The effects of leaf toughness, trichome characteristics, nutrient and water availability, and secondary chemistry on diet selection by the leafcutting ant Atta cephalotes were determined using individuals of 49 woody plant species from a tropical deciduous forest in Costa Rica. The palatability of plants was measured by presenting whole leaves to leafcutting ant foragers on trails and counting the number of pieces cut from leaves. The effectiveness of physical barriers to leaf cutting was evaluated by measuring the speed with which ants were able to cut leaves of varying toughness, trichome densities, and trichome lengths. Plant secondary chemistry was highly correlated with the relative palatability of leaves, while palatability was only marginally correlated with trichome characteristics and plant nutritional quality and was uncorrelated with leaf toughness and water content. Highly palatable leaves generally contained hydrolyzable tannins, while most unpalatable leaves possessed nonpolar extracts found to deter leaf cutting in laboratory bioassays. Within the subset of highly palatable plants, leaf protein content was positively correlated with the number of pieces cut, indicating that secondary chemistry and nutrient availability interact to determine palatability. Decreases in plant palatability over time paralleled decreases in protein content. Ant size was the most important variable determining the rate at which ants cut leaves. Trichome characteristics were uncorrelated with the rate at which leaves were cut, while leaf toughness explained a small amount of the variance in cutting rate. The results suggest that leaf toughness and trichome characteristics within the ranges encountered by leafcutting ants in this tropical deciduous forest do not generally deter attack by ants. Instead, the results are consistent with the idea that leafcutting ants use a combination of plant secondary chemistry and nutrient availability to evaluate the quality of leaves.
Howard, J.J. (1988). Leafcutting ant diet selection: relative influence of leaf chemistry and physical features. Ecology 69 (1): 250-260.