Honey bee workers show a strong tendency to visit flowers of only one type during a foraging trip.

As first described by Aristotle, honey bee workers show a strong tendency to visit flowers of only one type during a foraging trip. It is known that workers rapidly learn a flower colour when rewarded with artificial nectar (sucrose solution). However, some previous studies report that the degree of constancy after training is unaffected by reward quantity and quality when bees are tested in an array of artificial flowers of two easily distinguished colours, such as blue and yellow. One possible reason for this surprising result is that large reward volumes were compared. This is likely to mask the abilities of foragers to make adaptive decisions under more realistic conditions.

To test this possibility, Grüter et al. (2011) offered untrained honey bee workers ecologically relevant rewards (0.5, 1 or 2 ul of 0.5 or 1 mol 1-1 sucrose solution) on one or two consecutive yellow or blue artificial flowers and then recorded which flowers the bees subsequently landed on in an array of 40 empty flowers. The results showed that an increase in all three factors (volume, concentration and number of rewards) significantly increased constancy (proportion of visits to flowers of the trained colour) and persistence (number of flowers visited) during the foraging bout. Constancy for the least rewarding situation was 75.9% compared with 98.6% for the most rewarding situation.

These results clearly show that honey bee workers do become more constant to blue or yellow with increasing nectar rewards, provided that the rewards used are ecologically realistic. As the most rewarding conditions led to nearly 100% constancy, further reward increases during training would not have been able to further increase constancy. This explains why previous studies comparing large rewards found no effect of reward on constancy.

Gegear and Laverty (2004) assessed the flower constancy of Italian honey bees by presenting individual foragers with a mixed array of equally rewarding yellow and blue flowers after they were trained to visit each colour in succession. All honey bees showed a high degree of flower constancy to one colour and rarely visited the alternate colour.

The flower constancy of honey bees on successive days was studied by removing and identifying pollen collected by marked pollen-gatherers. Removing pollen from the bees decreased their tendency to collect it later; chilling the bees before marking did not influence their foraging behaviour; neither treatment affected their constancy to the kind of pollen collected. No sequence of nectar or pollen collection with age of bee was found. In each experiment most of the bees collected only a few of the pollens available to them. Bees collecting the most common pollens tended to be the most constant. In general, the proportion of bees collecting their original pollen decreased as the number of foraging days increased, and only about half were doing so after one week; the rate of decrease differed in different experiments. No bee regularly collected different pollens at different times of the day. When the pollen they were accustomed to collect was unavailable for a day, most foraged for nectar only or remained at home.

Most bees that changed to another pollen probably did so when the pollen they had previously collected was scarce or unattractive for longer periods. About 6% of the loads each contained more than one species of pollen. Bees that collected mixed loads were more inclined to do so later; probably they were dissatisfied with the crops they were working and were sampling others. When a colony was moved to another site with similar flora, the bees tended to visit the same species as before, but when one species predominated, bees that had not visited it before tended to do so (Free 1963).

Pernal and Currie (2002) evaluated the influence of pollen-based cues on the foraging decisions made by honey bees using a series of two-choice bioassays, performed within a highly controlled indoor environment. They examined behaviours related to the choice and collection of pollen by foragers among six floral species and three artificial substrates (pollen analogues). First, they evaluated the responses of honey bees to the odours produced by different pollens (or pollen analogues) and pollen lipid extracts. Honey bees displayed similar levels of preference to the odours produced by all pollen species over those of pollen analogues, with a similar pattern of response shown to their extracts. They then evaluated behaviours of foragers in response to variation in particle size, using soybean meal that was ground and sifted to create a hierarchy of particle size classes. Bees preferred particle sizes below 150 um, but the greatest response was shown for those particle sizes below 45 um. They also assayed the effect of varying protein content on the foraging decisions made by bees by mixing soy flour with different proportions of cellulose powder.

Foragers, however, were incapable of discriminating protein content. They determined changes in the response of foragers to different levels of handling time using different sized screens through which bees were forced to crawl to reach an attractive pollen odour source. In these tests, pollen-seeking behaviours were seen to decrease with increases in handling time. When odour was presented simultaneously with other stimuli, it was the primary and overriding cue used by bees to select pollen.

These results suggest that individual honey bee foragers do not discriminate among pollen sources based on intrinsic differences in quality, but instead evaluate cues that may increase their efficiency of collection and recruitment to such a food resource.

Three distinct predictive models of foraging ecology have been developed. Optimal diet theory predicts the behaviour which should maximize joules gained per joule expended in searching for food. A joule is a unit of work or energy equal to the work done by a force of one newton acting through a distance of one meter. This entails obtaining the greatest reward while traveling the shorter distance. Minimal uncertainty theory describes the foraging pattern which maximizes the probability of obtaining a reward, but the average reward may not be maximized. Individual constancy theory indicates that foragers restrict visits largely to a single floral type (Wells and Wells 1983).

Experiments using honey bees and artificial flower patches were designed to test three alternative foraging ecology models: optimal diet, minimal uncertainty, and individual constancy. Honey bee responses to a mixed colour flower patch and to flower morph associated differences in reward quantity, quality and frequency were measured (Wells and 

Wells 1983). Each honey bee visiting a patch of randomly distributed blue and yellow flowers was constant to one colour, even though that behaviour was suboptimal. When reward quantity was unequal between the two flower morphs each bee was constant to one colour even though that behaviour often resulted in suboptimal reward. When reward frequency was higher in one flower morph than in the other each bee was constant to one colour, even though that behaviour often failed to maximize reward or minimize uncertainty.

Although each of their experiments had the potential to refute the individual constancy model of honey bee foraging ecology, none did. Optimal diet and minimal uncertainty theories failed to predict honey bee foraging behaviour and, under the conditions defined by their experiments,
are refuted.

Honey bees, visiting artificial flower patches, were used as a model system to study the effects of sugar type (sucrose, glucose, fructose, and mixed monosaccharide), caloric reward, and floral colour on nectarivore foraging behaviour. Observed behaviour was compared to the predictions of various (sometimes contradictory) foraging models. Bees drank indiscriminately from flowers in patches with a blue-white flower dimorphism when caloric values of rewards were equal (e.g. 1 M sucrose in both colours; 1 M sucrose versus 2 M monosaccharide of either type), but when nectar caloric rewards were unequal, they switched to the flower colour with the calorically greater reward. In yellow-blue dimorphic flower patches, on the other hand, bees did not maximize caloric reward. Rather, bees were individually constant, some to blue, others to yellow, regardless of the sugar types or energy content of the rewards provided in the two
flower morphs.

These results suggest that optimal foraging theory (maximization of net caloric gain per unit time) is a robust predictor of behaviour with regard to the sugar types common to nectars; such optimal foraging is, however, limited by a superstructure of individual constancy (Wells et al. 1992).

Honey bees were trained to artificial floral arrays to investigate their discrimination of, and constancy to, UV (ultraviolet) as a floral colour. The floral arrays enabled colour to be varied while other floral characteristics (odour, height, etc.) remained constant. Bees were trained to an array that had only one colour morph, and then were tested on arrays to which had been added increasing frequencies of an opposing floral colour-morph. Colours tested were yellow, UV and ‘bee purple’ (yellow plus UV). Bees trained to UV or bee purple remained constant to those colours when opposing colour morphs were inserted; bees trained to yellow were variable in their constancy.

It was concluded that honey bees cannot only discriminate between and remain constant to flowers when the sole floral cue difference is the presence or absence of UV reflectance, but also show a greater constancy to flowers having at least some UV reflectance (Jones et al. 1986).

Honey bee forager use of flower pigment patterns (patterns) was examined in the context of a repetitive decision process of flower choice made within-visits that occurred over several trips to the flower patch (among-visits). Petrikin and Wells (1995) examined whether foragers can utilize pattern information alone as the basis for a complex foraging strategy, and if they can, which it is (e.g., maximization, risk aversion, individual constancy)? Three experiments were performed: 1) Blue-White Radial- pattern versus BlueWhite Bilateral-pattern, 2) Blue-White Radial-patterns with reversed colour placement, and 3) Blue versus White flowers (control). When rewards were identical in flower morphs bees foraged randomly. When rewards differed between flower morphs, bees utilized flower pattern to restrict flower visitation to the morph offering the greater caloric reward.

Forager behaviour thus conformed only to expectations of the energy maximization model. Forager error rate (choice of the flower morph offering the lower caloric reward) within pattern dimorphic flower patches, however, was 32 percent - over three times that observed when only a colour dimorphism existed. Bees changed flower morph preference usually on their first visit to the flower patch after rewards were altered, often after visiting just one flower with the lower caloric reward. Increasing accuracy in choosing the more rewarding flower morph on return trips was not generally observed, as might be expected, with gradual learning or a prolonged conditioning response.

Honey bee forager use of flower pigment patterns was examined in the context of a repetitive decision process of flower choice made within-visits that occurred over several trips to the flower patch. Lamb and Wells (1995) examined whether foragers can utilize shape (three-dimensional form) information alone as the basis for a complex foraging strategy, and if they can, which strategy is used (e.g., energy maximization, risk aversion, individual constancy)? Horizontal two-dimensional, Vertical two dimensional, and L-Shape three-dimensional flowers were used in dimorphic artificial flower patches. When rewards were identical in flower morphs, bees showed no uniform preference. Some bees foraged randomly, while other bees had flower morph preferences, but not all to the same flower type. When rewards differed between flower morphs, bees utilized flower form to restrict flower visitation to the morph offering the greater caloric reward.

Forager behaviour thus conformed only to expectations of the energy maximization model. Forager “error” rate (choice of the flower morph offering the lower caloric reward) within flower-form dimorphic patches, however, was approximately three times that observed when only a colour
dimorphism existed.

Honey bees are adept at regulating pollen stores in the colonies based on the needs of the colony. Mechanisms for regulation of pollen foraging are complex and remain controversial. Sagili and Pankiw (2007) used a novel approach to test the two competing hypotheses of pollen foraging regulation. They manipulated nurse bee biosynthesis of brood food using a protease inhibitor that interferes with midgut protein digestion, significantly decreasing the amount of protein extractable from hypopharyngeal glands. Experimental colonies were given equal amounts of protease inhibitor-treated and untreated pollen. Colonies receiving protease inhibitor treatment had significantly lower hypopharyngeal gland protein content than controls. There was no significant difference in the ratio of pollen to nonpollen foragers between the treatments. Their results supported the  pollen foraging effort predictions generated from the direct independent effects of pollen on the regulation of pollen foraging and did not support the prediction that nurse bees regulate pollen foraging through amount of hypopharyngeal gland protein biosynthesis.

Fewell and Winston (1992) examined interactions between individual foraging behaviour and pollen storage levels in the hive. Colonies responded to low pollen storage conditions by increasing pollen intake rates 54% relative to high pollen storage conditions, demonstrating a direct relationship between pollen storage levels and foraging effort. Approximately 80% of the difference in pollen intake rates was accounted for by variation in individual foraging effort, via changes in foraging activity and individual pollen load size. An additional 20% resulted from changes in the  proportion of the foraging population collecting pollen. Under both high and low pollen storage treatments, colonies returned to pollen storage levels to pre-experimental levels within 16 days suggesting that honey bees regulate pollen storage levels around a homeostatic set point. They also found a direct relationship between pollen storage levels and colony brood production, demonstrating the potential for cumulative changes in individual foraging decisions to affect colony fitness.

Honey bee foraging patterns were studied on artificial flower patches to determine if given individuals could change behaviours under differing conditions. Two types of flower patches were used; those simulating a population of flowers, dimorphic for colour and grids simulating a single colour-dimorphic inflorescence. In the simulated population of flowers, bees were individually constant to colour over a range of reward volumes and flower patch sizes. Each bee remained individually constant to a flower morph when visiting a population-type grid but changed to random visitation on the simulated inflorescence. On the simulated inflorescence, with morphs providing unequal qualities of reward, most bees foraged on the higher molarity morph. Most, but not all bees, failed to minimize uncertainty on the simulated inflorescence. On the simulated inflorescence, bees failed to optimize when one morph provided a greater reward volume than did the other. In the population of flowers bees flew from flower to flower, whereas, they walked on the simulated inflorescence
(Wells and Wells 1984).

When presented with an artificial flower patch of blue and yellow pedicellate flowers, individual honey bees became constant to one of the two flower colours, rarely even sampling the alternative colour. Some bees visited only blue flowers while others visited only yellow flowers. Hill et al. (1997) described the onset of constancy for bees that had had no experience with the experimental apparatus. In 3,020 visits, bees failed to land on or drink from the flower colour on which they landed only 17 times. This behaviour was not modified by quality or quantity of reward, training to the experimental site, group effects or presence of odour during trials.

However, when they trained bees to a target painted with two colours and then forced them to sample monomorphic flower patches in sequence, all bees visited the only colour present: yellow or blue. When they subsequently offered these same bees yellow and blue flowers simultaneously (rewarded choices), they became constant. Eleven of 23 bees showed constancy to the less rewarding flower morph without even sampling the alternative. Those bees failed to sample even though they had previously been forced to visit the alternative flower morph, which offered a reward with twice the calories/ volume. Constancy is thus spontaneous in honey bees, but it can be hidden by some experimental protocols designed to study learning. 

 References

Fewell, J.H. and M.L. Winston 1992. Colony state and regulation of pollen foraging in the honey bee, Apis mellifera L. Behav. Ecol. Sociobiol. 30: 387-393.

Free, J.B. 1963. The flower constancy of honeybees. J. Anim. Ecol.
32: 119-131.

Gegear, R.J. and T.M. Laverty 2004. Effect of a colour dimorphism on the flower constancy of honey bees and bumble bees. Canad. J. zool.
82: 587-593.

Grüter, C. , H. Moore, N. Firmin, H. Helanterä and F.L.W. Ratnieks 2011. Flower constancy in honey bee workers (Apis mellifera) depends on ecologically realistic rewards. J. Exp. Biol. 214: 1397-1402.

Hill, P.S.M., P.H. Wells and H. Wells 1997. Spontaneous flower constancy and learning in honey bees as a function of colour. Anim. Behav.
54: 615-627.

Jones, C.E., C.L. Scannell, K.J. Kramer and W.E. Sawyer 1986. Honeybee constancy to ultraviolet floral reflectance. J. Apic. Res. 25: 220-226.

Lamb, J.M. and H. Wells 1995. Honey bee (Apis mellifera) use of flower form in making foraging choices. J. Kansas Entomol. soc. 68: 388-398.

Pernal, S.F. and R.W. Currie 2002. Discrimination and preferences for pollen-based cues by foraging honeybees, Apis mellifera L. Anim. Behav. 63: 369-390.

Petrikin, J. and H. Wells 1995. Honey bee (Apis mellifera) use of flower pigment patterns in making foraging choices. J. Kansas Entomol. soc.
68: 377-387.

Sagili, R.R. and T. Pankiw 2007. Effects of protein-constrained brood food on honey bee (Apis mellifera L.) pollen foraging and colony growth. Behav. Ecol. Sociobiol. 61: 1471-1478.

Wells, H. and P.H. Wells 1983. Honey bee foraging ecology: optimal diet, minimal uncertainity or individual constancy? J. Anim. Ecol. 52: 829-836.

Wells, P.H. and H. Wells 1984. Can honey bees change foraging patterns? Ecol. Entomol. 9: 467-473.

Wells, H., P.S. Hill and P.H. Wells 1992. Nectarivore foraging ecology: rewards differing in sugar types. Ecol. Entomol. 17: 280-288.

The ABK thanks Clarence for permission to reprint his articles from BEE CULTURE.