1. Worker reproduction is blocked by worker-worker policing. Worker bees have the potential to lay eggs that will produce males, but less than 0.1% of the males produced in a colony derive from worker-laid eggs. Workers are, therefore, effectively sterile and this promotes cooperation by creating a setting in which each worker can best promote her genes by helping her mother (the queen) reproduce. What prevents the workers from successfully rearing sons is a process of mutual policing: the workers in a queenright colony will attack a nestmate worker whose ovaries become active, and will recognize and eat any eggs laid by workers. This worker-worker policing is favored in honey bee colonies because the queen mates with multiple males, hence each worker is surrounded mainly by half sisters and each worker is more closely related to the male offspring of the queen (her brothers, r = 0.25) than those of her half sisters (her half-nephews, r = 0.125).
2. Queen-worker signalling is mutualistic, not manipulative. For many years it was thought that what keeps the workers in a honey bee colony from laying eggs is a chemical signal from the queen, the so-called "queen substance" pheromone. We now understand that what ultimately prevents worker reproduction is worker-worker policing, and that the function of the queen's pheromone is to inform the workers of her presence. Then if the queen should die the workers can quickly sense her demise and begin rearing a replacement queen from the eggs or young larvae remaining in the nest.
3. A colony effectively monitors its environment for rich food sources. A honey bee colony functions as a large, diffuse, amoeboid entity which can extend itself over great distances and in multiple directions simultaneously to tap the flower patches in the surrounding environment. More specifically, we now know that the range over which a colony searches for flower patches extends 6 or more kilometers from the hive, and that the probability of a colony detecting any given flower patch within 2 km of the hive is greater than 0.5. Exactly how this impressive search ability arises from the activities of scout bees remains unknown.
4. A colony's foragers distribute themselves among nectar sources in a way that optimizes the colony's energy collection. The process by which this arises does not involve any bee having a synoptic view of a colony's foraging operation. Instead, each forager operates with surprisingly limited information. Each employed forager advertises her work site with waggle dances in accordance with the desirability of her site and each unemployed forager follows one of these dances to find a suitable work site. Employed foragers visiting highly desirable forage sites persist at these sites and produce vigorous dances which arouse unemployed foragers to join them, whereas employed foragers visiting less desirable sites will tend to refrain from announcing them to the unemployed foragers and may even cease working at these sites. In this way a huge mass of information which is dispersed among the employed foragers is neatly exploited without the need to convey it to a central authority. Moreover, the resulting distribution of forages among nectar sources is very close to the "ideal free distribution," which is optimal and what would be produced if each bee had complete information about the foraging opportunities.
5. A colony adjusts its selectivity among nectar sources in relation to forage abundance. Honey bee colonies will exploit lower profitability nectar sources during times of scarcity than during times of plenty. Thus they continue to acquire food when forage is scarce, but they utilize low-yield sources only when necessary. The mechanism by which the selectivity is adjusted is that the foragers adjust the threshold level of food-source profitability for producing waggle dances in relation to how long they must search to find a receiver bee when they return to the hive with a load of nectar. When forage is sparse and the colony's nectar influx is low, returning foragers find receiver bees quickly and have a low dance threshold, hence food sources with low profitability are exploited. When forage is abundant, the search time and thus the dance threshold rises, hence only highly profitable food sources are exploited.
6. A colony adjusts the percentage of its workers engaged in nectar collection in relation to forage abundance. The availability of nectar goes through booms and busts, and a colony responds adaptively by adjusting the number of bees working outside the hive. The percentage of a colony's worker force that is engaged in foraging varies between 10 and 40 percent. One of the mechanisms by which foragers are activated when nectar becomes plentiful following a dearth is the production of shaking signals by the first bees to discover that nectar is once again plentiful. On their first trips back with food, they run around inside the hive shaking other workers in a stereotyped manner, and this stimulates the inactive foragers to move toward the nest entrance. Here they will encounter waggle dances that steer them to profitable flower patches.
7. A colony adjusts the percentage of its workers engaged in nectar reception in relation to forage abundance. When a colony boosts its nectar collecting rate colony (as described above) it also needs to increase its nectar processing rate to prevent a bottleneck in the overall process of nectar acquisition. The principal mechanism by which additional receiver bees are activated is the production of tremble dances by returning foragers when they experience long search times to find receiver bees. These striking dances are performed throughout the nest and activate primarily unemployed, middle-aged bees to the task of nectar reception. Tremble dances and waggle dances therefore play complementary roles in keeping a colony's rates of nectar collecting and processing well matched. Both dances are produced by returning foragers, by whereas waggle dances are produced when the search time is low and raise the collecting rate, tremble dances are produced when the search time is high and raise the processing rate.
8. A colony controls tightly its investment in comb construction. Beeswax comb is energetically costly and a colony builds comb only when the need for additional comb is certain. This is when the colony is experiencing a high nectar intake and has nearly full honey combs. The mechanism by which the bees responsible for comb building sense that it is time to start building more comb remains unknown.
9. The control of the type of comb constructed-worker comb or drone comb- is decentralized. A colony precisely regulates the percentage of drone comb (built of large cells for rearing drones) in the nest at about 15%. How is this accomplished? Although the queen of a colony would seem to be well suited to sensing her colony's need for additional drone comb, because she walks throughout the nest and constantly measures the cells she encounters, it is clear that she plays no role in this control process. Rather this decision is made collectively by the workers.
10. The collection of pollen is controlled by negative feedback from the nurse bees to the foragers. A colony regulates its pollen collection to maintain a small reserve supply of pollen, about 1 kg, in the nest. The mechanism by which this is accomplished involves the nurse bees feeding the foragers proteinaceous food when the pollen reserve is large, but withholding this when the pollen reserve is small. When the foragers are not given this proteinaceous food, it is likely that they begin to feel hungry for protein and this stimulates them to gather pollen rather than nectar or water.
11. The collection of water is controlled by negative feedback from the receiver bees to the foragers. A colony controls its water collection in relation to its need for water to cool the hive and feed the brood. Although we do not know yet how foragers decide when to start collecting water, we have shown how they decide whether to continue or to stop collecting water based on the ease of unloading to receiver bees inside the nest. The greater the colony's need for water, the quicker a water collector is able to start unloading, the sooner she can end her unloading, the fewer encounters she has with bees refusing her water load, and the more bees she has unloading her simultaneously.
12. A colony chooses its future home site carefully and well. When a colony needs to find a new nest site, in the course of swarming, typically it will evaluate a dozen or more potential sites, each of which is judged with respect to at least 6 variables (cavity volume; entrance size, height, direction, and proximity to the cavity floor; and presence of combs in the cavity). Also, the selection process generally lasts for several days with the result that the colony does not choose the first acceptable site that it finds (satisficing strategy), but reliably chooses the best site that it finds (best-of-N strategy).
13. The decision-making process during nest-site selection utilizes a curious means of consensus building. A colony takes up residence in a new home site only after its scout bees have reached complete agreement about the chosen site. How do the scouts achieve a consensus? Curiously, the consensus-building process does not rely upon scout bees switching allegiances among sites, rather it relies upon a rapid turnover among the bees functioning as scouts. This rapid turnover, coupled with strong recruitment only to sites high in quality, yields an agreement among the scouts for one of the high-quality sites. Any mediocre site quickly loses its supporters, and sooner or later one of the superior sites achieves an insurmountable lead in the "competition" for scout bees. The beauty of this means of decision-making is that it produces a reliable selection of one of the high-quality sites but does not require any of the participants to have broad knowledge of the alternatives under consideration.
14. Queen honey bees may mate multiply as a means of coping with parasites. The curious promiscuity of queen bees, which mate with and utilize the sperm of a dozen or more males, has long been a mystery. With my colleagues Paul Sherman and Kern Reeve, I have suggested that this polyandry is advantageous for it endows a queen with a great deal of genetic variability among her workers, and this could prove vital in resisting one of the many parasite-based diseases to which honey bee colonies are susceptible. This hypothesis remains to be tested.
15. Predation has had a pervasive influence on the functional organization of honey bee colonies. Exploring E.O. Wilson's metaphor of a social insect colony as a factory-fortress, I compared the defense strategies of colonies of three species of honey bee that live in southern Asia. This fieldwork revealed that the colonial defense strategy of each species comprises a broad constellation of traits, including colony size and mobility; nest site and size; and worker size, metabolic rate, and fighting behavior. This comparative study demonstrates the value of viewing colonies of honey bees as functionally integrated units.