Adaptation, genetic drift, pleiotropy, and history in the evolution of bee foraging behavior

Abstract

Our goal in this review is to determine whether particular behavioral traits represent actual adaptations in the context of foraging. Social bees are our chosen study system because they provide a convenient and tractable biological system with which to study the potential adaptiveness of a wide range of foraging traits, such as flower constancy, floral color preference, learning to associate floral color as a predictor of reward, traplining, and communication about food sources. This variety of behavioral traits allows us to demonstrate the strengths, and weaknesses, of applying five approaches (four experimental and one theoretical) to the study of foraging at the species, population, and colony level. (1) The comparative approach allows us to contrast behavioral traits of extant species with those of their common ancestor. We correlated differences in floral color preference between closely related species (and populations), with a known phylogeny, with features in each bee's respective environment. (2) Reciprocal transplant experiments allowed us to test for local adaptation. We compared the relative foraging performance of distinct bee populations in both of their respective native environments. (3) Manipulating the foraging environment to eliminate specific behavioral traits permitted a direct comparison of animals' foraging performance in their normal and experimentally manipulated environment, allowing us to quantify the effect of the trait in question (traplining) on foraging performance. (4) Manipulating the foraging phenotype to eliminate specific behavioral traits is another valuable approach. Unless suitable behavioral mutants, knockouts, or molecular techniques to selectively block gene expression are available, creating such artificial foraging phenotypes is only possible for a very small number of specific traits, e.g. the honeybee dance language. (5) Integrating biologically realistic modeling with experimental studies allows us to test predictions about the adaptive significance of foraging related traits not amenable to experimental manipulation, and to identify the ranges over which these traits might affect fitness. Do these approaches provide evidence that foraging behaviors are adaptive? In some cases, we show that forager behavior has indeed been tuned to function adaptively in a given niche: interestingly the adaptive benefits of such behavioral traits are often strongly context dependent. However in other cases, the observed patterns of behavior were more parsimoniously explained by chance evolutionary processes, or by the historical conditions under which bees operated in their evolutionary past.