Key points are not available for this paper at this time.
Simple models representing nutrient-dependent growth of plant prey are used to study herbivores and edible and "inedible" plants. Three types of inedible plants are considered: those that have no direct interaction with herbivores, those that interfere with the ingestion of edible plants but are not themselves consumed, and those that are consumed but provide no nutritional benefit to herbivores. A necessary condition for coexistence at equilibrium applies for all types of inedible plants, which requires that edible plants be better competitors than inedible plants in the absence of herbivores but worse in their presence. All models predict compensatory changes in relative densities of edible and inedible plants at equilibrium as herbivore loss rate changes. If inedible plants do not interact with herbivores, then only inedible (not edible) plant biomass increases with nutrient supply; otherwise, biomasses of both edible and inedible plants increase with nutrient supply. If inedible plants interfere with ingestion of edible plants but are not consumed, then equilibria are stable in the parameter space examined, for a parameterized model of a planktonic system. In the other cases examined, equilibria are unstable in most of the parameter space examined.
James P. Grover (Mon,) studied this question.