Some current approaches to the evolution of plant–herbivore interactions

Abstract ‘Coevolution’ between plants and herbivorous arthropods has several meanings: cospeciation, reciprocal adaptation, and a history of ‘escape‐and‐radiation’. Few well documented examples of each are known. Most evolutionary research on insect‐plant interactions concerns the adaptations of insects to plants or of plants to insects, but little of it expressly addresses reciprocal adaptation. Modern phylogenetic research confirms that host associations in many clades of specialized insects are evolutionarily conservative. An example from leaf beetles ( Ophraella ) is presented, in which the historical conservatism of host shifts is mirrored by patterns of paucity of genetic variation in the ability to use novel hosts. In several species of Ophraella , genetic variation was more often discerned in responses to plants closely related to the insect’s normal host plant than to more distantly related plants. Thus availability of genetic variation might bias the evolution of host shifts, and account for the phylogenetic pattern. The difficult problem of showing that chemical and other resistance features of plants evolved for their defensive function is slowly yielding to investigation. One difficulty is that most insect herbivores are thought to be usually too rare to impose appreciable selection. Insects collectively could exert quite strong selection on resistance characters if these characters have genetically correlated effects across many species of herbivores (i.e., if selection were ‘diffuse’). Surprisingly, the little available evidence does not suggest that pervasive genetic correlations are common. I conclude that more, and more multidisciplinary, research is needed to understand the evolution of plant resistance and defense.