allocation. in birds and mammals, p. 301-337. In J. R. Krebs Princeton Univ. Press, Princeton, NJ. and N. B. Davies [eds.], Behavioural ecology.An CREEL,S. 1990a. The future componentsof inclusive evolutionary approach. Blackwell Scientific, Ox- fitness:accounting for interactionsbetween mem- ford, England. bers of overlappinggenerations. Anim. Behav. 40: EMLEN,S. T., ANDL. W. ORING. 1977. Ecology,sex- 127-134. ual selection,and the evolution of mating systems. CREEL,S. 1990b. How to measure inclusive fitness. Science 1971215-223. Proc. Roval Sot. London B 241:229-231. EMLEN, S. T., AND P. H. WREGE. 1989. A test of CROOK,J. H. ’ 1965. The adaptive significanceof avi- alternatehypotheses for helpingbehavior in White- an social organizations. Symp. Zool. Sot. Lond. fronted Bee-eatersof Kenya. Behav. Ecol. Socio- 14:181-218. biol. 25:303-3 19. DARWTN,C. 1859. On the origin of species.Murray, EMLEN, S. T., H. K. REEVE,P. W. SHERMAN,F. L. London. WREGE, F.L.W. RATNIEKS, AND J. SHELL- DARWIN,C. 1871. The descentof man and selection MAN-REEVE.199 1. Adaptive versusnonadaptive in relationto sex.Modem Library, Random House, explanationsof behavior: the case of alloparental New York. helping. Am. Nat. 138:259-270. HISTORICAL PERSPECTIVES 241

ENDLER,J. A. 1986. Natural selection in the wild. ing behaviour in birds:a departurefrom functional Princeton Univ. Press, Princeton, NJ. explanations,D. 79-98. In P.P.G. Batesonand P. FISHER,R. A. 1915. The evolution of sexual prefer- H.. Klopfer [ehs.], Perspectivesin ethology. Ple- ence. EugenicsRev. 7: 184-I 92. num Press, New York. FISHER,R. A: 1930. The genetical theory of natural KALELA,0. 1954. Uber den revierbesitz bei vogeln selection. Clarendon Press, Oxford. und saugetierenals populationsokologischerfak- FRETWFLL.S. D. 1972. Pooulationsin a seasonalen- tor. Ann. Zool. Sot. ‘Vanamo’ 16:1-48. vironment. Princeton University Press,Princeton, KIRKPATRICK,M. 1982. Sexualselection and the evo- NJ. lution of female choice. Evolution 36:1-12. FRETWELL,S. D., AND H. L. LUCAS. 1970. On terri- KLOMP,H. 1972. Regulation of the size of bird pop- torial behaviour and other factorsinfluencing hab- ulations by means of territorial behaviour. Neth- itat distribution in birds. Acta Biotheor. 19:16- erlands J. Zool. 22:4561188. 36. KLOPFER, P. H. 1962. Behavioral aspectsof ecology. GASTON,A. J. 1978. Demography of the JungleBab- Prentice-Hall, Ennlewood Cliffs. NJ. bler Turdoides striatus. J. An. Ecol. 47:845-870. KLOPFER,P. H. [ED.:. 1970. Behavioral ecology. GILL, F. B., AND L. L. WOLF. 1975. Economics of Dickenson Publishing, Belmont, CA. feeding territoriality in the Golden-winged Sun- KLOPFER,P. H., AND J. P. HAILMAN. 1967. An in- bird. Ecology 56:333-345. troduction to animal behavior. Ethology’s first GOULD,S. J., AND R. C. LEWONTIN. 1979. The span- century, 1st ed. Prentice-Hall, Englewood Cliffs, drels of San Marco and the Panaglossianpara- NJ. digm: a critique of the adaptationis; programme. KOENIG, W. D., AND F. A. PITELKA. 1981. Ecological Proc. Roval Sot. London B 205:581-598. factors and kin selection in the evolution of co- HAMILTON,w. D. 1963. The evolution of altruistic operative breeding in birds, p. 26 l-280. In R. D. behaviour. Am. Nat. 97:354-356. Alexander and D. W. Tinkle teds.], Natural selec- HAMILTON,W. D. 1964. The genetical evolution of tion and social behavior: recent researchand new social behaviour. I. and II. J. Theoret. Biol. 7: l- theory. Chiron Press, New York. 52. KOENIG,W. D., F. A. PITELKA,W. J. CARMEN,R. L. HARPER,D.G.C. 1982. Competitive foragingin Mal- MUMME. AND M. T. STANBACK.1992. The evo- lards: ‘ideal free’ ducks. Anim. Behav. 30:575- lution ofdelayed dispersalin cooperativebreeders. 584. Quart. Rev. Biol. 67: 11l-l 50. HATCHWELL,B. J., AND N. B. DAVIES. 1992a. An KOMDEUR,J. 1992. Importance of habitat saturation experimental study of mating competition in mo- and territory quality for the evolution of cooper- nogamous and polyandrous Dunnocks, Prunella ative breeding in the SeychellesWarbler. Nature modularis: II. Influence of removal and replace- 358:493-495. ment experiments on mating systems.Anim. Be- hav. 43~61 l-622. KONISHI,M. 1985. Bird song:from behavior to neu- ron. Ann. Rev. Neurosci. 8: 125-l 70. HATCHWELL,B. J., AND N. B. DAVIES. 1992b. An experimental study of mating competition in mo- KREBS,J. R., ANDN. B. DAVIES[EDS.]. 1978. Behav- nogamous and polyandrous Dunnocks, Prune/la ioural ecology.An evolutionary approach, 1st ed. modularis: II. Mate guarding and copulations. Sinauer Associates,Sunderland, MA. Anim. Behav. 43:595-609. KREBS, J. R., AND N. B. DAVIES[EDS.]. 1984. Behav- HEINROTH,0. 1911. Beitrgge zur biologie, nament- ioural ecology.An evolutionary approach, 2nd ed. lich ethologieund psychologieder anatiden. Verh. Sinauer Associates,Sunderland, MA. Int. Omithol. Kongr. (Berlin) 5:589-702. (Trans- KREBS, J. R., ANDN. B. DAVIES. 1987. An introduc- lated in Burghardt 1985) tion to behavioural ecology, 2nd ed. Sinauer As- HINDE, R. A. 1966. Animal behaviour. A synthesis sociates,Sunderland, MA. of ethology and comparative psychology, 1st ed. KREBS,J. R., AND N. B. DAVIES[EDS.]. 1991. Behav- McGraw-Hill Book, New York. ioural ecology, 3rd ed. Blackwell Scientific Pub- HINDE, R. A. 1970. Animal behaviour. 2nd ed. Mc- lications, Oxford, England. Craw-Hill, New York. LACK,D. 1947. The significanceof clutch size, parts HUXLEY.J. S. 1942. Evolution. the modem svnthesis. I-II. Ibis 89:302-352. All& & Unwin, London. LACK,D. 1948a. Natural selectionand family size in INGLE, D., AND D. CREWS. 1985. Vertebrate neu- the starling. Evolution 2:95-l 10. roethology:definitions and paradigms.Ann. Rev. LACK,D. 1948b. The significanceof clutch size, part Neurosci. 8:457-494. III. Ibis 90:25-45. JAMIESON,I. 1989. Behavioral heterochronyand the LACK,D. 1954a. The evolution of reproductiverates, evolution of birds’ helping at the nest: an unse- p. 143-156. In J. Huxley, A. C. Hardy, and E. B. lected consequenceof communal breeding?Am. Ford [eds.], Evolution as a process.Allen and Un- Nat. 133:39&lO6. win, London. JAMIESON.I. 1991. The unselectedhvoothesis ,. for the LACK, D. 1954b. The natural regulation of animal evolution of helping behavior: too much or too numbers., 1st ed. Clarendon Press, Oxford, En- little emphasison naturalselection? Am. Nat. 1138: gland. 27 l-282. LACK, D. 1966. Population studies of birds. Clar- JAMIESON,I., ANDJ. L. CRAIG. 1987. Critique ofhelp- endon Press, Oxford, England. 242 HISTORICAL PERSPECTIVES

LACK, D. 1968. Ecological adaptations for breeding of biology. Harvard Univ. Press,Cambridge, MA. in birds. Methuen, London. MILINSKI, M. 1979. An evolutionarily stable feeding LANDE,R. 1980. Sexual dimorphism, sexual selec- strategyin sticklebacks.Z. f. Tierpsychol. 5 1:36- tion, and adaptation in polygeniccharacters. Evo- 40. lution 34:292-305. MILINSKI,M., ANDG. A. PARKER. 1991. Competition LANYON,S. M. 1992. Interspecific brood parasitism for resources,p. 137-168. In J. R. Krebs and N. in blackbirds (Icterinae): a phylogenetic perspec- B. Davies [eds.], Behavioural ecology. An evolu- tive. Science 255:77-79. tionary approach. Blackwell Scientific, Oxford, LECROY,M. K. 1981. The genus Paradisea: display England. and evolution. Am. Mus. Novitates 27 14:l-52. MOORE,B. 1920. The scopeof ecology.Ecology 1:3-5. LIGON, J. D. 1983. Cooperation and reciprocity in MORRIS,D. 1967. The naked ape: a zoologist’s study avian social systems.Am. Nat. 12 1:366-384. of the human animal. McGraw-Hill, New York. LIGON, J. D., AND S. H. LICON. 1978. Communal MORSE,D. H. 1980. Behavioral mechanismsin ecol- breedingin Green Woodhoopoesas a casefor rec- ogy. Harvard Univ. Press, Cambridge, MA. iprocity. Nature 276~496-498. MUMME, R. L. 1992. Do helpers increasereproduc- LIGON,J. D., AND S. H. LIGON. 1983. Reciprocity in tive success?An experimental analysisin the Flor- the Green Woodhoopoe (Phoeniculus purpureus). ida Scrub Jay. Behav. Ecol. Sociobiol. 31:3 19- Anim. Behav. 31:480-489. 328. LORENZ,K. 1937. Der Kumpan in der Umwelt des MUMME, R. L., W. D. KOENIG,AND F.L.W. RATNIEKS. Vogels. J. f. Ornithol. 83:137-213, 289-413. 1989. Helping behaviour,reproductive value, and (Translated in Lorenz 1970) the future component of inclusive fitness. Anim. LORENZ,K. 1941. Vergleichende Bewegungsstudien Behav. 38:331-343. bei Anatiden. J. f. Ornithol. 89: 194-294. fTrans- MYERS,H. W. 1915. A late nesting record for the lated in Avicultural Magazine volumes 57-59) California woodpecker. Condor 17:183-l 85. LORENZ,K. 1963. On aggression.Harcourt, Brace & NICE, M. M. 1943. Studies in the life history of the World, New York. Song Sparrow. Dover, New York. LORENZ,K. 1970. Studies in animal and human be- ORIANS,G. H. 1961. The ecology of blackbird (Age- haviour, Vol. 1. Harvard Univ. Press,Cambridge, luius) social systems.Ecol. Monogr. 3 1:285-312. MA. ORIANS,G. H. 1969. On the evolution of mating Lorr, D. F. 1991. Intraspecificvariation in the social systems in birds and mammals. Am. Nat. 103: systems of wild vertebrates. Cambridge Univ. 589-603. Press, Cambridge, England. ORIANS,G. H., C. E. ORIANS,AND K. J. ORIANS. 1977. MACARTHUR.R.. AND E. PIANKA. 1966. On the ov- Helpers at the nest in some Argentine blackbirds, timal use of a patchy environment. Am. Nat. ldo: p. 137-I 5 1. In B. Stonehouseand C. Perrins [eds.]. 603-609. Evolutionary ecology.University Park Press,Lon- MANGEL, M., AND C. W. CLARKE. 1988. Dynamic don and New York. modeling in behavioral ecology. Princeton Univ. OSTER,G. F., AND E. 0. WILSON. 1978. Caste and Press, Princeton, NJ. ecology in the social insects. Monogr. Pop. Biol. MARLER,P., AND D. R. GRIFFIN. 1973. The 1973 12:352. Nobel prize for physiologyand medicine. Science PARRY,V. 1973. The auxiliary social system and its 182:464-466. effect on territory and breeding in kookaburras. MARLER,P., AND W. J. HAMILTON,III. 1966. Mech- Emu 73:81-100. anisms of animal behavior. John Wiley & Sons, PRUETT-JONES,S. G., AND M. J. LEWIS. 1990. Sex New York. ratio and habitat limitation promote delayed dis- MARZLUFF,J. M., AND R. P. BALDA. 1990. Pinyon persalin superbfairy-wrens. Nature 348:541-542. Jays:making the best ofa bad situation by helping, PULLIAM,H. R., AND T. CARACO. 1984. Living in p. 199-237. In P. B. Stacey and W. D. Koenig groups: is there an optimal group size?, p. 122- [eds.], Cooperative breeding in birds: long-term 147. In J. R. Krebs and N. B. Davies [eds.], Be- studiesof ecologyand behavior. Cambridge Uni- havioural ecology:an evolutionary approach. Si- versity Press, Cambridge, England. nauer Assoc., Sunderland, MA. MAYNARDSMITH, J. 1964. Group selection and kin PYKE, G. H. 1979. The economics of territory size selection. Nature 20 1:1145-I 147. and time budget in the Golden-winged Sunbird. MAYR, E. 1942. Systematicsand the origin of spe- Am. Nat. 114:131-145. ties, 1st ed. (reprinted 1964). Dover, New York. PYKE, G. H., H. R. PULLIAM,AND E. L. CHARNOV. MAYR, E. 1958. Behavior and systematics,p. 341- 1977. Optimal foragingtheory. Quart. Rev. Biol. 362. In A. Roe and G. G. Simvson leds.1.Behavior 52:137-154. and evolution. Yale Univ. Press,New Haven, CT. RABENOLD,K. N. 1985. Cooperation in breeding by MAYR, E. 1963. Animal speciesand evolution, 1st nonreproductive wrens: kinship, reciprocity and ed. Harvard Univ. Press, Cambridge, MA. demography. Behav. Ecol. Sociobiol. 77: l-l 8. MAYR, E. 1982. The growth of biological thought, REYER,H.-U. 1984. Investment and relatedness:a 1st ed. Harvard Un&. Press, Cambridge, MAY cost/benefitanalysis of breedingand helping in the MAYR. E. 1993. What was the evolutionarv svnthe- Pied Kingfisher. Anim. Behav. 32: 1163-I 178. sis?Trends Ecol. Evol. 8:31-34. . . RICHMAN, A. D., AND T. PRICE. 1992. Evolution of MAYR, E., AND W. B. PROVINE. 1980. The evolu- ecologicaldifferences in the Old World Leaf War- tionary synthesis:perspectives on the unification blers. Nature 355:817-821. HISTORICAL PERSPECTIVES 243

RITTER, W. E. 1938. The California woodpeckerand TINBERGEN,N. 1959. Comparative studiesof the be- I. Univ. of California Press, Berkeley. havior of gulls (Laridae): a progressreport. Be- ROSENZWEIG,M. L. 1991. Habitat selectionand pop- haviour 15:l-70. ulation interactions: the search for mechanism. TRIERS, R. L. 1971. The evolution of reciprocal al- Am. Nat. 137, Supplement:%-S28. truism. Quart. Rev. Biol. 46:35-57. ROWLEY,I. 1965. The life history of the superb Blue VERNER,J., AND M. F. WILLSON. 1966. The influence Wren, Malurus cyaneus. Emu 641251-297. of habitats on mating systemsof North American SAKAGAMI, S. F.. AND Y. MAETA. 1987. Sociality, passerinebirds. Ecology 47:143-147. induced and/or natural, in the basically solitary VLECK,C. M., N. A. MAYS,AND J. W. DAWSON. 1991. small carpenterbees (Cerutinn), p. 1-16. In Y. Ito, Hormonal correlates of parental and helping be- J. L. Brown. and J. Kikkawa leds.1.Animal soci- havior in cooperatively breeding Harris’ Hawks eties: theories and facts. Japan. Sci. Sot. Press, (Parabuteo unicinctus). Auk 108:638-648. Tokyo. WATSON,A., AND R. Moss. 1970. Dominance, spac- SCHLEIDT,W. M. [ED.]. 1988. Der kreis urn Konrad ing behaviour and aggressionin relation to pop- Lorenz-ideen, hypothesen, ansichen [The circle ulation limitation in vertebrates, p. 167-218. In around Konrad Lorenz-ideas, hypotheses,opin- A. Watson [ed.], Animal populations in relation ions]. Parey, Berlin. to their food resources.Blackwell Scientific, Ox- SCHOENER,T. W. 1987. A brief history of optimal ford, England. foraging theory, p. S-67. In A. C. Kamili J. R. WEST-EBERHARD,M. J. 1983. Sexualselection, social Krebs. and H. R. Pulliam leds.1,Foraging behav- competition, and speciation.Quart. Rev. Biol. 58: ior. Plenum Press, New Y&k. _ - - 155-183. SELANDER,R. K. 1964. Speciation in wrens of the WEST-EBERHARD,M. J. 1987. Flexible strategy and genus Campy/orhynchus. Univ. Calif. Publ. 2001. socialevolution, p. 35-5 1. In Y. Ito, J. L. Brown, 74: l-224. and J. Kikkawa [eds.], Animal societies:theories SIBLY, R. M., AND R. H. SMITH[EDS.]. 1985. Behav- and facts. Japan Sci. Sot. Press, Tokyo. ioural ecology. Ecologicalconsequences of adap- WEST-EBERHARD,M. J. 1989. Phenotypic plasticity tive behaviour. Blackwell Scientific, Oxford, En- and the origins of diversity. Ann. Rev. Ecol. Sys- gland. tem. 20:249-278. SIMPSON,G. G. 1944. Tempo and mode in evolution, WHITMAN,C. 0. 1899. Animal behavior. Biol. Lect. 1st ed. Columbia Univ. Press, New York. Marine Biol. Lab., Woods Hole 7:285-338. SKUTCH,A. F. 1935. Helpers at the nest. Auk 52: WHITMAN,C. 0. 1919. The behavior of pigeons.Publ. 257-273. Carnegie Inst. 257: 1-161. SKUTCH,A. F. 1953. The White-throated Magpie- WIENS,J. 1966. On group selection and Wynne-Ed- Jay. Wilson Bull. 65:68-74. wards’ hypothesis.Amer. Sci. 541273-287. SKUTCH,A. F. 1959. Life history of the Groove-billed WILLIAMS,G. C. 1966. Adaptation and natural se- Ani. Auk 76:281-317. lection. Princeton Univ. Press, Princeton, NJ. SKUTCH,A. F. 196I. Helpers among birds. Condor WILSON,E. 0. 1975. Sociobiology. Harvard Univ. 63: 198-226. Press,Cambridge, MA. SKUTCH,A. F. 1987. Helpers at birds’ nests.A world- WOLF, L. L., AND F. R. HAINSWORTH.197 1. Time wide survey of cooperative breeding and related and energy budgets of territorial hummingbirds. behavior. Univ. of Iowa Press, Iowa City. Ecology 52:980-988. STACEY,P., AND W. D. KOENIG [EDS.]. 1990. Coop- WOOLFENDEN,G. E. 1975. Florida ScrubJay he!pers erative breeding in birds: long-term studies of at the nest. Auk 92: l-l 5. ecology and behavior. Cambridge Univ. Press, WOOLFENDEN,G. E., AND J. W. FITZPATRICK.1978. Cambridge, England. The inheritanceofterritory in group-breedingbirds. THORPE,W. H. 1979. The originsand rise ofethology: Bioscience28: 104-108. the science of the natural behaviour of animals. WOOLFENDEN,G. E., AND J. W. FITZPATRICK.1984. Heinemann Educational, London. The Florida Scrub Jay: demography of a cooper- TIGER,L. 1969. Men in groups.Random House, New ative-breeding bird. Princeton Univ. Press, York. Princeton, NJ. TIGER, L., AND R. Fox. 1971. The imperial animal. WYNNE-EDWARDS,V. C. 1962. Animal dispersion in Holt, Rinehart, Winston, New York. relation to social behavior. Hafner, New York. TINBERGEN,N. 1951. The study of instinct. Oxford YAMASHINA,M. 1938. A sociable breeding habit Univ. Press,Oxford, England. among timaliine birds. Proc. Intemat. Omithol. TINBERGEN,N. 1953a. The Herring Gull’s world. Ba- Congr. 9:453-456. sic Books, New York. (Also reprinted in 1989 by ZAHAVI, A. 1975. Mate selection-a selection for a Lyons and Burford) handicap. J. Theoret. Biol. 53:205-214. TINBERGEN,N. 1953b. Social behaviour in animals, 1st ed. Methuen, London.