Arthropod-Plant Interactions

, Volume 6, Issue 4, pp 561–569

Why ant pollination is rare: new evidence and implications of the antibiotic hypothesis

Original Paper

DOI: 10.1007/s11829-012-9201-8

Cite this article as:
Dutton, E.M. & Frederickson, M.E. Arthropod-Plant Interactions (2012) 6: 561. doi:10.1007/s11829-012-9201-8


The antibiotic hypothesis proposes that ant pollination is rare at least in part because the cuticular antimicrobial secretions of ants are toxic to pollen grains. We tested this hypothesis by comparing the effects of ants and bees on pollen in two regions: a tropical rainforest in Amazonian Peru and temperate forests and old fields in Canada. We found support for three predictions that follow from the antibiotic hypothesis. (1) For all 10 ant and 11 plant species in our study, contact with ants significantly reduced pollen germination, confirming the generality of this effect. (2) Contact with two bee species did not have similar effects; pollen exposed to bees germinated as well as control pollen. (3) Consistent with the presumed greater abundance of entomopathogens in the tropics, which may have selected for stronger antibiotic secretions in tropical ants, tropical ants had more negative effects on pollen than temperate ants. We speculate that the antibiotic hypothesis contributes not only to the rarity but also to the biogeography of ant pollination, and we discuss whether the negative effects of ants on pollen have resulted in selection for floral defenses against ants.


Ant–plant interactionsCuticular antimicrobial secretionsDisease resistanceInsect immunityHymenopteraTropical rainforests

Copyright information

© Springer Science+Business Media B.V. 2012

Authors and Affiliations

  1. 1.University of OttawaOttawaCanada
  2. 2.Department of Ecology and Evolutionary BiologyUniversity of TorontoTorontoCanada