Bibliography: pages 47-55. / The statement that similar sorts of plants often have similar medicinal properties (Le. contain similar chemicals) is at least 300 years old, while the concept probably dates back several thousand years (Stace, 1980). Indeed, a knowledge of the secondary chemicals produced by different plant species has played an important role since the early stages of man's cultural, and probably evolutionary, development (e.g. Leopold and Ardrey, 1972). Until recently, however, the biological role of this large group of compounds has remained largely obscure, with many plant physiologists regarding them as waste products and of no possible survival value to plants (Harborne, 1982). In recent years a tremendous increase in attention paid to these compounds has led to the suggestion that plant secondary substances evolved as herbivore deterrents, and diversified with the plants producing them as herbivores evolved various means of coping with their toxicity. In what follows I present an overview of events leading to the establishment of this theory, and a history of the discovery that some insect herbivores have turned the toxicity of plant secondary compounds to their own advantage, thereby becoming exclusively associated with toxic food plants. It will be seen that although this theory of plant-herbivore coevolution has become generally accepted, it has yet to be tested for a wide range of herbivores and chemical types. From this view I have taken a detailed look at a specific group of herbivores and plant toxins: Lepidoptera feeding on cyanide-producing plants.