Ants on Plants: Distribution of visitation across plant taxa and plant reproductive condition

Background/Question/Methods: Many plants offer extrafloral nectar rewards to attract ants that may subsequently protect the plant against natural enemies (herbivores).  The phylogenetic, phenologic and geographical distribution of plant visitation by ants is not well understood, although there is evidence that extrafloral nectar production can be a widespread within desert plant communities.  To explore the visitation phenomenon, we monitored native Sonoran Desert plant species in the arboretum of the Arizona Sonora Desert Museum outside Tucson, Arizona, USA.  The frequency with which plants acted as a foraging surface for ants was inspected for 92 plant species in 12 families (n >1200 inspections). The presence/absence of ants on each plant was described based on a 30 second inspection of the plant (or an inspection of the entire plant if less time was required).  Surveys were repeated in each of the five seasons of the Sonoran Desert (winter, spring, dry summer, monsoon, autumn) across 2011-2012.  Because reward production in some plants is known to change quantitatively and/or qualitatively during certain stages when protection is particularly valuable, we recorded the presence/absence of flower buds and flowers on each plant at the time of inspection.

Results/Conclusions: The average frequency of ant visitation increases by approximately four-fold during reproduction.  This pattern was robust when based on paired comparisons within the 38 species where both reproductive and non-reproductive stages have been inspected (global species average = 24 and 6% of inspections, respectively; paired t = 3.3, df = 38, p = 0.001) and when limited to the species with at least 5 inspections of both reproductive and non-reproductive individuals; paired t = 2.72, df = 6, p = 0.03).  Frequent visitation to non-reproductive individuals (ants detected in at least 20% of inspections) was limited to two cactus genera (Ferocactus wislizeni, F. rectispinus, F. peninsulae and F. cylindraceus and Opuntia wilcoxii, O. leptocaulis). We conclude that enhanced visitation of plant by ants may be a common feature of plant reproduction in desert environments, and infer that this interaction could influence ant and plant communities in contexts where plants are protected by ants, ants are nourished by plants, and/or ants interact with pollinators or pollen.  We also speculate that plant species that provide rewards throughout the year (i.e., even when not producing buds and flowers) have a particular capacity to modify adjacent ant communities.