Unclusterable, underdispersed arrangement of insect-pollinated plants in pollinator niche space.

Pollinators can mediate facilitative or competitive relationships between plant species, but the relative importance of these two conflicting phenomena in shaping community-wide pollinator resource use remains unexplored. This article examines the idea that the arrangement of large samples of plant species in Hutchinsonian pollinator niche space (n-dimensional hypervolume whose axes represent pollinator types) can help to evaluate the comparative importance of facilitation and competition as drivers of pollinator resource use at the community level. Pollinator composition data were gathered for insect-pollinated plants from the Sierra de Cazorla mountains (southeastern Spain), comprising ~95% of widely distributed insect-pollinated species. The following questions were addressed at regional (45 sites, 221 plant species) and local (1 site, 73 plant species) spatial scales: (1) Do plant species clusters occur in pollinator niche space? Four pollinator niche spaces differing in dimensionality were considered, the axes of which were defined by insect orders, families, genera, and species. (2) If all plant species form a single, indivisible cluster, are they overdispersed or underdispersed within the cluster relative to a random arrangement? "Clusterability" tests failed to reject the null hypothesis that there was only one pollinator-defined plant species cluster in pollinator niche space, irrespective of spatial scale, pollinator niche space, or pollinator importance measurement (proportions of pollinator individuals or flowers visited by each pollinator type). Observed means of interspecific dissimilarity in pollinator composition were smaller than randomly simulated values in the order-, family-, and genus-defined pollinator niche spaces. This finding revealed an underdispersed arrangement of plant species in each of these pollinator niche spaces. In the undisturbed montane habitats studied, arrangement of insect-pollinated plant species in the various niche spaces defined by pollinator composition did not support a major role for interspecific competition as a force shaping community-wide pollinator resource use by plants, but rather suggested a situation closer to the facilitation-dominated extreme in a hypothetical competition-facilitation gradient. Results also highlight the importance of investigations on complete or nearly complete insect-pollinated plant communities for suggesting and testing novel hypotheses on the ecology and evolution of plant-pollinator systems.