FEEDING PREFERENCES OF A GENERALIST SALT-MARSH CRAB - RELATIVE IMPORTANCE OF MULTIPLE PLANT TRAITS

Few studies have evaluated the relative importance of multiple plant t raits to herbivore diet choice, especially with an experimental approa ch. Moreover, although circumstantial evidence points to plant toughne ss and silica content as important determinents of diet choice, few st udies have experimentally demonstrated that these factors actually det er feeding by herbivores. We examined feeding preferences of a general ist saltmarsh crab, Armases cinereum, for all the common angiosperms i n its habitat. We took an experimental approach to evaluating the impo rtance of toughness, secondary chemistry, silica, salt, and protein in determining feeding preferences. Consumption of plants by Armases in two experiments was correlated with decreasing toughness. Consumption was more equitable when plants were ground up and reconstituted in aga r discs, with reduced differences in toughness. In four pairwise choic e tests, Armases always preferred the softer plant of the pair, but in three of four cases exhibited no preference when the same plants were presented in reconstituted discs. Several plant extracts significantl y stimulated or deterred feeding by Armases, but these effects were no t consistent with and could not be used to predict overall preferences for fresh plants. Salt in artificial diets stimulated feeding by Arma ses. Silica had no effect on feeding in two experiments and stimulated feeding in a third. Armases was mildly stimulated to feed by high lev els of protein in artificial diets, but these levels were above those found in plants, and the protein content of plants did not correlate w ith feeding preferences. Our results point towards plant toughness as most important in determining feeding choices of Armases. Since Armase s is omnivorous, it may escape dietary constraints that affect feeding choices of pure herbivores. The dogma that silica defends plants agai nst herbivory needs to be re-examined with additional experimental stu dies.