The definition of eusociality

We describe more precise definitions for the term "eusociality"and other social systems. Our criterion for eusociality is thepresence of castes, which are groups of individuals that becomeirreversibly behaviorally distinct at some point prior to reproductivematurity. Eusocial societies are characterized by two traits:(1) helping by individuals of the less-reproductive caste, and(2) either behavioral totipotency of only the more reproductivecaste (facultative eusociality) or totipotency of neither caste(obligate eusociality). We define "cooperative breeding" asalloparental care without castes. Cooperatively breeding societiesmay comprise two types, semisocial (distribution of lifetimereproductive success bimodal), and quasisocial (distributionof lifetime reproductive success unimodal), but this hypothesisrequires empirical analysis. Our definitions conceptually unifystudies of arthropod and vertebrate sociality.     

Dispersal polymorphism and physiological condition of males and females in the ant, Formica truncorum

This study deals with dispersal behavior of sexuals and intraspecificvariation in queen numbers. The specific questions are: (1)Is there an association between male and female dispersal behaviorand the number of queens in a colony? (2) Is there an associationbetween individual behavior and physiological condition? (3)Do males and females from monogyne (one queen per colony) andpolygyne (several functional queens per colony) colonies differwith respect to size, weight, and physiological condition? Theresults show that both males and females are more prone to dispersein monogyne than in polygyne colonies. Moreover, males and femalesof both monogyne and polygyne colonies show dispersal polymorphism,suggesting that an increased tendency of reproductive femalesto stay in the maternal colony may cause monogyne colonies toswitch to polygyny. The propensity to disperse is associatedwith the physiological condition of individuals. Larger andheavier females containing more fat and glycogen preferentiallydisperse, whereas smaller ones with less fat and glycogen moreeasily dealate and mate without a previous nuptial flight. Maledispersal correlates positively to larger size and higher levelsof glycogen; fat contents do not increase during maturation.The females produced in monogyne colonies are larger, heavier,and contain more fat and glycogen than those produced in polygynecolonies. The males produced in monogyne colonies have relativelylonger wings and are heavier than those produced in polygynecolonies. However, there are no differences in size and fatcontents between males from monogyne and polygyne colonies.     

Energetic constraints and foraging efficiency

Previous research considers foraging options that differ interms of their gross rate of gain b and rate of energy expenditurec. This research argues that maximizing efficiency b/c willmaximize net energetic gain when there is an upper limit onthe amount of energy that can be assimilated. This analysisdoes not include the expenditure during the time for which theanimal is unable to forage because of this constraint. Whenthis expenditure is included, maximizing efficiency is no longeroptimal. Instead the best feeding option is the one with thehighest value of b/(c – c₁), where c, is the metabolicrate when the animal is not foraging.     

Sexual selection for large male size in a polyandrous butterfly: the effect of body size on male versus female reproductive success in Pieris napi

Sexual selection theory predicts that the larger sex shouldbe that for which fitness increases at the faster rate withsize. In butterflies, as in most invertebrates, females areusually the larger sex, but previous comparative analysis hasshown that relative male size increases with female polyandryamong butterflies. In agreement with this pattern, males arelarger than females in the strongly polyandrous green-veinedwhite butterfly, Pieris napi L., and in this article we assessthe size dependence of reproductive success in both sexes. Inan experiment where virgin males and females were released inthe field, we found no strong association between size and malemating success. However, laboratory experiments showed thatthere was a strong correlation between size and the ejaculatethat the male delivered to the female at mating and that largeejaculates delayed female remating for a longer time comparedto small ejaculates. Moreover, female P. napi utilize male-derivednutrients received at mating to increase their fecundity. Hence,large males sire more offspring both by way of donating morenutrients to female egg production and by way of delaying femaleremating (given that the last male to mate with the female willfather most of the offspring). Laboratory experiments showedthat the association between size and fecundity was low, ornonexistent, among P. napi females allowed to mate only once.However, weak size dependence was found for polyandrous females.We hypothesize that size dependence of female fecundity maybe especially weak among polyandrous butterflies because a fundamentalsource of variation in fecundity relates to their ability tofind nutrient giving males, an ability which may be unrelatedto female size. According to this hypothesis there is a causalassociation between weak size dependence of female fecundityand polyandry, and a strong size dependence of male reproductivesuccess that may underlie the comparative pattern of positivecorrelation between relative male size and polyandry.     

Suppressed breeding in the field vole (Microtus agrestis): an adaptation to cyclically fluctuating predation risk

The densities of microtine rodents and their main predators,small mustelids, fluctuate synchronously in 3–5-year cyclesin central and northern Fennoscandia. Predation by small mustelidshas been suggested as one of the driving forces in microtinecyclicity, causing deep synchronous declines of several volespecies. We studied experimentally the effects of small mustelidson mating behavior, foraging, and breeding in nonwintered fieldvoles (Microtus agrestis) originating from a cyclic population.By using mustelid odors, we simulated a crash phase environmentwith high predation risk for breeding pairs of voles. In ourexperiments, 87% of the female field voles suppressed breedingwhen exposed to mustelid odors. Both female and male behaviorchanged, and no mating behaviors were observed under the simulatedpredation risk. Weights of both sexes decreased when exposedto mustelid odor, probably due to decreased foraging; weightsof the control females increased due to pregnancy; and no weightchanges occurred in control males. Decreased breeding and foragingpossibilities under high predation pressure may form the basisfor the ultimate explanation for breeding suppression. Thereare at least two different mechanisms for breeding suppression:either mating does not take place or malnutrition in femalesdoes not' allow breeding to occur. Delayed breeding under highrisk of predation, for whatever reason, could increase the probabilityof individuals, especially that of the females, to survive overthe crash to the next, safer breeding season when their youngwould have better possibilities to survive.     

Dynamic discarding decisions: foraging theory for high-grading in a trawl fishery

The decision by fishermen to discard or retain fish of low valueto make room for more valuable fish in the hold of a boat (high-grading)is similar to diet choice problems faced by natural foragers.In our study, we apply the rationale of diet choice theory tohigh-grading behavior in the Oregon trawl fishery by treatingfishermen as foragers who must decide how much of each net'shaul to "ingest" before searching for more prey. We derive astate-dependent, temporal model of discarding behavior withina fishing trip. This optimization considers the availabilityof differently valued fish, trip quotas set by the regulatoryagency, and the risk of premature trip termination due to lossof gear or injury. The results indicate that those parametersaffect discarding behavior through their effect on the probabilityof exceeding the allowable catch, which we consider analogousto gut capacity. High-grading (partial prey consumption) occurredthroughout many simulated trips. The predictions were consistentwith the trends in discarding observed in the Oregon trawl fleet.Behavioral models such as ours can be useful to fishery managersby providing a means to explore the potential responses of fishermento new regulations before they are implemented.     

Investigating structure and temporal scale in social organizations using identified individuals

Studies of individually identified animals can produce substantialdata sets containing information on the structure and temporalscale of social organizations. However, methods of analyzingsuch data are not well established. Important features of asocial organization are revealed by plotting the rate of persistenceof the associations between pairs of individuals over a rangeof time lags (lagged association rate). The consistency of long-termrelationships can be characterized using the rate of associationof pairs of individuals between their first and last observedassociations (intermediate association rate). A hierarchicalseries of models featuring exponentially decaying lagged associationrates may be fitted to these data. This technique retrievedthe essential parameters of five simulated social organizationsand, when used on real data, portrayed the essential featuresof the patterns of temporal change in relationships betweenanimals. The method should be especially useful for analyzingfissionfusion societies containing 10–10, 000 individuallyidentifiable animals.     

Conflict between sexes in the water strider, Gerris lacustris: a test of two hypotheses for male guarding behavior

We studied the effect of operational sex ratio on female reluctanceand male persistence to mate as well as on the length of copulationand postcopulatory guarding in Gerris lacustris by adding fivesurplus males or females to the basin with a pair in tandem.In the control treatment, a pair alone was tested. Accordingto the copulatory guarding hypothesis (CGH), males should prolongmating and guard females in the presence of surplus males. Accordingto the convenience polyandry hypothesis (CPH), females shouldshow lower levels of resistance to prolonged mating in the presenceof surplus males because the mating male protects the femaleagainst harassment from other males. As expected on the basisof both the CGH and CPH, mating (copulation + guarding) averagedlonger in the male-biased treatment. The behavior of males andfemales during mating suggested that both hypotheses hold true:females showed less resistance to prolonged mating (as predictedfrom CPH), and male behavior suggested stronger efforts to stayon the female when surplus males were present (as predictedfrom CGH). Comparisons of the treatment with surplus femaleswith the results from the mating pair without surplus individualssuggested that the capabilities of water striders in tandemto assess the sex of nearby nonmating striders are limited.     

The location of landmark leks in the small heath butterfly, Coenonympha pamphilus: evidence against the hot-spot model

Two major theories have been developed to explain the locationand evolution of leks. According to the hot-spot model leksare male initiated and occur because males aggregate at locationsof maximum female home range overlap. The alternative femalepreference model, on the other hand, assumes that leks are femaleinitiated, and that females prefer to mate with males that aggregate.Small heath butterfly leks are situated close to landmarks suchas trees and bushes in open grassland. Censuses of wild populationsand experiments with artificial landmarks showed that theirattractiveness increased with height and width measured at themiddle of tree height. Landmark width at field layer heightwas negatively correlated with attractiveness, which suggeststhat a cornet shape is preferred to a pyramid shape. Togetherthese landmark size variables explained 60% of variation inlek size. Release experiments showed that landmarks were approachedby receptive virgin females, but mated females were indifferentto them. Landmark use was correlated to microclimatic conditions.Males preferred wider landmarks during cooler weather. Theyalso competed for the lee side of landmarks. There was no correlationbetween the location of landmarks and the dispersion of matedfemales or emergence sites of females. Hence, receptive femaledispersion was largely determined by their mate-locating behavior,which in turn appeared to result from male dispersion. Leksmay initially have developed around easily detected visual cuesthat offered a more suitable microclimate for male activity.Females visiting these locations would find a suitable matemore rapidly and save valuable time for oviposition. Althoughnonadaptive evolution cannot be excluded, taken together theresults support the female preference model for lek evolution.     

Testosterone effects on the immune system and parasite infestations in the barn swallow (Hirundo rustica): an experimental test of the immunocompetence hypothesis

The immunocompetence hypothesis predicts that testosterone (T)enhances the expression of male secondary sexual characterswhile exerting a suppressive effect on the immune system therebyexposing hosts to higher intensities of parasite infestations.In a natural population of barn swallow (Hirundo rustica) males,the intensity of infestation by some ectoparasites was negativelycorrelated with tail length and was positively correlated withimmunoglobulin levels, but no clear relationship was observedbetween immune responses (leukocyte counts, immunoglobulins)and tail length. Males implanted with T had higher intensitiesof parasite infestations at the time of recapture than controlmales, and T-implanted males experienced an increase in countsof eosinophils. In T-implanted males, immunoglobulin levelsinitially decreased and then increased as time from implantationelapsed. Among T-implanted males, those with longer tails hada smaller increase in eosinophil counts, tended to experiencea smaller increase of parasite infestations, and were more likelyto survive until the following breeding season than those withshorter tails. The relationships between parasite burden, immunesystem, and exaggeration of tail length in the natural populationof males are consistent with some aspects of the immunocompetencehandicap hypothesis. The results from the manipulation of Tplasma levels are also partly consistent with the hypothesis,since T-implantation resulted in higher levels of parasite infestations,but contradict the assumption of an obligatory immunosuppressiveeffect of T. Higher activation of the immune system of T-implantedmales indicate that high T plasma levels imposed a two-foldcost because of the effects on parasites and the immune responseto parasites, and this suggests that the effect of T on parasitesmight not be mediated by the immune system of the host. Theresults of the manipulation of T plasma levels support the handicapversion of the immunocompetence hypothesis since high quality,long-tailed males paid less in terms of activation of the immunesystem, change in parasite infestations, and chances of survivalthan low-quality, short-tailed males.     

Low frequency of extrapair paternity in the polygynous great reed warbler, Acrocephalus arundinaceus

We carried out DNA fingerprinting on 553 young (130 broods)great reed warblers (Acrocephalus arundnaceus) in 1987–1991.In the study population, where 40% of the males become polygynous,there was a low frequency of extrapair fertilizations (EPF).When data from all five years were pooled, 3.1% of the youngwere sired by extrapair males (EPF-males) and 5.4% of the broodscontained extrapair young. We found no cases of extrapair maternity;young with 6–17 mismatched DNA bands (n= 17) had highband sharing with their putative mothers (range = 0.52–0.72)but low band sharing with their putative fathers (range = 0.24–0.40).In broods exposed to EPF, on average 53% of the young were siredby EPF-males. We found the genetic father to each of the illegitimateyoung. In all cases the same EPF-male sired all extrapair youngin a brood. Broods containing EPF-young tended to be initiatedlate during the breeding season. Breeding attempts were ratherevenly distributed over two months, thus this breeding asynchronywould have facilitated EPFs. There was no difference in EPFfrequency between broods where the pair males had left theirfemales unguarded during parts of their fertile periods andbroods where males guarded throughout the fertile periods. Nestswith extrapair young had significantly shorter mean distanceto the closest male neighbor and more male neighbors within100 m than nests without extrapair young. We found no indicationthat females engaged in EPF to get parental care from the EPF-males,or because they were forced to copulate with extrapair males.The low frequency of EPF suggested that females did not seekgenetic diversity to their brood. We cannot rule out the possibilitythat females engaged in EPF to insure fertility. However, datasupporting this hypothesis were weak. Instead, our data supportthe conclusion that females engaged in EPF to increase the geneticquality of their offspring, and that females may have used malesong repertoire size as a cue when choosing EPF partners.     

Forum: The eusociality continuum

Eusocial societies are traditionally characterized by a reproductivedivision of labor, an overlap of generations, and cooperativecare of the breeders' young. Eusociality was once thought tooccur only in termites, ants, and some bee and wasp species,but striking evolutionary convergences have recently becomeapparent between the societies of these insects and those ofcooperatively breeding birds and mammals. These parallels haveblurred distinctions between cooperative breeding and eusociality,leading to calls for either drastically restricting or expandingusage of these terms. We favor the latter approach. Cooperativebreeding and eusociality are not discrete phenomena, but ratherform a continuum of fundamentally similar social systems whosemain differences lie in the distribution of lifetime reproductivesuccess among group members. Therefore we propose to array vertebrateand invertebrate cooperative breeders along a common axis, representinga standardized measure of reproductive variance, and to dropsuch (loaded) terms as "primitive" and "advanced" eusociality.The terminology we propose unites all occurrences of alloparentalhelping of kin under a single theoretical umbrella (e.g., Hamilton'srule). Thus, cooperatively breeding vertebrates can be regardedas eusocial, just as eusocial invertebrates are cooperativebreeders. We believe this integrated approach will foster potentiallyrevealing cross-taxon comparisons, which are essential to understandingsocial evolution in birds, mammals, and insects.     

Predicting group size in primates: foraging costs and predation risks

We present a direct test of the long-standing hypothesis thatfood competition limits primate group size. Group size is acritical social variable because it constrains most other aspectsof social organization. We develop a simple population-specificindex of indirect feeding competition based on daily foragingcosts. This index explains nearly two-thirds of between-populationvariation in mean group sizes of mostly fruit-eating (but notof mostly leaf-eating) primates. Group size is also significantlyrelated to body size and terrestriality (or use of open country),which are suspected correlates of predation risk, although feedingcompetition remains an important predictor of group size evenwhen these correlates are controlled. Phylogeny also appearsto be important: the differences between observed mean populationgroup sizes and those predicted using ecological factors aremost positive for the Old World monkeys and most negative forthe lemuroids in our sample. The weak relationship between groupsize and feeding competition found for folivorous species maybe explained either by the energetic constraints of a leafydiet or by limits to group size imposed by infanticide as ahabitual male reproductive strategy.     

Impact of habitat fragmentation on activity and hunting behavior in the tawny owl, Strix aluco

I examined the effect of woodland fragmentation on activityand hunting behavior in the tawny owl. I made three predictionson the effect of fragmentation, two of which were upheld andone rejected. Over 2 years, 24 owls were radio tracked in thenonbreeding season. Eight of these were in continuous woodlandand the rest in highly fragmented woodland. For each owl, Irecorded data on small mammal availability, diet, home rangesize, perch time, and interperch distance. In accordance withthe first two predictions, owls in fragmented woodlands hadlonger interperch distances and perch times, but contrary tothe third prediction, perch times were positively correlatedwith small mammal density. This was thought to be due to owlsaltering their hunting to other prey types when small mammalswere scarce. Within fragmented woodlands, perch times were thesame between the sexes, but males had slightly greater interperchdistances than females. Male owls in fragmented woods flew 40%further per hour than males in continuous woodland. For datathroughout the night, food availability and area of trees inhome range accounted for 65% of the variation in perch times.For data before 2200 h, area of trees alone accounted for thesame amount of variation. The data suggest that habitat fragmentation,particularly through reducing woodland area, can greatly influenceowl activity and hunting behavior.     

Correlates of male quality in the field cricket, Gryllus campestris L.: age, size, and symmetry determine pairing success in field populations

Sexual selection in field crickets (Gryllidae) is well documented.Several studies have identified male traits, such as age andbody size, that influence pairing success in the field. HereI show how these traits covary with male quality by examiningthe degree of fluctuating asymmetry in male Gryllus campestris.Older males were both larger and more symmetrical. Principalcomponents analysis suggests that these three variables wereessentially measures of the same trait, male quality. A comparisonof paired and calling males showed that males of high qualitywere more successful in obtaining mates; paired males were older,larger, and more symmetrical. The area of the harp covariedwith morphological traits and determined the carrier frequencyof the male's call. Information related to male quality wastherefore available for female discrimination.     

Black-and-white plumage in male pied flycatchers (Ficedula hypoleuca) reduces the risk of predation from sparrowhawks (Accipiter nisus) during the breeding season

Bright colors in birds might signal that they are undesirableas prey (aposematic), an idea that has been difficult to test.When stuffed pied flycatchers Ficedula hypoleuca are exposedto migrating sparrowhawks Accipiter nisus in spring or in autumn,the hawks attack cryptic females more often than bright males.To achieve better statistical control and to assess whethermale plumage also reduces predation risk in the breeding seasonand in the nesting habitat of the pied flycatcher, I placedpairs of male and female flycatcher mounts in similar positionsnear 22 nests of sparrowhawks. The hawks attacked mainly femalemounts, verifying that the preference is real. The sparrowhawkscaught at least 19 live pied flycatchers; 12 young, 5 adultmales, 1 adult female, and 1 female or young. Hawks that caughtan adult male seemed to prefer attacking female mounts. I discussthree interpretations of these results, suggesting that black-and-whitemale flycatchers may benefit from being a novel and aberrantprey, at least early in the breeding season.     

Predation, sensitivity, and sex: why female black rhinoceroses outlive males

Among sexually dimorphic, polygynous mammals, adult femalestend to outlive males and respond more strongly to predatorsthan males. We asked whether a monomorphic, polygynous speciesvirtually immune to predation due to large size (black rhinoceros,Diceros bicornis) conforms to this pattern. Data on 193 interactionswith lions (Panthera leo) and spotted hyenas (Crocuta crocuta)in two nonhabituated populations in Namibia studied from 1991to 1993 revealed that: (1) females were more vigilant or aggressivethan males to either of the potential predators; and (2) whethersolitary or with calves, females attacked more often than males.Although solitary females tended to be more aggressive to lionsthan to hyenas, neither females with calves or males seemedto discriminate between the two carnivores. We also simulatedthe behavior of human predators (poachers) during 69 encounterswith rhinoceroses. While both sexes abandoned local sites becauseof our presence, females ran farther than males, covering upto 40 km in a day. These findings implicate a behavioral mechanismto explain why secondary sex ratios favor females - males aremore prone to human predation, a prediction consistent withdata from 12 populations throughout Africa. Black rhinocerosesappear to be an unanticipated exception to the well-establishedpattern of male-biased mortality in polygynous mammals; in theabsence of intense human predation (a recent event), male mortalityfails to exceed that of females, suggesting that intrasexualcompetition in a polygynous mammal: may not be the primary causeof unbalanced secondary sex ratios. Our results on the causesof sex differences in mortality and in responsiveness to differentpredators reinforce the relevance of behavioral ecology to conservation;such information is necessary for planning how best to minimizenegative human influences on the few remaining wild Africanrhinos.     

Tail loss reduces home range size and access to females in male lizards, Psammodromus algirus

Numerous lizard species use caudal autotomy as an antipredatordevice even though there must be significant costs during theperiod of tail regeneration. Strategies used by tailless individualsto enhance survival in natural populations are still poorlyunderstood. We experimentally examine tail loss in large, dominantmales of Psammodromus algirus in the middle of the breedingseason in the field. We report data showing home range reductionof large dominant males after autotomy, reduction in the numberof females in the home ranges of manipulated males, and a potentialincrease in mating opportunities of small subordinate maleswith complete tails. We conclude that changes in home rangeuse because of desertion of areas with less cover can resultin decreased predation risk at the cost of decreased accessto females.     

Direct selection on mate choice: female redlip blennies pay more for better mates

Sexual selection may be reduced by costs of mate choice suchas predation risk and energetic expense. Despite their theoreticalimportance, such costs have rarely been quantified or relatedto the quality of mate obtained. We performed such a study witha wild population of redlip blennies in Barbados. In this coralreef fish, females traveled 0.2–12 m from their feedingterritories to the territories of nesting males, where theyspawned and left their eggs in the males' care. Previous studieshave shown that larger and older males are better parents andare preferred by females. Here, we found that long-distancetrips by females were made primarily toward larger males, andthe extra distance traveled beyond the nearest male was correlatedwith the improvement in size of male obtained. Comparisons withfeeding budgets while the females remained on their home territoriessuggested that costs from lost time and energy were minor, andno predation attempts on females were observed. However, onlonger trips the females faced more aggressive attacks fromdamselfishes, whose territories they crossed. In contrast, noattacks were observed from damselfish when the blennies remainedon their own territories. These attacks appeared to be responsiblefor a greater incidence of scars on females than on males. Thisdifference between the sexes became more pronounced as spawningprogressed, and then decreased as females healed during thenonspawning interval. Thus, female blennies endure a cost ofinterspecific harassment which is correlated with the qualityof mate obtained. This cost may restrain sexual selection inthis species.     

Reproductive success of polygynous male corn buntings (Miliaria calandra) as confirmed by DNA fingerprinting

Despite a sex ratio approximating to unity, female corn buntingswere not equally distributed among males. In 1989 and 1990,41.2% of 50 males were monogamously paired, 29.4% were polygynous,and 23.5% were unpaired. Polygynous males usually paired withtwo females, although in 1990 three males were trigamous. Polygynousmales fledged more offspring from their territories than didmonogamous males, mainly because they had more mates. The fledgingsuccess per nesting female was slightly higher in territoriesof polygynous males, but not significantly so. DNA fingerprintingwas used to confirm the true paternity of 44 offspring from15 broods and the true maternity of 50 offspring from 16 broods.A further 12 offspring from three broods for which neither putativeparent was available were also fingerprinted. Actual reproductivesuccess of parents was close to that inferred from observationsof number of young raised. There was only one brood, containingtwo chicks (4.5% of offspring, or in 6.7% of broods), wherethe chicks were not fathered by the male defending the territory.However, this nest was close to the territory boundary, andthe defending male may have been assigned incorrectly. Therewere no cases of intraspecific brood parasitism (n = 16 broods).The copulation rate was low, and extrapair copulation attemptswere rare, probably because of the poor chances of sneakingonto a neighbor's territory undetected and the costs of leavinga territory unguarded.