Natural Selection and Social Behavior: Recent Research and New Theory

272 Pages·1981·19.094 MB·English

by  Richard D. Alexander, Donald W. Tinkle

SELECTION
~-ATURAL
1
AND
SOCIAL BEHAVIOR
,..,--- -
-4SEP1981
81/23862
CHIRO~
PRESS
INCORPORATED
New York & Concord
Distributed Outside North America
by
BLACKWELL SCIENTIFIC PUBLICATIONS
Oxford London Edinburgh Melbourne
l.
This volume is dedicated to George C. Williams, whose pioneering efforts to return our
Copyright© 1981 by Chiron Press, Incorporated. Printed in the United States of America. attention .to the c~cial que~tion of what are the units of selection and whose ideas regarding
This work, in whole or in part, may not be copied, reproduced, or transmitted in any form or that crucial quest10n, applied to such areas of research as senescence, sterile castes in the
by any means without permission in writing from Chiron Press, Incorporated, ~ith the sole social insects, reproductive effort, sexuality, sex ratios, among many others, in large part
exception that brief excerpts may be quoted in reviews for inclusion in periodicals, news fostered the revolution responsible for the research presented here.
papers, or broadcasts.
Sales: Chiron Press, Incorporated
Publishers Storage & Shipping Corp.
2352 Main Street
Concord, Massachusetts 01742
Editorial: Chiron Press, Incorporated
24 West 96th Street
New York, New York 10025
Distributed throughout the World excluding North America by:
Blackwell Scientific Publications
Osney Mead, Oxford OX2 OEL
8 John Street, London WClN 2ES
9 Forrest Road, Edinburgh EHI 2QH
214 Berkeley Street, Carlton, Victoria 3053, Australia
Library of Congress Catalogue Card Number 80-65758
ISBN 0-913462-08-X Blackwell ISBN 0-632-00624-2
First Printing
In Memorium Donald W Tinkle, 1930-1980
r
Donald W. Tinkle passed away on February 21, 1980. Without his efforts this volume
would not have appeared. He was largely responsible for the symposium from which it de
rived, and he read and criticized carefully all of the chapters. As the authors themselves
know, his unusual intelligence and perspicacity are reflected in the structure of many argu
ments throughout the book. His excellence in field research, teaching, and administration
also affected in more direct ways, and sometimes profoundly, the lives of most of the authors
represented here.
RDA
[
II
Our acquaintance with Donald Tinkle was unhappily brief and all too slight. We knew,
of course, that he was a distinguished biologist, with a broad knowledge of evolution and an
outstanding ability to devise experimental measurements of ecological theory. We also knew
his reputation for being a gifted and imaginative administrator and a superb teacher. Our
lasting impressions of him are his keen perception and intelligence and his open and gen
erous spirit.
Since his death, we have talked about him with several of his students and friends.
From knowledge of him far greater than ours, they bear multiple witness to these qualities.
As a scientist, Donald Tinkle was highly productive, publishing some eighty papers and
leaving more than a dozen manuscripts. He had a reverence for the classics and an under
standing of what was important historically. In his thirties and forties, he displayed a willing
ness to go on learning new techniques and the capacity to master them. He was a
well-rounded critic of both science and writing, and though unyielding, his criticism was
inevitably friendly and constructive.
The quality of friendliness particularly impressed his students, to whom his door was
always open. He seems to have given them the impression that he was learning from them,
that he was their guide but they were his equals. He was absolutely lacking in pettiness and
was absolutely fair. It all seems to have worked. One cannot fail to be struck by the excel
lence of his students. They may be his most remarkable legacy.
The Publisher
Contents
xi
Introduction Part Four COOPERATIVE BREEDING IN BIRDS
Part One EUSOCIALITY IN INSECTS 13 Altruism, Kinship, and Reciprocity in the White-Fronted 217
Intragroup Selection and the Evolution of Insect Societies 3 Bee-Eater
Mary Jane West-Eberhard Stephen T. Emlen
2 Individual Strategies of Inclusive-Fitness-Maximizing in Polistes 14 Demographic Patterns and Communal Breeding in the Green 231
fuscatus Foundresses 18 Woodhoopoe, Phoeniculus purpureus
Katharine M. Noonan J. David Ligon
3 Kin SJeolaecnt iEo.n S atnrads Ssmataelnlnit e Nests in Polistes exclamans 45 15 Kin SJeelrercatimo nL .a nBdr oInwdni vainddu aEl sStheelerc Rti.o Bn rionw Bna bblers 244
257
16 Selfish Behavior by Florida Scrub Jay Helpers
Part Two CRVOEEPRRRRTOEEBLDARUATCETTSEI VSO EF SMUACLCEE SASN IDN FVEAMRAIOLUES 17 EcoloGgliecna lE F. aWctooorslf eanndde Kn in Selection in the Evolution of 261
Cooperative Breeding in Birds
4 Male Age-Size Distribution and Male Mating Success in Bullfrogs 61 Walter D. Koenigp,nd Frank A. Pitelka
5 The TRiimchinagr do Df R. eHparwodaurcdt ion and Its Behavioral Consequences for 78 Part Five SMOACMIAMLA BLESH AVIOR AND LIFE HISIDRIES IN
Mottled Sculpins, Cottus bairdi
6 PhenJoetyrrpyic F C. Dororwelnatheosw oefr M analde L Ruetphreord Burcotiwvne Success in the Lizard, 96 18 NPreapiroiJteio sDhmno Lagn .( dSH cCoiouoorgipdlaeanread:t iCvyen Bormeyesd ilnugd oinv itchiea nBulsa)c k-tailed 283
Sceloporus jarrovi
Douglas E. Ruby 19 Reproductive Competition and Infanticide in Belding's 311
7 Population Structure and Social Behavior: Models for Testing the 108 Ground Squirrels and Other Animals
Ecological Significance of Song Dialects in Birds Paul W. Sherman
8 The ERvoobleurtti oBn. oPfa Syenxeu al Indistinguishability 121 20 CSpoerrceielast iEoxnhsi Abimtinogn gT wLoif eE Hxtirsetmorey FCohrmarsa cotef rMisotincos goafm Mya mmalian 332
Nancy Burley D. G. Kleiman
9 The EJvaoclku Wtio nB roafd Lbeukrsy 138 21 SPorimmeaG tAele pMnp Hlaitcaianutgsifo aanntsed ro ,S fCo Ccaoiramol lpS Duyt.se tSer amMusno ddeerlss ,t oa nthde M Sitcuhdaye ol fC hapman 345
Part Three· PARENTAL STRATEGIES IN VERTEBRATES Part Six THE PROBLEM OF SEXUALITY
1101 NFPiaastrhueernKsKat laee lnSn Bnetewleethohcato Rivdoi.Dno M ra. oncWKfd e Matlhlysaee l eE vanodlu Ftieomn aolfe I Fnrtoergssp ecific Brood Care in 117834 2223 FPGlouelncytomuWmaoitelrilp oiPahnami rsol mifDa Em .a nsHe vuFaitrasmo,c intalomtnordsen nS,i nteP axtehn twede irMCt hAoa .tei hnHvetoe eRlnnvaedended cr QAseo uonnetf,ae Sgnaeo nxnd i Nsta ncy A. Moran33 6832
12 Reproductive Cost and The Sex Ratio in Red-Winged Blackbirds 198 John Hartung '
Kent L. Fiala
[
r
Part Seven HUMANS
J. .
24 Abuse and Neglect of Children in Evolutionary Perspective 405 Introduction
Martin Daly & Margo/. Wilson
25 Paternal Confidence and Dowry Competition: A Biocultural
Analysis of Purdah 417
Mildred Dickemann
26 Uterine vs. Agnatic Kinship Variability and Associated Cousin
Marriage Preferences: An Evolutionary Biological Analysis 439
Mark Flinn
27 · Why Lineage Exogamy? 476 This volume results from a symposium held at The University of Michigan, Ann Arbor,
William Irons in October of 1978 and sponsored by the National Science Foundation. The papers included
were, for the most part, presented at the symposium, though a few additional ones were
28 Terminological Kinship, Genealogical Relatedness and Village requested fox the publication.
Fissioning among the Yanomamo Indians 490 The occasion for the symposium was the fiftieth anniversary of the University of Mich
Napoleon A. Chagnon igan Museum of Zoology in its present structure. It seemed to us that no better commemora
tion could be planned than a general discussion of the questions raised by the revolution in
29 Evolution, Culture, and Human Behavior: evolutionary biology that has occurred during the past two decades. The part of that revolu-
Some General Considerations 509 ~ ~ tion currently attracting the most attention concerns the evolutionary basis of social behavior
in all parts of the animal kingdom-hence the title of the symposium and of this volume.
Richard D. Alexander Museums have always played a special role in the nurture of evolutionary biology, and
so it is not surprising that most of the participants in this symposium are now, or have been in
the past, associated with such institutions. We are pleased that about half of the contributing
authors have direct connections to the University of Michigan Museum of Zoology.
In planning the symposium we were particularly interested in including field studies
designed to test significant predictions from recent new theory about the evolution of social
behavior. As the list of authors demonstrates, such studies-as well as important theoretical
contributions-are often the work of young investigators.
We believe that the papers included here represent most of the topics that have sparked
the recent interest in behavioral evolution. Read in sequence, the papers provide an excellent
overview of current research and theory. Differences of opinion and approach are obvious
and are often provocative and stimulating. We have not tried to eliminate such differences,
feeling instead that each paper should stand on its own merits. We also believe that this is the
first major volume of original papers devoted almost wholly to research stimulated princi
pally by George C. Williams and William D. Hamilton, who stressed two main ideas: first, it
is valuable to identify the level (gene, inoividual, population, species) at which natural se
lection acts most consistently and powerfully and, second, natural selection can favor contri
butions to genetic reproduction not only through descendant but also through nondescendant
relatives. The importance of these two ideas is apparent throughout the volume.
The organization of the volume is partly taxonomic and partly by subject. We thought it
appropriate to begin with the social insects, for their sterile castes have, since Darwin, been
a focal point in the understanding of natural selection. What, after all, could be more chal
lenging to a theory of evolution based on differential reproduction than explaining the exist
ence of individuals that normally produce no offspring of their own? The currently intensive
study of cooperative breeding in birds, represented here by several investigations, involves
obvious parallels, because helpers sometimes die without producing offspring; however, the
conclusions reached in studies of social insects and cooperatively breeding birds often di
verge intriguingly. Nevertheless, in both cases the emerging picture suggests that two cru
cial variables are genetic relatedqess and fluctuations in the availability of breeding habitat.
I
xi
L
Nearly all of the investigators in this symposium, including those interested in caste
systems and cooperative breeding, have sought to measure the reproductive success
of individuals in systems of sexual competition and parental care. Data on this long
neglected problem are presented for insects, fish, frogs, lizards, birds, and mammals,
including humans.
Sexuality can be viewed as involving a kind of proto-social cooperative behavior.
Among prominent questions in evolutionary biology at present, the evolutionary raison
d'etre of sexuality is fairly described as the most difficult. It is fitting, therefore, that this
volume should include two papers with promising new ideas on this question.
Finally, we are particularly pleased with the section on human sociality, for it shows
clearly that the theory of natural selection, which has for so long guided research at all levels
of inquiry in biology, has significant implications for the study of human behavior and social
systems as well.
Eusocialty in Insects
xii
L
1 .
lntragroup Selection and the Evolution of
Insect Societies
Mary Jane West-Eberhard
Theories of insect sociality have recently become very complicated and rather nar
rowly focused on the problem of worker altruism or helping behavior. Current discussions
often seem overly preoccupied with theoretical fine points and with the debate over whether
insect societies are despotic matriarchies, families united by kin selection and (when hyme
nopteran) the genetic strictures of haplodiploidy, or tense communes of selfish mutualists.
As a result, students of social insects are sometimes distracted from the broader picture. So I
shall begin by reconsidering the idea that led to the present burst of interest in social
phenomena-lest it be lost in specialized controversies that, like the bickering in a love
affair gone sour, tend to make us forget what the excitement was about in the first place.
The insight that has transformed present-day evolutionary analyses of sociality is a
simple extension or rethinking of how natural selection applies to interacting individuals. It
states that an important influence on social behavior, whether antagonistic or collaborative,
is the degree to which interacting individuals are likely to carry copies of the same genetic
alleles (Hamilton, 1964). The idea can be summarized by two related principles.
The first concerns genetic similarity and social collaboration. It says that gene fre
quencies can be affected not only by differential direct reproduction (production of off
spring) but also by indirect effects (Fisher, 1930) such as helping offspring or other
relatives. For traits affecting indirect reproduction to evolve, a sufficient proportion of the
affected (helped) individuals must be co-carriers of the genetic allele(s) for helping borne
by the helping individual (Hamilton, 1964). Such selection could well be called "co
carrier'' selection. Because relatives are the genetic co-carriers most obviously associated
in nature, it is usually called' 'kin'' selection (Maynard Smith, 1964), even though alleles
may be co-carried for reasons other than dose genealogical connections (see Hamilton,
1975, p. 141).
The second principle involves genetic dissimilarity and social competition. It states
that as long as individuals (i.e., members of a group) are genetically different they are ex
pected to compete and express conflicts of interest under natural selection. Therefore group
selection arguments that regard the social group as a single unit of selection will sometimes
be misleading because of heterogeneities, even within highly integrated colonies, or fami
lies (see Williams, 1966). Individuals of such groups may show a high degree of ''commu
nity of 0interests" (sensu Alexander and Borgia, 1978) because of genetic kinship,
manipulation, and mutual benefits, but individualistic behavior and the means of suppress
ing it should continue to evolve as long as intragroup genetic heterogeneity persists. Illus
trative examples in highly social insects are given below.
In keeping. with these two principles I shall examine some special consequences of
intragroup competition for the orpnization and evolution of insect societies. I will use par-
3
L
West-Eberhard 5
4 Eusociality in Insects
tic~lar examples from field studies of tropical social wasps to make the following general tives available to worker females losing out in social competition for available cells.
pomts. Helping relatives is a way of salvaging some reproductive success in a situation where alter
natives involving direct reproduction are closed or relatively unprofitable.
l. As orga?isi:n~ come to depen? on lif~ (or competition) in groups, the reproductive suc The queen-worker dimorphism in social insects is therefore fundamentally similar to
cess of md1v1dua~s dep~nds mcreasmgly on their ability to win in social competition otherfacultative polymorphisms. Such polymorphisms include the ''high-low'' (or major
r~ther than on th~lf ab1hty to confront other aspects of the environment. Social interac minor) dimorphisms of horned beetles (Eberhard, ms., demonstrates their facultative and
tions become~ kind of screening process determining reproductive success. competitive nature); the divergent behavioral strategies common in male-male competition
2. When group hfe or competition (e.g., in mating aggregations) is obligatory or highly formates [e.g., in beetles (Eberhard, ms.,); bullfrogs (Emlen, 1976; Howard, this volume);
advantageous-that is.' ~ndiv~d~als cannot go off and become ''winners'' independent solitary bees (Alcock et al., 1977); fish (Robertson and Warner, 1978); elephant seals (Le
o~ the group-co.mpetit10n w1thm the group may severely limit the access of some indi Beouf, 1974); and lekking birds Hogan-Warburg, 1966]; solitary vs. gregarious (migra
viduals t? essential.resources, producing "winners" and "losers" within the group. tory) phases in locusts (see Wynne-Edwards, 1962); and asexual vs. sexual reproduction in
3. Alternati~e strategies may then evolve that enable "losers" to salvage at least some certain plants and animals, e.g., strawberries and corals (Williams, 197 5). All of these phe
reproductive success.
nomena show the following characteristics:
4. Ev~n after a facultativ~ alternative has developed, the original pattern is likely to re
ma~n ~e more prod~ctive one, and how readily a costly or irreversible secondary alter 1. Strong local intraspecific competition for some resource (e.g., brood cells, mates,
nativ~ is adopt~d will depend on the extent to which an individual can expect to win by food, or growing space) essential to reproduction.
adoptmg the pnmary pattern in the future. 2. Greater success of some individuals (e.g. , larger, more established, or more aggressive
ones) at securing the resource via some pre-empting behavior pattern (e.g., fighting).
. In particul~, I will argue that worker behavior-in social insects has evolyed as an alter 3. Adoption of an alternative behavior pattern by individuals relatively unsuccessful at
native :e~roduct1ve stra~egy, and that the worker-queen functional dimorphism is thus basi the original pattern, enabling them to reproduce by another means (e.g., helping,
~ally sun~l~ to othe: ~nds of facultative alternative reproductive strategies under strong sneaking copulations, or invading a new region).
mtraspec1f1c competition. 4. Facultative (conditional) rather than genetic differentiation of morphs or strategies.
WORKER BEHAVIOR AS A SECONDARY REPRODUCTIVE [ THE EVOLUTION OF A REPRODUCTIVE "DIVISION OF LABOR"
ALTERNATIVE AS A PRODUCT OF SOCIAL COMPETITION
.social c~mpeti~io~ be~in~ ~n the so~ial insects with the advent of group living, what The winner and loser classes emerging under strong intra-group competition could
e~er its selective basis (m pnm1tively social wasps group living seems often to be associated originally be based on the variation (e.g., in size or aggressiveness) expected in any popula
with the advantage of re-using nests; other possible bases are discussed by Lin and Mi tion of solitary organisms. However once group life becomes highly advantageous or obli
chene~, 1972, and Alexander, 1974). In some nest sharing, casteless species, such as the gate, with differential reproductive success dependent on the outcome of social ·
sphec1d wasp, Trig.onop~is .c~meroni (Eberhard, 1974), each female provisions her own interactions, the characters determining social success become a special focus of natural
~ells as would a s~htary mdlVldu~l while tolerating the activities of others in close proxim selection and a specialization of the winner ''morph.' ' In the case of the social insects, these
ity and only occasionally competmg for an empty cell. In other species (reviewed in West
winner morphs are the queens.
Eber.hard, 19?8b) competition among nestmates results in at least temporary reproductive One favorable characteristic of a good alternative behavior pattern is that it be some-
dommance .with one or a few females gaining exclusive control of the nest. thing that is not done well by winners or is precluded by their specialization in the primary
In soc1all~ unspecial~zed group-living forms like Zethus miniatus, a eumenid wasp I pattern. That is, selection might be expected to create an ''opposite'' (mutually exclusive,
have observed m Colombia, a female who loses out in competition for a cell either builds or ''interfering'' -see Ghiselin, 197 4) alternative specialization allowing the original
anoth~r on the same n~st. or leaves the group to found a nest of her own. However, most losers to compete in a way in which they are better equipped to win (see Hamilton, 1978;
euso.cial (~orker-contamm~) f?r~s have two characteristics that severely limit the repro West-Eberhard, 1979). If increasing specialization in one alternative leads to decreasing
ductive o~tio~s of defeated mdlVlduals: group life is obligatory (solitary production of sex ability to perform the other, the two morphs are complementary. If they are performing
ual offspnn~ 1s ~~n?wn or rare), and reproduction within groups is dominated by one or a tasks essential for reproduction they may become mutually dependent.
few aggressive md1v1duals who~e activities curtail or completely suppress reproduction by The queen-worker functional dichotomy in the social insects is opposite and comple
others. Thus, a fem~le defeated m the attempt to control a cell is not only unable to leave the mentary in this way, and the opposite nature of queen and worker specializations must have
nest and reproduc~ mdependently, but her reproduction on the nest is not tolerated. facilitated their evolution. Being an effective queen entails being present on the nest, which
~n a few species (e.g., in the genus Polistes) defeated or subordinate females lurk at the both prevents egg laying by competitors and conserves energy and fat reserves for egg pro
margm~ of the nest or move from nest to nest, apparently attempting to usurp the positions duction. Foraging, on the other hand, requires leaving the nest, diversion of energy from
of dommant females (West-Eberhard, 1969; Gamboa, 1978). However, for reasons dis:. egg-production to locomotion, and increased mortality. The hungry brood of an egg layer
c~ssed elsewhere (Wes~-Eb.erhard~ 1978b), most insects in which group life is obligatory or who is preoccupied with defending her position against usurpers (see Gamboa, 1978) cre
highly advantag.eous l~ve m falillly groups. Individuals living among relatives have an ates an opportunity for indirect reproduction (via helping) by her relatives. Thus, although
added reproductive option: If the benefit/cost ratio is sufficiently high (see Hamilton, 1964 · queen-worker specialization can originate as and may ultimately produce a ''cooperative''
Wes~-Eberhard, 1975), they may reprod~ce indire~tly by contributing to the direct repro~ division of labor (sensu Ghis~lin, 1974), it can be fundamentally "competitive" in
ductive success of nestmates. Thus, helpmg a relative can be viewed as one of the alterna-
J.
West-Eberhard 7
6 Eusociality in Insects
oucrti gdiivne-rjsuisfitc aast iios nt huen ddievri csoiomnp oeft iltaiboon.r arising in capitalistic human societies through prod cmoomrep eptriotolorsn gtoe dm (oseneo palosloiz Feo irts. yCtho,r r1e9sp78o,n odinn Mgl.y ,a zthteec eag).g -Wguhaernd tihnegr ev iigs ial blaercgoem eemse mrgoernec aen odf
The basic ''division of labor'' between the sexes-between large, nutritive female young, unsuppressed females on a nest with developing competing queens, competition is
galalmy deteepse anndde nsmt malolr, pmhos.t iAlen md,a lliek oen thees -dliivkiesiwonis oef i lnavboolrv iend s oopcpiaols iintes,e ccotsm, iptl ceamne bneta vriye,w meudt uas hmoousrts s. eOvenr eth, ea nodth tehre heagngd-g, uina rsdiinnggl ev-iqguile neena cro al osinnigelse, naegwgrleys lsaivide e cgogm cpaent iltaosrts a sw liothn gd aesv ethlorepe-
cphaoauvpsiunelsga taionordnig ,c ionlenaastdeediqn ugve inatoc ecgso amomfp deeittveit iisdoiionmn ao onrfpd lh adibsisomrru a(prsteeiv efeu P rstaherelkerec drt iiosectn u awsls.ie,t dh1 iin9n 7 Wa2 )ve. asTtr-iEhabeb leeer vhioasolrudgt,ai o1mn9e7atr9iyc. i(pnsregoe ao cWvIhnaee rwrisesta,-s sE bpabun setd pru hetnhcageireud esa,g rwgd1-ie9tgd6hu 9 nta)her. ewdD ielnygug-grl iv-angiiggdu i aeltr hgadigrsiesn vgabir ogvetii horg" ciatlchb,a esos evi"noegtnr.u at alflirygd heitanitngeg"n abqmyu eooetnhnges rc dofeomm ·pn aeolteti stao. trAtsa sci ske sraacprae
lated bending behavior and the queen dance indicate, queens are potentially
hyperaggressive females and are recognized as such by nestmates. Evidently the threat of
COMPETITION FOR REPRODUCTIVE PRIORITY IN their presence near a newly laid egg suffices to protect it.
METAPOLYBIA AND SYNOECA: THE SIGNIFICANCE OF These observations indicate that bending and egg-guarding in social wasps are func
SOCIAL RITUALS AND DISPLAYS tionally analogous to the displays and plumage of lekking male birds (see Selander, 1974;
Armstrong, 1965; and Wynne-Edwards, 1962, for many examples) that compete socially
vAdiosd uma aArale lrssee u apmllrtr ooesdaropudchcyiaot ilmlv obeege nsyhtu aiaocvncnioedesr dds ,m, i ostaphnylec asbeyee s ciuononctmdeiareelar ce instliaetoebxnruoasarc ablt tieseoce naolsenm cdbet eic oacon osm mp(peFelci esicxahrl ieo tfriro,c cea1uvl9 sed3 neo0 tf;'e' nerTmaxhttiuornramaavnlpa tsssgeo alonenf,c t ti,min'o'd snai.s. ; fTsr1oat9hrtroi7 emnn7 sgg)ap ,,tt e ehawcssei. ia imtIlnhnia z b omjeiondrarde td fsiorin saorgp nma lld ae f tykefhiswses ,ah mair nse(i sd nipenioaev tr rihH dt( eouoi nwarf dl tnasheo ser,pn e 1s- sc)c9po r6roleeo6nepn)snr,ii ioberndislgte uud cpfaeotrlsioircvz cremeeisdbso. es dsI dtion nhom sehfoir itnceghai,hea ns- lcss oiteucna cisakacenelec dscst sosct fhmo(urWmlpe arepeettes iptctt-riiotEooimndob uneeisrc ts htvoeiao eprrrndaey. ,
Wicrmnehgopea nrsreadt.B- snsEMe deanbh nyetea d rlcovha Slbibaooysrosnreder ro,oalv yetb1aic s9otraeie7no rls9 vnauo)tes.rfe di idnf n iiadgntmnie ctenawase ts(oer-La dtn .he()etRa (oetVitr ctrmehohseipapniriricidd nabsagl,ie o p,1il onoP9lgto7yeil8rbei)assi cti iinatnnire oae wne ws,sho a Pi sicsonphil sm ys,m boiMliucaiilnreat iitil)-an qiip lnutlohsueleeesy ctnbrrte iasgas.t pre oaTesu czththpteesess c defoaui irsdsnecpec ucestsioc osRimeneids ppbtcLioloiceaisruincaptleletaoy e rmdq lebiyunero eelcdteecohesntut sdarcni poomtieprmnsrer saobnm taatdaauctu,l hl ar ttpi mioan-e nqgorcdsu nop.e g naeoe vtahnbekoy ss ictpr.da)eoea,nplr notcarsconog.ed imgeA,ur spewcv.es toiitsItviiinhdnve age atan h gqcccegutoei rev moeoeivstpnfsyi espni,v t oesiwetsseontihemtreemisosna.s n t t Tseo·c us hroa cieuitvhsu nor aecetielxdhssrup a oelerodntanaf ecdbn tsiayteto n ruimsinorsetettelehise cr ecsaw rptac iobmetoiuycnooil n eodntssov gb i iea(pnpe vocop.osolgaiyirr.td ,
abh1nee9cdr7eo 3( mtuB,h seo1austt9ah 7tlt hlhM8yeeb)ye ) sst .ca oeaTplvenohe rlubeyagsebl g ciqia noul taneraysaenie-nddgrs e r So(oryaenunsd p ot m theoceacgo naemny tefhp osaeetusr t ni.i3 ntd5i o t nihnnee si Mstps oev itpenaur pysloaw tsliaytorbronminass g) o,c, b ooasnnneltdryav i toenhdnien e (g sso tenfaeu kw meWhsei ecrasohrteu -e Esvh bewiegnorhthru;k aaterlhdlreys, iawsnpta gpnl .oki gnianhengt da wd tS oahy etnsinomo ewwec ooiatrn hks eepamr pps.p e,a tcryque unl iereaeewrln as sctl ioovtywmeplnbyiecs s;ai nslal anytchd tas iiivtnt ce iPo.m noInmt Zr yaobbsboitaist lh sew c oariftoth tbt hhtaheelesi sem q (sauppriegcerkicsni .ea so nsob,d sf ai .scn) otqiemnur aebMcestne iastvan eopd vmo iwlpoyhovbeseiian t
winner (which in older colonies may be a small group of persistent queens) gains the entire ments of the workers.
nest and worker staff; defeated females are completely excluded from direct reproduction,
because in these species females are not known to reproduce outside of groups.
The reduction in queen number in a newly founded colony of M. aztecoides is de
scribed by West-Eberhard (1978a). Over a period of six months 30 queens were eliminated,
mainly during two episodes of strong social competition. These periods were characterized 15
by an increased frequency of certain displays such as "bending" by queens, and the
"queen dance" performed by workers (West-Eberhard, 1978a). Bending behavior seems
to function as a threat in both species. Queens direct it at approaching individuals and per
form it more vigorously near other queens, sometimes with aggressive jerking and shoving.
The workers' queen dance, performed when they encounter queens walking on the comb,
appears to represent arrested or inhibited aggressiveness. During episodes of queen elimi ....
nation this worker display intensifies until it becomes an attack and results in the elimina 5
tion of queens that fail to respond aggressively, apparently functioning as a test of >
0
dominance.
Egg-guarding behavior is another pattern associated with reproductive competition in IJ-:
a variety of social wasps, including Polistesfuscatus (West-Eberhard, 1969), Protopolybia ~ 5
pumila (Naumann, 1970), Polybia occidentalis, andMetapolybia azteca (Forsyth, 1978) as
well as M. aztecoides and S. surinama. Figure 1-1 shows the increasing length of time
newly laid eggs are guarded as competition gets stronger in a colony of M. aztecoides: As Figure 1-1. Duration of post-oviposition
the colony gets older, the number of competing queens goes down, but those present have vigil and age of queens in a colony of M etapoly-
increasingly developed ovaries and are increasingly aggressive toward each other. In effect, 5 bia aztecoides. Numeral above each point indi-
available egg-laying space is decreased for each female by the increased capacity of her QUEEN AGE (MONTHS) cates number of queens present.
8 Eusociality in Insects West-Eberhard 9
CASTE FLEXIBILITY IN SYNOECA AND METAPOLYBIA: though I have seen idle Metapolybia females perform nest duties briefly and then resume
A GENERAL MODEL OF ROLE DETERMINATION BASED queenlike behavior, I have never seen a forager later become a queen.
The changes in social role observed in M. aztecoides and S. surinama are summarized
ON INCLUSIVE REPRODUCTIVE VALUE in Figure 1-3. Figure 1-4 shows a hypothetical scheme to explain the movement toward
"indirect" reproduction and ultimately irreversible worker behavior (foraging) by a de
Social competition is particularly dramatic in the swarming polybiine wasps because feated swarm queen. According to this scheme a female begins to work only when the cost
these are specialized social insects with relatively large colonies (a high payoff for win of aid, in terms of her own future direct reproductive success (RS), drops below the benefit
ning), yet castes are flexible and determined by contests among adults rather than by ma (the augmented RS of aided queens devalued"By their relatedness to the helper). Among
nipulations of larvae. other factors, the cost of aid is directly proportional to the female's probability of reproduc
The ultimate selective basis for the caste flexibility of tropical polybiines is apparently tion. This probability rises as long as the individual is an egg-layer with an increasingly
predation. Predation on colonies is frequent (Forsyth, 1978) and devastating, causing developed ovary but drops off sharply when she is defeated in social competition (setting
wasps to abandon nest and brood and abscond to a new site, so that the long process of the point of downward inflection of the cost curve in Figure l '-4). The cost of helping a
rearing replacement workers and, subsequently, sexual offspring is sometimes reinitiated defeated queen does not drop immediately to zero because there must remain some proba
repeatedly before the cycle is completed (see West-Eberhard, 1978a). The caste flexibility bility ofreproduction should the dominant queen(s) die. Meanwhile, the benefit of working
of these wasps enables individuals to respond to the erratically fluctuating reproductive rises with the increasing shortage of workers in the newly founded colony (see Figure 1-5)
opportunities and colony maintenance crises associated with the repeatedly truncated re
productive cycle. Perhaps significantly, the most marked (least flexible) morphological
castes in the Polybiini are present in species such as Stelopolybia areata (Jeanne and Fagen, DIRECT REPRODUCTION
1974), that have very large colonies. Large colonies are presumably better defended and
better able to sustain a worker population even if repeatedly forced to abscond. \' QUEEN' A'"B-\) J SOLE EGG-LAYER
In Metapolybia and Synoeca all newly emerged females are potential queens (West ONE OF SEVERAL COMPETING EGG-LAYERS
Eberhard, 1978a). Whether a given female becomes a queen or a worker depends on her .l C\oJ
IDLE
situation relative to others. If there is a queen or group of queens present during her first
days of adult life she first enters an idle period, during which she is occasionally attacked \ NURSE, BUILDER (TASKS AT NEST)
and actively dominated by the queen(s). Then, at the age of 6-10 days, she begins to build 'WORKER' \.. FORAGER
and care for the brood, and at 10-15 days she begins to forage (West-Eberhard, 1978a; \ E
Forsyth, 1978). If the colony happens to become queenless during the initial idle period INDIRECT REPRODUCTION
young non-worker females become queens (see West.,.Eberhard, 1978a, and Figure 1-2).
Young queens present in a swarm or in a group of replacement queens become increas Figure 1-3. Behavioral changes observed in Metapolybia aztecoides and Synoeca surinama.
ingly aggressive and competitive as they grow older. Eventually all but one (in small colo
nies) or a few (in large colonies) leave in swarms or become workers (West-Eberhard,
1978a). As in the case of newly emerged females, defeated queens usually pass through a
period of idleness and take up nest building and brood attendance before foraging. Al-
Q = PRE -REPRODUCTIVE QUEEN
W = WORKER
I = IDLE CALLOW
x y
Q
<"' Q Q AGE
:>0i5: I QQ QQ w ww LEAGYEGR IDLE DNUETSIETS FORAGE
~ Q Q w w w w w
DAY: 4 5 6 7 8 9 10 11 (CAPTURED) Figure 1-4. Hypothesized cost/benefit explanation for changes in the behavior of a defeated
BI UILD FORAGE swarm queen. The line below graph represents the behavioral role changes typically undergone by a
M etapo{ybta or Synoeca queen defeated at age x (the sequence shown is invariable but the length of time
Figure 1-2. Emergence time and role when captured of females produced in a queenless colony of spent in each role is highly variable). The cost curve (C) reflects the changes in inclusive reproductive
Synoeca surinama. The queen that laid the eggs producing these females was removed four days before value that would presumably be experienced by such a female (see text). The indirect benefit (B) or
they began to emerge. Each letter represents one individual. Building by an offspring female was first payoff for performing aid increases gradually as the brood grows and swarm workers die off. The dot
observed on day 8 (day 1 being day of emergence of first offspring females). Foraging was first ob ted line (B ) represents the level to which the benefit for aid is raised as a result of the individual's own
served Oil\ day 11. ''Pre-reproductive queens'' performed the bending display and were danced to by probabilit~ of direct reproduction in the aided colony. This value (Bd), like the cost of aid, reflects
workers (see text). "Workers" foraged or added pulp to the nest. reproductive value. Low cost aid (nest duties) begin when B; + Bd >C.