Abstract

  There are many different aspects involved in the behavior of foraging within the insect order of Hymenoptera. Methods of  locating food and of communicating resource locations to hive and nest-mates relate to distance, direction, and the vertical measurement of the height of the food source. Bees describe the distance to a food source by a measurement of  energy expended on the trip. Direction is described by angles relative to the position of the sun. This information is incorporated into an elaborate dance routine and passed throughout the colony.

  Foraging  behaviors in other groups within Order Hymenoptera are often just as intricate. Harvester ant colonies use fan shaped dispersal techniques to cover large areas of foraging range. These areas are used alternately from day to day at the command of  a group of workers called patrollers.

  Colony fitness can greatly effect the foraging behaviors of workers. If the colony’s well-being is in a state of decline because of a lack of resource, foragers will take on bigger work loads, forage for longer periods of time, and enlist fellow workers to aid in acquiring whatever is missing.

 Introduction

  This review article is a compilation of examples of many aspects of Hymenopteran life that each contribute to the foraging behaviors of  bees, wasps and ants. Components such as food availability, colony size, colony health, host availability, and foraging area are all important factors that influence decisions such as when to forage, how much food to find, and what to look for.  The theories involved in communication such as path integration, use of landmarks, and fluorescent visual cues all contribute to the foraging aspects of where to look and sometimes where not to look for food. All of these situations are found within this review article.

 Transfer of Food Source Information Between Hive-Mates

  In many Hymenopteran societies, in this case the dancing honey bee, (Hymenoptera: Apis mellifera), foraging worker bees known as scouts leave the hive to locate adequate food sources, either pollen or nectar. Upon finding these sources, the scout bees return to their hive to enlist  other worker bees, known as recruits, to go to the described site and gather food. The scout bee will convey information about the site that they have found using elaborate methods of dancing and wing buzzing. The dances contain information about distance in relation to the amount of energy expended on the trip, direction in relation to angles relative to the position of the sun, and of the height of the food source. The exact methods used by the bees to remember the locations of food that they have found is sometimes questioned.
 In 1986 J.L. Gould  asked if bees used cognitive maps to locate described food sources rather than utilizing path integration methods. This viewpoint led to an experiment performed by Wolfgang H. Kirchned and Ulrich Braun to reaffirm the previously accepted idea that bees use path integration as their primary technique of locating and remembering a food source.

  Kirchner and Braun used a new experimental approach to test the cognitive map hypothesis versus the path integration concept. They attached small magnets to the dorsal side of the thorax of foraging bees. This enabled the insects to be “temporarily tethered” while in a wind tunnel.  The bees were trained to feed at several different feeding stations within the laboratory. Their theory was that if the scout bees used cognitive maps of the area around their hive to remember where the food was when relating information, the  fact that they were tethered to create the illusion that they had flown a longer distance should make no difference in their dance, as if they had flown to the feeders without interference. If the bees used path integration methods, the effects of being tethered in the wind tunnel should be described to the other bees through the dances as being a increase in distance to the food station.

  The experimental results unequivocally displayed that cognitive maps are not the methods used by Apis Mellifera to remember or to communicate the locations of food sources. Even the presence of known landmarks were not taken into account by the bees as they related travel information to the rest of the hive. This is an issue that will continue to cause disagreement between Entomologists who are interested in this topic.

 Landmark Influences on Distance Estimation to Food Sources

  Although it is widely accepted that honey bees use path integration methods to remember and to describe the locations of food sources, the aspect of landmark use may have authority as well. Lars Chittka, Karl Geiger and Jan Kunze performed an experiment to determine the strength of landmark use in addition to the use of  the sun compass and measured energy investment. Their theory stated that it was unlikely that the mechanisms of path integration alone would “reliably guide an insect during long distance orientation in natural conditions.”   Chittka, Geiger, and Kunze felt that disruption in path integration methods such as “wind or thick cloud cover” as well as mistakes simply made by the bee itself would have to be compensated for by use of other navigational methods, namely landmarks.

  Chittka, Geiger, and Kunze created an experiment in which the honey bee Apis mellifera  was the subject in a series of tests which differed in food source location relative to a trained site, each with varying degrees of landmark occurrence.
 The results of these trials seemed to indicate that bees use both path integration and landmarks as a method of finding food sources. Apparently this topic is still open for discussion.

 Foraging Behaviors in Parasitoid Hymenopterans

  A study was done by Alex R. Kraaijeveld and Jacques J. M. van Alphen in 1994 on the attack rates by parasitoid Hymenopterans upon Drosophila melanogaster larvae at different stages of larval development. Three different methods of foraging by the Hymenopterans were described in relation to this experiment. One method was called “vibrotaxis”, described as when the parasitoid detected movement of the Drosophila larvae. The second method was labeled “ovipositor searching” and was explained as when the wasp randomly probed at the ground using her ovipositor. The third method was termed “antennal searching”. This was described as when the parasitoid “uses the tips of her antennae to locate the larvae.”

  Although other species of wasp were mentioned in the paper, three select parasitoid species were used in the experiment, each of whom used a different foraging technique. These three species were selected because they are the natural predators of Drosophila larvae in Silwood Park, England, where the experiment took place. They were named as Asobara tabida, (vibrotaxis), Leptopilina heterotoma (ovipositor searching) and Tanycarpa punctata (antennal searching). Each wasp had varying degrees of attempts and successes of oviposition relating to stages of the larval development and varying degrees of  larval movement. The Drosophila larvae were separated in terms of how they themselves foraged for food. “Sitter” was the term used to describe larvae that stayed in one general place, and “rovers” were larvae that moved around much more relative to the sitters.

  The attack rates corresponded as they had been expected to between the parasitoid foraging tactics and the mobility of the larvae. Results from the experiment varied according to wasp success, however, because of defensive methods utilised by the larvae, such as the ability to encapsulate the parasitoid eggs.

  A field study done by Edward F. Connor and Michele J. Cargain in 1994 posed new questions about some aspects of foraging behaviors in parasitoid Hymenopterans. In this case, a parasitoid wasp named Closterocerus tricinctus was observed for her foraging behavior on a leaf-mining moth called Cameraria hamadryadella. Connor and Cargain sought to answer the following questions: “Do parasites forage in density-dependent, density-independent and inversely density-dependent ways, or is the range of parasite foraging behavior more limited?”  A field study was performed during which the behaviors of C. tricinctus were observed as the wasps traveled from leaf to leaf in search of suitable hosts; the larvae of C. hamadryadella. Host larvae are found on upper leaf surfaces in structures called leaf-mines.

  The study took into account the number of leaf-mines found on each leaf in relation to the foraging behaviors enlisted by C. tricinctus. Behaviors taken into consideration included: the frequency of landings by C. tricinctus on multiply-mined leaves compared to less densely mined leaves, the time spent on each leaf by the wasps, the handling time spent and the search time spent. Also measured were the rates of successful landings on leaf-mines, attempted parasitism, and success rates of the wasp.

  The results of this field study suggested that the selection of leaves that the wasps foraged upon appeared to be density-dependent.  The proportion of leaf-mines visited per leaf was found to be inversely density-dependent, thus making the “overall spatial pattern of visitation by C. tricinctus to mines of C. hamadryadella inversely density-dependent.”

 Effects On Foraging in Relation to the State of the Colony

  In social Hymenopterans, the health and success of the colony is of up most importance. High levels of reproductive performance and survival rates are main goals of a social Hymenopteran society. It is presumed that the “behavioral activities of individual workers should be integrated with colony state.”   Some studies done in the past show that hierarchy within the worker social structure plays a role in behavioral changes.  Some questions have been raised about the changes in foraging behavior due to the fitness of the colony. If the “nutritional status of the colony is low, should individual workers work harder at the task of foraging, or should more workers be enlisted, or both? What type of foraging plan should workers assume should the colony begin to decline?”

  An analysis done in 1993 by Mark L. Winston and Ron C. Ydenberg takes these questions into account. Variations in colony size caused behavioral changes in Pheidole; when a group of workers were removed from the colony, the remaining workers took on increased activity levels in order to compensate.

  Another study done in relation to foraging behavior and colony status was accomplished by  R. C. Plowright, James D. Thomson, L. P. Lefkovitch and C. M. S. Plowright in 1993.  The subject of this experiment was a North American bumble bee, Bombus terricola. The main issue was the relationship between larval biomass of the colony and the effect it had on pollen and nectar foraging. Pollen is collected for the purpose of feeding larvae, thus you would assume that the larger the larval population, the more demand there would be for pollen , therefore foragers would proceed to seek out and gather larger quantities of pollen. Plowright et al challenge this notion, saying that this notion is “only correlational, (and) cannot by itself be taken as evidence that pollen collection is regulated by demand for it from within the colony.” They went on to state that “it might be equally argued that since the demands for pollen and for honey generally march hand in hand, the relationship between colony pollen intake and larval biomass in Pendrel’s  data is merely a by-product of an even stronger relationship between larval biomass and the intenity [sic] of nectar foraging.”

  To answer this uncertainty, Plowright et al set up an experiment using two colonies of Bombus terricola, both produced at the same time.  Eleven days later, the number of adult worker bees was counted, and each bee marked with a numbered tag. At this point, each hive was alternately robbed of either it’s pollen or honey reserves in a manner such as that “colony A was deprived of pollen on 4, 6, and 8 July and of honey on 3, 5, and 7 July, whereas colony B was deprived of pollen on 3, 5, 7 July and of honey on 4, 6, and 8 July.”

  When one food source was taken away, the experimenters placed a surplus of the other in its proper place in the hive. For example, when honey was taken away, a surplus of pollen was placed in the pollen cylinders. Plowright et al kept track of which worker bees were leaving and returning, at what times this was happening, what they had brought back and how much. This experiment lasted for six days. After analyzing the data, Plowright et al found that overall, the worker bees brought back more pollen on the days when their pollen stores were taken than they did when they were deprived of nectar, as expected. It was pointed out that the bees did not take more foraging trips during times of deprivation, they simply increased the amount of pollen taken back per foraging trip.

 Foraging Ranges and Territoriality

  An experiment developed by Deborah M. Gordon raised several questions regarding the factors contributing to the size, location, and movement of foraging ranges, as well as the degrees of territorial behavior in the harvester ant, Pogonomyrmex barbatus. A correlation between these aspects and the age of the colony were the main focuses of the experiment.

  Pogonomyrmex barbatus colonies can live for fifteen to twenty years, reaching an average stable size of 12,000 workers when it is about five years old. After a colony reaches two years of age, the foraging range tends to stay at that relative size for the rest of the colony’s life.  The size of the foraging range increases most rapidly during the colony’s first and second year. This information led Gordon to conclude that  the “increase of foraging area appears to be related to colony growth rather than overall colony size.”

  Territoriality also appears to be linked to the age of the colony. At about five years of age, the colony begins to produce reproductive offspring. Younger colonies, especially those between their third and fourth year have been noted to be more defensive of their foraging range than the ants of older colonies. This leads to the conclusion that territorial behavior regarding foraging area depends upon colony age, thus where the attention of the colony is being focused; either on reproduction or on conflicts with their neighboring colonies over foraging space .

  The size of a colony’s foraging range can vary throughout the life of the colony. This fluctuation may result from varying degrees of resource abundance within the area, tying in with the costs and benefits of defense.

  The foraging behaviors of the harvester ant include a widespread fan-like dispersal of workers who move slowly as they search for seeds. The foragers occupy different areas from one day to the next.  “A mature colony may have up to eight habitual foraging directions, of which it uses about three to five a day.”  Each morning, before the foragers become active, a group of workers known as patrollers evaluate the foraging range and decide which foraging area will be used that day.

  The Effects of Food Source Familiarity in Scouts and Recruits

  The prior experience of a leaf cutter ant (Atta colombica) with certain plant resources likely effects the acceptance or rejection of the food source by the ant. A study was organized in Costa Rica by J.J. Howard, M. L. Henneman, G. Cronin, J.A. Fox and G. Hormiga to shed some light on the subject. The intensity of recruitment behavior by scouts of this species depends on the quantity and quality of the food source as well as any prior experience that the scout has had with the plant in question. The scouts were found to show recruitment behavior faster when they had encountered a familiar plant source remembered to be palatable than they did when encountering unfamiliar plants.
 The experiment was set up using three food sources. Three sets of ant colonies were observed. One set was conditioned to one type of flower, Aphelandra golfodulcensis, made familiar to the ants because it had been placed within their foraging territory. The second set of colonies was familiarized with a second flower type, Caryocar costaricense, and the third set was familiarized  primarily with stipules of Ficus. Both the first set and the second set of ant colonies had access to Ficus within their dominions as well as their respective flowers, but the third set had access to neither floral resource.

  After adequate time had passed for the ants to familiarize themselves with their given food source, Howard et al proceeded to laden trails with each of the three food sources in varying mixtures or by themselves. Careful observations of scout and recruit behavior were made as the ants found or were led to familiar and/or unfamiliar food sources. Atta colombica scouts from all three sets of colonies recruited other workers within 45 minutes after encountering Ficus stipule patches. None of the ants familiarized with Caryocar recruited other workers to Aphelandra. However, one scout from a colony familiarized with Aphelandra cut a piece of Caryocar and recruited other ants to the site in record time.  After several exposures and a day to think about it, other scouts accustomed to Aphelandra eventually accepted Caryocar as a suitable food source.

  Individual ants are capable of making their own choices, as shown above where the Aphelandra scout  took such an unexpected liking to Caryocar.  Recruits can and do make independent decisions which can vary from their scout’s preferred choice while foraging. Independent decision making can  “reveal much about the integration of individual behavior patterns into complex and efficient foraging systems.”

 Types of Food Sources and Their Effects on Foraging Behaviors

  James R. Hagler and Stephen L. Buchmann tested the role of phenolic compounds and the effects that they have on the foraging behaviors of the honey bee, Apis mellifera. Phenolics are found in many nectars and are thought to have several possible effects. One possible function may be that their fluorescent properties are used as a visual cue to inform insects of phenolic presence. Another theory that phenolics may cause carbohydrates and amino acids to be rendered less or non-metabolizable, therefore deterring non-pollinating insects. Perhaps they impart an unfavorable taste to nectar for non-pollinators and the right taste for adapted pollinators.”

  In the experiment, several phenolic-rich solutions were offered to worker honey bees for foraging. The amounts of liquid taken from each container was analyzed and related to the type of solution that the container had consisted of. The results of this experiment  showed that A. mellifera avoided phenolic-rich solutions and opted for the 30% sucrose control, with one interesting exception- the honey bees seemed to enjoy a very low concentration of salt cedar (a phenolic containing substance).
 

 References

Kirchner, W. H. & Braun, U. 1994. Dancing honey bees indicate the location of food sources using path integration rather than cognitive maps. Animal Behaviour, 48, 1437-1441.

Dukas, R. 1995. Transfer and interference in bumblebee learning. Animal Behaviour, 49, 1481-1490.

Chittka, L., Geiger, K. & Kunze, J. 1995. The influences of landmarks on distance estimation of honey bees. Animal Behaviour, 50, 23-31.

Villalobos, E. M. & Shelly, T. E. 1996. Temporal and spatial variation in the foraging behavior of honey bees (Hymenoptera: Apidae) at Chinese violets. Florida Entomologist, 79 (3), September. 398-407.

Kraaijeveld, A. R. & van Alphen, J. M. 1995. Foraging behavior and encapsulation ability of Drosophila melanogaster larvae: correlated polymorphisms? (Diptera: Drosophilidae). Journal of Insect Behavior, 8 (3), 305-313.

Schmid-Hempel, P., Winston, M. L. & Ydenberg, R. C. 1993. Invitation paper (C. P. Alexander Fund): Foraging of individual workers in relation to colony state in the social Hymenoptera. The Canadian Entomologist, 125, 129-160.

Plowright, R. C., Thomson, J. D., Lefkovitch, L. P. & Plowright, C. M. S. 1993. An experimental study of the effect of colony resource level manipulation on foraging for pollen by worker bumble bees (Hymenoptera: Apidae). Canadian Journal of Zoology, 71,1393-1396.

Hagler, J. R. & Buchmann, S. L. 1993. Honey bee (Hymenoptera: Apidae) foraging responses to phenolic-rich nectars. Journal of the Kansas Entomological Society, 66 (2), 223-230.

Gordon, D. M. 1995. The development of an ant colony’s foraging range. Animal Behaviour, 49,649-659.

Howard, J. J., Henneman, M. L., Cronin, G., Fox, J. A. & Hormiga, G. 1996. Conditioning of scouts and recruits during foraging by a leaf-cutting ant, Atta colombica. Animal Behaviour, 52, 299-306.

Connor, E. F. & Cargain, M. J. 1994. Density-related foraging behaviour in Closterocerus tricinctus, a parasitoid of the leaf-mining moth, Cameraria hamadryadella. Ecological Entomology, 19, 327-334.