Colony Dispersal and the Evolution of Queen Morphology in Social Hymen Opt Era

8/8/2019 Colony Dispersal and the Evolution of Queen Morphology in Social Hymen Opt Era

1/32

Annu. Rev. Entomol. 2001. 46:60130Copyright c 2001 by Annual Reviews. All rights reserved

COLONY DISPERSAL AND THE EVOLUTION OFQUEEN MORPHOLOGY IN SOCIAL HYMENOPTERA

Christian PeetersCentre National de la Recherche Scientique UMR 7625, Laboratoire dEcologie,Universit e Pierre-et-Marie Curie, 75005 Paris, France; e-mail: [email protected]

Fuminori Ito Laboratory of Insect Ecology, Faculty of Agriculture, Kagawa University,Takamatsu 760-8522, Japan; e-mail: [email protected]

Key Words gamergate, colony ssion, morphology, ight, Ponerinae

s Abstract Social Hymenoptera show two contrasting strategies of colony repro-duction. A reproductive female can raise the rst generation of brood alone (inde-pendent foundation), or a colony can divide into autonomous parts in which thereproductive female is helped by sterile relatives (ssion, budding, swarming). Inindependent-founding ants, queens can histolize their ight muscles after dispersal;in many species, large ight muscles and metabolic reserves reduce or eliminate theneed for risky foraging trips during the vulnerable solitary stage. Colony division isa derived strategy, and we review the selective pressures leading to its occurrence inthe different social taxa. In various ants, ssion coexists with independent founda-tion, and alate queens are retained. However, in ants exhibiting obligate ssion (e.g.all army ants and many Ponerinae), queens are permanently wingless (ergatoid), orthe queen caste is missing altogether. When reproductive females are ightless, dis-persal distances and colonization ability are reduced, and there are extensive modi-cations in mating behavior and resource allocation. We focus on the characteristicsof ssion in the phylogenetically primitive ants Ponerinae in which both ergatoidqueens and gamergates occur. The ground-living habits of ants have permitted ex-tensive changes in the phenotypes of their reproductive females, unlike in wasps andbees.

CONTENTS

OVERVIEW AND PERSPECTIVES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 602CASTE AND DISPERSAL OF INDIVIDUALS . . . . . . . . . . . . . . . . . . . . . . . . . . 603

Flight and Mating Behavior in Ants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 604How Do Flightless Reproductive Ants Find Mates? . . . . . . . . . . . . . . . . . . . . . . . 605

byCNRS-multi-siteon03/07/09.Forpersonaluseonly.


2/32

602 PEETERS ITO

COLONY DIVISION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 608How Do Insect Colonies Fission? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 609What is the Difference Between Fission and Budding? . . . . . . . . . . . . . . . . . . . . 611

Obligate Fission and the Evolution of Flightless Reproductive Females in Ants . . . 611Budding Coexists with Independent Foundation . . . . . . . . . . . . . . . . . . . . . . . . . 613INSIGHTS FROM PHYLOGENETICALLY PRIMITIVE ANTS . . . . . . . . . . . . 617

Frequent Occurrence of Wingless Reproductive Females . . . . . . . . . . . . . . . . . . . 617Dispersal Dimorphism and Evolution of Gamergate Reproduction . . . . . . . . . . . . 619Fission and the Pattern of Reproductive Investment . . . . . . . . . . . . . . . . . . . . . . . 619Fission and Nest Relocation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 621

SYNTHESIS AND FUTURE PERSPECTIVES . . . . . . . . . . . . . . . . . . . . . . . . . . . 622

OVERVIEW AND PERSPECTIVESThe ight ability of insects confers important advantages for dispersal and hasgreatly contributed to their evolutionary success (123). Yet in many species wingshavebeen lost in a proportionof individuals. Zera & Denno(134) recently reviewedthe occurrence of dispersal polymorphism in solitary insects. Ants and termitesalso provide an outstanding example of such polymorphism, because the majorityof individuals are completely wingless. Furthermore, in social wasps and bees,most of the ying adult females never disperse from their natal nest.

We review colony reproduction and dispersal in the social Hymenoptera, to-gether with their relationship with the patterns of morphological specializationamong the adult females. Larvae in many insects have the capacity to initiate di-vergent developmental pathways in response to various factors. Such sensitivityto environmental, nutritional, and social cues underlies the evolution of dissimilarfemale phenotypes found in a proportion of social wasps and bees and in all of theants and termites (127). Queen and worker castes are morphologically adapted forthe efcient performance of reproduction and helper activities, respectively. Castedivergence reaches an extreme in the ants and termites as a consequence of living

on the ground, because workers never have wings. To give proper emphasis to thesignicance of morphological specialization, we restrict the terms caste, queen,and worker to the outcomes of pre-adult differentiation (92).

New insect societies can begin in two distinct ways (41); one is indepen-dent colony foundation (ICF), in which a single reproductive female (or a fewin the case of pleometrosis) can raise the rst generation of brood alone. In theants, the wing muscles can be metabolized by a foundress after dispersal; thisinuences her success rate greatly because it reduces the need for risky for-aging. The other strategy is colony ssion or buddingone or several repro-

ductive females are helped by a group of sterile relatives, as part of a colonymoves away and becomes autonomous. In the ants and termites, this colony di-



3/32

ANT QUEEN MORPHOLOGY AND DISPERSAL 603

all adult females retain the ability to y, and thus ssion has little apparenteffect on dispersal distance. A further peculiarity of ants is that queens have be-come permanently wingless in a large number of species (all of the army ants for

example). Colony ssion is obligatory in these, unlike other ants, in which ssionoccurs as an alternative to ICF and alate queens (AQ) remain essential.

Colony ssion has replaced independent foundation in a large proportion of so-cial Hymenopterans (also in the termites; 109). The mode of colony reproductionis inuenced by ecological conditions, for example increased density of coloniesleads to stronger between-colony competition for resources, which selects againstICF(8,76).However, differentdegrees of queen specialization also affect theselec-tive pressures for colony ssion. Queens in higher ants (subfamilies Formicinae,Dolichoderinae, and Myrmicinae) have large wing muscles and body reserves that

increase their rate of success during ICF. This difference is striking when review-ing the available data on ICF and morphological specialization in social wasps andbees. Furthermore, because the characteristics of ssion are little understood inants, we make extensive comparisons with wasps in which the exposed nature of aerial nests makes it possible to study the behavior of individuals.

Understanding the evolution of ightless reproductive females in ants requiresthat we examine the tremendous variation in the extent of morphological diver-gence between AQ and workers. In some ants, either workers or intercastes(queen-worker intermediates) can reproduce sexually and supplement or replace

alate queens. This is unlike species in which queens are permanently wingless,and thus it is crucial to realize that ightless reproductives refers to females withdiverse developmental origins. Heinze & Tsuji (35) have drawn attention to thereplacement of AQ by wingless reproductives, and we want to explore this furtherby distinguishing between different selective contexts characteristic of the varioustaxonomic groups. We focus on ightless reproductive females in the phylogenet-ically primitive subfamily Ponerinae, in which closely related species often haveAQ. The limited caste dimorphism typical of this subfamily is of great relevanceto the evolution of novel reproductive strategies. In addition, the small size of

colonies (affecting the strategy of reproductive investment) and the simple natureof nests (affecting the likelihood of emigration and eventual fragmentation) mustbe considered.

Termites are excluded from our review, but they exhibit many parallels with theants.

CASTE AND DISPERSAL OF INDIVIDUALS

Ants, social wasps, and bees live in extended families in which most daughters donot attempt to disperse and breed. This is true even when all adult females remain



4/32

604 PEETERS ITO

wasp genus Ropalidia , all adult females can leave their natal nest and attempt toreproduce elsewhere (26).

Subsequent to the evolution of castes, only the queens disperse and mate (to-

gether with the males). In vespine wasps, some polistine wasps, honeybees, andstingless bees, workers are unable to store sperm (their spermatheca is no longerfunctional) and thus cannot produce diploid offspring. Similarly in the majority of ants, workers cannot mate, except in some Ponerinae (87). Social insects generallyexhibit strong seasonal periodicity in the production and release of sexuals. Thestudy of mating behavior can provide valuable information on dispersal charac-teristics. Unlike in social wasps and bees, the search for sexual mates is the onlyopportunity for ight in ants. In ants performing ICF, mating is intimately linkedwith both dispersal and the establishment of new colonies, and in many species

nuptial ights have evolved. This is seldom true in species performing ssion.

Flight and Mating Behavior in AntsMating behavior is very diverse, but H olldobler & Bartz (38) recognized two majorstrategies. In the female-calling syndrome, isolated queens are stationary on theground or low vegetation, often in the vicinity of their natal nests. Males y aroundsearching for individuals, and queens often release sex pheromones (12,40).Chances of successful mate location are enhanced because both sexes are active

above ground simultaneously. Female-calling occurs in taxonomically varied ants(42), including ponerine and leptothoracine ants with small colonies and limitedqueen-worker dimorphism (12, 22, 30, 31), as well as the myrmicine Carebaravidua with a highly dimorphic queen caste and colonies consisting of tens of thousands of workers (98).

On the other hand, the male-aggregation syndrome involves males and AQnding each other during nuptial ights that occur away from the nests. Withinminutes of one another, colonies of genera such as Lasius , Pogonomyrmex , orCamponotus release thousands of sexuals (37), and nuptial ights are of very li-

mited duration. Copulation takes place either in the air or on the ground underneaththe swarm (15, 98). Male-aggregation is widespread in higher subfamilies, but italso occurs in relatively few ponerines and Myrmecia (10, 30, 79, 130).

Nuptial ights do not exist in wasps. Individual patrolling ights are observed,or males defend territories (see 96 for a review in Polistes ). In some species of Ropalidia , males leave their nests for several hours every day, presumably attempt-ing to mate, and they return afterwards (26). This indicates an ability to forageindependently and live longer, compared with male ants. In several species of Vespula (all with dimorphic castes), males form loose, conspecic aggregations

around prominent vegetation and other landmarks while ying back and forththroughout the area (29).



5/32


6/32

606 PEETERS ITO

When Ant Queens Are on Their Own: With or WithoutMetabolic Reserves?

In a proportion of species, foundresses need to forage outside their shelter at regularintervals to feed the rst generation of larvae. This semiclaustral foundation isoften thebasal condition (Figure1) and is exhibited predominantly by ponerine andmyrmeciine ants having weak queen-worker divergence (30, 90). Flight musclesare small, and their histolysis is evidently not a sufcient source of amino acids forthe developing larvae. Furthermore, ponerine larvae cannot be fed by regurgitatedsecretions, and instead they eat some of the queens eggs or pieces of prey broughtinto the nest (90). Semiclaustral ICF is linked with poor success rates due to therisksof brood parasitism and predation on foraging queens, although few empirical

data are available.



7/32


In contrast, in species with more pronounced caste dimorphism, queens canraise their rst broodwithoutobtaining food outside thenest. Various body reservesmust be accumulated before the nuptial ight to sustain the claustral foundress and

her immature daughters for several months. Virgin queens carry large quantitiesof lipids and carbohydrates (56), which function as a ight fuel, while fat storesare depleted during the founding period. In addition, storage proteins, recentlydiscovered in the queens of higher ants, are synthesized while the new queens arein the parental nest (129). Stronger ight muscles are presumably needed to carrythis extra body weight, and these tissues can later be digested to provide additionalproteins for oogenesis. A hallmark of the specialization of ant queens for claustralICF is indeed their voluminous thoraces. A further adaptation common in higherants is that smaller than average (nanitic) workers are reared during the rst

generations after foundation (42,95). As a result, foraging outside the nest canbegin earlier; nanitics are not found in most Ponerinae.

Fully claustral foundation is thus associated with the production of more ex-pensive queens. In Lasius niger , the dry weight of queens increased from 4 mg ateclosion to 15.5 mg at the time of nuptial ight, whereas males remained approx-imately constant at 0.9 mg (7). In the re ant Solenopsis invicta , the dry weight of maturing queens increased by 484% in monogynous colonies, and their fat contentincreased by 776% (57). Adequate metabolic reserves mean that foundresses neverneed to leave their underground chamber, and thus the success rate of incipient

colonies is increased. Nevertheless, no matter how specialized the queens are forICF, they experience high mortality during dispersal, especially species exhibitingthe male-aggregation syndrome. The large majority of AQ die within hours of leaving the natal nest, destroyed by predators in ight (e.g. birds and robber ies)and on the ground (e.g. lizards, birds, and ants; 42, 98). Natural elements also taketheir toll (desiccation during ight or drowning for example). This tremendousmortality (99% of dispersing AQ in Pogonomyrmex ; 27a) partially explains whyso many female sexuals must be produced by the colonies of ants with matingaggregations.

Are Social Wasps and Bees Specialized forIndependent Foundation?

Wasp foundresses cannot isolate themselves from the outside world because of theexposed nature of their edgling nests (except Vespinae, which are cavity-nesters).They need to forage regularly and their brood is then vulnerable to predators andparasites. This is true for taxa both without and with dimorphic castes. In polis-tine wasps such as Polistes , Belonogaster , and most Ropalidia species, individual

females have the option of starting colonies alone, but usually they cooperate withrelated foundresses. They can also attempt to take over a nest initiated by a con-



8/32

608 PEETERS ITO

brood (29, 69); usurpation attempts of established colonies are also common. It ispossible that queens have metabolic reserves, because storage proteins have beendocumented in Polistes foundresses (J Hunt, N Buck, D Wheeler, manuscript in

preparation). Various lines of evidence indicate that the success rate during ICF islow. In a detailed observation of 59 newly founded nests of Vespa analis , 47.5%lost their queens on foraging trips before emergence of the rst workers (69). InVespa , a link has been documented between the more regular supply of food thatis possible once several workers begin foraging, and the increased rate of brooddevelopment compared with rst generations (69). Bees with dimorphic queensand workers (e.g. Apis species and Meliponinae) do not exhibit ICF (72).

Fully claustral foundation is thus an adaptation that is unique to the higher ants.It is possible partly owing to an exceptional modication of the thorax in order to

maximize the quantity of metabolic reserves that can be accumulated while in thenatal nest (J Liebig & C Peeters, manuscript in preparation). One can argue that thehuge ight muscles of some ant foundresses would be too expensive to maintainand operate on a daily basis. Opposed to this are the ight muscles of queens inwasps and bees that remain essential to escape predators and move colonies. Theirreproductive apparatus is highly specialized in some species, but the allometricrelationship between thorax volume and other body parts has not changed muchfrom that of solitary wasps. The production of numerous specialized queens seemspossible only in species with large colonies (see 7a for theoretical background).

Wasps and bees that perform ICF do not produce large numbers of sexuals, but thisdifference is partly a consequence of the much bigger size of ant colonies. Onlyin some vespine wasps do colonies exceed several thousands of adults (135), andthese can produce more than 1000 new queens annually (69).

COLONY DIVISION

Reproductive and infertile females cooperate to start new colonies in a large num-ber of taxonomically diverse social hymenopterans. This is accompanied by a greatdecrease in the mortality of reproductive females, who take fewer risks than theindependent foundresses of other species. Newly divided colonies are good com-petitors from the very beginning, because there is a sufcient number of adultsto build safe nests, rear brood, and forage efciently. The widespread replacementof ICF by colony ssion suggests the existence of great benets linked with thelatter strategy. Fission is obligate in 28 of 35 genera of social wasps (126) as wellas in all honeybees and stingless bees. Fission is also found in all the subfamiliesof ants. Fission makes it possible for all the obvious advantages of group living tobe retained throughout colony ontogeny, unlike ICF, which obligatorily involves asolitary stage. The social wasps and bees are ideal models to investigate the ecolog-ical pressures dictating the switch from ICF to colony ssion because this switch



9/32


consequences on both gene ow and colonization. In a heterogeneous environmentconsisting of a set of discrete patches that persist for some nite time period, move-ment between patches can only be accomplished by ight (99). Ecological con-

ditions selecting for either ICF or colony division in ants have been reviewed byseveral authors (4, 8, 9, 32, 35, 41, 76, 100). Once a favorable site is found by ants,it can be efciently exploited by daughter colonies remaining in the same area.

How Do Insect Colonies Fission?Colony ssion is obligate in thousands of ant species, yet it has seldom been docu-mented in the eld. Episodes of nest relocation are relatively common, but it is noteasy to establish when ssion is taking place. In contrast to wasps and honeybeesin which the exposed colonies can be monitored, in ants it is necessary to exca-vate one or both nests involved in an emigration, and compare their demography.Unlike ICF, ssion events are unpredictable in time and require long-term inves-tigations of groups of neighboring nests with marked nestmates. The few reportspublished ( Dolichoderus cuspidatus , Monomorium sp., Pachycondyla marginata ,some army ants; 11, 28, 58, 65) have all been chance observations of single ssionevents. Attempts to trigger ssion in the laboratory have seldom been successful(in contrast with investigations of nest emigration, which can be initiated simplyby manipulating the temperature and humidity of nest sites). Although geneticmarkers can help identify and compare patrilines and matrilines in neighboringcolonies suspected of originating from a ssion event (83), observational data re-main essential to understand how it came about. We are almost completely ignorantof the behavioral mechanisms leading to the segregation of workers and brood intotwo units that eventually become autonomous.

The best knowledge of ssion is in army ants. They have no xed nest, and it ispossible to monitor their above-ground bivouacs. However, these are atypical ants,due to their enormous colonies (several hundred thousands to millions of adults)and nomadic lifestyles. In Eciton hamatum , colonies consist of 50,000250,000workers, and early in the dry season they produce about 1500 males and < 10new queens. The latter become the center of attraction for a large fraction of theworkers, which congregate away from the old queen and her cluster of workerswithin the bivouac. Once the males emerge from their cocoons, an emigration istriggered, and two raid systems set off in opposite directions. This leads to theseparation of the old and the new queens, as well as their respective clusters of workers. All but one virgin queen are abandoned by the workers. The males yfrom the parental bivouac in search of foreign colonies. In some cases, the oldqueen is replaced during the ssion process (23, 42, 106).

In contrast, the behavioral mechanisms involved in colony ssion have beenstudied in Apis bees, several genera of stingless bees, and polistine wasps, and



10/32


11/32


is physogastric and can no longer y. A similar pattern prevails in Hypotrigonaspecies from Africa (19).

What is the Difference Between Fission and Budding?Colony division in ants is widely termed budding if associated with secondarypolygyny, whereas ssion is used for monogynous species. Franks & H olldobler(23) originally made this distinction and emphasized dispersal distance, becausethey thought that ssion was restricted to the huge colonies of army ants, whereasbudding was a mechanism for the formation of colonies with multiple nests that re-main interconnected (polydomy). Both terms have been successively redened(8, 42). According to Bourke & Franks (8), ssion occurs when a monogynouscolony produces a new generation of sexuals and then divides into two monogy-nous units. Budding occurs in species in which queens and workers leave poly-gynous colonies to form new societies (which can also be polygynous). Fission isassociated with the annual production of very few new queens, which is not thecase with budding (in which secondary reproductives are often involved). Thus twocontrasting and highly idealized processes of colony division are currently recog-nized, with distinct theoretical implications for sex allocation (17, 81). Pamilo (81)did not consider the number of reproductive females, but he re-emphasized thatbudding is linked with limited dispersal of daughter colonies. The latter charac-teristic leads to local resource competition, which restricts the success of daughtercolonies. However, precise data about the distance between newly divided coloniesare generally lacking in ants, and this distance is likely to be species-specic andopportunistic. We later discuss queenless species of ants in which the spatial char-acteristics of colonydivision areapparentlynot affectedby monogyny or polygyny;colony size seems to have a more important inuence. The semantic distinctionbetween ssion and budding is ingrained in the ant literature, but it may be best todescribe colony division on a case-by-case basis as involving one or more elementscharacteristic of what are currently termed ssion and budding. In the wasps andbees swarming is used with a more general denition (females from more thanone generation move to new sites in a coordinated group and initiate nests; 77).One difference that we emphasize is that colony division is obligate in a proportionof species, but it exists as an alternative to ICF in others. Thus in many Ponerinaeand all army ants, ssion is the exclusive mode of colony reproduction (and it in-volves mostly monogynous colonies, but also colonies with multiple gamergates).In contrast, the colonies of various higher ants begin by ICF, but once they reacha certain size and become polygynous, colonies can multiply by budding.

Obligate Fission and the Evolution of FlightlessReproductive Females in Ants



12/32

612 PEETERS ITO

trade-off should ensue (128). This argument is based on solitary insects, but it canbe applied to a social context. The resources of individual colonies are necessarilylimited and the nutrients not used for construction and maintenance of the ight

apparatus in female reproductives can be available to rear a greater number of workers. Indeed, the worker population of a colony determines whether it candivide successfully (63). For ant species in which semiclaustral ICF is selectedagainst, the production of AQ is too costly, provided that outbreeding is possiblethrough the males.

Ergatoid Queens In a large proportion of species in the subfamilies Ponerinaeand Myrmeciinae, the queens lack wings and ight muscles, and consequentlytheir thorax is simplied like theworkers (88, 114, 116).These apterousqueens aretermed ergatoid, and externally they are often not conspicuously different fromworkers (except for the enlarged abdomen in several species) (Figure 2). Impor-tantly, colonies produce only few ergatoid queens (EQ) annually. EQ occur alsoin all of the army ants (Table 1), which are characterized by a syndrome of convergently evolved traits (5, 23, 28). Queen-worker dimorphism in army ants isoften very pronounced, especially since queens can be physogastric (the interseg-mental membranes of the abdomen stretch to accommodate the enlarged ovaries,

Figure 2 Three types of reproductive females in the ponerine genus Gnamptogenys . Alate queen



13/32


TABLE 1 Distribution of wingless reproductive females in all of the ant subfamilies

Subfamily Genera (N) with Type of female b

(total genera [N]: ref. 6) wingless reproductives in [N] genera

Aenictinae [1] a 1 EQ [1]

Aneuretinae [1] 0 -

Apomyrminae [1] 0 -

Cerapachyinae [5] a 3 EQ or I [3]c

Dolichoderinae [22] 3 EQ or I [3] c

Dorylinae [1] a 1 EQ [1]

Ecitoninae [5] a 5 EQ [5]

Formicinae [49] 2 BQ [1]; EQ or I [1] cLeptanillinae [3] 2 EQ [2]

Myrmeciinae [2] 2 BQ [2]; EQ [1]

Myrmicinae [157] 20 BQ [5]; EQ or I [19] c

Ponerinae [42] 19 EQ [13]; G [10]; I [2]

Pseudomyrmecinae [3] 1 EQ or I [1] c

aArmy ants belonging to the doryline section (5).bBQ: brachypterous queen, EQ: ergatoid queen, I: reproductive intercaste, G: gamergate.cEQ or I denotes ambiguity in classication, on the basis of published descriptions.

and the sclerites separate from one another, instead of normally overlapping) (28;Table 2). This allows for a dramatic increase in the rate of egg laying becausehundreds of oocytes can mature simultaneously. EQ never perform ICF (but see117), and ssion is obligate.

Evolutionary Loss of Queens In about 100 species of ants the queen caste isabsent altogether, and one or a few workers in a colony mate and lay fertilized eggs.These are termed gamergates to emphasize that they differ both in dispersal abil-ity and fecundity from queens (87, 89). Furthermore, gamergates cannot performICF, and ssion is obligate. Queenless species are all restricted to the subfamilyPonerinae, where workers in some taxa have retained a functional spermathecaowing to limited divergence from the ancestral monomorphic females.

Budding Coexists with Independent Foundation

Colony division occurs in addition to claustral ICF in a substantial number of antspecies belonging to the subfamilies Myrmicinae, Formicinae, and Dolichoderi-



14/32

614 PEETERS ITO

TABLE 2 Colony size and reproductive specialization in ant species with ergatoid queens

Number of No. ovariolesSpecies a workers b in queens Physogastry? Reference

Subfamily MYRMECIINAE Myrmecia froggatti c 38 12 79/ovary no 52

Subfamily PONERINAETribe Amblyoponini

Onychomyrmex hedleyi 543 184 3/ovary no 88

Tribe EctatomminiGnamptogenys bicolor c 89 6/ovary no F. Ito, unpubl.

Tribe Ponerini

Anochetus faurei c 391, 476 67/ovary no 120 Leptogenys distinguenda 35 104 Unknown yes 67 Leptogenys diminuta 277 116 4/ovary yes 51 Leptogenys kraepelini 21 7 3/ovary no 48Pachycondyla analis c 583 174 3032/ovary no 62, 88

( = Megaponera foetens )

Subfamily LEPTANILLINAE Leptanilla japonica 100200 1417/ovary yes 68

Subfamily ECITONINAE

Labidus praedator 106

Unknown yes 28 Eciton burchelli 1570 104 1300/ovary yes 28 Eciton hamatum 525 104 Unknown yes 28 Neivamyrmex nigrescens 114 104 500/ovary yes 28

Subfamily DORYLINAE Dorylus wilverthi 1520 106 15000/queen yes 28, 42 Aenictus laeviceps 611 104 Unknown yes 28

Subfamily MYRMICINAE Myrmecina sp. Ac 130 96 2/ovary no 47a

aAll species are monogynous with the exception of Myrmecina sp. A.bMean standard deviation given if at least 3 colonies were excavated, otherwise sizes of individual colonies.cCongeneric species have alate queens.

polygyny is central to the mechanism of colony division in these species (112), incontrast to the strict monogyny characterizing ants with EQ. Secondary polygynyand budding are the basis of the high invasive success of tramp ants (84).

Secondary Polygyny Involving AQ Instead of attempting ICF, newly matedqueens can seek adoption in an established nest, either their natal colony or an



15/32


In Formica lugubris , some queens take part in nuptial ights, whereas others stayin their natal colony and mate with nestmate males (15). In the Formica rufa group,a proportion of female sexuals from each colony seem to shed their wings prema-

turely, suggesting a mixed dispersal strategy (100). This dispersal polymorphismis affected by the distribution and persistence of suitable habitats (8), as in solitaryinsects (134). Rosengren et al (100) argued that, in Formica ants, risky long-rangedispersal and low probability of successful ICF are key factors selecting for poly-gyny (through adoption of young queens) and budding. Predation, resource short-age, and habitat patchiness are costs affecting dispersal (9, 35). In US populationsof Solenopsis invicta , once appropriate nesting habitats are saturated, decreasedsuccess with ICF selects for queens seeking re-adoption, and polygynous coloniesmultiply by budding (101, 103). In the monogynous Cataglyphis cursor , AQ do

not y away and instead mate near their natal nest, which they re-enter. This resultsin temporary polygyny that triggers budding (59). ICF is thought to be uncommonyet AQ are retained.

In the higher ants, the occurrence of budding is associated with physiologi-cal and morphological modications in AQ, but not wing loss. Keller & Passera(56) showed that, between the time of emergence and the time of mating, AQ inspecies performing ICF accumulate much higher amounts of fat (295% increaseon average) than AQ doing ssion (48% on average). Furthermore, when the mat-ing ight disappears in various species, queens have lower amounts of glycogen

and free sugars (used as energy for ying) (85). Similarly, using queen-workerratios in thorax volume to estimate the relative investment in sexuals, a compara-tive study revealed much variability among congeneric species, depending on theoccurrence of either ICF or budding (110). The same trend is also found withinspecies. In Formica truncorum , the propensity to disperse is directly linked tothe physiological condition of individual queens (112). In S. invicta , colonies areeither monogynous with ICF or polygynous with budding, and the increase in ab-domen weight of female sexuals (prior to dispersal) is 70% higher in ICF colonies,representing mostly accumulation of fat (57).

Alternative modes of colony reproduction can also be associated with a sizepolymorphism among AQ (reviewed in 104). In Solenopsis geminata , macrogynesare both larger and heavier than microgynes, and only the former attempt ICF,while microgynes disperse at a different time of the year and are attracted toexisting colonies where they may replace a dying queen (71). There is no secondarypolygyny and no budding in this species, unlike in Leptothorax rugatulus withsimilarly dimorphic queens (105). Importantly, the smaller queens in both of thesespecies retain normal ight ability, in contrast to the brachypterous (short-winged)queens of other species (11).

Secondary Polygyny Involving Wingless Reproductives In contrast to the adop-



16/32

616 PEETERS ITO

originate from a blending of the queen and worker developmental pathways. Insubfamilies Myrmicinae and Dolichoderinae especially, a graded series of inter-castes connect AQ and workers (14, 88). In contrast, EQ represent a true queen

caste morphologically distinct from workers. Importantly, intercastes have a func-tional spermatheca ( Harpagoxenus canadensis is one exception; 13), unlike theworkers. Therefore, intercastes can produce diploid offspring. Intercastes are pro-duced erratically in most ants, but in others they occur in large numbers and havea reproductive function (Table 1), often as part of a life history that includes AQ.

Technomyrmex albipes (Dolichoderinae) produces polydomous colonies con-sisting of up to several millions of adults (131). AQ outbreed and initiate coloniesalone. When colonies reach a certain size, numerous and morphologically vari-able intercastes are reared, and most copulate with nestmate wingless males

polygyny ensues. In several myrmicine ants also, both AQ and intercastes repro-duce (11, 14,32, 34,78). Single AQ also start monogynous colonies independentlyand larger colonies are polygynous, although there is no inbreeding. In Myrmecinanipponica , intercaste colonies are very abundant at high latitudes where suitablehabitats are patchily distributed (78). In Leptothorax sp. A, queen colonies areabundant in extendedboreal forests while intercastecolonies occur in isolated habi-tat patches (32). In Hypoponera , species in forests have AQ only (and monogyny),while species in disturbed habitats tend to have intercastes and polygyny (132).

Reproductive intercastes seem an adaptation to enlarge established colonies,

and create new ones, without incurring dispersal costs. This is analogous to theevolution of EQ, which dispenses with the need of producing expensive queens.However, reproductive intercastes typically coexist with AQ, whereas EQ com-pletely replace AQ. Furthermore, reproduction by intercastes is always associatedwith secondary polygyny, whereas reproduction by EQ is usually monogynous.Nonetheless, distinguishing these two evolutionary endpoints is difcult in somespecies, partly because of incomplete evidence. One criterion is that EQ are mor-phologically invariant within species, unlike intercastes (88).

AQ Remain Crucial for Dispersal Polymorphism Colony reproduction is notdependent on ICF in various species of higher ants, but AQ continue to be produced.Nonetheless, diverse physiological adaptations are exhibited by AQ that donot disperse (e.g. less metabolic reserves, poor ying ability, and small size),presumably to reduce reproductive investment of colonies. It seems generally truethat secondary reproductives (microgynes and intercastes) are less fecund thanfounding queens, which may explain why they are polygynous.

Hamilton & May (29a) pointed out that adaptations for dispersal remain ad-vantageous even in stable and saturated habitats. The relative success of dispersal

varies with local conditions (e.g. the cost/benet ratio of staying in a restrictedpatch increases with time), and the need for exibility explains why permanently



17/32


with many ponerine and army ants in which the ability to disperse by ight andperform semiclaustral ICF has been lost permanently.

INSIGHTS FROM PHYLOGENETICALLYPRIMITIVE ANTS

The ant subfamily Ponerinae ( 1300 species) is characterized by the retention of a relatively large proportion of ancestral morphological characters. These ants livemostly in tropical and subtropical regions and are predators, either opportunisticin their choice of prey or highly prey specic. Colonies are generally small(101103 adults) and often occur at very low densities. Many species nest in preex-isting structures that are modied only slightly, and they relocate whenever spacebecomes limiting or otherwise unsuitable (90).

Queen-workerdimorphismis least pronouncedin thePonerinae(also in Myrme-ciinae), and this appears linked to a great diversity of reproductive modications.Queens are EQ in many species (Table 2), or workers reproduce sexually (inspecies with and without AQ; Table 3), making this taxonomic group ideal for acomparative study of adaptations for ssion. Colony ssion is obligate in specieswithout ying queens, and ponerine ants can enrich our knowledge about theproximate and ultimate issues associated with ssion. In contrast to army ants,many species have < 100 workers per colony, which may imply a different patternof ssion (89).

Frequent Occurrence of Wingless Reproductive FemalesThere are 42 genera of Ponerine ants arranged in ve tribes (6), and wingless repro-ductive females are known from 19 genera (in four tribes). A reliable phylogenyof Ponerinae is currently lacking, but recent molecular data indicate that it is para-phyletic (111). The existence of AQ is the ancestral state, given the occurrence of

wings in most Aculeates. We can attempt a conservative estimate of the numberof independent losses of AQ; modern revisions are available for most genera, andthey are monophyletic. Since EQ species occur together with AQ species in 11genera, wing loss must have evolved 11 times. Another two genera have onlyEQ. In parallel, queenless species occur together with AQ species in six genera.In Rhytidoponera species, AQ have disappeared in three separate species groupsdifferentiated by using morphological as well as molecular data (97). The frequentdisappearance of AQ in the Ponerinae contrasts with their persistence in othersubfamilies.

AQ are seldom very divergent from workers in the Ponerinae, with respect toboth their fecundity and their ability to start new colonies independently. As a



18/32

618 PEETERS ITO

TABLE 3 Colony size and nesting preference of a selection of ponerine ants exhibiting gamer-gate reproduction

Species (Monogynous/ Queens Number of workers Nestingpolygynous) or not? (N colonies collected) site a References

Tribe Amblyoponini (7 genera) Amblyopone reclinata (P) G 96 51 (N = 22) 2 46

Tribe Ectatommini (8 genera)Gnamptogenys menadensis (P) Q + G 113 75 (N = 37) 4 27 Rhytidoponera confusa (P) Q + G 164 208 (N = 64)b 2/ 3 124 Rhytidoponera sp. 12 (P) G 577 281 (N = 9) 1 82, 86, 113a

Tribe Platythyreini (2 genera)

Platythyrea quadridenta (P) Q + G 19 15 (N = 13) 3 47Platythyrea schultzei (M) G 21 10 (N = 7) 2 119Platythyrea lamellosa (M) G 115 83 (N = 25) 1 122

Tribe Ponerini (22 genera) Diacamma sp. from Japan (M) G 118 88 (N = 22) 2 25 D. ceylonense (M) G 231 109 (N = 21) 1 K Vedham &

R Gadagkarunpubl.

Dinoponera australis (M) G 13 6 (N = 37) 1 80 D. quadriceps (M) G 82 29 (N = 17) 1 74 Harpegnathos saltator (P) Q + G 65 40 (N = 59) 2 94 Leptogenys peuqueti (P) G 30 24 (N = 13) 2 48 and unpubl. Leptogenys schwabi (M) G 184 76 (N = 17) 2 19aPachycondyla Q + G 9.4 5.5 (N = 12) 2 44 and unpubl.

( Bothroponera ) sp. (M)P. ( Bothroponera ) sublaevis (M) G 9.6 3.9 (N = 37) 1 36P. ( Hagensia ) havilandi (M) G 28 8 (N = 9) 2 118Streblognathus aethiopicus (M) G 36 21 (N = 19) 1/ 2 C Peeters &

R Creweunpubl.

aNesting sites: 1, underground ( > 50 cm); 2, underground ( < 50 cm); 3, rotten wood lying on ground; 4, arboreal.bGamergate colonies only.

by the less risky process of ssion in the Ponerinae (90). Nevertheless, manyPonerinae keep both AQ and ICF. In a survey of 89 ponerine species from theoriental tropics, 48 species had AQ, 20 species had EQ, 12 species had AQ toge-ther with gamergates, and 9 were permanently queenless (49).

It is evident that AQ remain valuable even though restricted to semiclaus-

tral ICF. Foundation success may be improved in some species by more pro-nounced queen-worker dimorphism, as well as by specic behavioral adaptations.



19/32


sufcient to compensate for a low success rate. However, the number of femalesexuals reared is generally limited by colony size. In Paltothyreus tarsatus in whichcaste dimorphism is relatively pronounced, the bigger colonies (1000 4000 work-

ers) produce hundreds of AQ annually (10). In species having AQ and workersthat differ little in dry weight and size (e.g. Harpegnathos saltator ), dozens of AQare released from the small colonies ( < 100 workers) each year (94).

Dispersal Dimorphism and Evolution of Gamergate Reproduction

Sexual reproduction by workers is an unusual type of social structure in the ants.Recent data on a few species in which both alate queens and workers can mateand produce diploid offspring (Q + G) (Table 3) give a valuable insight into theevolution of gamergates. Colonies are independently founded by AQ, and when thefoundress dies, several gamergates begin to reproduce ( H. saltator , Pachycondylaspp., Gnamptogenys menadensis , R. confusa ; 27, 49, 93, 124). The life historyof Q + G species parallels that of various higher ants in which newly mated AQor intercastes reproduce in established colonies (initially begun by ICF), leadingto secondary polygyny. Thus the evolution of gamergate reproduction appearsstrongly associated with the adaptive benets of secondary polygyny (e.g. in-creased colony lifespan and resource inheritance; 75), and it is the preferred option

in species having workers able to reproduce sexually.Among Q + G species, the frequency of queen production and the proportionof queen colonies vary greatlyqueen colonies are very rare in several species,while more than 50% have queens in H. saltator and R. confusa (27, 49, 94, 124).Gamergate reproduction without queens (so called queenless species) seems to bethe extremein a continuum ofQ + G species; ICF is thenno longerpossible. In fact,mostqueenless antsoccur ingenerawith AQ orQ + G species. Only in three genera( Diacamma, Dinoponera and Streblognathus ) are all species queenless; these areall monogynous, which appears to be a derived character in gamergate species. In

the near future, molecular data may unravel the phylogenetic relationships amongcongeneric species; once the extant ancestors of Q + G species are known, we caninvestigate the reproductive abilities of their workers.

Fission and the Pattern of Reproductive InvestmentBoth EQ and gamergate reproduction seem to be evolutionary responses to theexistence of AQ that are not very efcient for ICF. As a consequence of switchingto obligate ssion, EQ or gamergates reproduce instead of AQ (90). It remainsunclear what determines the occurrence of gamergates as opposed to EQ. Workers

in EQ species are usually reproductively degenerate (50), and thus gamergatereproduction is impossible. The genus Leptogenys (175 species) is one of the



20/32

620 PEETERS ITO

forests, and colonies nest in short-lived sites (dead branches fallen on the ground,under stones or in compressed leaf litter). The great majority of species haveEQ (including Leptogenys ergatogyna , which has a ight thorax but no wings),

whereas one species has AQ (3) and a few have gamergates (Table 3). In theoriental tropics, most gamergate species of Leptogenys occur in disturbed areas,whereas EQ species occur mostly in forests. All gamergate species are polygynous(19a, 48), whereas EQ species are monogynous with three known exceptions (48).There is little caste dimorphism in most species of Leptogenys , with the overallbody length of EQ generally the same as that of workers. Accordingly, theseEQ have a low egg-laying rate and colonies have < 500 workers (Figure 3). In therainforests of the oriental tropics, 7 of 10 EQ species have < 40 workers per colony.In contrast, a few oriental species have colonies with several tens of thousands of

workers and they behave like army ants; they lack xed nests and raid in swarms of about 3000 workers (67, 130a). In such species, EQ have a high fecundity whichis reected by their considerably enlarged abdomen (Table 2). Despite severaldetailed investigations, no data are available on the annual production of EQ,although males are occasionally found (48). In conclusion, the most importantdifference between gamergate and EQ species is that the latter have regularlyevolved larger colony sizes (Figure 3). This seems the case in other ponerinegenera as well (90). Larger colonies allow for sophisticated foraging strategiesand a modied ecological niche.

Leptogenys sp . 35L. parvula

Leptogenys sp .13L. kraepelini

Leptogenys sp . 21

L. mjobergi L. longensis

L. attenuata L. diminuta L. castanea

L. nitida Leptogenys sp .12

L. distinguenda Leptogenys sp . 24Leptogenys sp . 22

L. peuqueti

L. schwabi

0 10 100 1000 10000 100000

Species withergatoid queens

Species withgamergates

787

3

3

10

4

4

5

8

6

4

13

10

1317

1

Colony size



21/32


The success rate of a reproductive female that is accompanied by nestmateworkers is greatly increased relative to ICF. Fission is absolutely dependent onthe production of a sufcient number of workers so that both post-ssion colonies

can function as viable units. A minimum colony size is important in species thatare specialized predators on large arthropods, such as Amblyopone (45) and Lep-togenys . Workers thus represent the major portion of reproductive investment(63, 76). Selection for successful ssion may favor an increase in colony size,and this is possible since investment in female sexuals is considerably reducedor eliminated. However, in species having only a few dozens of workers in theircolonies (Tables 2 and 3), producing a sufcient number of new workers to supportssion may be a formidable task. Studies of both army ants and honeybees haveled to the general expectation that colonies can undergo ssion every year, but this

may not be valid for several ants. Ponerine species with small colonies may beable to divide only if they have managed to enlarge sufciently, and this probablyoccurs at very irregular intervals. The annual production of males is their onlypredictable means of spreading their genes in the population. In some years it maybe better not to divide at all (81). The rst priority of individual colonies is toreplace dying nestmates. The potentially long and indeterminate lifespan of theircolonies makes it likely that there will be chances of ssion in the future.

Fission and Nest RelocationVery little is known about ssion in ponerine ants with EQ or gamergates. However,various ponerine species frequently shift nest sites (24, 25, 58, 62, 66, 67, 109a),and the behavioral mechanisms involved in nest emigration can give us an insightinto the process of ssion. It remains to be determined whether ssion results fromaccidental fragmentation during nest relocation, or if it is an organized process.The great variability in colony sizes of individual species (Tables 2 and 3) suggeststhat the threshold at which ssion occurs is not rigid.

Colony emigration often occurs as a result of mechanical disturbance to nests,

altered microclimate, ooding, or predation (reviewed in 90). Emigrations are or-ganized by sophisticated communication among workers, including tactile signals,stridulation, and pheromone trails (42). Specialized behaviors exist to carry adultsand different brood stages (25, 73) over considerable ground distances. Emigra-tions can be initiated by specialized foragers that scout for suitable nesting sites,and, if successful, return to the colony and proceed to recruit nestmates, often us-ing pheromones (24, 66). In M. foetens , there are no trails but emigrating coloniesare led by a single worker, and regular bivouacs are used to regroup workers andbrood (61).

It has been assumed that the size of the colony fragment that starts an indepen-dent existence affects the distance over which dispersal takes place (8), but this



22/32

622 PEETERS ITO

relocated over distances of 59 m (mean 19 m) (67, 130a), while Leptogenyssp. 13 (near kraepelini ) changed nests every few days (median 2.5 days) and moved0.44.4 m (109a) (colony sizes in Table 2). Emigration distance is also likely to

be inuenced by the size of individual ants. We assume that dispersal distancesduring ssion will be similar to these relocation distances.

Fission is obligate both in species with EQ and in species with gamergates, butthey differ in one important respect. In the latter, ssion needs not be precededby the production of specialized sexuals, and thus we can expect it to occur moreopportunistically than in EQ species. Since several queenless ants are polygynous,it will be of interest to compare their characteristics of ssion with that in monog-ynous species (both gamergate and EQ species). Fission of monogynous coloniesoften results in the differentiation of a new gamergate, as in Diacamma (89). In

ergatoid species, the production of female sexuals and ssion are two separateprocesses. Field data collected by both the authors of this review over 15 years in-dicate that colonies are almost never found with new EQ (either as pupae or newlyeclosed adults), although males are regularly found. Thus we remain ignorant of the behavioral interactions that accompany the fragmentation of a colony.

SYNTHESIS AND FUTURE PERSPECTIVES

Colony ssion (or budding) is more widespread in the ants than is generallythought. Many monogynous ants exhibit ssion, not only army ants with theirtypically large colonies. Indeed, ponerine ants with only dozens or hundreds of workers per colony (Tables 2 and 3) reproduce exclusively by ssion. A commonmisconception is that, with the exception of army ants, ssion is more typical of polygynous species (54). This may stem from the absence from temperate regionsof themajority of specieswith EQ. A surveyof 24 Europeanant species showed thatssion occurs more commonly in polygynous species, but all of these had AQ (54).

We have reviewed evidence that ssion is the only mode of colony reproduction

in a proportion of species, while in others ssion is an alternative to ICF. Ecologicalconditions (e.g. degree of habitat saturation) selecting for either of these have beenstudied in only a few ant genera ( Leptothorax and Formica ; 32, 100). Ponerineants seem appropriate for comparative studies investigating population ecology,and whether colony size affects dispersal distance during colony division. Dospecies with multiple gamergates disperse as much as monogynous queenlessspecies (given similar colony sizes)?

In the ants, ICF or ssion can involve two morphological specializations thatare impossible in social wasps and bees. At one extreme, queens in various species

have huge ight thoraces in addition to other physiological adaptations (metabolicreserves accumulated before dispersal). The annual production of hundreds or



23/32


new and distant habitats. The irreversible loss of ying ability by some queensrepresents the other extreme. Ergatoid queens are the ultimate specialization forcolony ssion. They cannot survive alone, and their success depends on the num-

ber of adult workers that accompany them. The workers thus represent most of the reproductive investment; only a handful of EQ need be produced because anestablished colony can only divide once or twice each year. In many species EQare highly fertile, and colonies are dramatically larger. The loss of the queen casteis another adaptation for ssion, but it is restricted to a small number of ponerinesin which workers can reproduce sexually.

Unlike obligate ssion, budding is linked with a different set of adaptations inthe reproductives. Alate queens continue to be produced, and secondary polygynyresults when a foundress is succeeded by a new generation of reproductivesthese

canhavedifferentdevelopmental originsdependingon thetaxa (AQ readopted aftermating, intercastes, or gamergates). Females attempt to mate in the vicinity of theirnests, and the vagaries of aerial dispersal are dispensed with. Female mortalityreaches a minimum when they copulate inside their own colonies.

Social life affects dispersal within and connectedness between populations andso inuences thepattern of geographicdifferentiation (83). Studies of geneticstruc-turing in populations using molecular markers can provide valuable informationon the presumed cost of obligate colony ssion in ants. This is of special interestin species where queens have secondarily become wingless, or disappeared alto-

gether. Only the ying males can then contribute to gene mixing in the population,and the roles of males and females in causing gene ow can be separated, e.g.queenless species of Diacamma (J-B Andr e, C Peeters, C Doums, submitted forpublication) and Rhytidoponera (18, 113, 113b). Since colonies with wingless re-productive females cannot escape deteriorating habitats, such species are likely tohave a high extinction rate. Obligate ssion must also have an important inuenceon biogeography and speciation rate (125).

ACKNOWLEDGMENTS

We thank A Bourke, M Cobb, R Gadagkar, G Robinson, K Ross, P Ward, andD Wheeler for valuable comments on earlier drafts of this paper. Our work issupported by the French National Agency for Scientic Research (CNRS) and theJapan Ministry of Education, Science and Culture (Grant for Overseas Research).

Visit the Annual Reviews home page at www.AnnualReviews.org

LITERATURE CITED

1. Deleted in proof 2 Deleted in proof

Formicidae) in the Ethiopian region.Bull Br Mus Nat Hist Entomol 31:



24/32

624 PEETERS ITO

of dispersal strategy in the Monomoriumsalomonis group (Hymenoptera: Formici-dae). J. Nat. Hist. 20:26772

5. Bolton B. 1990. Army ants reassessed: thephylogeny and classication of the dory-line section (Hymenoptera: Formicidae). J. Nat. Hist. 24:133964

6. Bolton B. 1995. A taxonomic and zoo-geographical census of the extant ant taxa(Hymenoptera: Formicidae). J. Nat. Hist.29:103756

7. Boomsma JJ, Isaaks JA. 1985. Energy in-vestment and respiration in queens and

males of Lasius niger (Hymenoptera:Formicidae). Behav. Ecol. Sociobiol. 18:1927

7a. Bourke AFG. 1999. Colony size, so-cial complexity and reproductive con-ict in social insects. J. Evol. Biol. 12:24557

8. Bourke AFG, Franks NR. 1995. Social Evolution in Ants . Princeton, NJ: Prince-ton Univ. Press. 529 pp.

9. Bourke AFG, Heinze J. 1994. The ecol-ogy of communal breeding: the case of multiple-queen leptothoracine ants. Phi-los. Trans. R. Soc. London Ser. B 345:35972

10. Braun U, H olldobler B, Peeters C. 1994.Colony life cycle and sex ratio of the antPaltothyreus tarsatus in Ivory Coast. Proc. Int. Congr. Int. Union Study Soc. Insects,12th, Paris, France , p. 330. Paris: Univ.Paris-Nord (Abstr.)

11. BrieseDT. 1983.Differentmodesof repro-ductive behaviour (including a descriptionof colony ssion) in a species of Chelaner (Hymenoptera: Formicidae). Insectes Soc.30:30816

12. Buschinger A. 1983. Sexual behavior andslave raiding of the dulotic ant, Harpagox-enus sublaevis (Nyl.) under eld condi-tions (Hymenoptera: Formicidae). InsectesSoc. 30:23540

13. Buschinger A, Alloway TM. 1978. Caste

14. Buschinger A, Heinze J. 1992. Polymor-phism of female reproductives in ants. In Biology and Evolution of Social Insects ,

ed. J Billen, pp. 1123. Leuven, Belgium:Leuven Univ. Press15. Cherix D, Chautems D, Fletcher D,

Fortelius W, Gris G, et al 1991. Alter-native reproductive strategies in Formicalugubris Zett. (Hymenoptera Formici-dae). Ethol. Ecol. Evol. 1:6166

16. Choe J, Crespi B, eds. 1997. The Evolutionof Social Behavior in Insects and Arach-nids . New York: Cambridge Univ. Press.

541 pp.17. Crozier RH, Pamilo P. 1996. Evolution of Social Insect Colonies: Sex Allocation and Kin Selection . Oxford, UK: Oxford Univ.Press. 306 pp.

18. Crozier RH, Pamilo P, Crozier YC.1984. Relatedness and microgeographicgenetic variation in Rhytidoponera mayri ,an Australian arid-zone ant. Behav. Ecol.Sociobiol. 15:14350

19. Darchen R. 1977. Lessaimage chezles Hypotrigones au Gabon: dynamiquede quelques populations. Apidologie 8:3359

19a. Davies SJ, Villet MH, Blomeeld TM,Crewe RM. 1994. Reproduction and di-vision of labour in Leptogenys schwabiForel (Hymenoptera: Formicidae), apolygynous, queenless ponerine ant. Ethol. Ecol. Evol. 6:50717

20. Dejean A, Lachaud J-P. 1994. Ecologyand behavior of the seed-eating poner-ine ant Brachyponera senaarensis (Mayr). Insectes Soc. 41:191210

21. Dussmann O, Peeters C, Holldobler B.1996. Morphology and reproductive be-haviour of intercastes in the ponerineant Pachycondyla obscuricornis . InsectesSoc. 43:42125

22. Francoeur A, Loiselle R, Buschinger A.1985. Biosyst ematique de la tribu Lep-tothoracini (Formicidae, Hymenoptera).



25/32


23. Franks NR, H olldobler B. 1987. Sexualcompetition during colony reproductionin army ants. Biol. J. Linn. Soc. 30:

2294324. Freitas AV. 1995. Nest relocation andprey specialization in the ant Leptog-enys propefalcigera Roger (Formicidae:Ponerinae) in an urban area in southeast-ern Brazil. Insectes Soc. 42:45356

25. Fukumoto Y, Abe T. 1983. Social organi-zation of colony movement in the tropi-cal ponerine ant, Diacamma rugosum (LeGuillou). J. Ethol. 1:1018

26. Gadagkar R. 1991. Belonogaster , Mischocyttarus , Parapolybia , andindependent-founding Ropalidia . SeeRef. 102, pp. 14990

27. Gobin B, Peeters C, Billen J. 1998.Colony reproduction and arboreal lifein the ponerine ant Gnamptogenysmenadensis . Neth. J. Zool. 48:5363

27a. Gordon DM, Kulig A. 1996. Founding,foraging and ghting: colony size and the

spatial distribution of harvester ant nests. Ecology 77:2393409

28. Gotwald WH Jr. 1995. Army Ants: The Biology of Social Predation . Ithaca, NY:Cornell Univ. Press. 302 pp.

29. Greene A. 1991. Dolichovespula andVespula . See Ref. 102, pp. 263305

29a. Hamilton WD, May RM. 1977. Dispersalin stable habitats. Nature 269:57881

30. Haskins CP. 1970. Researches in the biol-ogy and social behavior of primitive ants.In Development and Evolution of Behav-ior , ed. L Aronson,E Tobach,D Lehrman,J Rosenblatt, pp. 35588. San Francisco:Freeman

31. Haskins CP. 1978. Sexual calling be-havior in highly primitive ants. Psyche85:40715

32. Heinze J. 1993. Habitat structure, disper-salstrategiesandqueennumberintwobo-real Leptothorax ants. Oecologia 96:3239

34. Heinze J, Buschinger A. 1987. Queenpolymorphism in a non-parasitic Leptotho-rax species (Hymenoptera: Formicidae).

Insectes Soc. 34:284335. Heinze J, Tsuji K. 1995. Ant reproductivestrategies. Res. Popul. Ecol. 37:13549

36. Higashi S, Ito F, Sugiura N, OhkawaraK. 1994. Workers age regulating the lin-ear dominance hierarchy in the queen-less ponerine ant Pachycondyla sublaevis(Hymenoptera: Formicidae). Anim. Behav.47:17984

37. Holldobler B. 1976. The behavioral ecol-

ogy of mating in harvester ants (Hy-menoptera: Formicidae: Pogonomyrmex ). Behav. Ecol. Sociobiol. 1:40523

38. Holldobler B, Bartz SH. 1985. Sociobi-ology of reproduction in ants. In Exper-imental Behavioral Ecology and Socio-biology , ed. B Holldobler, M Lindauer,pp. 23757. Stuttgart: Gustav FischerVerlag

39. Holldobler B, Carlin N. 1985. Colony

founding, queen dominance and oligo-gyny in the Australian meat ant Iridomyr-mex purpureus . Behav. Ecol. Sociobiol.18:4558

40. Holldobler B, Haskins CP. 1977. Sexualcalling behavior in primitive ants. Science195:79394

41. Holldobler B, Wilson EO. 1977. The num-ber of queens: an important trait in ant evo-lution. Naturwissenschaften 64:815

42. Holldobler B, Wilson EO. 1990. The Ants .Cambridge,MA:Harv. Univ. Press. 732pp.

43. Inoue T, Sakagami SF, Salmah S, YamaneS. 1984. The process of colony multiplica-tion in the Sumatran stingless bee Trigonalaeviceps . Biotropica 16:10011

44. Ito F. 1993. Functional monogyny anddominance hierarchy in the queenlessponerine ant Pachycondyla sp. in WestJava, Indonesia. Ethology 95:12640

45. Ito F. 1993. Observation of group re-cruitment to prey in a primitive poner-



26/32

626 PEETERS ITO

46. Ito F. 1993. Social organization in a prim-itive ponerine ant: queenless reproduc-tion, dominance hierarchy and functional

polygyny in Amblyopone sp. (reclinatagroup). J. Nat. Hist. 27:13152447. Ito F. 1994. Colony composition of the

twoMalaysianPonerineants, Platythyreaquadridenta and P. tricuspidata: sexualreproduction by workers and productionof queens. Psyche 101:20918

47a. Ito F. 1996. Colony characteristics of an Indonesian myrmecine ant Myrmecinasp. (Hymenoptera, Formicidae, Myrmic-

inae): polygynous reproduction by er-gatoid queens. Ann. Entomol. Soc. Am.89:55054

48. Ito F. 1997. Colony characteristics andmorphological caste differentiation in theponerine ant genus Leptogenys in orientaltropics. Ethol. Ecol. Evol. 8:33543

49. Ito F. 1998. Diversity of colony structurein oriental ponerine ants. See Ref. 107, p.226

49a. Ito F. 1999. Male behavior and regula-tion of worker mating in a ponerine ant,Pachycondyla ( Bothroponera ) sp. (Hy-menoptera: Formicidae). J. Insect Behav.12:19398

50. Ito F, Ohkawara K. 1994. Spermath-eca size differentiation between queensand workers in primitive ants: re-lationship with reproductive structureof colonies. Naturwissenschaften 81:13840

51. Ito F, Ohkawara K. 2000. Productionand behavior of ergatoid queens in twospecies of the oriental ponerine genus Leptogenys (diminuta -group). Ann. Ento-mol. Soc. Am. 93:86973

52. Ito F, Sugiura N, Higashi S. 1994. Workerpolymorphism in the red-head bulldogant (Hymenoptera: Formicidae), with de-scription of nest structure and colonycomposition. Ann. Entomol. Soc. Am.87:33741

54. Keller L. 1991. Queen number, mode of colony founding, and queen reproductivesuccess in ants (Hymenoptera Formicidae).

Ethol. Ecol. Evol. 3:3071655. Keller L ed. 1993. Queen Number and Sociality in Insects . Oxford, UK: OxfordUniv. Press. 439 pp.

56. Keller L, Passera L. 1989. Size and fatcontent of gynes in relation to the modeof colony founding in ants (Hymenoptera;Formicidae). Oecologia 80:23640

57. Keller L, Ross KG. 1993. Phenotypic plas-ticity and cultural transmission of alter-

native social organizations in the re antSolenopsis invicta . Behav. Ecol. Sociobiol.33:12129

58. Leal IR, Oliveira PS. 1995. Behavioralecology of the neotropical termite-huntingant Pachycondyla ( = Termitopone) margi-nata : colony founding, group-raiding andmigratory patterns. Behav. Ecol. Sociobiol.37:37383

59. Lenoir A, Qu erard L, Pondicq N, Berton F.1988. Reproduction anddispersal in theantCataglyphis cursor (Hymenoptera, Formi-cidae). Psyche 95:2144

60. Deleted in proof 61. Longhurst C, Howse PE. 1979. Foraging,

recruitment and emigration in Megapon-era foetens (Fab.) (Hymenoptera: Formi-cidae) from the Nigerian Guinea savanna. Insectes Soc. 26:20415

62. LonghurstC, HowseP. 1979. Some aspectsof the biology of the males of Megaponera foetens (Fab.) (Hymenoptera: Formicidae). Insectes Soc. 26:8591

63. Macevicz S. 1979. Some consequences of Fishers sex ratio principle for social Hy-menoptera that reproduce by colony s-sion. Am. Nat. 113:36371

64. Makino S. 1989. Usurpation and nest re-building in Polistes riparius : two waysto reproduce after the loss of the origi-nalnest (Hymenoptera:Vespidae). InsectesSoc. 36:11628



27/32


cuspidatus : an ant with a novel mode of life. Behav. Ecol. Sociobiol. 17:17184

66. Maschwitz U, Schonegge P. 1983. For-

age communication, nest moving recruit-ment, and prey specialization in the ori-ental ponerine Leptogenys . Oecologia 57:17582

67. Maschwitz U,Steghaus-Kovac S,GaubeR,Hanel H. 1989. A South East Asian poner-ine ant of the genus Leptogenys (Hym.,Form.) with army ant life habits. Behav. Ecol. Sociobiol. 24:30516

68. Masuko K. 1990. Behavior and ecology

of the enigmatic ant Leptanilla japonicaBaroni Urbani (Hymenoptera: Formicidae:Leptanillinae). Insectes Soc. 37:3157

69. Matsuura M. 1991. Vespa and Provespa.See Ref.102, pp. 23262

70. Matsuura M. 1999. Size and compositionof swarmingcolonies in Provespa anomala(Hymenoptera, Vespidae), a nocturnal so-cial wasps. Insectes Soc. 46:21923

71. McInnes DA, Tschinkel WR. 1995. Queendimorphism and reproductive strategiesin the re ant Solenopsis geminata (Hy-menoptera: Formicidae). Behav. Ecol. So-ciobiol. 36:36775

72. Michener CD. 1974. The Social Behavior of the Bees: A Comparative Study . Cam-bridge, MA: Harv. Univ. Press

73. Moglich M, Holldobler B. 1974. Socialcarrying behavior and division of laborduring nest moving in ants. Psyche 81:21936

74. Monnin T, Peeters C. 1998. Monogyny andregulation of worker mating in the queen-less ant Dinoponera quadriceps . Anim. Be-hav. 55:299306

75. Myles TG. 1988. Resource inheritance insocial evolution from termites to man. InThe Ecology of Social Behavior , ed. CNSlobodchikoff, pp. 379423. San Diego,CA: Academic

76. Nonacs P. 1993. The effects of polygynyand colony life history on optimal sex in-

determination in eusocial wasps (Hy-menoptera: Vespidae). Annu. Rev. Ento-mol. 43:32346

78. Ohkawara K, Ito F, Higashi S. 1993.Production and reproductive function of intercastes in Myrmecina graminicola nip- ponica colonies (Hymenoptera; Formici-dae). Insectes Soc. 40:110

79. Ortius D, Lechner K. 1997. Nuptial ightin two ponerine ants, Pachycondyla im- pressa and P. fauveli , overlooking MachuPicchu, Peru. Stud. Neotrop. Fauna Envi-ron. 32:22729

80. Paiva RV, Brand ao CR. 1995. Nests,worker population, and reproductive statusof workers, in the giant queenless poner-ine ant Dinoponera Roger (HymenopteraFormicidae). Ethol. Ecol. Evol. 7:297312

81. Pamilo P. 1991. Evolution of colony char-acteristics in social insects. I. Sex alloca-tion. Am. Nat. 137:83107

82. Pamilo P, Crozier RH, Fraser J. 1985.Inter-nest interactions, nest autonomy, andreproductive specialization in an Aus-tralian arid-zone ant, Rhytidoponera sp.12.Psyche 92:21736

83. Pamilo P, Gertsch P, Thoren P, Sepp a P.1997. Molecular population genetics of so-cial insects. Annu. Rev. Ecol. Syst. 28:125

84. Passera L. 1994. Characteristics of trampspecies. In Exotic Ants : Biology , Impact ,and Control of Introduced Species , ed.DF Williams, pp. 2324 Boulder, CO:Westview

85. Passera L, Keller L. 1990. Loss of mat-ing ight and shift in the pattern of carbohydrate storage in sexuals of ants(Hymenoptera; Formicidae). J. Comp.Physiol. B 160:20711

86. Peeters C. 1987. The reproductive divisionof labour in the queenless ponerine ant Rhytidoponera sp. 12. Insectes Soc. 34:7586

87. Peeters C. 1991. The occurrence of sexual



28/32

628 PEETERS ITO

88. Peeters C. 1991. Ergatoid queens and in-tercastes in ants: two distinct adult formswhich look morphologically intermedi-

atebetween workersandwingedqueens. Insectes Soc. 38:11589. Peeters C. 1993. Monogyny and poly-

gyny in ponerine ants with or withoutqueens. See Ref. 55, pp. 23461

90. Peeters C. 1997. Morphologicallyprimitive ants: comparative review of social characters, and the importance of queen-worker dimorphism. See Ref. 16,pp. 37291

91. Peeters C, Crewe R. 1986. Male biol-ogy in the queenless ponerine ant Oph-thalmopone berthoudi (Hymenoptera:Formicidae). Psyche 93:27784

92. Peeters C, Crozier RH. 1988. Caste andreproduction in ants: not all mated egg-layers are queens. Psyche 95:28388

93. Peeters C, H olldobler B. 1995. Repro-ductive cooperation between queens andtheir mated workers: the complex lifehistory of an ant with a valuable nest.Proc. Natl Acad. Sci. USA 92:1097779

94. Peeters C, Liebig J, H olldobler B. 2000.Sexual reproduction by both queens andworkers in the ponerine ant Harpeg-nathos saltator . Insectes Soc. 47: Inpress

95. Porter SD, Tschinkel WR. 1986. Adap-tive value of nanitic workers innewly founded Red Imported FireAnt colonies. Ann. Entomol. Soc. Am.79:72326

96. Reeve HK. 1991. Polistes. See Ref. 102,pp. 99148

97. Reichel H. 1998. Molecular phylogenyof Rhytidoponera . See Ref. 107, pp. 390

98. Robertson HG, Villet M. 1989. Matingbehaviour in three species of myrmicineants (Hymenoptera: Formicidae). J. Nat. Hist. 23:76773

99. Roff DA. 1994. Habitat persistence andthe evolution of wing dimorphism in in-

W. 1993. Monogyny and polygyny inFormica ants: the result of alternativedispersal tactics. See Ref. 55, pp.

30833101. Ross KG, Keller L. 1995. Ecology andevolution of social organization: insightsfrom re ants and other highly eusocialants. Annu. Rev. Ecol. Syst. 26:63156

102. Ross KG, Matthews RW, eds. 1991. TheSocial Biology of Wasps . Ithaca, NY:Cornell Univ. Press. 678 pp.

103. Ross KG, Vargo EL, Keller L. 1996. So-cial evolution in a new environment: the

case of introduced re ants. Proc. Natl Acad. Sci. USA 93:302125

104. Ruppell O, Heinze J. 1999. Alterna-tive reproductive tactics in females: thecase of size polymorphismin wingedantqueens. Insectes Soc. 46:617

105. Ruppell O, Heinze J., H olldobler B.1998. Size dimorphism in the queens of the North American Leptothorax rugat-ulus (Emery). Insectes Soc. 45:6777

106. Schneirla TC. 1956. A preliminary sur-vey of colony division and related pro-cesses in two species of terrestrial armyants. Insectes Soc. 3:4969

107. Schwarz M, Hogendoorn K. 1998. So-cial insects at the turn of the mille-nium, Proc . Int. Congr. Int. Union StudySoc. Insects, 13th, Adelaide, Austr.Adelaide, Aust.: Flinders Univ.

108. Seeley TD. 1995. The Wisdom of the Hive: The Social Physiology of Honey Bee Colonies . Cambridge, Boston: Harv.Univ. Press. 295 pp.

109. Shellman-Reeve JS.1997.Thespectrumof eusociality in termites. See Ref. 16,pp. 5293

109a. Steghaus-Kovac S, Maschwitz U. 1993.Predation on earwigs: a novel diet spe-cialization within the genus Leptogenys(Formicidae: Ponerinae). Insectes Soc.40:33740

110. Stille M. 1996. Queen/worker thorax



29/32


111. SullenderBW, Johnson MJ.1998.A pre-liminary molecular phylogeny for theFormicidae. See Ref. 107, p. 460

112. Sundstr om L. 1995. Dispersal poly-morphism and physiological condi-tion of males and females in the antFormica truncorum . Behav. Ecol. 6:13239

113. Tay WT, Cook JM, Rowe DJ, CrozierRH. 1997. Migration between nestsin the Australian arid-zone ant Rhyti-doponera sp. 12 revealed by DGGEanalyses of mitochondrial DNA. Mol.

Ecol. 6:40311113a. Tay WT, Crozier RH. 2000. Nestmate

interactions and egg-laying behaviour inthe queenless ponerine ant Rhytidopon-era sp. 12. Insectes Soc. 47:13340

113b. Tay WT, Crozier RH. 2000. Microsatel-lite analysis of gamergate relatedness of the queenless ponerine ant Rhytidopon-era sp. 12. Insectes Soc. 47:18892

114. Tinaut A, Heinze J. 1992. Wing reduc-tion in ant queens from arid habitats. Naturwissenschaften 79:8485

115. Deleted in proof 116. Villet M. 1989. A syndrome leading to

ergatoid queens in ponerine ants (Hy-menoptera: Formicidae). J. Nat. Hist.23:82532

117. Villet M. 1991. Colony foundation inPlectroctena mandibularis F. Smith, andthe evolution of ergatoid queens in Plec-troctena (Hymenoptera: Formicidae). J. Nat. Hist. 25:97983

118. Villet M. 1991. Social biology of Ha-gensia havilandi (Forel 1901) (Hy-menoptera: Formicidae) and the originof queenlessness in ponerine ants. Trop. Zool. 5:195206

119. Villet M. 1991. Social differentiationand division of labour in the queen-less ant Platythyrea schultzei Forel1910 (Hymenoptera: Formicidae). Trop.Zool 4:1329

biology of ponerine ants (Hymenoptera:Formicidae). J. Nat. Hist. 25:160310

121. Villet M, Crewe R, Robertson H. 1989.

Mating behavior and dispersal in Pal-tothyreus tarsatus Fabr. (Hymenoptera:Formicidae). J. Insect Behav. 2:41317

122. Villet M, Hart A, Crewe R. 1990. Socialorganization of Platythyrea lamellosa(Roger) (Hymenoptera: Formicidae). I.Reproduction. S. Afr. J. Zool. 25:25053

123. Wagner DL, Liebherr JK. 1992. Flight-lessness in insects. Trends Ecol. Evol.7:21620

124. Ward PS. 1983. Genetic relatedness andcolony organization in a species com-plex of ponerine ants. I. Phenotypic andgenotypic composition of colonies. Be-hav. Ecol. Sociobiol. 12:28599

125. Ward PS. 1989. Genetic and socialchanges associated with ant speciation.In The Genetics of Social Evolution , ed.MN Breed, RE Page, pp. 12348. Boul-der, CO: Westview

126. West-Eberhard MJ. 1982. The natureandevolution of swarmingin tropical so-cial wasps (Vespidae, Polistinae,Polybi-ini). In Social Insects in the Tropics ,ed. P Jaisson, 1:97128. Paris: Univ.Paris-Nord

127. Wheeler DE. 1986. Developmental andphysiological determinants of caste insocial Hymenoptera: evolutionary im-plications. Am. Nat. 128:1334

128. Wheeler DE. 1996. The role of nourish-ment in oogenesis. Annu. Rev. Entomol.41:40731

129. Wheeler DE, Buck NA. 1995. Stor-age proteins in ants during developmentand colony founding. J. Insect Physiol.41:88594

130. Wheeler WM. 1916. The marriage-ight of a bull-dog ant ( Myrmecia san-guinea F. Smith). J. Anim. Behav. 6:7073

130a. Witte V, Maschwitz U. 2000. Raiding



30/32

630 PEETERS ITO

(Hymenoptera: Formicidae). InsectesSoc. 47:7683

131. Yamauchi K, Furukawa T, Kinomura K,

Takamine H, Tsuji K. 1991. Secondarypolygyny by inbred wingless sexuals inthe dolichoderine ant Technomyrmex al-bipes . Behav. Ecol. Sociobiol. 29:31319

132. Yamauchi K, Kimura Y, Corbara B,Kinomura K, Tsuji K. 1996. Dimorphicergatoid males and their reproductivebehavior in the ponerine ant Hypoponerabondroiti . Insectes Soc. 43:11930

133. Deleted in proof 134. Zera AJ, Denno RF. 1997. Physiol-

ogy and ecology of dispersal polymor-

phism in insects. Annu. Rev. Entomol. 42:20731135. Zucchi R, Sakagami SF, Noll FB,

Mechi MR, Mateus S. et al. 1995. Agelaia vicina , a swarm-founding polis-tine with the largest colony size amongwasps and bees (Hymenoptera: Vespi-dae). J. New York Entomol. Soc. 103:12937



31/32

Annual Review of EntomologyVolume 46, 2001

CONTENTS

BIOGEOGRAPHY AND COMMUNITY STRUCTURE OF NORTHAMERICAN SEED-HARVESTER ANTS, Robert A. Johnson 1

MATING BEHAVIOR AND CHEMICAL COMMUNICATION IN THEORDER HYMENOPTERA, M. Ayasse, R. J. Paxton, J. Teng 31

INSECT BIODEMOGRAPHY, James R. Carey 79

PREDICTING ST. LOUIS ENCEPHALITIS VIRUS EPIDEMICS:Lessons from Recent, and Not So Recent, Outbreaks, Jonathan F. Day 111

EVOLUTION OF EXCLUSIVE PATERNAL CARE IN ARTHOPODS, Douglas W. Tallamy 139MATING STRATEGIES AND SPERMIOGENESIS IN IXODIDTICKS, Anthony E. Kiszewski, Franz-Rainer Matuschka, AndrewSpielman 167

GENETIC AND PHYSICAL MAPPING IN MOSQUITOES: MolecularApproaches, David W. Severson, Susan E. Brown, Dennis L. Knudson 183

INSECT ACID-BASE PHYSIOLOGY, Jon F. Harrison 221EVOLUTION AND BEHAVIORAL ECOLOGY OFHETERONOMOUS APHELINID PARASITOIDS, Martha S. Hunter,

James B. Woolley 251

SPECIES TRAITS AND ENVIRONMENTAL CONSTRAINTS:Entomological Research and the History of Ecological Theory, Bernhard Statzner, Alan G. Hildrew, Vincent H. Resh 291

Genetic Transformation Systems in Insects, Peter W. Atkinson, AlexandraC. Pinkerton, David A. O'Brochta 317

TESTS OF REPRODUCTIVE-SKEW MODELS IN SOCIAL INSECTS, H. Kern Reeve, Laurent Keller 347

BIOLOGY AND MANAGEMENT OF GRAPE PHYLLOXERA, Jeffrey Granett, M. Andrew Walker, Laszlo Kocsis, Amir D. Omer 387

MODELS OF DIVISION OF LABOR IN SOCIAL INSECTS, Samuel N. Beshers, Jennifer H. Fewell 413

POPULATION GENOMICS: Genome-Wide Sampling of InsectPopulations, William C. Black IV, Charles F. Baer, Michael F. Antolin,

Nancy M. DuTeau 441

THE EVOLUTION OF COLOR VISION IN INSECTS, Adriana D. Briscoe, Lars Chittka 471

METHODS FOR MARKING INSECTS: Current Techniques and FutureProspects, James R. Hagler, Charles G. Jackson 511

RESISTANCE OF DROSOPHILA TO TOXINS, Thomas G. Wilson 545


32/32

CHEMICAL ECOLOGY AND SOCIAL PARASITISM IN ANTS, A. Lenoir, P. D'Ettorre, C. Errard, A. Hefetz 573

COLONY DISPERSAL AND THE EVOLUTION OF QUEENMORPHOLOGY IN SOCIAL HYMENOPTERA, Christian Peeters,Fuminori Ito 601

JOINING AND AVOIDANCE BEHAVIOR IN NONSOCIAL INSECTS, Ronald J. Prokopy, Bernard D. Roitberg 631

BIOLOGICAL CONTROL OF LOCUSTS AND GRASSHOPPERS, C. J. Lomer, R. P. Bateman, D. L. Johnson, J. Langewald, M. Thomas 667

NEURAL LIMITATIONS IN PHYTOPHAGOUS INSECTS:Implications for Diet Breadth and Evolution of Host Affiliation, E. A.

Bernays 703

FOOD WEBS IN PHYTOTELMATA: ""Bottom-Up"" and ""Top-Down"" Explanations for Community Structure, R. L. Kitching 729

Date post:	10-Apr-2018
Category:	Documents
Upload:	prosen21
View:	215 times
Download:	0 times

Colony Dispersal and the Evolution of Queen Morphology in Social Hymen Opt Era

Documents