The impact of pollinator community diversity on trait-matching in plant-pollinator interactions is poorly studied, even though many mutualisms involve multiple interaction partners. We studied ten communities in which one to three species of oil-collecting Rediviva bee pollinate the long-spurred, oil-producing flowers of Diascia ‘floribunda’ to examine how pollinator diversity affects covariation of functional traits across sites and trait-matching within sites. Floral spur-length was significantly correlated with weighted grand mean foreleg length of the local bee community but not with foreleg length of individual bee species. The closeness of trait-matching varied among populations and was inversely related to pollinator community diversity. For all bee species, trait-matching was closest at sites characterized by exclusive, pairwise interactions. Reduced trait-matching associated with increased community diversity for individual pollinator species, but close matching at the community level support the importance of community context for shaping interacting traits of flowers and pollinators.

The dataset contains the data of forleg length and spur length of all collected Rediviva bees and Diascia flowers at each site. 

Forelegs of 2–56 females of each Rediviva species per site were measured. Bees were collected with a sweep net and stored in 96% ethanol. The right foreleg was removed and foreleg segments (disti + mediotarsus, basitarsus, tibia, femur, trochanter, coxa) were measured to the nearest 0.01 mm using a Keyence VHX 500 F digital microscope following the method described by Steiner and Whitehead (1990). Foreleg length was calculated as the sum of the individual foreleg segments.

The right spurs of freshly collected flowers of Diascia floribunda were longitudinally sectioned prior to measurement. Length of spur entrance area, proximal and apical elaiophores fields and the gap between fields were measured to the nearest 0.01 mm using a Dino-Lite Digital Microscope AD-413T as per the methods described in Hollens et al. (2017). Total spur length was calculated as the sum of the individual segments.

Hollens H, van der Niet T, Cozien R, Kuhlmann M. 2017 A spur-ious inference: Pollination is not more specialized in long-spurred than in spurless species in Diascia-Rediviva mutualisms. Flora 232, 73–82. (doi:10.1016/j.flora.2016.12.006)

Steiner KE, Whitehead VB. 1990 Pollinator adaptation to oil-secreting flowers – Rediviva and Diascia. Evolution 44, 1701–1707. (doi:10.1111/j.1558-5646.1990.tb03857.x)