Another approach has been to document plant– insect associations in the all-too-frequent absence of relevant body–fossil data in more stratigraphically focused intervals that span an extinction event (Labandeira et al. 2002a, 2002b; Fig. 5c). Such ecological data provide an alternative framework for assessing insect diversity, origination, and extinction patterns, specifically through the effects of herbivorous insects on plants. However, such an approach is relevant for only primary consumers of plants, herbivorous insects, which constitute 50–60% of all extant species (Price 1997), but they also constitute the majority of body–fossil occurrences in many insect-bearing deposits. A major extinction of herbivores, however, should have major, cascading consequences on dependent insects such as predators, parasitoids, and parasites (Godfray et al. 1999). Although never explicitly applied to extinction in the fossil record, many molecular studies of coevolved plant and insect taxa from the present have provided estimates of the temporal ranges for the period of initial association of host–herbivore systems. A sufficient number of these temporal ranges, gleaned from various studies in the plant–insect associational literature, can document temporal gaps that indicate the presence of significant disruptive events separating clusters of associations linking older associated plant and insect clades from newer such clades (Labandeira 2002a; Fig. 5d).