Susceptibility to herbivores depends on recent history of both the plant and animal

Physical stress to seaweeds and hunger stress of herbivores can influence the outcome of chemically mediated seaweed-herbivore interactions. The unpalatable brown seaweed Dictyota ciliolata produces the diterpenoid secondary metabolites pachydictyol A, dictyol B acetate, and dictyodial. At natural concentrations, pachydictyol A deterred the sea urchin Arbacia punctulata but did not inhibit feeding by the pinfish Lagodon rhomboides or the amphipod Ampithoe longimana until concentrations were 2.5-5 times natural levels. Dictyol B acetate deterred the urchin, amphipod, and pinfish at, or far below, natural concentrations. Dictyodial was too unstable to assay directly, but indirect experiments suggested that natural concentrations deterred the urchin, but not the pinfish or amphipod. Mild desiccation of D. ciliolata reduced concentrations of the different secondary metabolites by 7-38% and plants became 2.6-3.4 times more susceptible to urchin and amphipod grazing. The combined concentrations of pachydictyol A and dictyol B acetate found in undesiccated Dictyota ciliolata deterred feeding by urchins, but this deterrent effect was lost at concentrations found in the desiccated plants. Desiccated and undesiccated plants did not differ in nutritive value (as measured by protein and total N content) or toughness. Thus, desiccated plants became more palatable because chemical defenses were lost, not because nutritive value was increased. The stress of near-surface ultraviolet radiation also caused significant physiological changes in Dictyota ciliolata. UV-exposed blades bleached, senesced, and grew 84% less than blades protected from UV radiation. Tissue loss and minimal growth of UV-stressed plants constrained our sample sizes, but the limited assays that could be run suggested that UV stress may lower chemical defenses and increase plant susceptibility to herbivores. Because many previous investigations of herbivore feeding patterns used animals that had been starved for days before an assay, we tested the effects of this commonly used procedure on feeding discrimination. Recently fed urchins always avoided food containing natural concentrations of pachydictyol A during separate feeding trials performed on each of four consecutive days. In contrast, urchins deprived of food for 3 d before this assay did not avoid the treated food on days 1 and 2 of feeding trials, but did avoid it on days 3 and 4 after their hunger was reduced by feeding during days 1 and 2. If we had used only starved urchins (a common procedure in previous investigations), we could have concluded, with apparent justification, that urchins were unaffected by pachydictyol A (if the assays were run for only 1-2 d) or that they needed 2 d of exposure to the compound in order to learn to avoid it. Both of these conclusions would have been incorrect.