Abstract.(1) One hundred and two plants were offered to locust nymphs after a 5 h period without food. Large meals were only taken on grasses and onJuncus.Most dicotyledons were totally rejected.(2) All the plants which were relatively unpalatable yielded an extract which reduced the amount eaten, but no inhibition of feeding was induced by extracts of plants which were freely eaten. Very few of the extracts increased the amount eaten.(3) Over 100 chemicals, mostly non‐nutritional, known to occur in leaves, were tested for their inhibitory effect on feeding. Some had no effect, but most progressively inhibited feeding as their concentration was increased until no feeding at all occurred. Chemicals of many classes had this effect, but alkaloids and monoterpenoids were the most consistently deterrent at the lowest concentrations.(4) It is concluded that the failure to eat most plants results from the presence in the leaves of one or more chemicals in amounts which inhibit feeding. Grasses and a few other plants are readily acceptable because they do not contain deterrent chemicals in sufficient quantity to limit feedin