Start making scents: the challenge of integrating chemistry into pollination ecology

Although research on plant volatiles and pollination ecology has grown explosively over the past 15 years, there remains little dialogue between these fields. Here I examine the historical and cultural reasons for this impasse, focusing on the ways that questions in each field are addressed, and the potential for productive cross-talk. The specialization-generalization debate in pollination has cast doubt on the importance of sensory biology in mediating plant-pollinator interactions on the community scale. However, chemical 'filters' of volatile or nectar-borne repellents are likely to explain the absence of specific interactions in plant-pollinator webs. In addition, the omission of plant volatiles from path analyses measuring the relative impacts of herbivores and pollinators on plant fitness may be one reason for large unexplained variance terms in such models. Floral scent functions in concert with visual and gustatory cues by attracting pollinators from a distance, increasing approaches and landings, and mediating outcrossing rates through changes in visitation frequency and duration. All dimensions of floral chemistry, including ontogenetic and diel variation in scent emissions, have the potential to respond to balancing selection between herbivores and pollinators. The available data reveal that chemical aspects of floral phenotypes are important across the specialization-generalization spectrum, and thus are widely applicable to mainstream pollination ecology.