Suitability of stressed and vigorous plants to various insect herbivores

We conducted a controlled experiment to test the plant vigor and the plant stress hypotheses. The two hypotheses associate plant physiological conditions to insect feeding mode and performance. We exposed tomato, Lycopersicon esculentum, to different types of growing conditions: optimal (vigorous plants), resource based stress (water and/or nutrient deficit), and physical stress (punched hole in terminal leaflets). Plant performance, foliar nutritional value for insects and chemical defenses were analyzed after 14 d. These plants were offered to insects belonging to distinct feeding guilds: the silverleaf whitefly, Bemisia argentifolii, a phloem feeder; the leafminer, Liriomyza trifolii; and the corn earworm, Heliothis zea, a leaf chewing caterpillar. The experimental conditions generated a gradient of plant growth in the following order: optimal (vigorous) > control = hole punched > no fertilizer > no water > no water and no fertilizer, The last two treatments resulted in plants with poor nutritional value (based on %water, C/N, %N) and higher levels of defensive compounds (i.e., peroxidase and total phenolics) compared with control and the vigorous plants. Hole-punching neither affected plant growth nor any of the phytochemicals measured. In a choice experiment adult whitefly ovipositioning was not affected by either vigor or punching but was reduced on the other plants (P<0.01). Leafminer feeding and oviposition and corn earworm larval growth rates were higher on the vigorous plants and lower on the punched, no fertilizer, no water, and no water and no fertilizer host plants (P<0.01). Regardless of insect species or bioassay method, the results in the tomato system support the plant vigor hypothesis that predicts positive association between insect performance and plant growth. The results contradict the plant stress hypothesis that rank stressed plants as better hosts for insects. The mechanisms involved are a combination of poor nutritional value and chemical defenses. We demonstrate a negative association between plant growth and chemical defense. However, induced response triggered by hole-punching was not cost effective to the plants.