Variation in reward quality and pollinator attraction: the consumer does not always get it right

• Published in: AoB plants 2015-04, Vol.7
• Main Authors: Carr, David E., Haber, Ariela I., LeCroy, Kathryn A., Lee, De'Ashia E., Link, Rosabeth I.
• Format: Article
• Language: English
• Subjects: Bees; Inbreeding; Proteins; SPECIAL ISSUE: Using Non-Model Systems to Explore Plant–Pollinator and Plant–Herbivore Interactions
• ISSN: 2041-2851; 2041-2851
• DOI: 10.1093/aobpla/plv034

Bees depend on pollen as the primary protein source for their larvae and should be strongly selected to identify cues associated with the most rewarding flowers. We examined the ability of bumble bees (Bombus impatiens) to identify the most rewarding foraging opportunities using arrays of live monkeyflowers (Mimulus guttatus), artificial plants, and pairwise olfactory tests. Bees could identify pollen rewards by scent and tended to visit the most rewarding artificial flowers. They seemed less able to identify the best pollen sources when foraging on live plants. We suggest that live plants may provide conflicting or deceptive signals to pollinators. Abstract Nearly all bees rely on pollen as the sole protein source for the development of their larvae. The central importance of pollen for the bee life cycle should exert strong selection on their ability to locate the most rewarding sources of pollen. Despite this importance, very few studies have examined the influence of intraspecific variation in pollen rewards on the foraging decisions of bees. Previous studies have demonstrated that inbreeding reduces viability and hence protein content in Mimulus guttatus (seep monkeyflower) pollen and that bees strongly discriminate against inbred in favour of outbred plants. We examined whether variation in pollen viability could explain this preference using a series of choice tests with living plants, artificial plants and olfactometer tests using the bumble bee Bombus impatiens. We found that B. impatiens preferred to visit artificial plants provisioned with fertile anthers over those provisioned with sterile anthers. They also preferred fertile anthers when provided only olfactory cues. These bumble bees were unable to discriminate among live plants from subpopulations differing dramatically in pollen viability, however. They preferred outbred plants even when those plants were from subpopulations with pollen viability as low as the inbred populations. Their preference for outbred plants was evident even when only olfactory cues were available. Our data showed that bumble bees are able to differentiate between anthers that provide higher rewards when cues are isolated from the rest of the flower. When confronted with cues from the entire flower, their choices are independent of the quality of the pollen reward, suggesting that they are responding more strongly to cues unassociated with rewards than to those correlated with rewards. If so, this suggests that a sen