Geographic mosaic theory of coevolution in plant and pollinator interaction: Process, evidence, prospect

Abstract

Coevolution refers to evolutionary changes in the traits of two or more interacting species caused by reciprocal natural selection. Geographic mosaic theory of coevolution (GMC) proposes that the dynamics of coevolution between pairs or group of species often occur at different geographic scales. There are three hypotheses on the coevolutionary process. (1) Different evolutionary interaction trajectories among populations are likely to generate a selection mosaic. (2) A mixture of coevolutionary hotspots and coldspots. Hot spots are the subset of communities in which much of the coevolutionary change occurs and coldspots refer to areas where coevolutionary selection hardly occurs or completely does not occur. (3) There is a continual population remixing of the range of coevolving traits, resulting from the selection mosaic, coevolutionary hotspots, gene flow, random genetic drift, and local extinction of subpopulations. Studies of GMC between plants and pollinators indicate that shifts of effective pollinators among different geographic populations may cause divergence in floral traits (i.e., pollination ecotype). Changes of pollinator assemblages or species components have been illustrated coevolving with floral morphology or reward among populations from different altitudes or habitats. Here, recent case studies of GMC between plants and pollinators are grouped into five categories: Floral morphology and pollinator components, corolla tube/nectar spur and pollinator feeding structure, floral color, floral scents, floral rewards and pollinator assemblages. Floral tube or nectar spur length is positively related to pollinator body structures, such as the proboscis and leg length. The adaptation of flower color and pollinator vision provides a series of examples for the GMC. With the increase of altitude, flower color changed from blue and ultraviolet blue adapted to bee vision to blue-green and green adapted to fly vision. Recent study indicated that the difference in cycad odor components related to a shift of beetle species in different regions, suggesting a potential coevolutionary relationship between flower odor and pollinators. Studies showed that where bees foraged more nectar, flowers secreted more nectar, suggesting that floral reward amounts and types coevolve with pollinators among different geographic populations. An examination of GMC in plant-pollinator interactions involves three steps in general: (1) Detection of the relationship between floral traits and pollinator shift or body structure. (2) Transplanting experiments to detect differences in seed set between native phenotypes and transplanted phenotypes. (3) Single pollinator visit experiment to detect pollen output differences between native and transplanted phenotypes. Experimental evolutionary biology not only can test whether floral variation is related to pollinator shifts, but also make people aware of the impact of the reduction or loss of pollinators on the evolutionary consequences. To illustrate GMC between plants and pollinators, it is essential to isolate the evolutionary changes in traits caused by biotic factors or abiotic factors in future. It would be highly appreciated if genome scans and analysis are used in further studies to explore the genome-wide patterns of GMC between plants and pollinators.