Toleration games: compensatory growth by plants in response to enemy attack is an evolutionarily stable strategy.

Damage to plants from natural enemies is a ubiquitous feature of the natural world. Accordingly, plants have evolved a variety of strategies to deal with attack from enemies including the ability to simply tolerate attack. Tolerance often involves some form of compensatory response, such as the regrowth of tissues following damage. While ecological models of defence are common, there has been less effort to make predictions about the evolutionary stability of tolerance. Here, we present and experimentally test a game theoretic model of tolerance to herbivory. Plants in the model have a vector strategy which includes both root and shoot production, and herbivores in the model have a scalar strategy which is time spent foraging. The evolutionarily stable strategy (ESS) is the set of root growth, shoot growth and herbivore foraging which simultaneously maximizes all player's fitness. Compensatory growth is not guaranteed, but it may emerge as an ESS if it maximizes plant fitness. We also experimentally tested the model predictions using wheat and simulated herbivory by clipping 0, 15, 30, 45 or 60 % of shoot production, and measured root, shoot and fruit production at senescence. The model predicted that compensatory growth was often an ESS when herbivores were either above- or below-ground. Plants in the experiment followed model predictions. Specifically, plants produced more tissues than expected based on damage, and for 15 % damage this allowed them to maintain equal fitness compared to undamaged plants. The model allows for above- and below-ground herbivory to be modelled, and predicts their impact on whole plant growth and reproduction. For example, we can predict the effects of shoot damage on root growth. When combined with other advances in predicting plant ecology with evolutionary game theory, we anticipate that this will be a valuable tool for generating further testable hypotheses.