Super-generalist flowers attract high numbers of bees and even higher numbers of non-bee flower visitors

Background/Question/Methods

Floral resources are important for a great diversity of life: up to 300,000 animals are regular flower visitors. Flower visitors range from obligate to opportunistic, and from efficient to inefficient pollinators. Plant generality, defined as the number of insects visiting a plant species for floral resources, can range from low on specialist flowers, to quite high for generalist flowers. Generalist flowers could have great impacts on the spread of flower-dwelling microbes such as insect pathogens, viruses, and nectar-inhabiting microbes. Thus, the variationin generality among plant species in a community could have great impact on the spread of such microbes. However, the variation in generality of flowers for different taxa of flower visitors is largely unknown. Using Charles Robertson’s historical flower visitor dataset, I evaluated how the distribution of generality among plant species varies when considering their interactions with different groups of flower visitors. Specifically, I compared the observed plant generality distributions 1) to a Poisson distribution, and 2) among groups of flower visitor species: pollen-gathering-specialists (non-parasitic bees) and other orders of flower visitors (other Hymenoptera, Coleoptera, Diptera, and Lepidoptera) using the Kolmogorov-Smirnov test (K-S test). 

Results/Conclusions

Plant generality distributions for each group deviate from the Poisson distribution, indicating that interactions between plant species and flower visitor species do not occur at random. In all groups, there are plants with more species of visitors than expected by chance; some plants were visited by >100 species. These plants could be considered super-generalists.

The K-S tests show that all distributions are significantly different from each other (p<0.05), but the plant generality distribution for non-parasitic bees is most different than all of the others (D>0.4). Most plant species were visited by a modest number of pollen-collecting bee species, with relatively few moderate-strong generalists. In contrast, plant generality with respect to other flower visiting insect groups (parasitic bees and other Hymenoptera, Coleoptera, Diptera) was highly skewed: many plants had zero visitors whereas a few had very large numbers of visitors.

The presence of such super-generalist plants could have important community implications, especially for the transmission of flower-borne microbes. I am currently testing this hypothesis in the field by 1) determining whether different flower visitors disperse unique nectar microbe communities, and 2) surveying plant communities to see if more generalist plants are associated with a more diverse nectar microbe community.