Trapline foraging by bumble bees: III. Temporal patterns of visitation and foraging success at single plants

We analyzed the temporal structure of visitation by bumble beeworkers to a single Penstemon strictus plant growing in an arrayof conspecifics. When tested against a null distribution usinga randomization model, the observed pattern of arrivals forthe whole group of bees was random, but departures were clusteredin time. Certain individuals visited the plant repeatedly andfrequently throughout the day. These showed significantly regulararrival and departure schedules, which were likely producedby traplining. We explored whether these more frequent and regularforagers gained a higher reward than random or incidental plantvisitors. Using an analytical model, Possingham predicted thata dominating forager that visited a simple, renewing resourcein a regular pattern would garner higher and less variable rewardsthan random visitors. Inspired by these results, but interestedin plant-level visitation, we constructed a simulation modelof resource dynamics for a multiflowered plant with high visitation.The model incorporates the observed visitation schedules ofall bees and independent reward dynamics for each flower onthe plant. We calculated the rewards that observed bees wouldhave collected given a range of resource-renewal parameters.More frequent visitors did not return to the plant when whole-plantresource levels were higher, but these visitors did get greaterrewards. Their increased reward resulted from greater foragingefficiency, primarily through selecting (on average) more rewardingflowers than those selected by less frequent, random visitors.     

Trapline foraging by bumblebees: I. Persistence of flight-path geometry

By setting out arrays of potted plants of Penstemon strictus,I tested whether freely foraging bumblebee (Bombus spp. ) workerswould establish regular foraging routes that reflected the geometryof the array. They did, passing through an asymmetrical arrayin a pattern that minimized interplant flight distances. Afterthe array was changed to a symmetrical pattern, however, theexperienced bees continued to show their previous asymmetricalflight patterns. New bees without experience on the asymmetricalarray showed no asymmetry on the symmetrical array. I term thispersistence of flight-path geometry "trapline holdover, " anddiscuss its implications for the study of animals' learningand foraging behavior.     

Heritable allometric variation in bumble bees: opportunities for colony-level selection of foraging ability

Different characters of an organism may be correlated if genes control the allometric relationship between them. If genetic variation exists for such genes then the allometric relation itself is potentially subject to change by selection. In social insects allometric relations represent colony-level characters. If colonies differ in these relations and this variation leads to differential productivity among colonies, then selection on allometric relations can operate at the level of the colony. We assessed the extent of heritable, between-colony variation for the allometric coefficients relating proboscis ( = glossa) length to wing length for two bumble bee species (Bombus huntii and B. occidentalis). We found that in both species colonies did not differ significantly in slope (b) but did differ significantly in intercept (a) of the regression of glossa length on wing length. Within-colony variation of the intercept was estimated by randomly constituting groups of five workers from each colony and calculating the regression for each group. The intraclass correlation was then calculated from the between- and within-colony mean squares. We found significant intraclass correlations in both species, giving heritabilities of 0.5 ± 0.3 in B. hunti and 0.7 ± 0.3 in B. occidentalis. If this allometric relation affects colony foraging success and foraging environments vary geographically, then the intercept should exhibit corresponding geographic variation. We tested this prediction by comparing intercepts calculated using wild-caught B. vagans workers from Alberta, Ontario and Maine. We found that the intercepts did differ significantly between sites, with the bees from Alberta having a significantly smaller intercept than the bees from eastern North America. Our results illustrate the opportunity for selection on an allometric relation that directly affects the foraging success of individual bumble bee colonies.     

Foraging dynamics of bumble bees: correlates of movements within and between plant species

What rules determine whether bumble bees continue exploitingplants of the species just visited or switch to another species?To tackle this question, we recorded handling times and flighttimes from bees foraging in a natural meadow containing fiveplant species. Inter- and intra-specific plant distances werequantified. The bee-subjective colors of the five species weredetermined; two of these species had similar colors and structures,while three species were distinct from all others. The followingrules were identified: (1) The decision to switch species wascorrelated with previous flower handling time, which we assumeis a function of the reward amount received at the flower. Aftershort handling times, the probability of switching to anotherspecies increased, whereas it decreased after long handlingtimes. This difference became even greater if the bee had hada run of several short or several long handling times. (2) Constantflights (those between flowers of the same species) and transitionflights (those between flowers of different species) followedstereotyped temporal patterns independent of the distances betweenflowers. Constant flights within five plant species consistentlyhad median durations of about 2 seconds, whereas median transitiontimes between species took 3–6 seconds. (3) This temporalrule broke down, however, if the flowers of two species hadsimilar colors, in which case transition flights had equal dynamicsas constant flights. (4) Bees switched more frequently fromrare than from common species but even more frequently betweensimilar species. We conclude that the bees' choices were determinedby a set of rules that guided them to stay with the currentplant species as long as flowers were rewarding and availablewithin close distance but to switch to another species if flowersoffered low rewards or were not encountered at close range     

The combined effect of color and odor on flower choice behavior of bumble bees in flower mimicry systems

Food-deceptive flowers are pollinated by animals that expecta reward but are cheated. Such plants profit from their similarityto rewarding plants and should develop signals that hinderdiscrimination. We use artificial rewarding model flowers andnonrewarding mimicking flowers that present similar visual cues. We test how additional scent cues change flower choiceof the mimic by bumble bees (Bombus terrestris) in two situations:(1) both flower types are simultaneously present and can becompared by the pollinator, and (2) both flower types are encounteredsuccessively in the absence of each other. We find that insituation 1, discrimination learning is greater if scents areused as cues for identifying the mimic, whether the mimic hasa different scent or if it is scentless while the model isscented. In situation 2, a generalization task, a scented mimicis avoided faster than a scentless one. Discrimination of themimic is poorest if it has the same scent as the model, thusdemonstrating a potential for scent mimicry, which has not yet been proved to exist among differently rewarding flowers. Thus,the best strategy for a mimic would be to have the same scentas the model, but this strategy may not be used due to evolutionaryconstraints. Alternatively, if there are several potentialmodels, then having no scent would be a better strategy thanmimicking just one of the models. In situation 1 flower discriminationby color cues is enhanced in the mere presence of scent, comparedto unscented controls, even if the scent does not provide a distinguishable cue itself. The results indicate that the presenceof scent may enhance color discrimination by improving attentiontowards visual cues and/or that combined color/odor cues maylead to better memory formation and retrieval.     

Response of traplining bumble bees to competition experiments: shifts in feeding location and efficiency

Summary In field experiments, we mapped the regular foraging routes (traplines) of marked bumble bees visiting Aralia hispida. When other bees were removed to create localized competitive vacuums, the marked bees shifted their feeding activity toward the removal areas. Bees foraging in these competitive vacuums probed more flowers per inflorescence than control bees. One bee's foraging was studied intensively before and after its local competitors were removed. Compared to four nestmates foraging elsewhere, the focal bee's trip times were shorter (p<0.005) and its food collection rate was marginally higher (P=0.064) during the removal, although all the bees foraged similarly before the removal. These observations indicate that traplining bumble bees opportunistically modify their use of space in response to the activities of other bees in a highly competitive environment.     

The genetic control of seven isozymic loci in Vicia faba L. Identification of lines and estimates of outcrossing rates between plants pollinated by bumble bees

A simplified and non-destructive method using starch gel electrophoresis has been developed on seeds to identify inbred lines of Vicia faba and assess outcrossing rates and gene dispersal in pollination experiments. Six enzyme systems (Alcohol dehydrogenase, Aspartate aminotransferase, Glucose-6-phosphate isomerase, Isocitrate dehydrogenase, Phosphogluconate dehydrogenase and Shikimate dehydrogenase) were analysed from parental lines, crosses performed between lines bearing dissimilar isozyme patterns in pollination cages with Bombus and F₂ progenies obtained from manual selfing of F₁ hybrids. The allozymes at each of the seven studied loci segregated in the expected Mendelian fashion and behaved in a co-dominant manner except for the Adh-2 locus where the only variant was a null allele. No evidence of genetic linkage was observed between at least 13 of the 15 pairs of the studied loci. Percentage of cross fertilisation by Bombus between seven pairs of inbred lines ranged between 1.7% and 28.3%. Pollen transfer between a donor line and a recipient line by two species of Bombus did not lead to differences in outcrossing rates (both about 8%). The new PGD marker with two loci at three alleles each is particularly discriminating and valuable in pollination studies and breeding of V. faba.     

Consistent mixing of near and distant resources in foraging bouts by the solitary mason bee Osmia lignaria

Female bees are usually confronted with a choice among severalflower species that differ in their location and abundance withinthe community, and in the efficiency with which their pollenand nectar can be harvested. We investigated the effects ofdistance and flower density of two flower species on pollencollection by providing nest locations for the mason bee Osmialignaria in natural settings. Distance weakly affected pollenuse; on average, bees nesting near a flower species tended tocollect more of its pollen than did bees nesting at a greaterdistance. Flower density did not predictably impact pollen use,and use did not track changes in density during the season.Bees consistently mixed pollen from more distant species, despitesubstantial added foraging costs, and also mixed when one specieswas an order of magnitude less abundant than the other. Beesrequire nectar as well as pollen to feed their offspring, andour preliminary data suggest that the efficiencies of pollenand nectar collection are inversely related between the twoflower species, which would favor visitation to both species.Bees appear to collect some pollen from the low-pollen, high-nectarplant while visiting it for nectar. Thus, a nectar-collectingconstraint may favor collecting pollen from mixtures of species.     

Niche differentiation in nectar-collecting stingless bees: the influence of morphology, floral choice and interference competition

1. Dozens of social bee species, most of them stingless bees, occur sympatrically in the tropics. The proximate mechanisms through which they partition their resources are, apart from aggressive interactions between Trigona species, not well studied. In the work reported here, niche differentiation at patch level was studied, using two species of Melipona that occur sympatrically in the Central Pacific part of Costa Rica. Foragers of Melipona beecheii are known to collect more concentrated nectar than do Melipona fasciata foragers, even from the same plant species. This observation raises the question of what mechanism leads to such partitioning of nectar resources? To address this question, the roles of bee morphology, floral preferences, and interference competition in partitioning of nectar sources were studied. 2. It was shown experimentally that the feeding rate of both species was highest at 60% nectar concentration. Melipona fasciata preferred 60–70% concentrations to less concentrated solutions, whereas M. beecheii ignored 20% solutions and visited the other solutions equally often. Both species preferred sucrose to glucose and fructose. Melipona beecheii, with a yellowish coloured body, preferred sunny patches, whereas M. fasciata, with a dark brown body, preferred shady patches. Interference competition between the species occurred when they visited the same sugar-water feeder: M. fasciata was dominant over M. beecheii. 3. The nectar foraging strategies of the two species can be summarised as follows: M. beecheii is able to visit sunlit patches, due to its lighter body colour, and is thus able to collect nectar of optimal or near optimal sugar concentration (40–65%). Individuals do not actively choose nectar with a high concentration of sugar, but the average sugar concentration in bee-collected nectar will typically be high in sunlit patches. Melipona fasciata, on the other hand, avoids sunlit patches, but actively chooses the richest nectar in shady patches and tries to dominate such patches. Thus, niche differentiation occurs according to the radiation regime at flower patches, and interference competition will occur rarely under natural conditions. The role of similar mechanisms in other closely related stingless bees is discussed.     

Searching at different spatial scales: the foraging behaviour of the aphid parasitoid Aphidius rosae in rose bushes

The searching patterns of the aphid parasitoid Aphidius rosae were analysed at three different spatial scales: leaves, shoots and bushes. Parasitoid females searched aphid-infested leaves and shoots intensively and remained on average more than twice as long on infested than on uninfested shoots. Patch times and oviposition numbers per shoot were highly variable both between females and for different shoot visits within females. However, at the shoot and bush level low oviposition numbers were generally found. The time spent on different behavioural patterns (searching, resting, feeding, host handling) changed significantly during subsequent shoot visits of individual females but oviposition success was not influenced by this change. Parasitoids searched individual leaves and shoots mainly by walking, while moving between shoots occurred exclusively by flight. The travel time between shoots (i.e. flight time) accounted for less than 1% of the residence time in a bush. At the bush level foraging was characterized by a high ability to localize infested shoots and consequently little time was wasted in searching on uninfested shoots. The pattern of resource exploitation of individual females was consistent with the distribution of A. rosae larvae in field samples taken from individual rose bushes.     

Optimal traits when there are several costs: the interaction of mortality and energy costs in determining foraging behavior

I analyzed the interaction of different types of costs in determiningoptimal behavior using mathematical models. The analysis concentrateson foraging behavior and asks (1) whether the cost factor thathas the greatest effect on fitness generally has the greatesteffect on optimal trait values and (2) whether increasing thesize of one type of cost makes the optimal behavior absolutelyor relatively more sensitive to that cost. The foraging costsconsidered are energy expenditure, predation risk, and othermortality factors. It is shown that increasing the magnitudeof one cost often decreases the relative and absolute sensitivityof the optimal foraging strategy to that cost. The relativefitness effects of different costs generally differ from therelative sensitivities of the optimal strategies to the costfactors. Researchers should therefore measure the shapes ofcost curves rather than their average magnitudes to determinewhich of several costs can be ignored in cost-benefit analyses.     

On the economics of sit-and-wait foraging: site selection and assessment

We present two models of optimal resource exploitation for sit-and-waitforagers. The first model assumes immediate recognition of sitequality and that site quality does not change over time. Thismodel predicts a forager's minimum acceptable site quality.We present a graphical analysis to show how (1) the distributionof site qualities, (2) the travel time between sites, (3) costof search, and (4) expected duration of the foraging processinfluence the minimum acceptable rate. Our second model allowssite qualities to change and relaxes the assumption of immediaterecognition. This model defines conditions of (1) state duration,(2) recognition time, (3) site abundance, and (4) cost of searchwhere the optimal policy is to stay put in a site regardlessof experience. We discuss the implications of these models forthe design and interpretation of field experiments of site useand habitat selection.     

Assessment of scale-dependent foraging behaviour in southern elephant seals incorporating the vertical dimension: a development of the First Passage Time method

1. Identifying the spatial scales at which top marine predators forage is important for understanding oceanic ecosystems. Several methods quantify how individuals concentrate their search effort along a given path. Among these, First-Passage Time (FPT) analysis is particularly useful to identify transitions in movement patterns (e.g. between searching and feeding). This method has mainly been applied to terrestrial animals or flying seabirds that have little or no vertical component to their foraging, so we examined the differences between classic FPT and a modification of this approach using the time spent at the bottom of a dive for characterizing the foraging activity of a diving predator: the southern elephant seal. 2. Satellite relayed data loggers were deployed on 20 individuals during three successive summers at the Kerguelen Islands, providing a total of 72 978 dives from eight juvenile males and nine adult females. 3. Spatial scales identified using the time spent at the bottom of a dive ( = 68.2 +/- 42.1 km) were smaller than those obtained by the classic FPT analysis ( = 104.7 +/- 67.3 km). Moreover, foraging areas identified using the new approach clearly overlapped areas where individuals increased their body condition, indicating that it accurately reflected the foraging activity of the seals. 4. These results suggest that incorporating the vertical dimension into FPT provides a different result to the surface path alone. Close to the Antarctic continent, within the pack-ice, sinuosity of the path could be explained by a high sea-ice concentration (restricting elephant seal movements), and was not necessarily related to foraging activity. 5. Our approach distinguished between actual foraging activity and changes in behaviour induced by the physical environment like sea ice, and could be applied to other diving predators. Inclusion of diving parameters appears to be essential to identify the spatial scale of foraging areas of diving animals.     

Plant structure and the foraging success of Aphidius rhopalosiphi (Hymenoptera: Aphidiidae)

Abstract. 1. A study of searching behaviour on wheat and host preference of Aphidius rhopalosiphi (DeStefani-Perez) is described.
2. Parasitoids divided their time equally between the leaves but spent very little time on the ear.
3. After contact with honeydew or an aphid host, parasitoids were arrested in a particular area and increased the time spent searching.
4. Aphids feeding on the ear were parasitized less successfully, since their position between the grains protected them from parasitoid attacks.
5. A.rhopalosiphi exhibited no preference for Metopolophium dirhodum (Walker) or Sitobion avenae (Fabricius), although the handling time for the latter was significantly longer; this resulted in fewer S.avenae being parasitized when it was abundant. Parasitoids did not switch between hosts in these experiments.
6. As a result of its searching behaviour, A.rhopalosiphi will encounter and parasitize M.dirhodum feeding on the leaves more frequently than S.avenae, which feeds on the ear. This will limit the parasitoid's ability to regulate populations of the cereal aphid S. avenae.     

Sexual cannibalism and sperm competition in the golden orb-web spider Nephila plumipes (Araneoidea): female and male perspectives

Mating systems are frequently shaped by conflicts over reproductive interests between males and females. Sexual cannibalism canbe an especially dramatic manifestation of such conflicts.However, the resolutions of this conflict differ among sexuallycannibalistic spider species. Cannibalism may be in the interestof both sexes when females consume males as a foraging decisionto improve fecundity and/or males sacrifice their bodies toincrease fertilization success. In other species, females exertsequential choice of partner by selectively terminating copulationthrough cannibalism while males fail to obtain a paternityadvantage. Here, we investigate the adaptive value of cannibalismin the orb-web spider Nephila plumipes where 60% of males donot survive copulation. Virgin females in poor condition aremore frequently cannibalistic and more likely to kill largemales, but the frequency of cannibalism among mated femalesis not influenced by these factors. Instead, males that matewith mated females increase their fertilization success bybeing cannibalized. Cannibalized males generally mate for longer,but longer copulations correspond with increased paternity onlyin mated females. The amount of sperm from particular malesthat a female stored was not influenced by any of the measuredvariables. The number of sperm stored was not related to paternity,nor was there any detectable reduction in sperm number afterfemales had reproduced. Our data suggest that the conflict between the sexes differs between virgin and mated females.Females should always cannibalize a male, but males only gainfrom cannibalism when mating with mated females, not when matingwith virgin females. Interestingly, the frequencies of cannibalismare not different in matings with virgin or mated females.     

Benefits of recruitment in honey bees: effects of ecology and colony size in an individual-based model

Why do some social insects have sophisticated recruitment systems,while other species do not communicate about food source locationsat all? To answer this question, it is necessary to identifythe social or ecological factors that make recruitment adaptiveand thus likely to evolve. We developed an individual-basedmodel of honey bee foraging to quantify the benefits of recruitmentunder different spatial distributions of nondepleting resourcepatches and with different colony sizes. Benefits of recruitmentwere strongly dependent on resource patch quality, density,and variability. Communication was especially beneficial ifpatches were poor, few, and variable. A sensitivity analysisof the model showed that under conditions of high resource densityrecruitment could even become detrimental, especially if foragingduration was short, tendency to scout was high, or recruitsneeded a long time to find communicated locations. Colony size,a factor often suspected to influence recruitment evolution,had no significant effect. These results may explain the recentexperimental findings that in honey bees, benefits of waggledance recruitment seem to vary seasonally and with habitat.They may also explain why some, but not other, species of socialbees have evolved a strategy to communicate food locations tonest mates.     

A life in the fast lane: energetics and foraging strategies of the great cormorant

Body insulation is critically important for diving marine endotherms. However,cormorants have a wettable plumage, which leads to poor insulation. Despitethis, these birds are apparently highly successful predatorsin most aquatic ecosystems. We studied the theoretical influenceof water temperature, dive depth, foraging techniques, and preyavailability on the energetic costs of diving, prey search time,daily food intake, and survival in foraging, nonbreeding greatcormorants (Phalacrocorax carbo). Our model was based on fieldmeasurements and on data taken from the literature. Water temperatureand dive depth influenced diving costs drastically, with predicted increasesof up to 250% and 258% in males and females, respectively. Changes inwater temperature and depth conditions may lead to an increaseof daily food intake of 500-800 g in males and 440-780 g infemales. However, the model predicts that cormorant foragingparameters are most strongly influenced by prey availability,so that even limited reduction in prey density makes birds unableto balance energy needs and may thus limit their influence onprey stocks. We discuss the ramifications of these results withregard to foraging strategies, dispersal, population dynamics,and intraspecific competition in this avian predator and pointout the importance of this model species for our understandingof foraging energetics in diving endotherms.