Estimating the Impact of Lygus hesperus on Cotton: The Insect, Plant, and Human Observer as Sources of Variability

Unexplained variation in the relationship between herbivore densities and the shortterm appearance of crop damage is sometimes observed in pest management. Here we used a Þeld survey of commercial cotton Þelds and a linked questionnaire for cooperating pest control advisors to document the existence of such unexplained variation in the impact of the western tarnished plant bug, Lygus hesperus, on upland cotton,Gossypium hirsutum. L. hesperus feeds preferentially on ßower buds (“squares”), and theplantmay respond to this damagewith abscission of the square.Weexplored four classes of factors that might contribute to unexplained variation in square abscission. First, misperceptions by the human observer (i.e., sampling problems)may play a role, because commercial Þeld scouts signiÞcantly underestimate densities of L. hesperus nymphs. Second, we found no support for the hypothesis that variable behavior expressed byL. hesperus contributes to unexplained variation in square abscission. L. hesperus seems to generate relatively predictable levels of square feeding damage; the variation that was observed was unrelated to grower categorization of Þelds as exhibiting normal versus unexpectedly high or low levels of square abscission. Third, variable plant responses to damagemay instead be the key source of unexplained square abscission. Younger plants and plants with higher petiole phosphate concentrations expressed increased sensitivity to L. hesperus feeding; these correlations must, however, be tested experimentally before deÞnitive conclusions are drawn. Fourth, another arthropod might be generating damage that was erroneously being attributed to L. hesperus. The omnivoreGeocoris pallenswas a candidate agent of cryptic damage to squares; however, an experiment showed thatG. pallens generated only trivial square damage andnodetectable increase in square abscission. Thus, this study has focused our attention away from the arthropod side of the interaction and toward the host plant as the primary source of greater than expected square abscission generated by L. hesperus.